chapter O xidative phosphorylation is the culmination of energy- yielding metabolism in aerobic organisms. All oxi- dative steps in the degradation of carbohydrates, fats, and amino acids converge at this final stage of cellular respiration, in which the energy of oxidation drives the synthesis of ATP. Photophosphorylation is the means by which photosynthetic organisms capture the energy of sunlight—the ultimate source of energy in the bio- sphere—and harness it to make ATP. Together, oxida- tive phosphorylation and photophosphorylation account for most of the ATP synthesized by most organisms most of the time. In eukaryotes, oxidative phosphorylation occurs in mitochondria, photophosphorylation in chloroplasts. Oxidative phosphorylation involves the reduction of O 2 to H 2 O with electrons donated by NADH and FADH 2 ; it occurs equally well in light or darkness. Photophosphor- ylation involves the oxidation of H 2 O to O 2 , with NADP H11001 as ultimate electron acceptor; it is absolutely dependent on the energy of light. Despite their differ- ences, these two highly efficient energy-converting processes have fundamentally similar mechanisms. Our current understanding of ATP synthesis in mi- tochondria and chloroplasts is based on the hypothesis, introduced by Peter Mitchell in 1961, that transmem- brane differences in proton concentration are the reser- voir for the energy extracted from biological oxidation reactions. This chemiosmotic theory has been ac- cepted as one of the great unifying principles of twen- tieth century biology. It provides insight into the processes of oxidative phosphorylation and photophos- phorylation, and into such apparently disparate energy transductions as active transport across membranes and the motion of bacterial flagella. Oxidative phosphorylation and photophosphoryla- tion are mechanistically similar in three respects. (1) Both 19 690 OXIDATIVE PHOSPHORYLATION AND PHOTOPHOSPHORYLATION If an idea presents itself to us, we must not reject it simply because it does not agree with the logical deductions of a reigning theory. —Claude Bernard, An Introduction to the Study of Experimental Medicine, 1813 The aspect of the present position of consensus that I find most remarkable and admirable, is the altruism and generosity with which former opponents of the chemiosmotic hypothesis have not only come to accept it, but have actively promoted it to the status of a theory. —Peter Mitchell, Nobel Address, 1978 OXIDATIVE PHOSPHORYLATION 19.1 Electron-Transfer Reactions in Mitochondria 691 19.2 ATP Synthesis 704 19.3 Regulation of Oxidative Phosphorylation 716 19.4 Mitochondrial Genes: Their Origin and the Effects of Mutations 719 19.5 The Role of Mitochondria in Apoptosis and Oxidative Stress 721 PHOTOSYNTHESIS: HARVESTING LIGHT ENERGY 19.6 General Features of Photophosphorylation 723 19.7 Light Absorption 725 19.8 The Central Photochemical Event: Light-Driven Electron Flow 730 19.9 ATP Synthesis by Photophosphorylation 740 8885d_c19_690-750 3/1/04 11:32 AM Page 690 mac76 mac76:385_reb: processes involve the flow of electrons through a chain of membrane-bound carriers. (2) The free energy made available by this “downhill” (exergonic) electron flow is coupled to the “uphill” transport of protons across a proton-impermeable membrane, conserving the free energy of fuel oxidation as a transmembrane electro- chemical potential (p. 391). (3) The transmembrane flow of protons down their concentration gradient through specific protein channels provides the free energy for synthesis of ATP, catalyzed by a membrane protein complex (ATP synthase) that couples proton flow to phosphorylation of ADP. We begin this chapter with oxidative phosphoryla- tion. We first describe the components of the electron- transfer chain, their organization into large functional complexes in the inner mitochondrial membrane, the path of electron flow through them, and the proton movements that accompany this flow. We then consider the remarkable enzyme complex that, by “rotational catalysis,” captures the energy of proton flow in ATP, and the regulatory mechanisms that coordinate oxida- tive phosphorylation with the many catabolic pathways by which fuels are oxidized. With this understanding of mitochondrial oxidative phosphorylation, we turn to photophosphorylation, looking first at the absorption of light by photosynthetic pigments, then at the light- driven flow of electrons from H 2 O to NADP H11001 and the molecular basis for coupling electron and proton flow. We also consider the similarities of structure and mech- anism between the ATP synthases of chloroplasts and mitochondria, and the evolutionary basis for this con- servation of mechanism. OXIDATIVE PHOSPHORYLATION 19.1 Electron-Transfer Reactions in Mitochondria The discovery in 1948 by Eugene Kennedy and Albert Lehninger that mitochondria are the site of oxidative phosphorylation in eukaryotes marked the beginning of the modern phase of studies in biological energy transduc- tions. Mitochondria, like gram- negative bacteria, have two membranes (Fig. 19–1). The outer mitochondrial membrane is readily permeable to small molecules (M r H110215,000) and ions, which move freely through transmembrane chan- nels formed by a family of inte- gral membrane proteins called porins. The inner membrane is impermeable to most small molecules and ions, including protons (H H11001 ); the only species that cross this membrane do so through specific transporters. The inner membrane bears the compo- nents of the respiratory chain and the ATP synthase. The mitochondrial matrix, enclosed by the inner membrane, contains the pyruvate dehydrogenase com- plex and the enzymes of the citric acid cycle, the fatty 19.1 Electron-Transfer Reactions in Mitochondria 691 Outer membrane Freely permeable to small molecules and ions ATP synthase (F o F 1 ) Cristae Impermeable to most small molecules and ions, including H H11001 Contains: Contains: Ribosomes Porin channels ? Respiratory electron carriers (Complexes I–IV) ? ADP-ATP translocase ? ATP synthase (F o F 1 ) ? Other membrane transporters ? Pyruvate dehydrogenase complex ? Citric acid cycle enzymes ? Amino acid oxidation enzymes ? DNA, ribosomes ? Many other enzymes ? ATP, ADP, P i , Mg 2H11001 , Ca 2H11001 , K H11001 ? Many soluble metabolic intermediates Inner membrane Matrix ? Fatty acid -oxidation enzymes H9252 Albert L. Lehninger, 1917–1986 FIGURE 19–1 Biochemical anatomy of a mitochondrion. The convo- lutions (cristae) of the inner membrane provide a very large surface area. The inner membrane of a single liver mitochondrion may have more than 10,000 sets of electron-transfer systems (respiratory chains) and ATP synthase molecules, distributed over the membrane surface. Heart mitochondria, which have more profuse cristae and thus a much larger area of inner membrane, contain more than three times as many sets of electron-transfer systems as liver mitochondria. The mitochon- drial pool of coenzymes and intermediates is functionally separate from the cytosolic pool. The mitochondria of invertebrates, plants, and mi- crobial eukaryotes are similar to those shown here, but with much vari- ation in size, shape, and degree of convolution of the inner membrane. 8885d_c19_690-750 3/1/04 11:32 AM Page 691 mac76 mac76:385_reb: acid H9252-oxidation pathway, and the pathways of amino acid oxidation—all the pathways of fuel oxidation ex- cept glycolysis, which takes place in the cytosol. The selectively permeable inner membrane segregates the intermediates and enzymes of cytosolic metabolic path- ways from those of metabolic processes occurring in the matrix. However, specific transporters carry pyruvate, fatty acids, and amino acids or their H9251-keto derivatives into the matrix for access to the machinery of the citric acid cycle. ADP and P i are specifically transported into the matrix as newly synthesized ATP is transported out. Electrons Are Funneled to Universal Electron Acceptors Oxidative phosphorylation begins with the entry of elec- trons into the respiratory chain. Most of these electrons arise from the action of dehydrogenases that collect electrons from catabolic pathways and funnel them into universal electron acceptors—nicotinamide nucleotides (NAD H11001 or NADP H11001 ) or flavin nucleotides (FMN or FAD). Nicotinamide nucleotide–linked dehydroge- nases catalyze reversible reactions of the following gen- eral types: Reduced substrate H11001 NAD H11001 oxidized substrate H11001 NADH H11001 H H11001 Reduced substrate H11001 NADP H11001 oxidized substrate H11001 NADPH H11001 H H11001 Most dehydrogenases that act in catabolism are specific for NAD H11001 as electron acceptor (Table 19–1). Some are z y z y in the cytosol, others are in mitochondria, and still oth- ers have mitochondrial and cytosolic isozymes. NAD-linked dehydrogenases remove two hydrogen atoms from their substrates. One of these is transferred as a hydride ion ( : H H11002 ) to NAD H11001 ; the other is released as H H11001 in the medium (see Fig. 13–15). NADH and NADPH are water-soluble electron carriers that associ- ate reversibly with dehydrogenases. NADH carries elec- trons from catabolic reactions to their point of entry into the respiratory chain, the NADH dehydrogenase com- plex described below. NADPH generally supplies elec- trons to anabolic reactions. Cells maintain separate pools of NADPH and NADH, with different redox po- tentials. This is accomplished by holding the ratios of [reduced form]/[oxidized form] relatively high for NADPH and relatively low for NADH. Neither NADH nor NADPH can cross the inner mitochondrial membrane, but the electrons they carry can be shuttled across in- directly, as we shall see. Flavoproteins contain a very tightly, sometimes covalently, bound flavin nucleotide, either FMN or FAD (see Fig. 13–18). The oxidized flavin nucleotide can ac- cept either one electron (yielding the semiquinone form) or two (yielding FADH 2 or FMNH 2 ). Electron transfer occurs because the flavoprotein has a higher reduction potential than the compound oxidized. The standard reduction potential of a flavin nucleotide, un- like that of NAD or NADP, depends on the protein with which it is associated. Local interactions with functional groups in the protein distort the electron orbitals in the flavin ring, changing the relative stabilities of oxidized and reduced forms. The relevant standard reduction Chapter 19 Oxidative Phosphorylation and Photophosphorylation692 TABLE 19–1 Some Important Reactions Catalyzed by NAD(P)H-Linked Dehydrogenases Reaction * Location ? NAD-linked H9251-Ketoglutarate H11001 CoA H11001NAD H11001 succinyl-CoA H11001 CO 2 H11001 NADH H11001 H H11001 M L-Malate H11001 NAD H11001 oxaloacetate H11001 NADH H11001 H H11001 M and C Pyruvate H11001 CoA H11001 NAD H11001 acetyl-CoA H11001 CO 2 H11001 NADH H11001 H H11001 M Glyceraldehyde 3-phosphate H11001 P i H11001 NAD H11001 1,3-bisphosphoglycerate H11001 NADH H11001 H H11001 C Lactate H11001 NAD H11001 pyruvate H11001 NADH H11001 H H11001 C H9252-Hydroxyacyl-CoA H11001 NAD H11001 H9252-ketoacyl-CoA H11001 NADH H11001 H H11001 M NADP-linked Glucose 6-phosphate H11001 NADP H11001 6-phosphogluconate H11001 NADPH H11001 H H11001 C NAD- or NADP-linked L-Glutamate H11001 H 2 O H11001 NAD(P) H11001 H9251-ketoglutarate H11001 NH 4 H11001 H11001 NAD(P)H M Isocitrate H11001 NAD(P) H11001 H9251-ketoglutarate H11001 CO 2 H11001 NAD(P)H H11001 H H11001 M and C z y z y z y z y z y z y z y z y z y * These reactions and their enzymes are discussed in Chapters 14 through 18. ? M designates mitochondria; C, cytosol. 8885d_c19_690-750 3/1/04 11:32 AM Page 692 mac76 mac76:385_reb: potential is therefore that of the particular flavoprotein, not that of isolated FAD or FMN. The flavin nucleotide should be considered part of the flavoprotein’s active site rather than a reactant or product in the electron- transfer reaction. Because flavoproteins can participate in either one- or two-electron transfers, they can serve as intermediates between reactions in which two elec- trons are donated (as in dehydrogenations) and those in which only one electron is accepted (as in the reduction of a quinone to a hydroquinone, described below). Electrons Pass through a Series of Membrane-Bound Carriers The mitochondrial respiratory chain consists of a series of sequentially acting electron carriers, most of which are integral proteins with prosthetic groups capable of accepting and donating either one or two electrons. Three types of electron transfers occur in oxidative phosphorylation: (1) direct transfer of electrons, as in the reduction of Fe 3H11001 to Fe 2H11001 ; (2) transfer as a hydro- gen atom (H H11001 H11001 e H11002 ); and (3) transfer as a hydride ion (:H H11002 ), which bears two electrons. The term reducing equivalent is used to designate a single electron equiv- alent transferred in an oxidation-reduction reaction. In addition to NAD and flavoproteins, three other types of electron-carrying molecules function in the res- piratory chain: a hydrophobic quinone (ubiquinone) and two different types of iron-containing proteins (cyto- chromes and iron-sulfur proteins). Ubiquinone (also called coenzyme Q, or simply Q) is a lipid-soluble ben- zoquinone with a long isoprenoid side chain (Fig. 19–2). The closely related compounds plastoquinone (of plant chloroplasts) and menaquinone (of bacteria) play roles analogous to that of ubiquinone, carrying electrons in membrane-associated electron-transfer chains. Ubiqui- none can accept one electron to become the semi- quinone radical ( H11080 QH) or two electrons to form ubiquinol (QH 2 ) (Fig. 19–2) and, like flavoprotein carriers, it can act at the junction between a two-electron donor and a one-electron acceptor. Because ubiquinone is both small and hydrophobic, it is freely diffusible within the lipid bilayer of the inner mitochondrial membrane and can shuttle reducing equivalents between other, less mobile electron carriers in the membrane. And because it car- ries both electrons and protons, it plays a central role in coupling electron flow to proton movement. The cytochromes are proteins with characteristic strong absorption of visible light, due to their iron- containing heme prosthetic groups (Fig. 19–3). Mito- chondria contain three classes of cytochromes, desig- nated a, b, and c, which are distinguished by differences in their light-absorption spectra. Each type of cyto- chrome in its reduced (Fe 2H11001 ) state has three absorp- tion bands in the visible range (Fig. 19–4). The longest- wavelength band is near 600 nm in type a cytochromes, near 560 nm in type b, and near 550 nm in type c. To distinguish among closely related cytochromes of one type, the exact absorption maximum is sometimes used in the names, as in cytochrome b 562 . The heme cofactors of a and b cytochromes are tightly, but not covalently, bound to their associated pro- teins; the hemes of c-type cytochromes are covalently attached through Cys residues (Fig. 19–3). As with the flavoproteins, the standard reduction potential of the heme iron atom of a cytochrome depends on its inter- action with protein side chains and is therefore differ- ent for each cytochrome. The cytochromes of type a and b and some of type c are integral proteins of the inner mitochondrial membrane. One striking exception is the cytochrome c of mitochondria, a soluble protein that associates through electrostatic interactions with the outer surface of the inner membrane. We encountered cytochrome c in earlier dis- cussions of protein structure (see Fig. 4–18). In iron-sulfur proteins, first discovered by Helmut Beinert, the iron is present not in heme but in association with inorganic sulfur atoms or with the sulfur atoms of Cys residues in the protein, or both. These iron-sulfur (Fe-S) 19.1 Electron-Transfer Reactions in Mitochondria 693 ? O HCCH R OH CH 3 CH 3 O (CH 2 O CH 3 O CH 3 CH 2 ) 10 Ubiquinone (Q) (fully oxidized) Semiquinone radical ( ? QH) Ubiquinol (QH 2 ) (fully reduced) H H11001 H11001 e H11002 O CH 3 CH 3 O CH 3 O H H11001 H11001 e H11002 OH OH R CH 3 CH 3 O CH 3 O FIGURE 19–2 Ubiquinone (Q, or coenzyme Q). Complete reduction of ubiquinone requires two electrons and two protons, and occurs in two steps through the semiquinone radical intermediate. Helmut Beinert 8885d_c19_690-750 3/1/04 11:32 AM Page 693 mac76 mac76:385_reb: centers range from simple structures with a single Fe atom coordinated to four Cys OSH groups to more com- plex Fe-S centers with two or four Fe atoms (Fig. 19–5). Rieske iron-sulfur proteins (named after their dis- coverer, John S. Rieske) are a variation on this theme, in which one Fe atom is coordinated to two His residues rather than two Cys residues. All iron-sulfur proteins participate in one-electron transfers in which one iron atom of the iron-sulfur cluster is oxidized or reduced. At least eight Fe-S proteins function in mitochondrial electron transfer. The reduction potential of Fe-S pro- teins varies from H110020.65 V to H110010.45 V, depending on the microenvironment of the iron within the protein. In the overall reaction catalyzed by the mitochon- drial respiratory chain, electrons move from NADH, suc- cinate, or some other primary electron donor through flavoproteins, ubiquinone, iron-sulfur proteins, and cy- tochromes, and finally to O 2 . A look at the methods used to determine the sequence in which the carriers act is instructive, as the same general approaches have been used to study other electron-transfer chains, such as those of chloroplasts. Chapter 19 Oxidative Phosphorylation and Photophosphorylation694 Fe N N N N CH 3 CH 3 CH 2 CH 2 COO H11002 CH 2 CH CH 2 CH CH 3 Heme A (in a-type cytochromes) Fe N N N N CH 3 CH 3 CH 3 CH 3 CH 2 CH 2 COO H11002 CH 2 CH 2 CHO CH 2 CHCH 2 Iron protoporphyrin IX (in b-type cytochromes) COO H11002 CH 3 OH Cys S Cys Fe N N N N CH 3 CH 3 CH 3 CH 3 CH 2 CH 2 COO H11002 CH 2 CH 2 Heme C (in c-type cytochromes) COO H11002 CH 3 CH 2 CH S CH CH 2 CH CH 3 CH 3 CH 3 CH 3 COO H11002 FIGURE 19–3 Prosthetic groups of cytochromes. Each group consists of four five-membered, nitrogen-containing rings in a cyclic structure called a porphyrin. The four nitrogen atoms are coordinated with a central Fe ion, either Fe 2H11001 or Fe 3H11001 . Iron protoporphyrin IX is found in b-type cytochromes and in hemoglobin and myoglobin (see Fig. 4–17). Heme c is covalently bound to the protein of cytochrome c through thioether bonds to two Cys residues. Heme a, found in the a-type cytochromes, has a long isoprenoid tail attached to one of the five-membered rings. The conjugated double-bond system (shaded pink) of the porphyrin ring accounts for the absorption of visible light by these hemes. 100 Relative light absorption (%) 50 300 400 500 600 Wavelength (nm) Oxidized cyt c Reduced cyt c H9251 H9252 H9253 0 FIGURE 19–4 Absorption spectra of cytochrome c (cyt c) in its oxi- dized (red) and reduced (blue) forms. Also labeled are the character- istic H9251, H9252, and H9253 bands of the reduced form. 8885d_c19_690-750 3/1/04 11:32 AM Page 694 mac76 mac76:385_reb: First, the standard reduction potentials of the in- dividual electron carriers have been determined ex- perimentally (Table 19–2). We would expect the carri- ers to function in order of increasing reduction potential, because electrons tend to flow spontaneously from carriers of lower EH11032H11034 to carriers of higher EH11032H11034. The order of carriers deduced by this method is NADH → Q → cytochrome b → cytochrome c 1 → cytochrome c → cytochrome a → cytochrome a 3 → O 2 . Note, how- ever, that the order of standard reduction potentials is not necessarily the same as the order of actual reduc- tion potentials under cellular conditions, which depend on the concentration of reduced and oxidized forms (p. 510). A second method for determining the sequence 19.1 Electron-Transfer Reactions in Mitochondria 695 S (c) Cys Cys Fe S S S S S S S Fe Fe Fe Cys Cys S (b) SCys Cys Fe S Fe S S CysSCysS Cys S S S Cys (a) Cys Cys Fe Protein (d) FIGURE 19–5 Iron-sulfur centers. The Fe-S centers of iron-sulfur proteins may be as simple as (a), with a single Fe ion surrounded by the S atoms of four Cys residues. Other centers include both inorganic and Cys S atoms, as in (b) 2Fe-2S or (c) 4Fe-4S centers. (d) The ferredoxin of the cyanobacterium Anabaena 7120 has one 2Fe-2S center (PDB ID 1FRD); Fe is red, inorganic S 2 is yellow, and the S of Cys is orange. (Note that in these designations only the inorganic S atoms are counted. For example, in the 2Fe-2S center (b), each Fe ion is actually surrounded by four S atoms.) The exact standard reduction potential of the iron in these centers depends on the type of center and its interaction with the associated protein. TABLE 19–2 Standard Reduction Potentials of Respiratory Chain and Related Electron Carriers Redox reaction (half-reaction) E H11032H11034 (V) 2H H11001 H11001 2e H11002 8n H 2 H110020.414 NAD H11001 H11001 H H11001 H11001 2e H11002 8n NADH H110020.320 NADP H11001 H11001 H H11001 H11001 2e H11002 8n NADPH H110020.324 NADH dehydrogenase (FMN) H11001 2H H11001 H11001 2e H11002 8n NADH dehydrogenase (FMNH 2 ) H110020.30 Ubiquinone H11001 2H H11001 H11001 2e H11002 8n ubiquinol 0.045 Cytochrome b (Fe 3H11001 ) H11001 e H11002 8n cytochrome b (Fe 2H11001 ) 0.077 Cytochrome c 1 (Fe 3H11001 ) H11001 e H11002 8n cytochrome c 1 (Fe 2H11001 ) 0.22 Cytochrome c (Fe 3H11001 ) H11001 e H11002 8n cytochrome c (Fe 2H11001 ) 0.254 Cytochrome a (Fe 3H11001 ) H11001 e H11002 8n cytochrome a (Fe 2H11001 ) 0.29 Cytochrome a 3 (Fe 3H11001 ) H11001 e H11002 8n cytochrome a 3 (Fe 2H11001 ) 0.35 H5007 1 2 H5007 O 2 H11001 2H H11001 H11001 2e H11002 8n H 2 O 0.8166 8885d_c19_690-750 3/1/04 11:32 AM Page 695 mac76 mac76:385_reb: of electron carriers involves reducing the entire chain of carriers experimentally by providing an electron source but no electron acceptor (no O 2 ). When O 2 is suddenly introduced into the system, the rate at which each electron carrier becomes oxidized (measured spectroscopically) reveals the order in which the car- riers function. The carrier nearest O 2 (at the end of the chain) gives up its electrons first, the second carrier from the end is oxidized next, and so on. Such exper- iments have confirmed the sequence deduced from standard reduction potentials. In a final confirmation, agents that inhibit the flow of electrons through the chain have been used in com- bination with measurements of the degree of oxidation of each carrier. In the presence of O 2 and an electron donor, carriers that function before the inhibited step become fully reduced, and those that function after this step are completely oxidized (Fig. 19–6). By using sev- eral inhibitors that block different steps in the chain, in- vestigators have determined the entire sequence; it is the same as deduced in the first two approaches. Electron Carriers Function in Multienzyme Complexes The electron carriers of the respiratory chain are or- ganized into membrane-embedded supramolecular complexes that can be physically separated. Gentle treatment of the inner mitochondrial membrane with detergents allows the resolution of four unique electron- carrier complexes, each capable of catalyzing electron transfer through a portion of the chain (Table 19–3; Fig. 19–7). Complexes I and II catalyze electron transfer to ubiquinone from two different electron donors: NADH (Complex I) and succinate (Complex II). Complex III carries electrons from reduced ubiquinone to cyto- chrome c, and Complex IV completes the sequence by transferring electrons from cytochrome c to O 2 . We now look in more detail at the structure and function of each complex of the mitochondrial respira- tory chain. Complex I: NADH to Ubiquinone Figure 19–8 illustrates the relationship between Complexes I and II and ubiquinone. Complex I, also called NADH:ubiquinone oxidore- ductase or NADH dehydrogenase, is a large enzyme composed of 42 different polypeptide chains, including an FMN-containing flavoprotein and at least six iron- sulfur centers. High-resolution electron microscopy shows Complex I to be L-shaped, with one arm of the L in the membrane and the other extending into the ma- trix. As shown in Figure 19–9, Complex I catalyzes two simultaneous and obligately coupled processes: (1) the Chapter 19 Oxidative Phosphorylation and Photophosphorylation696 NADH Q Cyt c 1 Cyt (a H11001 a 3 ) O 2 rotenone antimycin A CN H11002 or CO Cyt cCyt bNADH Q O 2 NADH Q O 2 Cyt c 1 Cyt cCyt b Cyt c 1 Cyt Cyt cCyt b Cyt (a H11001 a 3 ) a H11001 a 3 )( FIGURE 19–6 Method for determining the sequence of electron carriers. This method measures the effects of inhibitors of electron transfer on the oxidation state of each carrier. In the presence of an electron donor and O 2 , each inhibitor causes a characteristic pattern of oxidized/reduced carriers: those before the block become reduced (blue), and those after the block become oxidized (pink). TABLE 19–3 The Protein Components of the Mitochondrial Electron-Transfer Chain Enzyme complex/protein Mass (kDa) Number of subunits * Prosthetic group(s) I NADH dehydrogenase 850 43 (14) FMN, Fe-S II Succinate dehydrogenase 140 4 FAD, Fe-S III Ubiquinone cytochrome c oxidoreductase 250 11 Hemes, Fe-S Cytochrome c ? 13 1 Heme IV Cytochrome oxidase 160 13 (3–4) Hemes; Cu A ,Cu B * Numbers of subunits in the bacterial equivalents in parentheses. ? Cytochrome c is not part of an enzyme complex; it moves between Complexes III and IV as a freely soluble protein. 8885d_c19_696 3/1/04 1:58 PM Page 696 mac76 mac76:385_reb: exergonic transfer to ubiquinone of a hydride ion from NADH and a proton from the matrix, expressed by NADH H11001 H H11001 H11001 Q On NAD H11001 H11001 QH 2 (19–1) and (2) the endergonic transfer of four protons from the matrix to the intermembrane space. Complex I is there- fore a proton pump driven by the energy of electron transfer, and the reaction it catalyzes is vectorial: it moves protons in a specific direction from one location (the matrix, which becomes negatively charged with the departure of protons) to another (the intermembrane space, which becomes positively charged). To empha- size the vectorial nature of the process, the overall re- action is often written with subscripts that indicate the location of the protons: P for the positive side of the in- ner membrane (the intermembrane space), N for the negative side (the matrix): NADH H11001 5H H11001 N H11001 Q On NAD H11001 H11001 QH 2 H11001 4H H11001 P (19–2) 19.1 Electron-Transfer Reactions in Mitochondria 697 Osmotic rupture Inner membrane fragments Outer membrane fragments discarded ATP synthase IV III II I I II III IV ATP synthase NADH Q Suc- cinate Q Q Cyt c Cyt c O 2 ATP ADP H11001 P i Reactions catalyzed by isolated fractions in vitro Solubilization with detergent followed by ion-exchange chromatography Treatment with digitonin FIGURE 19–7 Separation of functional complexes of the respiratory chain. The outer mitochondrial membrane is first removed by treat- ment with the detergent digitonin. Fragments of inner membrane are then obtained by osmotic rupture of the mitochondria, and the frag- ments are gently dissolved in a second detergent. The resulting mix- ture of inner membrane proteins is resolved by ion-exchange chro- matography into different complexes (I through IV) of the respiratory chain, each with its unique protein composition (see Table 19–3), and the enzyme ATP synthase (sometimes called Complex V). The isolated Complexes I through IV catalyze transfers between donors (NADH and succinate), intermediate carriers (Q and cytochrome c), and O 2 , as shown. In vitro, isolated ATP synthase has only ATP-hydrolyzing (ATPase), not ATP-synthesizing, activity. III Intermembrane space Matrix Fe-S Fe-S FAD Glycerol 3-phosphate (cytosolic) glycerol 3-phosphate dehydrogenase FAD FMN NADH NAD + Succinate ETF:Q oxidoreductase acyl-CoA dehydrogenase ETF (FAD) Fe-S (FAD) Fatty acyl–CoA FAD Q FIGURE 19–8 Path of electrons from NADH, succinate, fatty acyl–CoA, and glycerol 3-phosphate to ubiquinone. Electrons from NADH pass through a flavoprotein to a series of iron-sulfur proteins (in Complex I) and then to Q. Electrons from succinate pass through a flavoprotein and several Fe-S centers (in Complex II) on the way to Q. Glycerol 3-phosphate donates electrons to a flavoprotein (glycerol 3-phosphate dehydrogenase) on the outer face of the inner mito- chondrial membrane, from which they pass to Q. Acyl-CoA dehydro- genase (the first enzyme of H9252 oxidation) transfers electrons to electron- transferring flavoprotein (ETF), from which they pass to Q via ETF:ubiquinone oxidoreductase. 8885d_c19_690-750 3/1/04 11:32 AM Page 697 mac76 mac76:385_reb: Amytal (a barbiturate drug), rotenone (a plant product commonly used as an insecticide), and piericidin A (an antibiotic) inhibit electron flow from the Fe-S centers of Complex I to ubiquinone (Table 19–4) and therefore block the overall process of oxidative phosphorylation. Ubiquinol (QH 2 , the fully reduced form; Fig. 19–2) diffuses in the inner mitochondrial membrane from Complex I to Complex III, where it is oxidized to Q in a process that also involves the outward movement of H H11001 . Complex II: Succinate to Ubiquinone We encountered Complex II in Chapter 16 as succinate dehydroge- nase, the only membrane-bound enzyme in the citric acid cycle (p. 612). Although smaller and simpler than Complex I, it contains five prosthetic groups of two types and four different protein subunits (Fig. 19–10). Subunits C and D are integral membrane proteins, each with three transmembrane helices. They contain a heme group, heme b, and a binding site for ubiquinone, the final electron acceptor in the reaction catalyzed by Complex II. Subunits A and B extend into the matrix (or the cytosol of a bacterium); they contain three 2Fe-2S centers, bound FAD, and a binding site for the substrate, succinate. The path of electron transfer from the succinate-binding site to FAD, then through the Fe-S centers to the Q-binding site, is more than 40 ? long, but none of the individual electron-transfer distances exceeds about 11 ?—a reasonable distance for rapid electron transfer (Fig. 19–10). Chapter 19 Oxidative Phosphorylation and Photophosphorylation698 Complex I Intermembrane space (P side) Matrix (N side) 2H + 4H + Fe-S FMN NADH NAD + H + H11001 2e – 2e – N-2 Q QH 2 Matrix arm Membrane arm FIGURE 19–9 NADH:ubiquinone oxidoreductase (Complex I). Com- plex I catalyzes the transfer of a hydride ion from NADH to FMN, from which two electrons pass through a series of Fe-S centers to the iron- sulfur protein N-2 in the matrix arm of the complex. Electron transfer from N-2 to ubiquinone on the membrane arm forms QH 2 , which dif- fuses into the lipid bilayer. This electron transfer also drives the ex- pulsion from the matrix of four protons per pair of electrons. The de- tailed mechanism that couples electron and proton transfer in Complex I is not yet known, but probably involves a Q cycle similar to that in Complex III in which QH 2 participates twice per electron pair (see Fig. 19–12). Proton flux produces an electrochemical potential across the inner mitochondrial membrane (N side negative, P side positive), which conserves some of the energy released by the electron-transfer reactions. This electrochemical potential drives ATP synthesis. TABLE 19–4 Agents That Interfere with Oxidative Phosphorylation or Photophosphorylation Type of interference Compound * Target/mode of action Inhibition of electron transfer Cyanide Carbon monoxide Antimycin A Blocks electron transfer from cytochrome b to cytochrome c 1 Myxothiazol Rotenone Amytal Piericidin A DCMU Competes with Q B for binding site in PSII Inhibition of ATP synthase Aurovertin Inhibits F 1 Oligomycin Venturicidin DCCD Blocks proton flow through F o and CF o Uncoupling of phosphorylation FCCP from electron transfer DNP Hydrophobic proton carriers Valinomycin K H11001 ionophore Thermogenin In brown fat, forms proton-conducting pores in inner mitochondrial membrane Inhibition of ATP-ADP exchange Atractyloside Inhibits adenine nucleotide translocase * DCMU is 3-(3,4-dichlorophenyl)-1,1-dimethylurea; DCCD, dicyclohexylcarbodiimide; FCCP, cyanide-p-trifluoromethoxyphenylhydrazone; DNP, 2,4-dinitrophenol. Inhibit cytochrome oxidase Prevent electron transfer from Fe-S center to ubiquinone Inhibit F o and CF o ? ? ? ? ? ? ? ? ? ? ? ? ? ? 8885d_c19_690-750 3/1/04 11:32 AM Page 698 mac76 mac76:385_reb: flavoprotein acyl-CoA dehydrogenase (see Fig. 17–8), involves transfer of electrons from the substrate to the FAD of the dehydrogenase, then to electron-transferring flavoprotein (ETF), which in turn passes its electrons to ETF : ubiquinone oxidoreductase (Fig. 19–8). This enzyme transfers electrons into the respiratory chain by reducing ubiquinone. Glycerol 3-phosphate, formed ei- ther from glycerol released by triacylglycerol breakdown or by the reduction of dihydroxyacetone phosphate from glycolysis, is oxidized by glycerol 3-phosphate dehydrogenase (see Fig. 17–4). This enzyme is a flavo- protein located on the outer face of the inner mito- chondrial membrane, and like succinate dehydrogenase and acyl-CoA dehydrogenase it channels electrons into the respiratory chain by reducing ubiquinone (Fig. 19–8). The important role of glycerol 3-phosphate de- hydrogenase in shuttling reducing equivalents from cytosolic NADH into the mitochondrial matrix is de- scribed in Section 19.2 (see Fig. 19–28). The effect of each of these electron-transferring enzymes is to con- tribute to the pool of reduced ubiquinone. QH 2 from all these reactions is reoxidized by Complex III. Complex III: Ubiquinone to Cytochrome c The next respi- ratory complex, Complex III, also called cytochrome bc 1 complex or ubiquinone:cytochrome c oxidore- ductase, couples the transfer of electrons from ubiquinol (QH 2 ) to cytochrome c with the vectorial transport of protons from the matrix to the intermem- brane space. The determination of the complete struc- ture of this huge complex (Fig. 19–11) and of Complex IV (below) by x-ray crystallography, achieved between 1995 and 1998, were landmarks in the study of mito- chondrial electron transfer, providing the structural framework to integrate the many biochemical observa- tions on the functions of the respiratory complexes. Based on the structure of Complex III and detailed biochemical studies of the redox reactions, a reasonable model has been proposed for the passage of electrons 19.1 Electron-Transfer Reactions in Mitochondria 699 The heme b of Complex II is apparently not in the direct path of electron transfer; it may serve instead to reduce the frequency with which electrons “leak” out of the system, moving from succinate to mo- lecular oxygen to produce the reactive oxygen species (ROS) hydrogen peroxide (H 2 O 2 ) and the superoxide radical ( H11080 O 2 H11002 ) described in Section 19.5. Humans with point mutations in Complex II subunits near heme b or the quinone-binding site suffer from hereditary para- ganglioma. This inherited condition is characterized by benign tumors of the head and neck, commonly in the carotid body, an organ that senses O 2 levels in the blood. These mutations result in greater production of ROS and perhaps greater tissue damage during succinate oxidation. ■ Other substrates for mitochondrial dehydrogenases pass electrons into the respiratory chain at the level of ubiquinone, but not through Complex II. The first step in the H9252 oxidation of fatty acyl–CoA, catalyzed by the FIGURE 19–10 Structure of Complex II (succinate dehydrogenase) of E. coli (PDB ID 1NEK). The enzyme has two transmembrane sub- units, C (green) and D (blue); the cytoplasmic extensions contain sub- units B (orange) and A (purple). Just behind the FAD in subunit A (gold) is the binding site for succinate (occupied in this crystal structure by the inhibitor oxaloacetate, green). Subunit B has three sets of Fe-S cen- ters (yellow and red); ubiquinone (yellow) is bound to subunit C; and heme b (purple) is sandwiched between subunits C and D. A cardi- olipin molecule is so tightly bound to subunit C that it shows up in the crystal structure (gray spacefilling). Electrons move (blue arrows) from succinate to FAD, then through the three Fe-S centers to ubiquinone. The heme b is not on the main path of electron transfer but protects against the formation of reactive oxygen species (ROS) by electrons that go astray. Substrate binding site Cytoplasm (N side) C B D A FAD Fe-S centers Periplasm (P side) Cardiolipin Ubiquinone Heme bQH 2 8885d_c19_690-750 3/1/04 11:32 AM Page 699 mac76 mac76:385_reb: and protons through the complex. The net equation for the redox reactions of this Q cycle (Fig. 19–12) is QH 2 H11001 2 cyt c 1 (oxidized) H11001 2H H11001 N O n Q H11001 2 cyt c 1 (reduced) H11001 4H H11001 P (19–3) The Q cycle accommodates the switch between the two- electron carrier ubiquinone and the one-electron carri- ers—cytochromes b 562 , b 566 , c 1 , and c—and explains the measured stoichiometry of four protons translocated per pair of electrons passing through the Complex III to cytochrome c. Although the path of electrons through this segment of the respiratory chain is complicated, the net effect of the transfer is simple: QH 2 is oxidized to Q and two molecules of cytochrome c are reduced. Cytochrome c (see Fig. 4–18) is a soluble protein of the intermembrane space. After its single heme accepts an electron from Complex III, cytochrome c moves to Complex IV to donate the electron to a binuclear cop- per center. Complex IV: Cytochrome c to O 2 In the final step of the respiratory chain, Complex IV, also called cytochrome oxidase, carries electrons from cytochrome c to mo- lecular oxygen, reducing it to H 2 O. Complex IV is a large enzyme (13 subunits; M r 204,000) of the inner mito- chondrial membrane. Bacteria contain a form that is much simpler, with only three or four subunits, but still capable of catalyzing both electron transfer and proton pumping. Comparison of the mitochondrial and bacter- ial complexes suggests that three subunits are critical to the function (Fig. 19–13). Mitochondrial subunit II contains two Cu ions com- plexed with the OSH groups of two Cys residues in a binuclear center (Cu A ; Fig. 19–13b) that resembles the 2Fe-2S centers of iron-sulfur proteins. Subunit I con- tains two heme groups, designated a and a 3 , and an- other copper ion (Cu B ). Heme a 3 and Cu B form a sec- ond binuclear center that accepts electrons from heme a and transfers them to O 2 bound to heme a 3 . Chapter 19 Oxidative Phosphorylation and Photophosphorylation700 FIGURE 19–11 Cytochrome bc 1 complex (Complex III). The com- plex is a dimer of identical monomers, each with 11 different sub- units. (a) Structure of a monomer. The functional core is three sub- units: cytochrome b (green) with its two hemes (b H and b L , light red); the Rieske iron-sulfur protein (purple) with its 2Fe-2S centers (yellow); and cytochrome c 1 (blue) with its heme (red) (PDB ID 1BGY). (b) The dimeric functional unit. Cytochrome c 1 and the Rieske iron-sulfur pro- tein project from the P surface and can interact with cytochrome c (not part of the functional complex) in the intermembrane space. The complex has two distinct binding sites for ubiquinone, Q N and Q P , which correspond to the sites of inhibition by two drugs that block oxidative phosphorylation. Antimycin A, which blocks electron flow from heme b H to Q, binds at Q N , close to heme b H on the N (matrix) side of the membrane. Myxothiazol, which prevents electron flow from QH 2 to the Rieske iron-sulfur protein, binds at Q P , near the 2Fe-2S center and heme b L on the P side. The dimeric structure is essential to the function of Complex III. The interface between monomers forms two pockets, each containing a Q P site from one monomer and a Q N site from the other. The ubiquinone intermediates move within these sheltered pockets. Complex III crystallizes in two distinct conformations (not shown). In one, the Rieske Fe-S center is close to its electron acceptor, the heme of cytochrome c 1 , but relatively distant from cytochrome b and the QH 2 -binding site at which the Rieske Fe-S center receives elec- trons. In the other, the Fe-S center has moved away from cytochrome c 1 and toward cytochrome b. The Rieske protein is thought to oscil- late between these two conformations as it is reduced, then oxidized. (a) Intermembrane space (P side) Matrix (N side) Cytochrome c 1 Cytochrome b Rieske iron- sulfur protein 2Fe-2S (b) Cytochrome c 1 Cytochrome c Rieske iron- sulfur protein 2Fe-2S center Cytochrome b (P side) (N side) b L Q P Q N b H c 1 Heme 8885d_c19_690-750 3/1/04 11:32 AM Page 700 mac76 mac76:385_reb: Electron transfer through Complex IV is from cyto- chrome c to the Cu A center, to heme a, to the heme a 3 –Cu B center, and finally to O 2 (Fig. 19–14). For every four electrons passing through this complex, the enzyme consumes four “substrate” H H11001 from the matrix (N side) in converting O 2 to 2H 2 O. It also uses the energy of this redox reaction to pump one proton outward into the in- termembrane space (P side) for each electron that passes through, adding to the electrochemical potential produced by redox-driven proton transport through Complexes I and III. The overall reaction catalyzed by Complex IV is 4 Cyt c (reduced) H11001 8H H11001 N H11001 O 2 O n 4 cyt c (oxidized) H11001 4H H11001 P H11001 2H 2 O (19–4) This four-electron reduction of O 2 involves redox cen- ters that carry only one electron at a time, and it must occur without the release of incompletely reduced intermediates such as hydrogen peroxide or hydroxyl free radicals—very reactive species that would damage cellular components. The intermediates remain tightly bound to the complex until completely converted to water. The Energy of Electron Transfer Is Efficiently Conserved in a Proton Gradient The transfer of two electrons from NADH through the respiratory chain to molecular oxygen can be written as NADH H11001 H H11001 H11001 H5007 1 2 H5007 O 2 O n NAD H11001 H11001 H 2 O (19–5) This net reaction is highly exergonic. For the redox pair NAD H11001 /NADH, EH11032H11034 is H110020.320 V, and for the pair O 2 /H 2 O, EH11032H11034 is 0.816 V. The H9004EH11032H11034 for this reaction is therefore 1.14 V, and the standard free-energy change (see Eqn 13–6, p. 510) is H9004GH11032H11034 H11005 H11002n H9004EH11032H11034 (19–6) H11005H110022(96.5 kJ/V H11080 mol)(1.14 V) H11005H11002220 kJ/mol (of NADH) This standard free-energy change is based on the as- sumption of equal concentrations (1 M) of NADH and 19.1 Electron-Transfer Reactions in Mitochondria 701 b H b L Cyt c 1 Cyt c Oxidation of first QH 2 Fe-S Q 2H + ? Q – QH 2 Q b H b L Cyt c 1 Cyt c Fe-S 2H + Q Oxidation of second QH 2 Matrix (N side) Intermembrane space (P side) QH 2 2H + ? Q – QH 2 QH 2 cyt c 1 (oxidized)H11001 QH 2 2 cyt c 1 (oxidized)H11001 2H P H11001 H11001 cyt c 1 (reduced)H11001 Q 2 cyt c 1 (reduced)H11001H11001 H11001 ? Q H11002 ? Q H11002 H11001 4H P H11001 2H P H11001 2H N H11001 H11001 2H N H11001 QH 2 QH 2 cyt c 1 (oxidized)H11001 QH11001H11001cyt c 1 (reduced)H11001 Net equation: FIGURE 19–12 The Q cycle. The path of electrons through Complex III is shown by blue arrows. On the P side of the membrane, two mol- ecules of QH 2 are oxidized to Q near the P side, releasing two pro- tons per Q (four protons in all) into the intermembrane space. Each QH 2 donates one electron (via the Rieske Fe-S center) to cytochrome c 1 , and one electron (via cytochrome b) to a molecule of Q near the N side, reducing it in two steps to QH 2 . This reduction also uses two protons per Q, which are taken up from the matrix. 8885d_c19_690-750 3/1/04 11:32 AM Page 701 mac76 mac76:385_reb: NAD H11001 . In actively respiring mitochondria, the actions of many dehydrogenases keep the actual [NADH]/[NAD H11001 ] ratio well above unity, and the real free-energy change for the reaction shown in Equation 19–5 is therefore substantially greater (more negative) than H11002220 kJ/mol. A similar calculation for the oxidation of succinate shows that electron transfer from succinate (EH11032H11034 for fumarate/succinate H11005 0.031 V) to O 2 has a smaller, but still negative, standard free-energy change of about H11002150 kJ/mol. Chapter 19 Oxidative Phosphorylation and Photophosphorylation702 (a) (b) FIGURE 19–13 Critical subunits of cytochrome oxidase (Complex IV). The bovine complex is shown here (PDB ID 1OCC). (a) The core of Complex IV, with three subunits. Subunit I (yellow) has two heme groups, a and a 3 (red), and a copper ion, Cu B (green sphere). Heme a 3 and Cu B form a binuclear Fe-Cu center. Subunit II (blue) contains two Cu ions (green spheres) complexed with the OSH groups of two Cys residues in a binuclear center, Cu A , that resembles the 2Fe-2S cen- ters of iron-sulfur proteins. This binuclear center and the cytochrome c–binding site are located in a domain of subunit II that protrudes from the P side of the inner membrane (into the intermembrane space). Subunit III (green) seems to be essential for Complex IV function, but its role is not well understood. (b) The binuclear center of Cu A . The Cu ions (green spheres) share electrons equally. When the center is reduced they have the formal charges Cu 1H11001 Cu 1H11001 ; when oxidized, Cu 1.5H11001 Cu 1.5H11001 . Ligands around the Cu ions include two His (dark blue), two Cys (yellow), an Asp (red), and Met (orange) residues. Subunit II Subunit I 4H + 4H + (pumped) Subunit III 4H + (substrate) 4e – 2H 2 O O 2 4Cyt c Cu B Cu A Fe-Cu center a 3 a Intermembrane space (P side) Matrix (N side) FIGURE 19–14 Path of electrons through Complex IV. The three pro- teins critical to electron flow are subunits I, II, and III. The larger green structure includes the other ten proteins in the complex. Electron trans- fer through Complex IV begins when two molecules of reduced cy- tochrome c (top) each donate an electron to the binuclear center Cu A . From here electrons pass through heme a to the Fe-Cu center (cy- tochrome a 3 and Cu B ). Oxygen now binds to heme a 3 and is reduced to its peroxy derivative (O 2 2H11002 ) by two electrons from the Fe-Cu center. Delivery of two more electrons from cytochrome c (making four elec- trons in all) converts the O 2 2H11002 to two molecules of water, with con- sumption of four “substrate” protons from the matrix. At the same time, four more protons are pumped from the matrix by an as yet unknown mechanism. 8885d_c19_690-750 3/1/04 11:32 AM Page 702 mac76 mac76:385_reb: Much of this energy is used to pump protons out of the matrix. For each pair of electrons transferred to O 2 , four protons are pumped out by Complex I, four by Com- plex III, and two by Complex IV (Fig. 19–15). The vec- torial equation for the process is therefore NADH H11001 11H H11001 N H11001 H5007 1 2 H5007 O 2 O n NAD H11001 H11001 10H H11001 P H11001 H 2 O (19–7) The electrochemical energy inherent in this difference in proton concentration and separation of charge rep- resents a temporary conservation of much of the energy of electron transfer. The energy stored in such a gradi- ent, termed the proton-motive force, has two com- ponents: (1) the chemical potential energy due to the difference in concentration of a chemical species (H H11001 ) in the two regions separated by the membrane, and (2) the electrical potential energy that results from the separation of charge when a proton moves across the membrane without a counterion (Fig. 19–16). As we showed in Chapter 11, the free-energy change for the creation of an electrochemical gradient by an ion pump is H9004G H11005 RT ln H20898H20899 H11001 Z H9004H9274 (19–8) where C 2 and C 1 are the concentrations of an ion in two regions, and C 2 H11022 C 1 ; Z is the absolute value of its elec- trical charge (1 for a proton), and H9004H9274 is the transmem- brane difference in electrical potential, measured in volts. For protons at 25 H11034C, ln H20898H20899 H11005 2.3(log [H H11001 ] P H11002 log [H H11001 ] N ) H11005 2.3(pH N H11002 pH P ) H11005 2.3 H9004pH and Equation 19–8 reduces to H9004G H11005 2.3RT H9004pH H11001H9004H9274 (19–9) H11005 (5.70 kJ/mol)H9004pH H11001 (96.5 kJ/V H11080 mol)?H9274 In actively respiring mitochondria, the measured ?H9274 is 0.15 to 0.20 V and the pH of the matrix is about 0.75 C 2 H5007 C 1 C 2 H5007 C 1 units more alkaline than that of the intermembrane space, so the calculated free-energy change for pump- ing protons outward is about 20 kJ/mol (of H H11001 ), most of which is contributed by the electrical portion of the electrochemical potential. Because the transfer of two electrons from NADH to O 2 is accompanied by the out- ward pumping of 10 H H11001 (Eqn 19–7), roughly 200 kJ of the 220 kJ released by oxidation of a mole of NADH is conserved in the proton gradient. When protons flow spontaneously down their elec- trochemical gradient, energy is made available to do work. In mitochondria, chloroplasts, and aerobic bacte- ria, the electrochemical energy in the proton gradient drives the synthesis of ATP from ADP and P i . We return to the energetics and stoichiometry of ATP synthesis driven by the electrochemical potential of the proton gradient in Section 19.2. 19.1 Electron-Transfer Reactions in Mitochondria 703 Intermembrane space (P side) Matrix (N side) 4H + 1 – 2 O 2 + 2H + H 2 O Succinate Fumarate 4H + II NADH + H + NAD + Cyt c IV 2H + III I Q FIGURE 19–15 Summary of the flow of electrons and protons through the four complexes of the respiratory chain. Electrons reach Q through Complexes I and II. QH 2 serves as a mobile carrier of elec- trons and protons. It passes electrons to Complex III, which passes them to another mobile connecting link, cytochrome c. Complex IV then transfers electrons from reduced cytochrome c to O 2 . Electron flow through Complexes I, III, and IV is accompanied by proton flow from the matrix to the intermembrane space. Recall that electrons from H9252 oxidation of fatty acids can also enter the respiratory chain through Q (see Fig. 19–8). N side [H H11001 ] N H11005 C 1 OH H11002 OH H11002 OH H11002 OH H11002 OH H11002 OH H11002 OH H11002 H H11001 H H11001 H H11001 H H11001 H H11001 H H11001 H H11001 P side [H H11001 ] P H11005 C 2 H9004G H11005 RT ln (C 2 /C 1 ) H11001 Z? H9004 H9274 H11005 2.3RT H9004pH H11001 ? ? H9274 H H11001 Proton pump H9274 FIGURE 19–16 Proton-motive force. The inner mitochondrial mem- brane separates two compartments of different [H H11001 ], resulting in dif- ferences in chemical concentration (H9004pH) and charge distribution (H9004H9274) across the membrane. The net effect is the proton-motive force (H9004G), which can be calculated as shown here. This is explained more fully in the text. 8885d_c19_690-750 3/1/04 11:32 AM Page 703 mac76 mac76:385_reb: Plant Mitochondria Have Alternative Mechanisms for Oxidizing NADH Plant mitochondria supply the cell with ATP during periods of low illumination or darkness by mechanisms entirely analogous to those used by nonphotosynthetic organisms. In the light, the principal source of mito- chondrial NADH is a reaction in which glycine, produced by a process known as photorespiration, is converted to serine (see Fig. 20–21): 2 Glycine H11001 NAD H11001 88n serine H11001 CO 2 H11001 NH 3 H11001 NADH H11001 H H11001 For reasons discussed in Chapter 20, plants must carry out this reaction even when they do not need NADH for ATP production. To regenerate NAD H11001 from unneeded NADH, plant mitochondria transfer electrons from NADH directly to ubiquinone and from ubiquinone directly to O 2 , bypassing Complexes III and IV and their proton pumps. In this process the energy in NADH is dissipated as heat, which can sometimes be of value to the plant (Box 19–1). Unlike cytochrome oxidase (Complex IV), the alternative QH 2 oxidase is not inhibited by cyanide. Cyanide-resistant NADH oxidation is therefore the hall- mark of this unique plant electron-transfer pathway. SUMMARY 19.1 Electron-Transfer Reactions in Mitochondria ■ Chemiosmotic theory provides the intellectual framework for understanding many biological energy transductions, including oxidative phosphorylation and photophosphorylation. The mechanism of energy coupling is similar in both cases: the energy of electron flow is conserved by the concomitant pumping of protons across the membrane, producing an electrochemical gradient, the proton-motive force. ■ In mitochondria, hydride ions removed from substrates by NAD-linked dehydrogenases donate electrons to the respiratory (electron-transfer) chain, which transfers the electrons to molecular O 2 , reducing it to H 2 O. ■ Shuttle systems convey reducing equivalents from cytosolic NADH to mitochondrial NADH. Reducing equivalents from all NAD-linked dehydrogenations are transferred to mito- chondrial NADH dehydrogenase (Complex I). ■ Reducing equivalents are then passed through a series of Fe-S centers to ubiquinone, which transfers the electrons to cytochrome b, the first carrier in Complex III. In this complex, electrons take two separate paths through two b-type cytochromes and cytochrome c 1 to an Fe-S center. The Fe-S center passes electrons, one at a time, through cytochrome c and into Complex IV, cytochrome oxidase. This copper-containing enzyme, which also contains cytochromes a and a 3 , accumulates electrons, then passes them to O 2 , reducing it to H 2 O. ■ Some electrons enter this chain of carriers through alternative paths. Succinate is oxidized by succinate dehydrogenase (Complex II), which contains a flavoprotein that passes electrons through several Fe-S centers to ubiquinone. Electrons derived from the oxidation of fatty acids pass to ubiquinone via the electron-transferring flavoprotein. ■ Plants also have an alternative, cyanide-resistant NADH oxidation pathway. 19.2 ATP Synthesis How is a concentration gradient of protons transformed into ATP? We have seen that electron transfer releases, and the proton-motive force conserves, more than enough free energy (about 200 kJ) per “mole” of elec- tron pairs to drive the formation of a mole of ATP, which requires about 50 kJ (see Box 13–1). Mi- tochondrial oxidative phospho- rylation therefore poses no thermodynamic problem. But what is the chemical mechanism that couples proton flux with phosphorylation? The chemiosmotic model, proposed by Peter Mitchell, is the paradigm for this mecha- nism. According to the model (Fig. 19–17), the electrochemi- cal energy inherent in the difference in proton concen- tration and separation of charge across the inner mito- chondrial membrane—the proton-motive force—drives the synthesis of ATP as protons flow passively back into the matrix through a proton pore associated with ATP synthase. To emphasize this crucial role of the proton- motive force, the equation for ATP synthesis is some- times written ADP H11001 P i H11001 nH H11001 P O n ATP H11001 H 2 O H11001 nH H11001 N (19–10) Mitchell used “chemiosmotic” to describe enzymatic re- actions that involve, simultaneously, a chemical reaction and a transport process. The operational definition of “coupling” is shown in Figure 19–18. When isolated mi- tochondria are suspended in a buffer containing ADP, P i , and an oxidizable substrate such as succinate, three easily measured processes occur: (1) the substrate is oxidized (succinate yields fumarate), (2) O 2 is consumed, and (3) ATP is synthesized. Oxygen consumption and ATP synthesis depend on the presence of an oxidizable substrate (succinate in this case) as well as ADP and P i . Chapter 19 Oxidative Phosphorylation and Photophosphorylation704 Peter Mitchell, 1920–1992 8885d_c19_690-750 3/1/04 11:32 AM Page 704 mac76 mac76:385_reb: Because the energy of substrate oxidation drives ATP synthesis in mitochondria, we would expect in- hibitors of the passage of electrons to O 2 (such as cyanide, carbon monoxide, and antimycin A) to block ATP synthesis (Fig. 19–18a). More surprising is the find- ing that the converse is also true: inhibition of ATP syn- thesis blocks electron transfer in intact mitochondria. This obligatory coupling can be demonstrated in isolated mitochondria by providing O 2 and oxidizable substrates, but not ADP (Fig. 19–18b). Under these conditions, no ATP synthesis can occur and electron transfer to O 2 does not proceed. Coupling of oxidation and phosphor- ylation can also be demonstrated using oligomycin or venturicidin, toxic antibiotics that bind to the ATP syn- thase in mitochondria. These compounds are potent in- hibitors of both ATP synthesis and the transfer of elec- trons through the chain of carriers to O 2 (Fig. 19–18b). Because oligomycin is known to interact not directly with the electron carriers but with ATP synthase, it follows that electron transfer and ATP synthesis are obligately coupled; neither reaction occurs without the other. Chemiosmotic theory readily explains the depend- ence of electron transfer on ATP synthesis in mitochon- dria. When the flow of protons into the matrix through the proton channel of ATP synthase is blocked (with oligomycin, for example), no path exists for the return of protons to the matrix, and the continued extrusion of protons driven by the activity of the respiratory chain generates a large proton gradient. The proton-motive force builds up until the cost (free energy) of pumping 19.2 ATP Synthesis 705 NADH + H + NAD + Succinate Fumarate Cyt c + – ADP + P i ATP + + + + + + + + + + + + + + + – – – – – – – – – – – – 4H + 4H + 2H + H + Chemical potential ?pΗ (inside alkaline) ATP synthesis driven by proton-motive force Electrical potential ?w (inside negative) + + – –O 2 + H 2 2H + O 2 1 II IV I III F o F 1 Intermembrane space Matrix Q FIGURE 19–17 Chemiosmotic model. In this simple representation of the chemiosmotic theory applied to mitochondria, electrons from NADH and other oxidizable substrates pass through a chain of carriers arranged asymmet- rically in the inner membrane. Electron flow is accompanied by proton transfer across the membrane, producing both a chemical gradient (H9004pH) and an electrical gradient (H9004H9274). The inner mitochondrial membrane is imper- meable to protons; protons can reenter the matrix only through proton-specific channels (F o ). The proton-motive force that drives protons back into the matrix provides the energy for ATP synthesis, catalyzed by the F 1 complex associated with F o . O 2 consumed Add ADP H11001 P i Add succinate Time(b) ATP synthesized Add venturicidin or oligomycin Add DNP Uncoupled O 2 consumed Add ADP H11001 P i Add succinate Time(a) ATP synthesized Add CN H11002 FIGURE 19–18 Coupling of electron transfer and ATP synthesis in mitochondria. In experiments to demonstrate coupling, mitochondria are suspended in a buffered medium and an O 2 electrode monitors O 2 consumption. At intervals, samples are removed and assayed for the presence of ATP. (a) Addition of ADP and P i alone results in little or no increase in either respiration (O 2 consumption; black) or ATP synthe- sis (red). When succinate is added, respiration begins immediately and ATP is synthesized. Addition of cyanide (CN H11002 ), which blocks electron transfer between cytochrome oxidase and O 2 , inhibits both respiration and ATP synthesis. (b) Mitochondria provided with succinate respire and synthesize ATP only when ADP and P i are added. Subsequent ad- dition of venturicidin or oligomycin, inhibitors of ATP synthase, blocks both ATP synthesis and respiration. Dinitrophenol (DNP) is an un- coupler, allowing respiration to continue without ATP synthesis. 8885d_c19_690-750 3/1/04 11:32 AM Page 705 mac76 mac76:385_reb: Chapter 19 Oxidative Phosphorylation and Photophosphorylation706 BOX 19–1 THE WORLD OF BIOCHEMISTRY Hot, Stinking Plants and Alternative Respiratory Pathways Many flowering plants attract insect pollinators by re- leasing odorant molecules that mimic an insect’s nat- ural food sources or potential egg-laying sites. Plants pollinated by flies or beetles that normally feed on or lay their eggs in dung or carrion sometimes use foul- smelling compounds to attract these insects. One family of stinking plants is the Araceae, which includes philodendrons, arum lilies, and skunk cab- bages. These plants have tiny flowers densely packed on an erect structure, the spadix, surrounded by a modified leaf, the spathe. The spadix releases odors of rotting flesh or dung. Before pollination the spadix also heats up, in some species to as much as 20 to 40 H11034C above the ambient temperature. Heat produc- tion (thermogenesis) helps evaporate odorant mole- cules for better dispersal, and because rotting flesh and dung are usually warm from the hyperactive me- tabolism of scavenging microbes, the heat itself might also attract insects. In the case of the eastern skunk cabbage (Fig. 1), which flowers in late winter or early spring when snow still covers the ground, thermogen- esis allows the spadix to grow up through the snow. How does a skunk cabbage heat its spadix? The mitochondria of plants, fungi, and unicellular eukary- otes have electron-transfer systems that are essen- tially the same as those in animals, but they also have an alternative respiratory pathway. A cyanide- resistant QH 2 oxidase transfers electrons from the ubiquinone pool directly to oxygen, bypassing the two proton-translocating steps of Complexes III and IV (Fig. 2). Energy that might have been conserved as ATP is instead released as heat. Plant mitochondria also have an alternative NADH dehydrogenase, insen- sitive to the Complex I inhibitor rotenone (see Table 19–4), that transfers electrons from NADH in the ma- trix directly to ubiquinone, bypassing Complex I and its associated proton pumping. And plant mitochon- dria have yet another NADH dehydrogenase, on the external face of the inner membrane, that transfers electrons from NADPH or NADH in the intermem- brane space to ubiquinone, again bypassing Complex I. Thus when electrons enter the alternative respira- tory pathway through the rotenone-insensitive NADH dehydrogenase, the external NADH dehydrogenase, or succinate dehydrogenase (Complex II), and pass to O 2 via the cyanide-resistant alternative oxidase, en- ergy is not conserved as ATP but is released as heat. A skunk cabbage can use the heat to melt snow, pro- duce a foul stench, or attract beetles or flies. FIGURE 1 Eastern skunk cabbage. FIGURE 2 Electron carriers of the inner membrane of plant mitochondria. Electrons can flow through Complexes I, III, and IV, as in animal mitochondria, or through plant-specific alterna- tive carriers by the paths shown with blue arrows. Heat IV I Intermembrane space Q Matrix Cyt c NAD + NAD(P) + External NAD(P)H dehydrogenase Alternative oxidase Alternative NADH dehydrogenase III H 2 O NADH NAD(P)H 1 – 2 O 2 8885d_c19_706 3/1/04 1:59 PM Page 706 mac76 mac76:385_reb: protons out of the matrix against this gradient equals or exceeds the energy released by the transfer of electrons from NADH to O 2 . At this point electron flow must stop; the free energy for the overall process of electron flow coupled to proton pumping becomes zero, and the sys- tem is at equilibrium. Certain conditions and reagents, however, can un- couple oxidation from phosphorylation. When intact mi- tochondria are disrupted by treatment with detergent or by physical shear, the resulting membrane fragments can still catalyze electron transfer from succinate or NADH to O 2 , but no ATP synthesis is coupled to this respiration. Certain chemical compounds cause uncoupling without disrupting mitochondrial structure. Chemical uncouplers include 2,4-dinitrophenol (DNP) and carbonylcyanide-p- trifluoromethoxyphenylhydrazone (FCCP) (Table 19–4; Fig. 19–19), weak acids with hydrophobic properties that permit them to diffuse readily across mitochondrial membranes. After entering the matrix in the protonated form, they can release a proton, thus dissipating the proton gradient. Ionophores such as valinomycin (see Fig. 11–45) allow inorganic ions to pass easily through membranes. Ionophores uncouple electron transfer from oxidative phosphorylation by dissipating the electrical contribution to the electrochemical gradient across the mitochondrial membrane. 19.2 ATP Synthesis 707 A prediction of the chemiosmotic theory is that, be- cause the role of electron transfer in mitochondrial ATP synthesis is simply to pump protons to create the elec- trochemical potential of the proton-motive force, an ar- tificially created proton gradient should be able to re- place electron transfer in driving ATP synthesis. This has been experimentally confirmed (Fig. 19–20). Mito- chondria manipulated so as to impose a difference of proton concentration and a separation of charge across the inner membrane synthesize ATP in the absence of an oxidizable substrate; the proton-motive force alone suffices to drive ATP synthesis. FIGURE 19–19 Two chemical uncouplers of oxidative phosphoryla- tion. Both DNP and FCCP have a dissociable proton and are very hydrophobic. They carry protons across the inner mitochondrial mem- brane, dissipating the proton gradient. Both also uncouple photo- phosphorylation (see Fig. 19–57). FIGURE 19–20 Evidence for the role of a proton gradient in ATP syn- thesis. An artificially imposed electrochemical gradient can drive ATP synthesis in the absence of an oxidizable substrate as electron donor. In this two-step experiment, (a) isolated mitochondria are first incu- bated in a pH 9 buffer containing 0.1 M KCl. Slow leakage of buffer and KCl into the mitochondria eventually brings the matrix into equi- librium with the surrounding medium. No oxidizable substrates are present. (b) Mitochondria are now separated from the pH 9 buffer and resuspended in pH 7 buffer containing valinomycin but no KCl. The change in buffer creates a difference of two pH units across the inner mitochondrial membrane. The outward flow of K H11001 , carried (by vali- nomycin) down its concentration gradient without a counterion, cre- ates a charge imbalance across the membrane (matrix negative). The sum of the chemical potential provided by the pH difference and the electrical potential provided by the separation of charges is a proton- motive force large enough to support ATP synthesis in the absence of an oxidizable substrate. N NH H11001 H H11001 N H11002 2,4-Dinitrophenol (DNP) Carbonylcyanide-p- trifluoromethoxyphenylhydrazone (FCCP) OH NO 2 O H11002 H11001 H H11001 NO 2 NO 2 NO 2 N C CC NN C CC NN O CFF F O CFF F [K H11001 ] H11005 [Cl H11002 ] H11005 0.1 M [H H11001 ] H11005 10 H110029 M [H H11001 ] H11005 10 H110029 M F o F 1 pH lowered from 9 to 7; valinomycin present; no K H11001 (a) [K H11001 ] H11021 [Cl H11002 ] [H H11001 ] H11005 10 H110029 M (b) [H H11001 ] H11005 10 H11002 7 M ADPH11001 P i K H11001 K H11001 H11001 H11001 H11001 H11001 H11001 H11001 H11001 H11002 H11002 H11002 H11002 H11002 H11002 H11002 H11002 ATP Matrix Intermembrane space 8885d_c19_690-750 3/1/04 11:32 AM Page 707 mac76 mac76:385_reb: ATP Synthase Has Two Functional Domains, F o and F 1 Mitochondrial ATP synthase is an F-type ATPase (see Fig. 11–39; Table 11–3) similar in structure and mech- anism to the ATP synthases of chloroplasts and eubac- teria. This large enzyme complex of the inner mito- chondrial membrane catalyzes the formation of ATP from ADP and P i , accompanied by the flow of protons from the P to the N side of the mem- brane (Eqn 19–10). ATP syn- thase, also called Complex V, has two distinct components: F 1 , a peripheral membrane protein, and F o (o denoting oligomycin-sensitive), which is integral to the membrane. F 1 , the first factor recognized as essential for oxidative phos- phorylation, was identified and purified by Efraim Racker and his colleagues in the early 1960s. In the laboratory, small membrane vesicles formed from inner mitochondrial membranes carry out ATP syn- thesis coupled to electron transfer. When F 1 is gently extracted, the “stripped” vesicles still contain intact res- piratory chains and the F o portion of ATP synthase. The vesicles can catalyze electron transfer from NADH to O 2 but cannot produce a proton gradient: F o has a proton pore through which protons leak as fast as they are pumped by electron transfer, and without a proton gra- dient the F 1 -depleted vesicles cannot make ATP. Iso- lated F 1 catalyzes ATP hydrolysis (the reversal of syn- thesis) and was therefore originally called F 1 ATPase. When purified F 1 is added back to the depleted vesicles, it reassociates with F o , plugging its proton pore and restoring the membrane’s capacity to couple electron transfer and ATP synthesis. ATP Is Stabilized Relative to ADP on the Surface of F 1 Isotope exchange experiments with purified F 1 reveal a remarkable fact about the enzyme’s catalytic mecha- nism: on the enzyme surface, the reaction ADP H11001 P i ATP H11001 H 2 O is readily reversible—the free-energy change for ATP synthesis is close to zero! When ATP is hydrolyzed by F 1 in the presence of 18 O-labeled water, the P i released contains an 18 O atom. Careful meas- urement of the 18 O content of P i formed in vitro by F 1 - catalyzed hydrolysis of ATP reveals that the P i has not one, but three or four 18 O atoms (Fig. 19–21). This in- dicates that the terminal pyrophosphate bond in ATP is cleaved and re-formed repeatedly before P i leaves the enzyme surface. With P i free to tumble in its binding site, each hydrolysis inserts 18 O randomly at one of the z y Chapter 19 Oxidative Phosphorylation and Photophosphorylation708 Efraim Racker, 1913–1991 18 O 18 O ADP ATP H11001 H 2 18 O H11001 18 OP 18 O Enzyme (F 1 ) (a) ADP a-Arg 376 b-Arg 182 b-Glu 181 b-Lys 155 Mg 2+ FIGURE 19–21 Catalytic mechanism of F 1 . (a) 18 O-exchange exper- iment. F 1 solubilized from mitochondrial membranes is incubated with ATP in the presence of 18 O-labeled water. At intervals, a sample of the solution is withdrawn and analyzed for the incorporation of 18 O into the P i produced from ATP hydrolysis. In minutes, the P i contains three or four 18 O atoms, indicating that both ATP hydrolysis and ATP synthesis have occurred several times during the incubation. (b) The likely transition state complex for ATP hydrolysis and synthesis in ATP synthase (derived from PDB ID 1BMF). The H9251 subunit is shown in green, H9252 in gray. The positively charged residues H9252-Arg 182 and H9251-Arg 376 coordinate two oxygens of the pentavalent phosphate intermediate; H9252- Lys 155 interacts with a third oxygen, and the Mg 2H11001 ion (green sphere) further stabilizes the intermediate. The blue sphere represents the leav- ing group (H 2 O). These interactions result in the ready equilibration of ATP and ADP H11001 P i in the active site. (b) 8885d_c19_690-750 3/1/04 11:32 AM Page 708 mac76 mac76:385_reb: four positions in the molecule. This exchange reaction occurs in unenergized F o F 1 complexes (with no proton gradient) and with isolated F 1 —the exchange does not require the input of energy. Kinetic studies of the initial rates of ATP synthesis and hydrolysis confirm the conclusion that H9004GH11032H11034 for ATP synthesis on the enzyme is near zero. From the meas- ured rates of hydrolysis (k 1 H11005 10 s H110021 ) and synthesis (k H110021 H11005 24 s H110021 ), the calculated equilibrium constant for the reaction Enz-ATP Enz–(ADP H11001 P i ) is KH11032 eq H11005H11005 H110052.4 From this KH11032 eq , the calculated apparent H9004GH11032H11034 is close to zero. This is much different from the KH11032 eq of about 10 5 (H9004GH11032H11034 H11005 H1100230.5 kJ/mol) for the hydrolysis of ATP free in solution (not on the enzyme surface). What accounts for the huge difference? ATP syn- thase stabilizes ATP relative to ADP H11001 P i by binding ATP more tightly, releasing enough energy to counterbalance the cost of making ATP. Careful measurements of the binding constants show that F o F 1 binds ATP with very high affinity (K d ≤ 10 H1100212 M) and ADP with much lower affinity (K d ≈ 10 H110025 M). The difference in K d corresponds to a difference of about 40 kJ/mol in binding energy, and this binding energy drives the equilibrium toward for- mation of the product ATP. The Proton Gradient Drives the Release of ATP from the Enzyme Surface Although ATP synthase equilibrates ATP with ADP H11001 P i , in the absence of a proton gradient the newly syn- thesized ATP does not leave the surface of the enzyme. 24 s H110021 H5007 10 s H110021 k H110021 H5007 k 1 z y It is the proton gradient that causes the enzyme to re- lease the ATP formed on its surface. The reaction co- ordinate diagram of the process (Fig. 19–22) illustrates the difference between the mechanism of ATP synthase and that of many other enzymes that catalyze ender- gonic reactions. For the continued synthesis of ATP, the enzyme must cycle between a form that binds ATP very tightly and a form that releases ATP. Chemical and crystallo- graphic studies of the ATP synthase have revealed the structural basis for this alternation in function. Each H9252 Subunit of ATP Synthase Can Assume Three Different Conformations Mitochondrial F 1 has nine subunits of five different types, with the composition H9251 3 H9252 3 H9253H9254H9255. Each of the three H9252 subunits has one catalytic site for ATP synthesis. The crystallographic determination of the F 1 structure by John E. Walker and colleagues revealed structural de- tails very helpful in explaining the catalytic mechanism of the enzyme. The knoblike portion of F 1 is a flattened sphere, 8 nm high and 10 nm across, consisting of al- ternating H9251 and H9252 subunits arranged like the sections of an orange (Fig. 19–23a–c). The polypeptides that make up the stalk in the F 1 crystal structure are asymmetri- cally arranged, with one domain of the single H9253 subunit making up a central shaft that passes through F 1 , and another domain of H9253 associated primarily with one of the three H9252 subunits, designated H9252-empty (Fig. 19–23c). Although the amino acid sequences of the three H9252 sub- units are identical, their conformations differ, in part because of the association of the H9253 subunit with just one of the three. The structures of the H9254 and H9255 subunits are not revealed in these crystallographic studies. The conformational differences among H9252 subunits extend to differences in their ATP/ADP-binding sites. 19.2 ATP Synthesis 709 G (kJ/mol) Reaction coordinate 80 60 40 20 0 ? P ADPH11001P i ES E H11001 S E ADPH11001P i [E ATP] Typical enzyme ATP synthase ATP (in solution) FIGURE 19–22 Reaction coordinate diagrams for ATP synthase and for a more typical enzyme. In a typical enzyme-catalyzed reaction (left), reaching the transition state (?) between substrate and product is the major energy barrier to overcome. In the reaction catalyzed by ATP synthase (right), release of ATP from the enzyme, not formation of ATP, is the major energy barrier. The free-energy change for the formation of ATP from ADP and P i in aqueous solution is large and positive, but on the enzyme surface, the very tight binding of ATP provides sufficient binding energy to bring the free energy of the enzyme-bound ATP close to that of ADP H11001 P i , so the reaction is readily reversible. The equilibrium constant is near 1. The free energy required for the release of ATP is provided by the proton-motive force. 8885d_c19_690-750 3/1/04 11:32 AM Page 709 mac76 mac76:385_reb: When researchers crystallized the protein in the pres- ence of ADP and App(NH)p, a close structural analog of ATP that cannot be hydrolyzed by the ATPase activ- ity of F 1 , the binding site of one of the three H9252 subunits was filled with App(NH)p, the second was filled with Chapter 19 Oxidative Phosphorylation and Photophosphorylation710 b-ATP b-ADP b-empty a-ADP a-empty a-ATP (c) ATP ADP (b) H9251 H9251H9251 H9252 H9252 H9252 H9253 (a) FIGURE 19–23 Mitochondrial ATP synthase complex. (a) Structure of the F 1 complex, deduced from crystallographic and biochemical studies. In F 1 , three H9251 and three H9252 subunits are arranged like the seg- ments of an orange, with alternating H9251 (shades of gray) and H9252 (shades of purple) subunits around a central shaft, the H9253 subunit (green). (b) Crystal structure of bovine F 1 (PDB ID 1BMF), viewed from the side. Two H9251 subunits and one H9252 subunit have been omitted to reveal the central shaft (H9253 subunit) and the binding sites for ATP (red) and ADP (yellow) on the H9252 subunits. The H9254 and H9255 subunits are not shown here. (c) F 1 viewed from above (that is, from the N side of the membrane), showing the three H9252 and three H9251 subunits and the central shaft (H9253 sub- unit, green). Each H9252 subunit, near its interface with the neighboring H9251 subunit, has a nucleotide-binding site critical to the catalytic activity. The single H9253 subunit associates primarily with one of the three H9251H9252 pairs, forcing each of the three H9252 subunits into slightly different con- formations, with different nucleotide-binding sites. In the crystalline enzyme, one subunit (H9252-ADP) has ADP (yellow) in its binding site, the next (H9252-ATP) has ATP (red), and the third (H9252-empty) has no bound nu- cleotide. (d) Side view of the F o F 1 structure. This is a composite, in which the crystallographic coordinates of bovine mitochondrial F 1 (shades of purple and gray) have been combined with those of yeast mitochondrial F o (shades of yellow and orange) (PDB ID 1QO1). Sub- units a, b, H9254, and H9255 were not part of the crystal structure shown here. (e) The F o F 1 structure, viewed end-on in the direction P side to N side. The major structures visible in this cross section are the two trans- membrane helices of each of ten c subunits arranged in concentric circles. (f) Diagram of the F o F 1 complex, deduced from biochemical and crystallographic studies. The two b subunits of F o associate firmly with the H9251 and H9252 subunits of F 1 , holding them fixed relative to the membrane. In F o , the membrane-embedded cylinder of c subunits is attached to the shaft made up of F 1 subunits H9253 and H9255. As protons flow through the membrane from the P side to the N side through F o , the cylinder and shaft rotate, and the H9252 subunits of F 1 change conforma- tion as the H9253 subunit associates with each in turn. John E. Walker Nonhydrolyzable CH 2 O P P H NO O OO O H N N NH 2 N N H OHOH HH P O H11002 O H11002 O H11002 O H11002 H9252H9253H9251 H9252-H9253 bondH9252 App(NH)p (H9252,H9253-imidoadenosine 5H11032-triphosphate) 8885d_c19_690-750 3/1/04 11:32 AM Page 710 mac76 mac76:385_reb: F 1 F o (d) ADP, and the third was empty. The corresponding H9252 subunit conformations are designated H9252-ATP, H9252-ADP, and H9252-empty (Fig. 19–23c). This difference in nucleo- tide binding among the three subunits is critical to the mechanism of the complex. The F o complex making up the proton pore is composed of three subunits, a, b, and c, in the propor- tion ab 2 c 10–12 . Subunit c is a small (M r 8,000), very hydrophobic polypeptide, consisting almost entirely of two transmembrane helices, with a small loop extend- ing from the matrix side of the membrane. The crystal structure of the yeast F o F 1 , solved in 1999, shows the arrangement of the c subunits. The yeast complex has ten c subunits, each with two transmembrane helices roughly perpendicular to the plane of the membrane and arranged in two concentric circles (Fig. 19–23d, e). The inner circle is made up of the amino-terminal helices of each c subunit; the outer circle, about 55 ? in diame- ter, is made up of the carboxyl-terminal helices. The H9255 and H9253 subunits of F 1 form a leg-and-foot that projects from the bottom (membrane) side of F 1 and stands firmly on the ring of c subunits. The schematic drawing in Figure 19–23f combines the structural information from studies of bovine F 1 and yeast F o F 1 . Rotational Catalysis Is Key to the Binding-Change Mechanism for ATP Synthesis On the basis of detailed kinetic and binding studies of the reactions catalyzed by F o F 1 , Paul Boyer proposed a rotational catalysis mechanism in which the three active sites of F 1 take turns catalyzing ATP synthesis 19.2 ATP Synthesis 711 (e) H9251 H9280 H9251H9251 H9252 H9252 H9252 H9253 H9254 b 2 c 10 H + a N side P side ADP + P i ATP (f) 8885d_c19_690-750 3/1/04 11:32 AM Page 711 mac76 mac76:385_reb: (Fig. 19–24). A given H9252 sub- unit starts in the H9252-ADP con- formation, which binds ADP and P i from the surrounding medium. The subunit now changes conformation, assum- ing the H9252-ATP form that tightly binds and stabilizes ATP, bringing about the ready equilibration of ADP H11001 P i with ATP on the enzyme surface. Finally, the subunit changes to the H9252-empty conformation, which has very low affinity for ATP, and the newly synthesized ATP leaves the en- zyme surface. Another round of catalysis begins when this subunit again assumes the H9252-ADP form and binds ADP and P i . The conformational changes central to this mecha- nism are driven by the passage of protons through the F o portion of ATP synthase. The streaming of protons through the F o “pore” causes the cylinder of c subunits and the attached H9253 subunit to rotate about the long axis of H9253, which is perpendicular to the plane of the mem- brane. The H9253 subunit passes through the center of the H9251 3 H9252 3 spheroid, which is held stationary relative to the membrane surface by the b 2 and H9254 subunits (Fig. 19–23f). With each rotation of 120H11034, H9253 comes into con- tact with a different H9252 subunit, and the contact forces that H9252 subunit into the H9252-empty conformation. The three H9252 subunits interact in such a way that when one assumes the H9252-empty conformation, its neigh- bor to one side must assume the H9252-ADP form, and the other neighbor the H9252-ATP form. Thus one complete ro- tation of the H9253 subunit causes each H9252 subunit to cycle through all three of its possible conformations, and for each rotation, three ATP are synthesized and released from the enzyme surface. One strong prediction of this binding-change model is that the H9253 subunit should rotate in one direction when F o F 1 is synthesizing ATP and in the opposite direction when the enzyme is hydrolyzing ATP. This prediction was confirmed in elegant experiments in the laborato- ries of Masasuke Yoshida and Kazuhiko Kinosita, Jr. The rotation of H9253 in a single F 1 molecule was observed mi- croscopically by attaching a long, thin, fluorescent actin polymer to H9253 and watching it move relative to H9251 3 H9252 3 im- mobilized on a microscope slide, as ATP was hydrolyzed. When the entire F o F 1 complex (not just F 1 ) was used in a similar experiment, the entire ring of c subunits ro- tated with H9253 (Fig. 19–25). The “shaft” rotated in the pre- dicted direction through 360H11034. The rotation was not smooth, but occurred in three discrete steps of 120H11034. As calculated from the known rate of ATP hydrolysis by one F 1 molecule and from the frictional drag on the long actin polymer, the efficiency of this mechanism in con- verting chemical energy into motion is close to 100%. It is, in Boyer’s words, “a splendid molecular machine!” Chemiosmotic Coupling Allows Nonintegral Stoichiometries of O 2 Consumption and ATP Synthesis Before the general acceptance of the chemiosmotic model for oxidative phosphorylation, the assumption was that the overall reaction equation would take the following form: xADP H11001 xP i H11001 H5007 1 2 H5007 O 2 H11001 H H11001 H11001 NADH 88n xATP H11001 H 2 O H11001 NAD H11001 (19–11) with the value of x—sometimes called the P/O ratio or the P/2e H11546 ratio—always an integer. When intact mito- Chapter 19 Oxidative Phosphorylation and Photophosphorylation712 ATP ATP ADP +P i H9252 H9251 H9252 H9251 H9252 H9251 ATP ATP ADP +P i H9252 H9251 H9252 H9251 H9252 H9251 ATP ATP ADP +P i H9252 H9251 H9252 H9251 H9252 H9251 3 H P + 3 H P + 3 H N + 3 H N + 3 H N + 3 H P + FIGURE 19–24 Binding-change model for ATP synthase. The F 1 com- plex has three nonequivalent adenine nucleotide–binding sites, one for each pair of H9251 and H9252 subunits. At any given moment, one of these sites is in the H9252-ATP conformation (which binds ATP tightly), a second is in the H9252-ADP (loose-binding) conformation, and a third is in the H9252- empty (very-loose-binding) conformation. The proton-motive force causes rotation of the central shaft—the H9253 subunit, shown as a green arrowhead—which comes into contact with each H9251H9252 subunit pair in succession. This produces a cooperative conformational change in which the H9252-ATP site is converted to the H9252-empty conformation, and ATP dissociates; the H9252-ADP site is converted to the H9252-ATP conforma- tion, which promotes condensation of bound ADP H11001 P i to form ATP; and the H9252-empty site becomes a H9252-ADP site, which loosely binds ADP H11001 P i entering from the solvent. This model, based on experimental findings, requires that at least two of the three catalytic sites alternate in activity; ATP cannot be released from one site unless and until ADP and P i are bound at the other. Paul Boyer 8885d_c19_690-750 3/1/04 11:32 AM Page 712 mac76 mac76:385_reb: ADP + P i ATP H9251 H9251 H9254 H9252 H9253 Ni complex His residues His residues Avidin F o F 1 a b c Actin filament chondria are suspended in solution with an oxidizable substrate such as succinate or NADH and are provided with O 2 , ATP synthesis is readily measurable, as is the decrease in O 2 . Measurement of P/O, however, is com- plicated by the fact that intact mitochondria consume ATP in many reactions taking place in the matrix, and they consume O 2 for purposes other than oxidative phosphorylation. Most experiments have yielded P/O (ATP to H5007 1 2 H5007O 2 ) ratios of between 2 and 3 when NADH was the electron donor, and between 1 and 2 when suc- cinate was the donor. Given the assumption that P/O should have an integral value, most experimenters agreed that the P/O ratios must be 3 for NADH and 2 for succinate, and for years those values have appeared in research papers and textbooks. With introduction of the chemiosmotic paradigm for coupling ATP synthesis to electron transfer, there was no theoretical requirement for P/O to be integral. The relevant questions about stoichiometry became, how many protons are pumped outward by electron transfer from one NADH to O 2 , and how many protons must flow inward through the F o F 1 complex to drive the synthe- sis of one ATP? The measurement of proton fluxes is technically complicated; the investigator must take into account the buffering capacity of mitochondria, non- productive leakage of protons across the inner mem- brane, and use of the proton gradient for functions other than ATP synthesis, such as driving the transport of sub- strates across the inner mitochondrial membrane (de- scribed below). The consensus values for number of pro- tons pumped out per pair of electrons are 10 for NADH and 6 for succinate. The most widely accepted experi- mental value for number of protons required to drive the synthesis of an ATP molecule is 4, of which 1 is used in transporting P i , ATP, and ADP across the mitochon- drial membrane (see below). If 10 protons are pumped out per NADH and 4 must flow in to produce 1 ATP, the proton-based P/O ratio is 2.5 for NADH as the electron donor and 1.5 (6/4) for succinate. We use the P/O val- ues of 2.5 and 1.5 throughout this book, but the values 3.0 and 2.0 are still common in the biochemical litera- ture. The final word on proton stoichiometry will prob- ably not be written until we know the full details of the F o F 1 reaction mechanism. The Proton-Motive Force Energizes Active Transport Although the primary role of the proton gradient in mi- tochondria is to furnish energy for the synthesis of ATP, the proton-motive force also drives several transport processes essential to oxidative phosphorylation. The inner mitochondrial membrane is generally imperme- able to charged species, but two specific systems trans- port ADP and P i into the matrix and ATP out to the cy- tosol (Fig. 19–26). The adenine nucleotide translocase, integral to the inner membrane, binds ADP 3H11002 in the intermembrane space and transports it into the matrix in exchange for an ATP 4H11002 molecule simultaneously transported outward (see Fig. 13–1 for the ionic forms of ATP and ADP). Be- cause this antiporter moves four negative charges out for every three moved in, its activity is favored by the 19.2 ATP Synthesis 713 FIGURE 19–25 Rotation of F o and H9253 experimentally demonstrated. F 1 genetically engineered to contain a run of His residues adheres tightly to a microscope slide coated with a Ni complex; biotin is co- valently attached to a c subunit of F o . The protein avidin, which binds biotin very tightly, is covalently attached to long filaments of actin la- beled with a fluorescent probe. Biotin-avidin binding now attaches the actin filaments to the c subunit. When ATP is provided as sub- strate for the ATPase activity of F 1 , the labeled filament is seen to ro- tate continuously in one direction, proving that the F o cylinder of c subunits rotates. In another experiment, a fluorescent actin filament was attached directly to the H9253 subunit. The series of fluorescence mi- crographs shows the position of the actin filament at intervals of 133 ms. Note that as the filament rotates, it makes a discrete jump about every eleventh frame. Presumably the cylinder and shaft move as one unit. 8885d_c19_690-750 3/1/04 11:32 AM Page 713 mac76 mac76:385_reb: transmembrane electrochemical gradient, which gives the matrix a net negative charge; the proton-motive force drives ATP-ADP exchange. Adenine nucleotide translocase is specifically inhibited by atractyloside, a toxic glycoside formed by a species of thistle. If the transport of ADP into and ATP out of mitochondria is inhibited, cytosolic ATP cannot be regenerated from ADP, explaining the toxicity of atractyloside. A second membrane transport system essential to oxidative phosphorylation is the phosphate translo- case, which promotes symport of one H 2 PO 4 H11002 and one H H11001 into the matrix. This transport process, too, is fa- vored by the transmembrane proton gradient (Fig. 19–26). Notice that the process requires movement of one proton from the P to the N side of the inner mem- brane, consuming some of the energy of electron trans- fer. A complex of the ATP synthase and both translo- cases, the ATP synthasome, can be isolated from mitochondria by gentle dissection with detergents, sug- gesting that the functions of these three proteins are very tightly integrated. Shuttle Systems Indirectly Convey Cytosolic NADH into Mitochondria for Oxidation The NADH dehydrogenase of the inner mitochondrial membrane of animal cells can accept electrons only from NADH in the matrix. Given that the inner membrane is not permeable to NADH, how can the NADH generated by glycolysis in the cytosol be reoxidized to NAD H11001 by O 2 via the respiratory chain? Special shuttle systems carry reducing equivalents from cytosolic NADH into mitochondria by an indirect route. The most active NADH shuttle, which functions in liver, kidney, and heart mitochondria, is the malate-aspartate shuttle (Fig. 19–27). The reducing equivalents of cytosolic NADH are first transferred to cytosolic oxaloacetate to yield malate, catalyzed by cytosolic malate dehydroge- nase. The malate thus formed passes through the inner membrane via the malate–H9251-ketoglutarate transporter. Within the matrix the reducing equivalents are passed to NAD H11001 by the action of matrix malate dehydrogenase, forming NADH; this NADH can pass electrons directly to the respiratory chain. About 2.5 molecules of ATP are generated as this pair of electrons passes to O 2 . Cy- tosolic oxaloacetate must be regenerated by transami- nation reactions and the activity of membrane trans- porters to start another cycle of the shuttle. Skeletal muscle and brain use a different NADH shuttle, the glycerol 3-phosphate shuttle (Fig. 19–28). It differs from the malate-aspartate shuttle in that it delivers the reducing equivalents from NADH to ubiquinone and thus into Complex III, not Complex I (Fig. 19–8), providing only enough energy to synthesize 1.5 ATP molecules per pair of electrons. The mitochondria of plants have an externally ori- ented NADH dehydrogenase that can transfer electrons directly from cytosolic NADH into the respiratory chain at the level of ubiquinone. Because this pathway by- passes the NADH dehydrogenase of Complex I and the associated proton movement, the yield of ATP from cy- tosolic NADH is less than that from NADH generated in the matrix (Box 19–1). SUMMARY 19.2 ATP Synthesis ■ The flow of electrons through Complexes I, III, and IV results in pumping of protons across the inner mitochondrial membrane, making the matrix alkaline relative to the intermembrane space. This proton gradient provides the energy (in the form of the proton-motive force) for ATP synthesis from ADP and P i by ATP synthase (F o F 1 complex) in the inner membrane. Chapter 19 Oxidative Phosphorylation and Photophosphorylation714 Intermembrane space Matrix Adenine nucleotide translocase (antiporter) ATP synthase Phosphate translocase (symporter) ATP 4H11002 H11001 H11001 H11001 H11001 H11002 H11002 H11002 H11001 H11001 H11001 H11002 H11002 H11002 H11001 H11001 H11002 H11002 H11001 H11001 H11002 H11002 H11002 H H11001 H 2 PO 4 H H11001 ATP 4H11002 ADP 3H11002 H H11001 H 2 PO 4 H H11001 ADP 3H11002 H11002 H11002 FIGURE 19–26 Adenine nucleotide and phosphate translocases. Transport systems of the inner mitochondrial membrane carry ADP and P i into the matrix and newly synthesized ATP into the cytosol. The adenine nucleotide translocase is an antiporter; the same protein moves ADP into the matrix and ATP out. The effect of replacing ATP 4H11002 with ADP 3H11002 is the net efflux of one negative charge, which is favored by the charge difference across the inner membrane (outside positive). At pH 7, P i is present as both HPO 4 2H11002 and H 2 PO 4 H11002 ; the phosphate translocase is specific for H 2 PO 4 H11002 . There is no net flow of charge dur- ing symport of H 2 PO 4 H11002 and H H11001 , but the relatively low proton con- centration in the matrix favors the inward movement of H H11001 . Thus the proton-motive force is responsible both for providing the energy for ATP synthesis and for transporting substrates (ADP and P i ) in and prod- uct (ATP) out of the mitochondrial matrix. All three of these transport systems can be isolated as a single membrane-bound complex (ATP synthasome). 8885d_c19_690-750 3/1/04 11:32 AM Page 714 mac76 mac76:385_reb: Intermembrane space Oxaloacetate Aspartate Oxaloacetate malate dehydrogenase malate dehydrogenase aspartate aminotransferase aspartate aminotransferase Glutamate H9251-Ketoglutarate Glutamate H9251-Ketoglutarate Glutamate-aspartate transporter Malate– H9251-ketoglutarate transporter 2 3 NAD + NADH 4 5 Matrix NAD + NADHH + + 6 1 Aspartate Malate Malate OH C COO H11002 CH 2 OOC H11002 H NH 3 H11001 NH 3 H11001 NH 3 H11001 NH 3 H11001 C COO H11002 CH 2 CH 2 H O C COO H11002 CH 2 CH 2 C COO H11002 CH 2 H C COO H11002 CH 2 CH 2 H C COO H11002 CH 2 H OH C COO H11002 CH 2 H O C COO H11002 CH 2 CH 2 O C COO H11002 CH 2 OOC H11002 O C COO H11002 CH 2 OOC H11002 OOC H11002 OOC H11002 OOC H11002 OOC H11002 OOC H11002 OOC H11002 OOC H11002 H + + FIGURE 19–27 Malate-aspartate shuttle. This shuttle for transporting reducing equivalents from cytosolic NADH into the mitochondrial ma- trix is used in liver, kidney, and heart. 1 NADH in the cytosol (in- termembrane space) passes two reducing equivalents to oxaloacetate, producing malate. 2 Malate crosses the inner membrane via the malate–H9251-ketoglutarate transporter. 3 In the matrix, malate passes two reducing equivalents to NAD H11001 , and the resulting NADH is oxi- dized by the respiratory chain. The oxaloacetate formed from malate cannot pass directly into the cytosol. 4 It is first transaminated to as- partate, which 5 can leave via the glutamate-aspartate transporter. 6 Oxaloacetate is regenerated in the cytosol, completing the cycle. 19.2 ATP Synthesis 715 Q Matrix NADH + H + NAD + Glycolysis Dihydroxyacetone phosphate Glycerol 3- phosphate FAD FADH 2 III CH 2 OH CH 2 CHOH O P P – – –– CH 2 OH CH 2 C O – – – – – – O mitochondrial glycerol 3-phosphate dehydrogenase cytosolic glycerol 3-phosphate dehydrogenase FIGURE 19–28 Glycerol 3-phosphate shuttle. This alternative means of moving reducing equivalents from the cytosol to the mitochondrial matrix operates in skeletal muscle and the brain. In the cytosol, dihydroxyacetone phosphate accepts two reducing equivalents from NADH in a reaction catalyzed by cytosolic glycerol 3-phosphate dehydrogenase. An isozyme of glycerol 3-phosphate dehydrogenase bound to the outer face of the inner membrane then transfers two reducing equivalents from glycerol 3-phosphate in the intermembrane space to ubiquinone. Note that this shuttle does not involve membrane transport systems. 8885d_c19_690-750 3/1/04 11:32 AM Page 715 mac76 mac76:385_reb: ■ ATP synthase carries out “rotational catalysis,” in which the flow of protons through F o causes each of three nucleotide-binding sites in F 1 to cycle from (ADP H11001 P i )–bound to ATP-bound to empty conformations. ■ ATP formation on the enzyme requires little energy; the role of the proton-motive force is to push ATP from its binding site on the synthase. ■ The ratio of ATP synthesized per H5007 1 2 H5007O 2 reduced to H 2 O (the P/O ratio) is about 2.5 when elec- trons enter the respiratory chain at Complex I, and 1.5 when electrons enter at CoQ. ■ Energy conserved in a proton gradient can drive solute transport uphill across a membrane. ■ The inner mitochondrial membrane is impermeable to NADH and NAD H11001 , but NADH equivalents are moved from the cytosol to the matrix by either of two shuttles. NADH equivalents moved in by the malate-aspartate shuttle enter the respiratory chain at Complex I and yield a P/O ratio of 2.5; those moved in by the glycerol 3-phosphate shuttle enter at CoQ and give a P/O ratio of 1.5. 19.3 Regulation of Oxidative Phosphorylation Oxidative phosphorylation produces most of the ATP made in aerobic cells. Complete oxidation of a molecule of glucose to CO 2 yields 30 or 32 ATP (Table 19–5). By comparison, glycolysis under anaerobic conditions (lactate fermentation) yields only 2 ATP per glucose. Clearly, the evolution of oxidative phosphorylation pro- vided a tremendous increase in the energy efficiency of catabolism. Complete oxidation to CO 2 of the coenzyme A derivative of palmitate (16:0), which also occurs in the mitochondrial matrix, yields 108 ATP per palmitoyl- CoA (see Table 17–1). A similar calculation can be made for the ATP yield from oxidation of each of the amino acids (Chapter 18). Aerobic oxidative pathways that result in electron transfer to O 2 accompanied by oxida- tive phosphorylation therefore account for the vast majority of the ATP produced in catabolism, so the reg- ulation of ATP production by oxidative phosphorylation to match the cell’s fluctuating needs for ATP is ab- solutely essential. Oxidative Phosphorylation Is Regulated by Cellular Energy Needs The rate of respiration (O 2 consumption) in mitochon- dria is tightly regulated; it is generally limited by the availability of ADP as a substrate for phosphorylation. Dependence of the rate of O 2 consumption on the avail- ability of the P i acceptor ADP (Fig. 19–18b), the ac- ceptor control of respiration, can be remarkable. In some animal tissues, the acceptor control ratio, the ratio of the maximal rate of ADP-induced O 2 consump- tion to the basal rate in the absence of ADP, is at least ten. The intracellular concentration of ADP is one meas- ure of the energy status of cells. Another, related meas- ure is the mass-action ratio of the ATP-ADP system, [ATP]/([ADP][P i ]). Normally this ratio is very high, so the ATP-ADP system is almost fully phosphorylated. When the rate of some energy-requiring process (pro- tein synthesis, for example) increases, the rate of break- down of ATP to ADP and P i increases, lowering the mass-action ratio. With more ADP available for oxida- tive phosphorylation, the rate of respiration increases, causing regeneration of ATP. This continues until the mass-action ratio returns to its normal high level, at which point respiration slows again. The rate of oxida- tion of cellular fuels is regulated with such sensitivity and precision that the [ATP]/([ADP][P i ]) ratio fluctuates only slightly in most tissues, even during extreme vari- ations in energy demand. In short, ATP is formed only as fast as it is used in energy-requiring cellular activities. Chapter 19 Oxidative Phosphorylation and Photophosphorylation716 Process Direct product Final ATP Glycolysis 2 NADH (cytosolic) 3 or 5 * 2 ATP 2 Pyruvate oxidation (two per glucose) 2 NADH (mitochondrial matrix) 5 Acetyl-CoA oxidation in citric acid cycle 6 NADH (mitochondrial matrix) 15 (two per glucose) 2 FADH 2 3 2 ATP or 2 GTP 2 Total yield per glucose 30 or 32 TABLE 19–5 ATP Yield from Complete Oxidation of Glucose * The number depends on which shuttle system transfers reducing equivalents into the mitochondrion. 8885d_c19_690-750 3/1/04 11:32 AM Page 716 mac76 mac76:385_reb: An Inhibitory Protein Prevents ATP Hydrolysis during Ischemia We have already encountered ATP synthase as an ATP- driven proton pump (see Fig. 11–39; Table 11–3), cat- alyzing the reverse of ATP synthesis. When a cell is is- chemic (deprived of oxygen), as in a heart attack or stroke, electron transfer to oxygen ceases, and so does the pumping of protons. The proton-motive force soon collapses. Under these conditions, the ATP synthase could operate in reverse, hydrolyzing ATP to pump pro- tons outward and causing a disastrous drop in ATP lev- els. This is prevented by a small (84 amino acids) pro- tein inhibitor, IF 1 , which simultaneously binds to two ATP synthase molecules, inhibiting their ATPase activ- ity (Fig. 19–29). IF 1 is inhibitory only in its dimeric form, which is favored at pH lower than 6.5. In a cell starved for oxygen, the main source of ATP becomes glycolysis, and the pyruvic or lactic acid thus formed lowers the pH in the cytosol and the mitochondrial matrix. This fa- vors IF 1 dimerization, leading to inhibition of the ATPase activity of ATP synthase, thereby preventing wasteful hydrolysis of ATP. When aerobic metabolism resumes, production of pyruvic acid slows, the pH of the cytosol rises, the IF 1 dimer is destabilized, and the inhibition of ATP synthase is lifted. Uncoupled Mitochondria in Brown Fat Produce Heat There is a remarkable and instructive exception to the general rule that respiration slows when the ATP supply is adequate. Most newborn mammals, including humans, have a type of adipose tissue called brown fat in which fuel oxidation serves not to produce ATP but to gener- ate heat to keep the newborn warm. This specialized adipose tissue is brown because of the presence of large numbers of mitochondria and thus large amounts of cytochromes, whose heme groups are strong absorbers of visible light. The mitochondria of brown fat are like those of other mammalian cells in all respects, except that they have a unique protein in their inner membrane. Thermo- genin, also called the uncoupling protein (Table 19–4), provides a path for protons to return to the matrix without passing through the F o F 1 complex (Fig. 19–30). 19.3 Regulation of Oxidative Phosphorylation 717 FIGURE 19–29 Structure of bovine F 1 -ATPase in a complex with its regulatory protein IF 1 . (Derived from PDB ID 1OHH) Two F 1 molecules are viewed here as in Figure 19–23c. The inhibitor IF 1 (red) binds to the H9251H9252 interface of the subunits in the diphosphate (ADP) conformation (H9251ADP and H9252ADP), freezing the two F 1 complexes and thereby blocking ATP hydrolysis (and synthesis). (Parts of IF 1 that failed to resolve in crystals of F 1 are shown in white outline as they occur in crystals of isolated IF 1 .) This complex is stable only at the low cytosolic pH characteristic of cells that are producing ATP by glycolysis; when aerobic metabolism resumes, the cytosolic pH rises, the inhibitor is destabilized, and ATP synthase becomes active. I II III IV Cyt c Uncoupling protein (thermogenin) ADP + P i ATP H + H + H + F o F 1 Intermembrane space Matrix Heat FIGURE 19–30 Heat generation by uncoupled mitochondria. The un- coupling protein (thermogenin) of brown fat mitochondria, by pro- viding an alternative route for protons to reenter the mitochondrial matrix, causes the energy conserved by proton pumping to be dissi- pated as heat. 8885d_c19_690-750 3/1/04 11:32 AM Page 717 mac76 mac76:385_reb: As a result of this short-circuiting of protons, the en- ergy of oxidation is not conserved by ATP formation but is dissipated as heat, which contributes to maintaining the body temperature of the newborn. Hibernating an- imals also depend on uncoupled mitochondria of brown fat to generate heat during their long dormancy (see Box 17–1). ATP-Producing Pathways Are Coordinately Regulated The major catabolic pathways have interlocking and concerted regulatory mechanisms that allow them to function together in an economical and self-regulating manner to produce ATP and biosynthetic precursors. The relative concentrations of ATP and ADP control not only the rates of electron transfer and oxidative phos- phorylation but also the rates of the citric acid cycle, pyruvate oxidation, and glycolysis (Fig. 19–31). When- ever ATP consumption increases, the rate of electron transfer and oxidative phosphorylation increases. Si- multaneously, the rate of pyruvate oxidation via the cit- ric acid cycle increases, increasing the flow of electrons into the respiratory chain. These events can in turn evoke an increase in the rate of glycolysis, increasing the rate of pyruvate formation. When conversion of ADP to ATP lowers the ADP concentration, acceptor control slows electron transfer and thus oxidative phosphoryla- tion. Glycolysis and the citric acid cycle are also slowed, because ATP is an allosteric inhibitor of the glycolytic enzyme phosphofructokinase-1 (see Fig. 15–18) and of pyruvate dehydrogenase (see Fig. 16–18). Phosphofructokinase-1 is also inhibited by citrate, the first intermediate of the citric acid cycle. When the cycle is “idling,” citrate accumulates within mitochon- dria, then spills into the cytosol. When the concentra- tions of both ATP and citrate rise, they produce a con- certed allosteric inhibition of phosphofructokinase-1 that is greater than the sum of their individual effects, slowing glycolysis. SUMMARY 19.3 Regulation of Oxidative Phosphorylation ■ Oxidative phosphorylation is regulated by cellular energy demands. The intracellular [ADP] and the mass-action ratio [ATP]/([ADP][P i ]) are measures of a cell’s energy status. Chapter 19 Oxidative Phosphorylation and Photophosphorylation718 Glucose P i Glycolysis Glucose 6-phosphate Fructose 1,6-bisphosphate AMP, ADP ATP, citrate hexokinase phosphofructokinase-1 multistep Phosphoenolpyruvate ADP ATP, NADH Pyruvate AMP, ADP, NAD H11001 ATP, NADH Acetyl-CoA ADP ATP, NADH Citrate ADP ATP pyruvate kinase H9251-Ketoglutarate ATP, NADH Succinyl-CoA Oxaloacetate pyruvate dehydrogenase complex citrate synthase ADP, P i Respiratory chain H 2 O ADP H11001 P i ATP O 2 1 2 NADH NAD H11001 isocitrate dehydrogenase multistep -ketoglutarate dehydrogenase H9251 Citric acid cycle Oxidative phosphory- lation FIGURE 19–31 Regulation of the ATP-producing pathways. This di- agram shows the interlocking regulation of glycolysis, pyruvate oxi- dation, the citric acid cycle, and oxidative phosphorylation by the rel- ative concentrations of ATP, ADP, and AMP, and by NADH. High [ATP] (or low [ADP] and [AMP]) produces low rates of glycolysis, pyruvate oxidation, acetate oxidation via the citric acid cycle, and oxidative phosphorylation. All four pathways are accelerated when the use of ATP and the formation of ADP, AMP, and P i increase. The interlock- ing of glycolysis and the citric acid cycle by citrate, which inhibits glycolysis, supplements the action of the adenine nucleotide system. In addition, increased levels of NADH and acetyl-CoA also inhibit the oxidation of pyruvate to acetyl-CoA, and a high [NADH]/[NAD H11001 ] ra- tio inhibits the dehydrogenase reactions of the citric acid cycle (see Fig. 16–18). 8885d_c19_690-750 3/1/04 11:32 AM Page 718 mac76 mac76:385_reb: ■ In ischemic (oxygen-deprived) cells, a protein inhibitor blocks ATP hydrolysis by the ATP synthase operating in reverse, preventing a drastic drop in [ATP]. ■ In brown fat, which is specialized for the production of metabolic heat, electron transfer is uncoupled from ATP synthesis and the energy of fatty acid oxidation is dissipated as heat. ■ ATP and ADP concentrations set the rate of electron transfer through the respiratory chain via a series of interlocking controls on respiration, glycolysis, and the citric acid cycle. 19.4 Mitochondrial Genes: Their Origin and the Effects of Mutations Mitochondria contain their own genome, a circular, double-stranded DNA molecule. Each of the hundreds or thousands of mitochondria in a typical cell has about five copies of this genome. The human mitochondrial chromosome (Fig. 19–32) contains 37 genes (16,569 bp), including 13 that encode subunits of proteins of the respiratory chain (Table 19–6); the remaining genes code for rRNA and tRNA molecules essential to the protein-synthesizing machinery of mitochondria. About 900 different mitochondrial proteins are encoded by nu- clear genes, synthesized on cytoplasmic ribosomes, then imported and assembled within the mitochondria (Chapter 27). Mutations in Mitochondrial Genes Cause Human Disease A growing number of human diseases can be at- tributed to mutations in mitochondrial genes. Many of these diseases, those known as the mitochon- drial encephalomyopathies, affect primarily the brain and skeletal muscle (both heavily dependent on an abundant supply of ATP). These diseases are invariably inherited from the mother, because a developing em- bryo derives all its mitochondria from the mother’s egg. The rare disease Leber’s hereditary optic neuropa- thy (LHON) affects the central nervous system, in- cluding the optic nerves, causing bilateral loss of vision in early adulthood. A single base change in the mito- chondrial gene ND4 (Fig. 19–32a) changes an Arg residue to a His residue in a polypeptide of Complex I, and the result is mitochondria partially defective in elec- tron transfer from NADH to ubiquinone. Although these mitochondria can produce some ATP by electron trans- fer from succinate, they apparently cannot supply suf- ficient ATP to support the very active metabolism of neurons. One result is damage to the optic nerve, lead- ing to blindness. A single base change in the mitochon- drial gene for cytochrome b, a component of Complex III, also produces LHON, demonstrating that the pathol- ogy results from a general reduction of mitochondrial function, not specifically from a defect in electron trans- fer through Complex I. Myoclonic epilepsy and ragged-red fiber dis- ease (MERRF) is caused by a mutation in the mito- chondrial gene that encodes a transfer RNA specific for lycine (lysyl-tRNA). This disease, characterized by un- controllable muscular jerking, apparently results from defective production of several of the proteins whose synthesis involves mitochondrial tRNAs. Skeletal mus- cle fibers of individuals with MERRF have abnormally shaped mitochondria that sometimes contain paracrys- talline structures (Fig. 19–32b). Mutations in the mito- chondrial lysyl-tRNA gene are also one of the causes of adult-onset (type II) diabetes mellitus. Other mutations in mitochondrial genes are believed to be responsible for the progressive muscular weakness that character- izes mitochondrial myopathy and for enlargement and deterioration of the heart muscle in hypertrophic cardio- myopathy. According to one hypothesis on the progres- sive changes that accompany aging, the accumulation of mutations in mitochondrial DNA during a lifetime of exposure to DNA-damaging agents such as H11080 O 2 H11002 (see below) results in mitochondria that cannot supply suf- ficient ATP for normal cellular function. Mitochondrial disease can also result from mutations in any of the 900 nuclear genes that encode mitochondrial proteins. ■ 19.4 Mitochondrial Genes: Their Origin and the Effects of Mutations 719 Number Number of subunits encoded Complex of subunits by mitochondrial DNA I NADH dehydrogenase H1102243 7 II Succinate dehydrogenase 4 0 III Ubiquinone:cytochrome c oxidoreductase 11 1 IV Cytochrome oxidase 13 3 V ATP synthase 8 2 TABLE 19–6 Respiratory Proteins Encoded by Mitochondrial Genes in Humans 8885d_c19_690-750 3/1/04 11:32 AM Page 719 mac76 mac76:385_reb: Chapter 19 Oxidative Phosphorylation and Photophosphorylation720 Menaquinone n H11005 7–9 CH 3 (CH 2 CH CH 2 ) n C CH 3 H O O (b) Cytosol (N side) Bacterial inner (plasma) membrane Periplasmic space (P side) Cyt b Cyt o Cu 2Fe-2S FAD NADH, succinate, or glycerol 4Fe-4S Q O 2 H + H + (a) FIGURE 19–33 Bacterial respiratory chain. (a) Shown here are the respiratory carriers of the inner membrane of E. coli. Eubacteria con- tain a minimal form of Complex I, containing all the prosthetic groups normally associated with the mitochondrial complex but only 14 polypeptides. This plasma membrane complex transfers electrons from NADH to ubiquinone or to (b) menaquinone, the bacterial equivalent of ubiquinone, while pumping protons outward and creating an elec- trochemical potential that drives ATP synthesis. FIGURE 19–32 Mitochondrial genes and mutations. (a) Map of human mitochondrial DNA, showing the genes that encode proteins of Complex I, the NADH dehydroge- nase (ND1 to ND6); the cytochrome b of Complex III (Cyt b); the subunits of cytochrome oxidase (Complex IV) (COI to COIII); and two subunits of ATP synthase (ATPase6 and ATPase8). The colors of the genes correspond to those of the complexes shown in Figure 19–7. Also included here are the genes for ribosomal RNAs (rRNA) and for a number of mitochondrion-specific transfer RNAs; tRNA specificity is indicated by the one-letter codes for amino acids. Arrows indicate the positions of mutations that cause Leber’s heredi- tary optic neuropathy (LHON) and myoclonic epilepsy and ragged- red fiber disease (MERRF). Numbers in parentheses indicate the posi- tion of the altered nucleotides (nucleotide 1 is at the top of the circle and numbering proceeds counterclockwise). (b) Electron micrograph of an abnormal mitochondrion from the muscle of an individual with MERRF, showing the paracrystalline protein inclusions sometimes pres- ent in the mutant mitochondria. 0/16,569 12S rRNA 16S rRNA ND1 ND2 COI COII COIII ATPase6 ND3 ND4L ND4 ND5 ND6 Cyt b F V L I M W A N C Y S D K G (a) R H L S E P T LHON (15,257) Q LHON (3,460) LHON (4,160) LHON (11,778) MERRF (8,344) ATPase8 Complex I Complex III Complex IV ATP synthase Transfer RNA Ribosomal RNA Control region of DNA (b) 8885d_c19_690-750 3/1/04 11:32 AM Page 720 mac76 mac76:385_reb: Mitochondria Evolved from Endosymbiotic Bacteria The existence of mitochondrial DNA, ribosomes, and tRNAs supports the hypothesis of the endosymbiotic origin of mitochondria (see Fig. 1–36), which holds that the first organisms capable of aerobic metabolism, in- cluding respiration-linked ATP production, were prokar- yotes. Primitive eukaryotes that lived anaerobically (by fermentation) acquired the ability to carry out oxidative phosphorylation when they established a symbiotic re- lationship with bacteria living in their cytosol. After much evolution and the movement of many bacterial genes into the nucleus of the “host” eukaryote, the endosymbiotic bacteria eventually became mitochondria. This hypothesis presumes that early free-living prokaryotes had the enzymatic machinery for oxidative phosphorylation and predicts that their modern prokaryotic descendants must have respiratory chains closely similar to those of modern eukaryotes. They do. Aerobic bacteria carry out NAD-linked electron trans- fer from substrates to O 2 , coupled to the phosphoryla- tion of cytosolic ADP. The dehydrogenases are located in the bacterial cytosol and the respiratory chain in the plasma membrane. The electron carriers are similar to some mitochondrial electron carriers (Fig. 19–33). They translocate protons outward across the plasma mem- brane as electrons are transferred to O 2 . Bacteria such as Escherichia coli have F o F 1 complexes in their plasma membranes; the F 1 portion protrudes into the cytosol and catalyzes ATP synthesis from ADP and P i as protons flow back into the cell through the proton channel of F o . The respiration-linked extrusion of protons across the bacterial plasma membrane also provides the driv- ing force for other processes. Certain bacterial trans- port systems bring about uptake of extracellular nutri- ents (lactose, for example) against a concentration gradient, in symport with protons (see Fig. 11–42). And the rotary motion of bacterial flagella is provided by “proton turbines,” molecular rotary motors driven not by ATP but directly by the transmembrane electro- chemical potential generated by respiration-linked pro- ton pumping (Fig. 19–34). It appears likely that the chemiosmotic mechanism evolved early, before the emergence of eukaryotes. SUMMARY 19.4 Mitochondrial Genes: Their Origin and the Effects of Mutations ■ A small proportion of human mitochondrial proteins (13 proteins) are encoded in the mitochondrial genome and synthesized within mitochondria. About 900 mitochondrial proteins are encoded in nuclear genes and imported into mitochondria after their synthesis. ■ Mutations in the genes that encode components of the respiratory chain, whether in the mitochondrial genes or in the nuclear genes that encode mitochondrial proteins, cause a variety of human diseases, which often affect muscle and brain most severely. ■ Mitochondria most likely arose from aerobic prokaryotes that entered into an endosymbiotic relationship with ancestral eukaryotes. 19.5 The Role of Mitochondria in Apoptosis and Oxidative Stress Besides their central role in ATP synthesis, mitochon- dria also participate in processes associated with cellu- lar damage and death. Apoptosis is a controlled process by which cells die for the good of the organism, while the organism conserves the molecular components (amino acids, nucleotides, and so forth) of the dead cells. Apoptosis may be triggered by an external signal, acting at a receptor in the plasma membrane, or by in- ternal events such as a viral infection. When a cell re- ceives a signal for apoptosis, one consequence is an in- crease in the permeability of the outer mitochondrial membrane, allowing escape of the cytochrome c nor- mally confined in the intermembrane space (see Fig. 12–50). The released cytochrome c activates one of the proteolytic enzymes (caspase 9) responsible for protein degradation during apoptosis. This is a dramatic case of 19.5 The Role of Mitochondria in Apoptosis and Oxidative Stress 721 Flagellum Outer membrane Inner (plasma) membrane Rotary motor Peptidoglycan and periplasmic space Electron-transfer chain H H11001 H H11001 FIGURE 19–34 Rotation of bacterial flagella by proton-motive force. The shaft and rings at the base of the flagellum make up a rotary mo- tor that has been called a “proton turbine.” Protons ejected by elec- tron transfer flow back into the cell through the turbine, causing ro- tation of the shaft of the flagellum. This motion differs fundamentally from the motion of muscle and of eukaryotic flagella and cilia, for which ATP hydrolysis is the energy source. 8885d_c19_690-750 3/1/04 11:32 AM Page 721 mac76 mac76:385_reb: one protein (cytochrome c) playing two very different roles in the cell. Mitochondria are also involved in the cell’s response to oxidative stress. As we have seen, several steps in the path of oxygen reduction in mitochondria have the potential to produce highly reactive free radicals that can damage cells. The passage of electrons from QH 2 to cytochrome b L through Complex III, and passage of elec- trons from Complex I to QH 2 , involve the radical H11080 Q H11002 as an intermediate. The H11080 Q H11002 can, with a low probability, pass an electron to O 2 in the reaction O 2 H11001 e H11002 On H11080 O 2 H11002 The superoxide free radical thus generated, H11080 O 2 H11002 , is very reactive and can damage enzymes, membrane lipids, and nucleic acids. Antimycin A, an inhibitor of Complex III, may act by occupying the Q N site (Fig. 19–11), thus blocking the Q cycle and prolonging the binding of H11080 Q H11002 to the Q P site; this would increase the likelihood of su- peroxide radical formation and cellular damage. From 0.1% to as much as 4% of the O 2 used by actively respir- ing mitochondria forms H11080 O 2 H11002 —more than enough to have lethal effects on a cell unless the free radical is quickly disposed of. To prevent oxidative damage by H11080 O 2 H11002 , cells have sev- eral forms of the enzyme superoxide dismutase, which catalyzes the reaction 2 H11080 O 2 H11002 H11001 2H H11001 88n H 2 O 2 H11001 O 2 The hydrogen peroxide (H 2 O 2 ) generated by this reac- tion is rendered harmless by the action of glutathione peroxidase (Fig. 19–35). This enzyme is remarkable for the presence of a selenocysteine residue (see Fig. 3–8a), in which an atom of selenium replaces the sulfur atom normally present in the thiol of the side chain. The se- lenol group (OSeH) is more acidic than the thiol (OSH); its pKa is about 5, so at neutral pH, the selenocysteine side chain is essentially fully ionized (OCH 2 Se H11002 ). Gluta- thione reductase recycles oxidized glutathione to its re- duced form, using electrons from the NADPH formed by nicotinamide nucleotide transhydrogenase or by the pentose phosphate pathway (see Fig. 14–20). Reduced glutathione also serves in keeping protein sulfhydryl groups in their reduced state, preventing some of the deleterious effects of oxidative stress (Fig. 19–35). SUMMARY 19.5 The Role of Mitochondria in Apoptosis and Oxidative Stress ■ Mitochondrial cytochrome c, released into the cytosol, participates in activation of one of the proteases (caspase 9) involved in apoptosis. ■ Reactive oxygen species produced in mitochondria are inactivated by a set of protective enzymes, including superoxide dismutase and glutathione peroxidase. Chapter 19 Oxidative Phosphorylation and Photophosphorylation722 Nicotinamide nucleotide transhydrogenase NADH NADPH GSSG H 2 O 2 H 2 O 2 GSH S S 2 GSH Enz inactive active oxidative stress protein thiol reduction GSSG NADP + glutathione reductase glutathione peroxidase superoxide dismutase NAD + O 2 O 2 NAD + Inner mitochondrial membrane Q III IV I Cyt c . – SH SH FIGURE 19–35 Mitochondrial production and disposal of super- oxide. Superoxide radical, ?O 2 H11002 , is formed in side reactions at Complexes I and III, as the partially reduced ubiquinone radical (?Q H11002 ) donates an electron to O 2 . The reactions shown in blue defend the cell against the damaging effects of superoxide. Reduced glutathione (GSH; see Fig. 22–27) donates electrons for the reduction of hydrogen peroxide (H 2 O 2 ) and of oxidized Cys residues (OSOSO) in proteins, and GSH is regenerated from the oxidized form (GSSG) by reduction with NADPH. 8885d_c19_690-750 3/1/04 11:32 AM Page 722 mac76 mac76:385_reb: FIGURE 19–37 The light reactions of photosynthesis generate energy- rich NADPH and ATP at the expense of solar energy. These products are used in the carbon- assimilation reactions, which occur in light or darkness, to reduce CO 2 to form trioses and more complex compounds (such as glucose) derived from trioses. Light reactions NADP + Carbon-assimilation reactions ADP + P i NADPH ATP H 2 O O 2 CO 2 Carbohydrate Photosynthetic cells CarbohydrateO 2 CO 2 H 2 O Heterotrophic cells FIGURE 19–36 Solar energy as the ultimate source of all biological energy. Photosynthetic organisms use the energy of sunlight to manufacture glucose and other organic products, which hetero- trophic cells use as energy and carbon sources. PHOTOSYNTHESIS: HARVESTING LIGHT ENERGY We now turn to another reaction sequence in which the flow of electrons is coupled to the synthesis of ATP: light-driven phosphorylation. The capture of solar en- ergy by photosynthetic organisms and its conversion to the chemical energy of reduced organic compounds is the ultimate source of nearly all biological energy. Pho- tosynthetic and heterotrophic organisms live in a bal- anced steady state in the biosphere (Fig. 19–36). Pho- tosynthetic organisms trap solar energy and form ATP and NADPH, which they use as energy sources to make carbohydrates and other organic compounds from CO 2 and H 2 O; simultaneously, they release O 2 into the at- mosphere. Aerobic heterotrophs (humans, for example, as well as plants during dark periods) use the O 2 so formed to degrade the energy-rich organic products of photosynthesis to CO 2 and H 2 O, generating ATP. The CO 2 returns to the atmosphere, to be used again by pho- tosynthetic organisms. Solar energy thus provides the driving force for the continuous cycling of CO 2 and O 2 through the biosphere and provides the reduced substrates—fuels, such as glucose—on which nonpho- tosynthetic organisms depend. Photosynthesis occurs in a variety of bacteria and in unicellular eukaryotes (algae) as well as in vascular plants. Although the process in these organisms differs in detail, the underlying mechanisms are remarkably similar, and much of our understanding of photosyn- thesis in vascular plants is derived from studies of sim- pler organisms. The overall equation for photosynthesis in vascular plants describes an oxidation-reduction re- action in which H 2 O donates electrons (as hydrogen) for the reduction of CO 2 to carbohydrate (CH 2 O): light CO 2 H11001 H 2 O 888n O 2 H11001 (CH 2 O) 19.6 General Features of Photophosphorylation Unlike NADH (the major electron donor in oxidative phosphorylation), H 2 O is a poor donor of electrons; its standard reduction potential is 0.816 V, compared with H110020.320 V for NADH. Photophosphorylation differs from oxidative phosphorylation in requiring the input of en- ergy in the form of light to create a good electron donor and a good electron acceptor. In photophosphorylation, electrons flow through a series of membrane-bound car- riers including cytochromes, quinones, and iron-sulfur proteins, while protons are pumped across a membrane to create an electrochemical potential. Electron trans- fer and proton pumping are catalyzed by membrane complexes homologous in structure and function to Complex III of mitochondria. The electrochemical po- tential they produce is the driving force for ATP syn- thesis from ADP and P i , catalyzed by a membrane-bound ATP synthase complex closely similar to that of oxida- tive phosphorylation. Photosynthesis in plants encompasses two pro- cesses: the light-dependent reactions, or light re- actions, which occur only when plants are illuminated, and the carbon-assimilation reactions (or carbon- fixation reactions), sometimes misleadingly called the dark reactions, which are driven by products of the light reactions (Fig. 19–37). In the light reactions, chlorophyll and other pigments of photosynthetic cells absorb light energy and conserve it as ATP and NADPH; simultane- ously, O 2 is evolved. In the carbon-assimilation reac- tions, ATP and NADPH are used to reduce CO 2 to form triose phosphates, starch, and sucrose, and other prod- ucts derived from them. In this chapter we are con- cerned only with the light-dependent reactions that lead to the synthesis of ATP and NADPH. The reduction of CO 2 is described in Chapter 20. 19.6 General Features of Photophosphorylation 723 8885d_c19_690-750 3/1/04 11:32 AM Page 723 mac76 mac76:385_reb: Photosynthesis in Plants Takes Place in Chloroplasts In photosynthetic eukaryotic cells, both the light-de- pendent and the carbon-assimilation reactions take place in the chloroplasts (Fig. 19–38), membrane- bounded intracellular organelles that are variable in shape and generally a few micrometers in diameter. Like mitochondria, they are surrounded by two membranes, an outer membrane that is permeable to small molecules and ions, and an inner membrane that encloses the in- ternal compartment. This compartment contains many flattened, membrane-surrounded vesicles or sacs, the thylakoids, usually arranged in stacks called grana (Fig. 19–38b). Embedded in the thylakoid membranes (commonly called lamellae) are the photosynthetic pigments and the enzyme complexes that carry out the light reactions and ATP synthesis. The stroma (the aqueous phase enclosed by the inner membrane) con- tains most of the enzymes required for the carbon- assimilation reactions. Light Drives Electron Flow in Chloroplasts In 1937 Robert Hill found that when leaf extracts con- taining chloroplasts were illuminated, they (1) evolved O 2 and (2) reduced a nonbiological electron acceptor added to the medium, according to the Hill reaction: light 2H 2 O H11001 2A 888n 2AH 2 H11001 O 2 where A is the artificial electron acceptor, or Hill reagent. One Hill reagent, the dye 2,6-dichlorophenol- indophenol, is blue when oxidized (A) and colorless when reduced (AH 2 ), making the reaction easy to fol- low. When a leaf extract supplemented with the dye was illuminated, the blue dye became colorless and O 2 was evolved. In the dark, neither O 2 evolution nor dye re- duction took place. This was the first evidence that absorbed light energy causes electrons to flow from H 2 O to an electron acceptor. Moreover, Hill found that CO 2 was neither required nor reduced to a stable form un- der these conditions; O 2 evolution could be dissociated from CO 2 reduction. Several years later Severo Ochoa showed that NADP H11001 is the biological electron acceptor in chloroplasts, according to the equation light 2H 2 O H11001 2NADP H11001 888n 2NADPH H11001 2H H11001 H11001 O 2 To understand this photochemical process, we must first consider the more general topic of the effects of light absorption on molecular structure. SUMMARY 19.6 General Features of Photophosphorylation ■ The light reactions of photosynthesis are those directly dependent on the absorption of light; the resulting photochemistry takes electrons from H 2 O and drives them through a series of membrane-bound carriers, producing NADPH and ATP. ■ The carbon-assimilation reactions of photosynthesis reduce CO 2 with electrons from NADPH and energy from ATP. OH Reduced form (colorless) Oxidized form (blue) Dichlorophenolindophenol OH O OH N Cl ClCl Cl NH Chapter 19 Oxidative Phosphorylation and Photophosphorylation724 Grana (thylakoids) Stroma (b) Outer membrane Inner membrane Thylakoids (a) FIGURE 19–38 Chloroplast. (a) Schematic diagram. (b) Electron mi- crograph at high magnification showing grana, stacks of thylakoid membranes. 8885d_c19_690-750 3/1/04 11:32 AM Page 724 mac76 mac76:385_reb: 19.7 Light Absorption Visible light is electromagnetic radiation of wavelengths 400 to 700 nm, a small part of the electromagnetic spec- trum (Fig. 19–39), ranging from violet to red. The en- ergy of a single photon (a quantum of light) is greater at the violet end of the spectrum than at the red end; shorter wavelength (and higher frequency) corresponds to higher energy. The energy, E, in a “mole” of photons (1 einstein, or 6 H11003 10 23 photons) of visible light is 170 to 300 kJ, as given by the Planck equation: E H11005 hH9263 where h is Planck’s constant (6.626 H11003 10 H1100234 J H11554 s) and H9263 is the wavelength. These amounts of energy are almost an order of magnitude greater than the 30 to 50 kJ re- quired to synthesize a mole of ATP from ADP and P i . When a photon is absorbed, an electron in the ab- sorbing molecule (chromophore) is lifted to a higher energy level. This is an all-or-nothing event; to be ab- sorbed, the photon must contain a quantity of energy (a quantum) that exactly matches the energy of the elec- tronic transition. A molecule that has absorbed a photon is in an excited state, which is generally unstable. An electron lifted into a higher-energy orbital usually re- turns rapidly to its normal lower-energy orbital; the ex- cited molecule decays to the stable ground state, giving up the absorbed quantum as light or heat or us- ing it to do chemical work. Light emission accompany- ing decay of excited molecules (called fluorescence) is always at a longer wavelength (lower energy) than that of the absorbed light (see Box 12–2). An alterna- tive mode of decay important in photosynthesis involves direct transfer of excitation energy from an excited mol- ecule to a neighboring molecule. Just as the photon is a quantum of light energy, so the exciton is a quantum of energy passed from an excited molecule to another molecule in a process called exciton transfer. Chlorophylls Absorb Light Energy for Photosynthesis The most important light-absorbing pigments in the thy- lakoid membranes are the chlorophylls, green pigments with polycyclic, planar structures resembling the proto- porphyrin of hemoglobin (see Fig. 5–1), except that Mg 2H11001 , not Fe 2H11001 , occupies the central position (Fig. 19–40). The four inward-oriented nitrogen atoms of chlorophyll are coordinated with the Mg 2H11001 . All chlorophylls have a long phytol side chain, esterified to a carboxyl-group sub- stituent in ring IV, and chlorophylls also have a fifth five- membered ring not present in heme. The heterocyclic five-ring system that surrounds the Mg 2H11001 has an extended polyene structure, with al- ternating single and double bonds. Such polyenes char- acteristically show strong absorption in the visible re- gion of the spectrum (Fig. 19–41); the chlorophylls have unusually high molar extinction coefficients (see Box 3–1) and are therefore particularly well-suited for ab- sorbing visible light during photosynthesis. Chloroplasts always contain both chlorophyll a and chlorophyll b (Fig. 19–40a). Although both are green, their absorption spectra are sufficiently different (Fig. 19–41) that they complement each other’s range of light absorption in the visible region. Most plants con- tain about twice as much chlorophyll a as chlorophyll b. The pigments in algae and photosynthetic bacteria include chlorophylls that differ only slightly from the plant pigments. Chlorophyll is always associated with specific binding proteins, forming light-harvesting com- plexes (LHCs) in which chlorophyll molecules are fixed in relation to each other, to other protein com- plexes, and to the membrane. The detailed structure of one light-harvesting complex is known from x-ray crystallography (Fig. 19–42). It contains seven mole- cules of chlorophyll a, five of chlorophyll b, and two of the accessory pigment lutein (see below). 19.7 Light Absorption 725 380 CyanBlueViolet Green Yellow Orange Red Wavelength (nm) Energy (kJ/einstein) 300 430 500 560 600 650 750 240 200 170 Wavelength Type of radiation Gamma rays Visible light X rays UV Infrared Microwaves Radio waves Thousands of metersH110211 millimeter 1 meterH110211 nm 100 nm FIGURE 19–39 Electromagnetic radiation. The spectrum of electromagnetic radiation, and the energy of photons in the visible range of the spectrum. One einstein is 6 H11003 10 23 photons. 8885d_c19_725 3/1/04 1:59 PM Page 725 mac76 mac76:385_reb: Cyanobacteria and red algae employ phycobilins such as phycoerythrobilin and phycocyanobilin (Fig. 19–40b) as their light-harvesting pigments. These open- chain tetrapyrroles have the extended polyene system found in chlorophylls, but not their cyclic structure or central Mg 2H11001 . Phycobilins are covalently linked to spe- cific binding proteins, forming phycobiliproteins, which associate in highly ordered complexes called phy- cobilisomes (Fig. 19–43) that constitute the primary light-harvesting structures in these microorganisms. Chapter 19 Oxidative Phosphorylation and Photophosphorylation726 A CH 2 N Mg H O C III IIIIV G D B M 0 ; ; H H CH 2 CH 3 CH 2 CH 3 CH CH 3 N NN D CH 3 O CH 3 O CH 2 B D O C O CH 2 CH 3 CH 3 CH 3 CH 3 G CH 3 phytol side chain b-Carotene O C CH 3 in bacteriochlorophyll CHO in chlorophyll b Saturated bond in bacteriochlorophyll A GJ CH 3 CH 3 CH 3 H 3 CCH 3 CH 3 CH 3 CH 3 CH 3 CH 3 Phycoerythrobilin A A A A COO H11002 N CH 3 CH 3 CH 2 COO H11002 CH 2 CH 2 CH 3 CH 2 N H N H CH 3 CH O A CH 3 N H CH 3 CH O Unsaturated bond in phycocyanobilin in phycocyanobilinB CH 2 CH 3 Chlorophyll a G G (a) (b) (c) Lutein (xanthophyll) CH OH 3 HC 3 HC 3 H 3 C CH 3 CH 3 CH 3 CHHO 3 CH 3 CH 3 (d) FIGURE 19–40 Primary and secondary photopigments. (a) Chloro- phylls a and b and bacteriochlorophyll are the primary gatherers of light energy. (b) Phycoerythrobilin and phycocyanobilin (phycobilins) are the antenna pigments in cyanobacteria and red algae. (c) H9252- Carotene (a carotenoid) and (d) lutein (a xanthophyll) are accessory pigments in plants. The areas shaded pink are the conjugated systems (alternating single and double bonds) that largely account for the ab- sorption of visible light. 8885d_c19_690-750 3/1/04 11:32 AM Page 726 mac76 mac76:385_reb: 19.7 Light Absorption 727 Absorption 300 Sunlight reaching the earth Wavelength (nm) 400 500 600 700 Chlorophyll b -CaroteneH9252 Phycocyanin Chlorophyll a Phycoerythrin Lutein 800 FIGURE 19–41 Absorption of visible light by photopigments. Plants are green because their pigments absorb light from the red and blue regions of the spectrum, leaving primarily green light to be reflected or transmitted. Compare the absorption spectra of the pigments with the spectrum of sunlight reaching the earth’s surface; the combination of chlorophylls (a and b) and accessory pigments enables plants to harvest most of the energy available in sunlight. The relative amounts of chlorophylls and accessory pigments are characteristic of a particular plant species. Variation in the proportions of these pigments is responsible for the range of colors of photosyn- thetic organisms, from the deep blue-green of spruce needles, to the greener green of maple leaves, to the red, brown, or purple color of some species of multicellular algae and the leaves of some foliage plants favored by gardeners. FIGURE 19–42 A light-harvesting complex, LHCII. The functional unit is an LHC trimer, with 36 chlorophyll and 6 lutein molecules. Shown here is a monomer, viewed in the plane of the membrane, with its three transmembrane H9251-helical segments, seven chlorophyll a molecules (green), five chlorophyll b molecules (red), and two mole- cules of the accessory pigment lutein (yellow), which form an internal cross-brace. FIGURE 19–43 A phycobilisome. In these highly structured assem- blies found in cyanobacteria and red algae, phycobilin pigments bound to specific proteins form complexes called phycoerythrin (PE), phycocyanin (PC), and allophycocyanin (AP). The energy of photons absorbed by PE or PC is conveyed through AP (a phycocyanobilin- binding protein) to chlorophyll a of the reaction center by exciton transfer, a process discussed in the text. PE PE PC 550–650 nm PC AP AP Thylakoid membrane Exciton transfer Chlorophyll a reaction center 480–570 nm Light 8885d_c19_690-750 3/1/04 11:32 AM Page 727 mac76 mac76:385_reb: Accessory Pigments Extend the Range of Light Absorption In addition to chlorophylls, thylakoid membranes con- tain secondary light-absorbing pigments, or accessory pigments, called carotenoids. Carotenoids may be yel- low, red, or purple. The most important are H9252-carotene, which is a red-orange isoprenoid, and the yellow carotenoid lutein (Fig. 19–40c, d). The carotenoid pig- ments absorb light at wavelengths not absorbed by the chlorophylls (Fig. 19–41) and thus are supplementary light receptors. Experimental determination of the effectiveness of light of different colors in promoting photosynthesis yields an action spectrum (Fig. 19–44), often useful in identifying the pigment primarily responsible for a bi- ological effect of light. By capturing light in a region of the spectrum not used by other organisms, a photosyn- thetic organism can claim a unique ecological niche. For example, the phycobilins in red algae and cyanobacte- ria absorb light in the range 520 to 630 nm (Fig. 19–41), allowing them to occupy niches where light of lower or higher wavelength has been filtered out by the pigments of other organisms living in the water above them, or by the water itself. Chlorophyll Funnels the Absorbed Energy to Reaction Centers by Exciton Transfer The light-absorbing pigments of thylakoid or bacterial membranes are arranged in functional arrays called photosystems. In spinach chloroplasts, for example, each photosystem contains about 200 chlorophyll and 50 carotenoid molecules. All the pigment molecules in a photosystem can absorb photons, but only a few chloro- phyll molecules associated with the photochemical re- action center are specialized to transduce light into chemical energy. The other pigment molecules in a photosystem are called light-harvesting or antenna molecules. They absorb light energy and transmit it rapidly and efficiently to the reaction center (Fig. 19–45). The chlorophyll molecules in light-harvesting com- plexes have light-absorption properties that are subtly different from those of free chlorophyll. When isolated chlorophyll molecules in vitro are excited by light, the absorbed energy is quickly released as fluorescence and heat, but when chlorophyll in intact leaves is excited by visible light (Fig. 19–46, step 1 ), very little fluores- cence is observed. Instead, the excited antenna chloro- phyll transfers energy directly to a neighboring chloro- phyll molecule, which becomes excited as the first molecule returns to its ground state (step 2 ). This transfer of energy, exciton transfer, extends to a third, fourth, or subsequent neighbor, until one of a special pair of chlorophyll a molecules at the photochemical re- action center is excited (step 3 ). In this excited chloro- phyll molecule, an electron is promoted to a higher- energy orbital. This electron then passes to a nearby electron acceptor that is part of the electron-transfer chain, leaving the reaction-center chlorophyll with a Chapter 19 Oxidative Phosphorylation and Photophosphorylation728 (a) 400 20 Relative rate of photosynthesis Wavelength (nm) 40 60 80 100 0 500 600 700 (b) FIGURE 19–44 Two ways to determine the action spectrum for pho- tosynthesis. (a) Results of a classic experiment performed by T. W. En- glemann in 1882 to determine the wavelength of light that is most ef- fective in supporting photosynthesis. Englemann placed cells of a filamentous photosynthetic alga on a microscope slide and illuminated them with light from a prism, so that one part of the filament received mainly blue light, another part yellow, another red. To determine which algal cells carried out photosynthesis most actively, Englemann also placed on the microscope slide bacteria known to migrate toward re- gions of high O 2 concentration. After a period of illumination, the dis- tribution of bacteria showed highest O 2 levels (produced by photo- synthesis) in the regions illuminated with violet and red light. (b) Results of a similar experiment that used modern techniques (an oxygen electrode) for the measurement of O 2 production. An ac- tion spectrum (as shown here) describes the relative rate of photo- synthesis for illumination with a constant number of photons of dif- ferent wavelengths. An action spectrum is useful because, by comparison with absorption spectra (such as those in Fig. 19–41), it suggests which pigments can channel energy into photosynthesis. 8885d_c19_690-750 3/1/04 11:32 AM Page 728 mac76 mac76:385_reb: 19.7 Light Absorption 729 These molecules absorb light energy, transferring it between molecules until it reaches the reaction center. Antenna chlorophylls, bound to protein Carotenoids, other accessory pigments Light Reaction center Photochemical reaction here converts the energy of a photon into a separation of charge, initiating electron flow. FIGURE 19–45 Organization of photosystems in the thylakoid mem- brane. Photosystems are tightly packed in the thylakoid membrane, with several hundred antenna chlorophylls and accessory pigments surrounding a photoreaction center. Absorption of a photon by any of the antenna chlorophylls leads to excitation of the reaction center by exciton transfer (black arrow). Also embedded in the thylakoid mem- brane are the cytochrome b 6 f complex and ATP synthase (see Fig. 19–52). 1 2 3 4 5 The absorption of a photon has caused separation of charge in the reaction center. Antenna molecules Reaction-center chlorophyll * * * Electron acceptor Electron donor – + – + Light The electron hole in the reaction center is filled by an electron from an electron donor. The excited reaction- center chlorophyll passes an electron to an electron acceptor. This energy is transferred to a reaction-center chlorophyll, exciting it. The excited antenna molecule passes energy to a neighboring chlorophyll molecule (resonance energy transfer), exciting it. Light excites an antenna molecule (chlorophyll or accessory pigment), raising an electron to a higher energy level. FIGURE 19–46 Exciton and electron transfer. This generalized scheme shows conversion of the energy of an absorbed photon into separation of charges at the reaction center. The steps are further de- scribed in the text. Note that step 1 may repeat between succes- sive antenna molecules until the exciton reaches a reaction-center chlorophyll. The asterisk (*) represents the excited state of an antenna molecule. missing electron (an “electron hole,” denoted by H11001 in Fig. 19–46) (step 4 ). The electron acceptor acquires a negative charge in this transaction. The electron lost by the reaction-center chlorophyll is replaced by an electron from a neighboring electron-donor molecule (step 5 ), which thereby becomes positively charged. In this way, excitation by light causes electric charge separation and initiates an oxidation-reduction chain. 8885d_c19_690-750 3/1/04 11:32 AM Page 729 mac76 mac76:385_reb: SUMMARY 19.7 Light Absorption ■ Photophosphorylation in the chloroplasts of green plants and in cyanobacteria involves electron flow through a series of membrane-bound carriers. ■ In the light reactions of plants, absorption of a photon excites chlorophyll molecules and other (accessory) pigments, which funnel the energy into reaction centers in the thylakoid membranes. In the reaction centers, photo- excitation results in a charge separation that produces a strong electron donor (reducing agent) and a strong electron acceptor. 19.8 The Central Photochemical Event: Light-Driven Electron Flow Light-driven electron transfer in plant chloroplasts dur- ing photosynthesis is accomplished by multienzyme sys- tems in the thylakoid membrane. Our current picture of photosynthetic mechanisms is a composite, drawn from studies of plant chloroplasts and a variety of bacteria and algae. Determination of the molecular structures of bacterial photosynthetic complexes (by x-ray crystal- lography) has given us a much improved understanding of the molecular events in photosynthesis in general. Bacteria Have One of Two Types of Single Photochemical Reaction Center One major insight from studies of photosynthetic bacte- ria came in 1952 when Louis Duysens found that illumi- nation of the photosynthetic membranes of the purple bacterium Rhodospirillum rubrum with a pulse of light of a specific wavelength (870 nm) caused a temporary decrease in the absorption of light at that wavelength; a pigment was “bleached” by 870 nm light. Later studies by Bessel Kok and Horst Witt showed similar bleaching of plant chloroplast pigments by light of 680 and 700 nm. Furthermore, addition of the (nonbiological) electron ac- ceptor [Fe(CN) 6 ] 3H11002 (ferricyanide) caused bleaching at these wavelengths without illumination. These find- ings indicated that bleaching of the pigments was due to the loss of an electron from a photochemical reaction center. The pigments were named for the wavelength of maximum bleaching: P870, P680, and P700. Photosynthetic bacteria have relatively simple pho- totransduction machinery, with one of two general types of reaction center. One type (found in purple bacteria) passes electrons through pheophytin (chlorophyll lack- ing the central Mg 2H11001 ion) to a quinone. The other (in green sulfur bacteria) passes electrons through a quinone to an iron-sulfur center. Cyanobacteria and plants have two photosystems (PSI, PSII), one of each type, acting in tandem. Biochemical and biophysical studies have revealed many of the molecular details of reaction centers of bacteria, which therefore serve as prototypes for the more complex phototransduction systems of plants. The Pheophytin-Quinone Reaction Center (Type II Reaction Cen- ter) The photosynthetic machinery in purple bacteria consists of three basic modules (Fig. 19–47a): a single reaction center (P870), a cytochrome bc 1 electron- transfer complex similar to Complex III of the mito- chondrial electron-transfer chain, and an ATP synthase, also similar to that of mitochondria. Illumination drives electrons through pheophytin and a quinone to the cy- tochrome bc 1 complex; after passing through the com- plex, electrons flow through cytochrome c 2 back to the reaction center, restoring its preillumination state. This light-driven cyclic flow of electrons provides the energy for proton pumping by the cytochrome bc 1 complex. Powered by the resulting proton gradient, ATP synthase produces ATP, exactly as in mitochondria. The three-dimensional structures of the reaction centers of purple bacteria (Rhodopseudomonas viridis and Rhodobacter sphaeroides), deduced from x-ray crystallography, shed light on how phototransduction takes place in a pheophytin-quinone reaction center. The R. viridis reaction center (Fig. 19–48a) is a large protein complex containing four polypeptide subunits and 13 cofactors: two pairs of bacterial chlorophylls, a pair of pheophytins, two quinones, a nonheme iron, and four hemes in the associated c-type cytochrome. The extremely rapid sequence of electron transfers shown in Figure 19–48b has been deduced from physi- cal studies of the bacterial pheophytin-quinone centers, using brief flashes of light to trigger phototransduction and a variety of spectroscopic techniques to follow the flow of electrons through several carriers. A pair of bacteriochlorophylls—the “special pair,” designated (Chl) 2 —is the site of the initial photochemistry in the bacterial reaction center. Energy from a photon absorbed by one of the many antenna chlorophyll molecules sur- rounding the reaction center reaches (Chl) 2 by exciton transfer. When these two chlorophyll molecules—so close that their bonding orbitals overlap—absorb an ex- citon, the redox potential of (Chl) 2 is shifted, by an amount equivalent to the energy of the photon, con- verting the special pair to a very strong electron donor. The (Chl) 2 donates an electron that passes through a neighboring chlorophyll monomer to pheophytin (Pheo). This produces two radicals, one positively charged (the special pair of chlorophylls) and one negatively charged (the pheophytin): (Chl) 2 H11001 1 exciton 88n (Chl) 2 * (excitation) (Chl) 2 * H11001 Pheo 88n H11080 (Chl) 2 H11001 H11001 H11080 Pheo H11002 (charge separation) The pheophytin radical now passes its electron to a tightly bound molecule of quinone (Q A ), converting it to a semiquinone radical, which immediately donates its Chapter 19 Oxidative Phosphorylation and Photophosphorylation730 8885d_c19_690-750 3/1/04 11:32 AM Page 730 mac76 mac76:385_reb: extra electron to a second, loosely bound quinone (Q B ). Two such electron transfers convert Q B to its fully re- duced form, Q B H 2 , which is free to diffuse in the mem- brane bilayer, away from the reaction center: 2 H11080 Pheo H11002 H11001 2H H11001 H11001 Q B 88n 2 Pheo H11001 Q B H 2 (quinone reduction) The hydroquinone (Q B H 2 ), carrying in its chemical bonds some of the energy of the photons that originally excited P870, enters the pool of reduced quinone (QH 2 ) dissolved in the membrane and moves through the lipid phase of the bilayer to the cytochrome bc 1 complex. Like the homologous Complex III in mitochondria, the cytochrome bc 1 complex of purple bacteria carries electrons from a quinol donor (QH 2 ) to an electron ac- ceptor, using the energy of electron transfer to pump protons across the membrane, producing a proton- motive force. The path of electron flow through this complex is believed to be very similar to that through mitochondrial Complex III, involving a Q cycle (Fig. 19–12) in which protons are consumed on one side of the membrane and released on the other. The ultimate electron acceptor in purple bacteria is the electron- depleted form of P870, H11080 (Chl) 2 H11001 (Fig. 19–47a). Electrons move from the cytochrome bc 1 complex to P870 via a soluble c-type cytochrome, cytochrome c 2 . The electron- transfer process completes the cycle, returning the reaction center to its unbleached state, ready to absorb another exciton from antenna chlorophyll. A remarkable feature of this system is that all the chemistry occurs in the solid state, with reacting species held close together in the right orientation for reaction. The result is a very fast and efficient series of reactions. The Fe-S Reaction Center (Type I Reaction Center) Photo- synthesis in green sulfur bacteria involves the same three modules as in purple bacteria, but the process dif- fers in several respects and involves additional enzy- matic reactions (Fig. 19–47b). Excitation causes an electron to move from the reaction center to the cy- tochrome bc 1 complex via a quinone carrier. Electron transfer through this complex powers proton transport and creates the proton-motive force used for ATP syn- thesis, just as in purple bacteria and in mitochondria. 19.8 The Central Photochemical Event: Light-Driven Electron Flow 731 0.5 –1.0 –0.5 0 Cyt c 2 e – e – RC P870 P870* Proton gradient Purple bacteria (pheophytin-quinone type) (a) Green sulfur bacteria (Fe-S type) (b) Q Pheo Proton gradient Cyt bc 1 complex Excitons Cyt c 553 RC P840 P840* Q Fd e – Cyt bc 1 complex Fd-NAD reductase Excitons NAD H11001 NADH E H11032H11034 (volts) FIGURE 19–47 Functional modules of photosynthetic machinery in purple bacteria and green sulfur bacteria. (a) In purple bacteria, light energy drives electrons from the reaction center P870 through pheo- phytin (Pheo), a quinone (Q), and the cytochrome bc 1 complex, then through cytochrome c 2 back to the reaction center. Electron flow through the cytochrome bc 1 complex causes proton pumping, creating an elec- trochemical potential that powers ATP synthesis. (b) Green sulfur bac- teria have two routes for electrons driven by excitation of P840: a cyclic route passes through a quinone to the cytochrome bc 1 complex and back to the reaction center via cytochrome c, and a noncyclic route from the reaction center through the iron-sulfur protein ferredoxin (Fd), then to NAD H11001 in a reaction catalyzed by ferredoxin:NAD reductase. 8885d_c19_690-750 3/1/04 11:32 AM Page 731 mac76 mac76:385_reb: (a) Bacteriochlorophyll (2) ((Chl) 2 , the special pair) (270 ns) (3 ps) (200 ps) Hemes of c-type cytochrome Bacteriochlorophyll (2) (accessory pigments) Bacteriopheophytin (2) Q A (quinone) Q B (quinone) Fe Light 3 4 1 2 (b) (6 s)H9262 P side N side FIGURE 19–48 Photoreaction center of the purple bacterium Rhodopseudomonas viridis. (PDB ID 1PRC) (a) The system has four components: three subunits, H, M, and L (brown, blue, and gray, re- spectively), with a total of 11 transmembrane helical segments, and a fourth protein, cytochrome c (yellow), associated with the complex at the membrane surface. Subunits L and M are paired transmembrane proteins that together form a cylindrical structure with roughly bilat- eral symmetry about its long axis. Shown as space-filling models (and in (b) as ball-and-stick structures) are the prosthetic groups that par- ticipate in the photochemical events. Bound to the L and M chains are two pairs of bacteriochlorophyll molecules (green); one of the pairs (the “special pair,” (Chl) 2 ) is the site of the first photochemical changes after light absorption. Also incorporated in the system are a pair of pheophytin a (Pheo a) molecules (blue); two quinones, menaquinone (Q A ) and ubiquinone (Q B ) (orange and yellow), also arranged with bi- lateral symmetry; and a single nonheme Fe (red) located approximately on the axis of symmetry between the quinones. Shown at the top of the figure are four heme groups (red) associated with the c-type cy- tochrome of the reaction center. The reaction center of another pur- ple bacterium, Rhodobacter sphaeroides, is very similar, except that cytochrome c is not part of the crystalline complex. (b) Sequence of events following excitation of the special pair of bacteriochlorophylls (all components colored as in (a)), with the time scale of the electron transfers in parentheses. 1 The excited special pair passes an electron to pheophytin, 2 from which the electron moves rapidly to the tightly bound menaquinone, Q A . 3 This quinone passes electrons much more slowly to the diffusible ubiquinone, Q B , through the nonheme Fe. Meanwhile, 4 the “electron hole” in the special pair is filled by an electron from a heme of cytochrome c. However, in contrast to the cyclic flow of electrons in purple bacteria, some electrons flow from the reaction center to an iron-sulfur protein, ferredoxin, which then passes electrons via ferredoxin:NAD reductase to NAD H11001 , producing NADH. The electrons taken from the reaction center to reduce NAD H11001 are replaced by the ox- idation of H 2 S to elemental S, then to SO 4 2H11002 , in the re- action that defines the green sulfur bacteria. This oxi- dation of H 2 S by bacteria is chemically analogous to the oxidation of H 2 O by oxygenic plants. Kinetic and Thermodynamic Factors Prevent the Dissipation of Energy by Internal Conversion The complex construction of reaction centers is the product of evolutionary selection for efficiency in the photosynthetic process. The excited state (Chl) 2 * could in principle decay to its ground state by internal con- version, a very rapid process (10 picoseconds; 1 ps H11005 10 H1100212 s) in which the energy of the absorbed photon is converted to heat (molecular motion). Reaction centers are constructed to prevent the inefficiency that would result from internal conversion. The proteins of the re- action center hold the bacteriochlorophylls, bacterio- pheophytins, and quinones in a fixed orientation rela- tive to each other, allowing the photochemical reactions to take place in a virtually solid state. This accounts for the high efficiency and rapidity of the reactions; noth- ing is left to chance collision or random diffusion. Exci- ton transfer from antenna chlorophyll to the special pair of the reaction center is accomplished in less than 100 ps with H1102290% efficiency. Within 3 ps of the excitation of P870, pheophytin has received an electron and be- come a negatively charged radical; less than 200 ps later, the electron has reached the quinone Q B (Fig. 19–48b). The electron-transfer reactions not only are fast but are thermodynamically “downhill”; the excited special pair (Chl) 2 * is a very good electron donor (EH11032H11034 H110021 V), and each successive electron transfer is to an acceptor of substantially less negative EH11032H11034. The standard free-energy 732 8885d_c19_690-750 3/1/04 11:32 AM Page 732 mac76 mac76:385_reb: change for the process is therefore negative and large; recall from Chapter 13 that H9004GH11032H11034 H11005 H11002n H9004EH11032H11034; here, H9004EH11032H11034 is the difference between the standard reduction potentials of the two half-reactions (1) (Chl) 2 * 88n H11080 (Chl) 2 H11001 H11001 e H11002 EH11032H11034 ≈ H110021.0 V (2) Q H11001 2H H11001 H11001 2e H11002 88n QH 2 EH11032H11034 H11005 H110020.045 V Thus H9004EH11032H11034 H11005 H110020.045 V H11002 (H110021.0 V) ≈ 0.95 V and H9004GH11032H11034 H11005 H110022(96.5 kJ/V H11554 mol)(0.95 V) H11005H11002180 kJ/mol The combination of fast kinetics and favorable thermo- dynamics makes the process virtually irreversible and highly efficient. The overall energy yield (the percent- age of the photon’s energy conserved in QH 2 ) is H1102230%, with the remainder of the energy dissipated as heat. In Plants, Two Reaction Centers Act in Tandem The photosynthetic apparatus of modern cyanobacteria, algae, and vascular plants is more complex than the one- center bacterial systems, and it appears to have evolved through the combination of two simpler bacterial pho- tocenters. The thylakoid membranes of chloroplasts have two different kinds of photosystems, each with its own type of photochemical reaction center and set of antenna molecules. The two systems have distinct and complementary functions (Fig. 19–49). Photosystem II (PSII) is a pheophytin-quinone type of system (like the single photosystem of purple bacteria) containing roughly equal amounts of chlorophylls a and b. Excita- tion of its reaction center P680 drives electrons through the cytochrome b 6 f complex with concomitant move- ment of protons across the thylakoid membrane. Pho- tosystem I (PSI) is structurally and functionally related to the type I reaction center of green sulfur bac- teria. It has a reaction center designated P700 and a high ratio of chlorophyll a to chlorophyll b. Excited P700 passes electrons to the Fe-S protein ferredoxin, then to NADP H11001 , producing NADPH. The thylakoid membranes of a single spinach chloroplast have many hundreds of each kind of photosystem. These two reaction centers in plants act in tandem to catalyze the light-driven movement of electrons from H 2 O to NADP H11001 (Fig. 19–49). Electrons are carried be- tween the two photosystems by the soluble protein plastocyanin, a one-electron carrier functionally simi- lar to cytochrome c of mitochondria. To replace the elec- trons that move from PSII through PSI to NADP H11001 , cyanobacteria and plants oxidize H 2 O (as green sulfur 19.8 The Central Photochemical Event: Light-Driven Electron Flow 733 1.0 0 E H11032H11034 (volts) –1.0 P680* P680 P700* P700 Photosystem II Photosystem I NADP + Light A 0 A 1 Fe-S Fd Fd:NADP + oxidoreductase Plastocyanin Proton gradient PQ B PQ A Pheo Cyt b 6 f complex e – H 2 O O 2 Light PQ A = plastoquinone PQ B = second quinone A 0 = electron acceptor chlorophyll A 1 = phylloquinone NADPH O 2 evolving complex 1 2 FIGURE 19–49 Integration of photosys- tems I and II in chloroplasts. This “Z scheme” shows the pathway of electron transfer from H 2 O (lower left) to NADP H11001 (far right) in noncyclic photosynthesis. The position on the vertical scale of each electron carrier reflects its standard reduc- tion potential. To raise the energy of electrons derived from H 2 O to the energy level required to reduce NADP H11001 to NADPH, each electron must be “lifted” twice (heavy arrows) by photons absorbed in PSII and PSI. One photon is required per electron in each photosystem. After excitation, the high-energy electrons flow “downhill” through the carrier chains shown. Protons move across the thylakoid membrane during the water-splitting reaction and during electron transfer through the cytochrome b 6 f complex, producing the proton gradient that is central to ATP formation. The dashed arrow is the path of cyclic electron transfer (discussed later in the text), which involves only PSI; electrons return via the cyclic pathway to PSI, instead of reducing NADP H11001 to NADPH. 8885d_c19_690-750 3/1/04 11:32 AM Page 733 mac76 mac76:385_reb: D2 Lumen (P side) Stroma (N side) Pheo Pheo P680 2H 2 O4H + + O 2 D1 (Chl) 2 D 2 (Chl) 2 D 1 Tyr D PQ B Tyr Z Mn 4 PQ A F e 3 1 2 4 5 FIGURE 19–50 Photosystem II of the cyanobacterium Synechococcus elongates. The monomeric form of the complex shown here has two major transmembrane proteins, D1 and D2, each with its set of co- factors. Although the two subunits are nearly symmetric, electron flow occurs through only one of the two branches of cofactors, that on the right (on D1). The arrows show the path of electron flow from the Mn ion cluster (Mn 4 , purple) of the water-splitting enzyme to the quinone PQ B (orange). The photochemical events occur in the sequence indi- cated by the step numbers. Notice the close similarity between the positions of cofactors here and the positions in the bacterial photore- action center shown in Figure 19–48. The role of the Tyr residues is discussed later in the text. bacteria oxidize H 2 S), producing O 2 (Fig. 19–49, bottom left). This process is called oxygenic photosynthesis to distinguish it from the anoxygenic photosynthesis of purple and green sulfur bacteria. All O 2 -evolving photosynthetic cells—those of plants, algae, and cyanobacteria—contain both PSI and PSII; organisms with only one photosystem do not evolve O 2 . The dia- gram in Figure 19–49, often called the Z scheme be- cause of its overall form, outlines the pathway of elec- tron flow between the two photosystems and the energy relationships in the light reactions. The Z scheme thus describes the complete route by which electrons flow from H 2 O to NADP H11001 , according to the equation 2H 2 O H11001 2NADP H11001 H11001 8 photons On O 2 H11001 2NADPH H11001 2H H11001 For every two photons absorbed (one by each photo- system), one electron is transferred from H 2 O to NADP H11001 . To form one molecule of O 2 , which requires transfer of four electrons from two H 2 O to two NADP H11001 , a total of eight photons must be absorbed, four by each photosystem. The mechanistic details of the photochemical reac- tions in PSII and PSI are essentially similar to those of the two bacterial photosystems, with several important additions. In PSII, two very similar proteins, D1 and D2, form an almost symmetrical dimer, to which all the electron-carrying cofactors are bound (Fig. 19–50). Ex- citation of P680 in PSII produces P680*, an excellent electron donor that, within picoseconds, transfers an electron to pheophytin, giving it a negative charge ( H11080 Pheo H11002 ). With the loss of its electron, P680* is trans- formed into a radical cation, P680 H11001 . H11080 Pheo H11002 very rap- idly passes its extra electron to a protein-bound plas- toquinone, PQ A (or Q A ), which in turn passes its electron to another, more loosely bound plastoquinone, PQ B (or Q B ). When PQ B has acquired two electrons in two such transfers from PQ A and two protons from the solvent water, it is in its fully reduced quinol form, PQ B H 2 . The overall reaction initiated by light in PSII is 4P680 H11001 4H H11001 H11001 2PQ B H11001 4 photons On 4P680 H11001 H11001 2PQ B H 2 (19–12) Eventually, the electrons in PQ B H 2 pass through the cytochrome b 6 f complex (Fig. 19–49). The electron ini- tially removed from P680 is replaced with an electron obtained from the oxidation of water, as described below. The binding site for plastoquinone is the point of action of many commercial herbicides that kill plants by blocking electron transfer through the cytochrome b 6 f complex and preventing photosynthetic ATP production. The photochemical events that follow excitation of PSI at the reaction center P700 are formally similar to those in PSII. The excited reaction center P700* loses an electron to an acceptor, A 0 (believed to be a special form of chlorophyll, functionally homologous to the pheophytin of PSII), creating A 0 H11002 and P700 H11001 (Fig. 19–49, right side); again, excitation results in charge separation at the photochemical reaction center. P700 H11001 is a strong oxidizing agent, which quickly acquires an electron from plastocyanin, a soluble Cu-containing electron-transfer protein. A 0 H11002 is an exceptionally strong reducing agent that passes its electron through a chain of carriers that leads to NADP H11001 . First, phylloquinone (A 1 ) accepts an electron and passes it to an iron-sulfur protein (through three Fe-S centers in PSI). From here, the electron moves to ferredoxin (Fd), another iron-sulfur protein loosely associated with the thylakoid membrane. Spinach ferredoxin (M r 10,700) contains a 2Fe-2S center (Fig. 19–5) that undergoes one-electron oxidation and reduc- tion reactions. The fourth electron carrier in the chain is the flavoprotein ferredoxin : NADP H11001 oxidoreduc- tase, which transfers electrons from reduced ferredoxin (Fd red ) to NADP H11001 : 2Fd red H11001 2H H11001 H11001 NADP H11001 On 2Fdox H11001 NADPH H11001 H H11001 This enzyme is homologous to the ferredoxin:NAD re- ductase of green sulfur bacteria (Fig. 19–47b). Antenna Chlorophylls Are Tightly Integrated with Electron Carriers The electron-carrying cofactors of PSI and the light- harvesting complexes are part of a supramolecular com- plex (Fig. 19–51a), the structure of which has been solved crystallographically. The protein consists of three identical complexes, each composed of 11 different pro- teins (Fig. 19–51b). In this remarkable structure the many antenna chlorophyll and carotenoid molecules are Chapter 19 Oxidative Phosphorylation and Photophosphorylation734 8885d_c19_690-750 3/1/04 11:32 AM Page 734 mac76 mac76:385_reb: 19.8 The Central Photochemical Event: Light-Driven Electron Flow 735 FIGURE 19–51 The supramolecular complex of PSI and its associated antenna chlorophylls. (a) Schematic drawing of the essential proteins and cofactors in a single unit of PSI. A large number of antenna chloro- phylls surround the reaction center and convey to it (red arrows) the energy of photons they have absorbed. The result is excitation of the pair of chlorophyll molecules that constitute P700. Excitation of P700 greatly decreases its reduction potential, and it passes an electron through two nearby chlorophylls to phylloquinone (Q K ; also called A 1 ). Reduced phylloquinone is reoxidized as it passes two electrons, one at a time, to an Fe-S protein (F X ) near the N side of the mem- brane. From F X , electrons move through two more Fe-S centers (F A and F B ), then to the Fe-S protein ferredoxin in the stroma. Ferredoxin then donates electrons to NADP H11001 (not shown), reducing it to NADPH, one of the forms in which the energy of photons is trapped in chloro- plasts. (b) The trimeric structure (derived from PDB ID 1JBO), viewed from the thylakoid lumen perpendicular to the membrane, showing all protein subunits (gray) and cofactors (described in (c)). (c) A monomer of PSI with all the proteins omitted, revealing the antenna and reaction center chlorophylls (green with dark green Mg 2H11001 ions in the center), carotenoids (yellow), and Fe-S centers of the reaction center (space-filling red and orange structures). The proteins in the complex hold the components rigidly in orientations that maximize efficient exciton transfers between excited antenna molecules and the reaction center. Light Subunit B PSI Stroma (N side) Lumen (P side) Chl F X F A Subunit C F B (Chl) A 0 Chl (Chl) A 0 Q K Q K e – e – e – e – Subunit A (a) Ferredoxin Plastocyanin (Chl) 2 P 700 Exciton transfer (b) (c) 8885d_c19_690-750 3/1/04 11:32 AM Page 735 mac76 mac76:385_reb: Chapter 19 Oxidative Phosphorylation and Photophosphorylation736 Photosystem I Stacked membranes (granal lamellae) Photosystem II ATP synthase Cytochrome b 6 f complex Light-harvesting complex II Unstacked membranes (stromal lamellae) Stroma Lumen FIGURE 19–52 Localization of PSI and PSII in thylakoid membranes. Light- harvesting complex LHCII and ATP synthase are located in regions of the thylakoid membrane that are appressed (granal lamellae, in which several membranes are in contact) and in regions that are not appressed (stromal lamellae) and have ready access to ADP and NADP H11001 in the stroma. Photosystem II is present almost exclusively in the appressed regions, and photosystem I almost exclusively in nonappressed regions exposed to the stroma. LHCII is the “adhesive” that holds appressed lamellae together (see Fig. 19–53). LHCII –Thr–OH –Thr– Thylakoid membrane Appressed Nonappressed LHCII ATP P i ADP protein phosphatase protein kinase P FIGURE 19–53 Equalization of electron flow in PSI and PSII by mod- ulation of granal stacking. A hydrophobic domain of LHCII in one thylakoid lamella inserts into the neighboring lamella and closely ap- presses the two membranes. Accumulation of plastoquinol (not shown) stimulates a protein kinase that phosphorylates a Thr residue in the hydrophobic domain of LHCII, which reduces its affinity for the neigh- boring thylakoid membrane and converts appressed regions (granal lamellae) to nonappressed regions (stromal lamellae). A specific pro- tein phosphatase reverses this regulatory phosphorylation when the [PQ]/[PQH 2 ] ratio increases. precisely arrayed around the reaction center (Fig. 19–51c). The reaction center’s electron-carrying cofac- tors are therefore tightly integrated with antenna chlorophylls. This arrangement allows very rapid and ef- ficient exciton transfer from antenna chlorophylls to the reaction center. In contrast to the single path of electrons in PSII, the electron flow initiated by absorption of a photon is believed to occur through both branches of carriers in PSI. Spatial Separation of Photosystems I and II Prevents Exciton Larceny The energy required to excite PSI (P700) is less than that needed to excite PSII (P680) (shorter wavelength corresponds to higher energy). If PSI and PSII were physically contiguous, excitons originating in the an- tenna system of PSII would migrate to the reaction cen- ter of PSI, leaving PSII chronically underexcited and in- terfering with the operation of the two-center system. This “exciton larceny” is prevented by separation of PSI and PSII in the thylakoid membrane (Fig. 19–52). PSII is located almost exclusively in the tightly appressed membrane stacks of thylakoid grana (granal lamellae); its associated light-harvesting complex (LHCII) medi- ates the tight association of adjacent membranes in the grana. PSI and the ATP synthase complex are located almost exclusively in the thylakoid membranes that are not appressed (the stromal lamellae), where both have access to the contents of the stroma, including ADP and NADP H11001 . The cytochrome b 6 f complex is present throughout the thylakoid membrane. The association of LHCII with PSII is regulated by light intensity and wavelength. In bright sunlight (with a large component of blue light), PSII absorbs more light than PSI and produces reduced plastoquinone (plasto- quinol, PQH 2 ) faster than PSI can oxidize it. The re- sulting accumulation of PQH 2 activates a protein kinase that phosphorylates a Thr residue on LHCII (Fig. 19–53). Phosphorylation weakens the interaction of LHCII with PSII, and some LHCII dissociates and moves to the stro- mal lamellae; here it captures photons for PSI, speed- ing the oxidation of PQH 2 and reversing the imbalance between electron flow in PSI and PSII. In less intense light (in the shade, with more red light), PSI oxidizes PQH 2 faster than PSII can make it, and the resulting in- crease in PQ concentration triggers dephosphorylation of LHCII, reversing the effect of phosphorylation. 8885d_c19_690-750 3/1/04 11:32 AM Page 736 mac76 mac76:385_reb: Subunit IV Cyt b 6 b L b H Cyt f Rieske iron- sulfur protein Thylakoid lumen (P side) Q cycle e – e – Stroma (N side) Cu Plasto- cyanin PQ PQH 2 4H + 2 × 2H + (c) FIGURE 19–54 Electron and proton flow through the cytochrome b 6 f complex. (a) The crystal structure of the complex (PDB ID 1UM3) reveals the positions of the cofac- tors involved in electron transfers. In addition to the hemes of cytochrome b (heme b H and b L ; also called heme bN and b P , respectively, because of their proximity to the N and P sides of the bilayer) and that of cytochrome f (heme f ), there is a fourth (heme x) near heme b H , and there is a H9252-carotene of unknown function. Two sites bind plasto- quinone: the PQH 2 site near the P side of the bilayer, and the PQ site near the N side. The Fe-S center of the Rieske protein lies just outside the bilayer on the P side, and the heme f site is on a protein domain that extends well into the thylakoid lumen. (b) The complex is a homodimer arranged to create a cavern connecting the PQH 2 and PQ sites (compare with the structure of mitochondrial Complex III in Fig. 19–12). This cavern allows plastoquinone movement between the sites of its oxidation and reduction. (c) Plastoquinol (PQH 2 ) formed in PSII is oxidized by the cytochrome b 6 f complex in a series of steps like those of the Q cycle in the cytochrome bc 1 complex (Complex III) of mitochondria (see Fig. 19–11). One electron from PQH 2 passes to the Fe-S center of the Rieske protein (purple), the other to heme b L of cytochrome b 6 (green). The net effect is passage of electrons from PQH 2 to the soluble protein plastocyanin, which carries them to PSI. The Cytochrome b 6 f Complex Links Photosystems II and I Electrons temporarily stored in plastoquinol as a result of the excitation of P680 in PSII are carried to P700 of PSI via the cytochrome b 6 f complex and the soluble pro- tein plastocyanin (Fig. 19–49, center). Like Complex III of mitochondria, the cytochrome b 6 f complex (Fig. 19–54) contains a b-type cytochrome with two heme groups (designated b H and b L ), a Rieske iron-sulfur pro- tein (M r 20,000), and cytochrome f (for the Latin frons, “leaf”). Electrons flow through the cytochrome b 6 f com- plex from PQ B H 2 to cytochrome f, then to plastocyanin, and finally to P700, thereby reducing it. 19.8 The Central Photochemical Event: Light-Driven Electron Flow 737 Stroma (N side) Lumen (P side) Heme ? Fe-S center Plastocyanin H + 2H + Heme x PQ PQH 2 Heme b H Heme b L e – e – e – e – e – (a) (b) b-carotene 8885d_c19_737 3/1/04 2:10 PM Page 737 mac76 mac76:385_reb: what is available in a particular ecological niche. About 3 billion years ago, evolution of primitive photosynthetic bacteria (the progenitors of the modern cyanobacteria) produced a photosystem capable of taking electrons from a donor that is always available—water. Two water molecules are split, yielding four electrons, four protons, and molecular oxygen: 2H 2 O 88n 4H H11001 H11001 4e H11002 H11001 O 2 A single photon of visible light does not have enough energy to break the bonds in water; four photons are required in this photolytic cleavage reaction. The four electrons abstracted from water do not pass directly to P680 H11001 , which can accept only one elec- tron at a time. Instead, a remarkable molecular device, the oxygen-evolving complex (also called the water- splitting complex), passes four electrons one at a Chapter 19 Oxidative Phosphorylation and Photophosphorylation738 Like Complex III of mitochondria, cytochrome b 6 f conveys electrons from a reduced quinone—a mobile, lipid-soluble carrier of two electrons (Q in mitochondria, PQ B in chloroplasts)—to a water-soluble protein that carries one electron (cytochrome c in mitochondria, plastocyanin in chloroplasts). As in mitochondria, the function of this complex involves a Q cycle (Fig. 19–12) in which electrons pass, one at a time, from PQ B H 2 to cytochrome b 6 . This cycle results in the pumping of pro- tons across the membrane; in chloroplasts, the direction of proton movement is from the stromal compartment to the thylakoid lumen, up to four protons moving for each pair of electrons. The result is production of a pro- ton gradient across the thylakoid membrane as electrons pass from PSII to PSI. Because the volume of the flat- tened thylakoid lumen is small, the influx of a small num- ber of protons has a relatively large effect on lumenal pH. The measured difference in pH between the stroma (pH 8) and the thylakoid lumen (pH 5) represents a 1,000-fold difference in proton concentration—a pow- erful driving force for ATP synthesis. Cyanobacteria Use the Cytochrome b 6 f Complex and Cytochrome c 6 in Both Oxidative Phosphorylation and Photophosphorylation Cyanobacteria can synthesize ATP by oxidative phos- phorylation or by photophosphorylation, although they have neither mitochondria nor chloroplasts. The enzy- matic machinery for both processes is in a highly con- voluted plasma membrane (see Fig. 1–6). Two protein components function in both processes (Fig. 19–55). The proton-pumping cytochrome b 6 f complex carries electrons from plastoquinone to cytochrome c 6 in pho- tosynthesis, and also carries electrons from ubiquinone to cytochrome c 6 in oxidative phosphorylation—the role played by cytochrome bc 1 in mitochondria. Cytochrome c 6 , homologous to mitochondrial cytochrome c, carries electrons from Complex III to Complex IV in cyanobac- teria; it can also carry electrons from the cytochrome b 6 f complex to PSI—a role performed in plants by plas- tocyanin. We therefore see the functional homology be- tween the cyanobacterial cytochrome b 6 f complex and the mitochondrial cytochrome bc 1 complex, and between cyanobacterial cytochrome c 6 and plant plastocyanin. Water Is Split by the Oxygen-Evolving Complex The ultimate source of the electrons passed to NADPH in plant (oxygenic) photosynthesis is water. Having given up an electron to pheophytin, P680 H11001 (of PSII) must acquire an electron to return to its ground state in preparation for capture of another photon. In princi- ple, the required electron might come from any number of organic or inorganic compounds. Photosynthetic bac- teria use a variety of electron donors for this purpose— acetate, succinate, malate, or sulfide—depending on H 2 O 2 Fd red 2 Fd ox NADPH + H + NADP + PSII PSI Cyt b 6 f Complex III NAD + NADH + H + H 2 OO 2 NADH dehydrogenase Complex I e H11002 e H11002 Cyt a H11001 a 3 Complex IV 2 1 O 2 2 1 Plasto- quinone (a) (b) Photophosphorylation Oxidative phosphorylation Cyt c 6 Light Light FIGURE 19–55 Dual roles of cytochrome b 6 f and cytochrome c 6 in cyanobacteria. Cyanobacteria use cytochrome b 6 f, cytochrome c 6 , and plastoquinone for both oxidative phosphorylation and pho- tophosphorylation. (a) In photophosphorylation, electrons flow (top to bottom) from water to NADP H11001 . (b) In oxidative phosphorylation, elec- trons flow from NADH to O 2 . Both processes are accompanied by proton movement across the membrane, accomplished by a Q cycle. 8885d_c19_690-750 3/1/04 11:32 AM Page 738 mac76 mac76:385_reb: 19.8 The Central Photochemical Event: Light-Driven Electron Flow 739 time to P680 H11001 (Fig. 19–56). The immediate electron donor to P680 H11001 is a Tyr residue (often designated Z or Tyr z ) in protein subunit D1 of the PSII reaction center. The Tyr residue loses both a proton and an electron, generating the electrically neutral Tyr free radical, H11080 Tyr: 4P680 H11001 H11001 4 Tyr 88n 4P680 H11001 4 H11080 Tyr (19–13) The Tyr radical regains its missing electron and proton by oxidizing a cluster of four manganese ions in the water-splitting complex. With each single-electron transfer, the Mn cluster becomes more oxidized; four single-electron transfers, each corresponding to the ab- sorption of one photon, produce a charge of H110014 on the Mn complex (Fig. 19–56): 4 H11080 Tyr H11001 [Mn complex] 0 88n 4 Tyr H11001 [Mn complex] 4H11001 (19–14) In this state, the Mn complex can take four electrons from a pair of water molecules, releasing 4 H H11001 and O 2 : [Mn complex] 4H11001 H11001 2H 2 O 88n [Mn complex] 0 H11001 4H H11001 H11001 O 2 (19–15) Because the four protons produced in this reaction are released into the thylakoid lumen, the oxygen-evolving complex acts as a proton pump, driven by electron transfer. The sum of Equations 19–12 through 19–15 is 2H 2 O H11001 2PQ B H11001 4 photons On O 2 H11001 2PQ B H 2 (19–16) The water-splitting activity associated with the PSII reaction center has proved exceptionally difficult to pu- rify. A peripheral membrane protein (M r 33,000) on the lumenal side of the thylakoid membrane is believed to stabilize the Mn complex. In the crystal structure (PDB ID 1FE1; see Fig. 19–50), four Mn ions are clustered with precise geometry near a Tyr residue on the D1 sub- unit, presumably the one involved in water oxidation. Manganese can exist in stable oxidation states from H110012 to H110017, so a cluster of four Mn ions can certainly donate or accept four electrons. The mechanism shown in Fig- ure 19–56 is consistent with the experimental facts, but until the exact chemical structures of all the interme- diates of the Mn cluster are known, the detailed mech- anism remains elusive. SUMMARY 19.8 The Central Photochemical Event: Light-Driven Electron Flow ■ Bacteria have a single reaction center; in purple bacteria, it is of the pheophytin-quinone type, and in green sulfur bacteria, the Fe-S type. ■ Structural studies of the reaction center of a purple bacterium have provided information about light-driven electron flow from an excited special pair of chlorophyll molecules, through pheophytin, to quinones. Electrons then pass from quinones through the cytochrome bc 1 complex, and back to the photoreaction center. ■ An alternative path, in green sulfur bacteria, sends electrons from reduced quinones to NAD H11001 . FIGURE 19–56 Water-splitting activity of the oxygen-evolving com- plex. Shown here is the process that produces a four-electron oxidiz- ing agent—believed to be a multinuclear center with several Mn ions— in the water-splitting complex of PSII. The sequential absorption of four photons (excitons), each absorption causing the loss of one elec- tron from the Mn center, produces an oxidizing agent that can remove four electrons from two molecules of water, producing O 2 . The elec- trons lost from the Mn center pass one at a time to an oxidized Tyr residue in a PSII protein, then to P680 H11001 . Tyr Exciton Exciton Exciton Exciton e H11002 e H11002 H H11001 Lumen e H11002 e H11002 Mn complex 0 Mn complex 1H11001 Mn complex 2H11001 Mn complex 3H11001 Mn complex O 2 4H11001 H H11001 H H11001 H H11001 2H 2 O P680 8885d_c19_690-750 3/1/04 11:32 AM Page 739 mac76 mac76:385_reb: ■ Cyanobacteria and plants have two different photoreaction centers, arranged in tandem. ■ Plant photosystem I passes electrons from its excited reaction center, P700, through a series of carriers to ferredoxin, which then reduces NADP H11001 to NADPH. ■ The reaction center of plant photosystem II, P680, passes electrons to plastoquinone, and the electrons lost from P680 are replaced by electrons from H 2 O (electron donors other than H 2 O are used in other organisms). ■ The light-driven splitting of H 2 O is catalyzed by a Mn-containing protein complex; O 2 is produced. The reduced plastoquinone carries electrons to the cytochrome b 6 f complex; from here they pass to plastocyanin, and then to P700 to replace those lost during its photoexcitation. ■ Electron flow through the cytochrome b 6 f complex drives protons across the plasma membrane, creating a proton-motive force that provides the energy for ATP synthesis by an ATP synthase. 19.9 ATP Synthesis by Photophosphorylation The combined activities of the two plant photosystems move electrons from water to NADP H11001 , conserving some of the energy of absorbed light as NADPH (Fig. 19–49). Simultaneously, protons are pumped across the thy- lakoid membrane and energy is conserved as an elec- trochemical potential. We turn now to the process by which this proton gradient drives the synthesis of ATP, the other energy-conserving product of the light- dependent reactions. In 1954 Daniel Arnon and his colleagues discovered that ATP is generated from ADP and P i during photosynthetic electron transfer in illuminated spinach chloroplasts. Support for these findings came from the work of Albert Frenkel, who detected light-dependent ATP production in pigment- containing membranous struc- tures called chromatophores, derived from photosynthetic bacteria. Investigators concluded that some of the light energy captured by the photosynthetic systems of these organisms is transformed into the phosphate bond en- ergy of ATP. This process is called photophosphory- lation, to distinguish it from oxidative phosphorylation in respiring mitochondria. A Proton Gradient Couples Electron Flow and Phosphorylation Several properties of photosynthetic electron transfer and photophosphorylation in chloroplasts indicate that a proton gradient plays the same role as in mitochon- drial oxidative phosphorylation. (1) The reaction centers, electron carriers, and ATP-forming enzymes are located in a proton-impermeable membrane—the thy- lakoid membrane—which must be intact to support pho- tophosphorylation. (2) Photophosphorylation can be uncoupled from electron flow by reagents that promote the passage of protons through the thylakoid membrane. (3) Photophosphorylation can be blocked by venturi- cidin and similar agents that inhibit the formation of ATP from ADP and P i by the mitochondrial ATP synthase (Table 19–4). (4) ATP synthesis is catalyzed by F o F 1 complexes, located on the outer surface of the thylakoid membranes, that are very similar in structure and func- tion to the F o F 1 complexes of mitochondria. Electron-transferring molecules in the chain of carriers connecting PSII and PSI are oriented asym- metrically in the thylakoid membrane, so photoinduced electron flow results in the net movement of protons across the membrane, from the stromal side to the thy- lakoid lumen (Fig. 19–57). In 1966 André Jagendorf showed that a pH gradient across the thylakoid membrane (alkaline outside) could furnish the driving force to generate ATP. His early observations pro- vided some of the most im- portant experimental evi- dence in support of Mitchell’s chemiosmotic hypothesis. Jagendorf incubated chloroplasts, in the dark, in a pH 4 buffer; the buffer slowly penetrated into the inner compartment of the thylakoids, lowering their internal pH. He added ADP and P i to the dark suspension of chloroplasts and then suddenly raised the pH of the outer medium to 8, momentarily creating a large pH gra- dient across the membrane. As protons moved out of the thylakoids into the medium, ATP was generated from ADP and P i . Because the formation of ATP oc- curred in the dark (with no input of energy from light), this experiment showed that a pH gradient across the membrane is a high-energy state that, as in mitochon- drial oxidative phosphorylation, can mediate the trans- duction of energy from electron transfer into the chem- ical energy of ATP. Chapter 19 Oxidative Phosphorylation and Photophosphorylation740 Daniel Arnon, 1910–1994 André Jagendorf 8885d_c19_690-750 3/1/04 11:32 AM Page 740 mac76 mac76:385_reb: 19.9 ATP Synthesis by Photophosphorylation 741 Stroma (N side) PSII PSI Lumen (P side) Thylakoid membrane Light 2H H11001 NADP H11001 H11001H H11001 O 2 H11001 4H H11001 PQH 2 2H 2 O ADP H11001 P i ATP 2H H11001 NADPH Plasto- cyanin PQ CF O CF 1 Fd Cyt b 6 f complex Mn Light FIGURE 19–57 Proton and electron circuits in thylakoids. Electrons (blue arrows) move from H 2 O through PSII, through the intermediate chain of carriers, through PSI, and finally to NADP H11001 . Protons (red ar- rows) are pumped into the thylakoid lumen by the flow of electrons through the carriers linking PSII and PSI, and reenter the stroma through proton channels formed by the F o (designated CF o ) of ATP synthase. The F 1 subunit (CF 1 ) catalyzes synthesis of ATP. The Approximate Stoichiometry of Photophosphorylation Has Been Established As electrons move from water to NADP H11001 in plant chloro- plasts, about 12 H H11001 move from the stroma to the thy- lakoid lumen per four electrons passed (that is, per O 2 formed). Four of these protons are moved by the oxygen-evolving complex, and up to eight by the cy- tochrome b 6 f complex. The measurable result is a 1,000-fold difference in proton concentration across the thylakoid membrane (H9004pH H11005 3). Recall that the energy stored in a proton gradient (the electrochemical poten- tial) has two components: a proton concentration dif- ference (H9004pH) and an electrical potential (H9004H9274) due to charge separation. In chloroplasts, H9004pH is the dominant component; counterion movement apparently dissipates most of the electrical potential. In illuminated chloro- plasts, the energy stored in the proton gradient per mole of protons is H9004G H11005 2.3RT H9004pH H11001 Z H9004H9274 H11005H1100217 kJ/mol so the movement of 12 mol of protons across the thylakoid membrane represents conservation of about 200 kJ of energy—enough energy to drive the synthe- sis of several moles of ATP (H9004GH11032H11034 H11005 30.5 kJ/mol). Ex- perimental measurements yield values of about 3 ATP per O 2 produced. At least eight photons must be absorbed to drive four electrons from H 2 O to NADPH (one photon per electron at each reaction center). The energy in eight photons of visible light is more than enough for the syn- thesis of three molecules of ATP. ATP synthesis is not the only energy-conserving re- action of photosynthesis in plants; the NADPH formed in the final electron transfer is (like its close analog NADH) also energetically rich. The overall equation for noncyclic photophosphorylation (a term explained be- low) is 2H 2 O H11001 8 photons H11001 2NADP H11001 H11001 ~3ADP H11001 ~3P i O n O 2 H11001 ~3ATP H11001 2NADPH (19–17) Cyclic Electron Flow Produces ATP but Not NADPH or O 2 Using an alternative path of light-induced electron flow, plants can vary the ratio of NADPH to ATP formed in the light; this path is called cyclic electron flow to dif- ferentiate it from the normally unidirectional or non- cyclic electron flow from H 2 O to NADP H11001 , as discussed thus far. Cyclic electron flow (Fig. 19–49) involves only PSI. Electrons passing from P700 to ferredoxin do not continue to NADP H11001 , but move back through the cytochrome b 6 f complex to plastocyanin. The path of 8885d_c19_690-750 3/1/04 11:32 AM Page 741 mac76 mac76:385_reb: electrons matches that in green sulfur bacteria (Fig. 19–47b). Plastocyanin donates electrons to P700, which transfers them to ferredoxin when the plant is illumi- nated. Thus, in the light, PSI can cause electrons to cycle continuously out of and back into the reaction center of PSI, each electron propelled around the cycle by the en- ergy yielded by the absorption of one photon. Cyclic electron flow is not accompanied by net formation of NADPH or evolution of O 2 . However, it is accompanied by proton pumping by the cytochrome b 6 f complex and by phosphorylation of ADP to ATP, referred to as cyclic photophosphorylation. The overall equation for cyclic electron flow and photophosphorylation is simply light ADP H11001 P i 888n ATP H11001 H 2 O By regulating the partitioning of electrons between NADP H11001 reduction and cyclic photophosphorylation, a plant adjusts the ratio of ATP to NADPH produced in the light-dependent reactions to match its needs for these products in the carbon-assimilation reactions and other biosynthetic processes. As we shall see in Chap- ter 20, the carbon-assimilation reactions require ATP and NADPH in the ratio 3:2. The ATP Synthase of Chloroplasts Is Like That of Mitochondria The enzyme responsible for ATP synthesis in chloro- plasts is a large complex with two functional compo- nents, CF o and CF 1 (C denoting its location in chloroplasts). CF o is a transmembrane proton pore composed of several integral membrane proteins and is homologous to mitochondrial F o . CF 1 is a peripheral membrane protein complex very similar in subunit com- position, structure, and function to mitochondrial F 1 . Electron microscopy of sectioned chloroplasts shows ATP synthase complexes as knoblike projections on the outside (stromal or N) surface of thylakoid mem- branes; these complexes correspond to the ATP syn- thase complexes seen to project on the inside (matrix or N) surface of the inner mitochondrial membrane. Thus the relationship between the orientation of the ATP synthase and the direction of proton pumping is the same in chloroplasts and mitochondria. In both cases, the F 1 portion of ATP synthase is located on the more alkaline (N) side of the membrane through which protons flow down their concentration gradient; the di- rection of proton flow relative to F 1 is the same in both cases: P to N (Fig. 19–58). The mechanism of chloroplast ATP synthase is also believed to be essentially identical to that of its mito- chondrial analog; ADP and P i readily condense to form ATP on the enzyme surface, and the release of this enzyme-bound ATP requires a proton-motive force. Ro- tational catalysis sequentially engages each of the three H9252 subunits of the ATP synthase in ATP synthesis, ATP release, and ADP H11001 P i binding (Figs 19–24, 19–25). Chloroplasts Evolved from Endosymbiotic Bacteria Like mitochondria, chloroplasts contain their own DNA and protein-synthesizing machinery. Some of the polypeptides of chloroplast proteins are encoded by chloroplast genes and synthesized in the chloroplast; others are encoded by nuclear genes, synthesized out- side the chloroplast, and imported (Chapter 27). When plant cells grow and divide, chloroplasts give rise to new Chapter 19 Oxidative Phosphorylation and Photophosphorylation742 Mitochondrion Matrix (N side) Intermembrane space (P side) Bacterium (E. coli)Chloroplast H H11001 Thylakoid lumen (P side) Stroma (N side) ATP Cytosol (N side) Intermembrane space (P side) ATP ATP H H11001 H H11001 FIGURE 19–58 Comparison of the topology of proton movement and ATP synthase orientation in the membranes of mitochondria, chloroplasts, and the bacterium E. coli. In each case, orien- tation of the proton gradient relative to ATP synthase activity is the same. 8885d_c19_690-750 3/1/04 11:32 AM Page 742 mac76 mac76:385_reb: chloroplasts by division, during which their DNA is repli- cated and divided between daughter chloroplasts. The machinery and mechanism for light capture, electron flow, and ATP synthesis in photosynthetic bacteria are similar in many respects to those in the chloroplasts of plants. These observations led to the now widely ac- cepted hypothesis that the evolutionary progenitors of modern plant cells were primitive eukaryotes that en- gulfed photosynthetic bacteria and established stable endosymbiotic relationships with them (see Fig. 1–36). Diverse Photosynthetic Organisms Use Hydrogen Donors Other Than Water At least half of the photosynthetic activity on Earth occurs in microorganisms—algae, other photosynthetic eukaryotes, and photosynthetic bacteria. Cyanobacteria have PSII and PSI in tandem, and the PSII has an asso- ciated water-splitting activity resembling that of plants. However, the other groups of photosynthetic bacteria have single reaction centers and do not split H 2 O or produce O 2 . Many are obligate anaerobes and cannot tolerate O 2 ; they must use some compound other than H 2 O as electron donor. Some photosynthetic bacteria use inorganic compounds as electron (and hydrogen) donors. For example, green sulfur bacteria use hydrogen sulfide: light 2H 2 S H11001 CO 2 888n (CH 2 O) H11001 H 2 O H11001 2S These bacteria, instead of producing molecular O 2 , form elemental sulfur as the oxidation product of H 2 S. (They further oxidize the S to SO 4 2H11002 .) Other photosynthetic bacteria use organic compounds such as lactate as elec- tron donors: light 2 Lactate H11001 CO 2 888n (CH 2 O) H11001 H 2 O H11001 2 pyruvate The fundamental similarity of photosynthesis in plants and bacteria, despite the differences in the electron donors they employ, becomes more obvious when the equation of photosynthesis is written in the more gen- eral form light 2H 2 D H11001 CO 2 888n (CH 2 O) H11001 H 2 O H11001 2D in which H 2 D is an electron (and hydrogen) donor and D is its oxidized form. H 2 D may be water, hydrogen sul- fide, lactate, or some other organic compound, de- pending on the species. Most likely, the bacteria that first developed photosynthetic ability used H 2 S as their electron source, and only after the later development of oxygenic photosynthesis (about 2.3 billion years ago) did oxygen become a significant proportion of the earth’s atmosphere. With that development, the evolu- tion of electron-transfer systems that used O 2 as their ultimate electron acceptor became possible, leading to the highly efficient energy extraction of oxidative phos- phorylation. In Halophilic Bacteria, a Single Protein Absorbs Light and Pumps Protons to Drive ATP Synthesis The halophilic (“salt-loving”) bacterium Halobacterium salinarum, an archaebacterium derived from very an- cient evolutionary progenitors, traps the energy of sun- light in a process very different from the photosynthetic mechanisms we have described so far. This bacterium lives only in brine ponds and salt lakes (Great Salt Lake and the Dead Sea, for example), where the high salt concentration—which can exceed 4 M—results from water loss by evaporation; indeed, halobacteria cannot live in NaCl concentrations lower than 3 M. These or- ganisms are aerobes and normally use O 2 to oxidize organic fuel molecules. However, the solubility of O 2 is so low in brine ponds that sometimes oxidative metab- olism must be supplemented by sunlight as an alterna- tive source of energy. The plasma membrane of H. salinarum contains patches of the light-absorbing pigment bacteriorho- dopsin, which contains retinal (the aldehyde derivative of vitamin A; see Fig. 10–21) as a prosthetic group. When the cells are illuminated, all-trans-retinal bound to the bacteriorhodopsin absorbs a photon and under- goes photoisomerization to 13-cis-retinal. The restora- tion of all-trans-retinal is accompanied by the outward movement of protons through the plasma membrane. Bacteriorhodopsin, with only 247 amino acid residues, is the simplest light-driven proton pump known. The dif- ference in the three-dimensional structure of bacteri- orhodopsin in the dark and after illumination (Fig. 19–59a) suggests a pathway by which a concerted se- ries of proton “hops” could effectively move a proton across the membrane. The chromophore retinal is bound through a Schiff-base linkage to the H9255-amino group of a Lys residue. In the dark, the N of this Schiff base is pro- tonated; in the light, photoisomerization of retinal low- ers the pK a of this group and it releases its proton to a nearby Asp residue, triggering a series of proton hops that ultimately result in the release of a proton at the outer surface of the membrane (Fig. 19–59b). The electrochemical potential across the membrane drives protons back into the cell through a membrane ATP synthase complex very similar to that of mitochon- dria and chloroplasts. Thus, when O 2 is limited, halobac- teria can use light to supplement the ATP synthesized by oxidative phosphorylation. Halobacteria do not evolve O 2 , nor do they carry out photoreduction of NADP H11001 ; their phototransducing machinery is therefore much simpler than that of cyanobacteria or plants. Neverthe- less, the proton-pumping mechanism used by this simple protein may prove to be prototypical for the many other, more complex, ion pumps. Bacteriorhodopsin ■ 19.9 ATP Synthesis by Photophosphorylation 743 8885d_c19_690-750 3/1/04 11:32 AM Page 743 mac76 mac76:385_reb: FIGURE 19–59 Light-driven proton pumping by bacteriorhodopsin. (a) Bacteriorhodopsin (M r 26,000) has seven membrane-spanning H9251 helices (PDB ID 1C8R). The chromophore all-trans-retinal (purple) is covalently attached via a Schiff base to the H9255-amino group of a Lys residue deep in the membrane interior. Running through the protein are a series of Asp and Glu residues and a series of closely associated water molecules that together provide the transmembrane path for pro- tons (red arrows). Steps 1 through 5 indicate proton movements, described below. (b) In the dark (left panel), the Schiff base is protonated. Illumina- tion (right panel) photoisomerizes the retinal, forcing subtle confor- mational changes in the protein that alter the distance between the Schiff base and its neighboring amino acid residues. Interaction with these neighbors lowers the pK a of the protonated Schiff base, and the base gives up its proton to a nearby carboxyl group on Asp 85 (step 1 in (a)). This initiates a series of concerted proton hops between water molecules (see Fig. 2–14) in the interior of the protein, which ends with 2 the release of a proton that was shared by Glu 194 and Glu 204 near the extracellular surface. 3 The Schiff base reacquires a proton from Asp 96 , which 4 takes up a proton from the cytosol. 5 Finally, Asp 85 gives up its proton, leading to reprotonation of the Glu 204 -Glu 194 pair. The system is now ready for another round of proton pumping. (a) Cytosol Asp 96 Asp 85 Arg 82 Glu 204 Retinal Glu 194 H + H+ External medium 4 3 2 1 5 Thr 89 OH Asp 85 C C O OO Protonated Schiff base (high pK a ) Proton-release complex (protonated; high pK a ) Low pK a Retinal Dark O N H H11001 Lys 216 Arg 82 Glu 194 C HO O Glu 204 CN H NH 2 NH 2 H11001 H11001 N Lys 216 Leu 93 Val 49 Asp 85 C C O OO Conformational change lowers pK a of Schiff base pK a of proton- release complex lowered Higher pK a Light OH Arg 82 Tyr 83 Glu 204 C OO Glu 194 Proton release Proton transfer CNH NH 2 OH NH 2 H11002 H11002 H11002 H11002 Chapter 19 Oxidative Phosphorylation and Photophosphorylation744 (b) 8885d_c19_690-750 3/1/04 11:32 AM Page 744 mac76 mac76:385_reb: Chapter 19 Further Reading 745 SUMMARY 19.9 ATP Synthesis by Photophosphorylation ■ In plants, both the water-splitting reaction and electron flow through the cytochrome b 6 f complex are accompanied by proton pumping across the thylakoid membrane. The proton-motive force thus created drives ATP synthesis by a CF o CF 1 complex similar to the mitochondrial F o F 1 complex. ■ Flow of electrons through the photosystems produces NADPH and ATP, in the ratio of about 2:3. A second type of electron flow (cyclic flow) produces ATP only and allows variability in the proportions of NADPH and ATP formed. ■ The localization of PSI and PSII between the granal and stromal lamellae can change and is indirectly controlled by light intensity, optimizing the distribution of excitons between PSI and PSII for efficient energy capture. ■ Chloroplasts, like mitochondria, evolved from bacteria living endosymbiotically within early eukaryotic cells. The ATP synthases of eubacteria, cyanobacteria, mitochondria, and chloroplasts share a common evolutionary precursor and a common enzymatic mechanism. ■ Many photosynthetic microorganisms obtain electrons for photosynthesis not from water but from donors such as H 2 S. Key Terms chemiosmotic theory 690 nicotinamide nucleotide–linked dehydrogenases 692 flavoprotein 692 reducing equivalent 693 ubiquinone (coenzyme Q, Q) 693 cytochromes 693 iron-sulfur protein 693 Complex I 696 vectorial metabolism 697 Complex II 698 Complex III 699 cytochrome bc 1 complex 699 Q cycle 700 Complex IV 700 cytochrome oxidase 700 proton-motive force 703 ATP synthase 704 F 1 ATPase 708 rotational catalysis 711 P/O ratio 712 P/2e H11002 ratio 712 acceptor control 716 mass-action ratio 716 light-dependent reactions 723 light reactions 723 carbon-assimilation reactions 723 carbon-fixation reaction 723 thylakoid 724 stroma 724 exciton transfer 725 chlorophylls 725 light-harvesting complexes (LHCs) 725 accessory pigments 728 photosystem 728 photochemical reaction center 728 light-harvesting (antenna) molecules 728 photosystem II (PSII) 733 photosystem I (PSI) 733 oxygenic photosynthesis 734 oxygen-evolving complex (water- splitting complex) 738 photophosphorylation 740 cyclic electron flow 741 noncyclic electron flow 741 cyclic photophosphorylation 742 Terms in bold are defined in the glossary. Further Reading History and General Background Arnon, D.I. (1984) The discovery of photosynthetic phosphoryla- tion. Trends Biochem. Sci. 9, 258–262. Beinert, H. (1995) These are the moments when we live! From Thunberg tubes and manometry to phone, fax and FedEx. In Selected Topics in the History of Biochemistry: Personal Recollections, Comprehensive Biochemistry, Vol. 38, Elsevier Science Publishing Co., Inc., New York. An engaging personal account of the exciting period when the biochemistry of respiratory electron transfer was worked out. Blankenship, R.E. (2002) Molecular Mechanisms of Photosyn- thesis, Blackwell Science Inc., London. An intermediate-level discussion of all aspects of photosynthesis. Gray, M.W., Berger, G., & Lang, B.F. (1999) Mitochondrial evo- lution. Science 283, 1476–1481. Compact review of the endosymbiotic origin hypothesis and the evidence for and against it. Harold, F.M. (1986) The Vital Force: A Study in Bioenergetics, W. H. Freeman and Company, New York. A very readable synthesis of the principles of bioenergetics and their application to energy transductions. Heldt, H.-W. (1997) Plant Biochemistry and Molecular Biology, Oxford University Press, Oxford. A textbook of plant biochemistry with excellent discussions of photophosphorylation. Keilin, D. (1966) The History of Cell Respiration and Cy- tochrome, Cambridge University Press, London. An authoritative and absorbing account of the discovery of cy- tochromes and their roles in respiration, written by the man who discovered cytochromes. 8885d_c19_690-750 3/1/04 11:32 AM Page 745 mac76 mac76:385_reb: Chapter 19 Oxidative Phosphorylation and Photophosphorylation746 Mitchell, P. (1979) Keilin’s respiratory chain concept and its chemiosmotic consequences. Science 206, 1148–1159. Mitchell’s Nobel lecture, outlining the evolution of the chemios- motic hypothesis. Nicholls, D.G. & Ferguson, S.J. (2002) Bioenergetics 3, Acade- mic Press, Amsterdam. Up-to-date, comprehensive, well-illustrated treatment of all as- pects of mitochondrial and chloroplast energy transductions. A comprehensive, advanced treatise. Scheffler, I.E. (1999) Mitochondria, Wiley-Liss, New York. An excellent survey of mitochondrial structure and function. Slater, E.C. (1987) The mechanism of the conservation of energy of biological oxidations. Eur. J. Biochem. 166, 489–504. A clear and critical account of the evolution of the chemiosmotic model. OXIDATIVE PHOSPHORYLATION Respiratory Electron Flow Babcock, G.T. & Wickstr?m, M. (1992) Oxygen activation and the conservation of energy in cell respiration. Nature 356, 301–309. Advanced discussion of the reduction of water and pumping of protons by cytochrome oxidase. Brandt, U. (1997) Proton-translocation by membrane-bound NADH:ubiquinone-oxidoreductase (complex I) through redox- gated ligand conduction. Biochim. Biophys. Acta 1318, 79–91. Advanced discussion of models for electron movement through Complex I. Brandt, U. & Trumpower, B. (1994) The protonmotive Q cycle in mitochondria and bacteria. Crit. Rev. Biochem. Mol. Biol. 29, 165–197. Crofts, A.R. & Berry, E.A. (1998) Structure and function of the cytochrome bc 1 complex of mitochondria and photosynthetic bac- teria. Curr. Opin. Struct. Biol. 8, 501–509. Michel, H., Behr, J., Harrenga, A., & Kannt, A. (1998) Cy- tochrome c oxidase: structure and spectroscopy. Annu. Rev. Bio- phys. Biomol. Struct. 27, 329–356. Advanced review of Complex IV structure and function. Rottenberg, H. (1998) The generation of proton electrochemical potential gradient by cytochrome c oxidase. Biochim. Biophys. Acta 1364, 1–16. Tielens, A.G.M., Rotte, C., van Hellemond, J.J., & Martin, W. (2002) Mitochondria as we don’t know them. Trends Biochem. Sci. 27, 564–572. Intermediate-level discussion of the many organisms in which mitochondria do not depend on oxygen as the final electron donor. Tsukihara, T., Aoyama, H., Yamashita, E., Tomizaki, T., Yamaguchi, H., Shinzawa-Itoh, K., Nakashima, R., Yaono, R., & Yoshikawa, S. (1996) The whole structure of the 13-subunit oxidized cytochrome c oxidase at 2.8 ?. Science 272, 1136–1144. The solution by x-ray crystallography of the structure of this huge membrane protein. Xia, D., Yu, C.-A., Kim, H., Xia, J.-Z., Kachurin, A.M., Zhang, L., Yu, L., & Deisenhofer, J. (1997) Crystal structure of the cy- tochrome bc 1 complex from bovine heart mitochondria. Science 277, 60–66. Report revealing the crystallographic structure of Complex III. Yankovskaya, V., Horsefield, R., T?rnroth, S., Luna-Chavez, C., Myoshi, H., Léger, C., Byrne, B., Cecchini, G., & Iwata, S. (2003) Architecture of succinate dehydrogenase and reactive oxy- gen species generation. Science 299, 700–704. Coupling ATP Synthesis to Respiratory Electron Flow Abrahams, J.P., Leslie, A.G.W., Lutter, R., & Walker, J.E. (1994) The structure of F 1 -ATPase from bovine heart mitochondria determined at 2.8 ? resolution. Nature 370, 621–628. Bianchet, M.A., Hullihen, J., Pedersen, P.L., & Amzel, L.M. (1998) The 2.80 ? structure of rat liver F 1 -ATPase: configuration of a critical intermediate in ATP synthesis-hydrolysis. Proc. Natl. Acad. Sci. USA 95, 11,065–11,070. Research paper that provided important structural detail in support of the catalytic mechanism. Boyer, P.D. (1997) The ATP synthase—a splendid molecular ma- chine. Annu. Rev. Biochem. 66, 717–749. An account of the historical development and current state of the binding-change model, written by its principal architect. Cabezón, E., Montgomery, M.G., Leslie, A.G.W., & Walker, J.E. (2003) The structure of bovine F 1 -ATPase in complex with its regulatory protein IF 1 Nat. Struct. Biol. 10, 744–750. Hinkle, P.C., Kumar, M.A., Resetar, A., & Harris, D.L. (1991) Mechanistic stoichiometry of mitochondrial oxidative phosphoryla- tion. Biochemistry 30, 3576–3582. A careful analysis of experimental results and theoretical con- siderations on the question of nonintegral P/O ratios. Khan, S. (1997) Rotary chemiosmotic machines. Biochim. Bio- phys. Acta 1322, 86–105. Detailed review of the structures that underlie proton-driven rotary motion of ATP synthase and bacterial flagella. Sambongi, Y., Iko, Y., Tanabe, M., Omote, H., Iwamoto- Kihara, A., Ueda, I., Yanagida, T., Wada, Y., & Futai, M. (1999) Mechanical rotation of the c subunit oligomer in ATP synthase (F o F 1 ): direct observation. Science 286, 1722–1724. The experimental evidence for rotation of the entire cylinder of c subunits in F o F 1 . Stock, D., Leslie, A.G.W., & Walker, J.E. (1999) Molecular architecture of the rotary motor in ATP synthase. Science 286, 1700–1705. The first crystallographic view of the F o subunit, in the yeast F o F 1 . See also R. H. Fillingame’s editorial comment in the same issue of Science. Weber, J. & Senior, A.E. (1997) Catalytic mechanism of F 1 - ATPase. Biochim. Biophys. Acta 1319, 19–58. An advanced review of kinetic, structural, and biochemical evi- dence for the ATP synthase mechanism. Yasuda, R., Noji, H., Kinosita, K., Jr., & Yoshida, M. (1998) F 1 -ATPase is a highly efficient molecular motor that rotates with discrete 120H11034 steps. Cell 93, 1117–1124. Graphical demonstration of the rotation of ATP synthase. Regulation of Mitochondrial Oxidative Phosphorylation Brand, M.D. & Murphy, M.P. (1987) Control of electron flux through the respiratory chain in mitochondria and cells. Biol. Rev. Camb. Philos. Soc. 62, 141–193. An advanced description of respiratory control. 8885d_c19_690-750 3/1/04 11:32 AM Page 746 mac76 mac76:385_reb: Chapter 19 Further Reading 747 Harris, D.A. & Das, A.M. (1991) Control of mitochondrial ATP synthesis in the heart. Biochem. J. 280, 561–573. Advanced discussion of the regulation of ATP synthase by Ca 2H11001 and other factors. Klingenberg, M. & Huang, S.-G. (1999) Structure and function of the uncoupling protein from brown adipose tissue. Biochim. Biophys. Acta 1415, 271–296. Apoptosis and Mitochondrial Diseases Kroemer, G. (2003) Mitochondrial control of apoptosis: an intro- duction. Biochem. Biophys. Res. Commun. 304, 433–435. A short introduction to a collection of excellent papers in this journal issue. McCord, J.M. (2002) Superoxide dismutase in aging and disease: an overview. Meth. Enzymol. 349, 331–341. Newmeyer, D.D. & Ferguson-Miller, S. (2003) Mitochondria: re- leasing power for life and unleashing the machineries of death. Cell 112, 481–490. Intermediate-level review of the possible roles of mitochondria in apoptosis. Schapira, A.H.V. (2002) Primary and secondary defects of the mi- tochondrial respiratory chain. J. Inher. Metab. Dis. 25, 207–214. Wallace, D.C. (1999) Mitochondrial disease in man and mouse. Science 283, 1482–1487. PHOTOSYNTHESIS Harvesting Light Energy Cogdell, R.J., Isaacs, N.W., Howard, T.D., McLuskey, K., Fraser, N.J., & Prince, S.M. (1999) How photosynthetic bacteria harvest solar energy. J. Bacteriol. 181, 3869–3879. A short, intermediate-level review of the structure and function of the light-harvesting complex of the purple bacteria and exci- ton flow to the reaction center. Green, B.R., Pichersky, E., & Kloppstech, K. (1991) Chloro- phyll a/b-binding proteins: an extended family. Trends Biochem. Sci. 16, 181–186. An intermediate-level description of the proteins that orient chlorophyll molecules in chloroplasts. Kargul, J., Nield, J., & Barber, J. (2003) Three-dimensional re- construction of a light-harvesting Complex I–Photosystem I (LHCI- PSI) supercomplex from the green alga Chlamydomonas rein- hardtii. J. Biol. Chem. 278, 16,135–16,141. Zuber, H. (1986) Structure of light-harvesting antenna complexes of photosynthetic bacteria, cyanobacteria and red algae. Trends Biochem. Sci. 11, 414–419. Light-Driven Electron Flow Barber, J. (2002) Photosystem II: a multisubunit membrane pro- tein that oxidizes water. Curr. Opin. Struct. Biol. 12, 523–530. A short, intermediate-level summary of the structure of PSII. Barber, J. & Anderson, J.M. (eds) (2002) Photosystem II: mo- lecular structure and function. Philos. Trans. R. Soc. (Biol. Sci.) 357, 1321–1512. A collection of 16 papers on photosystem II. Chitnis, P.R. (2001) Photosystem I: function and physiology. Annu. Rev. Plant Physiol. Plant Mol. Biol. 52, 593–626. An advanced and lengthy review. Deisenhofer, J. & Michel, H. (1991) Structures of bacterial pho- tosynthetic reaction centers. Annu. Rev. Cell Biol. 7, 1–23. Description of the structure of the reaction center of purple bacteria and implications for the function of bacterial and plant reaction centers. Heathcote, P., Fyfe, P.K., & Jones, M.R. (2002) Reaction cen- tres: the structure and evolution of biological solar power. Trends. Biochem Sci. 27, 79–87. Intermediate-level review of photosystems I and II. Huber, R. (1990) A structural basis of light energy and electron transfer in biology. Eur. J. Biochem. 187, 283–305. Huber’s Nobel lecture, describing the physics and chemistry of phototransductions; an exceptionally clear and well-illustrated discussion, based on crystallographic studies of reaction centers. Jagendorf, A.T. (1967) Acid-base transitions and phosphorylation by chloroplasts. Fed. Proc. 26, 1361–1369. Report of the classic experiment establishing the ability of a proton gradient to drive ATP synthesis in the dark. Jordan, P., Fromme, P., Witt, H.T., Klukas, O., Saenger, W., & Krauss, N. (2001) Three-dimensional structure of cyanobacterial photosystem I at 2.5 ?. Nature 411, 909–917. Kamiya, N. & Shen, J.-R. (2003) Crystal structure of oxygen- evolving photosystem II from Thermosynechococcus vulcanus at 3.7 ? resolution. Proc. Natl. Acad. Sci. USA 100, 98–103. Kargul, J., Nield, J., & Barber, J. (2003) Three-dimensional re- construction of a light-harvesting complex I–photosystem I (LHCI-PSI) supercomplex from the green alga Chlamydomonas reinhardtii: insights into light harvesting for PSI. J. Biol. Chem. 278, 16,135–16,141. Leister, D. (2003) Chloroplast research in the genomic age. Trends Genet. 19, 47–56. Zouni, A., Witt, H.-T., Kern, J., Fromme, P., Krauss, N., Saenger, W., & Orth, P. (2001) Crystal structure of photosystem II from Synechococcus elongates at 3.8 ? resolution. Nature 409, 739–743. Water-Splitting Complex R?gner, M., Boekema, E.J., & Barber, J. (1996) How does pho- tosystem 2 split water? The structural basis of efficient energy conversion. Trends Biochem. Sci. 21, 44–49. Szalai, V.A. & Brudvig, G.W. (1998) How plants produce dioxy- gen. Am. Sci. 86, 542–551. A well-illustrated introduction to the oxygen-evolving complex of plants. Bacteriorhodopsin Luecke, H. (2000) Atomic resolution structures of bacteriorhodopsin photocycle intermediates: the role of discrete water molecules in the function of this light-driven ion pump. Biochim. Biophys. Acta 1460, 133–156. Advanced review of a proton pump that employs an internal chain of water molecules. Luecke, H., Schobert, B., Richter, H.-T., Cartailler, J.-P., & Lanyi, J.K. (1999) Structural changes in bacteriorhodopsin during ion transport at 2 angstrom resolution. Science 286, 255–264. This paper, accompanied by an editorial comment in the same Science issue, describes the model for H H11001 translocation by pro- ton hopping. 8885d_c19_690-750 3/1/04 11:32 AM Page 747 mac76 mac76:385_reb: Chapter 19 Oxidative Phosphorylation and Photophosphorylation748 Problems 1. Oxidation-Reduction Reactions The NADH dehy- drogenase complex of the mitochondrial respiratory chain pro- motes the following series of oxidation-reduction reactions, in which Fe 3H11001 and Fe 2H11001 represent the iron in iron-sulfur cen- ters, Q is ubiquinone, QH 2 is ubiquinol, and E is the enzyme: (1) NADH H11001 H H11001 H11001 E-FMN 88n NAD H11001 H11001 E-FMNH 2 (2) E-FMNH 2 H11001 2Fe 3H11001 88n E-FMN H11001 2Fe 2H11001 H11001 2H H11001 (3) 2Fe 2H11001 H11001 2H H11001 H11001 Q 88n 2Fe 3H11001 H11001 QH 2 Sum: NADH H11001 H H11001 H11001 Q 88n NAD H11001 H11001 QH 2 For each of the three reactions catalyzed by the NADH de- hydrogenase complex, identify (a) the electron donor, (b) the electron acceptor, (c) the conjugate redox pair, (d) the re- ducing agent, and (e) the oxidizing agent. 2. All Parts of Ubiquinone Have a Function In elec- tron transfer, only the quinone portion of ubiquinone under- goes oxidation-reduction; the isoprenoid side chain remains unchanged. What is the function of this chain? 3. Use of FAD Rather Than NAD H11545 in Succinate Oxi- dation All the dehydrogenases of glycolysis and the citric acid cycle use NAD H11001 (EH11032H11034 for NAD H11001 /NADH is H110020.32 V) as electron acceptor except succinate dehydrogenase, which uses covalently bound FAD (EH11032H11034 for FAD/FADH 2 in this en- zyme is 0.050 V). Suggest why FAD is a more appropriate elec- tron acceptor than NAD H11001 in the dehydrogenation of succinate, based on the EH11032H11034 values of fumarate/succinate (EH11032H11034 H11005 0.031), NAD H11001 /NADH, and the succinate dehydrogenase FAD/FADH 2 . 4. Degree of Reduction of Electron Carriers in the Respiratory Chain The degree of reduction of each carrier in the respiratory chain is determined by conditions in the mi- tochondrion. For example, when NADH and O 2 are abundant, the steady-state degree of reduction of the carriers decreases as electrons pass from the substrate to O 2 . When electron transfer is blocked, the carriers before the block become more reduced and those beyond the block become more oxidized (see Fig. 19–6). For each of the conditions below, predict the state of oxidation of ubiquinone and cytochromes b, c 1 , c, and a H11001 a 3 . (a) Abundant NADH and O 2 , but cyanide added (b) Abundant NADH, but O 2 exhausted (c) Abundant O 2 , but NADH exhausted (d) Abundant NADH and O 2 5. Effect of Rotenone and Antimycin A on Electron Transfer Rotenone, a toxic natural product from plants, strongly inhibits NADH dehydrogenase of insect and fish mi- tochondria. Antimycin A, a toxic antibiotic, strongly inhibits the oxidation of ubiquinol. (a) Explain why rotenone ingestion is lethal to some in- sect and fish species. (b) Explain why antimycin A is a poison. (c) Given that rotenone and antimycin A are equally effective in blocking their respective sites in the electron- transfer chain, which would be a more potent poison? Explain. 6. Uncouplers of Oxidative Phosphorylation In normal mitochondria the rate of electron transfer is tightly coupled to the demand for ATP. When the rate of use of ATP is rela- tively low, the rate of electron transfer is low; when demand for ATP increases, electron-transfer rate increases. Under these conditions of tight coupling, the number of ATP mole- cules produced per atom of oxygen consumed when NADH is the electron donor—the P/O ratio—is about 2.5. (a) Predict the effect of a relatively low and a relatively high concentration of uncoupling agent on the rate of elec- tron transfer and the P/O ratio. (b) Ingestion of uncouplers causes profuse sweating and an increase in body temperature. Explain this phenomenon in molecular terms. What happens to the P/O ratio in the pres- ence of uncouplers? (c) The uncoupler 2,4-dinitrophenol was once prescribed as a weight-reducing drug. How could this agent, in principle, serve as a weight-reducing aid? Uncoupling agents are no longer prescribed, because some deaths occurred following their use. Why might the ingestion of uncouplers lead to death? 7. Effects of Valinomycin on Oxidative Phosphoryla- tion When the antibiotic valinomycin is added to actively respiring mitochondria, several things happen: the yield of ATP decreases, the rate of O 2 consumption increases, heat is released, and the pH gradient across the inner mitochon- drial membrane increases. Does valinomycin act as an uncou- pler or an inhibitor of oxidative phosphorylation? Explain the experimental observations in terms of the antibiotic’s ability to transfer K H11001 ions across the inner mitochondrial membrane. 8. Mode of Action of Dicyclohexylcarbodiimide (DCCD) When DCCD is added to a suspension of tightly coupled, actively respiring mitochondria, the rate of electron transfer (measured by O 2 consumption) and the rate of ATP production dramatically decrease. If a solution of 2,4-dinitro- phenol is now added to the preparation, O 2 consumption re- turns to normal but ATP production remains inhibited. (a) What process in electron transfer or oxidative phos- phorylation is affected by DCCD? (b) Why does DCCD affect the O 2 consumption of mi- tochondria? Explain the effect of 2,4-dinitrophenol on the in- hibited mitochondrial preparation. (c) Which of the following inhibitors does DCCD most resemble in its action: antimycin A, rotenone, or oligomycin? 9. Compartmentalization of Citric Acid Cycle Compo- nents Isocitrate dehydrogenase is found only in the mito- chondrion, but malate dehydrogenase is found in both the cytosol and mitochondrion. What is the role of cytosolic malate dehydrogenase? 10. The Malate–H9251-Ketoglutarate Transport System The transport system that conveys malate and H9251-ketoglu- tarate across the inner mitochondrial membrane (see Fig. 19–27) is inhibited by n-butylmalonate. Suppose n-butyl- malonate is added to an aerobic suspension of kidney cells using glucose exclusively as fuel. Predict the effect of this in- hibitor on (a) glycolysis, (b) oxygen consumption, (c) lactate formation, and (d) ATP synthesis. 11. Cellular ADP Concentration Controls ATP Forma- tion Although both ADP and P i are required for the syn- 8885d_c19_690-750 3/1/04 11:32 AM Page 748 mac76 mac76:385_reb: 749Chapter 19 Problems thesis of ATP, the rate of synthesis depends mainly on the concentration of ADP, not P i . Why? 12. The Pasteur Effect When O 2 is added to an anaero- bic suspension of cells consuming glucose at a high rate, the rate of glucose consumption declines greatly as the O 2 is used up, and accumulation of lactate ceases. This effect, first ob- served by Louis Pasteur in the 1860s, is characteristic of most cells capable of both aerobic and anaerobic glucose catabolism. (a) Why does the accumulation of lactate cease after O 2 is added? (b) Why does the presence of O 2 decrease the rate of glucose consumption? (c) How does the onset of O 2 consumption slow down the rate of glucose consumption? Explain in terms of specific enzymes. 13. Respiration-Deficient Yeast Mutants and Ethanol Production Respiration-deficient yeast mutants (p H11002 ; “pe- tites”) can be produced from wild-type parents by treatment with mutagenic agents. The mutants lack cytochrome oxi- dase, a deficit that markedly affects their metabolic behavior. One striking effect is that fermentation is not suppressed by O 2 —that is, the mutants lack the Pasteur effect (see Prob- lem 12). Some companies are very interested in using these mutants to ferment wood chips to ethanol for energy use. Ex- plain the advantages of using these mutants rather than wild- type yeast for large-scale ethanol production. Why does the absence of cytochrome oxidase eliminate the Pasteur effect? 14. How Many Protons in a Mitochondrion? Electron transfer translocates protons from the mitochondrial matrix to the external medium, establishing a pH gradient across the inner membrane (outside more acidic than inside). The ten- dency of protons to diffuse back into the matrix is the driving force for ATP synthesis by ATP synthase. During oxidative phosphorylation by a suspension of mitochondria in a medium of pH 7.4, the pH of the matrix has been measured as 7.7. (a) Calculate [H H11001 ] in the external medium and in the matrix under these conditions. (b) What is the outside-to-inside ratio of [H H11001 ]? Comment on the energy inherent in this concentration difference. (Hint: See Eqn 11–3, p. 398.) (c) Calculate the number of protons in a respiring liver mitochondrion, assuming its inner matrix compartment is a sphere of diameter 1.5 H9262m. (d) From these data, is the pH gradient alone sufficient to generate ATP? (e) If not, suggest how the necessary energy for syn- thesis of ATP arises. 15. Rate of ATP Turnover in Rat Heart Muscle Rat heart muscle operating aerobically fills more than 90% of its ATP needs by oxidative phosphorylation. Each gram of tis- sue consumes O 2 at the rate of 10.0 H9262mol/min, with glucose as the fuel source. (a) Calculate the rate at which the heart muscle con- sumes glucose and produces ATP. (b) For a steady-state concentration of ATP of 5.0 H9262mol/g of heart muscle tissue, calculate the time required (in sec- onds) to completely turn over the cellular pool of ATP. What does this result indicate about the need for tight regulation of ATP production? (Note: Concentrations are expressed as micromoles per gram of muscle tissue because the tissue is mostly water.) 16. Rate of ATP Breakdown in Flight Muscle ATP pro- duction in the flight muscle of the fly Lucilia sericata results almost exclusively from oxidative phosphorylation. During flight, 187 ml of O 2 /hrjg of body weight is needed to main- tain an ATP concentration of 7.0 H9262mol/g of flight muscle. Assuming that flight muscle makes up 20% of the weight of the fly, calculate the rate at which the flight-muscle ATP pool turns over. How long would the reservoir of ATP last in the absence of oxidative phosphorylation? Assume that reducing equivalents are transferred by the glycerol 3-phosphate shut- tle and that O 2 is at 25 H11034C and 101.3 kPa (1 atm). 17. Transmembrane Movement of Reducing Equiva- lents Under aerobic conditions, extramitochondrial NADH must be oxidized by the mitochondrial electron-transfer chain. Consider a preparation of rat hepatocytes containing mitochondria and all the cytosolic enzymes. If [4- 3 H]NADH is introduced, radioactivity soon appears in the mitochondrial matrix. However, if [7- 14 C]NADH is introduced, no radioac- tivity appears in the matrix. What do these observations re- veal about the oxidation of extramitochondrial NADH by the electron-transfer chain? 18. NAD Pools and Dehydrogenase Activities Although both pyruvate dehydrogenase and glyceraldehyde 3-phosphate dehydrogenase use NAD H11001 as their electron acceptor, the two enzymes do not compete for the same cellular NAD pool. Why? 19. Photochemical Efficiency of Light at Different Wavelengths The rate of photosynthesis, measured by O 2 production, is higher when a green plant is illuminated with light of wavelength 680 nm than with light of 700 nm. How- ever, illumination by a combination of light of 680 nm and 700 nm gives a higher rate of photosynthesis than light of ei- ther wavelength alone. Explain. 20. Balance Sheet for Photosynthesis In 1804 Theodore de Saussure observed that the total weights of oxygen and dry organic matter produced by plants is greater than the weight of carbon dioxide consumed during photosynthesis. Where does the extra weight come from? 21. Role of H 2 S in Some Photosynthetic Bacteria Il- luminated purple sulfur bacteria carry out photosynthesis in the presence of H 2 O and 14 CO 2 , but only if H 2 S is added and O 2 is absent. During the course of photosynthesis, measured by formation of [ 14 C]carbohydrate, H 2 S is converted to ele- mental sulfur, but no O 2 is evolved. What is the role of the conversion of H 2 S to sulfur? Why is no O 2 evolved? 22. Boosting the Reducing Power of Photosystem I by Light Absorption When photosystem I absorbs red light at 700 nm, the standard reduction potential of P700 changes O 14 C H N NH 2 R [7- 14 C]NADH O 3 H C N 3 H NH 2 R [4- 3 H]NADH H 8885d_c19_690-750 3/1/04 11:32 AM Page 749 mac76 mac76:385_reb: Chapter 19 Oxidative Phosphorylation and Photophosphorylation750 from 0.40 V to about H110021.2 V. What fraction of the absorbed light is trapped in the form of reducing power? 23. Limited ATP Synthesis in the Dark In a laboratory experiment, spinach chloroplasts are illuminated in the ab- sence of ADP and P i , then the light is turned off and ADP and P i are added. ATP is synthesized for a short time in the dark. Explain this finding. 24. Mode of Action of the Herbicide DCMU When chloroplasts are treated with 3-(3,4-dichlorophenyl)-1,1- dimethylurea (DCMU, or diuron), a potent herbicide, O 2 evo- lution and photophosphorylation cease. Oxygen evolution, but not photophosphorylation, can be restored by addition of an external electron acceptor, or Hill reagent. How does DCMU act as a weed killer? Suggest a location for the inhibitory ac- tion of this herbicide in the scheme shown in Figure 19–49. Explain. 25. Bioenergetics of Photophosphorylation The steady- state concentrations of ATP, ADP, and P i in isolated spinach chloroplasts under full illumination at pH 7.0 are 120.0, 6.0, and 700.0 H9262m, respectively. (a) What is the free-energy requirement for the syn- thesis of 1 mol of ATP under these conditions? (b) The energy for ATP synthesis is furnished by light- induced electron transfer in the chloroplasts. What is the min- imum voltage drop necessary (during transfer of a pair of electrons) to synthesize ATP under these conditions? (You may need to refer to Eqn 13–6, p. 510.) 26. Light Energy for a Redox Reaction Suppose you have isolated a new photosynthetic microorganism that oxi- dizes H 2 S and passes the electrons to NAD H11001 . What wavelength of light would provide enough energy for H 2 S to reduce NAD H11001 under standard conditions? Assume 100% efficiency in the photochemical event, and use EH11032H11034 of H11002243 mV for H 2 S and H11002320 mV for NAD H11001 . See Figure 19–39 for energy equivalents of wavelengths of light. 27. Equilibrium Constant for Water-Splitting Reac- tions The coenzyme NADP H11001 is the terminal electron ac- ceptor in chloroplasts, according to the reaction 2H 2 O H11001 2NADP H11001 On 2NADPH H11001 2H H11001 H11001 O 2 Use the information in Table 19–2 to calculate the equilib- rium constant for this reaction at 25 H11034C. (The relationship between KH11032 eq and H9004GH11032H11034 is discussed on p. 492.) How can the chloroplast overcome this unfavorable equilibrium? 28. Energetics of Phototransduction During photosyn- thesis, eight photons must be absorbed (four by each photo- system) for every O 2 molecule produced: 2H 2 O H11001 2NADP H11001 H11001 8 photons 88n 2NADPH H11001 2H H11001 H11001 O 2 Assuming that these photons have a wavelength of 700 nm (red) and that the absorption and use of light energy are 100% efficient, calculate the free-energy change for the process. 29. Electron Transfer to a Hill Reagent Isolated spin- ach chloroplasts evolve O 2 when illuminated in the presence of potassium ferricyanide (a Hill reagent), according to the equation 2H 2 O H11001 4Fe 3H11001 88n O 2 H11001 4H H11001 H11001 4Fe 2H11001 where Fe 3H11001 represents ferricyanide and Fe 2H11001 , ferrocyanide. Is NADPH produced in this process? Explain. 30. How Often Does a Chlorophyll Molecule Absorb a Photon? The amount of chlorophyll a (M r 892) in a spinach leaf is about 20 H9262g/cm 2 of leaf. In noonday sunlight (average energy 5.4 J/cm 2 jmin), the leaf absorbs about 50% of the ra- diation. How often does a single chlorophyll molecule absorb a photon? Given that the average lifetime of an excited chloro- phyll molecule in vivo is 1 ns, what fraction of the chlorophyll molecules is excited at any one time? 31. Effect of Monochromatic Light on Electron Flow The extent to which an electron carrier is oxidized or reduced during photosynthetic electron transfer can sometimes be ob- served directly with a spectrophotometer. When chloroplasts are illuminated with 700 nm light, cytochrome f, plastocyanin, and plastoquinone are oxidized. When chloroplasts are illu- minated with 680 nm light, however, these electron carriers are reduced. Explain. 32. Function of Cyclic Photophosphorylation When the [NADPH]/[NADP H11001 ] ratio in chloroplasts is high, photo- phosphorylation is predominantly cyclic (see Fig. 19–49). Is O 2 evolved during cyclic photophosphorylation? Is NADPH produced? Explain. What is the main function of cyclic photo- phosphorylation? 8885d_c19_690-750 3/1/04 11:32 AM Page 750 mac76 mac76:385_reb: chapter W e have now reached a turning point in our study of cellular metabolism. Thus far in Part II we have de- scribed how the major metabolic fuels—carbohydrates, fatty acids, and amino acids—are degraded through con- verging catabolic pathways to enter the citric acid cy- cle and yield their electrons to the respiratory chain, and how this exergonic flow of electrons to oxygen is coupled to the endergonic synthesis of ATP. We now turn to anabolic pathways, which use chemical energy in the form of ATP and NADH or NADPH to synthesize cellular components from simple precursor molecules. Anabolic pathways are generally reductive rather than oxidative. Catabolism and anabolism proceed simulta- neously in a dynamic steady state, so the energy- yielding degradation of cellular components is counter- balanced by biosynthetic processes, which create and maintain the intricate orderliness of living cells. Plants must be especially versatile in their handling of carbohydrates, for several reasons. First, plants are autotrophs, able to convert inorganic carbon (as CO 2 ) into organic compounds. Second, biosynthesis occurs primarily in plastids, membrane-bounded organelles unique to plants, and the movement of intermediates be- tween cellular compartments is an important aspect of metabolism. Third, plants are not motile: they cannot move to find better supplies of water, sunlight, or nutri- ents. They must have sufficient metabolic flexibility to allow them to adapt to changing conditions in the place where they are rooted. Finally, plants have thick cell walls made of carbohydrate polymers, which must be assem- bled outside the plasma membrane and which constitute a significant proportion of the cell’s carbohydrate. The chapter begins with a description of the process by which CO 2 is assimilated into trioses and hexoses, then considers photorespiration, an important side re- action during CO 2 fixation, and the ways in which cer- tain plants avoid this side reaction. We then look at how the biosynthesis of sucrose (for sugar transport) and starch (for energy storage) is accomplished by mech- anisms analogous to those employed by animal cells to make glycogen. The next topic is the synthesis of the cellulose of plant cell walls and the peptidoglycan of bac- terial cell walls, illustrating the problems of energy- dependent biosynthesis outside the plasma membrane. Finally, we discuss how the various pathways that share pools of common intermediates are segregated within organelles yet integrated with one another. 20.1 Photosynthetic Carbohydrate Synthesis The synthesis of carbohydrates in animal cells always employs precursors having at least three carbons, all of which are less oxidized than the carbon in CO 2 . Plants 20 751 CARBOHYDRATE BIOSYNTHESIS IN PLANTS AND BACTERIA 20.1 Photosynthetic Carbohydrate Synthesis 751 20.2 Photorespiration and the C 4 and CAM Pathways 766 20.3 Biosynthesis of Starch and Sucrose 771 20.4 Synthesis of Cell Wall Polysaccharides: Plant Cellulose and Bacterial Peptidoglycan 775 20.5 Integration of Carbohydrate Metabolism in the Plant Cell 780 . . . the discovery of the long-lived isotope of carbon, carbon-14, by Samuel Ruben and Martin Kamen in 1940 provided the ideal tool for the tracing of the route along which carbon dioxide travels on its way to carbohydrate. —Melvin Calvin, Nobel Address, 1961 8885d_c20_751–786 2/18/04 1:56 PM Page 751 mac76 mac76:385_reb: and photosynthetic microorganisms, by contrast, can synthesize carbohydrates from CO 2 and water, reducing CO 2 at the expense of the energy and reducing power furnished by the ATP and NADPH that are generated by the light-dependent reactions of photosynthesis (Fig. 20–1). Plants (and other autotrophs) can use CO 2 as the sole source of the carbon atoms required for the biosyn- thesis of cellulose and starch, lipids and proteins, and the many other organic components of plant cells. By contrast, heterotrophs cannot bring about the net re- duction of CO 2 to achieve a net synthesis of glucose. Green plants contain in their chloroplasts unique enzymatic machinery that catalyzes the conversion of CO 2 to simple (reduced) organic compounds, a process called CO 2 assimilation. This process has also been called CO 2 fixation or carbon fixation, but we re- serve these terms for the specific reaction in which CO 2 is incorporated (fixed) into a three-carbon organic com- pound, the triose phosphate 3- phosphoglycerate. This simple product of photosynthesis is the precursor of more com- plex biomolecules, including sugars, polysaccharides, and the metabolites derived from them, all of which are synthe- sized by metabolic pathways similar to those of animal tis- sues. Carbon dioxide is assim- ilated via a cyclic pathway, its key intermediates constantly regenerated. The pathway was elucidated in the early 1950s by Melvin Calvin, Andrew Benson, and James A. Bassham, and is often called the Calvin cycle or, more descriptively, the photosynthetic carbon reduction cycle. Carbohydrate metabolism is more complex in plant cells than in animal cells or in nonphotosynthetic mi- croorganisms. In addition to the universal pathways of glycolysis and gluconeogenesis, plants have the unique reaction sequences for reduction of CO 2 to triose phos- phates and the associated reductive pentose phosphate pathway—all of which must be coordinately regulated to ensure proper allocation of carbon to energy pro- duction and synthesis of starch and sucrose. Key en- zymes are regulated, as we shall see, by (1) reduction of disulfide bonds by electrons flowing from photosys- tem I and (2) changes in pH and Mg 2H11001 concentration that result from illumination. When we look at other as- pects of plant carbohydrate metabolism, we also find en- zymes that are modulated by (3) conventional allosteric regulation by one or more metabolic intermediates and (4) covalent modification (phosphorylation). Plastids Are Organelles Unique to Plant Cells and Algae Most of the biosynthetic activities in plants (including CO 2 assimilation) occur in plastids, a family of self- reproducing organelles bounded by a double membrane and containing a small genome that encodes some of their proteins. Most proteins destined for plastids are encoded in nuclear genes, which are transcribed and translated like other nuclear genes; then the proteins are imported into plastids. Plastids reproduce by binary fission, replicating their genome (a single circular DNA molecule) and using their own enzymes and ribosomes to synthesize the proteins encoded by that genome. Chloroplasts (see Fig. 19–38) are the sites of CO 2 as- similation. The enzymes for this process are contained in the stroma, the soluble phase bounded by the inner chloroplast membrane. Amyloplasts are colorless plastids (that is, they lack chlorophyll and other pig- Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria752 Sucrose (transport) Starch (storage) Cellulose (cell wall) Hexose phosphates Pentose phosphates Metabolic intermediates DNA RNA Protein Lipid Triose phosphates ADP, NADP H11001 ATP, NADPH CO 2 , H 2 O Light-dependent reactions of photosynthesis FIGURE 20–1 Assimilation of CO 2 into biomass in plants. The light- driven synthesis of ATP and NADPH, described in Chapter 19, pro- vides energy and reducing power for the fixation of CO 2 into trioses, from which all the carbon-containing compounds of the plant cell are synthesized. The processes shown with red arrows are the focus of this chapter. Melvin Calvin, 1911–1997 8885d_c20_752 2/20/04 2:45 PM Page 752 mac76 mac76:385_reb: ments found in chloroplasts). They have no internal membranes analogous to the photosynthetic mem- branes (thylakoids) of chloroplasts, and in plant tissues rich in starch these plastids are packed with starch gran- ules (Fig. 20–2). Chloroplasts can be converted to pro- plastids by the loss of their internal membranes and chlorophyll, and proplastids are interconvertible with amyloplasts (Fig. 20–3). In turn, both amyloplasts and proplastids can develop into chloroplasts. The relative proportions of the plastid types depend on the type of plant tissue and on the intensity of illumination. Cells of green leaves are rich in chloroplasts, whereas amylo- plasts dominate in nonphotosynthetic tissues that store starch in large quantities, such as potato tubers. The inner membranes of all types of plastids are im- permeable to polar and charged molecules. Traffic across these membranes is mediated by sets of specific transporters. Carbon Dioxide Assimilation Occurs in Three Stages The first stage in the assimilation of CO 2 into biomole- cules (Fig. 20–4) is the carbon-fixation reaction: condensation of CO 2 with a five-carbon acceptor, ribu- lose 1,5-bisphosphate, to form two molecules of 3- phosphoglycerate. In the second stage, the 3-phos- phoglycerate is reduced to triose phosphates. Overall, three molecules of CO 2 are fixed to three molecules of ribulose 1,5-bisphosphate to form six molecules of glyc- eraldehyde 3-phosphate (18 carbons) in equilibrium with dihydroxyacetone phosphate. In the third stage, five of the six molecules of triose phosphate (15 car- bons) are used to regenerate three molecules of ribu- lose 1,5-bisphosphate (15 carbons), the starting mate- rial. The sixth molecule of triose phosphate, the net product of photosynthesis, can be used to make hex- oses for fuel and building materials, sucrose for trans- port to nonphotosynthetic tissues, or starch for storage. Thus the overall process is cyclical, with the continuous conversion of CO 2 to triose and hexose phosphates. Fructose 6-phosphate is a key intermediate in stage 3 of CO 2 assimilation; it stands at a branch point, leading either to regeneration of ribulose 1,5-bisphosphate or to synthesis of starch. The pathway from hexose phos- phate to pentose bisphosphate involves many of the same reactions used in animal cells for the conversion of pentose phosphates to hexose phosphates during the nonoxidative phase of the pentose phosphate pathway (see Fig. 14–22). In the photosynthetic as- similation of CO 2 , essentially the same set of reactions operates in the other direction, converting hexose phos- phates to pentose phosphates. This reductive pentose phosphate cycle uses the same enzymes as the oxida- tive pathway, and several more enzymes that make the reductive cycle irreversible. All 13 enzymes of the path- way are in the chloroplast stroma. 20.1 Photosynthetic Carbohydrate Synthesis 753 FIGURE 20–2 Amyloplasts filled with starch (dark granules) are stained with iodine in this section of Ranunculus root cells. Starch granules in various tissues range from 1 to 100 H9262m in diameter. FIGURE 20–3 Plastids: their origins and interconversions. All types of plastids are bounded by a double membrane, and some (notably the mature chloroplast) have extensive internal membranes. The in- ternal membranes can be lost (when a mature chloroplast becomes a proplastid) and resynthesized (as a proplastid gives rise to a pregranal plastid and then a mature chloroplast). Proplastids in nonphotosyn- thetic tissues (such as root) give rise to amyloplasts, which contain large quantities of starch. All plant cells have plastids, and these or- ganelles are the site of other important processes, including the syn- thesis of essential amino acids, thiamine, pyridoxal phosphate, flavins, and vitamins A, C, E, and K. Chloroplast ProplastidPregranal plastid Amyloplast 8885d_c20_751–786 2/18/04 1:56 PM Page 753 mac76 mac76:385_reb: Stage 1: Fixation of CO 2 into 3-Phosphoglycerate An im- portant clue to the nature of the CO 2 -assimilation mech- anisms in photosynthetic organisms came in the late 1940s. Calvin and his associates illuminated a suspen- sion of green algae in the presence of radioactive car- bon dioxide ( 14 CO 2 ) for just a few seconds, then quickly killed the cells, extracted their contents, and with the help of chromatographic methods searched for the metabolites in which the labeled carbon first appeared. The first compound that became labeled was 3-phos- phoglycerate, with the 14 C predominantly located in the carboxyl carbon atom. These experiments strongly suggested that 3-phosphoglycerate is an early interme- diate in photosynthesis. The many plants in which this three-carbon compound is the first intermediate are called C 3 plants, in contrast with the C 4 plants de- scribed below. The enzyme that catalyzes incorporation of CO 2 into an organic form is ribulose 1,5-bisphosphate carbox- ylase/oxygenase, a name mercifully shortened to ru- bisco. As a carboxylase, rubisco catalyzes the covalent attachment of CO 2 to the five-carbon sugar ribulose 1,5- bisphosphate and cleavage of the unstable six-carbon in- termediate to form two molecules of 3-phosphoglycerate, one of which bears the carbon introduced as CO 2 in its carboxyl group (Fig. 20–4). The enzyme’s oxygenase ac- tivity is discussed in Section 20.2. Plant rubisco, the crucial enzyme in the production of biomass from CO 2 , has a complex structure (Fig. 20–5a), with eight identical large subunits (M r 53,000; encoded in the chloroplast genome, or plastome), each containing a catalytic site, and eight identical small sub- units (M r 14,000; encoded in the nuclear genome) of uncertain function. The rubisco of photosynthetic bac- teria is simpler in structure, having two subunits that in many respects resemble the large subunits of the plant enzyme (Fig. 20–5b). This similarity is consistent with the endosymbiont hypothesis for the origin of chloro- plasts (p. 35). The plant enzyme has an exceptionally low turnover number; only three molecules of CO 2 are fixed per second per molecule of rubisco at 25 H11034C. To achieve high rates of CO 2 fixation, plants therefore need large amounts of this enzyme. In fact, rubisco makes up almost 50% of soluble protein in chloroplasts and is prob- ably one of the most abundant enzymes in the biosphere. Central to the proposed mechanism for plant ru- bisco is a carbamoylated Lys side chain with a bound Mg 2H11001 ion. The Mg 2H11001 ion brings together and orients the reactants at the active site (Fig. 20–6) and polar- izes the CO 2 , opening it to nucleophilic attack by the five-carbon enediolate reaction intermediate formed on the enzyme (Fig. 20–7). The resulting six-carbon inter- mediate breaks down to yield two molecules of 3- phosphoglycerate. Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria754 Ribulose 1,5- bisphosphate (3) C CHOH O P CHOH CH 2 O PCH 2 O CHOH PCH 2 O CHO Glyceraldehyde 3-phosphate (6) (3) H11001 H H11001 ATP (1) (6) ADP (3) NADP H11001 (6) (6)ADP (6) P i (6) (5) COO H11002 PCH 2 O 3-Phosphoglycerate (6) CHOH Energy production via glycolysis; starch or sugar synthesis Stage 3: Regeneration of acceptor Stage 1: Fixation Stage 2: Reduction (3) CO 2 ATP NADPH FIGURE 20–4 The three stages of CO 2 assimilation in photosynthetic organisms. Stoichiome- tries of three key intermediates (numbers in parentheses) reveal the fate of carbon atoms entering and leaving the cycle. As shown here, three CO 2 are fixed for the net synthesis of one molecule of glyceraldehyde 3-phosphate. This cycle is the photosynthetic carbon reduction cycle, or the Calvin cycle. 8885d_c20_751–786 2/18/04 1:56 PM Page 754 mac76 mac76:385_reb: 20.1 Photosynthetic Carbohydrate Synthesis 755 (b) (a) Top view Side view FIGURE 20–5 Structure of ribulose 1,5-bisphosphate carboxylase (rubisco). (a) Top and side view of a ribbon model of rubisco from spinach (PDB ID 8RUC). The enzyme has eight large subunits (blue) and eight small ones (gray), tightly packed into a structure of M r 550,000. Rubisco is present at a concentration of about 250 mg/mL in the chloroplast stroma, corresponding to an extraordinarily high concentration of active sites (~4 mM). Amino acid residues of the active site are shown in yellow, Mg 2H11001 in green. (b) Ribbon model of rubisco from the bacterium Rhodospirillum rubrum (PDB ID 9RUB). The subunits are in gray and blue. A Lys residue at the active site that is carboxylated to a carbamate in the active enzyme is shown in red. The substrate, ribulose 1,5-bisphosphate, is yellow; Mg 2+ is green. Asp 203 Glu 204 H 2 O (in CO 2 site) Carbamoyl-Lys 201 Ribulose 1,5-bisphosphate FIGURE 20–6 Central role of Mg 2H11545 in the catalytic mechanism of rubisco. (Derived from PDB ID 1RXO) Mg 2H11001 is coordinated in a roughly octahedral complex with six oxygen atoms: one oxygen in the carbamate on Lys 201 ; two in the carboxyl groups of Glu 204 and Asp 203 ; two at C-2 and C-3 of the substrate, ribulose 1,5-bisphosphate; and one in the other substrate, CO 2 . A water molecule occupies the Co2–binding site in this crystal structure. (Residue numbers refer to the spinach enzyme.) 8885d_c20_755 2/20/04 12:00 PM Page 755 mac76 mac76:385_reb: Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria756 Glu Asp Lys 201 N H O – OH H O O HC OH Enediolate intermediate C C C C + Mg 2+ O Glu Rubisco Asp Lys 201 N H O – O – H O O O CH 2 P CO 2 HC OH H 2 O Ribulose 1,5-bisphosphate 3-Phosphoglycerate Carbamoylated Lys side chain C C H C + N H His 294 b-Keto acid intermediate Hydrated intermediate 2 3 4 : : N Mg 2+ O CH 2 O H N H His 294 :N 5 1 O CH 2 P O CH 2 P O CH 2 P P O CH 2 P O CH 2 P Glu Asp Lys 201 N H O – – O O O OH HC OHC C C C + Mg 2+ H H O – OO C HCOH – OO C CH 2 O P Glu Asp Lys 201 N H O – – O O – O OH H O HC OHC C C C + Mg 2+ OH H O O CH 2 P O CH 2 P Glu Asp Lys 201 N H O OH C + O H – C Mg 2+ Lys 175 N + H H H – OO C HCOH CH 2 O P 3-Phosphoglycerate MECHANISM FIGURE 20–7 First stage of CO 2 assimilation: rubisco’s carboxylase activity. The CO 2 -fixation reaction is catalyzed by ribulose 1,5-bisphosphate carboxylase/oxygenase (rubisco). 1 Ribulose 1,5- bisphosphate forms an enediolate at the active site. 2 CO 2 , polarized by the proximity of the Mg 2H11001 ion, undergoes nucleophilic attack by the enediolate, producing a branched six-carbon sugar. 3 Hydroxy- lation at C-3 of this sugar, followed by aldol cleavage 4 , forms one molecule of 3-phosphoglycerate, which leaves the enzyme active site. 5 The carbanion of the remaining three-carbon fragment is proto- nated by the nearby side chain of Lys 175 , generating a second molecule of 3-phosphoglycerate. The overall reaction therefore accomplishes the combination of one CO 2 and one ribulose 1,5-bisphosphate to form two molecules of 3-phosphoglycerate, one of which contains the carbon atom from CO 2 (red). Rubisco Mechanism; Rubisco Tutorial 8885d_c20_756 2/20/04 12:01 PM Page 756 mac76 mac76:385_reb: As the catalyst for the first step of photosynthetic CO 2 assimilation, rubisco is a prime target for regula- tion. The enzyme is inactive until carbamoylated on the H9255 amino group of Lys 201 (Fig. 20–8). Ribulose 1,5- bisphosphate inhibits carbamoylation by binding tightly to the active site and locking the enzyme in the “closed” conformation, in which Lys 201 is inaccessible. Rubisco activase overcomes the inhibition by promoting ATP- dependent release of the ribulose 1,5-bisphosphate, ex- posing the Lys amino group to nonenzymatic car- bamoylation by CO 2 ; this is followed by Mg 2H11001 binding, which activates the rubisco. Rubisco activase in some species is activated by light through a redox mechanism (see Fig. 20–19). Another regulatory mechanism involves the “noc- turnal inhibitor” 2-carboxyarabinitol 1-phosphate, a nat- urally occurring transition-state analog (see Box 6–3) with a structure similar to that of the H9252-keto acid in- termediate of the rubisco reaction (Fig. 20–7; see also Fig. 20–20). This compound, synthesized in the dark in some plants, is a potent inhibitor of carbamoylated ru- bisco. It is either broken down when light returns or is expelled by rubisco activase, activating the rubisco. Stage 2: Conversion of 3-Phosphoglycerate to Glyceraldehyde 3-Phosphate The 3-phosphoglycerate formed in stage 1 is converted to glyceraldehyde 3-phosphate in two steps that are essentially the reversal of the corresponding steps in glycolysis, with one exception: the nucleotide cofactor for the reduction of 1,3-bisphosphoglycerate is NADPH rather than NADH (Fig. 20–9). The chloroplast stroma contains all the glycolytic enzymes except phos- phoglycerate mutase. The stromal and cytosolic enzymes are isozymes; both sets of enzymes catalyze the same reactions, but they are the products of different genes. CH 2 HO OH OH PO 3 2 H11002 CH 2 OH O CC 2-Carboxyarabinitol 1-phosphate C C H H O H11002 O 20.1 Photosynthetic Carbohydrate Synthesis 757 ADP HCOH HCOH rubisco activase CO 2 Ribulose 1,5-bisphosphate ATP-dependent removal of ribulose 1,5 bisphosphate uncovers e-amino group of Lys 201 . e-Amino group of Lys 201 is carbamoylated by CO 2 ; Mg 2+ binds to carbamoyl- Lys, activating rubisco. Rubisco with unmodified Lys 201 and bound ribulose 1,5-bisphosphate is inactive. O Mg 2+ Mg 2+ O O C CLys 201 NH 3 + Lys 201 NH 3 + + Lys 201 N H Rubisco ATP O CH 2 P O CH 2 P FIGURE 20–8 Role of rubisco activase in the carbamoylation of Lys 201 of rubisco. When the substrate ribulose 1,5-bisphosphate is bound to the active site, Lys 201 is not accessible. Rubisco activase couples ATP hydrolysis to expulsion of the bound sugar bisphosphate, exposing Lys 201 ; this Lys residue can now be carbamoylated with CO 2 in a re- action that is apparently not enzyme-mediated. Mg 2H11001 is attracted to and binds to the negatively charged carbamoyl-Lys, and the enzyme is thus activated. In the first step of stage 2, the stromal 3-phospho- glycerate kinase catalyzes the transfer of a phospho- ryl group from ATP to 3-phosphoglycerate, yielding 1,3-bisphosphoglycerate. Next, NADPH donates elec- trons in a reduction catalyzed by the chloroplast-specific isozyme of glyceraldehyde 3-phosphate dehydroge- nase, producing glyceraldehyde 3-phosphate and P i . Triose phosphate isomerase then interconverts glycer- aldehyde 3-phosphate and dihydroxyacetone phosphate. 8885d_c20_751–786 2/18/04 1:56 PM Page 757 mac76 mac76:385_reb: Most of the triose phosphate thus produced is used to regenerate ribulose 1,5-bisphosphate; the rest is either converted to starch in the chloroplast and stored for later use or immediately exported to the cytosol and con- verted to sucrose for transport to growing regions of the plant. In developing leaves, a significant portion of the triose phosphate may be degraded by glycolysis to pro- vide energy. Stage 3: Regeneration of Ribulose 1,5-Bisphosphate from Triose Phosphates The first reaction in the assimilation of CO 2 into triose phosphates consumes ribulose 1,5-bisphos- phate and, for continuous flow of CO 2 into carbohydrate, ribulose 1,5-bisphosphate must be constantly regener- ated. This is accomplished in a series of reactions (Fig. 20–10) that, together with stages 1 and 2, constitute the cyclic pathway shown in Figure 20–4. The product of Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria758 Sucrose ATP Starch Dihydroxyacetone phosphate Fructose 6-phosphate glyceraldehyde 3-phosphate dehydrogenase NADP + P i NADPH + H + ADP triose phosphate isomerase transaldolase fructose 1,6-bisphosphatase glycolysis P i –triose phosphate antiporter Stroma Cytosol CH 2 O C COO H5008 P C C CH 2 O P C O H OH OH H H HO CH 2 O C P C C CH 2 OH C O H OH OH H H HO CH 2 O CHOH P C CH 2 O CHOH P PO O CH CH 2 O CHOH P O CH 2 OH CH 2 O C P O 3-phosphoglycerate kinase Fructose 6-phosphate Glyceraldehyde 3-phosphate Fructose 1,6-bisphosphate Glyceraldehyde 3-phosphate Dihydroxyacetone phosphate Fructose 1,6-bisphosphate 1,3-Bisphosphoglycerate 3-Phosphoglycerate ATP FIGURE 20–9 Second stage of CO 2 assimilation. 3-Phosphoglycerate is converted to glyceraldehyde 3-phosphate (red arrows). Also shown are the alternative fates of the fixed carbon of glyceraldehyde 3-phosphate (blue arrows). Most of the glyceraldehyde 3-phosphate is recycled to ribulose 1,5-bisphosphate as shown in Figure 20–10. A small fraction of the “extra” glyceraldehyde 3-phosphate may be used immediately as a source of energy, but most is converted to sucrose for transport or is stored in the chloroplast as starch. In the latter case, glyceraldehyde 3-phosphate condenses with dihydroxyacetone phos- phate in the stroma to form fructose 1,6-bisphosphate, a precursor of starch. In other situations the glyceraldehyde 3-phosphate is converted to dihydroxyacetone phosphate, which leaves the chloroplast via a specific transporter (see Fig. 20–15) and, in the cytosol, can be degraded glycolytically to provide energy or used to form fructose 6-phosphate and hence sucrose. 8885d_c20_758 2/20/04 12:01 PM Page 758 mac76 mac76:385_reb: the first assimilation reaction (3-phosphoglycerate) thus undergoes transformations that regenerate ribu- lose 1,5-bisphosphate. The intermediates in this path- way include three-, four-, five-, six-, and seven-carbon sugars. In the following discussion, all step numbers re- fer to Figure 20–10. Steps 1 and 4 are catalyzed by the same enzyme, transaldolase. It first catalyzes the reversible conden- sation of glyceraldehyde 3-phosphate with dihydroxy- acetone phosphate, yielding fructose 1,6-bisphosphate (step 1 ); this is cleaved to fructose 6-phosphate and P i by fructose 1,6-bisphosphatase (FBPase-1) in step 2 . The reaction is strongly exergonic and essentially irreversible. Step 3 is catalyzed by transketolase, which contains thiamine pyrophosphate (TPP) as its prosthetic group (see Fig. 14–13a) and requires Mg 2H11001 . 20.1 Photosynthetic Carbohydrate Synthesis 759 Glyceraldehyde 3-phosphate Dihydroxyacetone phosphate transaldolase fructose 1,6-bisphosphatase transketolase ribulose 5-phosphate epimerase sedoheptulose 1,7-bisphosphatase ADP ADP ribulose 5-phosphate epimerase P i Fructose 1,6-bisphosphate Fructose 6-phosphate Erythrose 4-phosphateDihydroxyacetone phosphate Glyceraldehyde 3-phosphate Xylulose 5-phosphate Ribulose 5-phosphate Ribulose 5-phosphate ribose 5-phosphate isomerase ATP ATP ATP ADP ribulose 5-phosphate kinase transketolase Glyceraldehyde 3-phosphate Ribose 5-phosphate Ribulose 5-phosphate Ribulose 1,5-bisphosphate Ribulose 1,5- bisphosphate Sedoheptulose 1,7-bisphosphate Sedoheptulose 7-phosphate Xylulose 5-phosphate P i transaldolase 1 2 3 4 5 9 8 6 7 9 8 FIGURE 20–10 Third stage of CO 2 assimilation. This schematic diagram shows the interconversions of triose phosphates and pentose phosphates. Black dots represent the number of carbons in each compound. The starting materials are glyceraldehyde 3-phosphate and dihydroxyacetone phosphate. Reactions catalyzed by transaldolase ( 1 and 4 ) and transketolase ( 3 and 6 ) produce pentose phos- phates that are converted to ribulose 1,5-bisphosphate—ribose 5-phosphate by ribose 5-phosphate isomerase ( 7 ) and xylulose 5-phosphate by ribulose 5-phosphate epimerase ( 8 ). In step 9 , ribulose 5-phosphate is phosphorylated, regenerating ribulose 1,5- bisphosphate. The steps with blue arrows are exergonic and make the whole process irreversible: steps 2 fructose 1,6-bisphosphatase, 5 sedoheptulose bisphosphatase, and 9 ribulose 5-phosphate kinase. 8885d_c20_759 2/20/04 12:01 PM Page 759 mac76 mac76:385_reb: Transketolase catalyzes the reversible transfer of a 2-carbon ketol group (CH 2 OHOCOO) from a ketose phosphate donor, fructose 6-phosphate, to an aldose phosphate acceptor, glyceraldehyde 3-phosphate (Fig. 20–11a, b), forming the pentose xylulose 5-phosphate and the tetrose erythrose 4-phosphate. In step 4 , transaldolase acts again, combining erythrose 4-phos- phate with dihydroxyacetone phosphate to form the seven-carbon sedoheptulose 1,7-bisphosphate. An enzyme unique to plastids, sedoheptulose 1,7-bisphos- phatase, converts the bisphosphate to sedoheptulose 7-phosphate (step 5 ); this is the second irreversible reaction in the pathway. Transketolase now acts again, converting sedoheptulose 7-phosphate and glyceralde- hyde 3-phosphate to two pentose phosphates in step 6 (Fig. 20–11c). Figure 20–12 shows how a two-carbon fragment is temporarily carried on the transketolase cofactor TPP and condensed with the three carbons of glyceraldehyde 3-phosphate in step 6 . The pentose phosphates formed in the transketo- lase reactions—ribose 5-phosphate and xylulose 5-phos- phate—are converted to ribulose 5-phosphate (steps 7 and 8 ), which in the final step ( 9 ) of the cycle is phosphorylated to ribulose 1,5-bisphosphate by ribulose 5-phosphate kinase (Fig. 20–13). This is the third very exergonic reaction of the pathway, as the phosphate an- hydride bond in ATP is swapped for a phosphate ester in ribulose 1,5-bisphosphate. Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria760 CH 2 O CO CHOH P H11001 Ketose donor Aldose acceptor TPP transketolase Fructose 6-phosphate Glyceraldehyde 3-phosphate Xylulose 5-phosphate Erythrose 4-phosphate Sedoheptulose 7-phosphate Glyceraldehyde 3-phosphate (a) (b) (c) CH 2 OH R 2 C O R 1 CHOH H11001 R 2 R 1 CH 2 OH C O O C C H11001 HO HC CO CH 2 OH C H11001 C O C CH 2 OP OHH H OH CH 2 O HO OH P HC CO H11001 CH 2 OH C O C C C CH OHH CH 2 OP HO OH HC CO CH 2 OH C H CH 2 OP H11001 H C O C CH 2 OP OHH H CH OH OH OH H OHH HO OH Xylulose 5-phosphate HC CO CH 2 OH CH CH 2 OP H H H H Ribose 5-phosphate C O C C OHH OHH CH 2 OP OHCH H FIGURE 20–11 Transketolase-catalyzed reactions of the Calvin cycle. (a) General reaction catalyzed by transketolase: the transfer of a two-carbon group, carried temporarily on enzyme- bound TPP, from a ketose donor to an aldose acceptor. (b) Conversion of a hexose and a triose to a four-carbon and a five-carbon sugar (step 3 of Fig. 20–10). (c) Conversion of seven- carbon and three-carbon sugars to two pentoses (step 6 of Fig. 20–10). 8885d_c20_751–786 2/18/04 1:56 PM Page 760 mac76 mac76:385_reb: 20.1 Photosynthetic Carbohydrate Synthesis 761 Ribose 5-phosphate Xylulose 5-phosphate Glyceraldehyde 3-phosphate (aldose acceptor) Sedoheptulose 7-phosphate (ketose donor) Thiamine pyrophosphate (cofactor of transketolase) Carbanion, stabilized by resonance OH CH 2 OH C C C C C CH 2 O P OH OH O HHO H H H CH 2 O P P OHC C C OH OH H H H C HO CH 2 O P OHCH C HO OH CH 2 OH C C C CH 2 O P O HHO H OH H11002 C CH 2 OH CH 3 C RH11032 SN H11001 R OH C CH 2 OH CH 3 C RH11032 SN H11001 R H OH C OH H C OH H C OH H C O CH 2 OCH 2 C P H P RH11032R CH 2 CH 3 CH 3 CH 2 S N N NH 2 N H11001 FIGURE 20–12 TPP as a cofactor for transketolase. Transketolase transfers a two-carbon group from sedoheptulose 7-phosphate to glyceraldehyde 3-phosphate, producing two pentose phosphates (step 6 in Fig. 20–10). Thiamine pyrophosphate serves as a temporary carrier of the two-carbon unit and as an electron sink (see Fig. 14–13) to facilitate the reactions. O CHOH CH 2 O C OC HC H H HO Xylulose 5-phosphate ribose 5-phosphate isomerase ribose 5-phosphate epimerase P CHOH CH 2 OH OHC OHCH Ribose 5-phosphate PCH 2 O CHOH CH 2 O OC OHCH Ribulose 5-phosphate P CH 2 OH CHOH CH 2 O OC OHCH Ribulose 1,5-bisphosphate P ATP ADP CH 2 O P ribulose 5-phosphate kinase FIGURE 20–13 Regeneration of ribulose 1,5-bisphosphate. The starting material for the Calvin cycle, ribulose 1,5-bisphosphate, is regenerated from two pentose phosphates produced in the cycle. This pathway involves the action of an isomerase and an epimerase, then phosphorylation by a kinase, with ATP as phosphate group donor (steps 7 , 8 , and 9 of Fig. 20–10). 8885d_c20_751–786 2/18/04 1:56 PM Page 761 mac76 mac76:385_reb: Synthesis of Each Triose Phosphate from CO 2 Requires Six NADPH and Nine ATP The net result of three turns of the Calvin cycle is the conversion of three molecules of CO 2 and one molecule of phosphate to a molecule of triose phosphate. The stoi- chiometry of the overall path from CO 2 to triose phos- phate, with regeneration of ribulose 1,5-bisphosphate, is shown in Figure 20–14. Three molecules of ribulose 1,5-bisphosphate (a total of 15 carbons) condense with three CO 2 (3 carbons) to form six molecules of 3-phos- phoglycerate (18 carbons). These six molecules of 3- phosphoglycerate are reduced to six molecules of glyc- eraldehyde 3-phosphate (which is in equilibrium with dihydroxyacetone phosphate), with the expenditure of six ATP (in the synthesis of 1,3-bisphosphoglycerate) and six NADPH (in the reduction of 1,3-bisphospho- glycerate to glyceraldehyde 3-phosphate). The isozyme of glyceraldehyde 3-phosphate dehydrogenase present in chloroplasts can use NADP as its electron carrier and normally functions in the direction of 1,3-bisphospho- glycerate reduction. The cytosolic isozyme uses NAD, as does the glycolytic enzyme of animals and other eu- karyotes, and in the dark this isozyme acts in glycolysis to oxidize glyceraldehyde 3-phosphate. Both glycer- aldehyde 3-phosphate dehydrogenase isozymes, like all enzymes, catalyze the reaction in both directions. One molecule of glyceraldehyde 3-phosphate is the net product of the carbon assimilation pathway. The other five triose phosphate molecules (15 carbons) are rearranged in steps 1 to 9 of Figure 20–10 to form three molecules of ribulose 1,5-bisphosphate (15 car- bons). The last step in this conversion requires one ATP per ribulose 1,5-bisphosphate, or a total of three ATP. Thus, in summary, for every molecule of triose phos- phate produced by photosynthetic CO 2 assimilation, six NADPH and nine ATP are required. NADPH and ATP are produced in the light- dependent reactions of photosynthesis in about the same ratio (2:3) as they are consumed in the Calvin cy- cle. Nine ATP molecules are converted to ADP and phos- phate in the generation of a molecule of triose phos- phate; eight of the phosphates are released as P i and combined with eight ADP to regenerate ATP. The ninth phosphate is incorporated into the triose phosphate it- self. To convert the ninth ADP to ATP, a molecule of P i must be imported from the cytosol, as we shall see. In the dark, the production of ATP and NADPH by photophosphorylation, and the incorporation of CO 2 into triose phosphate (by the so-called dark reactions), cease. The “dark reactions” of photosynthesis were so named to distinguish them from the primary light- driven reactions of electron transfer to NADP H11001 and syn- thesis of ATP, described in Chapter 19. They do not, in Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria762 Glyceraldehyde 3-phosphate1 6 ATP 6ADP 6 NADPH + 6H + 6P i 6NADP + 3ADP 3 ATP 2P i 1,3-Bisphosphoglycerate6 3-Phosphoglycerate6 Ribulose 1,5-bisphosphate3 Ribulose 5-phosphate3 Glyceraldehyde 3-phosphate 5 CO 2 + H 2 O3 Dihydroxyacetone phosphate Glyceraldehyde 3-phosphate Dihydroxyacetone phosphate 6 FIGURE 20–14 Stoichiometry of CO 2 assimilation in the Calvin cycle. For every three CO 2 molecules fixed, one molecule of triose phosphate (glyceraldehyde 3-phosphate) is produced and nine ATP and six NADPH are consumed. 8885d_c20_751–786 2/18/04 1:56 PM Page 762 mac76 mac76:385_reb: fact, occur at significant rates in the dark and are thus more appropriately called the carbon-assimilation re- actions. Later in this section we describe the regula- tory mechanisms that turn on carbon assimilation in the light and turn it off in the dark. The chloroplast stroma contains all the enzymes necessary to convert the triose phosphates produced by CO 2 assimilation (glyceraldehyde 3-phosphate and di- hydroxyacetone phosphate) to starch, which is tem- porarily stored in the chloroplast as insoluble granules. Aldolase condenses the trioses to fructose 1,6-bisphos- phate; fructose 1,6-bisphosphatase produces fructose 6- phosphate; phosphohexose isomerase yields glucose 6- phosphate; and phosphoglucomutase produces glucose 1-phosphate, the starting material for starch synthesis (see Section 20.3). All the reactions of the Calvin cycle except those catalyzed by rubisco, sedoheptulose 1,7-bisphospha- tase, and ribulose 5-phosphate kinase also take place in animal tissues. Lacking these three enzymes, animals cannot carry out net conversion of CO 2 to glucose. A Transport System Exports Triose Phosphates from the Chloroplast and Imports Phosphate The inner chloroplast membrane is impermeable to most phosphorylated compounds, including fructose 6-phos- phate, glucose 6-phosphate, and fructose 1,6-bisphos- phate. It does, however, have a specific antiporter that catalyzes the one-for-one exchange of P i with a triose phosphate, either dihydroxyacetone phosphate or 3- phosphoglycerate (Fig. 20–15; see also Fig. 20–9). This antiporter simultaneously moves P i into the chloroplast, where it is used in photophosphorylation, and moves triose phosphate into the cytosol, where it can be used to synthesize sucrose, the form in which the fixed car- bon is transported to distant plant tissues. Sucrose synthesis in the cytosol and starch synthe- sis in the chloroplast are the major pathways by which the excess triose phosphate from photosynthesis is “har- vested.” Sucrose synthesis (described below) releases four P i molecules from the four triose phosphates re- quired to make sucrose. For every molecule of triose phosphate removed from the chloroplast, one P i is trans- ported into the chloroplast, providing the ninth P i men- tioned above, to be used in regenerating ATP. If this ex- change were blocked, triose phosphate synthesis would quickly deplete the available P i in the chloroplast, slow- ing ATP synthesis and suppressing assimilation of CO 2 into starch. The P i –triose phosphate antiport system serves one additional function. ATP and reducing power are needed in the cytosol for a variety of synthetic and energy- requiring reactions. These requirements are met to an as yet undetermined degree by mitochondria, but a sec- ond potential source of energy is the ATP and NADPH generated in the chloroplast stroma during the light reactions. However, neither ATP nor NADPH can cross the chloroplast membrane. The P i –triose phosphate antiport system has the indirect effect of moving ATP 20.1 Photosynthetic Carbohydrate Synthesis 763 Stroma Cytosol Chloroplast inner membrane P i –triose phosphate antiporter Dihydroxy- acetone phosphate Sucrose P i P i Dihydroxy- acetone phosphate photosynthesis Calvin cycle 9ADP + 8P i + P i 9ATP 9ATP 9ADP + 9P i Light FIGURE 20–15 The P i –triose phosphate antiport system of the inner chloroplast membrane. This transporter facilitates the exchange of cytosolic P i for stromal dihydroxyacetone phosphate. The products of photosynthetic carbon assimilation are thus moved into the cytosol where they serve as a starting point for sucrose biosynthesis, and P i required for photophosphorylation is moved into the stroma. This same antiporter can transport 3-phosphoglycerate and acts in the shuttle for exporting ATP and reducing equivalents (see Fig. 20–16). 8885d_c20_763 2/20/04 12:02 PM Page 763 mac76 mac76:385_reb: equivalents and reducing equivalents from the chloro- plast to the cytosol (Fig. 20–16). Dihydroxyacetone phosphate formed in the stroma is transported to the cytosol, where it is converted by glycolytic enzymes to 3-phosphoglycerate, generating ATP and NADH. 3- Phosphoglycerate reenters the chloroplast, completing the cycle. Four Enzymes of the Calvin Cycle Are Indirectly Activated by Light The reductive assimilation of CO 2 requires a lot of ATP and NADPH, and their stromal concentrations increase when chloroplasts are illuminated (Fig. 20–17). The light-induced transport of protons across the thylakoid membrane (Chapter 19) also increases the stromal pH from about 7 to about 8, and it is accompanied by a flow of Mg 2H11001 from the thylakoid compartment into the stroma, raising the [Mg 2H11001 ] from 1 to 3 mM to 3 to 6 mM. Several stromal enzymes have evolved to take advan- tage of these light-induced conditions, which signal the availability of ATP and NADPH: the enzymes are more active in an alkaline environment and at high [Mg 2H11001 ]. For example, activation of rubisco by formation of the carbamoyllysine is faster at alkaline pH, and high stro- mal [Mg 2H11001 ] favors formation of the enzyme’s active Mg 2H11001 complex. Fructose 1,6-bisphosphatase requires Mg 2H11001 and is very dependent on pH (Fig. 20–18); its activity increases more than 100-fold when pH and [Mg 2H11001 ] rise during chloroplast illumination. Four Calvin cycle enzymes are subject to a special type of regulation by light. Ribulose 5-phosphate kinase, fructose 1,6-bisphosphatase, sedoheptulose 1,7-bisphos- phatase, and glyceraldehyde 3-phosphate dehydroge- nase are activated by light-driven reduction of disulfide bonds between two Cys residues critical to their cat- alytic activities. When these Cys residues are disulfide- bonded (oxidized), the enzymes are inactive; this is the normal situation in the dark. With illumination, electrons flow from photosystem I to ferredoxin (see Fig. 19–49), which passes electrons to a small, soluble, disulfide- containing protein called thioredoxin (Fig. 20–19), in a reaction catalyzed by ferredoxin-thioredoxin re- ductase. Reduced thioredoxin donates electrons for the reduction of the disulfide bonds of the light-activated enzymes, and these reductive cleavage reactions are accompanied by conformational changes that increase enzyme activities. At nightfall, the Cys residues in the Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria764 NADH + H + Stroma Chloroplast inner membrane Cytosol P i –triose phosphate antiporter P i P i 3-Phosphoglycerate 1,3-Bisphosphoglycerate Dihydroxyacetone phosphate NAD + ADP P i P i phospho- glycerate kinase glyceraldehyde 3-phosphate dehydrogenase 1,3-Bisphosphoglycerate Glyceraldehyde 3-phosphate ADP Dihydroxyacetone phosphate 3-Phosphoglycerate P i –triose phosphate antiporter ATPATP Glyceraldehyde 3-phosphate NADPH + H + NADP + triose phosphate isomerase FIGURE 20–16 Role of the P i –triose phosphate antiporter in the transport of ATP and reducing equivalents. Dihydroxyacetone phosphate leaves the chloroplast and is converted to glyceraldehyde 3-phosphate in the cytosol. The cytosolic glycer- aldehyde 3-phosphate dehydrogenase and phosphoglycerate kinase reactions then produce NADH, ATP, and 3-phosphoglycerate. The latter reenters the chloroplast and is reduced to dihy- droxyacetone phosphate, completing a cycle that effectively moves ATP and reducing equivalents (NADPH/NADH) from chloroplast to cytosol. 8885d_c20_751–786 2/18/04 1:56 PM Page 764 mac76 mac76:385_reb: four enzymes are reoxidized to their disulfide forms, the enzymes are inactivated, and ATP is not expended in CO 2 assimilation. Instead, starch synthesized and stored during the daytime is degraded to fuel glycolysis at night. Glucose 6-phosphate dehydrogenase, the first en- zyme in the oxidative pentose phosphate pathway, is also regulated by this light-driven reduction mechanism, but in the opposite sense. During the day, when photo- synthesis produces plenty of NADPH, this enzyme is not needed for NADPH production. Reduction of a critical disulfide bond by electrons from ferredoxin inactivates the enzyme. 20.1 Photosynthetic Carbohydrate Synthesis 765 Light CO 2 -assimilation cycle Glyceraldehyde 3-phosphate CO 2 P i H + Mg 2+Photosynthetic electron transfer Stroma Thylakoid + H + ATP NADPH NADP + ADP + P i FIGURE 20–17 Source of ATP and NADPH. ATP and NADPH pro- duced by the light reactions are essential substrates for the reduction of CO 2 . The photosynthetic reactions that produce ATP and NADPH are accompanied by movement of protons (red) from the stroma into the thylakoid, creating alkaline conditions in the stroma. Magnesium ions pass from the thylakoid into the stroma, increasing the stromal [Mg 2H11001 ]. FIGURE 20–18 Activation of chloroplast fructose 1,6-bisphosphatase. Reduced fructose 1,6-bisphosphatase (FBPase-1) is activated by light and by the combination of high pH and high [Mg 2H11001 ] in the stroma, both of which are produced by illumination. FBPase-1 activity (units/mg) 0 150 [MgCl 2 ] (mM) 100 50 2015105 pH 8.0 pH 7.5 pH 7.0 0 FIGURE 20–19 Light activation of several enzymes of the Calvin cycle. The light activation is mediated by thioredoxin, a small, disulfide-containing protein. In the light, thioredoxin is reduced by electrons moving from photosystem I through ferredoxin (Fd) (blue arrows), then thioredoxin reduces critical disulfide bonds in each of the enzymes sedoheptulose 1,7-bisphosphatase, fructose 1,6- bisphosphatase, ribulose 5-phosphate kinase, and glyceraldehye 3-phosphate dehydrogenase, activating these enzymes. In the dark, the OSH groups undergo reoxidation to disulfides, inactivating the enzymes. Fd ox Fd red ferredoxin- thioredoxin reductase Thioredoxin HS SH Enzyme (active) O 2 (in dark) Photosystem I HS SH Enzyme (inactive) SS Thioredoxin SS Light 8885d_c20_765 2/20/04 12:02 PM Page 765 mac76 mac76:385_reb: SUMMARY 20.1 Photosynthetic Carbohydrate Synthesis ■ Photosynthesis in vascular plants takes place in chloroplasts. In the CO 2 -assimilating reactions (the Calvin cycle), ATP and NADPH are used to reduce CO 2 to triose phosphates. These reactions occur in three stages: the fixation reaction itself, catalyzed by rubisco; reduction of the resulting 3-phosphoglycerate to glyceraldehyde 3-phosphate; and regeneration of ribulose 1,5-bisphosphate from triose phosphates. ■ Rubisco condenses CO 2 with ribulose 1,5-bisphosphate, forming an unstable hexose bisphosphate that splits into two molecules of 3-phosphoglycerate. Rubisco is activated by covalent modification (carbamoylation of Lys 201 ) catalyzed by rubisco activase and is inhibited by a natural transition-state analog, whose concentration rises in the dark and falls during daylight. ■ Stromal isozymes of the glycolytic enzymes catalyze reduction of 3-phosphoglycerate to glyceraldehyde 3-phosphate; each molecule reduced requires one ATP and one NADPH. ■ Stromal enzymes, including transketolase and transaldolase, rearrange the carbon skeletons of triose phosphates, generating intermediates of three, four, five, six, and seven carbons and eventually yielding pentose phosphates. The pentose phosphates are converted to ribulose 5-phosphate, then phosphorylated to ribulose 1,5-bisphosphate to complete the Calvin cycle. ■ The cost of fixing three CO 2 into one triose phosphate is nine ATP and six NADPH, which are provided by the light-dependent reactions of photosynthesis. ■ An antiporter in the inner chloroplast membrane exchanges P i in the cytosol for 3-phosphoglycerate or dihydroxyacetone phosphate produced by CO 2 assimilation in the stroma. Oxidation of dihydroxyacetone phos- phate in the cytosol generates ATP and NADH, thus moving ATP and reducing equivalents from the chloroplast to the cytosol. ■ Four enzymes of the Calvin cycle are activated indirectly by light and are inactive in the dark, so that hexose synthesis does not compete with glycolysis—which is required to provide energy in the dark. 20.2 Photorespiration and the C 4 and CAM Pathways As we have seen, photosynthetic cells produce O 2 (by the splitting of H 2 O) during the light-driven reactions (Chapter 19) and use CO 2 during the light-independent processes (described above), so the net gaseous change during photosynthesis is the uptake of CO 2 and release of O 2 : CO 2 H11001 H 2 O 88n O 2 H11001 (CH 2 O) In the dark, plants also carry out mitochondrial res- piration, the oxidation of substrates to CO 2 and the conversion of O 2 to H 2 O. And there is another process in plants that, like mitochondrial respiration, consumes O 2 and produces CO 2 and, like photosynthesis, is driven by light. This process, photorespiration, is a costly side reaction of photosynthesis, a result of the lack of specificity of the enzyme rubisco. In this section we de- scribe this side reaction and the strategies plants use to minimize its metabolic consequences. Photorespiration Results from Rubisco’s Oxygenase Activity Rubisco is not absolutely specific for CO 2 as a substrate. Molecular oxygen (O 2 ) competes with CO 2 at the active site, and about once in every three or four turnovers, rubisco catalyzes the condensation of O 2 with ribulose 1,5-bisphosphate to form 3-phosphoglyc- erate and 2-phosphoglycolate (Fig. 20–20), a meta- bolically useless product. This is the oxygenase activ- ity referred to in the full name of the enzyme: ribulose 1,5-bisphosphate carboxylase/oxygenase. The reaction with O 2 results in no fixation of carbon and appears to be a net liability to the cell; salvaging the carbons from 2-phosphoglycolate (by the pathway outlined below) consumes significant amounts of cellular energy and releases some previously fixed CO 2 . Given that the reaction with oxygen is deleterious to the organism, why did the evolution of rubisco pro- duce an active site unable to discriminate well between CO 2 and O 2 ? Perhaps much of this evolution occurred before the time, about 2.5 billion years ago, when pro- duction of O 2 by photosynthetic organisms started to raise the oxygen content of the atmosphere. Before that time, there was no selective pressure for rubisco to dis- criminate between CO 2 and O 2 . The K m for CO 2 is about 9 H9262M, and that for O 2 is about 350 H9262M. The modern at- mosphere contains about 20% O 2 and only 0.04% CO 2 , so an aqueous solution in equilibrium with air at room temperature contains about 250 H9262M O 2 and 11 H9262M CO 2 — concentrations that allow significant O 2 “fixation” by ru- bisco and thus a significant waste of energy. The tem- perature dependence of the solubilities of O 2 and CO 2 is Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria766 8885d_c20_751–786 2/18/04 1:56 PM Page 766 mac76 mac76:385_reb: such that at higher temperatures, the ratio of O 2 to CO 2 in solution increases. In addition, the affinity of rubisco for CO 2 decreases with increasing temperature, exacer- bating its tendency to catalyze the wasteful oxygenase reaction. And as CO 2 is consumed in the assimilation re- actions, the ratio of O 2 to CO 2 in the air spaces of a leaf increases, further favoring the oxygenase reaction. The Salvage of Phosphoglycolate Is Costly The glycolate pathway converts two molecules of 2- phosphoglycolate to a molecule of serine (three carbons) and a molecule of CO 2 (Fig. 20–21). In the chloroplast, a phosphatase converts 2-phosphoglycolate to glycolate, which is exported to the peroxisome. There, glycolate is 20.2 Photorespiration and the C 4 and CAM Pathways 767 2-Phospho- glycolate CO 2 released in photorespiration Mitochondrion NAD + Serine Hydroxypyruvate H 2 O 2 O 2 Glyoxylate Glycine Glycolate glycolic acid oxidase Peroxisome Serine 2 Glycine -hydroxy acid reductase GlycolateGlycerate Calvin cycle ADP P i Chloroplast stroma NADH + H + NAD + NH 3 glycine decarboxylase [ NH 2 ] transamination + H + CH 2 O P COO H5008 CH 2 OH COO H5008 CH 2 OH COO H5008 CHO COO H5008 COO H5008 CH 2 NH 3 H11001 COO H5008 CH 2 NH 3 H11001 OH CH NH 3 H11001 COO H5008 CH 2 OOH C COO H5008 CH 2 Glycerate OHOH CH COO H5008 CH 2 OHOH CH COO H5008 CH 2 OH CH NH 3 H11001 COO H5008 CH 2 NADH H9251 O 2 Ribulose 1,5- bisphosphate 3-Phospho- glycerate ATP CH 2 O C OH HOHC CO CH 2 O CH 2 O C O HOH C CHOH CH 2 O Ribulose 1,5-bisphosphate Enediol form Enzyme-bound intermediate CH 2 O C OH H OH C COH CH 2 O O 2 O O H OH H5008 H 2 O CH 2 O C H11001 H5008 O C CH 2 O OHH O H5008 O C O 2-Phosphoglycolate 3-Phosphoglycerate P P P P P P P P FIGURE 20–20 Oxygenase activity of rubisco. Rubisco can incorpo- rate O 2 rather than CO 2 into ribulose 1,5-bisphosphate. The unstable intermediate thus formed splits into 2-phosphoglycolate (recycled as described in Fig. 20–21) and 3-phosphoglycerate, which can reenter the Calvin cycle. FIGURE 20–21 Glycolate pathway. This pathway, which salvages 2- phosphoglycolate (shaded pink) by its conversion to serine and even- tually 3-phosphoglycerate, involves three cellular compartments. Gly- colate formed by dephosphorylation of 2-phosphoglycolate in chloroplasts is oxidized to glyoxylate in peroxisomes and then transaminated to glycine. In mitochondria, two glycine molecules con- dense to form serine and the CO 2 released during photorespiration (shaded green). This reaction is catalyzed by glycine decarboxylase, an enzyme present at very high levels in the mitochondria of C 3 plants (see text). The serine is converted to hydroxypyruvate and then to glyc- erate in peroxisomes; glycerate reenters the chloroplasts to be phos- phorylated, rejoining the Calvin cycle. Oxygen (shaded blue) is con- sumed at two steps during photorespiration. 8885d_c20_751–786 2/18/04 1:56 PM Page 767 mac76 mac76:385_reb: oxidized by molecular oxygen, and the resulting aldehyde (glyoxylate) undergoes transamination to glycine. The hy- drogen peroxide formed as a side product of glycolate ox- idation is rendered harmless by peroxidases in the per- oxisome. Glycine passes from the peroxisome to the mitochondrial matrix, where it undergoes oxidative de- carboxylation by the glycine decarboxylase complex, an enzyme similar in structure and mechanism to two mito- chondrial complexes we have already encountered: the pyruvate dehydrogenase complex and the H9251-ketoglutarate dehydrogenase complex (Chapter 16). The glycine de- carboxylase complex oxidizes glycine to CO 2 and NH 3 , with the concomitant reduction of NAD H11001 to NADH and transfer of the remaining carbon from glycine to the cofactor tetrahydrofolate (Fig. 20–22). The one-carbon unit carried on tetrahydrofolate is then transferred to a second glycine by serine hydroxymethyltransferase, producing serine. The net reaction catalyzed by the glycine decarboxylase complex and serine hydrox- ymethyltransferase is 2 Glycine H11001 NAD H11001 H11001 H 2 O 88n serine H11001 CO 2 H11001 NH 3 H11001 NADH H11001 H H11001 The serine is converted to hydroxypyruvate, to glycer- ate, and finally to 3-phosphoglycerate, which is used to regenerate ribulose 1,5-bisphosphate, completing the long, expensive cycle (Fig. 20–21). In bright sunlight, the flux through the glycolate sal- vage pathway can be very high, producing about five times more CO 2 than is typically produced by all the ox- Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria768 FAD FAD FAD – OOC CH 2 NH + HC PLP H C S S H Glycine COO – H 3 N + CO 2 H 2 N H 2 O CH 2 N 5 ,N 10 -methylene H 4 F H 2 O NH 3 FAD H P 2 1 5 3 4 L T PLP S S H P L O HC T PLP S S H P L FADH 2 O HC T PLP H P L O HC T PLP S HS HS HS H P L O HC T NADH + H + NAD + H 4 F FIGURE 20–22 The glycine decarboxylase system. Glycine decar- boxylase in plant mitochondria is a complex of four types of subunits, with the stoichiometry P 4 H 27 T 9 L 2 . Protein H has a covalently attached lipoic acid residue that can undergo reversible oxidation. Step 1 is formation of a Schiff base between pyridoxal phosphate (PLP) and glycine, catalyzed by protein P (named for its bound PLP). In step 2 , protein P catalyzes oxidative decarboxylation of glycine, releas- ing CO 2 ; the remaining methylamine group is attached to one of the OSH groups of reduced lipoic acid. 3 Protein T (which uses tetrahy- drofolate (H 4 F) as cofactor) now releases NH 3 from the methylamine moiety and transfers the remaining one-carbon fragment to tetrahy- drofolate, producing N 5 ,N 10 -methylene tetrahydrofolate. 4 Protein L oxidizes the two OSH groups of lipoic acid to a disulfide, passing electrons through FAD to NAD H11001 5 , thus completing the cycle. The N 5 ,N 10 -methylene tetrahydrofolate formed in this process is used by serine hydroxymethyltransferase to convert a molecule of glycine to serine, regenerating the tetrahydrofolate that is essential for the reac- tion catalyzed by protein T. The L subunit of glycine decarboxylase is identical to the dihydrolipoyl dehydrogenase (E 3 ) of pyruvate dehy- drogenase and H9251-ketoglutarate dehydrogenase (see Fig. 16–6). 8885d_c20_751–786 2/18/04 1:56 PM Page 768 mac76 mac76:385_reb: idations of the citric acid cycle. To generate this large flux, mitochondria contain prodigious amounts of the glycine decarboxylase complex: the four proteins of the complex make up half of all the protein in the mito- chondrial matrix in the leaves of pea and spinach plants! In nonphotosynthetic parts of a plant, such as potato tu- bers, mitochondria have very low concentrations of the glycine decarboxylase complex. The combined activity of the rubisco oxygenase and the glycolate salvage pathway consumes O 2 and produces CO 2 —hence the name photorespiration. This pathway is perhaps better called the oxidative photosynthetic carbon cycle or C 2 cycle, names that do not invite comparison with respiration in mi- tochondria. Unlike mitochondrial respiration, “pho- torespiration” does not conserve energy and may actually inhibit net biomass formation as much as 50%. This inefficiency has led to evolutionary adaptations in the carbon-assimilation processes, particularly in plants that have evolved in warm climates. In C 4 Plants, CO 2 Fixation and Rubisco Activity Are Spatially Separated In many plants that grow in the tropics (and in temper- ate-zone crop plants native to the tropics, such as maize, sugarcane, and sorghum) a mechanism has evolved to circumvent the problem of wasteful photorespiration. The step in which CO 2 is fixed into a three-carbon prod- uct, 3-phosphoglycerate, is preceded by several steps, one of which is temporary fixation of CO 2 into a four- carbon compound. Plants that use this process are re- ferred to as C 4 plants, and the assimilation process as C 4 metabolism or the C 4 pathway. Plants that use the carbon-assimilation method we have described thus far, in which the first step is reaction of CO 2 with ribulose 1,5-bisphosphate to form 3-phosphoglycerate, are called C 3 plants. The C 4 plants, which typically grow at high light in- tensity and high temperatures, have several important characteristics: high photosynthetic rates, high growth rates, low photorespiration rates, low rates of water loss, and a specialized leaf structure. Photosynthesis in the leaves of C 4 plants involves two cell types: mesophyll and bundle-sheath cells (Fig. 20–23a). There are three variants of C 4 metabolism, worked out in the 1960s by Marshall Hatch and Rodger Slack (Fig. 20–23b). In plants of tropical origin, the first intermediate into which 14 CO 2 is fixed is oxaloacetate, a four-carbon compound. This reaction, which occurs in the cytosol of leaf mesophyll cells, is catalyzed by phosphoenolpyru- vate carboxylase, for which the substrate is HCO 3 H11002 , not CO 2 . The oxaloacetate thus formed is either reduced to malate at the expense of NADPH (as shown in Fig. 20–23b) or converted to aspartate by transamination: Oxaloacetate H11001 H9251-amino acid 88n L-aspartate H11001 H9251-keto acid 20.2 Photorespiration and the C 4 and CAM Pathways 769 The malate or aspartate formed in the mesophyll cells then passes into neighboring bundle-sheath cells through plasmodesmata, protein-lined channels that connect two plant cells and provide a path for move- ment of metabolites and even small proteins between cells. In the bundle-sheath cells, malate is oxidized and decarboxylated to yield pyruvate and CO 2 by the action of malic enzyme, reducing NADP H11001 . In plants that use aspartate as the CO 2 carrier, aspartate arriving in bundle-sheath cells is transaminated to form oxaloac- etate and reduced to malate, then the CO 2 is released by malic enzyme or PEP carboxykinase. As labeling ex- periments show, the free CO 2 released in the bundle- sheath cells is the same CO 2 molecule originally fixed into oxaloacetate in the mesophyll cells. This CO 2 is now fixed again, this time by rubisco, in exactly the same re- action that occurs in C 3 plants: incorporation of CO 2 into C-1 of 3-phosphoglycerate. The pyruvate formed by decarboxylation of malate in bundle-sheath cells is transferred back to the meso- phyll cells, where it is converted to PEP by an unusual enzymatic reaction catalyzed by pyruvate phosphate dikinase (Fig. 20–23b). This enzyme is called a dikinase because two different molecules are simultaneously phosphorylated by one molecule of ATP: pyruvate to PEP, and phosphate to pyrophosphate. The pyro- phosphate is subsequently hydrolyzed to phosphate, so two high-energy phosphate groups of ATP are used in regenerating PEP. The PEP is now ready to receive an- other molecule of CO 2 in the mesophyll cell. The PEP carboxylase of mesophyll cells has a high affinity for HCO 3 H11002 (which is favored relative to CO 2 in aqueous solution and can fix CO 2 more efficiently than can rubisco). Unlike rubisco, it does not use O 2 as an alternative substrate, so there is no competition be- tween CO 2 and O 2 . The PEP carboxylase reaction, then, serves to fix and concentrate CO 2 in the form of malate. Release of CO 2 from malate in the bundle-sheath cells yields a sufficiently high local concentration of CO 2 for rubisco to function near its maximal rate, and for sup- pression of the enzyme’s oxygenase activity. Once CO 2 is fixed into 3-phosphoglycerate in the bundle-sheath cells, the other reactions of the Calvin cy- cle take place exactly as described earlier. Thus in C 4 plants, mesophyll cells carry out CO 2 assimilation by the C 4 pathway and bundle-sheath cells synthesize starch and sucrose by the C 3 pathway. Three enzymes of the C 4 pathway are regulated by light, becoming more active in daylight. Malate dehy- drogenase is activated by the thioredoxin-dependent re- duction mechanism shown in Figure 20–19; PEP car- boxylase is activated by phosphorylation of a Ser residue; and pyruvate phosphate dikinase is activated by dephosphorylation. In the latter two cases, the de- tails of how light effects phosphorylation or dephos- phorylation are not known. 8885d_c20_751–786 2/18/04 1:56 PM Page 769 mac76 mac76:385_reb: The pathway of CO 2 assimilation has a greater en- ergy cost in C 4 plants than in C 3 plants. For each mol- ecule of CO 2 assimilated in the C 4 pathway, a molecule of PEP must be regenerated at the expense of two high- energy phosphate groups of ATP. Thus C 4 plants need five ATP molecules to assimilate one molecule of CO 2 , whereas C 3 plants need only three (nine per triose phosphate). As the temperature increases (and the affinity of rubisco for CO 2 decreases, as noted above), a point is reached (at about 28 to 30 H11034C) at which the gain in efficiency from the elimination of photorespira- tion more than compensates for this energetic cost. C 4 plants (crabgrass, for example) outgrow most C 3 plants during the summer, as any experienced gardener can attest. In CAM Plants, CO 2 Capture and Rubisco Action Are Temporally Separated Succulent plants such as cactus and pineapple, which are native to very hot, very dry environments, have an- other variation on photosynthetic CO 2 fixation, which reduces loss of water vapor through the pores (stom- ata) by which CO 2 and O 2 must enter leaf tissue. In- stead of separating the initial trapping of CO 2 and its fixation by rubisco across space (as do the C 4 plants), they separate these two events over time. At night, when the air is cooler and moister, the stomata open to allow entry of CO 2 , which is then fixed into oxaloacetate by PEP carboxylase. The oxaloacetate is reduced to malate and stored in the vacuoles, to protect cytosolic and plas- tid enzymes from the low pH produced by malic acid dissociation. During the day the stomata close, pre- venting the water loss that would result from high day- time temperatures, and the CO 2 trapped overnight in malate is released as CO 2 by the NADP-linked malic en- zyme. This CO 2 is now assimilated by the action of ru- bisco and the Calvin cycle enzymes. Because this method of CO 2 fixation was first discovered in stonecrops, perennial flowering plants of the family Crassulaceae, it is called crassulacean acid metabolism, and the plants are called CAM plants. Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria770 FIGURE 20–23 Carbon assimilation in C 4 plants. The C 4 pathway, in- volving mesophyll cells and bundle-sheath cells, predominates in plants of tropical origin. (a) Electron micrograph showing chloroplasts of adjacent mesophyll and bundle-sheath cells. The bundle-sheath cell contains starch granules. Plasmodesmata connecting the two cells are visible. (b) The C 4 pathway of CO 2 assimilation, which occurs through a four-carbon intermediate. NADPH + H + 3-PhosphoglycerateRibulose 1,5-bisphosphate AMP ATP P i Mesophyll cell PEP Pyruvate pyruvate phosphate dikinase PP i + + Malate NADP + malic enzyme Triose phosphates Bundle-sheath cell Plasma membranes Oxaloacetate Malate malate dehydrogenase NADPH + H + P i NADP + Pyruvate – Plasmodesmata PEP carboxylase H 2 O (in air) H + CO 2 HCO 3 CO 2 (b) (a) Mesophyll cell Plasmodesmata Bundle- sheath cell 8885d_c20_770 2/20/04 12:02 PM Page 770 mac76 mac76:385_reb: SUMMARY 20.2 Photorespiration and the C 4 and CAM Pathways ■ When rubisco uses O 2 rather than CO 2 as substrate, the 2-phosphoglycolate so formed is disposed of in an oxygen-dependent pathway. The result is increased consumption of O 2 —photorespiration or, more accurately, the oxidative photosynthetic carbon cycle or C 2 cycle. The 2-phosphoglycolate is converted to glyoxylate, to glycine, and then to serine in a pathway that involves enzymes in the chloroplast stroma, the peroxisome, and the mitochondrion. ■ In C 4 plants, the carbon-assimilation pathway minimizes photorespiration: CO 2 is first fixed in mesophyll cells into a four-carbon compound, which passes into bundle-sheath cells and releases CO 2 in high concentrations. The released CO 2 is fixed by rubisco, and the remaining reactions of the Calvin cycle occur as in C 3 plants. ■ In CAM plants, CO 2 is fixed into malate in the dark and stored in vacuoles until daylight, when the stomata are closed (minimizing water loss) and malate serves as a source of CO 2 for rubisco. 20.3 Biosynthesis of Starch and Sucrose During active photosynthesis in bright light, a plant leaf produces more carbohydrate (as triose phosphates) than it needs for generating energy or synthesizing pre- cursors. The excess is converted to sucrose and trans- ported to other parts of the plant, to be used as fuel or stored. In most plants, starch is the main storage form, but in a few plants, such as sugar beet and sugarcane, sucrose is the primary storage form. The synthesis of sucrose and starch occurs in different cellular com- partments (cytosol and plastids, respectively), and these processes are coordinated by a variety of regula- tory mechanisms that respond to changes in light level and photosynthetic rate. ADP-Glucose Is the Substrate for Starch Synthesis in Plant Plastids and for Glycogen Synthesis in Bacteria Starch, like glycogen, is a high molecular weight poly- mer of D-glucose in (H92511n4) linkage. It is synthesized in chloroplasts for temporary storage as one of the stable end products of photosynthesis, and for long-term stor- age it is synthesized in the amyloplasts of the nonpho- tosynthetic parts of plants—seeds, roots, and tubers (underground stems). The mechanism of glucose activation in starch syn- thesis is similar to that in glycogen synthesis. An acti- vated nucleotide sugar, in this case ADP-glucose, is formed by condensation of glucose 1-phosphate with ATP in a reaction made essentially irreversible by the pres- ence in plastids of inorganic pyrophosphatase (p. 502). Starch synthase then transfers glucose residues from ADP-glucose to preexisting starch molecules. Although it has generally been assumed that glucose is added to the nonreducing end of starch, as in glycogen synthesis (see Fig. 15–8), evidence now suggests that starch syn- thase has two equivalent active sites that alternate in in- serting a glucosyl residue onto the reducing end of the growing chain. This end remains covalently attached to the enzyme, first at one active site, then at the other (Fig. 20–24). Attachment to one active site effectively activates the reducing end of the growing chain for nu- cleophilic displacement of the enzyme by the attacking C-4 hydroxyl of a glucosyl moiety bound to the other ac- tive site, forming the (H92511n4) linkage characteristic of starch. The amylose of starch is unbranched, but amy- lopectin has numerous (H92511n6)-linked branches (see Fig. 7–15). Chloroplasts contain a branching enzyme, similar to glycogen-branching enzyme (see Fig. 15–9), that introduces the (H92511n6) branches of amylopectin. Taking into account the hydrolysis by inorganic py- rophosphatase of the PP i produced during ADP-glucose synthesis, the overall reaction for starch formation from glucose 1-phosphate is Starch n H11001 glucose 1-phosphate H11001 ATP 88n starch nH110011 H11001 ADP H11001 2P i H9004GH11032H11034H11005H1100250 kJ/mol Starch synthesis is regulated at the level of ADP-glucose formation, as discussed below. Many types of bacteria store carbohydrate in the form of glycogen (essentially highly branched starch), which they synthesize in a reaction analogous to that catalyzed by glycogen synthase in animals. Bacteria, like plant plastids, use ADP-glucose as the activated form of glucose, whereas animal cells use UDP-glucose. Again, the similarity between plastid and bacterial metabolism is consistent with the endosymbiont hypothesis for the origin of organelles (see Fig. 1–36). UDP-Glucose Is the Substrate for Sucrose Synthesis in the Cytosol of Leaf Cells Most of the triose phosphate generated by CO 2 fixation in plants is converted to sucrose (Fig. 20–25) or starch. In the course of evolution, sucrose may have been se- lected as the transport form of carbon because of its un- usual linkage between the anomeric C-1 of glucose and the anomeric C-2 of fructose. This bond is not hydrolyzed by amylases or other common carbohydrate-cleaving 20.3 Biosynthesis of Starch and Sucrose 771 8885d_c20_751–786 2/18/04 1:56 PM Page 771 mac76 mac76:385_reb: enzymes, and the unavailability of the anomeric carbons prevents sucrose from reacting nonenzymatically (as does glucose) with amino acids and proteins. Sucrose is synthesized in the cytosol, beginning with dihydroxyacetone phosphate and glyceraldehyde 3-phosphate exported from the chloroplast. After con- densation of two triose phosphates to form fructose 1,6- bisphosphate (catalyzed by aldolase), hydrolysis by fructose 1,6-bisphosphatase yields fructose 6-phosphate. Sucrose 6-phosphate synthase then catalyzes the reaction of fructose 6-phosphate with UDP-glucose to form sucrose 6-phosphate (Fig. 20–25). Finally, sucrose 6-phosphate phosphatase removes the phos- phate group, making sucrose available for export to other tissues. The reaction catalyzed by sucrose 6-phos- phate synthase is a low-energy process (H9004GH11032H11034 H11005 H110025.7 kJ/mol), but the hydrolysis of sucrose 6-phosphate to sucrose is sufficiently exergonic (H9004GH11032H11034 H11005 H1100216.5 kJ/mol) to make the overall synthesis of sucrose essentially irreversible. Sucrose synthesis is regulated and closely coordinated with starch synthesis, as we shall see. One remarkable difference between the cells of plants and animals is the absence in the plant cell cy- tosol of the enzyme inorganic pyrophosphatase, which catalyzes the reaction PP i H11001 H 2 O 88n 2P i H9004GH11032H11034 H11005 H1100219.2 kJ/mol For many biosynthetic reactions that liberate PP i , py- rophosphatase activity makes the process more favor- able energetically, tending to make these reactions ir- reversible. In plants, this enzyme is present in plastids but absent from the cytosol. As a result, the cytosol of leaf cells contains a substantial concentration of PP i — enough (~0.3 mM) to make reactions such as that cat- alyzed by UDP-glucose pyrophosphorylase (Fig. 15–7) readily reversible. Recall from Chapter 14 (p. 527) that the cytosolic isozyme of phosphofructokinase in plants uses PP i , not ATP, as the phosphoryl donor. Conversion of Triose Phosphates to Sucrose and Starch Is Tightly Regulated Triose phosphates produced by the Calvin cycle in bright sunlight, as we have noted, may be stored tem- porarily in the chloroplast as starch, or converted to su- crose and exported to nonphotosynthetic parts of the plant, or both. The balance between the two processes is tightly regulated, and both must be coordinated with the rate of carbon fixation. Five-sixths of the triose Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria772 Starch ADP-glucose – ADP X a – X b OH Each of the two reactive groups (X a , X b ) at the active site of starch synthase makes a nucleophilic attack on ADP-glucose, displacing ADP and forming a covalent attachment to C-1 of the glucose unit. The hydroxyl at C-4 of glucose 3 displaces X b from the disaccharide, forming a trisaccharide attached to X a . X b , now free, acquires glucose residue 4 from another ADP-glucose. The hydroxyl at C-4 of glucose 4 displaces X a , forming a tetrasaccharide, with its reducing end covalently attached to X b . Many repetitions of this sequence extend the oligosaccharide, adding glucose residues at its reducing end, with X a and X b alternately carrying the growing starch chain. When the chain reaches an appropriate length, it is separated from starch synthase. 1 ADP X a X b 3 ADP5 ADP2 2 1 : OH : OH X a X b ADP4 2 1 3 : OH Nonreducing end 5 : ADP6 X a X b 3 4 2 1 X a X b 34 2 1 The bond holding glucose residue 1 to X a undergoes nucleophilic attack by the OH at C-4 of glucose residue 2 on X b , forming an a(1 4)-disaccharide of residues 2 and 1. This remains attached through glucose 2 to X b . X a , now free, displaces ADP from another ADP- glucose and becomes attached to glucose 3. Starch synthase FIGURE 20–24 Starch synthesis. Starch synthesis proceeds by a two- site insertion mechanism, with ADP-glucose as the initial glucosyl donor. The two identical active sites on starch synthase alternate in displacing the growing chain from each other, and new glucosyl units are inserted at the reducing end of the growing chain. 8885d_c20_751–786 2/18/04 1:56 PM Page 772 mac76 mac76:385_reb: phosphate formed in the Calvin cycle must be recycled to ribulose 1,5-bisphosphate (Fig. 20–14); if more than one-sixth of the triose phosphate is drawn out of the cycle to make sucrose and starch, the cycle will slow or stop. However, insufficient conversion of triose phos- phate to starch or sucrose would tie up phosphate, leav- ing a chloroplast deficient in P i , which is also essential for operation of the Calvin cycle. The flow of triose phosphates into sucrose is reg- ulated by the activity of fructose 1,6-bisphosphatase (FBPase-1) and the enzyme that effectively reverses its action, PP i -dependent phosphofructokinase (PP-PFK-1; p. 527). These enzymes are therefore critical points for determining the fate of triose phosphates produced by photosynthesis. Both enzymes are regulated by fructose 2,6-bisphosphate (F2,6BP), which inhibits FBPase-1 and stimulates PP-PFK-1. In vascular plants, the con- centration of F2,6BP varies inversely with the rate of photosynthesis (Fig. 20–26). Phosphofructokinase-2, responsible for F2,6BP synthesis, is inhibited by dihy- droxyacetone phosphate or 3-phosphoglycerate and stimulated by fructose 6-phosphate and P i . During ac- tive photosynthesis, dihydroxyacetone phosphate is produced and P i is consumed, resulting in inhibition of PFK-2 and lowered concentrations of F2,6BP. This 20.3 Biosynthesis of Starch and Sucrose 773 O UDP-glucose sucrose 6-phosphate synthase 1 HOCH 2 HO H OH H OHO P UDP CH 2 O HOH H H HOH Fructose 6-phosphate UDP H HO O 2 HOCH 2 H HH 1 HO O CH 2 OH O HOH H H HOH Sucrose HOH O sucrose 6-phosphate phosphatase 2 HOCH 2 H OH HH 1 HO P i CH 2 OH O HOH H H HOH Sucrose 6-phosphate H O H11001 O CH 2 OH 2 H HO HO PCH 2 O CH 2 OH FIGURE 20–25 Sucrose synthesis. Sucrose is synthesized from UDP- glucose and fructose 6-phosphate, which are synthesized from triose phosphates in the plant cell cytosol by pathways shown in Figures 15–7 and 20–9. The sucrose 6-phosphate synthase of most plant species is allosterically regulated by glucose 6-phosphate and P i . FIGURE 20–26 Fructose 2,6-bisphosphate as regulator of sucrose syn- thesis. The concentration of the allosteric regulator fructose 2,6- bisphosphate in plant cells is regulated by the products of photosyn- thetic carbon assimilation and by P i . Dihydroxyacetone phosphate and 3-phosphoglycerate produced by CO 2 assimilation inhibit phospho- fructokinase-2 (PFK-2), the enzyme that synthesizes the regulator; P i stimulates PFK-2. The concentration of the regulator is therefore inversely proportional to the rate of photosynthesis. In the dark, the concentration of fructose 2,6-bisphosphate increases and stimulates the glycolytic enzyme PP i -dependent phosphofructokinase-1 (PP-PFK- 1), while inhibiting the gluconeogenic enzyme fructose 1,6- bisphosphatase (FBPase-1). When photosynthesis is active (in the light), the concentration of the regulator drops and the synthesis of fructose 6-phosphate and sucrose is favored. Chloroplast Fructose 2,6-bisphosphate ADP ATP PFK-2 FBPase-2 H 2 O P i Dark No photosynthesis P i Light Active photosynthesis 3-Phosphoglycerate and dihydroxyacetone phosphate Fructose 6-phosphate PP i P i P i FBPase-1 Triose phosphate Cytosol Dark Dark Fructose 1,6-bisphosphate PP-PFK-1 Sucrose Glycolysis Gluconeogenesis 8885d_c20_751–786 2/18/04 1:56 PM Page 773 mac76 mac76:385_reb: favors greater flux of triose phosphate into fructose 6- phosphate formation and sucrose synthesis. With this regulatory system, sucrose synthesis occurs when the level of triose phosphate produced by the Calvin cycle exceeds that needed to maintain the operation of the cycle. Sucrose synthesis is also regulated at the level of sucrose 6-phosphate synthase, which is allosterically activated by glucose 6-phosphate and inhibited by P i . This enzyme is further regulated by phosphorylation and dephosphorylation; a protein kinase phosphory- lates the enzyme on a specific Ser residue, making it less active, and a phosphatase reverses this inactiva- tion by removing the phosphate (Fig. 20–27). Inhibi- tion of the kinase by glucose 6-phosphate, and of the phosphatase by P i , strengthens the effects of these two compounds on sucrose synthesis. When hexose phos- phates are abundant, sucrose 6-phosphate synthase is activated by glucose 6-phosphate; when P i is elevated (as when photosynthesis is slow), sucrose synthesis is slowed. During active photosynthesis, triose phos- phates are converted to fructose 6-phosphate, which is rapidly equilibrated with glucose 6-phosphate by phosphohexose isomerase. Because the equilibrium lies far toward glucose 6-phosphate, as soon as fruc- tose 6-phosphate accumulates, the level of glucose 6-phosphate rises and sucrose synthesis is stimulated. The key regulatory enzyme in starch synthesis is ADP-glucose pyrophosphorylase (Fig. 20–28); it is activated by 3-phosphoglycerate (which accumulates during active photosynthesis) and inhibited by P i (which accumulates when light-driven condensation of ADP and P i slows). When sucrose synthesis slows, 3-phos- phoglycerate formed by CO 2 fixation accumulates, acti- vating this enzyme and stimulating the synthesis of starch. SUMMARY 20.3 Biosynthesis of Starch and Sucrose ■ Starch synthase in chloroplasts and amyloplasts catalyzes the addition of single glucose residues, donated by ADP-glucose, to the reducing end of a starch molecule by a two-step insertion mechanism. Branches in amylopectin are introduced by a second enzyme. ■ Sucrose is synthesized in the cytosol in two steps from UDP-glucose and fructose 1-phosphate. ■ The partitioning of triose phosphates between sucrose synthesis and starch synthesis is regulated by fructose 2,6-bisphosphate (F2,6BP), an allosteric effector of the enzymes that determine the level of fructose 6-phosphate. F2,6BP concentration varies inversely with the rate of photosynthesis, and F2,6BP inhibits the synthesis of fructose 6-phosphate, the precursor to sucrose. Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria774 Glucose 6-phosphate Glucose 6-phosphate (less active) (more active) CH 2 O SPS kinase H 2 O P i P i P i SPS phosphatase ADP ATP Sucrose 6-phosphate synthase Sucrose 6-phosphate synthase CH 2 OH P Photosynthesis Bright lightDim light or darkness ADP-glucose + PP i ADP-glucose pyrophosphorylase Glucose 1- phosphate + ATP 3-Phosphoglycerate slow ADP + P i ATP FIGURE 20–27 Regulation of sucrose phosphate synthase by phos- phorylation. A protein kinase (SPS kinase) specific for sucrose phos- phate synthase (SPS) phosphorylates a Ser residue in SPS, inactivating it; a specific phosphatase (SPS phosphatase) reverses this inhibition. The kinase is inhibited allosterically by glucose 6-phosphate, which also activates SPS allosterically. The phosphatase is inhibited by Pi, which also inhibits SPS directly. Thus when the concentration of glu- cose 6-phosphate is high as a result of active photosynthesis, SPS is activated and produces sucrose phosphate. A high P i concentration, which occurs when photosynthetic conversion of ADP to ATP is slow, inhibits sucrose phosphate synthesis. FIGURE 20–28 Regulation of ADP-glucose phosphorylase by 3- phosphoglycerate and P i . This enzyme, which produces the precursor for starch synthesis, is rate-limiting in starch production. The enzyme is stimulated allosterically by 3-phosphoglycerate (3-PGA) and inhib- ited by P i ; in effect, the ratio [3-PGA]/[P i ], which rises with increas- ing rates of photosynthesis, controls starch synthesis at this step. 8885d_c20_751–786 2/18/04 1:56 PM Page 774 mac76 mac76:385_reb: 20.4 Synthesis of Cell Wall Polysaccharides: Plant Cellulose and Bacterial Peptidoglycan Cellulose is a major constituent of plant cell walls, pro- viding strength and rigidity and preventing the swelling of the cell and rupture of the plasma membrane that might result when osmotic conditions favor water entry into the cell. Each year, worldwide, plants synthesize more than 10 11 metric tons of cellulose, making this simple polymer one of the most abundant compounds in the biosphere. The structure of cellulose is simple: lin- ear polymers of thousands of H9252(1n4)-linked D-glucose units, assembled into bundles of about 36 chains, which aggregate side by side to form a microfibril (Fig. 20–29). The biosynthesis of cellulose is less well understood than that of glycogen or starch. As a major component of the plant cell wall, cellulose must be synthesized from intracellular precursors but deposited and assembled outside the plasma membrane. The enzymatic machin- ery for initiation, elongation, and export of cellulose chains is more complicated than that needed to syn- thesize starch or glycogen (which are not exported). Bacteria face a similar set of problems when they syn- thesize the complex polysaccharides that make up their cell walls, and they may employ some of the same mech- anisms to solve these problems. Cellulose Is Synthesized by Supramolecular Structures in the Plasma Membrane The complex enzymatic machinery that assembles cel- lulose chains spans the plasma membrane, with one part positioned to bind the substrate, UDP-glucose, in the cytosol and another part extending to the outside, re- sponsible for elongating and crystallizing cellulose mol- ecules in the extracellular space. Freeze-fracture elec- tron microscopy shows these terminal complexes, also called rosettes, to be composed of six large parti- cles arranged in a regular hexagon (Fig. 20–30). Several proteins, including the catalytic subunit of cellulose synthase, make up the terminal complex. Cellulose synthase has not been isolated in its active form, but its amino acid sequence has been determined from the nu- cleotide sequence of the gene that encodes it. From the primary structure we can use hydropathy plots (see Fig. 11–11) to deduce that the enzyme has eight trans- membrane segments, connected by short loops on the outside, and several longer loops exposed to the cytosol. Much of the recent progress in understanding cellulose synthesis stems from genetic and molecular genetic studies of the plant Arabidopsis thaliana, which is es- pecially amenable to genetic dissection and whose genome has been sequenced. 20.4 Synthesis of Cell Wall Polysaccharides 775 Cellulose microfibrils in plant cell wall Microfibril b(1 4) Cellulose chains H HOH O O H CH 2 OH HOH O H HOH H CH 2 OH HOH O O FIGURE 20–29 Cellulose structure. The plant cell wall is made up in part of cellulose molecules arranged side by side to form paracrys- talline arrays—cellulose microfibrils. Many microfibrils combine to form a cellulose fiber, seen in the scanning electron microscope as a structure 5 to 12 nm in diameter, laid down on the cell surface in sev- eral layers distinguishable by the different orientations of their fibers. FIGURE 20–30 Rosettes. The outside surface of the plant plasma mem- brane in a freeze-fractured sample, viewed here with electron mi- croscopy, contains many hexagonal arrays of particles about 10 nm in diameter, believed to be composed of cellulose synthase molecules and associated enzymes. 8885d_c20_751–786 2/18/04 1:56 PM Page 775 mac76 mac76:385_reb: New cellulose chains appear to be initiated by the formation of a lipid-linked intermediate unlike anything involved in starch or glycogen synthesis. Glucose is trans- ferred from UDP-glucose to a membrane lipid, probably the plant sterol sitosterol (Fig. 20–31), on the inner face of the plasma membrane. Here, intracellular cellulose synthase adds several more glucose residues to the first one, in (H92521n4) linkage, forming a short oligosaccharide chain attached to the sitosterol (sitosterol dextrin). Next, the whole sitosterol dextrin flips across to the outer face of the plasma membrane, where most of the polysac- charide chain is removed by endo-1,4-H9252-glucanase. The shortened sitosterol dextrin primer now associates, per- haps covalently, with another form of cellulose synthase. Presumably this entire process occurs in the rosettes. Whether each of the 36 cellulose chains is initiated on its own lipid primer, or the primer recycles to start a num- ber of chains, is not yet clear. In either case, the second form of cellulose synthase extends the polymer to 500 to 15,000 glucose units, extruding it onto the outer surface of the cell. The action of the enzyme is processive: one enzyme molecule adds many glucose units before re- leasing the growing cellulose chain. The direction of chain growth (whether addition occurs at the reducing end or at the nonreducing end) has not been established. The finished cellulose is in the form of crystalline microfibrils (Fig. 20–29), each consisting of 36 separate cellulose chains lying side by side, all with the same (parallel) orientation of nonreducing and reducing ends. It seems likely that each particle in the rosette synthe- sizes six separate cellulose chains simultaneously and in parallel with the chains made by the other five particles, so that 36 polymers arrive together on the outer surface of the cell, already aligned and ready to crystallize as a microfibril of the cell wall. When the 36 polymers reach some critical length, their synthesis is terminated by an unknown mechanism; crystallization into a microfibril follows. In addition to its catalytic subunit, cellulose syn- thase may have subunits that mediate extrusion of the polysaccharide chain (the pore subunit) and crystal- lization of the polysaccharide chains outside the cell Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria776 Sitosterol OH UDP Glucose UDP Cytosol short oligosaccharide primer retained; sitosterol flips to cytosolic face Lengthening molecule exits cell Extracellular space endo-1,4-b- glucanase cellulose synthase : UDP cellulose synthase UDP n (( n (( O O Sitosterol dextrin 1 2 3 molecule flips from inner to outer leaflet 4 5 7 6 UDP UDP n (( n (( FIGURE 20–31 Lipid primer for cellulose synthesis. This proposed pathway begins with 1 the transfer of a glucosyl residue from UDP- glucose to a lipid “primer” (probably sitosterol) in the inner leaflet of the plasma membrane. After this initiation, 2 the chain of carbohy- drate is elongated by transfer of glucosyl residues from UDP-glucose, until 3 a critical length of oligosaccharide is reached. 4 The sitos- terol with its attached oligosaccharide now flips from the inner leaflet to the outer leaflet. 5 An endo-1,4-H9252-glucanase separates the grow- ing chain from a short oligonucleotide still attached to the lipid. As it is pushed out of the cell, 6 the lipid-free polymer of glucosyl residues (the glucan acceptor) is further extended by the addition of glucosyl residues from UDP-glucose, catalyzed by cellulose synthase. 7 The lipid-linked oligosaccharide returns to serve as the primer for another chain of cellulose. 8885d_c20_751–786 2/18/04 1:56 PM Page 776 mac76 mac76:385_reb: (the crystallization subunit). The potent herbicide CGA 325H11032615, which specifically inhibits cellulose synthesis, causes rosettes to fall apart; the small amount of cellu- lose still synthesized remains tightly, perhaps covalently, bound to the catalytic subunit of cellulose synthase. The inhibitor may act by dissociating the catalytic subunit from the pore and crystallization subunits, preventing the later stages of cellulose synthesis. The UDP-glucose used for cellulose synthesis is generated from sucrose produced during photosynthe- sis, by the reaction catalyzed by sucrose synthase (named for the reverse reaction): Sucrose H11001 UDP 88n UDP-glucose H11001 fructose In one proposed model, cellulose synthase spans the plasma membrane and uses cytosolic UDP-glucose as the precursor for extracellular cellulose synthesis. In another, a membrane-bound form of sucrose synthase forms a complex with cellulose synthase, feeding UDP-glucose from sucrose directly into cell wall synthesis (Fig. 20–32). In the activated precursor of cellulose (UDP- glucose), the glucose is H9251-linked to the nucleotide, but in the product (cellulose), glucose residues are H9252(1n4)- linked, so there is an inversion of configuration at the anomeric carbon (C-1) as the glycosidic bond forms. Glycosyltransferases that invert configuration are gen- erally assumed to use a single-displacement mechanism, with nucleophilic attack by the acceptor species at the anomeric carbon of the donor sugar (UDP-glucose). Certain bacteria (Acetobacter, Agrobacteria, Rhi- zobia, and Sarcina) and many simple eukaryotes also carry out cellulose synthesis, apparently by a mecha- nism similar to that in plants. If the bacteria use a mem- brane lipid to initiate new chains, it cannot be a sterol— bacteria do not contain sterols. Lipid-Linked Oligosaccharides Are Precursors for Bacterial Cell Wall Synthesis Like plants, many bacteria have thick, rigid extracellu- lar walls that protect them from osmotic lysis. The pep- tidoglycan that gives bacterial envelopes their strength and rigidity is an alternating linear copolymer of N- acetylglucosamine (GlcNAc) and N-acetylmuramic acid (Mur2Ac), linked by (H92521n4) glycosidic bonds and cross-linked by short peptides attached to the Mur2Ac (Fig. 20–33). During assembly of the polysaccharide backbone of this complex macromolecule, both GlcNAc and Mur2Ac are activated by attachment of a uridine 20.4 Synthesis of Cell Wall Polysaccharides 777 Crystallization subunit Sucrose Cytosol Fructose UDP UDP- glucose Cellulose synthase Sucrose synthase Pore subunit Catalytic subunit FIGURE 20–32 A plausible model for the structure of cellulose syn- thase. The enzyme complex includes a catalytic subunit with eight transmembrane segments and several other subunits that are presumed to act in threading cellulose chains through the catalytic site and out of the cell, and in the crystallization of 36 cellulose strands into the paracrystalline microfibrils shown in Figure 20–29. FIGURE 20–33 Peptidoglycan structure. This is the peptidoglycan of the cell wall of Staphylococcus aureus, a gram-positive bacterium. Peptides (strings of colored spheres) covalently link N-acetylmuramic acid residues in neighboring polysaccharide chains. Note the mixture of L and D amino acids in the peptides. Gram-positive bacteria such as S. aureus have a pentaglycine chain in the cross-link. Gram-negative bacteria, such as E. coli, lack the pentaglycine; instead, the terminal D-Ala residue of one tetrapeptide is attached directly to a neighbor- ing tetrapeptide through either L-Lys or a lysine-like amino acid, di- aminopimelic acid. Staphylococcus aureus Reducing end L-Ala D-Glu L-Lys D-Ala N-Acetylglucosamine (GlcNAc) Pentaglycine cross-link (b14) N-Acetylmuramic acid (Mur2Ac) 8885d_c20_751–786 2/18/04 1:56 PM Page 777 mac76 mac76:385_reb: nucleotide at their anomeric carbons. First, GlcNAc 1- phosphate condenses with UTP to form UDP-GlcNAc (Fig. 20–34, step 1 ), which reacts with phospho- enolpyruvate to form UDP-Mur2Ac (step 2 ); five amino acids are then added (step 3 ). The Mur2Ac-pentapep- tide moiety is transferred from the uridine nucleotide to the membrane lipid dolichol, a long-chain isoprenoid alcohol (see Fig. 10–22f) (step 4 ), and a GlcNAc residue is donated by UDP-GlcNAc (step 5 ). In many bacteria, five glycines are added in peptide linkage to the amino group of the Lys residue of the pentapeptide (step 6 ). Finally, this disaccharide decapeptide is added to the nonreducing end of an existing peptidoglycan molecule (step 7 ). A transpeptidation reaction cross- links adjacent polysaccharide chains (step 8 ), con- tributing to a huge, strong, macromolecular wall around the bacterial cell. Many of the most effective antibiotics in use today act by inhibiting reactions in the synthesis of the peptidoglycan (Box 20–1). Many other oligosaccharides and polysaccharides are synthesized by similar routes in which sugars are ac- tivated for subsequent reactions by attachment to nu- cleotides. In the glycosylation of proteins, for example (see Fig. 27–34), the precursors of the carbohydrate moieties include sugar nucleotides and lipid-linked oligosaccharides. Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria778 + UTPGlcNAc-1 UDP-GlcNAc + PP i PEP NADPH NADP + UDP-Mur2Ac UDP-GlcNAc L-Alanine 1 2 3 5 6 7 D-Glutamate L-Lysine D-Ala–D-Alanine Dolichol UDP PP 4 Dolichol UMP P P Dolichol 5 L-Glycine D-Alanine 8 transpeptidase PP DolicholPP Dolichol Existing peptidoglycan “primer” Another peptidoglycan chain Penicillins PP Mature, fully cross-linked peptidoglycan FIGURE 20–34 Synthesis of bacterial peptidoglycan. In the early steps of this pathway ( 1 through 4 ), N-acetylglucosamine (GlcNAc) and N-acetylmuramic acid (Mur2Ac) are activated by attachment of a uridine nucleotide (UDP) to their anomeric carbons and, in the case of Mur2Ac, of a long-chain isoprenyl alcohol (dolichol) through a phosphodiester bond. These acti- vating groups participate in the formation of glycosidic linkages; they serve as excellent leaving groups. After 5 , 6 assembly of a disaccharide with a peptide side chain (10 amino acid residues), 7 this precursor is transferred to the nonreducing end of an existing pepti- doglycan chain, which serves as a primer for the polymerization reaction. Finally, 8 in a transpeptida- tion reaction between the peptide side chains on two different peptidoglycan molecules, a Gly residue at the end of one chain displaces a terminal D-Ala in the other chain, forming a cross-link. This transpeptidation reaction is inhibited by the penicillins, which kill bacteria by weakening their cell walls. 8885d_c20_751–786 2/18/04 1:56 PM Page 778 mac76 mac76:385_reb: 20.4 Synthesis of Cell Wall Polysaccharides 779 BOX 20–1 BIOCHEMISTRY IN MEDICINE The Magic Bullet versus the Bulletproof Vest: Penicillin and H9252-Lactamase Because peptidoglycans are unique to bacterial cell walls, with no known homologous structures in mam- mals, the enzymes responsible for their synthesis are ideal targets for antibiotic action. Antibiotics that hit specific bacterial targets are sometimes called “magic bullets.” Penicillin and its many synthetic analogs have been used to treat bacterial infections since these drugs came into wide application in World War II. Penicillins and related antibiotics contain the H9252- lactam ring (Fig. 1), variously modified. All penicillins have a thiazolidine ring attached to the H9252-lactam, but they differ in the substituent at position 6, which ac- counts for the different pharmacological properties of the penicillins. For example, penicillin V is acid stable and can be administered orally, but methicillin is acid labile and must be given intravenously or intramus- cularly. However, methicillin resists breakdown by bacterial enzymes (H9252-lactamases) whereas many other penicillins do not. The H9252-lactams have many of the properties that make a good drug. First, they target a metabolic pathway present in bacteria but not in people. Second, they have half-lives in the body long enough to be clinically useful. Third, they reach thera- CCN H C H O C O R S CH 3 CH 3 CH COOH C N CH 3 CH 3 H 2 O C H H Enz Ser OH CN H C C O R S CH COOH C N H O Trans- peptidase b-Lactamase Ser O CH 3 CH 3 C HH CN H C C O R S CH COOH C N H O H11002 O (a) (b) Stably derivatized, inactive transpeptidase Inactive penicillin H CH 3 CH 3 C H H CN H C C O R S CH COOH C N H O Ser O Penicillin b-Lactamase FIGURE 2 H CN H C O C O R R groups S CH 3 CH 2 CH 3 COOH CH C N Side chain Thiazolidine ring General structure of penicillins O CH 2 OCH 3 OCH 3 Methicillin Penicillin V Penicillin G (Benzylpenicillin) 1 5 6 7 4 3 2 b-Lactam ring H C# ! ! FIGURE 1 peutic concentrations in most, if not all, tissues and organs. Finally, they are effective against a broad range of bacterial species. Penicillins block formation of the peptide cross- links in peptidoglycans, acting as mechanism-based (suicide) inhibitors. The normal catalytic mechanism of the target enzyme activates the inhibitor, which then covalently modifies a critical residue in the active site. Transpeptidases employ a reaction mechanism (involv- ing Ser residues) similar to that of chymotrypsin (see Fig. 6–21); the reaction activates H9252-lactams such as penicillin, which in turn inactivate the transpeptidases. After penicillin enters the transpeptidase active site, the proton on the hydroxyl group of an active-site Ser res- idue is abstracted to the nitrogen of the H9252-lactam ring, and the activated oxygen of the Ser hydroxyl attacks the carbonyl carbon at position 7 of the H9252-lactam, open- ing the ring and forming a stable penicilloyl-enzyme derivative that inactivates the enzyme (Fig. 2a). Widespread use of antibiotics has driven the se- lection and evolution of antibiotic resistance in many pathogenic bacteria. The most important mechanism of resistance is inactivation of the antibiotic by enzymatic hydrolysis of the lactam ring, catalyzed by bacterial (continued on next page) 8885d_c20_751–786 2/18/04 1:56 PM Page 779 mac76 mac76:385_reb: Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria780 SUMMARY 20.4 Synthesis of Cell Wall Polysaccharides: Plant Cellulose and Bacterial Peptidoglycan ■ Cellulose synthesis takes place in terminal complexes (rosettes) in the plasma membrane. Each cellulose chain begins as a sitosterol dextrin formed inside the cell. It then flips to the outside, where the oligosaccharide portion is transferred to cellulose synthase in the rosette and is then extended. Each rosette produces 36 separate cellulose chains simultaneously and in parallel. The chains crystallize into one of the microfibrils that form the cell wall. ■ Synthesis of the bacterial cell wall peptidoglycan also involves lipid-linked oligosaccharides formed inside the cell and flipped to the outside for assembly. 20.5 Integration of Carbohydrate Metabolism in the Plant Cell Carbohydrate metabolism in a typical plant cell is more complex in several ways than that in a typical animal cell. The plant cell carries out the same processes that generate energy in animal cells (glycolysis, citric acid cycle, and oxidative phosphorylation); it can generate hexoses from three- or four-carbon compounds by glu- coneogenesis; it can oxidize hexose phosphates to pen- tose phosphates with the generation of NADPH (the ox- idative pentose phosphate pathway); and it can produce a polymer of (H92511n4)-linked glucose (starch) and de- grade it to generate hexoses. But besides these carbo- hydrate transformations that it shares with animal cells, the photosynthetic plant cell can fix CO 2 into organic compounds (the rubisco reaction); use the products of fixation to generate trioses, hexoses, and pentoses (the Calvin cycle); and convert acetyl-CoA generated from fatty acid breakdown to four-carbon compounds (the glyoxylate cycle) and the four-carbon compounds to hexoses (gluconeogenesis). These processes, unique to the plant cell, are segregated in several compartments not found in animal cells: the glyoxylate cycle in gly- oxysomes, the Calvin cycle in chloroplasts, starch syn- thesis in amyloplasts, and organic acid storage in vac- uoles. The integration of events among these various compartments requires specific transporters in the membranes of each organelle, to move products from one organelle to another or into the cytosol. Gluconeogenesis Converts Fats and Proteins to Glucose in Germinating Seeds Many plants store lipids and proteins in their seeds, to be used as sources of energy and as biosynthetic pre- cursors during germination, before photosynthetic mechanisms have developed. Active gluconeogenesis in germinating seeds provides glucose for the synthesis of sucrose, polysaccharides, and many metabolites derived from hexoses. In plant seedlings, sucrose provides much of the chemical energy needed for initial growth. We noted earlier (Chapter 14) that animal cells can carry out gluconeogenesis from three- and four- carbon precursors, but not from the two acetyl carbons BOX 20–1 BIOCHEMISTRY IN MEDICINE (continued from previous page) H9252-lactamases, which provide bacteria with a bullet- proof vest (Fig. 2b). A H9252-lactamase forms a temporary covalent adduct with the carboxyl group of the opened H9252-lactam ring, which is immediately hydrolyzed, regen- erating active enzyme. One approach to circumvent- ing antibiotic resistance of this type is to synthesize penicillin analogs, such as methicillin, that are poor substrates for H9252-lactamases. Another approach is to administer along with antibiotics a H9252-lactamase in- hibitor such as clavulanate or sulbactam. Antibiotic resistance is a significant threat to pub- lic health. Some bacterial infections are now essen- tially untreatable with antibiotics. By the early 1990s, 20% to 40% of Staphylococcus aureus (the causative agent of “staph” infections) was resistant to methicillin, and 32% of Neisseria gonorrhoeae (the causative agent of gonorrhea) was resistant to penicillin. By 1986, 32% of Shigella (a pathogen responsible for se- vere forms of dysentery, some with a lethality of up to 15%) was resistant to ampicillin. Significantly, many of these pathogens are also resistant to many other antibiotics. In the future, we will need to develop new drugs that circumvent the bacterial resistance mech- anisms or that act on different bacterial targets. 8885d_c20_751–786 2/18/04 1:56 PM Page 780 mac76 mac76:385_reb: 20.5 Integration of Carbohydrate Metabolism in the Plant Cell 781 of acetyl-CoA. Because the pyruvate dehydrogenase reaction is effectively irreversible (pp. 602–603), animal cells have no way to convert acetyl-CoA to pyruvate or oxaloacetate. Unlike animals, plants and some micro- organisms can convert acetyl-CoA derived from fatty acid oxidation to glucose. Some of the enzymes essen- tial to this conversion are sequestered in glyoxysomes, where glyoxysome-specific isozymes of H9252-oxidation break down fatty acids to acetyl-CoA (see Fig. 16–22). The physical separation of the glyoxylate cycle and H9252-oxidation enzymes from the mitochondrial citric acid cycle enzymes prevents further oxidation of acetyl-CoA to CO 2 . Instead, the acetyl-CoA is converted to succi- nate in the glyoxylate cycle (see Fig. 16–20). The suc- cinate passes into the mitochondrial matrix, where it is converted by citric acid cycle enzymes to oxaloacetate, which moves into the cytosol. Cytosolic oxaloacetate is converted by gluconeogenesis to fructose 6-phos- phate, the precursor of sucrose. Thus the integration of reaction sequences in three subcellular compartments is required for the production of fructose 6-phosphate or sucrose from stored lipids. Because only three of the four carbons in each molecule of oxaloacetate are converted to hexose in the cytosol, about 75% of the carbon in the fatty acids stored as seed lipids is con- verted to carbohydrate by the combined pathways of Figure 20–35. The other 25% is lost as CO 2 in the conversion of oxaloacetate to phosphoenolpyruvate. Hydrolysis of storage triacylglycerols also produces glycerol 3-phosphate, which can enter the gluconeo- genic pathway after its oxidation to dihydroxyacetone phosphate (Fig. 20–36). Glucogenic amino acids (see Table 14–4) derived from the breakdown of stored seed proteins also yield precursors for gluconeogenesis, following transamina- tion and oxidation to succinyl-CoA, pyruvate, oxaloac- etate, fumarate, and H9251-ketoglutarate (Chapter 18)—all good starting materials for gluconeogenesis. Pools of Common Intermediates Link Pathways in Different Organelles Although we have described metabolic transformations in plant cells in terms of individual pathways, these pathways interconnect so completely that we should in- stead consider pools of metabolic intermediates shared among these pathways and connected by readily re- versible reactions (Fig. 20–37). One such metabolite pool includes the hexose phosphates glucose 1-phos- phate, glucose 6-phosphate, and fructose 6-phosphate; a second includes the 5-phosphates of the pentoses ri- bose, ribulose, and xylulose; a third includes the triose phosphates dihydroxyacetone phosphate and glycer- aldehyde 3-phosphate. Metabolite fluxes through these Glucose 6-phosphate Fructose 6-phosphate Sucrose Oxaloacetate Malate Succinate Acetyl-CoA Fatty acid oxidation Phosphoenolpyruvate Glyoxysome Mitochondrion Cytosol gluconeogenesis H11002 OOC CH 2 CH 2 COO H11002 CH 3 C O S-CoA Acetyl-CoA a b Citric acid cycle Glyoxylate cycle Succinate Fumarate Isocitrate Oxaloacetate Citrate Malate Succinyl-CoA Oxaloacetate CH 2 CO 2 C O H11002 OOC COO H11002 -Ketoglutarate Citrate Isocitrate Glyoxylate CH 3 C O S-CoA FIGURE 20–35 Conversion of stored fatty acids to sucrose in germi- nating seeds. This pathway begins in glyoxysomes. Succinate is pro- duced and exported to mitochondria, where it is converted to oxalo- acetate by enzymes of the citric acid cycle. Oxaloacetate enters the cytosol and serves as the starting material for gluconeogenesis and for the synthesis of sucrose, the transport form of carbon in plants. 8885d_c20_751–786 2/18/04 1:56 PM Page 781 mac76 mac76:385_reb: Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria782 CH 2 O C S-CoA CH 2 CH 3 O C O C O CH O C O CH 2 Triacylglycerol Fatty acids Glycerol Glycerol 3-phosphate NADH H11001 H H11001 NAD H11001 Acetyl-CoA Dihydroxyacetone phosphate P CH 2 OH H9252 oxidation lipase glycerol kinase O C O CHOH CH 2 OH CHOH CH 2 OH ATP ADP CH 2 OH CH 2 OP O glycerol 3-phosphate dehydrogenase FIGURE 20–36 Conversion of the glycerol moiety of triacylglycerols to sucrose in germinating seeds. The glycerol of triacylglycerols is ox- idized to dihydroxyacetone phosphate, which enters the gluconeo- genic pathway at the triose phosphate isomerase reaction. Starch ADP-Glucose UDP-Glucose Sucrose Glucose 1-phosphate Glucose 6-phosphate Frucose 6-phosphate 6-Phosphogluconate Fructose 1,6-bisphosphate Ribose 5-phosphate Glyceraldehyde 3-phosphate Dihydroxyacetone phosphate Ribulose 5-phosphate Xylulose 5-phosphate FIGURE 20–37 Pools of pentose phosphates, triose phosphates, and hexose phosphates. The compounds in each pool are readily intercon- vertible by reactions that have small standard free-energy changes. When one component of the pool is temporarily depleted, a new equilibrium is quickly established to replenish it. Movement of the sugar phosphates between intracellular compartments is limited; specific transporters must be present in an organelle membrane. pools change in magnitude and direction in response to changes in the circumstances of the plant, and they vary with tissue type. Transporters in the membranes of each organelle move specific compounds in and out, and the regulation of these transporters presumably influences the degree to which the pools mix. During daylight hours, triose phosphates produced in leaf tissue by the Calvin cycle move out of the chloro- plast and into the cytosolic hexose phosphate pool, where they are converted to sucrose for transport to nonphotosynthetic tissues. In these tissues, sucrose is converted to starch for storage or is used as an energy source via glycolysis. In growing plants, hexose phos- phates are also withdrawn from the pool for the syn- thesis of cell walls. At night, starch is metabolized by glycolysis to provide energy, essentially as in nonphoto- synthetic organisms, and NADPH and ribose 5-phos- phate are obtained through the oxidative pentose phos- phate pathway. SUMMARY 20.5 Integration of Carbohydrate Metabolism in the Plant Cell ■ Plants can synthesize sugars from acetyl-CoA, the product of fatty acid breakdown, by the combined actions of the glyoxylate cycle and gluconeogenesis. ■ The individual pathways of carbohydrate metabolism in plants overlap extensively; they share pools of common intermediates, including hexose phosphates, pentose phosphates, and triose phosphates. Transporters in the membranes of chloroplasts, mitochondria, amyloplasts, and peroxisomes mediate the movement of sugar phosphates between organelles. The direction of metabolite flow through the pools changes from day to night. 8885d_c20_751–786 2/18/04 1:56 PM Page 782 mac76 mac76:385_reb: Chapter 20 Further Reading 783 Key Terms Calvin cycle 752 plastids 752 chloroplast 752 amyloplast 752 carbon-fixation reaction 753 ribulose 1,5-bisphosphate 753 3-phosphoglycerate 753 pentose phosphate pathway 753 reductive pentose phosphate pathway 753 C 3 plants 754 ribulose 1,5-bisphosphate carboxylase/oxygenase (rubisco) 754 rubisco activase 757 transaldolase 759 transketolase 759 sedoheptulose 1,7-bisphosphate 760 ribulose 5-phosphate 760 carbon-assimilation reactions 763 thioredoxin 764 ferredoxin-thioredoxin reductase 764 photorespiration 766 2-phosphoglycolate 766 glycolate pathway 767 oxidative photosynthetic carbon cycle (C 2 cycle) 769 C 4 plants 769 phosphoenolpyruvate carboxylase 769 malic enzyme 769 pyruvate phosphate dikinase 769 CAM plants 770 nucleotide sugars 771 ADP-glucose 771 starch synthase 771 sucrose 6-phosphate synthase 772 fructose 2,6-bisphosphate 773 ADP-glucose pyrophosphorylase 774 cellulose synthase 775 peptidoglycan 777 metabolite pools 781 Terms in bold are defined in the glossary. Further Reading General References Blankenship, R.E. (2002) Molecular Mechanisms of Photosynthesis, Blackwell Science, Oxford. Very readable, well-illustrated, intermediate-level treatment of all aspects of photosynthesis, including the carbon metabolism covered in this chapter and the light-driven reactions described in Chapter 19. Buchanan, B.B., Gruissem, W., & Jones, R.L. (eds) (2000) Biochemistry and Molecular Biology of Plants, American Society of Plant Physiology, Rockville, MD. This wonderful book—up to date and authoritative—covers all aspects of plant biochemistry and molecular biology. The fol- lowing chapters cover carbohydrate synthesis in greater depth: Malkin, R. & Niyogi, K., Chapter 12, Photosynthesis (pp. 568–629); Dennis, D.T. & Blakeley, S.D., Chapter 13, Carbo- hydrate Metabolism (pp. 630–675); Siedow, J.N. & Day, D.A., Chapter 14, Respiration and Photorespiration (pp. 676–729). Heldt, H.-W. (1997) Plant Biochemistry and Molecular Biology, Oxford University Press, Oxford. Another excellent textbook of plant biochemistry. Especially useful are Chapter 6, Photosynthetic CO 2 Assimilation by the Calvin Cycle; Chapter 7, Photorespiration; and Chapter 9, Polysaccharides. Photosynthetic Carbohydrate Synthesis Andersson, I., Knight, S., Schneider, G., Lindqvist, Y., Lundqvist, T., Br?ndén, C.-I., & Lorimer, G.H. (1989) Crystal structure of the active site of ribulose-bisphosphate carboxylase. Nature 337, 229–234. Benson, A.A. (2002) Following the path of carbon in photosyn- thesis: a personal story—history of photosynthesis. Photosynth. Res. 73, 31–49. Cleland, W.W., Andrews, T.J., Gutteridge, S., Hartman, F.C., & Lorimer, G.H. (1998) Mechanism of rubisco—the carbamate as general base. Chem. Rev. 98, 549–561. Review with a special focus on the carbamate at the active site. Dietz, K.J., Link, G., Pistorius, E.K., & Scheibe, R. (2002) Redox regulation in oxygenic photosynthesis. Prog. Botany 63, 207–245. Flügge, U.-I. (1999) Phosphate translocaters in plastids. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50, 27–45. Review of the transporters that carry P i and various sugar phosphates across plastid membranes. Fridlyand, L.E. & Scheibe R. (1999) Homeostatic regulation upon changes of enzyme activities in the Calvin cycle as an example for general mechanisms of flux control: what can we expect from transgenic plants? Photosynth. Res. 61, 227–239. Hartman, F.C. & Harpel, M.R. (1994) Structure, function, regulation and assembly of D-ribulose-1,5-bisphosphate carboxylase/oxygenase. Annu. Rev. Biochem. 63, 197–234. Horecker, B.L. (2002) The pentose phosphate pathway. J. Biol. Chem. 277, 47,965–47,971. Portis, A.R., Jr. (2003) Rubisco activase: rubisco’s catalytic chaperon. Photosynth. Res. 75, 11–27. Structure, regulation, mechanism, and importance of rubisco activase. Raven, J. A. & Girard-Bascou, J. (2001) Algal model systems and the elucidation of photosynthetic metabolism. J. Phycol. 37, 943–950. Short, intermediate-level review. Schneider, G., Lindqvist, Y., Br?ndén, C.-I., & Lorimer, G. (1986) Three-dimensional structure of ribulose-1,5-bisphosphate carboxylase/oxygenase from Rhodospirillum rubrum at 2.9 ? resolution. EMBO J. 5, 3409–3415. Smith, A.M., Denyer, K., & Martin, C. (1997) The synthesis of the starch granule. Annu. Rev. Plant Physiol. Plant Mol. Biol. 48, 67–87. Review of the role of ADP-glucose pyrophosphorylase in the synthesis of amylose and amylopectin in starch granules. 8885d_c20_783 2/20/04 12:03 PM Page 783 mac76 mac76:385_reb: Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria784 Spreitzer, R.J. & Salvucci, M.E. (2002). Rubisco: structure, regulatory interactions, and possibilities for a better enzyme. Annu. Rev. Plant Biol. 53, 449–475. Advanced review on rubisco and rubisco activase. Stitt, M. (1995) Regulation of metabolism in transgenic plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 46, 341–368. Review of the genetic approaches to defining points of regulation in vivo. Tabita, F.R. (1999) Microbial ribulose 1,5-bisphosphate carboxy- lase/oxygenase: a different perspective. Photosynth. Res. 60, 1–28. Discussion of biochemical and genetic studies of the microbial rubisco, and comparison with the enzyme from plants. Wolosiuk, R., Ballicora, M., & Hagelin, K. (1993) The reduc- tive pentose phosphate cycle for photosynthetic CO 2 assimilation: enzyme modulation. FASEB J. 7, 622–637. Photorespiration and the C 4 and CAM Pathways Douce, R., Bourguignon, J., Neuburger, M., & Rébeillé, F. (2001) The glycine decarboxylase system: a fascinating complex. Trends Plant Sci. 6, 167–176. Intermediate-level description of the structure and the reaction mechanism of the enzyme. Douce, R. & Neuberger, M. (1999) Biochemical dissection of photorespiration. Curr. Opin. Plant Biol. 2, 214–222. Hatch, M.C. (1987) C 4 photosynthesis: a unique blend of modified biochemistry, anatomy and ultrastructure. Biochim. Biophys. Acta 895, 81–106. Intermediate-level review by one of the discoverers of the C 4 pathway. Tolbert, N.E. (1997) The C 2 oxidative photosynthetic carbon cycle. Annu. Rev. Plant Physiol. Plant Mol. Biol. 48, 1–25. A fascinating personal account of the development of our understanding of photorespiration, by one who was central in this development. Biosynthesis of Starch and Sucrose Ball, S.G., van de Wal, M.H.B.J., & Visser, R.G.F. (1998) Progress in understanding the biosynthesis of amylose. Trends Plant Sci. 3, 462–467. Delmer, D.P. (1999) Cellulose biosynthesis: exciting times for a difficult field of study. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50, 245–276. Doblin, M.S., Kurek, I., Javob-Wilk, D., & Delmer, D.P. (2002) Cellulose biosynthesis in plants: from genes to rosettes. Plant Cell Physiol. 43, 1407–1420. Errington, J., Daniel, R.A., & Scheffers, D.J. (2003) Cyto- kinesis in bacteria. Microbiol. Mol. Biol. Rev. 67, 52–65. Fernie, A.R., Willmitzer, L., & Trethewey, R.N. (2002) Sucrose to starch: a transition in molecular plant physiology. Trends Plant Sci. 7, 35–41. Intermediate-level review of the genes and proteins involved in starch synthesis in plant tubers. Huber, S.C. & Huber, J.L. (1996) Role and regulation of sucrose- phosphate synthase in higher plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 47, 431–444. Short review of factors that regulate this critical enzyme. Leloir, L.F. (1971) Two decades of research on the biosynthesis of saccharides. Science 172, 1299–1303. Leloir’s Nobel address, including a discussion of the role of sugar nucleotides in metabolism. Mukerjea, R., Yu, L., & Robyt, J.F. (2002) Starch biosynthesis: mechanism for the elongation of starch chains. Carbohydrate Res. 337, 1015–1022. Recent evidence that the starch chain grows at its reducing end. Synthesis of Cellulose and Peptidoglycan Delmer, D.P. (1999) Cellulose biosynthesis: exciting times for a difficult field of study. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50, 245–276. Dhugga, K.S. (2001) Building the wall: genes and enzyme complexes for polysaccharide synthases. Curr. Opin. Plant Biol. 4, 488–493. Peng, L., Kawagoe, Y., Hogan, P., & Delmer, D. (2002) Sitosterol-H9252-glucoside as primer for cellulose synthesis in plants. Science 295, 147–150. Recent evidence for the lipid-oligosaccharide intermediate in cellulose synthesis. Reid, J.S.G. (2000) Cementing the wall: cell wall polysaccharide synthesising enzymes. Curr. Opin. Plant Biol. 3, 512–516. Saxena, I.M. & Brown, R.M., Jr. (2000) Cellulose synthases and related enzymes. Curr. Opin. Plant Biol. 3, 523–531. Williamson, R.E., Burn, J.E., & Hocart, C.H. (2002) Towards the mechanism of cellulose synthesis. Trends Plant Sci. 7, 461–467. 1. Segregation of Metabolism in Organelles What are the advantages to the plant cell of having different organelles to carry out different reaction sequences that share inter- mediates? 2. Phases of Photosynthesis When a suspension of green algae is illuminated in the absence of CO 2 and then incubated with 14 CO 2 in the dark, 14 CO 2 is converted to [ 14 C]glucose for a brief time. What is the significance of this observation with regard to the CO 2 -assimilation process, and how is it related to the light reactions of photosynthesis? Why does the con- version of 14 CO 2 to [ 14 C]glucose stop after a brief time? 3. Identification of Key Intermediates in CO 2 Assim- ilation Calvin and his colleagues used the unicellular green alga Chlorella to study the carbon-assimilation reactions of photosynthesis. They incubated 14 CO 2 with illuminated sus- pensions of algae and followed the time course of appearance Problems 8885d_c20_751–786 2/18/04 1:56 PM Page 784 mac76 mac76:385_reb: Chapter 20 Problems 785 of 14 C in two compounds, X and Y, under two sets of condi- tions. Suggest the identities of X and Y, based on your un- derstanding of the Calvin cycle. (a) Illuminated Chlorella were grown with unlabeled CO 2 , then the light was turned off and 14 CO 2 was added (ver- tical dashed line in the graph below). Under these conditions, X was the first compound to become labeled with 14 C; Y was unlabeled. (b) Illuminated Chlorella cells were grown with 14 CO 2 . Illumination was continued until all the 14 CO 2 had disap- peared (vertical dashed line in the graph below). Under these conditions, X became labeled quickly but lost its radioactiv- ity with time, whereas Y became more radioactive with time. 4. Regulation of the Calvin Cycle Iodoacetate reacts irreversibly with the free OSH groups of Cys residues in proteins. Predict which Calvin cycle enzyme(s) would be inhibited by iodoacetate, and explain why. 5. Thioredoxin in Regulation of Calvin Cycle Enzymes Motohashi and colleagues * used thioredoxin as a hook to fish out from plant extracts the proteins that are activated by thioredoxin. To do this, they prepared a mutant thioredoxin in which one of the reactive Cys residues was replaced with a Ser. Explain why this modification was necessary for their experiments. Radioactivity 0 Time 14 CO 2 Y X Radioactivity CO 2 , light 0 Time 14 CO 2 , dark Y X 6. Comparison of the Reductive and Oxidative Pentose Phosphate Pathways The reductive pentose phosphate pathway generates a number of intermediates identical to those of the oxidative pentose phosphate pathway (Chapter 14). What role does each pathway play in cells where it is active? 7. Photorespiration and Mitochondrial Respiration Compare the oxidative photosynthetic carbon cycle (C 2 cy- cle), also called photorespiration, with the mitochondrial respiration that drives ATP synthesis. Why are both processes referred to as respiration? Where in the cell do they occur, and under what circumstances? What is the path of electron flow in each? 8. Rubisco and the Composition of the Atmosphere N. E. Tolbert ? has argued that the dual specificity of rubisco for CO 2 and O 2 is not simply a leftover from evolution in a low-oxygen environment. He suggests that the relative activ- ities of the carboxylase and oxygenase activities of rubisco actually have set, and now maintain, the ratio of CO 2 to O 2 in the earth’s atmosphere. Discuss the pros and cons of this hypothesis, in molecular terms and in global terms. How does the existence of C 4 organisms bear on the hypothesis? 9. Role of Sedoheptulose 1,7-Bisphosphatase What ef- fect on the cell and the organism might result from a defect in sedoheptulose 1,7-bisphosphatase in (a) a human hepato- cyte and (b) the leaf cell of a green plant? 10. Pathway of CO 2 Assimilation in Maize If a maize (corn) plant is illuminated in the presence of 14 CO 2 , after about 1 second, more than 90% of all the radioactivity in- corporated in the leaves is found at C-4 of malate, aspartate, and oxaloacetate. Only after 60 seconds does 14 C appear at C-1 of 3-phosphoglycerate. Explain. 11. Identifying CAM Plants Given some 14 CO 2 and all the tools typically present in a biochemistry research lab, how would you design a simple experiment to determine whether a plant was a typical C 4 plant or a CAM plant? 12. Chemistry of Malic Enzyme: Variation on a Theme Malic enzyme, found in the bundle-sheath cells of C 4 plants, carries out a reaction that has a counterpart in the citric acid cycle. What is the analogous reaction? Explain your choice. 13. The Cost of Storing Glucose as Starch Write the sequence of steps and the net reaction required to calculate the cost, in ATP molecules, of converting a molecule of cy- tosolic glucose 6-phosphate to starch and back to glucose 6- phosphate. What fraction of the maximum number of ATP molecules available from complete catabolism of glucose 6- phosphate to CO 2 and H 2 O does this cost represent? CH 2 O NAD C Inactive enzyme O ICH 2 O CH11001 O S Iodoacetate NAD H11001H11001 H11002 H11002 SH Cys Cys HI *Motohashi, K., Kondoh, A., Stumpp, M.T., & Hisabori, T. (2001) Com- prehensive survey of proteins targeted by chloroplast thioredoxin. Proc. Natl. Acad. Sci. USA 98, 11,224–11,229. ? Tolbert, N.E. (1994) The role of photosynthesis and photorespiration in regulating atmospheric CO 2 and O 2 . In Regulation of Atmospheric CO 2 and O 2 by Photosynthetic Carbon Metabolism (Tolbert, N.E. & Preiss, J., eds), pp. 8–33, Oxford University Press, New York. 8885d_c20_751–786 2/18/04 1:56 PM Page 785 mac76 mac76:385_reb: Chapter 20 Carbohydrate Biosynthesis in Plants and Bacteria786 14. Inorganic Pyrophosphatase The enzyme inorganic pyrophosphatase contributes to making many biosynthetic re- actions that generate inorganic pyrophosphate essentially ir- reversible in cells. By keeping the concentration of PP i very low, the enzyme “pulls” these reactions in the direction of PP i formation. The synthesis of ADP-glucose in chloroplasts is one reaction that is pulled in the forward direction by this mechanism. However, the synthesis of UDP-glucose in the plant cytosol, which produces PP i , is readily reversible in vivo. How do you reconcile these two facts? 15. Regulation of Starch and Sucrose Synthesis Su- crose synthesis occurs in the cytosol and starch synthesis in the chloroplast stroma, yet the two processes are intricately balanced. What factors shift the reactions in favor of (a) starch synthesis and (b) sucrose synthesis? 16. Regulation of Sucrose Synthesis In the regulation of sucrose synthesis from the triose phosphates produced during photosynthesis, 3-phosphoglycerate and P i play criti- cal roles (see Fig. 20–26). Explain why the concentrations of these two regulators reflect the rate of photosynthesis. 17. Sucrose and Dental Caries The most prevalent in- fection in humans worldwide is dental caries, which stems from the colonization and destruction of tooth enamel by a variety of acidifying microorganisms. These organisms synthesize and live within a water-insoluble network of dextrans, called dental plaque, composed of (H92511n6)-linked polymers of glucose with many (H92511n3) branch points. Poly- merization of dextran requires dietary sucrose, and the reaction is catalyzed by a bacterial enzyme, dextran-sucrose glucosyltransferase. (a) Write the overall reaction for dextran polymerization. (b) In addition to providing a substrate for the forma- tion of dental plaque, how does dietary sucrose also provide oral bacteria with an abundant source of metabolic energy? 18. Differences between C 3 and C 4 Plants The plant genus Atriplex includes some C 3 and some C 4 species. From the data in the plots below (species 1, black curve; species 2, red curve), identify which is a C 3 plant and which is a C 4 plant. Justify your answer in molecular terms that account for the data in all three plots. 19. C 4 Pathway in a Single Cell In typical C 4 plants, the initial capture of CO 2 occurs in one cell type, and the Calvin cycle reactions occur in another (see Fig. 20–23). Voznesen- skaya and colleagues ?? have described a plant, Bienertia cy- cloptera—which grows in salty depressions of semidesert in Central Asia—that shows the biochemical properties of a C 4 plant but unlike typical C 4 plants does not segregate the re- actions of CO 2 fixation into two cell types. PEP carboxylase and rubisco are present in the same cell. However, the cells have two types of chloroplasts, which are localized differently, as shown in the micrograph. One type, relatively poor in grana (thylakoids), is confined to the periphery; the more typical chloroplasts are clustered in the center of the cell, separated from the peripheral chloroplasts by large vacuoles. Thin cy- tosolic bridges pass through the vacuoles, connecting the pe- ripheral and central cytosol. 10 H9262m Uptake of CO 2 Uptake of CO 2 Uptake of CO 2 Light intensity Leaf temperature [CO 2 ] in intracellular space ?? Voznesenskaya, E.V., Fraceschi, V.R., Kiirats, O., Artyusheva, E.G., Freitag, H., & Edwards, G.E. (2002) Proof of C 4 photosynthesis with- out Kranz anatomy in Bienertia cycloptera (Chenopodiaceae). Plant J. 31, 649–662. In this plant, where would you expect to find (a) PEP carboxylase, (b) rubisco, and (c) starch granules? Explain your answers with a model for CO 2 fixation in these C 4 cells. 8885d_c20_786 2/20/04 12:03 PM Page 786 mac76 mac76:385_reb: chapter L ipids play a variety of cellular roles, some only re- cently recognized. They are the principal form of stored energy in most organisms and major constituents of cellular membranes. Specialized lipids serve as pig- ments (retinal, carotene), cofactors (vitamin K), deter- gents (bile salts), transporters (dolichols), hormones (vitamin D derivatives, sex hormones), extracellular and intracellular messengers (eicosanoids, phosphatidylino- sitol derivatives), and anchors for membrane proteins (covalently attached fatty acids, prenyl groups, and phosphatidylinositol). The ability to synthesize a vari- ety of lipids is essential to all organisms. This chapter describes the biosynthetic pathways for some of the most common cellular lipids, illustrating the strategies employed in assembling these water-insoluble products from water-soluble precursors such as acetate. Like other biosynthetic pathways, these reaction sequences are endergonic and reductive. They use ATP as a source of metabolic energy and a reduced electron carrier (usu- ally NADPH) as reductant. We first describe the biosynthesis of fatty acids, the primary components of both triacylglycerols and phos- pholipids, then examine the assembly of fatty acids into triacylglycerols and the simpler membrane phospho- lipids. Finally, we consider the synthesis of cholesterol, a component of some membranes and the precursor of steroids such as the bile acids, sex hormones, and adrenocortical hormones. 21.1 Biosynthesis of Fatty Acids and Eicosanoids After the discovery that fatty acid oxidation takes place by the oxidative removal of successive two-carbon (acetyl-CoA) units (see Fig. 17–8), biochemists thought the biosynthesis of fatty acids might proceed by a sim- ple reversal of the same enzymatic steps. However, as they were to find out, fatty acid biosynthesis and break- down occur by different pathways, are catalyzed by dif- ferent sets of enzymes, and take place in different parts of the cell. Moreover, biosynthesis requires the partici- pation of a three-carbon intermediate, malonyl-CoA, that is not involved in fatty acid breakdown. We focus first on the pathway of fatty acid synthe- sis, then turn our attention to regulation of the pathway and to the biosynthesis of longer-chain fatty acids, un- saturated fatty acids, and their eicosanoid derivatives. Malonyl-CoA Is Formed from Acetyl-CoA and Bicarbonate The formation of malonyl-CoA from acetyl-CoA is an irreversible process, catalyzed by acetyl-CoA carbox- ylase. The bacterial enzyme has three separate poly- peptide subunits (Fig. 21–1); in animal cells, all three CH 2 C O C S-CoAO Malonyl-CoA O H5008 LIPID BIOSYNTHESIS 21.1 Biosynthesis of Fatty Acids and Eicosanoids 787 21.2 Biosynthesis of Triacylglycerols 804 21.3 Biosynthesis of Membrane Phospholipids 808 21.4 Biosynthesis of Cholesterol, Steroids, and Isoprenoids 816 How the division of “spoils” came about I do not recall—it may have been by drawing lots. At any rate, David Shemin “drew” amino acid metabolism, which led to his classical work on heme biosynthesis. David Rittenburg was to continue his interest in protein synthesis and turnover, and lipids were to be my territory. —Konrad Bloch, on how his career turned to problems of lipid metabolism after the death of his mentor, Rudolf Schoen- heimer; article in Annual Review of Biochemistry, 1987 21 787 8885d_c21_787-832 2/26/04 9:35 AM Page 787 mac76 mac76:385_reb: activities are part of a single multifunctional polypep- tide. Plant cells contain both types of acetyl-CoA carboxylase. In all cases, the enzyme contains a biotin prosthetic group covalently bound in amide linkage to the H9255-amino group of a Lys residue in one of the three polypeptides or domains of the enzyme molecule. The two-step reaction catalyzed by this enzyme is very similar to other biotin-dependent carboxylation reac- tions, such as those catalyzed by pyruvate carboxylase (see Fig. 16–16) and propionyl-CoA carboxylase (see Fig. 17–11). The carboxyl group, derived from bicar- bonate (HCO 3 H11002 ), is first transferred to biotin in an ATP- dependent reaction. The biotinyl group serves as a tem- porary carrier of CO 2 , transferring it to acetyl-CoA in the second step to yield malonyl-CoA. Fatty Acid Synthesis Proceeds in a Repeating Reaction Sequence The long carbon chains of fatty acids are assembled in a repeating four-step sequence (Fig. 21–2). A saturated acyl group produced by this set of reactions becomes the substrate for subsequent condensation with an ac- tivated malonyl group. With each passage through the cycle, the fatty acyl chain is extended by two carbons. When the chain length reaches 16 carbons, the product Chapter 21 Lipid Biosynthesis788 OC HN NH S Biotin carboxylase N C O H5008 H11001HCO3 H5008 ATP ADP H11001 P i transcarboxylase O HN NH S Biotin arm Lys side arm NH C O C O H5008 O biotin carboxylase Malonyl-CoA CH 3 S-CoA C O OC HN NH S NH C O H11001 C O Acetyl-CoA CH 3 S-CoA C OO C NH CO O H5008 C O C H5008 O Malonyl-CoA S-CoA C O Acetyl-CoA CH 3 S-CoA C OO NH C O C H5008 O CH 2 O Lys Biotin carrier protein Biotin carrier protein Transcarboxylase Biotin FIGURE 21–1 The acetyl-CoA carboxylase reaction. Acetyl-CoA carboxylase has three functional regions: biotin carrier protein (gray); biotin carboxylase, which activates CO 2 by attaching it to a nitrogen in the biotin ring in an ATP-dependent reaction (see Fig. 16–16); and transcarboxylase, which transfers activated CO 2 (shaded green) from biotin to acetyl-CoA, producing malonyl-CoA. The long, flexible bi- otin arm carries the activated CO 2 from the biotin carboxylase region to the transcarboxylase active site, as shown in the diagrams below the reaction arrows. The active enzyme in each step is shaded blue. 8885d_c21_787-832 2/26/04 9:35 AM Page 788 mac76 mac76:385_reb: (palmitate, 16:0; see Table 10–1) leaves the cycle. Car- bons C-16 and C-15 of the palmitate are derived from the methyl and carboxyl carbon atoms, respectively, of an acetyl-CoA used directly to prime the system at the outset (Fig. 21–3); the rest of the carbon atoms in the chain are derived from acetyl-CoA via malonyl-CoA. Both the electron-carrying cofactor and the acti- vating groups in the reductive anabolic sequence differ from those in the oxidative catabolic process. Recall that in H9252 oxidation, NAD H11001 and FAD serve as electron ac- ceptors and the activating group is the thiol (OSH) group of coenzyme A (see Fig. 17–8). By contrast, the reducing agent in the synthetic sequence is NADPH and the activating groups are two different enzyme-bound OSH groups, as described below. All the reactions in the synthetic process are cat- alyzed by a multienzyme complex, fatty acid synthase. Although the details of enzyme structure differ in prokaryotes such as Escherichia coli and in eukary- otes, the four-step process of fatty acid synthesis is the same in all organisms. We first describe the process as it occurs in E. coli, then consider differences in enzyme structure in other organisms. The Fatty Acid Synthase Complex Has Seven Different Active Sites The core of the E. coli fatty acid synthase system con- sists of seven separate polypeptides (Table 21–1), and at least three others act at some stage of the process. The proteins act together to catalyze the formation of fatty acids from acetyl-CoA and malonyl-CoA. Through- out the process, the intermediates remain covalently at- tached as thioesters to one of two thiol groups of the synthase complex. One point of attachment is the OSH group of a Cys residue in one of the seven synthase pro- teins (H9252-ketoacyl-ACP synthase); the other is the OSH group of acyl carrier protein. 21.1 Biosynthesis of Fatty Acids and Eicosanoids 789 S H5008 O C CH 2 C O S O H Acetyl group (first acyl group) Fatty acid synthase dehydration CH 2 CH 3 C C O O H 2 O 3 Saturated acyl group, lengthened by two carbons SCH 3 C OO CH 2 C CO 2 1condensation SCH 3 C O OH CH 2 C H NADP H11001 4reduction NADPH H11001 H H11001 SCH 3 C O CC H CH 3 CH 2 Malonyl group S HS HS HS NADP H11001 2reduction NADPH H11001 H H11001 HS b a FIGURE 21–2 Addition of two carbons to a growing fatty acyl chain: a four-step sequence. Each malonyl group and acetyl (or longer acyl) group is activated by a thioester that links it to fatty acid synthase, a multienzyme complex described later in the text. 1 Condensation of an activated acyl group (an acetyl group from acetyl-CoA is the first acyl group) and two carbons derived from malonyl-CoA, with elimi- nation of CO 2 from the malonyl group, extends the acyl chain by two carbons. The mechanism of the first step of this reaction is given to il- lustrate the role of decarboxylation in facilitating condensation. The H9252-keto product of this condensation is then reduced in three more steps nearly identical to the reactions of H9252 oxidation, but in the re- verse sequence: 2 the H9252-keto group is reduced to an alcohol, 3 elimination of H 2 O creates a double bond, and 4 the double bond is reduced to form the corresponding saturated fatty acyl group. 8885d_c21_787-832 2/26/04 9:35 AM Page 789 mac76 mac76:385_reb: Acyl carrier protein (ACP) of E. coli is a small protein (M r 8,860) containing the prosthetic group 4H11541-phosphopantetheine (Fig. 21–4; compare this with the panthothenic acid and H9252-mercaptoethylamine moi- ety of coenzyme A in Fig. 8–41). Hydrolysis of thioesters is highly exergonic, and the energy released helps to make two different steps ( 1 and 5 in Fig. 21–5) in fatty acid synthesis (condensation) thermodynamically favorable. The 4H11032-phosphopante-theine prosthetic group of ACP is believed to serve as a flexible arm, tethering the growing fatty acyl chain to the surface of the fatty acid synthase complex while carrying the reaction intermediates from one enzyme active site to the next. Fatty Acid Synthase Receives the Acetyl and Malonyl Groups Before the condensation reactions that build up the fatty acid chain can begin, the two thiol groups on the en- zyme complex must be charged with the correct acyl groups (Fig. 21–5, top). First, the acetyl group of acetyl- CoA is transferred to the Cys OSH group of the H9252- ketoacyl-ACP synthase. This reaction is catalyzed by acetyl-CoA–ACP transacetylase (AT in Fig. 21–5). The second reaction, transfer of the malonyl group from malonyl-CoA to the OSH group of ACP, is catalyzed by malonyl-CoA–ACP transferase (MT), also part of the complex. In the charged synthase complex, the acetyl Chapter 21 Lipid Biosynthesis790 C H5008 O CH 3 CH 2 C O O CO 2 H11001 CH 3 4H H11001 4e H5008 CH 2 CH 2 O S CO S C COO H5008 O CH 2 S CO 2 H11001 CH 3 4H H11001 4e H5008 CH 2 CH 2 CH 2 CH 2 C S C O O CH 2 S CO 2 H11001 CH 3 4H H11001 4e H5008 CH 2 CH 2 CH 2 CH 2 CH 2 CH 2 C S C O O CH 2 S four more additions CH 3 CH 2 CH 2 CH 2 CH 2 CH 2 CH 2 CH 2 CH 2 CH 2 CH 2 CH 2 CH 2 CH 2 CH 2 C HS H11001 Palmitate Fatty acid synthase S HS COO H5008 COO H5008 COO H5008 FIGURE 21–3 The overall process of palmitate synthesis. The fatty acyl chain grows by two-carbon units donated by activated malonate, with loss of CO 2 at each step. The initial acetyl group is shaded yellow, C-1 and C-2 of malonate are shaded pink, and the carbon released as CO 2 is shaded green. After each two-carbon addition, reductions convert the growing chain to a saturated fatty acid of four, then six, then eight carbons, and so on. The final product is palmitate (16:0). TABLE 21–1 Proteins of the Fatty Acid Synthase Complex of E. coli Component Function Acyl carrier protein (ACP) Carries acyl groups in thioester linkage Acetyl-CoA–ACP transacetylase (AT) Transfers acyl group from CoA to Cys residue of KS H9252-Ketoacyl-ACP synthase (KS) Condenses acyl and malonyl groups (KS has at least three isozymes) Malonyl-CoA–ACP transferase (MT) Transfers malonyl group from CoA to ACP H9252-Ketoacyl-ACP reductase (KR) Reduces H9252-keto group to H9252-hydroxyl group H9252-Hydroxyacyl-ACP dehydratase (HD) Removes H 2 O from H9252-hydroxyacyl-ACP, creating double bond Enoyl-ACP reductase (ER) Reduces double bond, forming saturated acyl-ACP 8885d_c21_790 2/27/04 12:40 PM Page 790 mac116 mac116: and malonyl groups are very close to each other and are activated for the chain-lengthening process. The first four steps of this process are now considered in some detail; all step numbers refer to Figure 21–5. Step 1 Condensation The first reaction in the formation of a fatty acid chain is condensation of the activated acetyl and malonyl groups to form acetoacetyl-ACP, an acetoacetyl group bound to ACP through the phos- phopantetheine OSH group; simultaneously, a molecule of CO 2 is produced. In this reaction, catalyzed by H9252- ketoacyl-ACP synthase (KS), the acetyl group is transferred from the Cys OSH group of the enzyme to the malonyl group on the OSH of ACP, becoming the methyl-terminal two-carbon unit of the new acetoacetyl group. The carbon atom of the CO 2 formed in this reaction is the same carbon originally introduced into malonyl- CoA from HCO 3 H11002 by the acetyl-CoA carboxylase reaction (Fig. 21–1). Thus CO 2 is only transiently in covalent linkage during fatty acid biosynthesis; it is removed as each two-carbon unit is added. Why do cells go to the trouble of adding CO 2 to make a malonyl group from an acetyl group, only to lose the CO 2 during the formation of acetoacetate? Recall that in the H9252 oxidation of fatty acids (see Fig. 17–8), cleav- age of the bond between two acyl groups (cleavage of an acetyl unit from the acyl chain) is highly exergonic, so the simple condensation of two acyl groups (two acetyl-CoA molecules, for example) is highly ender- gonic. The use of activated malonyl groups rather than acetyl groups is what makes the condensation reactions thermodynamically favorable. The methylene carbon (C-2) of the malonyl group, sandwiched between car- bonyl and carboxyl carbons, is chemically situated to act as a good nucleophile. In the condensation step (step 1 ), decarboxylation of the malonyl group facilitates the nucleophilic attack of the methylene carbon on the thioester linking the acetyl group to H9252-ketoacyl-ACP synthase, displacing the enzyme’s OSH group. Coupling the condensation to the decarboxylation of the malonyl group renders the overall process highly exergonic. A similar carboxylation-decarboxylation sequence facili- tates the formation of phosphoenolpyruvate from pyru- vate in gluconeogenesis (see Fig. 14–17). By using activated malonyl groups in the synthesis of fatty acids and activated acetate in their degradation, the cell makes both processes energetically favorable, although one is effectively the reversal of the other. The extra energy required to make fatty acid synthesis favorable is provided by the ATP used to synthesize malonyl-CoA from acetyl-CoA and HCO 3 H11002 (Fig. 21–1). Step 2 Reduction of the Carbonyl Group The acetoacetyl- ACP formed in the condensation step now undergoes reduction of the carbonyl group at C-3 to form D-H9252- hydroxybutyryl-ACP. This reaction is catalyzed by H9252- ketoacyl-ACP reductase (KR) and the electron donor is NADPH. Notice that the D-H9252-hydroxybutyryl group does not have the same stereoisomeric form as the L-H9252- hydroxyacyl intermediate in fatty acid oxidation (see Fig. 17–8). Step 3 Dehydration The elements of water are now re- moved from C-2 and C-3 of D-H9252-hydroxybutyryl-ACP to yield a double bond in the product, trans-H9004 2 - butenoyl- ACP. The enzyme that catalyzes this dehydration is H9252- hydroxyacyl-ACP dehydratase (HD). Step 4 Reduction of the Double Bond Finally, the double bond of trans-H9004 2 -butenoyl-ACP is reduced (saturated) to form butyryl-ACP by the action of enoyl-ACP re- ductase (ER); again, NADPH is the electron donor. The Fatty Acid Synthase Reactions Are Repeated to Form Palmitate Production of the four-carbon, saturated fatty acyl–ACP completes one pass through the fatty acid synthase 21.1 Biosynthesis of Fatty Acids and Eicosanoids 791 FIGURE 21–4 Acyl carrier protein (ACP). The prosthetic group is 4H11032- phosphopantetheine, which is covalently attached to the hydroxyl group of a Ser residue in ACP. Phosphopantetheine contains the B vi- tamin pantothenic acid, also found in the coenzyme A molecule. Its OSH group is the site of entry of malonyl groups during fatty acid synthesis. H5008 O CH 2 C O CH 3 SH OP HN O CH 2 OC CH 3 CHOH CH 2 C O HN CH 2 CH 2 ACP Malonyl groups are esterified to the 4H11032-Phospho- pantetheine Pantothenic acid Ser side chain SH group. CH 2 8885d_c21_791 2/27/04 12:40 PM Page 791 mac116 mac116: Chapter 21 Lipid Biosynthesis792 KS AT MT KR HDER ACP HS S CoA-SH Malonyl-CoA KS AT MT KR HDER ACP S S KS AT MT KR HDER ACP S HS CO 2 condensation KS AT MT KR HDER ACP HS HS CoA-SH Acetyl-CoA 1 Fatty acid synthase complex charged with an acetyl and a malonyl group b-Ketobutyryl-ACP KS AT MT KR HDER ACP S b-Ketoacyl-ACP synthase Malonyl-CoA– ACP transferase Enoyl-ACP reductase b-Hydroxyacyl-ACP dehydratase Acetyl-CoA–ACP transacetylase b-Ketoacyl-ACP reductase KS AT MT KR HDER HS KS AT MT KR HDER ACP S HS reduction of double bond trans-D 2 -Butenoyl-ACP NADP H11001 NADPH H11001 H H11001 Butyryl-ACP KS AT MT KR HDER ACP SH S translocation of butyryl group to Cys on b-ketoacyl-ACP synthase (KS) KS AT MT KR HDER ACP S HS b-Hydroxybutyryl-ACP dehydration H 2 O reduction of b-keto group NADP H11001 NADPH H11001 H H11001 CH 3 C O S-CoA CH 3 C O CH 2 C O S-CoA C O H11002 O CH 2 C O CH 3 CH 2 CH 2 C O C O H11002 O CH 3 O C CH 2 C O CH 3 O C CH 2 C O CHCH 3 OH CH C O CHCH 3 C O CH 3 CH 2 CH 2 2 3 4 5 FIGURE 21–5 Sequence of events during synthesis of a fatty acid. The fatty acid synthase complex is shown schematically. Each segment of the disk represents one of the six enzymatic activities of the complex. At the center is acyl carrier protein (ACP), with its phosphopantetheine arm ending in an OSH. The enzyme shown in blue is the one that will act in the next step. As in Figure 21–3, the initial acetyl group is shaded yellow, C-1 and C-2 of malonate are shaded pink, and the carbon released as CO 2 is shaded green. Steps 1 to 4 are described in the text. 8885d_c21_787-832 2/26/04 9:35 AM Page 792 mac76 mac76:385_reb: complex. The butyryl group is now transferred from the phosphopantetheine OSH group of ACP to the Cys OSH group of H9252-ketoacyl-ACP synthase, which initially bore the acetyl group (Fig. 21–5). To start the next cycle of four reactions that lengthens the chain by two more car- bons, another malonyl group is linked to the now unoc- cupied phosphopantetheine OSH group of ACP (Fig. 21–6). Condensation occurs as the butyryl group, act- ing like the acetyl group in the first cycle, is linked to two carbons of the malonyl-ACP group with concurrent loss of CO 2 . The product of this condensation is a six- carbon acyl group, covalently bound to the phospho- pantetheine OSH group. Its H9252-keto group is reduced in the next three steps of the synthase cycle to yield the saturated acyl group, exactly as in the first round of re- actions—in this case forming the six-carbon product. Seven cycles of condensation and reduction pro- duce the 16-carbon saturated palmitoyl group, still bound to ACP. For reasons not well understood, chain elongation by the synthase complex generally stops at this point and free palmitate is released from the ACP by a hydrolytic activity in the complex. Small amounts of longer fatty acids such as stearate (18:0) are also formed. In certain plants (coconut and palm, for exam- ple) chain termination occurs earlier; up to 90% of the fatty acids in the oils of these plants are between 8 and 14 carbons long. We can consider the overall reaction for the syn- thesis of palmitate from acetyl-CoA in two parts. First, the formation of seven malonyl-CoA molecules: 7 Acetyl-CoA H11001 7CO 2 H11001 7ATP n 7 malonyl-CoA H11001 7ADP H11001 7P i (21–1) then seven cycles of condensation and reduction: Acetyl-CoA H11001 7 malonyl-CoA H11001 14NADPH H11001 14H H11001 n palmitate H11001 7CO 2 H11001 8 CoA H11001 14NADP H11001 H11001 6H 2 O (21–2) The overall process (the sum of Eqns 21–1 and 21–2) is 8 Acetyl-CoA H11001 7ATP H11001 14NADPH H11001 14H H11001 n palmitate H11001 8 CoA H11001 7ADP H11001 7P i H11001 14NADP H11001 H11001 6H 2 O (21–3) The biosynthesis of fatty acids such as palmitate thus requires acetyl-CoA and the input of chemical energy in two forms: the group transfer potential of ATP and the reducing power of NADPH. The ATP is required to at- tach CO 2 to acetyl-CoA to make malonyl-CoA; the NADPH is required to reduce the double bonds. We re- turn to the sources of acetyl-CoA and NADPH soon, but first let’s consider the structure of the remarkable enzyme complex that catalyzes the synthesis of fatty acids. 21.1 Biosynthesis of Fatty Acids and Eicosanoids 793 CH 2 C O S-CoA C O H5008 O CH 2 C O CH 3 CH 2 KS AT MT KR HDER ACP SH S Butyryl group CoA-SH Malonyl-CoA KS AT MT KR HDER ACP S S KS AT MT KR HDER ACP S SH CO 2 condensation -Ketoacyl-ACP CH 2 C O C O H5008 O CH 2 C O CH 2 CH 3 CH 2 C O CH 2 C O CH 2 CH 3 b FIGURE 21–6 Beginning of the second round of the fatty acid syn- thesis cycle. The butyryl group is on the Cys OSH group. The incoming malonyl group is first attached to the phosphopantetheine OSH group. Then, in the condensation step, the entire butyryl group on the Cys OSH is exchanged for the carboxyl group of the malonyl residue, which is lost as CO 2 (green). This step is analogous to step 1 in Fig- ure 21–5. The product, a six-carbon H9252-ketoacyl group, now contains four carbons derived from malonyl-CoA and two derived from the acetyl-CoA that started the reaction. The H9252-ketoacyl group then un- dergoes steps 2 through 4 , as in Figure 21–5. 8885d_c21_787-832 2/26/04 9:35 AM Page 793 mac76 mac76:385_reb: The Fatty Acid Synthase of Some Organisms Consists of Multifunctional Proteins In E. coli and some plants, the seven active sites for fatty acid synthesis (six enzymes and ACP) reside in seven separate polypeptides (Fig. 21–7, top). In these complexes, each enzyme is positioned with its active site near that of the preceding and succeeding enzymes of the sequence. The flexible pantetheine arm of ACP can reach all the active sites, and it carries the growing fatty acyl chain from one site to the next; intermediates are not released from the enzyme complex until it has formed the finished product. As we have seen in earlier chapters, this channeling of intermediates from one ac- tive site to the next increases the efficiency of the over- all process. The fatty acid synthases of yeast and of vertebrates are also multienzyme complexes, and their integration is even more complete than in E. coli and plants. In yeast, the seven distinct active sites reside in two large, multifunctional polypeptides, with three activities on the H9251 subunit and four on the H9252 subunit. In vertebrates, a single large polypeptide (M r 240,000) contains all seven enzymatic activities as well as a hydrolytic activ- ity that cleaves the finished fatty acid from the ACP-like part of the enzyme complex. The vertebrate enzyme functions as a dimer (M r 480,000) in which the two iden- tical subunits lie head-to-tail. The subunits appear to function independently. When all the active sites in one subunit are inactivated by mutation, palmitate synthe- sis is only modestly reduced. Fatty Acid Synthesis Occurs in the Cytosol of Many Organisms but in the Chloroplasts of Plants In most higher eukaryotes, the fatty acid synthase com- plex is found exclusively in the cytosol (Fig. 21–8), as are the biosynthetic enzymes for nucleotides, amino acids, and glucose. This location segregates synthetic processes from degradative reactions, many of which take place in the mitochondrial matrix. There is a cor- responding segregation of the electron-carrying cofac- tors used in anabolism (generally a reductive process) and those used in catabolism (generally oxidative). Usually, NADPH is the electron carrier for anabolic reactions, and NAD H11001 serves in catabolic reactions. In hepatocytes, the [NADPH]/[NADP H11001 ] ratio is very high (about 75) in the cytosol, furnishing a strongly reducing environment for the reductive synthesis of fatty acids and other biomolecules. The cytosolic [NADH]/[NAD H11001 ] ratio is much smaller (only about 8 H11003 10 H110024 ), so the NAD H11001 -dependent oxidative catabolism of glucose can take place in the same compartment, and at the same time, as fatty acid synthesis. The [NADH]/[NAD H11001 ] ratio in the mitochondrion is much higher than in the cytosol, because of the flow of electrons to NAD H11001 from the ox- idation of fatty acids, amino acids, pyruvate, and acetyl- CoA. This high mitochondrial [NADH]/[NAD H11001 ] ratio favors the reduction of oxygen via the respiratory chain. In hepatocytes and adipocytes, cytosolic NADPH is largely generated by the pentose phosphate pathway (see Fig. 14–21) and by malic enzyme (Fig. 21–9a). The NADP-linked malic enzyme that operates in the carbon-assimilation pathway of C 4 plants (see Fig. 20–23) is unrelated in function. The pyruvate produced in the reaction shown in Figure 21–9a reenters the mi- tochondrion. In hepatocytes and in the mammary gland of lactating animals, the NADPH required for fatty acid biosynthesis is supplied primarily by the pentose phos- phate pathway (Fig. 21–9b). In the photosynthetic cells of plants, fatty acid syn- thesis occurs not in the cytosol but in the chloroplast stroma (Fig. 21–8). This makes sense, given that NADPH is produced in chloroplasts by the light reac- tions of photosynthesis: Again, the resulting high [NADPH]/[NADP H11001 ] ratio provides the reducing environment that favors reduc- tive anabolic processes such as fatty acid synthesis. Acetate Is Shuttled out of Mitochondria as Citrate In nonphotosynthetic eukaryotes, nearly all the acetyl- CoA used in fatty acid synthesis is formed in mito- H 2 OO 22 1 H H11001 H11001NADPHH11001NADP H11001 H11001 light Chapter 21 Lipid Biosynthesis794 Bacteria, Plants Seven activities in seven separate polypeptides Yeast Seven activities in two separate polypeptides Vertebrates Seven activities in one large polypeptide KS AT MT KR HDER KS AT MT KR HDER ACP KS AT KR HDER ACP ACP MT FIGURE 21–7 Structure of fatty acid synthases. The fatty acid syn- thase of bacteria and plants is a complex of at least seven different polypeptides. In yeast, all seven activities reside in only two polypep- tides; the vertebrate enzyme is a single large polypeptide. 8885d_c21_787-832 2/26/04 9:35 AM Page 794 mac76 mac76:385_reb: chondria from pyruvate oxidation and from the catabo- lism of the carbon skeletons of amino acids. Acetyl-CoA arising from the oxidation of fatty acids is not a signifi- cant source of acetyl-CoA for fatty acid biosynthesis in animals, because the two pathways are reciprocally reg- ulated, as described below. The mitochondrial inner membrane is impermeable to acetyl-CoA, so an indirect shuttle transfers acetyl group equivalents across the inner membrane (Fig. 21–10). Intramitochondrial acetyl-CoA first reacts with oxaloacetate to form citrate, in the citric acid cycle re- action catalyzed by citrate synthase (see Fig. 16–7). Citrate then passes through the inner membrane on the citrate transporter. In the cytosol, citrate cleavage by citrate lyase regenerates acetyl-CoA in an ATP- dependent reaction. Oxaloacetate cannot return to the mitochondrial matrix directly, as there is no oxaloacetate transporter. Instead, cytosolic malate dehydrogenase re- duces the oxaloacetate to malate, which returns to the mitochondrial matrix on the malate–H9251-ketoglutarate transporter in exchange for citrate. In the matrix, malate is reoxidized to oxaloacetate to complete the shuttle. Al- ternatively, the malate produced in the cytosol is used to generate cytosolic NADPH through the activity of malic enzyme (Fig. 21–9a). Fatty Acid Biosynthesis Is Tightly Regulated When a cell or organism has more than enough meta- bolic fuel to meet its energy needs, the excess is gen- erally converted to fatty acids and stored as lipids such as triacylglycerols. The reaction catalyzed by acetyl-CoA 21.1 Biosynthesis of Fatty Acids and Eicosanoids 795 Animal cells, yeast cells Mitochondria Endoplasmic reticulum Cytosol Chloroplasts Peroxisomes Plant cells No fatty acid oxidation Fatty acid oxidation Acetyl-CoA production Ketone body synthesis Fatty acid elongation NADPH production (pentose phosphate pathway; malic enzyme) [NADPH]/[NADP H11001 ] high Isoprenoid and sterol synthesis (early stages) Fatty acid synthesis NADPH, ATP production [NADPH]/[NADP H11001 ] high Fatty acid synthesis Fatty acid oxidation ( Phospholipid synthesis Sterol synthesis (late stages) Fatty acid elongation Fatty acid desaturation Η 2 Ο 2 ) Catalase, peroxidase: Η 2 Ο 2 Η 2 Ο FIGURE 21–8 Subcellular localization of lipid metabolism. Yeast and vertebrate cells differ from higher plant cells in the compartmentation of lipid metabolism. Fatty acid synthesis takes place in the compart- ment in which NADPH is available for reductive synthesis (i.e., where the [NADPH]/[NADP H11001 ] ratio is high). Processes in red type are cov- ered in this chapter. malic enzyme pentose phosphate pathway CHOH COO H11002 CH 2 COO H11002 C CO 2 O C H11001 H 3 COO H11002 Malate Pyruvate NADP H11001 Glucose 6-phosphate Ribulose 5-phosphate H11001 H H11001 NADPH NADP H11001 NADPH (a) (b) NADP H11001 NADPH FIGURE 21–9 Production of NADPH. Two routes to NADPH, cat- alyzed by (a) malic enzyme and (b) the pentose phosphate pathway. 8885d_c21_787-832 2/26/04 9:35 AM Page 795 mac76 mac76:385_reb: carboxylase is the rate-limiting step in the biosynthesis of fatty acids, and this enzyme is an important site of regulation. In vertebrates, palmitoyl-CoA, the principal product of fatty acid synthesis, is a feedback inhibitor of the enzyme; citrate is an allosteric activator (Fig. 21–11a), increasing V max . Citrate plays a central role in diverting cellular metabolism from the consumption (oxidation) of metabolic fuel to the storage of fuel as fatty acids. When the concentrations of mitochondrial acetyl- CoA and ATP increase, citrate is transported out of mi- tochondria; it then becomes both the precursor of cyto- solic acetyl-CoA and an allosteric signal for the activation of acetyl-CoA carboxylase. At the same time, citrate in- hibits the activity of phosphofructokinase-1 (see Fig. 15–18), reducing the flow of carbon through glycolysis. Acetyl-CoA carboxylase is also regulated by cova- lent modification. Phosphorylation, triggered by the hormones glucagon and epinephrine, inactivates the enzyme and reduces its sensitivity to activation by cit- rate, thereby slowing fatty acid synthesis. In its active (dephosphorylated) form, acetyl-CoA carboxylase poly- merizes into long filaments (Fig. 21–11b); phosphoryla- tion is accompanied by dissociation into monomeric subunits and loss of activity. Chapter 21 Lipid Biosynthesis796 Matrix Cytosol NADH + H + Inner membrane Oxaloacetate Citrate transporter citrate synthase citrate lyase malate dehydrogenase Malate malic enzyme malate dehydrogenase α Malate– -ketoglutarate transporter Malate pyruvate carboxylase Oxaloacetate NAD + NADP + NAD + ADP + P i ADP + P i ATP CO 2 CO 2 CoA-SH Outer membrane Citrate Citrate CoAOSH pyruvate dehydrogenase Amino acids Acetyl-CoA Pyruvate Glucose Acetyl-CoA Fatty acid synthesis PyruvatePyruvate Pyruvate transporter NADH + H + NADPH + H + ATP FIGURE 21–10 Shuttle for transfer of acetyl groups from mitochon- dria to the cytosol. The mitochondrial outer membrane is freely per- meable to all these compounds. Pyruvate derived from amino acid catabolism in the mitochondrial matrix, or from glucose by glycolysis in the cytosol, is converted to acetyl-CoA in the matrix. Acetyl groups pass out of the mitochondrion as citrate; in the cytosol they are de- livered as acetyl-CoA for fatty acid synthesis. Oxaloacetate is reduced to malate, which returns to the mitochondrial matrix and is converted to oxaloacetate. An alternative fate for cytosolic malate is oxidation by malic enzyme to generate cytosolic NADPH; the pyruvate pro- duced returns to the mitochondrial matrix. 8885d_c21_787-832 2/26/04 9:35 AM Page 796 mac76 mac76:385_reb: The acetyl-CoA carboxylase of plants and bacteria is not regulated by citrate or by a phosphorylation- dephosphorylation cycle. The plant enzyme is activated by an increase in stromal pH and [Mg 2H11001 ], which occurs on illumination of the plant (see Fig. 20–18). Bacteria do not use triacylglycerols as energy stores. In E. coli, the primary role of fatty acid synthesis is to provide pre- cursors for membrane lipids; the regulation of this process is complex, involving guanine nucleotides (such as ppGpp) that coordinate cell growth with membrane formation (see Figs 8–42, 28–24). In addition to the moment-by-moment regulation of enzymatic activity, these pathways are regulated at the level of gene expression. For example, when animals in- gest an excess of certain polyunsaturated fatty acids, the expression of genes encoding a wide range of li- pogenic enzymes in the liver is suppressed. The detailed mechanism by which these genes are regulated is not yet clear. If fatty acid synthesis and H9252 oxidation were to pro- ceed simultaneously, the two processes would consti- tute a futile cycle, wasting energy. We noted earlier (see Fig. 17–12) that H9252 oxidation is blocked by malonyl-CoA, which inhibits carnitine acyltransferase I. Thus during fatty acid synthesis, the production of the first inter- mediate, malonyl-CoA, shuts down H9252 oxidation at the level of a transport system in the mitochondrial inner membrane. This control mechanism illustrates another advantage of segregating synthetic and degradative pathways in different cellular compartments. Long-Chain Saturated Fatty Acids Are Synthesized from Palmitate Palmitate, the principal product of the fatty acid syn- thase system in animal cells, is the precursor of other long-chain fatty acids (Fig. 21–12). It may be length- ened to form stearate (18:0) or even longer saturated fatty acids by further additions of acetyl groups, through the action of fatty acid elongation systems present in the smooth endoplasmic reticulum and in mitochon- dria. The more active elongation system of the ER ex- tends the 16-carbon chain of palmitoyl-CoA by two car- bons, forming stearoyl-CoA. Although different enzyme systems are involved, and coenzyme A rather than ACP is the acyl carrier in the reaction, the mechanism of elon- gation in the ER is otherwise identical to that in palmi- tate synthesis: donation of two carbons by malonyl-CoA, followed by reduction, dehydration, and reduction to the saturated 18-carbon product, stearoyl-CoA. 21.1 Biosynthesis of Fatty Acids and Eicosanoids 797 Malonyl-CoA acetyl-CoA carboxylase Acetyl-CoA Citrate Palmitoyl-CoA glucagon, epinephrine trigger phosphorylation/ inactivation insulin triggers activation citrate lyase (a) (b) 400 ? FIGURE 21–11 Regulation of fatty acid synthesis. (a) In the cells of vertebrates, both allosteric regulation and hormone-dependent cova- lent modification influence the flow of precursors into malonyl-CoA. In plants, acetyl-CoA carboxylase is activated by the changes in [Mg 2H11001 ] and pH that accompany illumination (not shown here). (b) Filaments of acetyl-CoA carboxylase (the active, dephosphorylated form) as seen with the electron microscope. Palmitate 16:0 elongation Stearate 18:0 desaturation Oleate 18:1(H9004 9 ) desaturation (in plants only) Linoleate 18:2(H9004 9,12 ) desaturation (in plants only) desaturation H9251-Linolenate 18:3(H9004 9,12,15 ) Other polyunsaturated fatty acids H9253-Linolenate 18:3(H9004 6,9,12 ) Eicosatrienoate 20:3(H9004 8,11,14 ) elongation Arachidonate 20:4(H9004 5,8,11,14 ) desaturation desaturation Palmitoleate 16:1(H9004 9 ) elongation Longer saturated fatty acids FIGURE 21–12 Routes of synthesis of other fatty acids. Palmitate is the precursor of stearate and longer-chain saturated fatty acids, as well as the monounsaturated acids palmitoleate and oleate. Mammals can- not convert oleate to linoleate or H9251-linolenate (shaded pink), which are therefore required in the diet as essential fatty acids. Conversion of linoleate to other polyunsaturated fatty acids and eicosanoids is out- lined. Unsaturated fatty acids are symbolized by indicating the num- ber of carbons and the number and position of the double bonds, as in Table 10–1. 8885d_c21_787-832 2/26/04 9:35 AM Page 797 mac76 mac76:385_reb: Desaturation of Fatty Acids Requires a Mixed-Function Oxidase Palmitate and stearate serve as precursors of the two most common monounsaturated fatty acids of animal tissues: palmitoleate, 16:1(H9004 9 ), and oleate, 18:1(H9004 9 ); both of these fatty acids have a single cis double bond between C-9 and C-10 (see Table 10–1). The double bond is introduced into the fatty acid chain by an ox- idative reaction catalyzed by fatty acyl–CoA desatu- rase (Fig. 21–13), a mixed-function oxidase (Box 21–1). Two different substrates, the fatty acid and NADH or NADPH, simultaneously undergo two-electron oxidations. The path of electron flow includes a cyto- chrome (cytochrome b 5 ) and a flavoprotein (cyto- chrome b 5 reductase), both of which, like fatty acyl–CoA desaturase, are in the smooth ER. Bacteria have two cytochrome b 5 reductases, one NADH-dependent and the other NADPH-dependent; which of these is the main electron donor in vivo is unclear. In plants, oleate is produced by a stearoyl-ACP desaturase in the chloro- plast stroma that uses reduced ferredoxin as the elec- tron donor. Mammalian hepatocytes can readily introduce dou- ble bonds at the H9004 9 position of fatty acids but cannot in- troduce additional double bonds between C-10 and the Chapter 21 Lipid Biosynthesis798 BOX 21–1 THE WORLD OF BIOCHEMISTRY Mixed-Function Oxidases, Oxygenases, and Cytochrome P-450 In this chapter we encounter several enzymes that carry out oxidation-reduction reactions in which molecular oxygen is a participant. The reaction that introduces a double bond into a fatty acyl chain (see Fig. 21–13) is one such reaction. The nomenclature for enzymes that catalyze re- actions of this general type is often confusing to stu- dents, as is the mechanism of the reactions. Oxidase is the general name for enzymes that catalyze oxida- tions in which molecular oxygen is the electron ac- ceptor but oxygen atoms do not appear in the oxidized product (however, there is an exception to this “rule,” as we shall see!). The enzyme that creates a double bond in fatty acyl–CoA during the oxidation of fatty acids in peroxisomes (see Fig. 17–13) is an oxidase of this type; a second example is the cytochrome oxidase of the mitochondrial electron-transfer chain (see Fig. 19–14). In the first case, the transfer of two electrons to H 2 O produces hydrogen peroxide, H 2 O 2 ; in the sec- ond, two electrons reduce H5007 1 2 H5007 O 2 to H 2 O. Many, but not all, oxidases are flavoproteins. Oxygenases catalyze oxidative reactions in which oxygen atoms are directly incorporated into the substrate molecule, forming a new hydroxyl or carboxyl group, for example. Dioxygenases cat- alyze reactions in which both oxygen atoms of O 2 are incorporated into the organic substrate mole- cule. An example of a dioxygenase is tryptophan 2, 3-dioxygenase, which catalyzes the opening of the five-membered ring of tryptophan in the catabolism of this amino acid. When this reaction takes place in the presence of 18 O 2 , the isotopic oxygen atoms are found in the two carbonyl groups of the product (shown in red). Monooxygenases, more abundant and more complex in their action, catalyze reactions in which only one of the two oxygen atoms of O 2 is incorpo- rated into the organic substrate, the other being re- duced to H 2 O. Monooxygenases require two sub- strates to serve as reductants of the two oxygen atoms of O 2 . The main substrate accepts one of the two oxy- gen atoms, and a cosubstrate furnishes hydrogen atoms to reduce the other oxygen atom to H 2 O. The general reaction equation for monooxygenases is AH H11001 BH 2 H11001 OOO 88n AOOH H11001 B H11001 H 2 O where AH is the main substrate and BH 2 the cosub- strate. Because most monooxygenases catalyze reac- tions in which the main substrate becomes hydroxy- lated, they are also called hydroxylases. They are COO H5008 CH 2 H NH 3 CH H11001 O 2 H NH C O CH COO H5008 NH 3 H11001 tryptophan 2,3-dioxygenase N-Formylkynurenine N C O CH 2 Tryptophan 8885d_c21_787-832 2/26/04 9:35 AM Page 798 mac76 mac76:385_reb: 21.1 Biosynthesis of Fatty Acids and Eicosanoids 799 also sometimes called mixed-function oxidases or mixed-function oxygenases, to indicate that they oxidize two different substrates simultaneously. (Note here the use of “oxidase”—a deviation from the gen- eral meaning of this term noted above.) There are different classes of monooxygenases, depending on the nature of the cosubstrate. Some use reduced flavin nucleotides (FMNH 2 or FADH 2 ), others use NADH or NADPH, and still others use H9251- ketoglutarate as the cosubstrate. The enzyme that hydroxylates the phenyl ring of phenylalanine to form tyrosine is a monooxygenase for which tetrahy- drobiopterin serves as cosubstrate (see Fig. 18–23). This is the enzyme that is defective in the human ge- netic disease phenylketonuria. The most numerous and most complex monooxy- genation reactions are those employing a type of heme protein called cytochrome P-450. This cytochrome is usually present in the smooth ER rather than the mitochondria. Like mitochondrial cytochrome oxi- dase, cytochrome P-450 can react with O 2 and bind carbon monoxide, but it can be differentiated from cy- tochrome oxidase because the carbon monoxide com- plex of its reduced form absorbs light strongly at 450 nm—thus the name P-450. Cytochrome P-450 catalyzes hydroxylation reac- tions in which an organic substrate, RH, is hydroxy- lated to ROOH, incorporating one oxygen atom of O 2 ; the other oxygen atom is reduced to H 2 O by reducing equivalents that are furnished by NADH or NADPH but are usually passed to cytochrome P-450 by an iron- sulfur protein. Figure 1 shows a simplified outline of the action of cytochrome P-450, which has interme- diate steps not yet fully understood. Cytochrome P-450 is actually a family of similar proteins; several hundred members of this protein family are known, each with a different substrate specificity. In the adrenal cortex, for example, a spe- cific cytochrome P-450 participates in the hydroxyla- tion of steroids to yield the adrenocortical hormones (see Fig. 21–47). Cytochrome P-450 is also important in the hydroxylation of many different drugs, such as barbiturates and other xenobiotics (substances for- eign to the organism), particularly if they are hy- drophobic and relatively insoluble. The environmen- tal carcinogen benzo[a]pyrene (found in cigarette smoke) undergoes cytochrome P-450–dependent hydroxylation during detoxification. Hydroxylation of xenobiotics makes them more soluble in water and allows their excretion in the urine. Unfortunately, hy- droxylation of some compounds converts them to toxic substances, subverting the detoxification system. Reactions described in this chapter that are cat- alyzed by mixed-function oxidases are those involved in fatty acyl–CoA desaturation (Fig. 21–13), leukotri- ene synthesis (Fig. 21–16), plasmalogen synthesis (Fig. 21–30), conversion of squalene to cholesterol (Fig. 21–37), and steroid hormone synthesis (Fig. 21–47). FIGURE 1 NADPH Oxidized Reduced RH Reduced OxidizedNADP H11001 O 2 H 2 O ROH cytochrome P-450 reductase (Fe–S) cytochrome P-450 2H 2 O Monounsaturated fatty acyl–CoA O CH 3 CH 2 CH S-CoA (CH 2 ) m fatty acyl– CoA desaturase H11001 H11001 H11001O 2 Cyt b 5 reductase (FADH 2 ) Cyt b 5 reductase (FAD) 2 Cyt b 5 (Fe 2H11001 ) 2 Cyt b 5 (Fe 3H11001 ) 2H H11001 NADP H11001 (CH 2 ) n CH 2 O CH 3 C S-CoA (CH 2 ) m (CH 2 ) n CCH Saturated fatty acyl–CoA NADPH H11001 H H11001 FIGURE 21–13 Electron transfer in the desaturation of fatty acids in vertebrates. Blue arrows show the path of electrons as two substrates—a fatty acyl–CoA and NADPH—undergo oxidation by molecular oxygen. These reactions take place on the lumenal face of the smooth ER. A similar pathway, but with different electron carriers, occurs in plants. 8885d_c21_787-832 2/26/04 9:35 AM Page 799 mac76 mac76:385_reb: methyl-terminal end. Thus mammals cannot synthesize linoleate, 18:2(H9004 9,12 ), or H9251-linolenate, 18:3(H9004 9,12,15 ). Plants, however, can synthesize both; the desaturases that introduce double bonds at the H9004 12 and H9004 15 positions are located in the ER and the chloroplast. The ER en- zymes act not on free fatty acids but on a phospholipid, phosphatidylcholine, that contains at least one oleate linked to the glycerol (Fig. 21–14). Both plants and bac- teria must synthesize polyunsaturated fatty acids to ensure membrane fluidity at reduced temperatures. Because they are necessary precursors for the syn- thesis of other products, linoleate and linolenate are es- sential fatty acids for mammals; they must be ob- tained from dietary plant material. Once ingested, linoleate may be converted to certain other polyunsat- urated acids, particularly H9253-linolenate, eicosatrienoate, and arachidonate (eicosatetraenoate), all of which can be made only from linoleate (Fig. 21–12). Arachidonate, 20:4(H9004 5,8,11,14 ), is an essential precursor of regulatory lipids, the eicosanoids. The 20-carbon fatty acids are synthesized from linoleate (and linolenate) by fatty acid elongation reactions analogous to those described on page 797. Eicosanoids Are Formed from 20-Carbon Polyunsaturated Fatty Acids Eicosanoids are a family of very potent biological sig- naling molecules that act as short-range messengers, af- fecting tissues near the cells that produce them. In re- sponse to hormonal or other stimuli, phospholipase A 2 , present in most types of mammalian cells, attacks mem- brane phospholipids, releasing arachidonate from the middle carbon of glycerol. Enzymes of the smooth ER then convert arachidonate to prostaglandins, begin- ning with the formation of prostaglandin H 2 (PGH 2 ), the immediate precursor of many other prostaglandins and of thromboxanes (Fig. 21–15a). The two reactions that lead to PGH 2 are catalyzed by a bifunctional enzyme, cyclooxygenase (COX), also called prostaglandin H 2 synthase. In the first of two steps, the cyclooxyge- nase activity introduces molecular oxygen to convert arachidonate to PGG 2 . The second step, catalyzed by the peroxidase activity of COX, converts PGG 2 to PGH 2 . Aspirin (acetylsalicylate; Fig. 21–15b) irre- versibly inactivates the cyclooxygenase activity of COX by acetylating a Ser residue and blocking the enzyme’s active site, thus inhibiting the synthesis of prostaglandins and thromboxanes. Ibuprofen, a widely used nonsteroidal antiinflammatory drug (NSAID; Fig. 21–15c), inhibits the same enzyme. The recent discov- ery that there are two isozymes of COX has led to the development of more precisely targeted NSAIDs with fewer undesirable side effects (Box 21–2). Thromboxane synthase, present in blood platelets (thrombocytes), converts PGH 2 to thrombox- ane A 2 , from which other thromboxanes are derived (Fig. 21–15a). Thromboxanes induce constriction of blood vessels and platelet aggregation, early steps in blood clotting. Low doses of aspirin, taken regularly, re- duce the probability of heart attacks and strokes by re- ducing thromboxane production. ■ Thromboxanes, like prostaglandins, contain a ring of five or six atoms; the pathway from arachidonate to these two classes of compounds is sometimes called the “cyclic” pathway, to distinguish it from the “linear” path- way that leads from arachidonate to the leukotrienes, which are linear compounds (Fig. 21–16). Leukotriene synthesis begins with the action of several lipoxygenases that catalyze the incorporation of molecular oxygen into arachidonate. These enzymes, found in leukocytes and in heart, brain, lung, and spleen, are mixed-function ox- idases that use cytochrome P-450 (Box 21–1). The var- ious leukotrienes differ in the position of the peroxide Chapter 21 Lipid Biosynthesis800 CH O C O 2 OP H11001 desaturase 1 CH 2 OC O O O N(CH 3 ) 3 H 2 C H 2 CH O H5008 C Phosphatidylcholine containing linolenate, 18:3(H9004 9,12,15 ) CH O C O 2 OP H11001 9 1 18 CH 2 OC O O O N(CH 3 ) 3 H 2 C H 2 CH O H5008 C Phosphatidylcholine containing oleate, 18:1(H9004 9 ) 9 desaturase CH O C O 2 OP H11001 1 12 CH 2 OC O O O N(CH 3 ) 3 H 2 C H 2 CH O H5008 C Phosphatidylcholine containing linoleate, 18:2( H9004 9,12 ) 9 18 12 1815 FIGURE 21–14 Action of plant desaturases. Desaturases in plants ox- idize phosphatidylcholine-bound oleate to polyunsaturated fatty acids. Some of the products are released from the phosphatidylcholine by hydrolysis. 8885d_c21_787-832 2/26/04 9:35 AM Page 800 mac76 mac76:385_reb: COO H5008 O OH Phospholipid containing arachidonate phospholipase A 2 Lysophospholipid Arachidonate, 20:4(H9004 5,8,11,14 ) aspirin, ibuprofen peroxidase activity of COX PGH 2 2O 2 COO H5008 O COO H5008 O OOH PGG 2 O Other prostaglandins Thromboxanes (a) cyclooxgenase activity of COX H11001 CH 3 Ser O C COO H5008 Ibuprofen O Acetylated, inactivated COX H11001 CH 3 Ser O OH C COO H5008 Aspirin (acetylsalicylate) O COX (c) OH CH 3 CH 2 COO H5008 CH 3 Salicylate (b) CH CH CH 3 Naproxen CH CH 3 COO H5008 O CH 3 21.1 Biosynthesis of Fatty Acids and Eicosanoids 801 FIGURE 21–15 The “cyclic” pathway from arachidonate to pros- taglandins and thromboxanes. (a) After arachidonate is released from phospholipids by the action of phospholipase A 2 , the cyclooxygenase and peroxidase activities of COX (also called prostaglandin H 2 syn- thase) catalyze the production of PGH 2 , the precursor of other prostaglandins and thromboxanes. (b) Aspirin inhibits the first reac- tion by acetylating an essential Ser residue on the enzyme. (c) Ibupro- fen and naproxen inhibit the same step, probably by mimicking the structure of the substrate or an intermediate in the reaction. 114 5 85 COO H5008 Arachidonate LTD 4 HOO 5-Hydroperoxyeicosatetraenoate (5-HPETE) 12-Hydroperoxyeicosatetraenoate (12-HPETE) OOH Leukotriene A 4 (LTA 4 ) LTC 4 20 Other leukotrienes 12 COO H5008 lipoxygenase COO H5008 11 O 2 lipoxygenase multistep multistep O 2 FIGURE 21–16 The “linear” pathway from arachidonate to leukotrienes. 8885d_c21_787-832 2/26/04 9:35 AM Page 801 mac76 mac76:385_reb: group introduced by the lipoxygenases. This linear path- way from arachidonate, unlike the cyclic pathway, is not inhibited by aspirin or other NSAIDs. Plants also derive important signaling molecules from fatty acids. As in animals, a key step in the initia- tion of signaling involves activation of a specific phos- pholipase. In plants, the fatty acid substrate that is re- leased is H9251-linolenate. A lipoxygenase then catalyzes the first step in a pathway that converts linolenate to jas- monate, a substance known to have signaling roles in insect defense, resistance to fungal pathogens, and pollen maturation. Jasmonate (see Fig. 12–28) also af- fects seed germination, root growth, and fruit and seed development. Chapter 21 Lipid Biosynthesis802 BOX 21–2 BIOCHEMISTRY IN MEDICINE Relief Is in (the Active) Site: Cyclooxygenase Isozymes and the Search for a Better Aspirin Each year, several thousand tons of aspirin (acetyl- salicylate) are consumed around the world for the re- lief of headaches, sore muscles, inflamed joints, and fever. Because aspirin inhibits platelet aggregation and blood clotting, it is also used in low doses to treat pa- tients at risk of heart attacks. The medicinal proper- ties of the compounds known as salicylates, including aspirin, were first described by western science in 1763, when Edmund Stone of England noted that bark of the willow tree Salix alba was effective against fevers, aches, and pains. By the 1830s, German chemists had purified the active components from wil- low and from another plant rich in salicylates, the meadowsweet, Spiraea ulmaria. However, salicylate itself was bitter-tasting and its use had some un- pleasant side effects, including severe stomach irrita- tion in some cases. To address these problems, Felix Hoffmann and Arthur Eichengrun synthesized acetyl- salicylate at the Bayer company in Germany in 1897. The new compound, with fewer side effects than sal- icylate, was marketed in 1899 under the trade name Aspirin (from a for acetyl and spir for Spirsaüre, the German word for the acid prepared from Spiraea). Within a few years, aspirin was in widespread use. Aspirin (now a generic name) is one of a number of nonsteroidal antiinflammatory drugs (NSAIDs); others include ibuprofen and naproxen (see Fig. 21–15), all now sold over the counter. Unfortunately, aspirin reduces but does not eliminate the side effects of salicylates. In some patients, aspirin itself can pro- duce stomach bleeding, kidney failure, and, in extreme cases, death. New NSAIDs with the beneficial effects of aspirin but without its side effects would be med- ically valuable. Aspirin and other NSAIDs inhibit the cyclooxyge- nase activity of prostaglandin H 2 synthase (also called COX, for cyclooxygenase), which adds molecular oxy- gen to arachidonate to initiate prostaglandin synthe- sis (see Fig. 21–15a). Prostaglandins regulate many physiological processes, including platelet aggrega- tion, uterine contractions, pain, inflammation, and the secretion of mucins that protect the gastric mucosa from acid and proteolytic enzymes in the stomach. The stomach irritation that is a common side effect of as- pirin use results from the drug’s interference with the secretion of gastric mucin. Mammals have two isozymes of prostaglandin H 2 synthase, COX-1 and COX-2. These have different functions but closely similar amino acid sequences (60% to 65% sequence identity) and similar reaction mechanisms at both of their catalytic centers. COX-1 is responsible for the synthesis of the prostaglandins that regulate the secretion of gastric mucin, and COX- 2 for the prostaglandins that mediate inflammation, pain, and fever. Aspirin inhibits both isozymes about equally, so a dose sufficient to reduce inflammation also risks stomach irritation. Much research is aimed at developing new NSAIDs that inhibit COX-2 specif- ically, and several such drugs have become available. The development of COX-2–specific inhibitors has been helped immensely by knowledge of the detailed three-dimensional structures of COX-1 and COX-2 (Fig. 1). Both proteins are homodimers. Each mono- mer (M r 70,000) has an amphipathic domain that penetrates but does not span the ER; this anchors the enzyme on the lumenal side of the ER (a very unusual topology—generally the hydrophobic regions of inte- gral membrane proteins span the entire bilayer). Both catalytic sites are on the globular domain protruding into the ER lumen. COX-1 and COX-2 have virtually identical tertiary and quaternary structures, but they differ subtly in a long, thin hydrophobic channel extending from the membrane interior to the lumenal surface. The chan- nel includes both catalytic sites and is presumed to be the binding site for the hydrophobic substrate, arachi- donate. Both COX-1 and COX-2 have been crystallized in the presence of several different bound NSAID compounds, defining the NSAID-binding site (Fig. 1). The bound drugs block the hydrophobic channel and prevent arachidonate entry. The subtle differences be- tween the channels of COX-1 and COX-2 have guided 8885d_c21_787-832 2/26/04 9:35 AM Page 802 mac76 mac76:385_reb: 21.1 Biosynthesis of Fatty Acids and Eicosanoids 803 SUMMARY 21.1 Biosynthesis of Fatty Acids and Eicosanoids ■ Long-chain saturated fatty acids are synthesized from acetyl-CoA by a cytosolic complex of six enzyme activities plus acyl carrier protein (ACP). The fatty acid synthase complex, which in some organisms consists of multifunctional polypeptides, contains two types of OSH groups (one furnished by the phosphopantetheine of ACP, the other by a Cys residue of H9252-ketoacyl-ACP synthase) that function as carriers of the fatty acyl intermediates. the design of NSAIDs that selectively fit COX-2 and therefore inhibit COX-2 more effectively than COX-1. Two of these drugs have been approved for use world- wide: celecoxib (Celebrex) for osteoarthritis and rheumatoid arthritis, and rofecoxib (Vioxx) for os- teoarthritis and acute musculoskeletal pain (Fig. 2). In clinical trials, these drugs have proven effective while significantly reducing the stomach irritation and other side effects of aspirin and other NSAIDs. The use of precise structural information about an en- zyme’s active site is a powerful tool in the develop- ment of better, more specific drugs. CF 3 S O H 2 N N N H 3 C Celecoxib (Celebrex) O O O Rofecoxib (Vioxx) S O H 3 C O (b) (a) FIGURE 1 Structures of COX-1 and COX-2. (a) COX-1 with an NSAID inhibitor (flurbiprofen, orange) bound (PDB ID 3PGH). The enzyme consists of two identical monomers (gray and blue) each with three domains: a mem- brane anchor consisting of four amphipathic helices; a second domain that somewhat resem- bles a domain of the epidermal growth factor; and the catalytic domain, which contains the cyclooxygenase and peroxidase activities, as well as the hydrophobic channel in which the sub- strate (arachidonate) binds. The heme that is part of the peroxidase active sites is shown in red; Tyr 385 , a key residue in the cyclooxygenase site, is turquoise. Other catalytically important residues include Arg 120 (dark blue), His 388 (green), and Ser 530 (yellow). Flurbiprofen blocks access to the enzyme active site. (b) A look at COX-1 and COX-2 side by side. The COX-1 en- zyme from sheep is shown at left, with bound ibuprofen (PDB ID 1EQG). The COX-2 enzyme from mouse is shown at right, with a similar inhibitor, Sc-558, bound (PDB ID 6COX). The inhibitors (red) are partially buried within the structures in these representations, which emphasize surface contours. COX-1 and COX-2 are very sim- ilar in structure, but enzymologists exploit the small differences in the structures of the active sites and the channels leading to them to design inhibitors specific for one enzyme or the other. FIGURE 2 Two COX-2–specific drugs that bind to COX-2 about 1,000 times better than to COX-1. F CH CH 3 COO H5008 Flurbiprofen 8885d_c21_787-832 2/26/04 9:35 AM Page 803 mac76 mac76:385_reb: ■ Malonyl-ACP, formed from acetyl-CoA (shuttled out of mitochondria) and CO 2 , condenses with an acetyl bound to the CysOSH to yield acetoacetyl-ACP, with release of CO 2 . This is followed by reduction to the D-H9252-hydroxy derivative, dehydration to the trans-H9004 2 - unsaturated acyl-ACP, and reduction to butyryl-ACP. NADPH is the electron donor for both reductions. Fatty acid synthesis is regulated at the level of malonyl-CoA formation. ■ Six more molecules of malonyl-ACP react successively at the carboxyl end of the growing fatty acid chain to form palmitoyl-ACP—the end product of the fatty acid synthase reaction. Free palmitate is released by hydrolysis. ■ Palmitate may be elongated to the 18-carbon stearate. Palmitate and stearate can be desaturated to yield palmitoleate and oleate, respectively, by the action of mixed-function oxidases. ■ Mammals cannot make linoleate and must obtain it from plant sources; they convert exogenous linoleate to arachidonate, the parent compound of eicosanoids (prostaglandins, thromboxanes, and leukotrienes), a family of very potent signaling molecules. 21.2 Biosynthesis of Triacylglycerols Most of the fatty acids synthesized or ingested by an or- ganism have one of two fates: incorporation into tria- cylglycerols for the storage of metabolic energy or in- corporation into the phospholipid components of membranes. The partitioning between these alternative fates depends on the organism’s current needs. During rapid growth, synthesis of new membranes requires the production of membrane phospholipids; when an or- ganism has a plentiful food supply but is not actively growing, it shunts most of its fatty acids into storage fats. Both pathways begin at the same point: the for- mation of fatty acyl esters of glycerol. In this section we examine the route to triacylglycerols and its regulation, and the production of glycerol 3-phosphate in the process of glyceroneogenesis. Triacylglycerols and Glycerophospholipids Are Synthesized from the Same Precursors Animals can synthesize and store large quantities of tri- acylglycerols, to be used later as fuel (see Box 17–1). Humans can store only a few hundred grams of glyco- gen in liver and muscle, barely enough to supply the body’s energy needs for 12 hours. In contrast, the total amount of stored triacylglycerol in a 70-kg man of average build is about 15 kg, enough to support basal energy needs for as long as 12 weeks (see Table 23–5). Triacylglycerols have the highest energy content of all stored nutrients—more than 38 kJ/g. Whenever carbo- hydrate is ingested in excess of the organism’s capacity to store glycogen, the excess is converted to triacyl- glycerols and stored in adipose tissue. Plants also man- ufacture triacylglycerols as an energy-rich fuel, mainly stored in fruits, nuts, and seeds. In animal tissues, triacylglycerols and glycerophos- pholipids such as phosphatidylethanolamine share two precursors (fatty acyl–CoA and L-glycerol 3-phosphate) and several biosynthetic steps. The vast majority of the glycerol 3-phosphate is derived from the glycolytic intermediate dihydroxyacetone phosphate (DHAP) by the action of the cytosolic NAD-linked glycerol 3-phosphate dehydrogenase; in liver and kidney, a small amount of glycerol 3-phosphate is also formed from glycerol by the action of glycerol kinase (Fig. 21–17). The other precursors of triacylglycerols are fatty acyl–CoAs, formed from fatty acids by acyl-CoA synthetases, the same enzymes responsible for the ac- tivation of fatty acids for H9252 oxidation (see Fig. 17–5). The first stage in the biosynthesis of triacylglycerols is the acylation of the two free hydroxyl groups of L- glycerol 3-phosphate by two molecules of fatty acyl–CoA to yield diacylglycerol 3-phosphate, more commonly called phosphatidic acid or phosphatidate (Fig. 21–17). Phosphatidic acid is present in only trace amounts in cells but is a central intermediate in lipid biosynthesis; it can be converted either to a triacylglycerol or to a glycerophospholipid. In the pathway to triacylglycerols, phosphatidic acid is hydrolyzed by phosphatidic acid phosphatase to form a 1,2-diacylglycerol (Fig. 21–18). Diacylglycerols are then converted to triacylglycerols by transesterification with a third fatty acyl–CoA. Triacylglycerol Biosynthesis in Animals Is Regulated by Hormones In humans, the amount of body fat stays relatively con- stant over long periods, although there may be minor short-term changes as caloric intake fluctuates. Carbo- hydrate, fat, or protein consumed in excess of energy needs is stored in the form of triacylglycerols that can be drawn upon for energy, enabling the body to with- stand periods of fasting. Biosynthesis and degradation of triacylglycerols are regulated such that the favored path de- pends on the metabolic resources and requirements of the moment. The rate of triacylglycerol biosynthesis is profoundly altered by the action of several hormones. Insulin, for example, promotes the conversion of car- bohydrate to triacylglycerols (Fig. 21–19). People with severe diabetes mellitus, due to failure of insulin se- cretion or action, not only are unable to use glucose properly but also fail to synthesize fatty acids from Chapter 21 Lipid Biosynthesis804 8885d_c21_787-832 2/26/04 9:35 AM Page 804 mac76 mac76:385_reb: O C P CHOH HO S-CoA CH 2 OH Glucose glycolysis Dihydroxyacetone phosphate glycerol 3-phosphate dehydrogenase glycerol kinase Glycerol CoA-SH L-Glycerol 3-phosphate acyl transferase Phosphatidic acid O H5008 CH 2 CH 2 OH C CH 2 O O P O O H5008 O H5008 CH 2 OH C CH 2 O P O O H5008 O H5008 H H ADP CH 2 OH R 2 C O ATP R 2 AMP R 1 S-CoA COO H5008 CoA-SH acyl-CoA synthetase C O ATP H11001 R 1 CoA-SH R 1 CO O C O CH 2 O O O H5008 acyl transferase PP AMP R 2 COO H5008 acyl-CoA synthetase ATP H11001 CoA-SH PP i H11001 H H11001 NAD H11001 NADH carbohydrates or amino acids. If the diabetes is un- treated, these individuals have increased rates of fat oxidation and ketone body formation (Chapter 17) and therefore lose weight. ■ An additional factor in the balance between biosyn- thesis and degradation of triacylglycerols is that ap- proximately 75% of all fatty acids released by lipolysis are reesterified to form triacylglycerols rather than used for fuel. This ratio persists even under starvation con- ditions, when energy metabolism is shunted from the use of carbohydrate to the oxidation of fatty acids. Some of this fatty acid recycling takes place in adipose tissue, with the reesterification occurring before release into the bloodstream; some takes place via a systemic cycle in which free fatty acids are transported to liver, recy- cled to triacylglycerol, exported again to the blood (transport of lipids in the blood is discussed in Section 21.4), and taken up again by adipose tissue after release from triacylglycerol by extracellular lipoprotein lipase 21.2 Biosynthesis of Triacylglycerols 805 FIGURE 21–17 Biosynthesis of phosphatidic acid. A fatty acyl group is activated by formation of the fatty acyl–CoA, then transferred to es- ter linkage with L-glycerol 3-phosphate, formed in either of the two ways shown. Phosphatidic acid is shown here with the correct stere- ochemistry at C-2 of the glycerol molecule. To conserve space in sub- sequent figures (and in Fig. 21–14), both fatty acyl groups of glyc- erophospholipids, and all three acyl groups of triacylglycerols, are shown projecting to the right. O CCH PCH 2 R 2 OC O R 1 O O O H5008 CH 2 O Glycerophospholipid attachment of head group (serine, choline, ethanolamine, etc.) phosphatidic acid phosphatase OC CH CH 2 R 3 OC O R 1 O O CH 2 O Triacylglycerol CCH CH 2 R 3 OC O R 1 O CH 2 O 1,2-Diacylglycerol OH O Head group R 2 C O S-CoA CoA-SH acyl transferase O CCH PO H5008 CH 2 R 2 OC O R 1 O O O H5008 CH 2 O Phosphatidic acid CR 2 FIGURE 21–18 Phosphatidic acid in lipid biosynthesis. Phosphatidic acid is the precursor of both triacylglycerols and glycerophospholipids. The mechanisms for head-group attachment in phospholipid synthe- sis are described later in this section. 8885d_c21_787-832 2/26/04 9:35 AM Page 805 mac76 mac76:385_reb: (Fig. 21–20; see also Fig. 17–1). Flux through this tri- acylglycerol cycle between adipose tissue and liver may be quite low when other fuels are available and the release of fatty acids from adipose tissue is limited, but as noted above, the proportion of released fatty acids that are reesterified remains roughly constant at 75% under all metabolic conditions. The level of free fatty acids in the blood thus reflects both the rate of release of fatty acids and the balance between the synthesis and breakdown of triacylglycerols in adipose tissue and liver. When the mobilization of fatty acids is required to meet energy needs, release from adipose tissue is stim- ulated by the hormones glucagon and epinephrine (see Figs 17–3, 17–12). Simultaneously, these hormonal sig- nals decrease the rate of glycolysis and increase the rate of gluconeogenesis in the liver (providing glucose for the brain, as further elaborated in Chapter 23). The re- leased fatty acid is taken up by a number of tissues, in- cluding muscle, where it is oxidized to provide energy. Much of the fatty acid taken up by liver is not oxidized but is recycled to triacylglycerol and returned to adi- pose tissue. The function of the apparently futile triacylglycerol cycle (futile cycles are discussed in Chapter 15) is not well understood. However, as we learn more about how the triacylglycerol cycle is sustained via metabolism in two separate organs and is coordinately regulated, some possibilities emerge. For example, the excess capacity in the triacylglycerol cycle (the fatty acid that is even- tually reconverted to triacylglycerol rather than oxi- dized as fuel) could represent an energy reserve in the bloodstream during fasting, one that would be more rap- idly mobilized in a “fight or flight” emergency than would stored triacylglycerol. The constant recycling of triacylglycerols in adipose tissue even during starvation raises a second question: what is the source of the glycerol 3-phosphate required for this process? As noted above, glycolysis is sup- pressed in these conditions by the action of glucagon and epinephrine, so little DHAP is available, and glyc- erol released during lipolysis cannot be converted di- rectly to glycerol 3-phosphate in adipose tissue, because these cells lack glycerol kinase (Fig. 21–17). So, how is sufficient glycerol 3-phosphate produced? The answer lies in a pathway discovered more than three decades ago and given little attention until recently, a pathway intimately linked to the triacylglycerol cycle and, in a larger sense, to the balance between fatty acid and car- bohydrate metabolism. Adipose Tissue Generates Glycerol 3-phosphate by Glyceroneogenesis Glyceroneogenesis is a shortened version of gluco- neogenesis, from pyruvate to DHAP (see Fig. 14–16), followed by conversion of the DHAP to glycerol 3- phosphate by cytosolic NAD-linked glycerol 3-phosphate dehydrogenase (Fig. 21–21). Glycerol 3-phosphate is subsequently used in triacylglycerol synthesis. Glycero- Chapter 21 Lipid Biosynthesis806 Dietary carbohydrates Glucose Dietary proteins Amino acids Acetyl-CoA Fatty acids Triacylglycerols Ketone bodies (acetoacetate, D- -hydroxybutyrate, acetone) insulin increased in diabetes H9252 FIGURE 21–19 Regulation of triacylglycerol synthesis by in- sulin. Insulin stimulates conversion of dietary carbohydrates and proteins to fat. Individuals with diabetes mellitus lack insulin; in uncontrolled disease, this results in diminished fatty acid synthesis, and the acetyl-CoA arising from catabolism of carbohydrates and pro- teins is shunted instead to ketone body production. People in severe ketosis smell of acetone, so the condition is sometimes mistaken for drunkenness (p. 909). Adipose tissue Glycerol 3-phosphate Glycerol 3-phosphate Fuel for tissues Glycerol Triacylglycerol Triacylglycerol Fatty acid Fatty acid Lipoprotein lipase LiverBlood Glycerol FIGURE 21–20 The triacylglycerol cycle. In mammals, triacylglycerol molecules are broken down and resynthesized in a triacylglycerol cy- cle during starvation. Some of the fatty acids released by lipolysis of triacylglycerol in adipose tissue pass into the bloodstream, and the re- mainder are used for resynthesis of triacylglycerol. Some of the fatty acids released into the blood are used for energy (in muscle, for ex- ample), and some are taken up by the liver and used in triacylglyc- erol synthesis. The triacylglycerol formed in the liver is transported in the blood back to adipose tissue, where the fatty acid is released by extracellular lipoprotein lipase, taken up by adipocytes, and reesteri- fied into triacylglycerol. 8885d_c21_787-832 2/26/04 9:35 AM Page 806 mac76 mac76:385_reb: neogenesis was discovered in the 1960s by Lea Reshef, Richard Hanson, and John Ballard, and simultaneously by Eleazar Shafrir and his coworkers, who were in- trigued by the presence of two gluconeogenic enzymes, pyruvate carboxylase and phosphoenolpyruvate (PEP) carboxykinase, in adipose tissue, where glucose is not synthesized. After a long period of inattention, interest in this pathway has been renewed by the demonstration of a link between glyceroneogenesis and late-onset (type 2) diabetes, as we shall see. Glyceroneogenesis has multiple roles. In adipose tis- sue, glyceroneogenesis coupled with reesterification of free fatty acids controls the rate of fatty acid release to the blood. In brown adipose tissue, the same pathway may control the rate at which free fatty acids are deliv- ered to mitochondria for use in thermogenesis (see Fig. 19–30). And in fasting humans, glyceroneogenesis in the liver alone supports the synthesis of enough glyc- erol 3-phosphate to account for up to 65% of fatty acids reesterified to triacylglycerol. Flux through the triacylglycerol cycle between liver and adipose tissue is controlled to a large degree by the activity of PEP carboxykinase, which limits the rate of both gluconeogenesis and glyceroneogenesis. Gluco- corticoid hormones such as cortisol (a biological steroid derived from cholesterol; see Fig. 21–46) and dexa- methasone (a synthetic glucocorticoid) regulate the levels of PEP carboxykinase reciprocally in the liver and adipose tissue. Acting through the glucocorticoid re- ceptor, these steroid hormones increase the expression of the gene encoding PEP carboxykinase in the liver, thus increasing gluconeogenesis and glyceroneogenesis (Fig. 21–22). Stimulation of glyceroneogenesis leads to an in- crease in the synthesis of triacylglycerol molecules in the liver and their release into the blood. At the same time, glucocorticoids suppress the expression of the gene encoding PEP carboxykinase in adipose tissue. This results in a decrease in glyceroneogenesis in adi- pose tissue; recycling of fatty acids declines as a result, and more free fatty acids are released into the blood. Thus glyceroneogenesis is regulated reciprocally in the liver and adipose tissue, affecting lipid metabolism in opposite ways: a lower rate of glyceroneogenesis in adi- pose tissue leads to more fatty acid release (rather than recycling), whereas a higher rate in the liver leads to more synthesis and export of triacylglycerols. The net result is an increase in flux through the triacylglycerol cycle. When the glucocorticoids are no longer present, flux through the cycle declines as the expression of PEP carboxykinase increases in adipose tissue and decreases in the liver. The recent attention given to glyceroneogenesis has arisen in part from the connection between this pathway and diabetes. High levels of free fatty acids in the blood interfere with glucose utilization in muscle and promote the insulin resistance that leads to type 2 diabetes. A new class of drugs called thiazolidine- diones have been shown to reduce the levels of fatty acids circulating in the blood and increase sensitivity to 21.2 Biosynthesis of Triacylglycerols 807 Pyruvate pyruvate carboxylase Oxaloacetate PEP carboxykinase Phosphoenolpyruvate multistep Dihydroxyacetone phosphate glycerol 3-phosphate dehydrogenase CH 2 OH CHOH CH 2 O H11002 O H11002 O O P Glycerol 3-phosphate Triacylglycerol synthesis FIGURE 21–21 Glyceroneogenesis. The pathway is essentially an ab- breviated version of gluconeogenesis, from pyruvate to dihydroxyace- tone phosphate (DHAP), followed by conversion of DHAP to glycerol 3-phosphate, which is used for the synthesis of triacylglycerol. C CH 2 OH O Cortisol O HC 3 HC 3 OH HO C CH 2 OH O Dexamethasone O F HC 3 HC 3 HC 3 OH HO CH 3 CH 3 CH 2 NN N H11002 O S Rosiglitazone (Avandia) Thiazolidinediones Pioglitazone (Actos) N O O N H11002 S O O O 8885d_c21_787-832 2/26/04 9:35 AM Page 807 mac76 mac76:385_reb: insulin. Thiazolidinediones bind to and activate a nu- clear hormone receptor called peroxisome proliferator- activated receptor H9253 (PPARH9253), leading to the induction in adipose tissue of PEP carboxykinase (Fig. 21–22); a higher activity of PEP carboxykinase then leads to in- creased synthesis of the precursors of glyceroneogene- sis. The therapeutic effect of thiazolidinediones is thus due, at least in part, to the increase in glyceroneogen- esis, which in turn increases the resynthesis of triacyl- glycerol in adipose tissue and reduces the release of free fatty acid from adipose tissue into the blood. ■ SUMMARY 21.2 Biosynthesis of Triacylglycerols ■ Triacylglycerols are formed by reaction of two molecules of fatty acyl–CoA with glycerol 3-phosphate to form phosphatidic acid; this product is dephosphorylated to a diacylglycerol, then acylated by a third molecule of fatty acyl–CoA to yield a triacylglycerol. ■ The synthesis and degradation of triacylglycerol are hormonally regulated. ■ Mobilization and recycling of triacylglycerol molecules results in a triacylglycerol cycle. Triacylglycerols are resynthesized from free fatty acids and glycerol 3-phosphate even during starvation. The dihydroxyacetone phosphate precursor of glycerol 3-phosphate is derived from pyruvate via glyceroneogenesis. 21.3 Biosynthesis of Membrane Phospholipids In Chapter 10 we introduced two major classes of mem- brane phospholipids: glycerophospholipids and sphin- golipids. Many different phospholipid species can be constructed by combining various fatty acids and polar head groups with the glycerol or sphingosine backbone (see Figs 10–8, 10–12). All the biosynthetic pathways follow a few basic patterns. In general, the assembly of phospholipids from simple precursors requires (1) syn- thesis of the backbone molecule (glycerol or sphingo- sine); (2) attachment of fatty acid(s) to the backbone through an ester or amide linkage; (3) addition of a hy- drophilic head group to the backbone through a phos- phodiester linkage; and, in some cases, (4) alteration or exchange of the head group to yield the final phospho- lipid product. In eukaryotic cells, phospholipid synthesis occurs primarily on the surfaces of the smooth endoplasmic reticulum and the mitochondrial inner membrane. Some newly formed phospholipids remain at the site of syn- thesis, but most are destined for other cellular locations. Chapter 21 Lipid Biosynthesis808 FIGURE 21–22 Regulation of glyceroneogenesis. (a) Gluco- corticoid hormones stimulate glyceroneogenesis and gluco- neogenesis in the liver, while suppressing glyceroneogenesis in the adipose tissue (by reciprocal regulation of the gene expressing PEP carboxykinase (PEPCK) in the two tissues); this increases the flux through the triacylglycerol cycle. The glycerol freed by the breakdown of triacylglycerol in adipose tissue is released to the blood and trans- ported to the liver, where it is primarily converted to glucose, although some is converted to glycerol 3-phosphate by glycerol kinase. (b) A class of drugs called thiazolidinediones are now used to treat type 2 diabetes. In this disease, high levels of free fatty acids in the blood interfere with glucose utilization in muscle and promote in- sulin resistance. Thiazolidinediones activate a nuclear receptor called peroxisome proliferator-activated receptor H9253 (PPARH9253), which induces the activity of PEP carboxykinase. Therapeutically, thiazolidinediones increase the rate of glyceroneogenesis, thus increasing the resynthe- sis of triacylglycerol in adipose tissue and reducing the amount of free fatty acid in the blood. Adipose tissue Glycerol 3-phosphate Glycerol 3-phosphate Fuel for tissues Glycerol Triacylglycerol Triacylglycerol Fatty acid Fatty acid LiverBlood Glycerol glycero- neogenesis (a) (b) Pyruvate Pyruvate DNA PEPCKPEPCK Glycerol 3-phosphate Glycerol 3-phosphate Fuel for tissues Glycerol Triacylglycerol Triacylglycerol Fatty acid Fatty acid Glycerol glycero- neogenesis Pyruvate DNA PEPCK Glucocorticoids Thiazolidinediones Lipoprotein lipase 8885d_c21_808 2/27/04 12:40 PM Page 808 mac116 mac116: The process by which water-insoluble phospholipids move from the site of synthesis to the point of their eventual function is not fully understood, but we con- clude this section by discussing some mechanisms that have emerged in recent years. Cells Have Two Strategies for Attaching Phospholipid Head Groups The first steps of glycerophospholipid synthesis are shared with the pathway to triacylglycerols (Fig. 21–17): two fatty acyl groups are esterified to C-1 and C-2 of L-glycerol 3-phosphate to form phosphatidic acid. Com- monly but not invariably, the fatty acid at C-1 is satu- rated and that at C-2 is unsaturated. A second route to phosphatidic acid is the phosphorylation of a diacyl- glycerol by a specific kinase. The polar head group of glycerophospholipids is at- tached through a phosphodiester bond, in which each of two alcohol hydroxyls (one on the polar head group and one on C-3 of glycerol) forms an ester with phos- phoric acid (Fig. 21–23). In the biosynthetic process, one of the hydroxyls is first activated by attachment of a nucleotide, cytidine diphosphate (CDP). Cytidine monophosphate (CMP) is then displaced in a nucle- ophilic attack by the other hydroxyl (Fig. 21–24). The CDP is attached either to the diacylglycerol, forming the activated phosphatidic acid CDP-diacylglycerol (strategy 1), or to the hydroxyl of the head group (strat- egy 2). Eukaryotic cells employ both strategies, whereas prokaryotes use only the first. The central importance of cytidine nucleotides in lipid biosynthesis was discov- ered by Eugene P. Kennedy in the early 1960s. 21.3 Biosynthesis of Membrane Phospholipids 809 Eugene P. Kennedy O CCH P CH 2 R 2 O C O R 1 O O O H5008 CH 2 O Glycerophospholipid CCH CH 2 O C O R 1 O CH 2 ODiacylglycerol O O Head group R 2 HHO OH P O O H5008 Head group HO H 2 O H 2 O phosphodiester alcohol Phosphoric acid alcohol FIGURE 21–23 Head-group attachment. The phospholipid head group is attached to a diacylglycerol by a phospho- diester bond, formed when phosphoric acid condenses with two alcohols, eliminating two molecules of H 2 O. OH Head group O CCH PO H5008 CH 2 R 2 O C O R 1 O O CH 2 O Glycerophospholipid O P O Rib Cytosine H5008 O O Strategy 2 CMP HO Head group O CCH PO H5008 CH 2 R 2 O C O R 1 O O CH 2 O CDP-diacylglycerol O P Rib Cytosine O H5008 O Strategy 1 Diacylglycerol activated with CDP Head group activated with CDP CMP CCH CH 2 R 2 O C O R 1 O O CH 2 O P O O H5008 O O Head group 1,2-Diacylglycerol FIGURE 21–24 Two general strategies for forming the phosphodiester bond of phospholipids. In both cases, CDP supplies the phosphate group of the phosphodiester bond. 8885d_c21_809 2/27/04 12:41 PM Page 809 mac116 mac116: Chapter 21 Lipid Biosynthesis810 OP O H11002 CDP-diacylglycerol O Rib Cytosine CR 1 O C O R 2 O CH O CCH CH 2 R 2 C O R 1 O O CH 2 P O H11002 O CTP CMP CCH CH 2 R 2 OC O R 1 O O CH 2 O P O H11002 OO H11002 O CH 2 O C CH 2 R 2 OO O O C O OR 1 O OO O CH 2 O P O O O NH 3 O CH O O H11002 CH 2 H11001 Phosphatidylethanolamine CHOH CH 2 OP O H11002 O Glycerol 3- phosphate cardiolipin PP i C CH CH 2 R 2 O C O R 1 O O CH 2 O PO CO 2 O CH O H11002 CH 2 Phosphatidylglycerol 3-phosphate OH CH 2 OP O H11002 O O H11002 PG 3-phosphate synthase Glycerol Cardiolipin Phosphatidylglycerol decarboxylase P i H 2 O PG 3-phosphate PS O C CHCH 2 R 2 OO O O C O OR 1 O OO O CH 2 O P O COO H11002 O O O NH 3 O CH O O H11002 CH 2 H11001 Phosphatidylserine CH 2 O C CH CH 2 OH R 2 O C O R 1 O CH 2 O O CH CH 2 OH CH 2 OP O H11002 O Phosphatidylglycerol CH 2 OOP O H11002 O CH 2 CR 2 C O R 1 O CH 2 O O CH CH 2 O phosphatase synthase (bacterial) Serine PS synthase CMP O O FIGURE 21–25 Origin of the polar head groups of phospholipids in E. coli. Initially, a head group (either serine or glycerol 3-phosphate) is attached via a CDP- diacylglycerol intermediate (strategy 1 in Fig. 21–24). For phospholipids other than phosphatidylserine, the head group is further modified, as shown here. In the enzyme names, PG represents phosphatidylglycerol; PS, phosphatidylserine. 8885d_c21_787-832 2/26/04 9:35 AM Page 810 mac76 mac76:385_reb: 21.3 Biosynthesis of Membrane Phospholipids 811 Phospholipid Synthesis in E. coli Employs CDP-Diacylglycerol The first strategy for head-group attachment is illus- trated by the synthesis of phosphatidylserine, phos- phatidylethanolamine, and phosphatidylglycerol in E. coli. The diacylglycerol is activated by condensation of phosphatidic acid with cytidine triphosphate (CTP) to form CDP-diacylglycerol, with the elimination of pyro- phosphate (Fig. 21–25). Displacement of CMP through nucleophilic attack by the hydroxyl group of serine or by the C-1 hydroxyl of glycerol 3-phosphate yields phos- phatidylserine or phosphatidylglycerol 3-phosphate, respectively. The latter is processed further by cleavage of the phosphate monoester (with release of P i ) to yield phosphatidylglycerol. Phosphatidylserine and phosphatidylglycerol can serve as precursors of other membrane lipids in bacte- ria (Fig. 21–25). Decarboxylation of the serine moiety in phosphatidylserine, catalyzed by phosphatidylserine decarboxylase, yields phosphatidylethanolamine. In E. coli, condensation of two molecules of phosphati- dylglycerol, with elimination of one glycerol, yields cardiolipin, in which two diacylglycerols are joined through a common head group. Eukaryotes Synthesize Anionic Phospholipids from CDP-Diacylglycerol In eukaryotes, phosphatidylglycerol, cardiolipin, and the phosphatidylinositols (all anionic phospholipids; see Fig. 10–8) are synthesized by the same strategy used for phospholipid synthesis in bacteria. Phosphatidylglycerol is made exactly as in bacteria. Cardiolipin synthesis in eukaryotes differs slightly: phosphatidylglycerol con- denses with CDP-diacylglycerol (Fig. 21–26), not an- other molecule of phosphatidylglycerol as in E. coli (Fig. 21–25). Phosphatidylinositol is synthesized by condensation of CDP-diacylglycerol with inositol (Fig. 21–26). Spe- cific phosphatidylinositol kinases then convert phos- phatidylinositol to its phosphorylated derivatives (see Fig. 10–17). Phosphatidylinositol and its phosphory- lated products in the plasma membrane play a central role in signal transduction in eukaryotes (see Figs 12–8, 12–19). OP O H11002 CDP-diacylglycerol O Rib Cytosine CR 1 O C O R 2 O CH O CH 2 O P O H11002 O Inositol CMP PI synthase CCH R 2 OC O R 1 O CH 2 O O CH 2 CHOH CH 2 O P O Phosphatidylglycerol CMP (eukaryotic) Cardiolipin C PO 2H11002 C O R 1 O O CH 2 O PO H O H11002 3 . CH 2 OO P O H11002 O CH 2 CR 2 C O R 1 O CH 2 O O CH O cardiolipin synthase O CH O OH H H OH H HH OH OH Phosphatidylinositol OH R 2 OH CH 2 CH 2 These groups can also be esterified with O H11002 FIGURE 21–26 Synthesis of cardiolipin and phosphatidylinositol in eukaryotes. These glycerophospholipids are synthesized using strategy 1 in Figure 21-24. Phosphatidylglycerol is synthesized as in bacteria (see Fig. 21–25). PI represents phosphatidylinositol. 8885d_c21_787-832 2/26/04 9:35 AM Page 811 mac76 mac76:385_reb: Eukaryotic Pathways to Phosphatidylserine, Phosphatidylethanolamine, and Phosphatidylcholine Are Interrelated Yeast, like bacteria, can produce phosphatidylserine by condensation of CDP-diacylglycerol and serine, and can synthesize phosphatidylethanolamine from phosphatidyl- serine in the reaction catalyzed by phosphatidylserine decarboxylase (Fig. 21–27). In mammalian cells, an al- ternative route to phosphatidylserine is a head-group exchange reaction, in which free serine displaces ethanolamine. Phosphatidylethanolamine may also be converted to phosphatidylcholine (lecithin) by the addition of three methyl groups to its amino group; S- adenosylmethionine is the methyl group donor (see Fig. 18–18) for all three methylation reactions. In mammals, phosphatidylserine is not synthesized from CDP-diacylglycerol; instead, it is derived from phosphatidylethanolamine via the head-group exchange reaction (Fig. 21–27). Synthesis of phosphatidylethan- olamine and phosphatidylcholine in mammals occurs by strategy 2 of Figure 21–24: phosphorylation and activa- tion of the head group, followed by condensation with Chapter 21 Lipid Biosynthesis812 methyltransferase Ethanolamine C CH 2 R 2 O C O R 1 O O CH 2 O PO O (CH 3 ) 3 CH O H11002 CH 2 C CH CO 2 R 2 O C O R 1 O O CH 2 O P COO H11002 O O NH 3 CH O H11002 CH 2 H11001 Phosphatidylcholine Phosphatidylserine phosphatidyl- serine decarboxylase Serine 3 adoMet CH 2 3 adoHcy C CH 2 R 2 O C O R 1 O O CH 2 O PO O NH 3 CH O H11002 CH 2 H11001 Phosphatidylethanolamine CH 2 N H11001 CH 2 phosphatidyl- ethanolamine– serine transferase FIGURE 21–27 The “salvage” pathway from phosphatidylserine to phosphatidylethanolamine and phosphatidylcholine in yeast. Phos- phatidylserine and phosphatidylethanolamine are interconverted by a reversible head-group exchange reaction. In mammals, phos- phatidylserine is derived from phosphatidylethanolamine by a re- versal of this reaction; adoMet is S-adenosylmethionine; adoHcy, S- adenosylhomocysteine. CMP CDP-choline– diacylglycerol phosphocholine transferase CDP-choline CytosineOP Diacylglycerol Rib CholineCH 2 HO CH 2 H11001 N(CH 3 ) 3 O O H11002 ADP choline kinase ATP CH 2 O H11002 O P CH 2 H11001 N(CH 3 ) 3 O H11002 O CH 2 OP CH 2 H11001 N(CH 3 ) 3 O H11002 O O PP i CTP-choline cytidylyl transferase Phosphocholine CTP Phosphatidylcholine CH 2 R 2 C O CH O O O H11002 R 1 CCH 2 O O CH 2 POO CH 2 H11001 N(CH 3 ) 3 FIGURE 21–28 Pathway for phosphatidylcholine synthesis from choline in mammals. The same strategy shown here (strategy 2 in Fig. 21–24) is also used for salvaging ethanolamine in phosphatidyle- thanolamine synthesis. 8885d_c21_787-832 2/26/04 9:35 AM Page 812 mac76 mac76:385_reb: diacylglycerol. For example, choline is reused (“sal- vaged”) by being phosphorylated then converted to CDP-choline by condensation with CTP. A diacylgly- cerol displaces CMP from CDP-choline, producing phosphatidylcholine (Fig. 21–28). An analogous sal- vage pathway converts ethanolamine obtained in the diet to phosphatidylethanolamine. In the liver, phos- phatidylcholine is also produced by methylation of phosphatidylethanolamine (with S-adenosylmethionine, as described above), but in all other tissues phos- phatidylcholine is produced only by condensation of diacylglycerol and CDP-choline. The pathways to phosphatidylcholine and phosphatidylethanolamine in various organisms are summarized in Figure 21–29. Although the role of lipid composition in membrane function is not entirely understood, changes in compo- sition can produce dramatic effects. Researchers have isolated fruit flies with mutations in the gene that en- codes ethanolamine kinase (analogous to choline kinase; Fig. 21–28). Lack of this enzyme eliminates one path- way for phosphatidylethanolamine synthesis, thereby reducing the amount of this lipid in cellular membranes. Flies with this mutation—those with the genotype eas- ily shocked—exhibit transient paralysis following elec- trical stimulation or mechanical shock that would not affect wild-type flies. Plasmalogen Synthesis Requires Formation of an Ether-Linked Fatty Alcohol The biosynthetic pathway to ether lipids, including plasmalogens and the platelet-activating factor (see Fig. 10–9), involves the displacement of an esteri- fied fatty acyl group by a long-chain alcohol to form the ether linkage (Fig. 21–30). Head-group attachment fol- lows, by mechanisms essentially like those used in for- mation of the common ester-linked phospholipids. Fi- nally, the characteristic double bond of plasmalogens (shaded blue in Fig. 21–30) is introduced by the action of a mixed-function oxidase similar to that responsible for desaturation of fatty acids (Fig. 21–13). The perox- isome is the primary site of plasmalogen synthesis. Sphingolipid and Glycerophospholipid Synthesis Share Precursors and Some Mechanisms The biosynthesis of sphingolipids takes place in four stages: (1) synthesis of the 18-carbon amine sphinga- nine from palmitoyl-CoA and serine; (2) attachment of a fatty acid in amide linkage to yield N-acylsphinga- nine; (3) desaturation of the sphinganine moiety to form N-acylsphingosine (ceramide); and (4) attach- ment of a head group to produce a sphingolipid such as a cerebroside or sphingomyelin (Fig. 21–31). The pathway shares several features with the pathways leading to glycerophospholipids: NADPH provides re- ducing power, and fatty acids enter as their activated CoA derivatives. In cerebroside formation, sugars enter as their activated nucleotide derivatives. Head-group attachment in sphingolipid synthesis has several novel aspects. Phosphatidylcholine, rather than CDP-choline, serves as the donor of phosphocholine in the synthesis of sphingomyelin. In glycolipids, the cerebrosides and gangliosides (see Fig. 10–12), the head-group sugar is attached di- rectly to the C-1 hydroxyl of sphingosine in glycosidic linkage rather than through a phosphodiester bond. The sugar donor is a UDP-sugar (UDP-glucose or UDP- galactose). 21.3 Biosynthesis of Membrane Phospholipids 813 Choline 3 adoHcy Mammals 3 adoMet Serine Ethanolamine Phosphatidyl- serine Bacteria and yeast CDP-diacyl- glycerol Serine CMP Phosphatidyl- choline CDP- ethanolamine Ethanolamine CMP CMP CDP-choline CO 2 decarboxylation Diacylglycerol Phosphatidyl- ethanolamine FIGURE 21–29 Summary of the pathways to phosphatidylcholine and phosphatidylethanolamine. Conversion of phosphatidylethanolamine to phosphatidylcholine in mammals takes place only in the liver. 8885d_c21_813 2/27/04 12:41 PM Page 813 mac116 mac116: H11001 O CH 2 OH C Dihydroxyacetone phosphate O H11002 PO O H11002 O OC O C O S-CoA CoA-SH R 2 COO H11002 Fatty acyl–CoA long-chain alcohol R 1 CH 2 CH 2 R 1 O PO O H11002 O CH 2 C R 1 CO O H11002 CH 2 fatty acyl group 1-Acyldihydroxyacetone 3-phosphate CH 2 OCH 2 R 2 O S-CoA CH 2 Saturated fatty alcohol OHCH 2 R 2 C PO O H11002 O CH 2 O H11002 R 2 PO O H11002 O CH 2 CH 2 O H11002 OCH 2 CH 2 CHOH NAD H11001 mixed-function oxidase 1-Alkylglycerol 3-phosphate A plasmalogen CR 3 R 2 O CH 2 O O CH CH 2 CH 2 CH 2 O PO H11002 O O H11002 Ethanolamine head- group attachment C CH 2 R 3 O R 2 O O CH O PO O O 2 CH O H11002 CH 2 H11001 CH 2 CH 2 CH NADP H11001 R 3 C O S-CoA CoA-SH NH 3 2H 2 O NADH H11001 H H11001 C CH 2 R 3 O R 2 O O CH 2 O PO O CH O H11002 CH 2 H11001 CH 2 CH 2 CH 2 NH 3 H11001 H H11001 NADPH CH 2 1-alkyldihydroxy- acetone 3-phosphate synthase 1-alkyldihydroxy- acetone 3-phosphate reductase 1-alkylglycerol 3-phosphate acyl transferase 1-Alkyl-2-acylglycerol 3-phosphate 1-Alkyldihydroxyacetone 3-phosphate 2NADP H11001 2 CoA-SH H11001 2H H11001 NADPH Polar Lipids Are Targeted to Specific Cellular Membranes After synthesis on the smooth ER, the polar lipids, in- cluding the glycerophospholipids, sphingolipids, and glycolipids, are inserted into specific cellular mem- branes in specific proportions, by mechanisms not yet understood. Membrane lipids are insoluble in water, so they cannot simply diffuse from their point of synthesis (the ER) to their point of insertion. Instead, they are delivered in membrane vesicles that bud from the Golgi complex then move to and fuse with the target mem- brane (see Fig. 11–23). Cytosolic proteins also bind phospholipids and sterols and transport them between cellular membranes. These mechanisms contribute to the establishment of the characteristic lipid composi- tions of organelle membranes (see Fig. 11–2). Chapter 21 Lipid Biosynthesis814 FIGURE 21–30 Synthesis of ether lipids and plasmalogens. The newly formed ether linkage is shaded pink. The intermediate 1-alkyl-2-acyl- glycerol 3-phosphate is the ether analog of phosphatidic acid. Mechanisms for attaching head groups to ether lipids are essentially the same as for their ester-linked analogs. The characteristic double bond of plasmalogens (shaded blue) is introduced in a final step by a mixed-function oxidase system similar to that shown in Figure 21–13. 8885d_c21_787-832 2/26/04 9:35 AM Page 814 mac76 mac76:385_reb: HO O H11002 H CH 3 C O O C CH 2 CoA-S (CH 2 ) 14 Palmitoyl-CoA Serine CoA-SH, CO 2 H9252-Ketosphinganine NADP H11001 Sphinganine Fatty acyl–CoA CoA-SH N-acylsphinganine O CH 2 C CH 2 OH C CH HO C OH N(CH 3 ) 3 H 2 3 1 NADPH H11001 H H11001 HO H 3 NH H11001 C CH 2 CH OH NH C O R mixed-function oxidase (animals) CHCH CH C NH R UDP-Glc UDP CH 2 HO H OH NH C O CHCH CH CR CH 2 CerebrosideCeramide, containing sphingosine O Glc Phosphatidylcholine Diacylglycerol head- group attachment Sphingomyelin CH 2 HO HNH C O CHCH CH CR O P O O H11001 CH 2 CH 3 (CH 2 ) 12 HH 3 N H11001 CH 3 (CH 2 ) 14 CH 3 (CH 2 ) 14 CH 3 (CH 2 ) 14 CH 3 (CH 2 ) 12 CH 3 (CH 2 ) 12 SUMMARY 21.3 Biosynthesis of Membrane Phospholipids ■ Diacylglycerols are the principal precursors of glycerophospholipids. ■ In bacteria, phosphatidylserine is formed by the condensation of serine with CDP-diacylglycerol; decarboxylation of phosphatidylserine produces phosphatidylethanolamine. Phosphatidylglycerol is formed by condensation of CDP-diacylglycerol with glycerol 3-phosphate, followed by removal of the phosphate in monoester linkage. ■ Yeast pathways for the synthesis of phosphatidylserine, phosphatidylethanolamine, and phosphatidylglycerol are similar to those in bacteria; phosphatidylcholine is formed by methylation of phosphatidylethanolamine. ■ Mammalian cells have some pathways similar to those in bacteria, but somewhat different routes for synthesizing phosphatidylcholine and phosphatidylethanolamine. The head-group alcohol (choline or ethanolamine) is activated as the CDP derivative, then condensed with diacylglycerol. Phosphatidylserine is derived only from phosphatidylethanolamine. ■ Synthesis of plasmalogens involves formation of their characteristic double bond by a mixed-function oxidase. The head groups of sphingolipids are attached by unique mechanisms. ■ Phospholipids travel to their intracellular destinations via transport vesicles or specific proteins. 21.3 Biosynthesis of Membrane Phospholipids 815 FIGURE 21–31 Biosynthesis of sphingolipids. Condensation of palmitoyl-CoA and serine followed by reduction with NADPH yields sphinganine, which is then acylated to N-acylsphinganine (a ceramide). In animals, a double bond (shaded pink) is created by a mixed-function oxidase, before the final addition of a head group: phosphatidylcholine, to form sphingomyelin; glucose, to form a cerebroside. 8885d_c21_787-832 2/26/04 9:35 AM Page 815 mac76 mac76:385_reb: 21.4 Biosynthesis of Cholesterol, Steroids, and Isoprenoids Cholesterol is doubtless the most publicized lipid, no- torious because of the strong correlation between high levels of cholesterol in the blood and the incidence of human cardiovascular diseases. Less well advertised is cholesterol’s crucial role as a component of cellular membranes and as a precursor of steroid hormones and bile acids. Cholesterol is an essential molecule in many animals, including humans, but is not required in the mammalian diet—all cells can synthesize it from simple precursors. The structure of this 27-carbon compound suggests a complex biosynthetic pathway, but all of its carbon atoms are provided by a single precursor—acetate (Fig. 21–32). The isoprene units that are the essential in- termediates in the pathway from acetate to cholesterol are also precursors to many other natural lipids, and the mechanisms by which isoprene units are polymerized are similar in all these pathways. We begin with an account of the main steps in the biosynthesis of cholesterol from acetate, then discuss the transport of cholesterol in the blood, its uptake by cells, the normal regulation of cholesterol synthesis, and its regulation in those with defects in cholesterol uptake or transport. We next consider other cellular compo- nents derived from cholesterol, such as bile acids and steroid hormones. Finally, an outline of the biosynthetic pathways to some of the many compounds derived from isoprene units, which share early steps with the path- way to cholesterol, illustrates the extraordinary versa- tility of isoprenoid condensations in biosynthesis. CCH 2 CHCH 2 CH 3 Isoprene Cholesterol Is Made from Acetyl-CoA in Four Stages Cholesterol, like long-chain fatty acids, is made from acetyl-CoA, but the assembly plan is quite different. In early experiments, animals were fed acetate labeled with 14 C in either the methyl carbon or the carboxyl car- bon. The pattern of labeling in the cholesterol isolated from the two groups of animals (Fig. 21–32) provided the blueprint for working out the enzymatic steps in cho- lesterol biosynthesis. Synthesis takes place in four stages, as shown in Figure 21–33: 1 condensation of three acetate units to form a six-carbon intermediate, mevalonate; 2 con- version of mevalonate to activated isoprene units; 3 polymerization of six 5-carbon isoprene units to form the 30-carbon linear squalene; and 4 cyclization of squalene to form the four rings of the steroid nucleus, with a further series of changes (oxidations, removal or migration of methyl groups) to produce cholesterol. Chapter 21 Lipid Biosynthesis816 COO H11002 CH 3 3 10 Acetate AB C D HO Cholesterol C C C C CC C C C C C C C C C C CC C C CC C C C C C 9 2 4 5 6 7 1 25 19 11 8 20 21 12 18 13 14 15 16 17 22 24 26 23 27 FIGURE 21–32 Origin of the carbon atoms of cholesterol. This can be deduced from tracer experiments with acetate labeled in the methyl carbon (black) or the carboxyl carbon (red). The individual rings in the fused-ring system are designated A through D. FIGURE 21–33 Summary of cholesterol biosynthesis. The four stages are discussed in the text. Isoprene units in squalene are set off by red dashed lines. COO H5008 3 CH 3 C CH 3 H5008 OOC Acetate O H5008 CH 2 CH 2 HO Mevalonate C CH 2 CH 2 OH CH 3 4 1 CH 2 2 CH 2 O PO P O O H5008 O O H5008 isoprene Activated isoprene 2 3 Squalene Cholesterol OH 8885d_c21_816 2/27/04 12:41 PM Page 816 mac116 mac116: FIGURE 21–34 Formation of mevalonate from acetyl-CoA. The ori- gin of C-1 and C-2 of mevalonate from acetyl-CoA is shown in pink. CH 2 OH 2NADP H11001 2 CH 3 C O CH 3 C O S-CoA Acetyl-CoA thiolase CoA-SH Acetoacetyl-CoA HMG-CoA reductase CH 3 C O CH 2 C O S-CoA CoA-SH CH 2 C O S-CoA CH 2 CH 3 S-CoA H9252-Hydroxy-H9252-methylglutaryl-CoA (HMG-CoA) COO H5008 CoA-SH H11001 2H H11001 CH 2 C CH 2 C COO H5008 OH OHCH 3 Mevalonate HMG-CoA synthase 3 1 5 2 4 NADPH2 Stage 1 Synthesis of Mevalonate from Acetate The first stage in cholesterol biosynthesis leads to the interme- diate mevalonate (Fig. 21–34). Two molecules of acetyl-CoA condense to form acetoacetyl-CoA, which condenses with a third molecule of acetyl-CoA to yield the six-carbon compound H9252-hydroxy-H9252-methylglu- taryl-CoA (HMG-CoA). These first two reactions are catalyzed by thiolase and HMG-CoA synthase, re- spectively. The cytosolic HMG-CoA synthase in this pathway is distinct from the mitochondrial isozyme that catalyzes HMG-CoA synthesis in ketone body formation (see Fig. 17–18). The third reaction is the committed and rate-limiting step: reduction of HMG-CoA to mevalonate, for which each of two molecules of NADPH donates two electrons. HMG-CoA reductase, an integral membrane protein of the smooth ER, is the major point of regulation on the pathway to cholesterol, as we shall see. Stage 2 Conversion of Mevalonate to Two Activated Isoprenes In the next stage of cholesterol synthesis, three phos- phate groups are transferred from three ATP molecules to mevalonate (Fig. 21–35). The phosphate attached to the C-3 hydroxyl group of mevalonate in the interme- diate 3-phospho-5-pyrophosphomevalonate is a good 21.4 Biosynthesis of Cholesterol, Steroids, and Isoprenoids 817 H5008 OOC CH 3 C O 3-Phospho-5- pyrophosphomevalonate H9004 3 -Isopentenyl pyrophosphate Dimethylallyl pyrophosphate Activated isoprenes CH 2 5 CH 3 CH 2 O H5008 OOC CH 3 C Mevalonate ADP ATP CH 2 CH 2 CH 2 OH OH H5008 OOC CH 3 C O 5-Phosphomevalonate ADP ATP CH 2 O H5008 CH 2 CH 2 OH O PO H5008 H5008 OOC CH 3 C O ADP ATP CH 2 O H5008 CH 2 CH 2 OH O P O H5008 O O H5008 O P O 5-Pyrophosphomevalonate O H5008 O PO H5008 O H5008 O P P O H5008 O O H5008 CO 2 , P i CH 3 C OCH 2 CH 2 CH 2 O O H5008 O PP O H5008 O O H5008 CH 3 C OCH 2 CH 2 O O H5008 O PP O H5008 O O H5008 CH 3142 mevalonate 5-phosphotransferase phosphomevalonate kinase pyrophospho- mevalonate decarboxylase pyrophospho- mevalonate decarboxylase FIGURE 21–35 Conversion of mevalonate to activated isoprene units. Six of these activated units combine to form squalene (see Fig. 21–36). The leaving groups of 3-phospho-5-pyrophosphomevalonate are shaded pink. The bracketed intermediate is hypothetical. 8885d_c21_817 2/27/04 1:45 PM Page 817 Mac113 mac113: leaving group; in the next step, both this phosphate and the nearby carboxyl group leave, producing a double bond in the five-carbon product, H9004 3 -isopentenyl pyrophosphate. This is the first of the two activated isoprenes central to cholesterol formation. Isomerization of H9004 3 -isopentenyl pyrophosphate yields the second acti- vated isoprene, dimethylallyl pyrophosphate. Synthe- sis of isopentenyl pyrophosphate in the cytoplasm of plant cells follows the pathway described here. However, plant chloroplasts and many bacteria use a mevalonate- independent pathway. This alternative pathway does not occur in animals, so it is an attractive target for the development of new antibiotics. Stage 3 Condensation of Six Activated Isoprene Units to Form Squalene Isopentenyl pyrophosphate and dimethylallyl pyrophosphate now undergo a head-to-tail condensa- tion, in which one pyrophosphate group is displaced and a 10-carbon chain, geranyl pyrophosphate, is formed (Fig. 21–36). (The “head” is the end to which pyrophos- phate is joined.) Geranyl pyrophosphate undergoes an- other head-to-tail condensation with isopentenyl pyro- Chapter 21 Lipid Biosynthesis818 H9004 3 -Isopentenyl pyrophosphate H5008 O O O H5008 O PP O H5008 O O H11001 Dimethylallyl pyrophosphate O O O H5008 O PP O H5008 O O H5008 prenyl transferase (head-to-tail) O O O H5008 O PP O H5008 O O H5008 Farnesyl pyrophosphate O O O H5008 O PP O H5008 O O H5008 PP i prenyl transferase (head-to-tail condensation) Geranyl pyrophosphateO O O H5008 O PP O H5008 O O H5008 H9004 3 -Isopentenyl pyrophosphate O O O H5008 O PP O H5008 O O H5008 Farnesyl pyrophosphate NADPH H11001 H H11001 NADP H11001 2 PP i squalene synthase (head-to-head) Squalene PP i FIGURE 21–36 Formation of squalene. This 30-carbon structure arises through successive condensations of activated isoprene (five-carbon) units. 8885d_c21_787-832 2/26/04 9:35 AM Page 818 mac76 mac76:385_reb: phosphate, yielding the 15-carbon intermediate farne- syl pyrophosphate. Finally, two molecules of farnesyl pyrophosphate join head to head, with the elimination of both pyrophosphate groups, to form squalene. The common names of these intermediates derive from the sources from which they were first isolated. Geraniol, a component of rose oil, has the aroma of gera- niums, and farnesol is an aromatic compound found in the flowers of the Farnese acacia tree. Many natural scents of plant origin are synthesized from isoprene units. Squalene, first isolated from the liver of sharks (genus Squalus), has 30 carbons, 24 in the main chain and 6 in the form of methyl group branches. Stage 4 Conversion of Squalene to the Four-Ring Steroid Nu- cleus When the squalene molecule is represented as in Figure 21–37, the relationship of its linear structure to the cyclic structure of the sterols becomes apparent. All 21.4 Biosynthesis of Cholesterol, Steroids, and Isoprenoids 819 Squalene Cholesterol squalene monooxygenase O 2 C 2 H 5 NADP H11001 Squalene 2,3-epoxide multistep (plants) cyclase Stigmasterol O cyclase (animals) Ergosterol multistep (fungi) Lanosterol multistep NADPH H11001 H H11001 HO H 2 O HO HO HO HO 3 2 H11001 FIGURE 21–37 Ring closure converts linear squalene to the condensed steroid nucleus. The first step in this sequence is catalyzed by a mixed-function oxidase (a monooxygenase), for which the co- substrate is NADPH. The product is an epoxide, which in the next step is cyclized to the steroid nucleus. The final product of these reactions in animal cells is cholesterol; in other organisms, slightly different sterols are produced, as shown. 8885d_c21_787-832 2/26/04 9:35 AM Page 819 mac76 mac76:385_reb: sterols have the four fused rings that form the steroid nucleus, and all are alcohols, with a hydroxyl group at C-3—thus the name “sterol.” The action of squalene monooxygenase adds one oxygen atom from O 2 to the end of the squalene chain, forming an epoxide. This enzyme is another mixed-function oxidase (Box 21–1); NADPH reduces the other oxygen atom of O 2 to H 2 O. The double bonds of the product, squalene 2,3-epoxide, are positioned so that a remarkable con- certed reaction can convert the linear squalene epox- ide to a cyclic structure. In animal cells, this cycliza- tion results in the formation of lanosterol, which contains the four rings characteristic of the steroid nucleus. Lanosterol is finally converted to cholesterol in a series of about 20 reactions that include the migration of some methyl groups and the removal of others. Elucidation of this extraordinary biosynthetic pathway, one of the most complex known, was ac- complished by Konrad Bloch, Feodor Lynen, John Cornforth, and George Popják in the late 1950s. Cholesterol is the sterol characteristic of animal cells; plants, fungi, and protists make other, closely re- lated sterols instead. They use the same synthetic path- way as far as squalene 2,3-epoxide, at which point the pathways diverge slightly, yielding other sterols, such as stigmasterol in many plants and ergosterol in fungi (Fig. 21–37). Cholesterol Has Several Fates Much of the cholesterol synthesis in vertebrates takes place in the liver. A small fraction of the cholesterol made there is incorporated into the membranes of he- patocytes, but most of it is exported in one of three forms: biliary cholesterol, bile acids, or cholesteryl es- ters. Bile acids and their salts are relatively hydrophilic cholesterol derivatives that are synthesized in the liver and aid in lipid digestion (see Fig. 17–1). Cholesteryl esters are formed in the liver through the action of acyl-CoA–cholesterol acyl transferase (ACAT). This enzyme catalyzes the transfer of a fatty acid from coenzyme A to the hydroxyl group of cholesterol (Fig. 21–38), converting the cholesterol to a more hy- drophobic form. Cholesteryl esters are transported in secreted lipoprotein particles to other tissues that use cholesterol, or they are stored in the liver. All growing animal tissues need cholesterol for membrane synthesis, and some organs (adrenal gland and gonads, for example) use cholesterol as a precur- sor for steroid hormone production (discussed below). Cholesterol is also a precursor of vitamin D (see Fig. 10–20a). Cholesterol and Other Lipids Are Carried on Plasma Lipoproteins Cholesterol and cholesteryl esters, like triacylglycerols and phospholipids, are essentially insoluble in water, yet must be moved from the tissue of origin to the tis- sues in which they will be stored or consumed. They are carried in the blood plasma as plasma lipoproteins, Chapter 21 Lipid Biosynthesis820 Konrad Bloch, Feodor Lynen, 1912–2000 1911–1979 John Cornforth George Popják Cholesteryl ester O Fatty acyl–CoA CoA-SH acyl-CoA–cholesterol acyl transferase (ACAT) Cholesterol HO O CR FIGURE 21–38 Synthesis of cholesteryl esters. Esterification con- verts cholesterol to an even more hydrophobic form for storage and transport. 8885d_c21_787-832 2/26/04 9:35 AM Page 820 mac76 mac76:385_reb: macromolecular complexes of specific carrier proteins, apolipoproteins, with various combinations of phos- pholipids, cholesterol, cholesteryl esters, and triacyl- glycerols. Apolipoproteins (“apo” designates the protein in its lipid-free form) combine with lipids to form several classes of lipoprotein particles, spherical complexes with hydrophobic lipids in the core and hydrophilic amino acid side chains at the surface (Fig. 21–39a). Dif- ferent combinations of lipids and proteins produce par- ticles of different densities, ranging from chylomicrons to high-density lipoproteins. These particles can be sep- arated by ultracentrifugation (Table 21–2) and visual- ized by electron microscopy (Fig. 21–39b). Each class of lipoprotein has a specific function, de- termined by its point of synthesis, lipid composition, and apolipoprotein content. At least nine different apolipo- proteins are found in the lipoproteins of human plasma (Table 21–3), distinguishable by their size, their reac- tions with specific antibodies, and their characteristic distribution in the lipoprotein classes. These protein components act as signals, targeting lipoproteins to spe- cific tissues or activating enzymes that act on the lipoproteins. Chylomicrons, discussed in Chapter 17 in con- nection with the movement of dietary triacylglycerols from the intestine to other tissues, are the largest of the lipoproteins and the least dense, containing a high 21.4 Biosynthesis of Cholesterol, Steroids, and Isoprenoids 821 Phospholipid monolayer Triacylglycerols Cholesteryl esters Free (unesterified) cholesterol ApoB-100 (a) Chylomicrons (H1100360,000) VLDL (H11003180,000) LDL (H11003180,000) HDL (H11003180,000) (b) FIGURE 21–39 Lipoproteins. (a) Structure of a low-density lipopro- tein (LDL). Apolipoprotein B-100 (apoB-100) is one of the largest sin- gle polypeptide chains known, with 4,636 amino acid residues (M r 513,000). (b) Four classes of lipoproteins, visualized in the electron microscope after negative staining. Clockwise from top left: chylomi- crons, 50 to 200 nm in diameter; VLDL, 28 to 70 nm; HDL, 8 to 11 nm; and LDL, 20 to 25 nm. For properties of lipoproteins, see Table 21–2. TABLE 21–2 Major Classes of Human Plasma Lipoproteins: Some Properties Composition (wt %) Lipoprotein Density (g/mL) Protein Phospholipids Free cholesterol Cholesteryl esters Triacylglycerols Chylomicrons H110211.006 2 9 1 3 85 VLDL 0.95–1.006 10 18 7 12 50 LDL 1.006–1.063 23 20 8 37 10 HDL 1.063–1.210 55 24 2 15 4 Source: Modified from Kritchevsky, D. (1986) Atherosclerosis and nutrition. Nutr. Int. 2, 290–297. 8885d_c21_787-832 2/26/04 9:35 AM Page 821 mac76 mac76:385_reb: proportion of triacylglycerols (see Fig. 17–2). Chylomi- crons are synthesized in the ER of epithelial cells that line the small intestine, then move through the lymphatic system and enter the bloodstream via the left subcla- vian vein. The apolipoproteins of chylomicrons include apoB-48 (unique to this class of lipoproteins), apoE, and apoC-II (Table 21–3). ApoC-II activates lipoprotein lipase in the capillaries of adipose, heart, skeletal muscle, and lactating mammary tissues, allowing the release of free fatty acids to these tissues. Chylomicrons thus carry di- etary fatty acids to tissues where they will be consumed or stored as fuel (Fig. 21–40). The remnants of chylo- microns (depleted of most of their triacylglycerols but still containing cholesterol, apoE, and apoB-48) move through the bloodstream to the liver. Receptors in the liver bind to the apoE in the chylomicron remnants and mediate their uptake by endocytosis. In the liver, the remnants release their cholesterol and are degraded in lysosomes. When the diet contains more fatty acids than are needed immediately as fuel, they are converted to tria- cylglycerols in the liver and packaged with specific apolipoproteins into very-low-density lipoprotein (VLDL). Excess carbohydrate in the diet can also be converted to triacylglycerols in the liver and exported as VLDLs (Fig. 21–40a). In addition to triacylglycerols, VLDLs contain some cholesterol and cholesteryl esters, as well as apoB-100, apoC-I, apoC-II, apoC-III, and apo- E (Table 21–3). These lipoproteins are transported in the blood from the liver to muscle and adipose tissue, where activation of lipoprotein lipase by apoC-II causes the release of free fatty acids from the VLDL triacyl- glycerols. Adipocytes take up these fatty acids, recon- vert them to triacylglycerols, and store the products in intracellular lipid droplets; myocytes, in contrast, pri- marily oxidize the fatty acids to supply energy. Most VLDL remnants are removed from the circulation by hepatocytes. The uptake, like that for chylomicrons, is Chapter 21 Lipid Biosynthesis822 Intestine Chylomicrons Free fatty acids Capillary lipoprotein lipase Mammary, muscle, or adipose tissue Blood plasma after fast (a) (b) HDL precursors (from liver and intestine) Extrahepatic tissues Reverse cholesterol transport HDL LDL Liver VLDL Chylomicron remnants VLDL remnants (IDL) Blood plasma after meal FIGURE 21–40 Lipoproteins and lipid transport. (a) Lipids are transported in the bloodstream as lipoproteins, which exist as sev- eral variants that have different functions, different protein and lipid compositions (see Tables 21–2, 21–3), and thus different densities. Dietary lipids are packaged into chylomicrons; much of their tria- cylglycerol content is released by lipoprotein lipase to adipose and muscle tissues during transport through capillaries. Chylomicron remnants (containing largely protein and cholesterol) are taken up by the liver. Endogenous lipids and cholesterol from the liver are delivered to adipose and muscle tissue by VLDL. Extraction of lipid from VLDL (along with loss of some apolipoproteins) gradually con- verts some of it to LDL, which delivers cholesterol to extrahepatic tissues or returns to the liver. The liver takes up LDL, VLDL rem- nants, and chylomicron remnants by receptor-mediated endocyto- sis. Excess cholesterol in extrahepatic tissues is transported back to the liver as HDL. In the liver, some cholesterol is converted to bile salts. (b) Blood plasma samples collected after a fast (left) and after a high-fat meal (right). Chylomicrons produced after a fatty meal give the plasma a milky appearance. 8885d_c21_787-832 2/26/04 9:35 AM Page 822 mac76 mac76:385_reb: receptor-mediated and depends on the presence of apoE in the VLDL remnants (Box 21–3 describes a link between apoE and Alzheimer’s disease). The loss of triacylglycerol converts some VLDL to VLDL remnants (also called intermediate density lipo- protein, IDL); further removal of triacylglycerol from VLDL produces low-density lipoprotein (LDL) (Table 21–2). Very rich in cholesterol and cholesteryl esters and containing apoB-100 as their major apoli- poprotein, LDLs carry cholesterol to extrahepatic tis- sues that have specific plasma membrane receptors that recognize apoB-100. These receptors mediate the up- take of cholesterol and cholesteryl esters in a process described below. The fourth major lipoprotein type, high-density lipoprotein (HDL), originates in the liver and small intestine as small, protein-rich particles that contain rel- atively little cholesterol and no cholesteryl esters (Fig. 21–40). HDLs contain apoA-I, apoC-I, apoC-II, and other apolipoproteins (Table 21–3), as well as the enzyme lecithin-cholesterol acyl transferase (LCAT), which catalyzes the formation of cholesteryl esters from lecithin (phosphatidylcholine) and cholesterol (Fig. 21–41). LCAT on the surface of nascent (newly form- ing) HDL particles converts the cholesterol and phos- phatidylcholine of chylomicron and VLDL remnants to cholesteryl esters, which begin to form a core, trans- forming the disk-shaped nascent HDL to a mature, spherical HDL particle. This cholesterol-rich lipoprotein then returns to the liver, where the cholesterol is un- loaded; some of this cholesterol is converted to bile salts. 21.4 Biosynthesis of Cholesterol, Steroids, and Isoprenoids 823 TABLE 21–3 Apolipoproteins of the Human Plasma Lipoproteins Apolipoprotein Molecular weight Lipoprotein association Function (if known) ApoA-I 28,331 HDL Activates LCAT; interacts with ABC transporter ApoA-II 17,380 HDL ApoA-IV 44,000 Chylomicrons, HDL ApoB-48 240,000 Chylomicrons ApoB-100 513,000 VLDL, LDL Binds to LDL receptor ApoC-I 7,000 VLDL, HDL ApoC-II 8,837 Chylomicrons, VLDL, HDL Activates lipoprotein lipase ApoC-III 8,751 Chylomicrons, VLDL, HDL Inhibits lipoprotein lipase ApoD 32,500 HDL ApoE 34,145 Chylomicrons, VLDL, HDL Triggers clearance of VLDL and chylomicron remnants Source: Modified from Vance, D.E. & Vance, J.E. (eds) (1985) Biochemistry of Lipids and Membranes. The Benjamin/Cummings Publishing Company, Menlo Park, CA. Cholesteryl ester lecithin-cholesterol acyl transferase (LCAT) Cholesterol CH 2 P CH O O CH 2 O H5008 R 1 O O C O CH 2 CH 2 N(CH 3 ) 3 H11001 H11001 Phosphatidylcholine (lecithin) HO CH 2 P R 2 CH O O CH 2 O H5008 C R 1 O O C O O O CH 2 CH 2 N(CH 3 ) 3 H11001 H11001 OH Lysolecithin O O CR 2 FIGURE 21–41 Reaction catalyzed by lecithin-cholesterol acyl trans- ferase (LCAT). This enzyme is present on the surface of HDL and is stimulated by the HDL component apoA-I. Cholesteryl esters accu- mulate within nascent HDLs, converting them to mature HDLs. 8885d_c21_787-832 2/26/04 9:35 AM Page 823 mac76 mac76:385_reb: HDL may be taken up in the liver by receptor- mediated endocytosis, but at least some of the choles- terol in HDL is delivered to other tissues by a novel mechanism. HDL can bind to plasma membrane recep- tor proteins called SR-BI in hepatic and steroidogenic tissues such as the adrenal gland. These receptors me- diate not endocytosis but a partial and selective trans- fer of cholesterol and other lipids in HDL into the cell. Depleted HDL then dissociates to recirculate in the bloodstream and extract more lipids from chylomicron and VLDL remnants. Depleted HDL can also pick up cholesterol stored in extrahepatic tissues and carry it to the liver, in reverse cholesterol transport pathways (Fig. 21–40). In one reverse transport path, interaction of nascent HDL with SR-BI receptors in cholesterol-rich cells triggers passive movement of cholesterol from the cell surface into HDL, which then carries it back to the liver. In a second pathway, apoA-I in depleted HDL in- teracts with an active transporter, the ABC1 protein, in a cholesterol-rich cell. The apoA-I (and presumably the HDL) is taken up by endocytosis, then resecreted with a load of cholesterol, which it transports to the liver. The ABC1 protein is a member of a large family of multidrug transporters, sometimes called ABC trans- porters because they all have ATP-binding cassettes; they also have two transmembrane domains with six transmembrane helices (Chapter 11). These proteins actively transport a variety of ions, amino acids, vita- mins, steroid hormones, and bile salts across plasma membranes. The CFTR protein that is defective in cys- tic fibrosis (see Box 11–3) is another member of this ABC family of multidrug transporters. Cholesteryl Esters Enter Cells by Receptor-Mediated Endocytosis Each LDL particle in the bloodstream contains apoB- 100, which is recognized by specific surface receptor proteins, LDL receptors, on cells that need to take up cholesterol. The binding of LDL to an LDL receptor ini- tiates endocytosis, which conveys the LDL and its re- ceptor into the cell within an endosome (Fig. 21–42). The endosome eventually fuses with a lysosome, which contains enzymes that hydrolyze the cholesteryl esters, releasing cholesterol and fatty acid into the cytosol. The apoB-100 of LDL is also degraded to amino acids that are released to the cytosol, but the LDL receptor es- capes degradation and is returned to the cell surface, to function again in LDL uptake. ApoB-100 is also present in VLDL, but its receptor-binding domain is not avail- able for binding to the LDL receptor; conversion of VLDL to LDL exposes the receptor-binding domain of apoB-100. This pathway for the transport of cholesterol in blood and its receptor-mediated endocytosis by target tissues was elucidated by Michael Brown and Joseph Goldstein. Chapter 21 Lipid Biosynthesis824 BOX 21–3 BIOCHEMISTRY IN MEDICINE ApoE Alleles Predict Incidence of Alzheimer’s Disease In the human population there are three common vari- ants, or alleles, of the gene encoding apolipoprotein E. The most common, accounting for about 78% of human apoE alleles, is APOE3; alleles APOE4 and APOE2 account for 15% and 7%, respectively. The APOE4 allele is particularly common in humans with Alzheimer’s disease, and the link is highly predictive. Individuals who inherit APOE4 have an increased risk of late-onset Alzheimer’s disease. Those who are ho- mozygous for APOE4 have a 16-fold increased risk of developing the disease; for those who do, the mean age of onset is just under 70 years. For people who inherit two copies of APOE3, by contrast, the mean age of onset of Alzheimer’s disease exceeds 90 years. The molecular basis for the association between apoE4 and Alzheimer’s disease is not yet known. Spec- ulation has focused on a possible role for apoE in sta- bilizing the cytoskeletal structure of neurons. The apoE2 and apoE3 proteins bind to a number of pro- teins associated with neuronal microtubules, whereas apoE4 does not. This may accelerate the death of neu- rons. Whatever the mechanism proves to be, these ob- servations promise to expand our understanding of the biological functions of apolipoproteins. Michael Brown and Joseph Goldstein 8885d_c21_787-832 2/26/04 9:35 AM Page 824 mac76 mac76:385_reb: Cholesterol that enters cells by this path may be in- corporated into membranes or reesterified by ACAT (Fig. 21–38) for storage within cytosolic lipid droplets. Accumulation of excess intracellular cholesterol is pre- vented by reducing the rate of cholesterol synthesis when sufficient cholesterol is available from LDL in the blood. The LDL receptor also binds to apoE and plays a significant role in the hepatic uptake of chylomicrons and VLDL remnants. However, if LDL receptors are un- available (as, for example, in a mouse strain that lacks the gene for the LDL receptor), VLDL remnants and chylomicrons are still taken up by the liver even though LDL is not. This indicates the presence of a back-up sys- tem for receptor-mediated endocytosis of VLDL rem- nants and chylomicrons. One back-up receptor is lipoprotein receptor-related protein (LRP), which binds to apoE as well as to a number of other ligands. Cholesterol Biosynthesis Is Regulated at Several Levels Cholesterol synthesis is a complex and energy- expensive process, so it is clearly advantageous to an organism to regulate the biosynthesis of cholesterol to complement dietary intake. In mammals, cholesterol production is regulated by intracellular cholesterol con- centration and by the hormones glucagon and insulin. The rate-limiting step in the pathway to cholesterol (and a major site of regulation) is the conversion of HMG- CoA to mevalonate (Fig. 21–34), the reaction catalyzed by HMG-CoA reductase. Regulation in response to cholesterol levels is me- diated by an elegant system of transcriptional regulation of the gene encoding HMG-CoA reductase. This gene, along with more than 20 other genes encoding enzymes that mediate the uptake and synthesis of cholesterol and 21.4 Biosynthesis of Cholesterol, Steroids, and Isoprenoids 825 LDL particle ApoB-100 Cholesteryl ester LDL receptor Golgi complex LDL receptor synthesis Endoplasmic reticulum Nucleus Cholesterol Cholesteryl ester droplet Fatty acids Amino acids Lysosome Endosome receptor-mediated endocytosis FIGURE 21–42 Uptake of cholesterol by receptor-mediated endocytosis. 8885d_c21_787-832 2/26/04 9:35 AM Page 825 mac76 mac76:385_reb: unsaturated fatty acids, is controlled by a small family of proteins called sterol regulatory element-binding proteins (SREBPs). When newly synthesized, these pro- teins are embedded in the ER. Only the soluble amino- terminal domain of an SREBP functions as a transcrip- tional activator, using mechanisms discussed in Chapter 28. However, this domain has no access to the nucleus and cannot participate in gene activation while it re- mains part of the SREBP molecule. To activate tran- scription of the HMG-CoA reductase gene and other genes, the transcriptionally active domain is separated from the rest of the SREBP by proteolytic cleavage. When cholesterol levels are high, SREBPs are inactive, secured to the ER in a complex with another protein called SREBP cleavage-activating protein (SCAP) (Fig. 21–43). It is SCAP that binds cholesterol and a number of other sterols, thus acting as a sterol sensor. When sterol levels are high, the SCAP-SREBP complex prob- ably interacts with another protein that retains the en- tire complex in the ER. When the level of sterols in the cell declines, a conformational change in SCAP causes release of the SCAP-SREBP complex from the ER- retention activity, and the complex migrates within vesi- cles to the Golgi complex. In the Golgi complex, SREBP is cleaved twice by two different proteases, the second cleavage releasing the amino-terminal domain into the cytosol. This domain travels to the nucleus and activates transcription of its target genes. The amino-terminal domain of SREBP has a short half-life and is rapidly degraded by proteasomes (see Fig. 27–42). When sterol levels increase sufficiently, the proteolytic release of SREBP amino-terminal domains is again blocked, and proteasome degradation of the existing active domains results in a rapid shut-down of the gene targets. Several other mechanisms also regulate cholesterol synthesis (Fig. 21–44). Hormonal control is mediated by covalent modification of HMG-CoA reductase itself. The enzyme exists in phosphorylated (inactive) and dephosphorylated (active) forms. Glucagon stimulates phosphorylation (inactivation), and insulin promotes dephosphorylation, activating the enzyme and favoring cholesterol synthesis. High intracellular concentrations of cholesterol activate ACAT, which increases esterifi- cation of cholesterol for storage. Finally, a high cellular cholesterol level diminishes transcription of the gene that encodes the LDL receptor, reducing production of the receptor and thus the uptake of cholesterol from the blood. Chapter 21 Lipid Biosynthesis826 Cytosol Golgi complex Golgi complex SCAP migration to Golgi complex SREBP Endoplasmic reticulum Sterol (binds SCAP, prevents release of SREBP) Cleavage by first protease Cleavage by second protease Transcription of target genes is activated released domain of SREBP migrates to nucleus N C N N N C C C N N DNA Nucleus C FIGURE 21–43 SREBP activation. Sterol regulatory element-binding proteins (SREBPs, shown in green) are embedded in the ER when first synthesized, in a complex with the protein SREBP cleavage-activating protein (SCAP, red). (N and C represent the amino and carboxyl ter- mini of the proteins.) When bound to SCAP, SREBPs are inactive. When sterol levels decline, the complex migrates to the Golgi complex, and SREBP is cleaved by two different proteases in succession. The liber- ated amino-terminal domain of SREBP migrates to the nucleus, where it activates transcription of sterol-regulated genes. insulin Acetyl-CoA -Hydroxy- -methyl- glutaryl-CoA Mevalonate Cholesterol (intracellular) HMG-CoA reductase glucagon X receptor- mediated endocytosis ACAT Cholesteryl esters LDL-cholesterol (extracellular) H9252H9252 stimulates proteolysis of HMG-CoA reductase multistep multistep FIGURE 21–44 Regulation of cholesterol formation balances syn- thesis with dietary uptake. Glucagon promotes phosphorylation (in- activation) of HMG-CoA reductase; insulin promotes dephosphoryla- tion (activation). X represents unidentified metabolites of cholesterol that stimulate proteolysis of HMG-CoA reductase. 8885d_c21_826 2/27/04 1:45 PM Page 826 Mac113 mac113: FIGURE 21–46 Some steroid hormones derived from cholesterol. The structures of some of these compounds are shown in Figure 10–19. Cholesterol Corticosterone (mineralocorticoid) Affects protein and carbohydrate metabolism; suppresses immune response, inflammation, and allergic responses. Pregnenolone Progesterone Aldosterone (mineralocorticoid) Testosterone Cortisol (glucocorticoid) Estradiol Male and female sex hormones. Influence secondary sexual char- acteristics; regulate female reproductive cycle. Regulate reabsorption of Na H11001 , Cl H5008 , HCO H5008 in the kidney. 3 Unregulated cholesterol production can lead to serious human disease. When the sum of choles- terol synthesized and cholesterol obtained in the diet exceeds the amount required for the synthesis of mem- branes, bile salts, and steroids, pathological accumula- tions of cholesterol in blood vessels (atherosclerotic plaques) can develop, resulting in obstruction of blood vessels (atherosclerosis). Heart failure due to oc- cluded coronary arteries is a leading cause of death in industrialized societies. Atherosclerosis is linked to high levels of cholesterol in the blood, and particularly to high levels of LDL-bound cholesterol; there is a negative cor- relation between HDL levels and arterial disease. In familial hypercholesterolemia, a human genetic disorder, blood levels of cholesterol are extremely high and severe atherosclerosis develops in childhood. These individuals have a defective LDL receptor and lack receptor-mediated uptake of cholesterol carried by LDL. Consequently, cholesterol is not cleared from the blood; it accumulates and contributes to the formation of ath- erosclerotic plaques. Endogenous cholesterol synthesis continues despite the excessive cholesterol in the blood, because extracellular cholesterol cannot enter the cell to regulate intracellular synthesis (Fig. 21–44). Two products derived from fungi, lovastatin and com- pactin, are used to treat patients with familial hyper- cholesterolemia. Both these compounds, and several synthetic analogs, resemble mevalonate (Fig. 21–45) and are competitive inhibitors of HMG-CoA reductase, thus inhibiting cholesterol synthesis. Lovastatin treat- ment lowers serum cholesterol by as much as 30% in individuals having one defective copy of the gene for the LDL receptor. When combined with an edible resin that binds bile acids and prevents their reabsorption from the intestine, the drug is even more effective. In familial HDL deficiency, HDL levels are very low; they are almost undetectable in Tangier disease. Both genetic disorders are the result of mutations in the ABC1 protein. Cholesterol-depleted HDL cannot take up cholesterol from cells that lack ABC1 protein, and cholesterol-poor HDL is rapidly removed from the blood and destroyed. Both familial HDL deficiency and Tang- ier disease are very rare (worldwide, fewer than 100 families with Tangier disease are known), but the exis- tence of these diseases establishes a role for ABC1 pro- tein in the regulation of plasma HDL levels. Because low plasma HDL levels correlate with a high incidence of coronary artery disease, the ABC1 protein may prove a useful target for drugs to control HDL levels. ■ Steroid Hormones Are Formed by Side-Chain Cleavage and Oxidation of Cholesterol Humans derive all their steroid hormones from choles- terol (Fig. 21–46). Two classes of steroid hormones are synthesized in the cortex of the adrenal gland: mineralocorticoids, which control the reabsorption of inorganic ions (Na H11001 , Cl H11002 , and HCO 3 H11002 ) by the kidney, and glucocorticoids, which help regulate gluconeogene- sis and reduce the inflamma- tory response. Sex hormones are produced in male and fe- male gonads and the pla- centa. They include proges- terone, which regulates the female reproductive cycle, and androgens (such as testosterone) and estrogens (such as estradiol), which influence the development of 21.4 Biosynthesis of Cholesterol, Steroids, and Isoprenoids 827 FIGURE 21–45 Inhibitors of HMG-CoA reductase. A comparison of the structures of mevalonate and four pharmaceutical compounds that inhibit HMG-CoA reductase. Mevalonate CH 3 CH 3 R 2 R 1 H11005 H R 1 H11005 CH 3 R 1 H11005 H R 1 H11005 H R 2 H11005 H R 2 H11005 CH 3 R 2 H11005 OH R 2 H11005 CH 3 Compactin Simvastatin (Zocor) Pravastatin (Pravachol) Lovastatin (Mevacor) R 1 COO H5008 OH HO H 3 C COO H5008 OH O O HO C O CH 3 Progesterone O 8885d_c21_787-832 2/26/04 9:35 AM Page 827 mac76 mac76:385_reb: secondary sexual characteristics in males and females, respectively. Steroid hormones are effective at very low concentrations and are therefore synthesized in rela- tively small quantities. In comparison with the bile salts, their production consumes relatively little cholesterol. Synthesis of steroid hormones requires removal of some or all of the carbons in the “side chain” on C-17 of the D ring of cholesterol. Side-chain removal takes place in the mitochondria of steroidogenic tissues. Re- moval involves the hydroxylation of two adjacent car- bons in the side chain (C-20 and C-22) followed by cleavage of the bond between them (Fig. 21–47). For- mation of the various hormones also involves the intro- duction of oxygen atoms. All the hydroxylation and oxy- genation reactions in steroid biosynthesis are catalyzed by mixed-function oxidases (Box 21–1) that use NADPH, O 2 , and mitochondrial cytochrome P-450. Intermediates in Cholesterol Biosynthesis Have Many Alternative Fates In addition to its role as an intermediate in cholesterol biosynthesis, isopentenyl pyrophosphate is the acti- vated precursor of a huge array of biomolecules with di- verse biological roles (Fig. 21–48). They include vita- mins A, E, and K; plant pigments such as carotene and the phytol chain of chlorophyll; natural rubber; many essential oils (such as the fragrant principles of lemon oil, eucalyptus, and musk); insect juvenile hormone, which controls metamorphosis; dolichols, which serve as lipid-soluble carriers in complex polysaccharide synthesis; and ubiquinone and plastoquinone, electron carriers in mitochondria and chloroplasts. Collectively, these molecules are called isoprenoids. More than Chapter 21 Lipid Biosynthesis828 Bile acids Vitamin DSteroid hormones Cholesterol Rubber Vitamin A Quinone electron carriers: ubiquinone, plastoquinone Dolichols Isoprene Phytol chain of chlorophyll Vitamin E Vitamin K Carotenoids Plant hormones abscisic acid and gibberellic acid H9004 3 -Isopentenyl pyrophosphate CH 2 C O CH 3 CH 2 CH 2 P O PO O H11002 O H11002 OO H11002 FIGURE 21–48 Overview of isoprenoid biosynthesis. The structures of most of the end products shown here are given in Chapter 10. FIGURE 21–47 Side-chain cleavage in the synthesis of steroid hor- mones. Cytochrome P-450 acts as electron carrier in this mixed- function oxidase system that oxidizes adjacent carbons. The process also requires the electron-transferring proteins adrenodoxin and adrenodoxin reductase. This system for cleaving side chains is found in mitochondria of the adrenal cortex, where active steroid pro- duction occurs. Pregnenolone is the precursor of all other steroid hormones (see Fig. 21–46). 17 22 20 Cholesterol Isocaproaldehyde Pregnenolone mixed-function oxidase cyt P-450 adrenodoxin (Fe–S) adrenodoxin reductase (flavoprotein) 2O 2 2H 2 O OH desmolase O 2 H11001 2H H11001 NADPH H11001 H 2 ONADP H11001 2NADP H11001 H11001 O 2 H C OCH 3 C NADPH H11001 H H11001 HO HO HO HO 20,22-Dihydroxycholesterol 8885d_c21_787-832 2/26/04 9:35 AM Page 828 mac76 mac76:385_reb: 20,000 different isoprenoid molecules have been dis- covered in nature, and hundreds of new ones are re- ported each year. Prenylation (covalent attachment of an isoprenoid; see Fig. 27–30) is a common mechanism by which pro- teins are anchored to the inner surface of cellular mem- branes in mammals (see Fig. 11–14). In some of these proteins the attached lipid is the 15-carbon farnesyl group; others have the 20-carbon geranylgeranyl group. Different enzymes attach the two types of lipids. It is possible that prenylation reactions target proteins to dif- ferent membranes, depending on which lipid is at- tached. Protein prenylation is another important role for the isoprene derivatives of the pathway to cholesterol. SUMMARY 21.4 Biosynthesis of Cholesterol, Steroids, and Isoprenoids ■ Cholesterol is formed from acetyl-CoA in a complex series of reactions, through the intermediates H9252-hydroxy-H9252-methylglutaryl-CoA, mevalonate, and two activated isoprenes, dimethylallyl pyrophosphate and isopentenyl pyrophosphate. Condensation of isoprene units produces the noncyclic squalene, which is cyclized to yield the steroid ring system and side chain. ■ Cholesterol synthesis is under hormonal control and is also inhibited by elevated concentrations of intracellular cholesterol, which acts through covalent modification and transcriptional regulation mechanisms. ■ Cholesterol and cholesteryl esters are carried in the blood as plasma lipoproteins. VLDL carries cholesterol, cholesteryl esters, and triacylglycerols from the liver to other tissues, where the triacylglycerols are degraded by lipoprotein lipase, converting VLDL to LDL. The LDL, rich in cholesterol and its esters, is taken up by receptor-mediated endocytosis, in which the apolipoprotein B-100 of LDL is recognized by receptors in the plasma membrane. HDL removes cholesterol from the blood, carrying it to the liver. Dietary conditions or genetic defects in cholesterol metabolism may lead to atherosclerosis and heart disease. ■ The steroid hormones (glucocorticoids, mineralocorticoids, and sex hormones) are produced from cholesterol by alteration of the side chain and introduction of oxygen atoms into the steroid ring system. In addition to cholesterol, a wide variety of isoprenoid compounds are derived from mevalonate through condensations of isopentenyl pyrophosphate and dimethylallyl pyrophosphate. ■ Prenylation of certain proteins targets them for association with cellular membranes and is essential for their biological activity. Chapter 21 Key Terms 829 Key Terms acetyl-CoA carboxylase 787 fatty acid synthase 789 acyl carrier protein (ACP) 790 fatty acyl-CoA desaturase 798 mixed-function oxidases 799 mixed-function oxygenases 799 cytochrome P-450 799 essential fatty acids 800 prostaglandins 800 cyclooxygenase (COX) 800 prostaglandin H 2 synthase 800 thromboxane synthase 800 thromboxanes 800 leukotrienes 800 glycerol 3-phosphate dehydrogenase 804 triacylglycerol cycle 806 glyceroneogenesis 806 thiazolidinediones 807 phosphatidylserine 811 phosphatidylglycerol 811 phosphatidylethanolamine 811 cardiolipin 811 phosphatidylcholine 812 plasmalogen 813 platelet-activating factor 813 cerebroside 813 sphingomyelin 813 gangliosides 813 isoprene 816 mevalonate 817 H9252-hydroxy-H9252-methylglutaryl-CoA (HMG-CoA) 817 thiolase 817 HMG-CoA synthase 817 HMG-CoA reductase 817 bile acids 820 cholesteryl esters 820 apolipoproteins 821 chylomicron 821 very-low-density lipoprotein (VLDL) 822 low-density lipoprotein (LDL) 823 high-density lipoprotein (HDL) 823 reverse cholesterol transport 824 LDL receptors 824 receptor-mediated endocytosis 824 atherosclerosis 827 lovastatin 827 mineralocorticoids 827 glucocorticoids 827 progesterone 827 androgens 827 estrogens 827 Terms in bold are defined in the glossary. 8885d_c21_787-832 2/26/04 9:35 AM Page 829 mac76 mac76:385_reb: Chapter 21 Lipid Biosynthesis830 Further Reading The general references in Chapters 10 and 17 are also useful. General Bell, S.J., Bradley, D., Forse, R.A., & Bistrian, B.R. (1997) The new dietary fats in health and disease. J. Am. Dietetic Assoc. 97, 280–286. Gotto, A.M., Jr. (ed.) (1987) Plasma Lipoproteins, New Comprehensive Biochemistry, Vol. 14 (Neuberger, A. & van Deenen, L.L.M., series eds), Elsevier Biomedical Press, Amsterdam. Twelve reviews cover the structure, synthesis, and metabolism of lipoproteins, regulation of cholesterol synthesis, and the enzymes LCAT and lipoprotein lipase. Hajjar, D.P. & Nicholson, A.C. (1995) Atherosclerosis. Am. Sci. 83, 460–467. A good description of the molecular basis of this disease and prospects for therapy. Hawthorne, J.N. & Ansell, G.B. (eds) (1982) Phospholipids, New Comprehensive Biochemistry, Vol. 4 (Neuberger, A. & van Deenen, L.L.M., series eds), Elsevier Biomedical Press, Amsterdam. This volume has excellent reviews of biosynthetic pathways to glycerophospholipids and sphingolipids, phospholipid transfer proteins, and bilayer assembly. Ohlrogge, J. & Browse, J. (1995) Lipid biosynthesis. Plant Cell 7, 957–970. A good summary of pathways for lipid biosynthesis in plants. Vance, D.E. & Vance, J.E. (eds) (1996) Biochemistry of Lipids, Lipoproteins, and Membranes, New Comprehensive Biochemistry, Vol. 31, Elsevier Science Publishing Co., Inc., New York. Excellent reviews of lipid structure, biosynthesis, and function. Biosynthesis of Fatty Acids and Eicosanoids Capdevila, J.H., Falck, J.R., & Estabrook, R.W. (1992) Cytochrome P450 and the arachidonate cascade. FASEB J. 6, 731–736. This issue contains 20 articles on the structure and function of various types of cytochrome P-450. Creelman, R.A. & Mullet, J.E. (1997) Biosynthesis and action of jasmonates in plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 48, 355–381. DeWitt, D.L. (1999) Cox-2-selective inhibitors: the new super aspirins. Mol. Pharmacol. 55, 625–631. A short, clear review of the topic discussed in Box 21–2. Drazen, J.M., Israel, E., & O’Byrne, P.M. (1999) Drug therapy: treatment of asthma with drugs modifying the leukotriene pathway. New Engl. J. Med. 340, 197–206. Lands, W.E.M. (1991) Biosynthesis of prostaglandins. Annu. Rev. Nutr. 11, 41–60. Discussion of the nutritional requirement for unsaturated fatty acids and recent biochemical work on pathways from arachidonate to prostaglandins; advanced level. Munday, M.R. (2002) Regulation of mammalian acetyl-CoA carboxylase. Biochem. Soc. Trans. 30, 1059–1064. Reshef, L., Olswang, Y., Cassuto, H., Blum, B., Croniger, C.M., Kalhan, S.C., Tilghman, S.M., & Hanson, R.M. (2003) Glyceroneogenesis and the triglyceride/fatty acid cycle. J. Biol. Chem. 278, 30,413–30,416. Smith, S. (1994) The animal fatty acid synthase: one gene, one polypeptide, seven enzymes. FASEB J. 8, 1248–1259. Smith, W.L., Garavito, R.M., & DeWitt, D.L. (1996) Prostaglandin endoperoxide H synthases (cyclooxygenases)-1 and -2. J. Biol. Chem. 271, 33,157–33,160. A concise review of the properties and roles of COX-1 and COX-2. Biosynthesis of Membrane Phospholipids Bishop, W.R. & Bell, R.M. (1988) Assembly of phospholipids into cellular membranes: biosynthesis, transmembrane movement and intracellular translocation. Annu. Rev. Cell Biol. 4, 579–610. Advanced review of the enzymology and cell biology of phospholipid synthesis and targeting. Dowhan, W. (1997) Molecular basis for membrane phospholipid diversity: why are there so many lipids? Annu. Rev. Biochem. 66, 199–232. Kennedy, E.P. (1962) The metabolism and function of complex lipids. Harvey Lect. 57, 143–171. A classic description of the role of cytidine nucleotides in phospholipid synthesis. Pavlidis, P., Ramaswami, M., & Tanouye, M.A. (1994) The Drosophila easily shocked gene: a mutation in a phospholipid synthetic pathway causes seizure, neuronal failure, and paralysis. Cell 79, 23–33. Description of the fascinating effects of changing the composition of membrane lipids in fruit flies. Raetz, C.R.H. & Dowhan, W. (1990) Biosynthesis and function of phospholipids in Escherichia coli. J. Biol. Chem. 265, 1235–1238. A brief review of bacterial biosynthesis of phospholipids and lipopolysaccharides. Biosynthesis of Cholesterol, Steroids, and Isoprenoids Bittman, R. (ed.) (1997) Subcellular Biochemistry, Vol. 28: Cholesterol: Its Functions and Metabolism in Biology and Medicine, Plenum Press, New York. Bloch, K. (1965) The biological synthesis of cholesterol. Science 150, 19–28. The author’s Nobel address; a classic description of cholesterol synthesis in animals. Brown, M.S., & Goldstein, J.L. (1999) A proteolytic pathway that controls the cholesterol content of membranes, cells, and blood. Proc. Natl. Acad. Sci. USA 96, 11,041–11,048. Chang, T.Y., Chang, C.C.Y., & Cheng, D. (1997) Acyl–coenzyme A: cholesterol acyltransferase. Annu. Rev. Biochem. 66, 613–638. Edwards, P.A. & Ericsson, J. (1999) Sterols and isoprenoids: signaling molecules derived from the cholesterol biosynthetic pathway. Annu. Rev. Biochem. 68, 157–185. 8885d_c21_787-832 2/26/04 9:35 AM Page 830 mac76 mac76:385_reb: Chapter 21 Problems 831 Gimpl, G., Burger, K., & Fahrenholz, F. (2002) A closer look at the cholesterol sensor. Trends Biochem. Sci. 27, 596–599. Goldstein, J.L. & Brown, M.S. (1990) Regulation of the mevalonate pathway. Nature 343, 425–430. Description of the allosteric and covalent regulation of the enzymes of the mevalonate pathway; includes a short discussion of the prenylation of Ras and other proteins. Knopp, R.H. (1999) Drug therapy: drug treatment of lipid disorders. New Engl. J. Med. 341, 498–511. Review of the use of HMG-CoA inhibitors and bile acid–binding resins to reduce serum cholesterol. Krieger, M. (1999) Charting the fate of the “good cholesterol”: identification and characterization of the high-density lipoprotein receptor SR-BI. Annu. Rev. Biochem. 68, 523–558. McGarvey, D.J. & Croteau, R. (1995) Terpenoid metabolism. Plant Cell 7, 1015–1026. A description of the amazing diversity of isoprenoids in plants. Olson, R.E. (1998) Discovery of the lipoproteins, their role in fat transport and their significance as risk factors. J. Nutr. 128 (2 Suppl.), 439S–443S. Brief, clear, historical background on studies of lipoprotein function. Russell, D.W. (2003) The enzymes, regulation, and genetics of bile acid synthesis. Annu. Rev. Biochem. 72, 137–174. Young, S.G. & Fielding, C.J. (1999) The ABCs of cholesterol efflux. Nat. Genet. 22, 316–318. A brief review of three papers in this journal issue that establish mutations in ABC1 as the cause of Tangier disease and familial HDL deficiency. 1. Pathway of Carbon in Fatty Acid Synthesis Using your knowledge of fatty acid biosynthesis, provide an expla- nation for the following experimental observations: (a) Addition of uniformly labeled [ 14 C]acetyl-CoA to a soluble liver fraction yields palmitate uniformly labeled with 14 C. (b) However, addition of a trace of uniformly labeled [ 14 C]acetyl-CoA in the presence of an excess of unlabeled malonyl-CoA to a soluble liver fraction yields palmitate labeled with 14 C only in C-15 and C-16. 2. Synthesis of Fatty Acids from Glucose After a per- son has ingested large amounts of sucrose, the glucose and fructose that exceed caloric requirements are transformed to fatty acids for triacylglycerol synthesis. This fatty acid syn- thesis consumes acetyl-CoA, ATP, and NADPH. How are these substances produced from glucose? 3. Net Equation of Fatty Acid Synthesis Write the net equation for the biosynthesis of palmitate in rat liver, start- ing from mitochondrial acetyl-CoA and cytosolic NADPH, ATP, and CO 2 . 4. Pathway of Hydrogen in Fatty Acid Synthesis Con- sider a preparation that contains all the enzymes and cofac- tors necessary for fatty acid biosynthesis from added acetyl- CoA and malonyl-CoA. (a) If [2- 2 H]acetyl-CoA (labeled with deuterium, the heavy isotope of hydrogen) and an excess of unlabeled malonyl-CoA are added as sub- strates, how many deuterium atoms are incorporated into every molecule of palmitate? What are their locations? Explain. (b) If unlabeled acetyl-CoA and [2- 2 H]malonyl-CoA are added as substrates, how many deuterium atoms are in- corporated into every molecule of palmitate? What are their locations? Explain. 5. Energetics of H9252-Ketoacyl-ACP Synthase In the condensation reaction catalyzed by H9252-ketoacyl-ACP synthase (see Fig. 21–5), a four-carbon unit is synthesized by the com- bination of a two-carbon unit and a three-carbon unit, with the release of CO 2 . What is the thermodynamic advantage of this process over one that simply combines two two-carbon units? 6. Modulation of Acetyl-CoA Carboxylase Acetyl- CoA carboxylase is the principal regulation point in the biosynthesis of fatty acids. Some of the properties of the en- zyme are described below. (a) Addition of citrate or isocitrate raises the V max of the enzyme as much as 10-fold. (b) The enzyme exists in two interconvertible forms that differ markedly in their activities: Protomer (inactive) filamentous polymer (active) Citrate and isocitrate bind preferentially to the filamentous form, and palmitoyl-CoA binds preferentially to the protomer. Explain how these properties are consistent with the reg- ulatory role of acetyl-CoA carboxylase in the biosynthesis of fatty acids. 7. Shuttling of Acetyl Groups across the Mitochon- drial Inner Membrane The acetyl group of acetyl-CoA, produced by the oxidative decarboxylation of pyruvate in the z y H5008 OOC C 2 H 2 H C O S-CoA 2 H C 2 H 2 H C O S-CoA Problems 8885d_c21_787-832 2/26/04 9:35 AM Page 831 mac76 mac76:385_reb: Chapter 21 Lipid Biosynthesis832 mitochondrion, is transferred to the cytosol by the acetyl group shuttle outlined in Figure 21-10. (a) Write the overall equation for the transfer of one acetyl group from the mitochondrion to the cytosol. (b) What is the cost of this process in ATPs per acetyl group? (c) In Chapter 17 we encountered an acyl group shuttle in the transfer of fatty acyl–CoA from the cytosol to the mito- chondrion in preparation for H9252 oxidation (see Fig. 17–6). One result of that shuttle was separation of the mitochondrial and cytosolic pools of CoA. Does the acetyl group shuttle also accomplish this? Explain. 8. Oxygen Requirement for Desaturases The biosyn- thesis of palmitoleate (see Fig. 21–12), a common unsatu- rated fatty acid with a cis double bond in the H9004 9 position, uses palmitate as a precursor. Can this be carried out under strictly anaerobic conditions? Explain. 9. Energy Cost of Triacylglycerol Synthesis Use a net equation for the biosynthesis of tripalmitoylglycerol (tri- palmitin) from glycerol and palmitate to show how many ATPs are required per molecule of tripalmitin formed. 10. Turnover of Triacylglycerols in Adipose Tissue When [ 14 C]glucose is added to the balanced diet of adult rats, there is no increase in the total amount of stored triacyl- glycerols, but the triacylglycerols become labeled with 14 C. Explain. 11. Energy Cost of Phosphatidylcholine Synthesis Write the sequence of steps and the net reaction for the biosynthesis of phosphatidylcholine by the salvage pathway from oleate, palmitate, dihydroxyacetone phosphate, and choline. Starting from these precursors, what is the cost (in number of ATPs) of the synthesis of phosphatidylcholine by the salvage pathway? 12. Salvage Pathway for Synthesis of Phosphatidyl- choline A young rat maintained on a diet deficient in me- thionine fails to thrive unless choline is included in the diet. Explain. 13. Synthesis of Isopentenyl Pyrophosphate If 2- [ 14 C]acetyl-CoA is added to a rat liver homogenate that is syn- thesizing cholesterol, where will the 14 C label appear in H9004 3 - isopentenyl pyrophosphate, the activated form of an isoprene unit? 14. Activated Donors in Lipid Synthesis In the biosyn- thesis of complex lipids, components are assembled by trans- fer of the appropriate group from an activated donor. For ex- ample, the activated donor of acetyl groups is acetyl-CoA. For each of the following groups, give the form of the acti- vated donor: (a) phosphate; (b) D-glucosyl; (c) phospho- ethanolamine; (d) D-galactosyl; (e) fatty acyl; (f) methyl; (g) the two-carbon group in fatty acid biosynthesis; (h) H9004 3 - isopentenyl. 15. Importance of Fats in the Diet When young rats are placed on a totally fat-free diet, they grow poorly, develop a scaly dermatitis, lose hair, and soon die—symptoms that can be prevented if linoleate or plant material is included in the diet. What makes linoleate an essential fatty acid? Why can plant material be substituted? 16. Regulation of Cholesterol Biosynthesis Cholesterol in humans can be obtained from the diet or synthesized de novo. An adult human on a low-cholesterol diet typically synthesizes 600 mg of cholesterol per day in the liver. If the amount of cho- lesterol in the diet is large, de novo synthesis of cholesterol is drastically reduced. How is this regulation brought about? 8885d_c21_787-832 2/26/04 9:35 AM Page 832 mac76 mac76:385_reb: chapter N itrogen ranks behind only carbon, hydrogen, and oxygen in its contribution to the mass of living sys- tems. Most of this nitrogen is bound up in amino acids and nucleotides. In this chapter we address all aspects of the metabolism of these nitrogen-containing com- pounds except amino acid catabolism, which is covered in Chapter 18. Discussing the biosynthetic pathways for amino acids and nucleotides together is a sound approach, not only because both classes of molecules contain nitrogen (which arises from common biological sources) but be- cause the two sets of pathways are extensively inter- twined, with several key intermediates in common. Cer- tain amino acids or parts of amino acids are incorporated into the structure of purines and pyrimidines, and in one case part of a purine ring is incorporated into an amino acid (histidine). The two sets of pathways also share much common chemistry, in particular a preponderance of reactions involving the transfer of nitrogen or one- carbon groups. The pathways described here can be intimidating to the beginning biochemistry student. Their complexity arises not so much from the chemistry itself, which in many cases is well understood, but from the sheer num- ber of steps and variety of intermediates. These path- ways are best approached by maintaining a focus on metabolic principles we have already discussed, on key intermediates and precursors, and on common classes of reactions. Even a cursory look at the chemistry can be rewarding, for some of the most unusual chemical transformations in biological systems occur in these pathways; for instance, we find prominent examples of the rare biological use of the metals molybdenum, sele- nium, and vanadium. The effort also offers a practical dividend, especially for students of human or veterinary medicine. Many genetic diseases of humans and animals have been traced to an absence of one or more enzymes of amino acid and nucleotide metabolism, and many pharmaceuticals in common use to combat infectious diseases are inhibitors of enzymes in these pathways— as are a number of the most important agents in cancer chemotherapy. Regulation is crucial in the biosynthesis of the nitrogen-containing compounds. Because each amino acid and each nucleotide is required in relatively small amounts, the metabolic flow through most of these path- ways is not nearly as great as the biosynthetic flow lead- ing to carbohydrate or fat in animal tissues. Because the different amino acids and nucleotides must be made in BIOSYNTHESIS OF AMINO ACIDS, NUCLEOTIDES, AND RELATED MOLECULES 22.1 Overview of Nitrogen Metabolism 834 22.2 Biosynthesis of Amino Acids 841 22.3 Molecules Derived from Amino Acids 854 22.4 Biosynthesis and Degradation of Nucleotides 862 Time passes rapidly when you are having fun. The thrill of seeing people get well who might otherwise have died of disease . . . cannot be described in words. The Nobel Prize was only the icing on the cake. —Gertrude Elion, quoted in an article in Science, 2002 22 833 8885d_c22_833-880 2/6/04 8:35 AM Page 833 mac76 mac76:385_reb: the correct ratios and at the right time for protein and nucleic acid synthesis, their biosynthetic pathways must be accurately regulated and coordinated with each other. And because amino acids and nucleotides are charged molecules, their levels must be regulated to maintain electrochemical balance in the cell. As discussed in ear- lier chapters, pathways can be controlled by changes in either the activity or the amounts of specific enzymes. The pathways we encounter in this chapter provide some of the best-understood examples of the regulation of enzyme activity. Control of the amounts of different enzymes in a cell (that is, of their synthesis and degra- dation) is a topic covered in Chapter 28. 22.1 Overview of Nitrogen Metabolism The biosynthetic pathways leading to amino acids and nucleotides share a requirement for nitrogen. Because soluble, biologically useful nitrogen compounds are gen- erally scarce in natural environments, most organisms maintain strict economy in their use of ammonia, amino acids, and nucleotides. Indeed, as we shall see, free amino acids, purines, and pyrimidines formed during metabolic turnover of proteins and nucleic acids are of- ten salvaged and reused. We first examine the pathways by which nitrogen from the environment is introduced into biological systems. The Nitrogen Cycle Maintains a Pool of Biologically Available Nitrogen The most important source of nitrogen is air, which is four-fifths molecular nitrogen (N 2 ). However, relatively few species can convert atmospheric nitrogen into forms useful to living organisms. In the biosphere, the meta- bolic processes of different species function interde- pendently to salvage and reuse biologically available ni- trogen in a vast nitrogen cycle (Fig. 22–1). The first step in the cycle is fixation (reduction) of atmospheric nitrogen by nitrogen-fixing bacteria to yield ammonia (NH 3 or NH 4 H11001 ). Although ammonia can be used by most living organisms, soil bacteria that derive their energy by oxidizing ammonia to nitrite (NO 2 H11002 ) and ultimately nitrate (NO 3 H11002 ) are so abundant and active that nearly all ammonia reaching the soil is oxidized to nitrate. This process is known as nitrification. Plants and many bac- teria can take up and readily reduce nitrate and nitrite through the action of nitrate and nitrite reductases. The ammonia so formed is incorporated into amino acids by plants. Animals then use plants as a source of amino acids, both nonessential and essential, to build their pro- teins. When organisms die, microbial degradation of their proteins returns ammonia to the soil, where nitri- fying bacteria again convert it to nitrite and nitrate. A balance is maintained between fixed nitrogen and at- mospheric nitrogen by bacteria that convert nitrate to N 2 under anaerobic conditions, a process called deni- trification (Fig. 22–1). These soil bacteria use NO 3 H11002 rather than O 2 as the ultimate electron acceptor in a se- ries of reactions that (like oxidative phosphorylation) generates a transmembrane proton gradient, which is used to synthesize ATP. Now let’s examine the process of nitrogen fixation, the first step in the nitrogen cycle. Nitrogen Is Fixed by Enzymes of the Nitrogenase Complex Only certain prokaryotes can fix atmospheric nitrogen. These include the cyanobacteria of soils and fresh and salt waters, other kinds of free-living soil bacteria such as Azotobacter species, and the nitrogen-fixing bacte- ria that live as symbionts in the root nodules of legu- minous plants. The first important product of nitrogen fixation is ammonia, which can be used by all organisms either directly or after its conversion to other soluble compounds such as nitrites, nitrates, or amino acids. The reduction of nitrogen to ammonia is an exer- gonic reaction: N 2 H11001 3H 2 88n 2NH 3 H9004GH11032H11034 H11005 H1100233.5 kJ/mol reduction by some anaerobic bacteria, most plants N 2 Ammonia NH 4 Nitrite NO 2 nitrogen fixation by some bacteria (e.g., Klebsiella, Azotobacter, Rhizobium) denitrification nitrification by soil bacteria (e.g., Nitrobacter) nitrification by soil bacteria (e.g., Nitrosomonas) synthesis in plants and microorganisms degradation by animals and microorganisms Amino acids and other reduced nitrogen-carbon compounds Nitrate NO 3 H11002 H11002 H11001 FIGURE 22–1 The nitrogen cycle. The total amount of nitrogen fixed annually in the biosphere exceeds 10 11 kg. Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules834 8885d_c22_833-880 2/6/04 8:35 AM Page 834 mac76 mac76:385_reb: The NmN triple bond, however, is very stable, with a bond energy of 930 kJ/mol. Nitrogen fixation therefore has an extremely high activation energy, and atmos- pheric nitrogen is almost chemically inert under normal conditions. Ammonia is produced industrially by the Haber process (named for its inventor, Fritz Haber), which requires temperatures of 400 to 500 H11034C and ni- trogen and hydrogen at pressures of tens of thousands of kilopascals (several hundred atmospheres) to provide the necessary activation energy. Biological nitrogen fix- ation, however, must occur at biological temperatures and at 0.8 atm of nitrogen, and the high activation bar- rier is overcome by other means. This is accomplished, at least in part, by the binding and hydrolysis of ATP. The overall reaction can be written N 2 H11001 10H H11001 H11001 8e H11002 H11001 16ATP 88n 2NH 4 H11001 H11001 16ADP H11001 16P i H11001 H 2 Biological nitrogen fixation is carried out by a highly conserved complex of proteins called the nitrogenase complex (Fig. 22–2), the crucial components of which are dinitrogenase reductase and dinitrogenase (Fig. 22–3). Dinitrogenase reductase (M r 60,000) is a dimer of two identical subunits. It contains a single 4Fe- 4S redox center (see Fig. 19–5), bound between the subunits, and can be oxidized and reduced by one elec- tron. It also has two binding sites for ATP/ADP (one site on each subunit). Dinitrogenase (M r 240,000), a tetramer with two copies of two different subunits, con- tains both iron and molybdenum; its redox centers have a total of 2 Mo, 32 Fe, and 30 S per tetramer. About half of the iron and sulfur is present as two bridged pairs of 4Fe-4S centers called P clusters; the remainder is pres- ent as part of a novel iron-molybdenum cofactor. A form of nitrogenase that contains vanadium rather than molybdenum has been discovered, and some bacterial species can produce both types of nitrogenase systems. The vanadium-containing enzyme may be the primary nitrogen-fixing system under some environmental con- ditions, but it is not yet as well characterized as the molybdenum-dependent enzyme. Nitrogen fixation is carried out by a highly reduced form of dinitrogenase and requires eight electrons: six for the reduction of N 2 and two to produce one mole- cule of H 2 as an obligate part of the reaction mecha- nism. Dinitrogenase is reduced by the transfer of elec- trons from dinitrogenase reductase (Fig. 22–2). The dinitrogenase tetramer has two binding sites for the re- ductase. The required eight electrons are transferred from reductase to dinitrogenase one at a time: a reduced reductase molecule binds to the dinitrogenase and transfers a single electron, then the oxidized reductase dissociates from dinitrogenase, in a repeating cycle. Each turn of the cycle requires the hydrolysis of two ATP molecules by the dimeric reductase. The immedi- ate source of electrons to reduce dinitrogenase reduc- tase varies, with reduced ferredoxin (p. 733; see also Fig. 19–5), reduced flavodoxin, and perhaps other sources playing a role. In at least one species, the ulti- mate source of electrons to reduce ferredoxin is pyru- vate (Fig. 22–2). The role of ATP in this process is somewhat un- usual. As you will recall, ATP can contribute not only chemical energy, through the hydrolysis of one or more of its phosphoanhydride bonds, but also binding en- ergy (pp. 196, 301), through noncovalent interactions that lower the activation energy. In the reaction car- ried out by dinitrogenase reductase, both ATP binding 22.1 Overview of Nitrogen Metabolism 835 8 Ferredoxin or 8 flavodoxin (oxidized) 8 Dinitrogenase reductase (oxidized) + 16ATP Dinitrogenase (oxidized) Dinitrogenase (reduced) 8 Dinitrogenase reductase (reduced) 8 Dinitrogenase reductase (oxidized) 8 Ferredoxin or 8 flavodoxin (reduced) N 2 2NH 4 8 Dinitrogenase reductase (reduced) + 16ATP 2H + H 2 8e H11002 8e H11002 8e H11002 16ADP + 16P i 16 ATP 8e H11002 4CoA + 4 pyruvate 4CO 2 + 4 acetyl-CoA + FIGURE 22–2 Nitrogen fixation by the nitrogenase complex. Elec- trons are transferred from pyruvate to dinitrogenase via ferredoxin (or flavodoxin) and dinitrogenase reductase. Dinitrogenase reductase re- duces dinitrogenase one electron at a time, with at least six electrons required to fix one molecule of N 2 . An additional two electrons are used to reduce 2 H H11001 to H 2 in a process that obligatorily accompanies nitrogen fixation in anaerobes, making a total of eight electrons re- quired per N 2 molecule. The subunit structures and metal cofactors of the dinitrogenase reductase and dinitrogenase proteins are described in the text and in Figure 22–3. 8885d_c22_833-880 2/6/04 8:35 AM Page 835 mac76 mac76:385_reb: and ATP hydrolysis bring about protein conformational changes that help overcome the high activation energy of nitrogen fixation. The binding of two ATP molecules to the reductase shifts the reduction potential (EH11032H11034) of this protein from H11002300 to H11002420 mV, an enhancement of its reducing power that is required to transfer elec- trons to dinitrogenase. The ATP molecules are then hy- drolyzed just before the actual transfer of one electron to dinitrogenase. Another important characteristic of the nitrogenase complex is an extreme lability in the presence of oxy- gen. The reductase is inactivated in air, with a half-life of 30 seconds; dinitrogenase has a half-life of 10 min- utes in air. Free-living bacteria that fix nitrogen cope with this problem in a variety of ways. Some live only anaerobically or repress nitrogenase synthesis when oxygen is present. Some aerobic species, such as Azo- tobacter vinelandii, partially uncouple electron trans- fer from ATP synthesis so that oxygen is burned off as rapidly as it enters the cell (see Box 19–1). When fix- ing nitrogen, cultures of these bacteria actually increase in temperature as a result of their efforts to rid them- selves of oxygen. The symbiotic relationship between leguminous plants and the nitrogen-fixing bacteria in their root nodules (Fig. 22–4) takes care of both the energy re- quirements and the oxygen lability of the nitrogenase complex. The energy required for nitrogen fixation was probably the evolutionary driving force for this plant-bacteria association. The bacteria in root nod- ules have access to a large reservoir of energy in the form of abundant carbohydrate and citric acid cycle intermediates made available by the plant. This may allow the bacteria to fix hundreds of times more ni- trogen than their free-living cousins can fix under con- ditions generally encountered in soils. To solve the oxygen-toxicity problem, the bacteria in root nodules are bathed in a solution of the oxygen-binding heme protein leghemoglobin, produced by the plant (al- though the heme may be contributed by the bacteria). Leghemoglobin binds all available oxygen so that it cannot interfere with nitrogen fixation, and efficiently delivers the oxygen to the bacterial electron-transfer system. The benefit to the plant, of course, is a ready supply of reduced nitrogen. The efficiency of the sym- biosis between plants and bacteria is evident in the enrichment of soil nitrogen brought about by legumi- nous plants. This enrichment is the basis of crop ro- tation methods, in which plantings of nonleguminous plants (such as maize) that extract fixed nitrogen from Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules836 (a) (b) (c) FIGURE 22–3 Enzymes and cofactors of the nitrogenase complex. (PDB ID 1N2C) (a) In this ribbon diagram, the dinitrogenase subunits are shown in gray and pink, the dinitrogenase reductase subunits in blue and green. The bound ADP is red. Note the 4Fe-4S complex (Fe atoms orange, S atoms yellow) and the iron-molybdenum cofactor (Mo black, homocitrate light gray). The P clusters (bridged pairs of 4Fe-4S complexes) are also shown. (b) The dinitrogenase complex cofactors without the protein (colors as in (a)). (c) The iron-molybdenum co- factor contains 1 Mo (black), 7 Fe (orange), 9 S (yellow), and one mol- ecule of homocitrate (gray). 8885d_c22_833-880 2/6/04 8:35 AM Page 836 mac76 mac76:385_reb: the soil are alternated every few years with plantings of legumes such as alfalfa, peas, or clover. Nitrogen fixation is the subject of intense study, be- cause of its immense practical importance. Industrial production of ammonia for use in fertilizers requires a large and expensive input of energy, and this has spurred a drive to develop recombinant or transgenic organisms that can fix nitrogen. Recombinant DNA techniques (Chapter 9) are being used to transfer the DNA that encodes the enzymes of nitrogen fixation into non-nitrogen-fixing bacteria and plants. Success in these efforts will depend on overcoming the problem of oxygen toxicity in any cell that produces nitrogenase. Ammonia Is Incorporated into Biomolecules through Glutamate and Glutamine Reduced nitrogen in the form of NH 4 H11001 is assimilated into amino acids and then into other nitrogen-containing bio- molecules. Two amino acids, glutamate and glutamine, provide the critical entry point. Recall that these same two amino acids play central roles in the catabolism of am- monia and amino groups in amino acid oxidation (Chap- ter 18). Glutamate is the source of amino groups for most other amino acids, through transamination reactions (the reverse of the reaction shown in Fig. 18–4). The amide nitrogen of glutamine is a source of amino groups in a wide range of biosynthetic processes. In most types of cells, and in extracellular fluids in higher organisms, one or both of these amino acids are present at higher concentrations—sometimes an order of magnitude or more higher—than other amino acids. An Escherichia coli cell requires so much glutamate that this amino acid is one of the primary solutes in the cytosol. Its concen- tration is regulated not only in response to the cell’s ni- trogen requirements but also to maintain an osmotic bal- ance between the cytosol and the external medium. The biosynthetic pathways to glutamate and gluta- mine are simple, and all or some of the steps occur in most organisms. The most important pathway for the 22.1 Overview of Nitrogen Metabolism 837 FIGURE 22–4 Nitrogen-fixing nodules. (a) Root nodules of bird’s-foot trefoil, a legume. The flower of this common plant is shown in the in- set. (b) Artificially colorized electron micrograph of a thin section through a pea root nodule. Symbiotic nitrogen-fixing bacteria, or bac- teroids (red), live inside the nodule cells, surrounded by the peribac- teroid membrane (blue). Bacteroids produce the nitrogenase complex that converts atmospheric nitrogen (N 2 ) to ammonium (NH 4 H11001 ); with- out the bacteroids, the plant is unable to utilize N 2 . The infected root cells provide some factors essential for nitrogen fixation, including leghemoglobin; this heme protein has a very high binding affinity for oxygen, which strongly inhibits nitrogenase. (The cell nucleus is shown in yellow/green. Not visible in this micrograph are other organelles of the infected root cell that are normally found in plant cells.) (a) (b) 2mH9262 8885d_c22_833-880 2/6/04 8:35 AM Page 837 mac76 mac76:385_reb: assimilation of NH 4 H11001 into glutamate requires two reactions. First, glutamine synthetase catalyzes the reaction of glutamate and NH 4 H11001 to yield glutamine. This reaction takes place in two steps, with enzyme-bound H9253-glutamyl phosphate as an intermediate (see Fig. 18–8): (1) Glutamate H11001 ATP 88n H9253-glutamyl phosphate H11001 ADP (2) H9253-Glutamyl phosphate H11001 NH 4 H11001 88n glutamine H11001 P i H11001 H H11001 Sum: Glutamate H11001 NH 4 H11001 H11001 ATP 88n glutamine H11001 ADP H11001 Pi H11001 H H11001 (22–1) Glutamine synthetase is found in all organisms. In ad- dition to its importance for NH 4 H11001 assimilation in bacte- ria, it has a central role in amino acid metabolism in mammals, converting toxic free NH 4 H11001 to glutamine for transport in the blood (Chapter 18). In bacteria and plants, glutamate is produced from glutamine in a reaction catalyzed by glutamate syn- thase. H9251-Ketoglutarate, an intermediate of the citric acid cycle, undergoes reductive amination with gluta- mine as nitrogen donor: H9251-Ketoglutarate H11001 glutamine H11001 NADPH H11001 H H11001 88n 2 glutamate H11001 NADP H11001 (22–2) The net reaction of glutamine synthetase and glutamate synthase (Eqns 22–1 and 22–2) is H9251-Ketoglutarate H11001 NH 4 H11001 H11001 NADPH H11001 ATP 88n L-glutamate H11001 NADP H11001 H11001 ADP H11001 P i Glutamate synthase is not present in animals, which, in- stead, maintain high levels of glutamate by processes such as the transamination of H9251-ketoglutarate during amino acid catabolism. Glutamate can also be formed in yet another, albeit minor, pathway: the reaction of H9251-ketoglutarate and NH 4 H11001 to form glutamate in one step. This is catalyzed by L-glutamate dehydrogenase, an enzyme present in all or- ganisms. Reducing power is furnished by NADPH: H9251-Ketoglutarate H11001 NH 4 H11001 H11001 NADPH 88n L-glutamate H11001 NADP H11001 H11001 H 2 O We encountered this reaction in the catabolism of amino acids (see Fig. 18–7). In eukaryotic cells, L-glutamate dehydrogenase is located in the mitochondrial matrix. The reaction equilibrium favors reactants, and the K m for NH 4 H11001 (~1 mM) is so high that the reaction probably makes only a modest contribution to NH 4 H11001 assimilation into amino acids and other metabolites. (Recall that the glutamate dehydrogenase reaction, in reverse (see Fig. 18–10), is one source of NH 4 H11001 destined for the urea cy- cle.) Concentrations of NH 4 H11001 high enough for the gluta- mate dehydrogenase reaction to make a significant con- tribution to glutamate levels generally occur only when NH 3 is added to the soil or when organisms are grown in a laboratory in the presence of high NH 3 concentra- tions. In general, soil bacteria and plants rely on the two- enzyme pathway outlined above (Eqns 22–1, 22–2). Glutamine Synthetase Is a Primary Regulatory Point in Nitrogen Metabolism The activity of glutamine synthetase is regulated in vir- tually all organisms—not surprising, given its central metabolic role as an entry point for reduced nitrogen. In enteric bacteria such as E. coli, the regulation is un- usually complex. The enzyme has 12 identical subunits of M r 50,000 (Fig. 22–5) and is regulated both alloster- ically and by covalent modification. Alanine, glycine, and at least six end products of glutamine metabolism are allosteric inhibitors of the enzyme (Fig. 22–6). Each in- hibitor alone produces only partial inhibition, but the ef- fects of multiple inhibitors are more than additive, and all eight together virtually shut down the enzyme. This control mechanism provides a constant adjustment of glutamine levels to match immediate metabolic re- quirements. Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules838 FIGURE 22–5 Subunit structure of glutamine synthetase as deter- mined by x-ray diffraction. (PDB ID 2GLS) (a) Side view. The 12 sub- units are identical; they are differently colored to illustrate packing and placement. (b) Top view, showing active sites (green). (b) (a) 8885d_c22_833-880 2/6/04 8:35 AM Page 838 mac76 mac76:385_reb: Superimposed on the allosteric regulation is inhibi- tion by adenylylation of (addition of AMP to) Tyr 397 , lo- cated near the enzyme’s active site (Fig. 22–7). This co- valent modification increases sensitivity to the allosteric inhibitors, and activity decreases as more subunits are adenylylated. Both adenylylation and deadenylylation are promoted by adenylyltransferase (AT in Fig. 22–7), part of a complex enzymatic cascade that re- sponds to levels of glutamine, H9251-ketoglutarate, ATP, and P i . The activity of adenylyltransferase is modulated by binding to a regulatory protein called P II , and the activ- ity of P II , in turn, is regulated by covalent modification (uridylylation), again at a Tyr residue. The adenylyl- transferase complex with uridylylated P II (P II -UMP) stimulates deadenylylation, whereas the same complex 22.1 Overview of Nitrogen Metabolism 839 Glutamate ATP ADP + P i NH 3 Glutamine CTP Histidine Glucosamine 6-phosphate AMP Tryptophan Carbamoyl phosphate Glycine Alanine glutamine synthetase FIGURE 22–6 Allosteric regulation of glutamine synthetase. The en- zyme undergoes cumulative regulation by six end products of gluta- mine metabolism. Alanine and glycine probably serve as indicators of the general status of amino acid metabolism in the cell. Enzyme OPCH 2 O H11002 H OH Adenine (a) (b) O O H OH H H O AMP Glutamine synthetase (inactive) Glutamine synthetase (active) Glutamate ADP + P i NH 3 Glutamine deadenylylation adenylylation PP i AT P II P i ADP AT P II UMP P II P II AT UMP H 2 OPP i uridylylation -KetoglutarateH9251 UT P i Gln ATP AT ATP ATP UTP UMP H17011 FIGURE 22–7 Second level of regulation of glutamine synthetase: covalent modifications. (a) An adenylylated Tyr residue. (b) Cascade leading to adenylylation (inactivation) of glutamine synthetase. AT rep- resents adenylyltransferase; UT, uridylyltransferase. Details of this cas- cade are discussed in the text. 8885d_c22_833-880 2/6/04 8:35 AM Page 839 mac76 mac76:385_reb: with deuridylylated P II stimulates adenylylation of glut- amine synthetase. Both uridylylation and deuridylyla- tion of P II are brought about by a single enzyme, uridy- lyltransferase. Uridylylation is inhibited by binding of glutamine and P i to uridylyltransferase and is stimulated by binding of H9251-ketoglutarate and ATP to P II . The regulation does not stop there. The uridylylated P II also mediates the activation of transcription of the gene encoding glutamine synthetase, thus increasing the cellular concentration of the enzyme; the deuridy- lylated P II brings about a decrease in transcription of the same gene. This mechanism involves an interaction of P II with additional proteins involved in gene regula- tion, of a type described in Chapter 28. The net result of this elaborate system of controls is a decrease in glu- tamine synthetase activity when glutamine levels are high, and an increase in activity when glutamine levels are low and H9251-ketoglutarate and ATP (substrates for the synthetase reaction) are available. The multiple layers of regulation permit a sensitive response in which glut- amine synthesis is tailored to cellular needs. Several Classes of Reactions Play Special Roles in the Biosynthesis of Amino Acids and Nucleotides The pathways described in this chapter include a vari- ety of interesting chemical rearrangements. Several of these recur and deserve special note before we progress to the pathways themselves. These are (1) transamina- tion reactions and other rearrangements promoted by enzymes containing pyridoxal phosphate; (2) transfer of one-carbon groups, with either tetrahydrofolate (usu- ally at the OCHO and OCH 2 OH oxidation levels) or S- adenosylmethionine (at the OCH 3 oxidation level) as cofactor; and (3) transfer of amino groups derived from the amide nitrogen of glutamine. Pyridoxal phosphate (PLP), tetrahydrofolate (H 4 folate), and S-adenosylme- thionine (adoMet) were described in some detail in Chapter 18 (see Figs 18–6, 18–17, and 18–18). Here we focus on amino group transfer involving the amide ni- trogen of glutamine. More than a dozen known biosynthetic reactions use glutamine as the major physiological source of amino groups, and most of these occur in the pathways out- lined in this chapter. As a class, the enzymes catalyzing these reactions are called glutamine amidotrans- ferases. All have two structural domains: one binding glutamine, the other binding the second substrate, which serves as amino group acceptor (Fig. 22–8). A conserved Cys residue in the glutamine-binding domain is believed to act as a nucleophile, cleaving the amide bond of glu- tamine and forming a covalent glutamyl-enzyme inter- mediate. The NH 3 produced in this reaction is not re- leased, but instead is transferred through an “ammonia channel” to a second active site, where it reacts with the second substrate to form the aminated product. The covalent intermediate is hydrolyzed to the free enzyme and glutamate. If the second substrate must be acti- vated, the usual method is the use of ATP to generate an acyl phosphate intermediate (ROOX in Fig. 22–8, with X as a phosphoryl group). The enzyme glutaminase acts in a similar fashion but uses H 2 O as the second sub- strate, yielding NH 4 H11001 and glutamate (see Fig. 18–8). Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules840 H 3 N Cys SH CH 2 CH 2 H + NH 3 channel Glutamine amidotransferase Glutamine- binding domain NH 3 - acceptor domain Glutamine Acceptor H 2 O Glutamate Activated substrate NH 2 CH : C COO – OH orR R 1 R 2 OC + O Glutamyl- enzyme intermediate 1 2 H 3 N Cys S CH 2 CH 2 CH C COO – + O NH 3 NH 2 OX or R R 1 R R 2 + + OC OX or H R 1 R 2 NHCH 2 O MECHANISM FIGURE 22–8 Proposed mechanism for glutamine amidotransferases. Each enzyme has two domains. The glutamine- binding domain contains structural elements conserved among many of these enzymes, including a Cys residue required for activity. The NH 3 -acceptor (second-substrate) domain varies. 1 The H9253-amido nitrogen of glutamine (red) is released as NH 3 in a reaction that prob- ably involves a covalent glutamyl-enzyme intermediate. The NH 3 trav- els through a channel to the second active site, where 2 it reacts with any of several acceptors. Two types of amino acceptors are shown. X represents an activating group, typically a phosphoryl group derived from ATP, that facilitates displacement of a hydroxyl group from ROOH by NH 3 . 8885d_c22_833-880 2/6/04 8:35 AM Page 840 mac76 mac76:385_reb: SUMMARY 22.1 Overview of Nitrogen Metabolism ■ The molecular nitrogen that makes up 80% of the earth’s atmosphere is unavailable to most living organisms until it is reduced. This fixation of atmospheric N 2 takes place in certain free- living bacteria and in symbiotic bacteria in the root nodules of leguminous plants. ■ The nitrogen cycle entails formation of ammonia by bacterial fixation of N 2 , nitrification of ammonia to nitrate by soil organisms, conversion of nitrate to ammonia by higher plants, synthesis of amino acids from ammonia by all organisms, and conversion of nitrate to N 2 by denitrifying soil bacteria. ■ Fixation of N 2 as NH 3 is carried out by the nitrogenase complex, in a reaction that requires ATP. The nitrogenase complex is highly labile in the presence of O 2 . ■ In living systems, reduced nitrogen is incorporated first into amino acids and then into a variety of other biomolecules, including nucleotides. The key entry point is the amino acid glutamate. Glutamate and glutamine are the nitrogen donors in a wide variety of biosynthetic reactions. Glutamine synthetase, which catalyzes the formation of glutamine from glutamate, is a main regulatory enzyme of nitrogen metabolism. ■ The amino acid and nucleotide biosynthetic pathways make repeated use of the biological cofactors pyridoxal phosphate, tetrahydrofolate, and S-adenosylmethionine. Pyridoxal phos- phate is required for transamination reactions involving glutamate and for other amino acid transformations. One-carbon transfers require S-adenosylmethionine and tetrahydrofolate. Glutamine amidotransferases catalyze reactions that incorporate nitrogen derived from glutamine. 22.2 Biosynthesis of Amino Acids All amino acids are derived from intermediates in gly- colysis, the citric acid cycle, or the pentose phosphate pathway (Fig. 22–9). Nitrogen enters these pathways by way of glutamate and glutamine. Some pathways are simple, others are not. Ten of the amino acids are just one or several steps removed from the common metabo- lite from which they are derived. The biosynthetic path- ways for others, such as the aromatic amino acids, are more complex. Organisms vary greatly in their ability to synthesize the 20 common amino acids. Whereas most bacteria and plants can synthesize all 20, mammals can synthesize only about half of them—generally those with simple pathways. These are the nonessential amino acids, not needed in the diet (see Table 18–1). The remain- der, the essential amino acids, must be obtained from food. Unless otherwise indicated, the pathways for the 20 common amino acids presented below are those op- erative in bacteria. 22.2 Biosynthesis of Amino Acids 841 Histidine Glutamine Proline Arginine Asparagine Methionine Threonine Lysine Pyruvate Tryptophan Phenylalanine Tyrosine Phosphoenolpyruvate Glycine Cysteine Serine 4 steps 4 steps Alanine Valine Leucine Isoleucine Erythrose 4- phosphate 3-Phosphoglycerate Glucose 6-phosphate Glucose GlutamateAspartate Oxaloacetate Ribose 5- phosphate -KetoglutarateH9251 Citrate FIGURE 22–9 Overview of amino acid biosynthesis. The carbon skeleton precursors derive from three sources: glycolysis (pink), the citric acid cycle (blue), and the pentose phosphate pathway (purple). 8885d_c22_833-880 2/6/04 8:35 AM Page 841 mac76 mac76:385_reb: A useful way to organize these biosynthetic path- ways is to group them into six families corresponding to their metabolic precursors (Table 22–1), and we use this approach to structure the detailed descriptions that follow. In addition to these six precursors, there is a notable intermediate in several pathways of amino acid and nucleotide synthesis—5-phosphoribosyl-1- pyrophosphate (PRPP): PRPP is synthesized from ribose 5-phosphate derived from the pentose phosphate pathway (see Fig. 14–21), in a reaction catalyzed by ribose phosphate pyro- phosphokinase: Ribose 5-phosphate H11001 ATP 88n 5-phosphoribosyl-1-pyrophosphate H11001 AMP This enzyme is allosterically regulated by many of the biomolecules for which PRPP is a precursor. H9251-Ketoglutarate Gives Rise to Glutamate, Glutamine, Proline, and Arginine We have already described the biosynthesis of gluta- mate and glutamine. Proline is a cyclized derivative of glutamate (Fig. 22–10). In the first step of proline synthesis, ATP reacts with the H9253-carboxyl group of glu- tamate to form an acyl phosphate, which is reduced by NADPH or NADH to glutamate H9253-semialdehyde. This intermediate undergoes rapid spontaneous cyclization and is then reduced further to yield proline. Arginine is synthesized from glutamate via or- nithine and the urea cycle in animals (Chapter 18). In principle, ornithine could also be synthesized from glu- tamate H9253-semialdehyde by transamination, but the spon- taneous cyclization of the semialdehyde in the proline pathway precludes a sufficient supply of this intermedi- ate for ornithine synthesis. Bacteria have a de novo biosynthetic pathway for ornithine (and thus arginine) that parallels some steps of the proline pathway but in- cludes two additional steps that avoid the problem of the spontaneous cyclization of glutamate H9253-semialdehyde (Fig. 22–10). In the first step, the H9251-amino group of glu- tamate is blocked by an acetylation requiring acetyl-CoA; Glutamate Glutamine Proline Arginine H9251-Ketoglutarate O P OH CH 2 O H11002 H11002 O H OH H H H O O O O PPO O O H11002 O H11002 O H11002 then, after the transamination step, the acetyl group is removed to yield ornithine. The pathways to proline and arginine are somewhat different in mammals. Proline can be synthesized by the pathway shown in Figure 22–10, but it is also formed from arginine obtained from dietary or tissue protein. Arginase, a urea cycle enzyme, converts arginine to orni- thine and urea (see Figs 18–10, 18–26). The ornithine is converted to glutamate H9253-semialdehyde by the enzyme ornithine H9254-aminotransferase (Fig. 22–11). The semi- aldehyde cyclizes to H9004 1 -pyrroline-5-carboxylate, which is then converted to proline (Fig. 22–10). The pathway for arginine synthesis shown in Figure 22–10 is absent in mammals. When arginine from dietary intake or pro- tein turnover is insufficient for protein synthesis, the ornithine H9254-aminotransferase reaction operates in the direction of ornithine formation. Ornithine is then con- verted to citrulline and arginine in the urea cycle. Serine, Glycine, and Cysteine Are Derived from 3-Phosphoglycerate The major pathway for the formation of serine is the same in all organisms (Fig. 22–12). In the first step, the hydroxyl group of 3-phosphoglycerate is oxidized by a 3-Phosphoglycerate Serine CysteineGlycine Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules842 TABLE 22–1 H9251-Ketoglutarate Glutamate Glutamine Proline Arginine 3-Phosphoglycerate Serine Glycine Cysteine Oxaloacetate Aspartate Asparagine Methionine* Threonine* Lysine* Pyruvate Alanine Valine* Leucine* Isoleucine* Phosphoenolpyruvate and erythrose 4-phosphate Tryptophan* Phenylalanine* Tyrosine ? Ribose 5-phosphate Histidine* Amino Acid Biosynthetic Families, Grouped by Metabolic Precursor *Essential amino acids. ? Derived from phenylalanine in mammals. 8885d_c22_833-880 2/6/04 8:35 AM Page 842 mac76 mac76:385_reb: ADP A O D B M O C H11002 O CH 2 COO H11002 CH S-CoA CoA-SH OO O O CH 3 O O C CH 2 acetylglutamate synthase Glutamate N-acetylglutamate dehydrogenase A O NH 3 COO H11002 CHOOCH 2 O B O CH 3 C A O D M CH 2 COO H11002 CH H11001 O O H 2 C O O C CH 2 O pyrroline carboxylate reductase NAD(P) H11001 A O D M H CH 2 COO H11002 CHOO NH 3 O CCH 2 O N-Acetylglutamate B O HN OCH 3 CO N-acetylglutamate kinase glutamate kinase N-Acetyl-H9253-glutamyl phosphate H9253-Glutamyl phosphate NAD(P)H NAD(P) H11001 H11001 H H11001 B O HN OCH 3 C A O D M O CH 2 COO H11002 CHOO O O CCH 2 O O P A O D M O CH 2 COO H11002 CH H11001 OO NH 3 O O CCH 2 O Glutamate H9253-semialdehyde (P5C) nonenzymatic N-Acetylornithine H 2 O CH 3 COO H11002 N-acetylornithinase H9004 1 -Pyrroline-5-carboxylate COO H11002 H11002 O OCH 2 H11001 NH 3 H C N H 2 CH COO H11002 H11001 O H 2 C CH 2 HC N H O CH glutamate dehydrogenase P i H f H B O HN OCH 3 C A O D M H CH 2 COO H11002 CHOO O CCH 2 O O N-Acetylglutamate H9253-semialdehyde Glutamate H9251-Ketoglutarate aminotransferase OCH 2 H 2 N H11001 OHN A O CH 2 COO H11002 CHOOCH 2 O H11001 OCH 2 H 3 N H11001 CH 2 Ornithine A O NH 3 COO H11002 CHOOCH 2 O H11001 OCH 2 H 3 N H11001 CH 2 G JH11001 H 2 N N H OC Proline Arginine ATP AMP H11001 PP i L-Citrulline argininosuccinate synthetase ATP H11001 aspartate P i ornithine carbamoyl- transferase Carbamoyl phosphate P NAD(P)H H11001 H H11001 ADP NAD(P) H11001 NAD(P)H H11001 H H11001 P i ATP Fumarate Argininosuccinate argininosuccinase Urea cycle FIGURE 22–10 Biosynthesis of proline and arginine from glutamate in bacteria. All five carbon atoms of proline arise from glutamate. In many organisms, glutamate dehydrogenase is unusual in that it uses either NADH or NADPH as a cofactor. The same may be true of other enzymes in these pathways. The H9253-semialdehyde in the proline path- way undergoes a rapid, reversible cyclization to H9004 1 -pyrroline-5- carboxylate (P5C), with the equilibrium favoring P5C formation. Cyclization is averted in the ornithine/arginine pathway by acetylation of the H9251-amino group of glutamate in the first step and removal of the acetyl group after the transamination. Although some bacteria lack arginase and thus the complete urea cycle, they can synthesize argi- nine from ornithine in steps that parallel the mammalian urea cycle, with citrulline and argininosuccinate as intermediates (see Fig. 18–10). Here, and in subsequent figures in this chapter, the reaction ar- rows indicate the linear path to the final products, without consider- ing the reversibility of individual steps. For example, the second step of the pathway leading to arginine, catalyzed by N-acetylglutamate dehydrogenase, is chemically similar to the glyceraldehyde 3-phos- phate dehydrogenase reaction in glycolysis (see Fig. 14–7) and is read- ily reversible. 843 8885d_c22_843 2/6/04 1:08 PM Page 843 mac76 mac76:385_reb: dehydrogenase (using NAD H11001 ) to yield 3-phosphohy- droxypyruvate. Transamination from glutamate yields 3- phosphoserine, which is hydrolyzed to free serine by phosphoserine phosphatase. Serine (three carbons) is the precursor of glycine (two carbons) through removal of a carbon atom by serine hydroxymethyltransferase (Fig. 22–12). Tetrahydrofolate accepts the H9252 carbon (C-3) of serine, which forms a methylene bridge between N-5 and N-10 to yield N 5 ,N 10 -methylenetetrahydrofolate (see Fig. 18–17). The overall reaction, which is reversible, also requires pyridoxal phosphate. In the liver of verte- brates, glycine can be made by another route: the re- verse of the reaction shown in Figure 18–20c, cat- alyzed by glycine synthase (also called glycine cleavage enzyme): CO 2 H11001 NH 4 H11001 H11001 N 5 ,N 10 -methylenetetrahydrofolate H11001 NADH H11001 H H11001 88n glycine H11001 tetrahydrofolate H11001 NAD H11001 Plants and bacteria produce the reduced sulfur re- quired for the synthesis of cysteine (and methionine, described later) from environmental sulfates; the path- way is shown on the right side of Figure 22–13. Sulfate is activated in two steps to produce 3-phosphoadeno- sine 5H11032-phosphosulfate (PAPS), which undergoes an eight-electron reduction to sulfide. The sulfide is then used in formation of cysteine from serine in a two-step pathway. Mammals synthesize cysteine from two amino acids: methionine furnishes the sulfur atom and serine furnishes the carbon skeleton. Methionine is first con- verted to S-adenosylmethionine (see Fig. 18–18), which can lose its methyl group to any of a number of accep- tors to form S-adenosylhomocysteine (adoHcy). This demethylated product is hydrolyzed to free homocys- teine, which undergoes a reaction with serine, catalyzed by cystathionine H9252-synthase, to yield cystathionine (Fig. 22–14). Finally, cystathionine H9253-lyase, a PLP- requiring enzyme, catalyzes removal of ammonia and cleavage of cystathionine to yield free cysteine. Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules844 Ornithine COO H11002 CHH 3 N CH 2 CH 2 CH 2 H11001 NH 3 H11001 ornithine H9254-aminotransferase -KetoglutarateH9251 Glutamate COO H11002 CHH 3 N CH 2 CH 2 C H O H11001 Glutamate H9253-semialdehyde H9004 1 -Pyrroline-5- carboxylate (P5C) H 2 C H 2 O H 2 O CH 2 CH CH COO H11002 N H H11001 FIGURE 22–11 Ornithine H9254-aminotransferase reaction: a step in the mammalian pathway to proline. This enzyme is found in the mitochondrial matrix of most tissues. Although the equilibrium favors P5C formation, the reverse reaction is the only mammalian pathway for synthesis of ornithine (and thus arginine) when arginine levels are insufficient for protein synthesis. FIGURE 22–12 Biosynthesis of serine from 3-phosphoglycerate and of glycine from serine in all organisms. Glycine is also made from CO 2 and NH 4 H11001 by the action of glycine synthase, with N 5 ,N 10 -methy- lenetetrahydrofolate as methyl group donor (see text). COO H11002 H C OH O H11001 CH 2 H H 3 N 3-Phosphoglycerate 3-Phosphohydroxypyruvate 3-Phosphoserine Glycine phosphoglycerate dehydrogenase phosphoserine aminotransferase phosphoserine phosphatase serine hydroxymethyl- transferase PLP H 4 folate N 5 ,N 10 -Methylene H 4 folate A O A C A H O OOH H O O COO H11002 C A A P O O COO H11002 C CH 2 NAD H11001 NADH H11001 H H11001 A A P OO O O COO H11002 C H11001 H 3 N Glutamate H9251-Ketoglutarate A A HOO COO H11002 C H11001 CH 2 OH H 3 N P i Serine A A HOO H 2 O H 2 O PO PO 8885d_c22_833-880 2/6/04 8:35 AM Page 844 mac76 mac76:385_reb: Three Nonessential and Six Essential Amino Acids Are Synthesized from Oxaloacetate and Pyruvate Alanine and aspartate are synthesized from pyruvate and oxaloacetate, respectively, by transamination from glutamate. Asparagine is synthesized by amidation of aspartate, with glutamine donating the NH 4 H11001 . These are nonessential amino acids, and their simple biosynthetic pathways occur in all organisms. Methionine, threonine, lysine, isoleucine, valine, and leucine are essential amino acids. Their biosynthetic pathways are complex and interconnected (Fig. 22–15). Aspartate Asparagine Methionine ThreonineLysine Oxaloacetate Isoleucine Pyruvate Alanine Valine Leucine 22.2 Biosynthesis of Amino Acids 845 ATP SO 4 2H11002 H11001 H9007 H11001 PP i ADP ATP CH 3 COO H11002 S 2H11002 H11001 H H11001 S 2H11002 3H11032-Phosphoadenosine 5H11032-phosphate (PAP) NADPH NADP H11001 ATP sulfurylase 3NADP H11001 3NADPH sulfide reductase PAPS reductase H11002 OSO O O H11002 P O O HH H OH OH H CH 2 O O H11002 Adenine Adenosine 5H11032-phosphosulfate (APS) H11002 OSO O O H11002 P O O HH H O SO 3 2H11002 P OH H CH 2 H11002 O O O O H11002 O H11002 Adenine 3H11032-Phosphoadenosine 5H11032-phosphosulfate (PAPS) Sulfite Sulfide COO H11002 CH 2 OH CHH 3 N H11001 Serine COO H11002 CH 2 SH CHH 3 N H11001 Cysteine COO H11002 CH 2 O CHH 3 N CH 3 CO H11001 O-Acetylserine CoA-SH H 3 CC serine acetyltransferase O-acetylserine (thiol) lyase S-CoA O FIGURE 22–13 Biosynthesis of cysteine from serine in bacteria and plants. The origin of reduced sulfur is shown in the pathway on the right. H11002 OOC HOCH 2 H11001 NH 3 H 2 O COO H11002 CH 2 SH HSC O CH CHCH 2 CH 2 S H11001 H11001 NH 3 Homocysteine Serine NH 4 H 2 O H11002 OOC H11001 NH 3 COO H11002 CH 2 CH CHCH 2 Cystathionine PLP CH 3 H11002 OOC H11001 NH 3 COO H11002 CHCH 2 CH 2 H11001 H9251-Ketobutyrate Cysteine cystathionine H9252-synthase cystathionine H9253-lyase H11001 H11001 NH 3 PLP FIGURE 22–14 Biosynthesis of cysteine from homocysteine and ser- ine in mammals. The homocysteine is formed from methionine, as de- scribed in the text. 8885d_c22_833-880 2/6/04 8:35 AM Page 845 mac76 mac76:385_reb: O C OH CH NH 2 CH 2 H11002 OOC NH 3 H11001 NADP H11001 NADPH H11001 H H11001 O H11002 O NH 3 H11001 Threonine O CHCH 3 COO H11002 P NH 3 H11001 ATP Aspartyl-H9252-phosphate O CH 2 C H C CH 2 CH COO H11002 NH 3 H11001 H COO H11002H11002 OOC Aspartate H9252-semialdehyde NADP NADPH H11001 H H11001 H COO H11002H11002 OOC N Dihydropicolinate H H 10 8 Pyruvate Pyruvate H11001 NH 3 COO H11002H11002 OOC N H9004 1 -Piperidine-2,6- dicarboxylate Lysine NH H 3 N 15 COO H11002H11002 OOC O N-Succinyl-2-amino- 6-keto-L-pimelate NH Succinate H9251-Ketoglutarate Glutamate NH 3 H11001 11 PLP COO H11002 COO H11002 10 H 2 O C H H H11001 N-Succinyl-L, L- H9251, H9280-diamino- pimelate (CH 2 ) 3 COO H11002 NH 3 H11001 L, L-H9251, H9280-Diamino- pimelate H 3 N COO H11002 C C H H H11001 COO H11002 meso-H9251, H9280-Diamino- pimelate H 3 N COO H11002 C C H H H11001 H 3 N H11001 H H11001 CH 2 CO 2 13 16 CAspartate CCHCH 2 COO H11002 NH 3 H11001 4 ADP CHCH 2 COO H11002 NH 3 H11001 CHCH H 2 O PLP CH 2 P Methionine ATP 1 ADP S CH COO H11002 NH 3 H11001 CH 2 OH CH 2 COO H11002 NH 3 H11001 CH 2 HomoserineCH 12 Succinyl-CoA H11001 CoA Succinate 6 Succinyl-CoA CoA CHCH 2 COO H11002 Succinate CH 2 O-Succinylhomoserine O H 2 O H COO H11002 NH 3 H11001 C CHCH 2 9 CH 2 CystathionineSH 2 C COO H11002 NH 3 H11001 Cysteine 7 Succinate CHCH 2 CH 2 HS COO H11002 NH 3 H11001 Homocysteine CH 2 3 COO H11002 O CH 2 Phosphohomoserine CH 3 OH NH 4 H11001 H11001 2 14 Succinate H 2 O N 5 -Methyl H 4 folate H 4 folate NH 3 H11001 5 P i H 2 O PLP P i NADPH H11001 H H11001 NADP H11001 O (CH 2 ) 3 (CH 2 ) 3 17 PLP PLP PLP Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules846 FIGURE 22–15 Biosynthesis of six essential amino acids from oxalo- acetate and pyruvate in bacteria: methionine, threonine, lysine, iso- leucine, valine, and leucine. Here, and in other multistep pathways, the enzymes are listed in the key. Note that L,L-H9251,H9280-diaminopimelate, the product of step 14 , is symmetric. The carbons derived from pyru- vate (and the amino group derived from glutamate) are not traced beyond this point, because subsequent reactions may place them at either end of the lysine molecule. 8885d_c22_833-880 2/6/04 8:35 AM Page 846 mac76 mac76:385_reb: 25 22 Glutamate CH 3 COO H11002 CO 2 OH NH 3 H11001 H9251-Ketoglutarate CH C Pyruvate 18 CH 3 H11002 Leucine O C CH 3 C CH 2 COO H11002 CH CO 2 NADH H11001 H H11001 NAD H11001 19 OH H9251-Acetolactate CH 3 CCH 3 COO H11002 O C OH H9252-Isopropylmalate CH 3 CH 3 O H9251-Isopropylmalate CH 3 COO H11002 21 19 Glutamate HO CH 3 COO H11002 H 2 O NAD(P)H H11001 H H11001 NAD(P) H11001 CH 3 O NH 3 H11001 H9251-Ketoglutarate C CH 2 C O COO H11002 CH 2 C CH 3 H9251-Ketobutyrate 18 Isoleucine CH 2 CH CH 3 CH 3 COO H11002 CH 19 OH H9251-Aceto-H9251- hydroxybutyrate CH 2 C CH 3 CH 3 COO H11002 O C H H9251, H9252-Dihydroxy- H9252-methylvalerate CH 2 C CH 3 CH 3 COO H11002 OH C 20 O H9251-Keto-H9252- methylvalerate CH 2 C CH 3 CH 3 COO H11002 H C C O COO H11002 CH 3 CH 3 TPP OH C PLP 24 Pyruvate PLP 21 19 Glutamate CH 3 COO H11002 NAD(P)H H11001 H H11001 NAD(P) H11001 O NH 3 H11001 H9251-Ketoglutarate C CH 3 C Valine CH 3 CHCH 3 COO H11002 CH OH H9251, H9252-Dihydroxy- isovalerate CH 3 CCH 3 COO H11002 OH C O H9251-Keto- isovalerate CH 3 CCH 3 COO H11002 H C 20 H 2 O PLP CH 3 CH CH 3 CH 2 CH CH CH CH COO H11002 COO H11002 CH 2 CH 3 COO H11002 H9251-Ketoisocaproate 23 CoA Acetyl -CoA COO H11002 OH H OH TPP 18 aspartokinase aspartate H9252-semialdehyde dehydrogenase homoserine dehydrogenase homoserine kinase threonine synthase homoserine acyltransferase cystathionine H9253-synthase cystathionine H9252-lyase methionine synthase dihydropicolinate synthase H9004 1 -piperidine-2,6-dicarboxylate dehydrogenase N-succinyl-2-amino-6-ketopimelate synthase succinyl diaminopimelate aminotransferase succinyl diaminopimelate desuccinylase diaminopimelate epimerase diaminopimelate decarboxylase threonine dehydratase (serine dehydratase) acetolactate synthase acetohydroxy acid isomeroreductase dihydroxy acid dehydratase valine aminotransferase H9251-isopropylmalate synthase isopropylmalate isomerase H9252-isopropylmalate dehydrogenase leucine aminotransferase 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 22.2 Biosynthesis of Amino Acids 847 8885d_c22_833-880 2/6/04 8:35 AM Page 847 mac76 mac76:385_reb: Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules848 O C CH 2 H 2 O COO H11002 CH P CH 2 O 3 O C CH 2 CHOH COO H11002 H11001 O C H P CHOH CH 2 O 7 H H HO OH OH C C H NAD H11001 P i 2 HO C H OH COO H11002 H HO Phosphoenolpyruvate (PEP) Erythrose 4-phosphate 2-Keto-3-deoxy-D- arabinoheptulosonate 7-phosphate 3-Dehydroquinate H 2 O P i 1 H OH H COO H11002 HO H NADP H11001 NADPH H11001 H H11001 4 HO H OH H COO H11002 HO H ADP 5 H OH H COO H11002 HO H PEP P i 6 HH COO H11002 HO H P i O O COO H11002 H 2 C C O CH 2 O C COO H11002 Shikimate Shikimate 3-phosphate 5-Enolpyruvylshikimate 3-phosphate Chorismate 3-Dehydroshikimate O O HO 3-Dehydroshikimate ATP P P P COO H11002 2-keto-3-deoxy-D-arabinoheptulosonate 7-phosphate synthase dehydroquinate synthase 3-dehydroquinate dehydratase shikimate dehydrogenase shikimate kinase 5-enolpyruvylshikimate 3-phosphate synthase chorismate synthase 1 2 3 4 5 6 7 to valine and isoleucine in pathways that begin with con- densation of two carbons of pyruvate (in the form of hydroxyethyl thiamine pyrophosphate; see Fig. 14–13) with another molecule of pyruvate (valine path) or with H9251-ketobutyrate (isoleucine path). The H9251-ketobutyrate is derived from threonine in a reaction that requires pyri- doxal phosphate (Fig. 22–15, step 17 ). An intermedi- ate in the valine pathway, H9251-ketoisovalerate, is the start- ing point for a four-step branch pathway leading to leucine (steps 22 to 25 ). FIGURE 22–16 Biosynthesis of chorismate, an intermediate in the synthesis of aromatic amino acids in bacteria and plants. All carbons are derived from either erythrose 4-phosphate (light purple) or phosphoenolpyruvate (pink). Note that the NAD H11001 required as a cofactor in step 2 is released unchanged; it may be transiently reduced to NADH during the reaction, with formation of an oxidized reaction intermediate. Step 6 is competitively inhibited by glyphosate ( H11002 COOOCH 2 ONHOCH 2 OPO 3 2H11002 ), the active ingredient in the widely used herbicide Roundup. The herbicide is relatively nontoxic to mammals, which lack this biosynthetic pathway. The chemical names quinate, shikimate, and chorismate are derived from the names of plants in which these intermediates have been found to accumulate. In some cases, the pathways in bacteria, fungi, and plants differ significantly. The bacterial pathways are outlined in Figure 22–15. Aspartate gives rise to methionine, threonine, and lysine. Branch points occur at aspartate H9252-semi- aldehyde, an intermediate in all three pathways, and at homoserine, a precursor of threonine and methionine. Threonine, in turn, is one of the precursors of isoleucine. The valine and isoleucine pathways share four en- zymes (Fig. 22–15, steps 18 to 21). Pyruvate gives rise 8885d_c22_848 2/6/04 1:08 PM Page 848 mac76 mac76:385_reb: 22.2 Biosynthesis of Amino Acids 849 OH PLP O B COO H11002 A H11001 H 2 O Serine PP i ! ' HO H H COO H11002 O C PRPP O O Glutamine Glutamate Pyruvate COO H11002 CH 2 H 2 O H11001 CO 2 5 1 NH 2 H HH H N H CH 2 O O O O P COO H11002 HO CH CH O HO OCOC O O CH 2 O O P A A B H A A A OH OH COO H11002 H N O POOCH 2 O O CH A OH OH N H NH 3 Glyceraldehyde 3-phosphate N H CH 2 O CH COO H11002 Tryptophan Indole-3-glycerol phosphate Enol-1-o-carboxyphenylamino-1- deoxyribulose phosphate N-(5H11032-Phosphoribosyl)- anthranilate Anthranilate Chorismate 2 C G H 3 4 e merase anthranilate synthase anthranilate phosphoribosyltransferase N-(5'-phosphoribosyl)-anthranilate isomerase indole-3-glycerol phosphate synthase tryptophan synthase 1 2 3 4 5 FIGURE 22–17 Biosynthesis of tryptophan from chorismate in bac- teria and plants. In E. coli, enzymes catalyzing steps 1 and 2 are subunits of a single complex. Chorismate Is a Key Intermediate in the Synthesis of Tryptophan, Phenylalanine, and Tyrosine Aromatic rings are not readily available in the environ- ment, even though the benzene ring is very stable. The branched pathway to tryptophan, phenylalanine, and ty- rosine, occurring in bacteria, fungi, and plants, is the main biological route of aromatic ring formation. It pro- ceeds through ring closure of an aliphatic precursor fol- lowed by stepwise addition of double bonds. The first four steps produce shikimate, a seven-carbon molecule derived from erythrose 4-phosphate and phospho- enolpyruvate (Fig. 22–16). Shikimate is converted to chorismate in three steps that include the addition of three more carbons from another molecule of phospho- enolpyruvate. Chorismate is the first branch point of the pathway, with one branch leading to tryptophan, the other to phenylalanine and tyrosine. In the tryptophan branch (Fig. 22–17), chorismate is converted to anthranilate in a reaction in which glu- tamine donates the nitrogen that will become part of the indole ring. Anthranilate then condenses with PRPP. The indole ring of tryptophan is derived from the ring carbons and amino group of anthranilate plus two car- bons derived from PRPP. The final reaction in the se- quence is catalyzed by tryptophan synthase. This en- zyme has an H9251 2 H9252 2 subunit structure and can be dissociated into two H9251 subunits and a H9252 2 subunit that catalyze different parts of the overall reaction: Indole-3-glycerol phosphate 88888n H9251 subunit indole H11001 glyceraldehyde 3-phosphate Indole H11001 serine 888888n tryptophan H11001 H 2 O H9252 2 subunit Phosphoenolpyruvate Erythrose 4-phosphate Phenylalanine Tyrosine Tryptophan Tyrosine H11001 8885d_c22_833-880 2/6/04 8:35 AM Page 849 mac76 mac76:385_reb: Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules850 CH 3 O COO H11002 P H11001 NH H11001 NH H11001 NH H Glyceraldehyde 3-phosphate N H CH CH CH 2 P Indole-3-glycerol phosphate O H C OH NH 3 CH CH Indole H11001 N H COO H11002 C PLP CH 2 P O O C H O H H 2 O CH 2 PLP H9251 tryptophan synthase H9251 subunits H9251 Quinonoid intermediate CH 3 COO H11002 P HO C H11001 H9252 N H CH 2 C H B O H Indole H 2 C H9252 Aldimine with tryptophan CH 2 H11001 N H N H H N H CH 3 P HO C N H CH 2 O H H9252 PLP-aminoacrylate adduct H11001 COO H11002 N H CH 2 C H9252 H11001 H OH Serine NH 3 CH H11001 COO H11002 H9252 CH 2 H 2 O N H Tryptophan CH 2 tryptophan synthase H9252 subunits Enzyme OH OH 1 2 3 5 4 B HB H9252H9252 MECHANISM FIGURE 22–18 Tryptophan synthase reaction. This en- zyme catalyzes a multistep reaction with several types of chemical re- arrangements. 1 An aldol cleavage produces indole and glyceralde- hyde 3-phosphate; this reaction does not require PLP. 2 Dehydration of serine forms a PLP-aminoacrylate intermediate. In steps 3 and 4 this condenses with indole, and 5 the product is hydrolyzed to re- lease tryptophan. These PLP-facilitated transformations occur at the H9252 carbon (C-3) of the amino acid, as opposed to the H9251-carbon reactions described in Figure 18–6. The H9252 carbon of serine is attached to the in- dole ring system. Tryptophan Synthase Mechanism The second part of the reaction requires pyridoxal phos- phate (Fig. 22–18). Indole formed in the first part is not released by the enzyme, but instead moves through a channel from the H9251-subunit active site to the H9252-subunit active site, where it condenses with a Schiff base inter- mediate derived from serine and PLP. Intermediate channeling of this type may be a feature of the entire pathway from chorismate to tryptophan. Enzyme active sites catalyzing different steps (sometimes not sequen- tial steps) of the pathway to tryptophan are found on single polypeptides in some species of fungi and bacte- ria, but are separate proteins in others. In addition, the activity of some of these enzymes requires a noncova- lent association with other enzymes of the pathway. These observations suggest that all the pathway en- zymes are components of a large, multienzyme complex in both prokaryotes and eukaryotes. Such complexes are generally not preserved intact when the enzymes are isolated using traditional biochemical methods, but ev- idence for the existence of multienzyme complexes is accumulating for this and a number of other metabolic pathways (p. 605). 8885d_c22_833-880 2/6/04 8:35 AM Page 850 mac76 mac76:385_reb: (see Figs 18–23, 18–24). Tyrosine is considered a con- ditionally essential amino acid, or as nonessential inso- far as it can be synthesized from the essential amino acid phenylalanine. Histidine Biosynthesis Uses Precursors of Purine Biosynthesis The pathway to histidine in all plants and bacteria dif- fers in several respects from other amino acid biosyn- thetic pathways. Histidine is derived from three pre- cursors (Fig. 22–20): PRPP contributes five carbons, the purine ring of ATP contributes a nitrogen and a carbon, and glutamine supplies the second ring nitrogen. The key steps are condensation of ATP and PRPP, in which N-1 of the purine ring is linked to the activated C-1 of the ribose of PRPP (step 1 in Fig. 22–20); purine ring opening that ultimately leaves N-1 and C-2 of adenine linked to the ribose (step 3 ); and formation of the im- idazole ring, a reaction in which glutamine donates a ni- trogen (step 5 ). The use of ATP as a metabolite rather than a high-energy cofactor is unusual— but not waste- ful, because it dovetails with the purine biosynthetic pathway. The remnant of ATP that is released after the transfer of N-1 and C-2 is 5-aminoimidazole-4-carbox- amide ribonucleotide (AICAR), an intermediate of purine biosynthesis (see Fig. 22–33) that is rapidly re- cycled to ATP. Amino Acid Biosynthesis Is under Allosteric Regulation The most responsive regulation of amino acid synthesis takes place through feedback inhibition of the first re- action in a sequence by the end product of the pathway. This first reaction is usually irreversible and catalyzed by an allosteric enzyme. As an example, Figure 22–21 shows the allosteric regulation of isoleucine synthesis from threonine (detailed in Fig. 22–15). The end prod- uct, isoleucine, is an allosteric inhibitor of the first reaction in the sequence. In bacteria, such allosteric modulation of amino acid synthesis occurs as a minute- to-minute response. Allosteric regulation can be considerably more com- plex. An example is the remarkable set of allosteric con- trols exerted on glutamine synthetase of E. coli (Fig. 22–6). Six products derived from glutamine serve as negative feedback modulators of the enzyme, and the overall effects of these and other modulators are more than additive. Such regulation is called concerted in- hibition. Ribose 5-phosphate Histidine 22.2 Biosynthesis of Amino Acids 851 FIGURE 22–19 Biosynthesis of phenylalanine and tyrosine from cho- rismate in bacteria and plants. Conversion of chorismate to prephen- ate is a rare biological example of a Claisen rearrangement. Phenylpyruvate O HO H COO H11002 amino- transferase 2 CH 2 H CO 2 H11001 OH H11002 O COO H11002 Prephenate H11002 OOC 1 COO H11002 H9251-Ketoglutarate C HO H Chorismate NADH H11001 H H11001 NAD H11001 C CH 2 COO H11002 O C CH COO H11002 O C 4-Hydroxyphenyl- pyruvate 3 CH 2 Phenylalanine CO 2 COO H11002 CH 2 CH 2 Glutamate OH Tyrosine amino- transferase COO H11002 H9251-Ketoglutarate CHCH 2 OH Glutamate NH 3 H11001H11001 NH 3 chorismate mutase prephenate dehydrogenase prephenate dehydratase 1 2 3 In plants and bacteria, phenylalanine and tyro- sine are synthesized from chorismate in pathways much less complex than the tryptophan pathway. The com- mon intermediate is prephenate (Fig. 22–19). The final step in both cases is transamination with glutamate. Animals can produce tyrosine directly from phenyl- alanine through hydroxylation at C-4 of the phenyl group by phenylalanine hydroxylase; this enzyme also participates in the degradation of phenylalanine 8885d_c22_851 2/6/04 1:08 PM Page 851 mac76 mac76:385_reb: Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules852 H C O O HC HHC OH C H O P O H11001 5-Phosphoribosyl- 1-pyrophosphate (PRPP) N HC N N NH 2 N P ATP 5 PP i P P P H C O CH 2 O HC H H C C O P H PRib P P 5-Aminoimidazole- 4-carboxamide ribonucleotide (AICAR) NH 2 3 H C O CH 2 O H C HHC OH C O P PRibNN HN C H N N H N 1 -5H11032-Phosphoribosyl-ATP H 2 O 7 H C O CH 2 O H C HHC OH C O P PRibNN HN C H N N H N 1 -5H11032-Phosphoribosyl-AMP RibNN H H 2 O C O C O H 2 N PRibNN CN N H H C C H HH C HO CH 2 O P N 1 -5H11032-Phosphoribosylformimino- 5-aminoimidazole-4- carboxamide ribonucleotide H 2 N RibNN NC O P H 2 N To purine biosynthesis C N OH P NH 3 HC H COH O N 1 -5H11032-Phosphoribulosyl- formimino-5-amino- imidazole-4-carboxamide ribonucleotide Histidine 4 Glutamate H9251-Ketoglutarate 1 OH PCH 2 HC HCOH O HC C H N N CH Imidazole glycerol 3-phosphate COO H11002 O P CH 2 C O HC C N Imidazole acetol 3-phosphate CH 2 HC CH 2 Glutamine Glutamate 6 9 2NADH H11001 2H H11001 NC H P C O C N NH 3 L-Histidinol phosphate CH 2 HC H 2 PP i 8 P i 2NAD H11001 H C O C N NH 3 L-Histidinol CH 2 HC H H11001H11001 H11001 H N CH H N CH CH 2 H N CH CH 2 H N CH CH 2 ATP phosphoribosyl transferase pyrophosphohydrolase phosphoribosyl-AMP cyclohydrolase phosphoribosylformimino-5-aminoimidazole- 4-carboxamide ribonucleotide isomerase glutamine amidotransferase imidazole glycerol 3-phosphate dehydratase L-histidinol phosphate aminotransferase histidinol phosphate phosphatase histidinol dehydrogenase 1 2 3 4 5 6 7 8 9 FIGURE 22–20 Biosynthesis of histidine in bacteria and plants. Atoms derived from PRPP and ATP are shaded red and blue, respectively. Two of the histidine nitrogens are derived from glutamine and glutamate (green). Note that the derivative of ATP remaining after step 5 (AICAR) is an intermediate in purine biosynthesis (see Fig. 22–33, step 9 ), so ATP is rapidly regenerated. 8885d_c22_833-880 2/6/04 8:35 AM Page 852 mac76 mac76:385_reb: Because the 20 common amino acids must be made in the correct proportions for protein synthesis, cells have developed ways not only of controlling the rate of synthesis of individual amino acids but also of coordi- nating their formation. Such coordination is especially well developed in fast-growing bacterial cells. Figure 22–22 shows how E. coli cells coordinate the synthesis of lysine, methionine, threonine, and isoleucine, all made from aspartate. Several important types of inhibi- tion patterns are evident. The step from aspartate to aspartyl-H9252-phosphate is catalyzed by three isozymes, each independently controlled by different modulators. This enzyme multiplicity prevents one biosynthetic end product from shutting down key steps in a pathway when other products of the same pathway are required. The steps from aspartate H9252-semialdehyde to homoser- ine and from threonine to H9251-ketobutyrate (detailed in Fig. 22–15) are also catalyzed by dual, independently controlled isozymes. One isozyme for the conversion of aspartate to aspartyl-H9252-phosphate is allosterically inhib- ited by two different modulators, lysine and isoleucine, whose action is more than additive—another example of concerted inhibition. The sequence from aspartate to isoleucine undergoes multiple, overlapping negative feedback inhibition; for example, isoleucine inhibits the conversion of threonine to H9251-ketobutyrate (as de- scribed above), and threonine inhibits its own forma- tion at three points: from homoserine, from aspartate H9252-semialdehyde, and from aspartate (steps 4 , 3 , and 1 in Fig. 22–15). This overall regulatory mechanism is called sequential feedback inhibition. Similar patterns are evident in the pathways lead- ing to the aromatic amino acids. The first step of the early pathway to the common intermediate chorismate is catalyzed by the enzyme 2-keto-3-deoxy-D-arabino- heptulosonate 7-phosphate (DAHP) synthase (step 1 in Fig. 22–16). Most microorganisms and plants have three DAHP synthase isozymes. One is allosterically in- hibited (feedback inhibition) by phenylalanine, another by tyrosine, and the third by tryptophan. This scheme helps the overall pathway to respond to cellular 22.2 Biosynthesis of Amino Acids 853 COO H11002 H11001 NH 3 CH 2 H9251-Ketobutyrate OO CH A A O O Isoleucine OCH OH CH 3 CH 3 CH COO H11002 O threonine dehydratase COO H11002 CH 2 OO O OCH 3 B C A CH 3 OCH H11001 NH 3 A Threonine 5 steps FIGURE 22–21 Allosteric regulation of isoleucine biosynthesis. The first reaction in the pathway from threonine to isoleucine is inhibited by the end product, isoleucine. This was one of the first examples of allosteric feedback inhibition to be discovered. The steps from H9251- ketobutyrate to isoleucine correspond to steps 18 through 21 in Fig- ure 22–15 (five steps because 19 is a two-step reaction). FIGURE 22–22 Interlocking regulatory mechanisms in the biosyn- thesis of several amino acids derived from aspartate in E. coli. Three enzymes (A, B, C) have either two or three isozyme forms, indicated by numerical subscripts. In each case, one isozyme (A 2 , B 1 , and C 2 ) has no allosteric regulation; these isozymes are regulated by changes in the amount synthesized (Chapter 28). Synthesis of isozymes A 2 and B 1 is repressed when methionine levels are high, and synthesis of isozyme C 2 is repressed when isoleucine levels are high. Enzyme A is aspartokinase; B, homoserine dehydrogenase; C, threonine dehy- dratase. Aspartate b-semialdehyde Lysine Methionine B 1 B 2 A 2 A 3 A 1 Aspartate Threonine C 1 C 2 a-Ketobutyrate Isoleucine Aspartyl-b-phosphate Homoserine 5 steps 3 steps 6 steps 8885d_c22_833-880 2/6/04 8:35 AM Page 853 mac76 mac76:385_reb: Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules854 requirements for one or more of the aromatic amino acids. Additional regulation takes place after the path- way branches at chorismate. For example, the enzymes catalyzing the first two steps of the tryptophan branch are subject to allosteric inhibition by tryptophan. SUMMARY 22.2 Biosynthesis of Amino Acids ■ Plants and bacteria synthesize all 20 common amino acids. Mammals can synthesize about half; the others are required in the diet (essential amino acids). ■ Among the nonessential amino acids, glutamate is formed by reductive amination of H9251-ketoglutarate and serves as the precursor of glutamine, proline, and arginine. Alanine and aspartate (and thus asparagine) are formed from pyruvate and oxaloacetate, respectively, by transamination. The carbon chain of serine is derived from 3-phosphoglycerate. Serine is a precursor of glycine; the H9252-carbon atom of serine is transferred to tetrahydrofolate. In microorganisms, cysteine is produced from serine and from sulfide produced by the reduction of environmental sulfate. Mammals produce cysteine from methionine and serine by a series of reactions requiring S-adenosylmethionine and cystathionine. ■ Among the essential amino acids, the aromatic amino acids (phenylalanine, tyrosine, and tryptophan) form by a pathway in which chorismate occupies a key branch point. Phosphoribosyl pyrophosphate is a precursor of tryptophan and histidine. The pathway to histidine is interconnected with the purine synthetic pathway. Tyrosine can also be formed by hydroxylation of phenylalanine (and thus is considered conditionally essential). The pathways for the other essential amino acids are complex. ■ The amino acid biosynthetic pathways are subject to allosteric end-product inhibition; the regulatory enzyme is usually the first in the sequence. Regulation of the various synthetic pathways is coordinated. 22.3 Molecules Derived from Amino Acids In addition to their role as the building blocks of proteins, amino acids are precursors of many specialized biomol- ecules, including hormones, coenzymes, nucleotides, alkaloids, cell wall polymers, porphyrins, antibiotics, pigments, and neurotransmitters. We describe here the pathways to a number of these amino acid derivatives. Glycine Is a Precursor of Porphyrins The biosynthesis of porphyrins, for which glycine is a major precursor, is our first example, because of the central importance of the porphyrin nucleus in heme proteins such as hemoglobin and the cytochromes. The porphyrins are constructed from four molecules of the monopyrrole derivative porphobilinogen, which itself is derived from two molecules of H9254-aminolevulinate. There are two major pathways to H9254-aminolevulinate. In higher eukaryotes (Fig. 22–23a), glycine reacts with succinyl- CoA in the first step to yield H9251-amino-H9252-ketoadipate, which is then decarboxylated to H9254-aminolevulinate. In plants, algae, and most bacteria, H9254-aminolevulinate is formed from glutamate (Fig. 22–23b). The glutamate is first esterified to glutamyl-tRNA Glu (see Chapter 27 on the topic of transfer RNAs); reduction by NADPH con- verts the glutamate to glutamate 1-semialdehyde, which is cleaved from the tRNA. An aminotransferase converts the glutamate 1-semialdehyde to H9254-aminolevulinate. In all organisms, two molecules of H9254-aminolevulinate condense to form porphobilinogen and, through a series of complex enzymatic reactions, four molecules of por- phobilinogen come together to form protoporphyrin (Fig. 22–24). The iron atom is incorporated after the protoporphyrin has been assembled, in a step catalyzed by ferrochelatase. Porphyrin biosynthesis is regulated in higher eukaryotes by the concentration of the heme product, which serves as a feedback inhibitor of early steps in the synthetic pathway. Genetic defects in the biosynthesis of porphyrins can lead to the accumulation of pathway intermediates, causing a variety of human diseases known collectively as porphyrias (Box 22–1). Heme Is the Source of Bile Pigments The iron-porphyrin (heme) group of hemoglobin, released from dying erythrocytes in the spleen, is degraded to yield free Fe 3H11001 and, ultimately, biliru- bin. This pathway is arresting for its capacity to inject color into human biochemistry. The first step in the two-step pathway, catalyzed by heme oxygenase (HO), converts heme to biliverdin, a linear (open) tetrapyrrole derivative (Fig. 22–25). The other products of the reaction are free Fe 2H11001 and CO. The Fe 2H11001 is quickly bound by ferritin. Carbon monox- ide is a poison that binds to hemoglobin (see Box 5–1), and the production of CO by heme oxygenase ensures that, even in the absence of environmental exposure, about 1% of an individual’s heme is complexed with CO. Biliverdin is converted to bilirubin in the second step, catalyzed by biliverdin reductase. You can monitor this reaction colorimetrically in a familiar in situ exper- iment. When you are bruised, the black and/or purple color results from hemoglobin released from damaged erythrocytes. Over time, the color changes to the green of biliverdin, and then to the yellow of bilirubin. Biliru- 8885d_c22_833-880 2/6/04 8:35 AM Page 854 mac76 mac76:385_reb: H11001 H11001 Glycine Glutamate CH 2 CO 2 COO H11002 Succinyl-CoA NH 3 H11001 CH 2 COO H11002 CH 2 CH H11001 C NH 3 O H9251-Amino-H9252- ketoadipate COO H11002 CH 2 S-CoA COO H11002 CH 2 CH 2 COO H11002 CH 2 C O CH 2 CH 2 C COO H11002 O CH 2 H9254-Aminolevulinate H11001 NH 3 CoA-SH H11001 NH 3 HC COO H11002 CH 2 COO H11002 CH 2 H11001 NH 3 HC CH 2 COO H11002 CH 2 H11001 NH 3 HC C ATP AMP PP i tRNA Glu tRNA Glu NADPH NADP H11001 tRNA Glu O C H O Glutamyl-tRNA Glu Glutamate 1-semialdehyde H9254-aminolevulinate synthase H9254-aminolevulinate synthase glutamyl-tRNA synthetase glutamyl-tRNA reductase glutamate-1-semialdehyde aminomutase (a) (b) 22.3 Molecules Derived from Amino Acids 855 FIGURE 22–23 Biosynthesis of H9254-aminolevulinate. (a) In mammals and other higher eukaryotes, H9254-aminolevulinate is synthesized from glycine and succinyl-CoA. The atoms furnished by glycine are shown in red. (b) In bacteria and plants, the precursor of H9254-aminolevulinate is glutamate. CH 3 CH 3 COO H11002 H11002 OOC COO H11002 Fe 2H11001 Fe 2H11001 CH 3 NH 3 H 3 N 8 H 2 O Pr Pr CH 3 CH 3 CH 3 CH 3 CH 3 NN N N O N H Heme CH 3 CH 3 CH 3 Pr Pr CH 3 N N Protoporphyrin Pr P NH NH HN HN Pr Pr Ac Ac Ac Ac HO Preuroporphyrinogen PrPr NH NH HN HN Pr Pr AcAc Ac Ac Uroporphyrinogen III PrPr NH NH HN HN Pr Pr Coproporphyrinogen III CH 3 CH 3 CH 3 CH 3 PrPr NH NH HN HN HN HN Protoporphyrinogen 1 6 2 84 H9254-Aminolevulinate 4 NH 4 H11001 3 H 2 O 5 4 4 CO 2 2 CO 2 7 Porphobilinogen Vinyl group porphobilinogen synthase uroporphyrinogen synthase uroporphyrinogen III cosynthase uroporphyrinogen decarboxylase coproporphyrinogen oxidase protoporphyrinogen oxidase ferrochelatase 1 2 3 4 5 6 7 FIGURE 22–24 Biosynthesis of heme from H9254-aminolevulinate. Ac rep- resents acetyl (OCH 2 COO H11002 ); Pr, propionyl (OCH 2 CH 2 COO H11002 ). 8885d_c22_833-880 2/6/04 8:35 AM Page 855 mac76 mac76:385_reb: bin is largely insoluble, and it travels in the bloodstream as a complex with serum albumin. In the liver, bilirubin is transformed to the bile pigment bilirubin diglu- curonide. This product is sufficiently water-soluble to be secreted with other components of bile into the small intestine, where microbial enzymes convert it to several products, predominantly urobilinogen. Some urobilino- gen is reabsorbed into the blood and transported to the kidney, where it is converted to urobilin, the compound that gives urine its yellow color (Fig. 22–25, left branch). Urobilinogen remaining in the intestine is converted (in another microbe-dependent reaction) to stercobilin (Fig. 22–25, right branch), which imparts the red-brown color to feces. Impaired liver function or blocked bile secretion causes bilirubin to leak from the liver into the blood, resulting in a yellowing of the skin and eyeballs, a con- dition called jaundice. In cases of jaundice, determina- tion of the concentration of bilirubin in the blood may be useful in the diagnosis of underlying liver disease. Newborn infants sometimes develop jaundice because they have not yet produced enough glucuronyl bilirubin transferase to process their bilirubin. A traditional treatment to reduce excess bilirubin, exposure to a fluo- rescent lamp, causes a photochemical conversion of bilirubin to compounds that are more soluble and easily excreted. These pathways of heme breakdown play significant roles in protecting cells from oxidative damage and in regulating certain cellular functions. The CO produced by heme oxygenase is toxic at high concentrations, but at the very low concentrations generated during heme degradation it appears to have some regulatory and/or signaling functions. It acts as a vasodilator, much the Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules856 FIGURE 22–25 Bilirubin and its breakdown products. M represents methyl; V, vinyl; Pr, propionyl; E, ethyl. For ease of comparison, these structures are shown in linear form, rather than in their correct stereochemical conformations. transport to intestine heme oxygenase CO Fe H110012 Heme O M V N H N H M Pr N MPr O M V N H O M V N H N H M Pr N H MPr H H O M V N H biliverdin reductase NADPH, H NADP H11001 Biliverdin H11001 Bilirubin (in blood) Bilirubin diglucuronide Bilirubin (in bile) transport to kidney Urobilinogen Urobilinogen (liver) transport in blood as complex with serum albumin glucuronyl- bilirubin transferase O M E N H N H M Pr N MPr O M E N H H O M E N H N H M Pr N MPr O M E N H H H H HH Urobilin Stercobilin 8885d_c22_833-880 2/6/04 8:35 AM Page 856 mac76 mac76:385_reb: same as (but less potent than) nitric oxide (discussed below). Low levels of CO also have some regulatory ef- fects on neurotransmission. Bilirubin is the most abun- dant antioxidant in mammalian tissues and is responsi- ble for most of the antioxidant activity in serum. Its protective effects appear to be especially important in the developing brain of newborn infants. The cell toxi- city associated with jaundice may be due to bilirubin levels in excess of the serum albumin needed to solu- bilize it. Given these varied roles of heme degradation prod- ucts, the degradative pathway is subject to regulation, mainly at the first step. Humans have at least three isozymes of heme oxygenase. HO-1 is highly regulated; the expression of its gene is induced by a wide range of stress conditions (shear stress, angiogenesis (uncon- trolled development of blood vessels), hypoxia, hyper- oxia, heat shock, exposure to ultraviolet light, hydrogen peroxide, and many other metabolic insults). HO-2 is found mainly in brain and testes, where it is continu- ously expressed. The third isozyme, HO-3, is not yet well characterized. ■ Amino Acids Are Precursors of Creatine and Glutathione Phosphocreatine, derived from creatine, is an im- portant energy buffer in skeletal muscle (see Fig. 13–5). Creatine is synthesized from glycine and arginine (Fig. 22–26); methionine, in the form of S-adenosylmethionine, acts as methyl group donor. Glutathione (GSH), present in plants, animals, and some bacteria, often at high levels, can be thought of as a redox buffer. It is derived from glycine, gluta- mate, and cysteine (Fig. 22–27). The H9253-carboxyl group of glutamate is activated by ATP to form an acyl phos- phate intermediate, which is then attacked by the H9251- amino group of cysteine. A second condensation reac- tion follows, with the H9251-carboxyl group of cysteine activated to an acyl phosphate to permit reaction with glycine. The oxidized form of glutathione (GSSG), pro- duced in the course of its redox activities, contains two glutathione molecules linked by a disulfide bond. Glutathione probably helps maintain the sulfhydryl groups of proteins in the reduced state and the iron of 22.3 Molecules Derived from Amino Acids 857 BOX 22–1 BIOCHEMISTRY IN MEDICINE Biochemistry of Kings and Vampires Porphyrias (listed at right) are a group of genetic dis- eases in which, because of defects in enzymes of the biosynthetic pathway from glycine to porphyrins, spe- cific porphyrin precursors accumulate in erythrocytes, body fluids, and the liver. The most common form is acute intermittent porphyria. Most affected individu- als are heterozygotes and are usually asymptomatic, because the single copy of the normal gene provides a sufficient level of enzyme function. However, certain nutritional or environmental factors (as yet poorly understood) can cause a buildup of H9254-aminolevulinate and porphobilinogen, leading to attacks of acute abdominal pain and neurological dysfunction. King George III, British monarch during the American Revolution, suffered several episodes of apparent madness that tarnished the record of this otherwise accomplished man. The symptoms of his condition suggest that George III suffered from acute intermit- tent porphyria. One of the rarer porphyrias results in an accu- mulation of uroporphyrinogen I, an abnormal isomer of a protoporphyrin precursor. This compound stains the urine red, causes the teeth to fluoresce strongly in ultraviolet light, and makes the skin abnormally sen- sitive to sunlight. Many individuals with this porphyria are anemic, because insufficient heme is synthesized. This genetic condition may have given rise to the vam- pire myths of folk legend. The symptoms of most porphyrias are now read- ily controlled with dietary changes or the administra- tion of heme or heme derivatives. H9254-Aminolevulinate Porphobilinogen Preuroporphyrinogen Urophorphyrinogen III Coproporphyrinogen Protoporphyrinogen Protoporphyrin Heme Doss porphyria Acute intermittent porphyria Congenital erythropoietic porphyria Porphyria cutanea tarda Hereditary coproporphyria Variegate porphyria Erythropoietic protoporphyria 8885d_c22_833-880 2/6/04 8:35 AM Page 857 mac76 mac76:385_reb: bound selenium (Se) atom in the form of selenocysteine (see Fig. 3–8a), which is essential for its activity. D-Amino Acids Are Found Primarily in Bacteria Although D-amino acids do not generally occur in proteins, they do serve some special functions in the structure of bacterial cell walls and peptide an- tibiotics. Bacterial peptidoglycans (see Fig. 20–23) con- tain both D-alanine and D-glutamate. D-Amino acids arise directly from the L isomers by the action of amino acid racemases, which have pyridoxal phosphate as cofactor (see Fig. 18–6). Amino acid racemization is uniquely important to bacterial metabolism, and enzymes such as Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules858 ATP Glutamate Cysteine H9253-Glu–Cys–Gly glutathione synthetase H9253-Glu Cys Gly O CCH N SH H11001 CH 2 (reduced) Glutathione (GSH) Glutathione (GSSG) (oxidized) Glycine CH CH 2 CH 2 C O N HH CH 2 NH 3 ATP ADP H11001 P i H5008 OOC COO H5008 ADP H11001 P i H9253-Glu–Cys–Gly H9253-Glu–Cys H9253-glutamyl cysteine synthetase (b) (a) S S FIGURE 22–27 Glutathione metabolism. (a) Biosynthesis of glu- tathione. (b) Reduced form of glutathione. heme in the ferrous (Fe 2H11001 ) state, and it serves as a re- ducing agent for glutaredoxin in deoxyribonucleotide synthesis (see Fig. 22–39). Its redox function is also used to remove toxic peroxides formed in the normal course of growth and metabolism under aerobic conditions: 2GSH H11001 ROOOOOH 88n GSSG H11001 H 2 O H11001 ROOH This reaction is catalyzed by glutathione peroxidase, a remarkable enzyme in that it contains a covalently A OP CH 2 H11001 N COO H11002 C A O H11002 A A NH 3 H11001 NH 3 A CH 3 A CH 2 A A CH H11001 NH 2 Arginine Glycine Ornithine Guanidinoacetate A P NH C A NH 2 A (CH 2 ) 3 H11001 NH 2 COO H11002 adoMet adoHcy Methionine Creatine Phosphocreatine methyltransferase creatine kinase amidinotransferase A P NH C A COO H11002 NH 2 A CH 2 A H11001 NH 2 O NH 2 ADP A A A A A A O P P O COO H11002 OP NH C O H11002 N CH 2 COO H11002 CH 3 NH 2 H11001 ATP FIGURE 22–26 Biosynthesis of creatine and phosphocreatine. Crea- tine is made from three amino acids: glycine, arginine, and methion- ine. This pathway shows the versatility of amino acids as precursors of other nitrogenous biomolecules. 8885d_c22_858 2/6/04 1:08 PM Page 858 mac76 mac76:385_reb: alanine racemase are prime targets for pharmaceutical agents. One such agent, L-fluoroalanine, is being tested as an antibacterial drug. Another, cycloserine, is used to treat tuberculosis. Because these inhibitors also affect some PLP-requiring human enzymes, how- ever, they have potentially undesirable side effects. ■ Aromatic Amino Acids Are Precursors of Many Plant Substances Phenylalanine, tyrosine, and tryptophan are converted to a variety of important compounds in plants. The rigid polymer lignin, derived from phenylalanine and tyro- sine, is second only to cellulose in abundance in plant tissues. The structure of the lignin polymer is complex and not well understood. Tryptophan is also the pre- cursor of the plant growth hormone indole-3-acetate, or auxin (Fig. 22–28a), which has been implicated in the regulation of a wide range of biological processes in plants. Phenylalanine and tyrosine also give rise to many commercially significant natural products, including the tannins that inhibit oxidation in wines; alkaloids such as morphine, which have potent physiological effects; and the flavoring of cinnamon oil (Fig. 22–28b), nutmeg, cloves, vanilla, cayenne pepper, and other products. Biological Amines Are Products of Amino Acid Decarboxylation Many important neurotransmitters are primary or secondary amines, derived from amino acids in simple pathways. In addition, some polyamines that form complexes with DNA are derived from the amino acid ornithine, a component of the urea cycle. A com- mon denominator of many of these pathways is amino acid decarboxylation, another PLP-requiring reaction (see Fig. 18–6). The synthesis of some neurotransmitters is illus- trated in Figure 22–29. Tyrosine gives rise to a family of catecholamines that includes dopamine, norepineph- rine, and epinephrine. Levels of catecholamines are correlated with, among other things, changes in blood pressure. The neurological disorder Parkinson’s disease is associated with an underproduction of dopamine, and it has traditionally been treated by administering L-dopa. Overproduction of dopamine in the brain may be linked to psychological disorders such as schizophrenia. Glutamate decarboxylation gives rise to H9253-amino- butyrate (GABA), an inhibitory neurotransmitter. Its underproduction is associated with epileptic seizures. H 3 NH HC O CH 2 NH 3 H11001 COO C OF C H11001 L-Fluoroalanine Cycloserine H5008 H 2 NCH GABA analogs are used in the treatment of epilepsy and hypertension. Levels of GABA can also be increased by administering inhibitors of the GABA-degrading enzyme GABA aminotransferase. Another important neuro- transmitter, serotonin, is derived from tryptophan in a two-step pathway. Histidine undergoes decarboxylation to histamine, a powerful vasodilator in animal tissues. Histamine is re- leased in large amounts as part of the allergic response, and it also stimulates acid secretion in the stomach. A growing array of pharmaceutical agents are being de- signed to interfere with either the synthesis or the ac- tion of histamine. A prominent example is the histamine receptor antagonist cimetidine (Tagamet), a structural analog of histamine: It promotes the healing of duodenal ulcers by inhibiting secretion of gastric acid. CH 3 C NHCH 2 N CH 2 CH 2 NH SNH CH 3 N C N 22.3 Molecules Derived from Amino Acids 859 COO H5008 CH C O CH 2 NH 3 H11001 COO H5008 CHCH COO H5008 amino- transferase decarboxylase CH 2 N H COO H5008 CH 2 CH NH 3 H11001 COO H5008 CH 2 N H N H Tryptophan Indole- 3-pyruvate Cinnamate Indole-3-acetate (auxin) Phenylalanine CO 2 phenylalanine ammonia lyase NH 3 (a) (b) FIGURE 22–28 Biosynthesis of two plant substances from amino acids. (a) Indole-3-acetate (auxin) and (b) cinnamate (cinnamon flavor). 8885d_c22_859 2/6/04 2:00 PM Page 859 mac76 mac76:385_reb: Polyamines such as spermine and spermidine, in- volved in DNA packaging, are derived from methionine and ornithine by the pathway shown in Figure 22–30. The first step is decarboxylation of ornithine, a precursor of arginine (Fig. 22–10). Ornithine decarboxylase, a PLP-requiring enzyme, is the target of several powerful inhibitors used as pharmaceutical agents (Box 22–2). ■ Arginine Is the Precursor for Biological Synthesis of Nitric Oxide A surprise finding in the mid-1980s was the role of ni- tric oxide (NO)—previously known mainly as a compo- nent of smog—as an important biological messenger. Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules860 COO H5008 CH CH 2 H 2 N H11001 HO Tyrosine H 2 O CH Ascorbate Dehydroascorbate dopamine CO 2 O 2 CH CH 2 NH 3 Dopa HO PLP CH 2 phenylethanolamine N-methyltransferase CH 2 NH 3 H11001 HO CH 3 Dopamine Tetrahydrobiopterin Dihydrobiopterin tryptophan H 2 O HO HO OH NH 3 H11001 Serotonin aromatic amino acid decarboxylase adoMet adoHcy NH 3 H11001 HO CHHO CH 2 OH CH 2 Epinephrine CO 2 PLP aromatic amino acid decarboxylase CHCH 2 H11001 H5008 OOC 5-Hydroxy- tryptophan CH 2 NH 3 H11001 CH 2 H5008 OOC CH 2 H9253-Aminobutyrate (GABA) HO COO H5008 CHCH 2 NH 3 H11001 COO H5008 Glutamate NH 3 COO H5008 CO 2 NH 3 H11001 PLP glutamate CH 2 CH 2 Norepinephrine CHCH 2 H11001 Histidine NH 3 COO H5008 N NH H 2 O Tetrahydrobiopterin Dihydrobiopterin tyrosine O 2 CH 2 H11001 NH 3 N H N H CO 2 H11001 PLP histidine HO CHCH 2 H11001 NH 3 COO H5008 Tryptophan N H CH 2 decarboxylase H9252-hydroxylase decarboxylase HO hydroxylase hydroxylase CH 2 N NH Histamine O 2 HO FIGURE 22–29 Biosynthesis of some neurotransmitters from amino acids. The key step is the same in each case: a PLP-dependent de- carboxylation (shaded in pink). This simple gaseous substance diffuses readily through membranes, although its high reactivity limits its range of diffusion to about a 1 mm radius from the site of syn- thesis. In humans NO plays a role in a range of physio- logical processes, including neurotransmission, blood clotting, and the control of blood pressure. Its mode of action is described in Chapter 12 (p. 434). Nitric oxide is synthesized from arginine in an NADPH-dependent reaction catalyzed by nitric oxide synthase (Fig. 22–31), a dimeric enzyme structurally re- lated to NADPH cytochrome P-450 reductase (see Box 21–1). The reaction is a five-electron oxidation. Each subunit of the enzyme contains one bound molecule of each of four different cofactors: FMN, FAD, tetrahydro- biopterin, and Fe 3H11001 heme. NO is an unstable molecule and cannot be stored. Its synthesis is stimulated by in- teraction of nitric oxide synthase with Ca 2H11001 -calmodulin (see Fig. 12–21). 8885d_c22_833-880 2/6/04 8:35 AM Page 860 mac76 mac76:385_reb: SUMMARY 22.3 Molecules Derived from Amino Acids ■ Many important biomolecules are derived from amino acids. Glycine is a precursor of por- phyrins. Degradation of iron-porphyrin (heme) generates bilirubin, which is converted to bile pigments, with several physiological functions. ■ Glycine and arginine give rise to creatine and phosphocreatine, an energy buffer. Glutathione, formed from three amino acids, is an important cellular reducing agent. ■ Bacteria synthesize D-amino acids from L-amino acids in racemization reactions requiring pyridoxal phosphate. ■ The aromatic amino acids give rise to many plant substances. The PLP-dependent decarboxylation of some amino acids yields important biological amines, including neurotransmitters. ■ Arginine is the precursor of nitric oxide, a biological messenger. Arginine COO H11002 CHH 3 N CH 2 CH 2 CH 2 NH NH 2 CNH 2 H11001 H11001 Hydroxyarginine NADPH, O 2 NADP H11001 , H 2 O NADPH, O 2 NADP H11001 , H 2 O COO H11002 CHH 3 N CH 2 CH 2 CH 2 NH NH 2 CNOH H11001 Citrulline COO H11002 CHH 3 N CH 2 CH 2 CH 2 NH NH 2 CO H11001 H11001 NO ? Nitric oxide 2 1 2 1 FIGURE 22–31 Biosynthesis of nitric oxide. Both steps are catalyzed by nitric oxide synthase. The nitrogen of the NO is derived from the guanidino group of arginine. H 3 N CO 2 CH H11001 S Methylthioadenosine Adenosine CH 2 CH 2 H11001 ornithine PLP CH 3 CH 2 H 3 N CH 2 CH 3 Adenosine H11001 H 3 N CH 2 (CH 2 ) 4 H11001 S COO Ornithine Putrescine Decarboxylated adoMet CO 2 adoMet PLP COO H11002 S H11001 CH 2 CH 2 CHCH 2 NH propylaminotransferase II Spermidine Adenosine H 3 N PP i H11001 P i S-Adenosylmethionine H 3 N H11001 H11001 NH 3 NH (CH 2 ) 3 Spermine AdenosineS propylaminotransferase I NH 3 H 3 N H11001 (CH 2 ) 4 H11001 NH 3 NH(CH 2 ) 3 (CH 2 ) 3 CH 3 H11001 NH 3 decarboxylase decarboxylase (CH 2 ) 4 Methionine ATP CH 3 H11001H11001 FIGURE 22–30 Biosynthesis of spermidine and spermine. The PLP- dependent decarboxylation steps are shaded in pink. In these reac- tions, S-adenosylmethionine (in its decarboxylated form) acts as a source of propylamino groups (shaded blue). 861 8885d_c22_833-880 2/6/04 8:35 AM Page 861 mac76 mac76:385_reb: 22.4 Biosynthesis and Degradation of Nucleotides As discussed in Chapter 8, nucleotides play a variety of important roles in all cells. They are the precursors of DNA and RNA. They are essential carriers of chem- ical energy—a role primarily of ATP and to some extent GTP. They are components of the cofactors NAD, FAD, S-adenosylmethionine, and coenzyme A, as well as of activated biosynthetic intermediates such as UDP-glucose and CDP-diacylglycerol. Some, such as cAMP and cGMP, are also cellular second messengers. Two types of pathways lead to nucleotides: the de novo pathways and the salvage pathways. De novo synthesis of nucleotides begins with their metabolic pre- cursors: amino acids, ribose 5-phosphate, CO 2 , and NH 3 . Salvage pathways recycle the free bases and nucleosides released from nucleic acid breakdown. Both types of Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules862 BOX 22–2 BIOCHEMISTRY IN MEDICINE Curing African Sleeping Sickness with a Biochemical Trojan Horse African sleeping sickness, or African trypanosomiasis, is caused by protists (single-celled eukaryotes) called trypanosomes (Fig. 1). This disease (and related trypanosome-caused diseases) is medically and eco- nomically significant in many developing nations. Until recently, the disease was virtually incurable. Vac- cines are ineffective, because the parasite has a novel mechanism to evade the host immune system. The cell coat of trypanosomes is covered with a sin- gle protein, which is the antigen to which the immune system responds. Every so often, however, by a process of genetic recombination (see Table 28–1), a few cells in the population of infecting trypanosomes switch to a new protein coat, not recognized by the immune system. This process of “changing coats” can occur hundreds of times. The result is a chronic cyclic infec- tion: the human host develops a fever, which subsides as the immune system beats back the first infection; trypanosomes with changed coats then become the seed for a second infection, and the fever recurs. This cycle can repeat for weeks, and the weakened person eventually dies. Some modern approaches to treating African sleeping sickness have been based on an under- standing of enzymology and metabolism. In at least one such approach, this involves pharmaceutical agents designed as mechanism-based enzyme inacti- vators (suicide inactivators; p. 211). A vulnerable point in trypanosome metabolism is the pathway of polyamine biosynthesis. The polyamines spermine and spermidine, used in DNA packaging, are required in large amounts in rapidly dividing cells. The first step in their synthesis is catalyzed by ornithine decarboxylase, a PLP-requiring enzyme (see Fig. FIGURE 1 Trypanosoma brucei rhodesiense, one of several trypanosomes known to cause African sleeping sickness. O H5008 CH OH H 3 N H H11001 CH 3 P O H 2 O CH C O CH 2 N H11001 NH Ornithine Putrescine O H5008 CH OH H 2 N H H11001 CH 3 P O (CH 2 ) 3 CH C O O CH 2 N NH 2 Pyridoxal phosphate H11001 CH OH H 2 N H CH 3 P O CH CH 2 N NH Schiff base (CH 2 ) 3 (CH 2 ) 3 CO 2 H 2 O H11001 H H H11001 H11001 FIGURE 2 Mechanism of ornithine decarboxylase reaction. 8885d_c22_862 2/6/04 1:09 PM Page 862 mac76 mac76:385_reb: pathways are important in cellular metabolism and both are presented in this section. The de novo pathways for purine and pyrimidine biosynthesis appear to be nearly identical in all living organisms. Notably, the free bases guanine, adenine, thymine, cytidine, and uracil are not intermediates in these pathways; that is, the bases are not synthesized and then attached to ribose, as might be expected. The purine ring structure is built up one or a few atoms at a time, attached to ribose throughout the process. The pyrimidine ring is synthesized as orotate, attached to ribose phosphate, and then converted to the common pyrimidine nucleotides required in nucleic acid synthe- sis. Although the free bases are not intermediates in the de novo pathways, they are intermediates in some of the salvage pathways. Several important precursors are shared by the de novo pathways for synthesis of pyrimidines and purines. 22.4 Biosynthesis and Degradation of Nucleotides 863 22–30). In mammalian cells, ornithine decarboxylase undergoes rapid turnover—that is, a constant round of enzyme degradation and synthesis. In some try- panosomes, however, the enzyme—for reasons not well understood—is stable, not readily replaced by newly synthesized enzyme. An inhibitor of ornithine decarboxylase that binds permanently to the enzyme would thus have little effect on human cells, which could rapidly replace inactivated enzyme, but would adversely affect the parasite. The first few steps of the normal reaction cat- alyzed by ornithine decarboxylase are shown in Fig- H11001 H 2 O DFMO F additional rearrangements H5008 OH H 2 N H H11001 CH 3 O CH C O O CH 2 N NH 2 Pyridoxal phosphate H11001 OH H 2 N H CH 3 P O CH CH 2 N NH Stuck! P C CH F C F O H5008 C O CH F C H11001 OH H 2 CH 3 P O CH CH 2 N H11001 NH F C EnzymeF H11001 Cys-S C H11001 OH H 2 N H CH 3 P O CH CH 2 N NH C Enzyme Schiff base O H5008 (CH 2 ) 3 (CH 2 ) 3 (CH 2 ) 3 N H HH (CH 2 ) 3 Cys-S CO 2 H11001 F H5008 ure 2. Once CO 2 is released, the electron movement is reversed and putrescine is produced (see Fig. 22–30). Based on this mechanism, several suicide in- activators have been designed, one of which is difluor- omethylornithine (DFMO). DFMO is relatively inert in solution. When it binds to ornithine decarboxylase, however, the enzyme is quickly inactivated (Fig. 3). The inhibitor acts by providing an alternative electron sink in the form of two strategically placed fluorine atoms, which are excellent leaving groups. Instead of electrons moving into the ring structure of PLP, the reaction results in displacement of a fluorine atom. The S of a Cys residue at the enzyme’s active site then forms a covalent complex with the highly reactive PLP-inhibitor adduct in an essentially irreversible re- action. In this way, the inhibitor makes use of the en- zyme’s own reaction mechanisms to kill it. DFMO has proved highly effective against African sleeping sickness in clinical trials and is now used to treat African sleeping sickness caused by T. brucei gambiense. Approaches such as this show great promise for treating a wide range of diseases. The design of drugs based on enzyme mechanism and structure can complement the more traditional trial- and-error methods of developing pharmaceuticals. FIGURE 3 Inhibition of ornithine decarboxylase by DFMO. 8885d_c22_863 2/6/04 1:09 PM Page 863 mac76 mac76:385_reb: Phosphoribosyl pyrophosphate (PRPP) is important in both, and in these pathways the structure of ribose is retained in the product nucleotide, in contrast to its fate in the tryptophan and histidine biosynthetic path- ways discussed earlier. An amino acid is an important precursor in each type of pathway: glycine for purines and aspartate for pyrimidines. Glutamine again is the most important source of amino groups—in five differ- ent steps in the de novo pathways. Aspartate is also used as the source of an amino group in the purine pathways, in two steps. Two other features deserve mention. First, there is evidence, especially in the de novo purine pathway, that the enzymes are present as large, multienzyme com- plexes in the cell, a recurring theme in our discussion of metabolism. Second, the cellular pools of nucleotides (other than ATP) are quite small, perhaps 1% or less of the amounts required to synthesize the cell’s DNA. Therefore, cells must continue to synthesize nucleotides during nucleic acid synthesis, and in some cases nu- cleotide synthesis may limit the rates of DNA replica- tion and transcription. Because of the importance of these processes in dividing cells, agents that inhibit nu- cleotide synthesis have become particularly important to modern medicine. We examine here the biosynthetic pathways of purine and pyrimidine nucleotides and their regulation, the formation of the deoxynucleotides, and the degra- dation of purines and pyrimidines to uric acid and urea. We end with a discussion of chemotherapeutic agents that affect nucleotide synthesis. De Novo Purine Nucleotide Synthesis Begins with PRPP The two parent purine nu- cleotides of nucleic acids are adenosine 5H11032-monophosphate (AMP; adenylate) and guano- sine 5H11032-monophosphate (GMP; guanylate), containing the purine bases adenine and gua- nine. Figure 22–32 shows the origin of the carbon and nitro- gen atoms of the purine ring system, as determined by John Buchanan using isotopic tracer experiments in birds. The detailed pathway of purine biosynthesis was worked out primarily by Buchanan and G. Robert Greenberg in the 1950s. In the first committed step of the pathway, an amino group donated by glutamine is attached at C-1 of PRPP (Fig. 22–33). The resulting 5-phosphoribosylamine is highly unstable, with a half-life of 30 seconds at pH 7.5. The purine ring is subsequently built up on this struc- ture. The pathway described here is identical in all or- ganisms, with the exception of one step that differs in higher eukaryotes as noted below. The second step is the addition of three atoms from glycine (Fig. 22–33, step 2 ). An ATP is consumed to activate the glycine carboxyl group (in the form of an acyl phosphate) for this condensation reaction. The added glycine amino group is then formylated by N 10 - formyltetrahydrofolate (step 3 ), and a nitrogen is con- tributed by glutamine (step 4 ), before dehydration and ring closure yield the five-membered imidazole ring of the purine nucleus, as 5-aminoimidazole ribonucleotide (AIR; step 5 ). At this point, three of the six atoms needed for the second ring in the purine structure are in place. To com- plete the process, a carboxyl group is first added (step 6 ). This carboxylation is unusual in that it does not re- quire biotin, but instead uses the bicarbonate generally present in aqueous solutions. A rearrangement transfers the carboxylate from the exocyclic amino group to po- sition 4 of the imidazole ring (step 7 ). Steps 6 and 7 are found only in bacteria and fungi. In higher eukary- otes, including humans, the 5-aminoimidazole ribonu- cleotide product of step 5 is carboxylated directly to carboxyaminoimidazole ribonucleotide in one step in- stead of two (step 6a). The enzyme catalyzing this re- action is AIR carboxylase. Aspartate now donates its amino group in two steps ( 8 and 9 ): formation of an amide bond, followed by elimination of the carbon skeleton of aspartate (as fu- marate). Recall that aspartate plays an analogous role in two steps of the urea cycle (see Fig. 18–10). The fi- nal carbon is contributed by N 10 -formyltetrahydrofolate (step 10 ), and a second ring closure takes place to yield the second fused ring of the purine nucleus (step 11). 864 Amide N of glutamine CO 2 C C C N Formate Aspartate Formate Glycine C N N C N FIGURE 22–32 Origin of the ring atoms of purines. This information was obtained from isotopic experiments with 14 C- or 15 N-labeled pre- cursors. Formate is supplied in the form of N 10 -formyltetrahydrofolate. FIGURE 22–33 (facing page) De novo synthesis of purine nucleotides: construction of the purine ring of inosinate (IMP). Each addition to the purine ring is shaded to match Figure 22–32. After step 2 , R sym- bolizes the 5-phospho-D-ribosyl group on which the purine ring is built. Formation of 5-phosphoribosylamine (step 1 ) is the first com- mitted step in purine synthesis. Note that the product of step 9 , AICAR, is the remnant of ATP released during histidine biosynthesis (see Fig. 22–20, step 5 ). Abbreviations are given for most interme- diates to simplify the naming of the pathway enzymes. Step 6a is the alternative path from AIR to CAIR occurring in higher eukaryotes. John Buchanan Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules 8885d_c22_833-880 2/6/04 8:35 AM Page 864 mac76 mac76:385_reb: H 2 H OP H OH N O Aspartate CH 2 O H OH OP P H H OH H P O CH 2 H H 4 ATP OH NH H 2 C H11001 O 5 ADP H11001 P i H R Glutamate OC N H H N Glycinamide ribonucleotide (GAR) N-Formylaminoimidazole- 4-carboxamide ribonucleotide (FAICAR) 11R Glutamine ATP H OC H H O R Formylglycinamide ribonucleotide (FGAR) PP i 1 Glutamine Glutamate ADP H11001 P i ATP H 2 C H 2 O H CH C CH R N Formylglycinamidine ribonucleotide (FGAM) 8 ADP H11001 P i ATP 6 5-Phospho-H9252- D-ribosylamine HCO 3 H5008 H C C C COO H5008 H5008 N H 2 C Glycine ATP 2 ADP H11001 P i 5-Aminoimidazole ribonucleotide (AIR) H C O N H 2 N C N N H H H 2 O C R COO H5008 5-Aminoimidazole-4-carboxamide ribonucleotide (AICAR) 10 7 N 10 -Formyl H 4 folate H 4 folate H N C C C R N H 2 N O H H 2 N O 5-Phosphoribosyl 1-pyrophosphate (PRPP) 3 N 10 -Formyl H 4 folate H 4 folate HC C O N N C R O C H 2 N HC C H R O N Carboxyamino- imidazole ribonucleotide (CAIR) N 5 -Carboxyaminoimidazole ribonucleotide (N 5 -CAIR) carboxamide -Succinyl-5-aminoimidazole-4- ribonucleotide (SAICAR) 9 Fumarate H 2 N C N C HC N C N N H NHC C N O OH H H H Inosinate (IMP) O C O H5008 PO CH 2 CH 2 O H NH 3 C C N N N C H 2 N N CH C CH R N ADP H11001 P i H5008 OO H5008 O C N AIR H ONC CO 2 6a glutamine-PRPP amidotransferase GAR synthetase GAR transformylase FGAR amidotransferase FGAM cyclase (AIR synthetase) N 5 -CAIR synthetase AIR carboxylase N 5 -CAIR mutase SAICAR synthetase SAICAR lyase AICAR transformylase IMP synthase 1 2 3 4 5 6 6a 7 8 9 10 11 865 8885d_c22_833-880 2/6/04 8:35 AM Page 865 mac76 mac76:385_reb: The first intermediate with a complete purine ring is inosinate (IMP). As in the tryptophan and histidine biosynthetic pathways, the enzymes of IMP synthesis appear to be organized as large, multienzyme complexes in the cell. Once again, evidence comes from the existence of sin- gle polypeptides with several functions, some catalyz- ing nonsequential steps in the pathway. In eukaryotic cells ranging from yeast to fruit flies to chickens, steps 1 , 3 , and 5 in Figure 22–33 are catalyzed by a mul- tifunctional protein. An additional multifunctional pro- tein catalyzes steps 10 and 11. In humans, a multi- functional enzyme combines the activities of AIR carboxylase and SAICAR synthetase (steps 6a and 8 ). In bacteria, these activities are found on separate pro- teins, but a large noncovalent complex may exist in these cells. The channeling of reaction intermediates from one enzyme to the next permitted by these com- plexes is probably especially important for unstable in- termediates such as 5-phosphoribosylamine. Conversion of inosinate to adenylate requires the insertion of an amino group derived from aspartate (Fig. 22–34); this takes place in two reactions similar to those used to introduce N-1 of the purine ring (Fig. 22–33, steps 8 and 9 ). A crucial difference is that GTP rather than ATP is the source of the high-energy phosphate in synthesizing adenylosuccinate. Guanylate is formed by the NAD H11001 -requiring oxidation of inosinate at C-2, fol- lowed by addition of an amino group derived from glu- tamine. ATP is cleaved to AMP and PP i in the final step (Fig. 22–34). Purine Nucleotide Biosynthesis Is Regulated by Feedback Inhibition Three major feedback mechanisms cooperate in regu- lating the overall rate of de novo purine nucleotide syn- thesis and the relative rates of formation of the two end products, adenylate and guanylate (Fig. 22–35). The first mechanism is exerted on the first reaction that is unique to purine synthesis—transfer of an amino group to PRPP to form 5-phosphoribosylamine. This reaction is catalyzed by the allosteric enzyme glutamine-PRPP amidotransferase, which is inhibited by the end prod- ucts IMP, AMP, and GMP. AMP and GMP act synergisti- cally in this concerted inhibition. Thus, whenever either AMP or GMP accumulates to excess, the first step in its biosynthesis from PRPP is partially inhibited. In the second control mechanism, exerted at a later stage, an excess of GMP in the cell inhibits formation of xanthylate from inosinate by IMP dehydrogenase, with- out affecting the formation of AMP (Fig. 22–35). Con- versely, an accumulation of adenylate inhibits formation of adenylosuccinate by adenylosuccinate synthetase, without affecting the biosynthesis of GMP. In the third mechanism, GTP is required in the conversion of IMP to AMP (Fig. 22–34, step 1 ), whereas ATP is required for conversion of IMP to GMP (step 4 ), a reciprocal arrangement that tends to balance the synthesis of the two ribonucleotides. The final control mechanism is the inhibition of PRPP synthesis by the allosteric regulation of ribose phosphate pyrophosphokinase. This enzyme is inhibited Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules866 P Inosinate (IMP) NH 2 Rib H 2 O H C H5008 OOC COO H5008 P Guanylate (GMP) N Rib P Adenylosuccinate NH Rib P NAD H11001 O Rib P Xanthylate (XMP) O Rib O GTP Aspartate GDP H11001 P i CH 2 NADH H11001 H H11001 Adenylate (AMP) Fumarate ATPGlu AMP H11001 PP i H 2 N Gln H 2 O N HN N N N HN N N N N N N N N N N H O N HN N adenylosuccinate synthetase adenylosuccinate lyase IMP dehydrogenase XMP-glutamine amidotransferase FIGURE 22–34 Biosynthesis of AMP and GMP from IMP. 8885d_c22_833-880 2/6/04 8:35 AM Page 866 mac76 mac76:385_reb: cytidylate synthetase ATP ADP H11001 P i Uridine 5H11032-triphosphate (UTP) kinases 2 ATP 2ADP Uridylate (UMP) orotidylate decarboxylase CO 2 Orotidylate orotate phosphoribosyl- transferase PRPP PP i Orotate dihydroorotate dehydrogenase NAD H11001 NADH H11001 H H11001 L-Dihydroorotate dihydroorotase H 2 O N-Carbamoylaspartate aspartate trans- carbamoylase Carbamoyl phosphate P i Aspartate Gln Glu Cytidine 5H11032-triphosphate (CTP) OCH 2 O H OH H HH OH O O PPP C N CH N C CH NH 2 OCH 2 O H OH H HH OH O O P P C N O C HN C CH O OCH 2 O H OH H HH OH O O C N CH HN C CH O C N H C HN C CH COO H5008 COO H5008 O O C N H CH HN C CH 2 COO H5008 O O O C N H CH H 2 N C CH 2 COO H5008 H5008 O 22.4 Biosynthesis and Degradation of Nucleotides 867 Ribose 5-phosphate ribose phosphate pyrophosphokinase (PRPP synthetase) glutamine-PRPP amidotransferase IMP GMP AMP XMP-glutamine amidotransferase IMP dehydrogenase adenylosuccinate synthetase adenylosuccinate lyase AMP GMP IMP ADP AMP GMP PRPP 5-Phosphoribosylamine Adenylosuccinate XMP 9 steps ATPADP FIGURE 22–35 Regulatory mechanisms in the biosynthesis of ade- nine and guanine nucleotides in E. coli. Regulation of these pathways differs in other organisms. FIGURE 22–36 De novo synthesis of pyrimidine nucleotides: biosyn- thesis of UTP and CTP via orotidylate. The pyrimidine is constructed from carbamoyl phosphate and aspartate. The ribose 5-phosphate is then added to the completed pyrimidine ring by orotate phosphori- bosyltransferase. The first step in this pathway (not shown here; see Fig. 18–11a) is the synthesis of carbamoyl phosphate from CO 2 and NH 4 H11001 , catalyzed in eukaryotes by carbamoyl phosphate synthetase II. by ADP and GDP, in addition to metabolites from other pathways of which PRPP is a starting point. Pyrimidine Nucleotides Are Made from Aspartate, PRPP, and Carbamoyl Phosphate The common pyrimidine ribonucleotides are cytidine 5H11032- monophosphate (CMP; cytidylate) and uridine 5H11032- monophosphate (UMP; uridylate), which contain the pyrimidines cytosine and uracil. De novo pyrimidine nu- cleotide biosynthesis (Fig. 22–36) proceeds in a some- what different manner from purine nucleotide synthe- sis; the six-membered pyrimidine ring is made first and then attached to ribose 5-phosphate. Required in this process is carbamoyl phosphate, also an intermediate in the urea cycle (see Fig. 18–10). However, as we noted 8885d_c22_833-880 2/6/04 8:35 AM Page 867 mac76 mac76:385_reb: in Chapter 18, in animals the carbamoyl phosphate re- quired in urea synthesis is made in mitochondria by car- bamoyl phosphate synthetase I, whereas the carbamoyl phosphate required in pyrimidine biosynthesis is made in the cytosol by a different form of the enzyme, car- bamoyl phosphate synthetase II. In bacteria, a sin- gle enzyme supplies carbamoyl phosphate for the syn- thesis of arginine and pyrimidines. The bacterial enzyme has three separate active sites, spaced along a channel nearly 100 ? long (Fig. 22–37). Bacterial carbamoyl phosphate synthetase provides a vivid illustration of the channeling of unstable reaction intermediates between active sites. Carbamoyl phosphate reacts with aspartate to yield N-carbamoylaspartate in the first committed step of pyrimidine biosynthesis (Fig. 22–36). This reaction is catalyzed by aspartate transcarbamoylase. In bacte- ria, this step is highly regulated, and bacterial aspartate transcarbamoylase is one of the most thoroughly stud- ied allosteric enzymes (see below). By removal of wa- ter from N-carbamoylaspartate, a reaction catalyzed by dihydroorotase, the pyrimidine ring is closed to form L-dihydroorotate. This compound is oxidized to the pyrimidine derivative orotate, a reaction in which NAD H11001 is the ultimate electron acceptor. In eukaryotes, the first three enzymes in this pathway—carbamoyl phosphate synthetase II, aspartate transcarbamoylase, and dihy- droorotase—are part of a single trifunctional protein. The protein, known by the acronym CAD, contains three identical polypeptide chains (each of M r 230,000), each with active sites for all three reactions. This suggests that large, multienzyme complexes may be the rule in this pathway. Once orotate is formed, the ribose 5-phosphate side chain, provided once again by PRPP, is attached to yield orotidylate (Fig. 22–36). Orotidylate is then decarboxy- lated to uridylate, which is phosphorylated to UTP. CTP is formed from UTP by the action of cytidylate syn- thetase, by way of an acyl phosphate intermediate (consuming one ATP). The nitrogen donor is normally glutamine, although the cytidylate synthetases in many species can use NH 4 H11001 directly. Pyrimidine Nucleotide Biosynthesis Is Regulated by Feedback Inhibition Regulation of the rate of pyrimidine nucleotide synthe- sis in bacteria occurs in large part through aspartate transcarbamoylase (ATCase), which catalyzes the first reaction in the sequence and is inhibited by CTP, the end product of the sequence (Fig. 22–36). The bacter- ial ATCase molecule consists of six catalytic subunits and six regulatory subunits (see Fig. 6–27). The cat- alytic subunits bind the substrate molecules, and the al- losteric subunits bind the allosteric inhibitor, CTP. The entire ATCase molecule, as well as its subunits, exists in two conformations, active and inactive. When CTP is not bound to the regulatory subunits, the enzyme is maximally active. As CTP accumulates and binds to the regulatory subunits, they undergo a change in confor- mation. This change is transmitted to the catalytic sub- units, which then also shift to an inactive conformation. ATP prevents the changes induced by CTP. Figure 22–38 shows the effects of the allosteric regulators on the ac- tivity of ATCase. Nucleoside Monophosphates Are Converted to Nucleoside Triphosphates Nucleotides to be used in biosynthesis are generally con- verted to nucleoside triphosphates. The conversion pathways are common to all cells. Phosphorylation of AMP to ADP is promoted by adenylate kinase, in the reaction 868 FIGURE 22–37 Channeling of intermediates in bacterial carbamoyl phosphate synthetase. (Derived from PDB ID 1M6V.) The reaction cat- alyzed by this enzyme is illustrated in Figure 18–11a. The large and small subunits are shown in gray and blue, respectively; the channel between active sites (almost 100 ? long) is shown as a yellow mesh. A glutamine molecule (green) binds to the small subunit, donating its amido nitrogen as NH 4 H11001 in a glutamine amidotransferase–type reac- tion. The NH 4 H11001 enters the channel, which takes it to a second active site, where it combines with bicarbonate in a reaction requiring ATP (bound ADP in blue). The carbamate then reenters the channel to reach the third active site, where it is phosphorylated to carbamoyl phos- phate (bound ADP in red). Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules 8885d_c22_833-880 2/6/04 8:35 AM Page 868 mac76 mac76:385_reb: ATP H11001 AMP 2ADP The ADP so formed is phosphorylated to ATP by the glycolytic enzymes or through oxidative phosphorylation. ATP also brings about the formation of other nu- cleoside diphosphates by the action of a class of en- zymes called nucleoside monophosphate kinases. These enzymes, which are generally specific for a par- ticular base but nonspecific for the sugar (ribose or de- oxyribose), catalyze the reaction ATP H11001 NMP ADP H11001 NDP The efficient cellular systems for rephosphorylating ADP to ATP tend to pull this reaction in the direction of products. Nucleoside diphosphates are converted to triphos- phates by the action of a ubiquitous enzyme, nucleo- side diphosphate kinase, which catalyzes the reac- tion NTP D H11001 NDP A NDP D H11001 NTP A This enzyme is notable in that it is not specific for the base (purines or pyrimidines) or the sugar (ribose or deoxyribose). This nonspecificity applies to both phos- phate acceptor (A) and donor (D), although the donor (NTP D ) is almost invariably ATP, because it is present in higher concentration than other nucleoside triphos- phates under aerobic conditions. Ribonucleotides Are the Precursors of Deoxyribonucleotides Deoxyribonucleotides, the building blocks of DNA, are derived from the corresponding ribonucleotides by di- rect reduction at the 2H11032-carbon atom of the D-ribose to form the 2H11032-deoxy derivative. For example, adenosine diphosphate (ADP) is reduced to 2H11032-deoxyadenosine z y z y z y diphosphate (dADP), and GDP is reduced to dGDP. This reaction is somewhat unusual in that the reduction oc- curs at a nonactivated carbon; no closely analogous chemical reactions are known. The reaction is catalyzed by ribonucleotide reductase, best characterized in E. coli, in which its substrates are ribonucleoside diphos- phates. The reduction of the D-ribose portion of a ribonu- cleoside diphosphate to 2H11032-deoxy-D-ribose requires a pair of hydrogen atoms, which are ultimately donated by NADPH via an intermediate hydrogen-carrying pro- tein, thioredoxin. This ubiquitous protein serves a sim- ilar redox function in photosynthesis (see Fig. 20–19) and other processes. Thioredoxin has pairs of OSH groups that carry hydrogen atoms from NADPH to the ribonucleoside diphosphate. Its oxidized (disulfide) form is reduced by NADPH in a reaction catalyzed by thioredoxin reductase (Fig. 22–39), and reduced thioredoxin is then used by ribonucleotide reductase to reduce the nucleoside diphosphates (NDPs) to de- oxyribonucleoside diphosphates (dNDPs). A second source of reducing equivalents for ribonucleotide re- ductase is glutathione (GSH). Glutathione serves as the reductant for a protein closely related to thioredoxin, 22.4 Biosynthesis and Degradation of Nucleotides 869 V max 1 2 V max 10 20 30 K 0.5 H11005 12 mM K 0.5 H11005 23 mM [Aspartate] (mM) Normal activity (no CTP) CTP H11001 ATP CTP M /min) V 0 ( H9262 FIGURE 22–38 Allosteric regulation of aspartate transcarbamoylase by CTP and ATP. Addition of 0.8 mM CTP, the allosteric inhibitor of ATCase, increases the K 0.5 for aspartate (lower curve) and the rate of conversion of aspartate to N-carbamoylaspartate. ATP at 0.6 mM fully reverses this effect (middle curve). glutathione reductase glutaredoxin reductase thioredoxin reductase (a) (b) dNDP NDP S S HS HS SH S S SH S S SH SH H 2 O Ribonucleotide reductase Glutaredoxin Glutaredoxin Thioredoxin Thioredoxin FADH 2 2GSH FADGSSG NADPH H11001 H H11001 NADP H11001 NADP H11001 NADPH H11001 H H11001 Ribonucleotide reductase FIGURE 22–39 Reduction of ribonucleotides to deoxyribonu- cleotides by ribonucleotide reductase. Electrons are transmitted (blue arrows) to the enzyme from NADPH by (a) glutaredoxin or (b) thiore- doxin. The sulfide groups in glutaredoxin reductase are contributed by two molecules of bound glutathione (GSH; GSSG indicates oxidized glutathione). Note that thioredoxin reductase is a flavoenzyme, with FAD as prosthetic group. 8885d_c22_833-880 2/6/04 8:35 AM Page 869 mac76 mac76:385_reb: glutaredoxin, which then transfers the reducing power to ribonucleotide reductase (Fig. 22–39). Ribonucleotide reductase is notable in that its reaction mechanism provides the best-characterized example of the involvement of free radicals in bio- chemical transformations, once thought to be rare in biological systems. The enzyme in E. coli and most eu- karyotes is a dimer, with subunits designated R1 and R2 (Fig. 22–40). The R1 subunit contains two kinds of regulatory sites, as described below. The two active sites of the enzyme are formed at the interface between the R1 and R2 subunits. At each active site, R1 con- tributes two sulfhydryl groups required for activity and R2 contributes a stable tyrosyl radical. The R2 subunit also has a binuclear iron (Fe 3H11001 ) cofactor that helps gen- erate and stabilize the tyrosyl radicals (Fig. 22–40). The tyrosyl radical is too far from the active site to interact directly with the site, but it generates another radical at the active site that functions in catalysis. A likely mechanism for the ribonucleotide reductase reaction is illustrated in Figure 22–41. The 3H11032-ribonu- cleotide radical formed in step 1 helps stabilize the cation formed at the 2H11032 carbon after the loss of H 2 O (steps 2 and 3 ). Two one-electron transfers accom- panied by oxidation of the dithiol reduce the radical cation (step 4 ). Step 5 is the reverse of step 1 , regenerat- ing the active site radical (ultimately, the tyrosyl radi- cal) and forming the deoxy product. The oxidized dithiol is reduced to complete the cycle (step 6 ). In E. coli, likely sources of the required reducing equivalents for this reaction are thioredoxin and glutaredoxin, as noted above. Four classes of ribonucleotide reductase have been reported. Their mechanisms (where known) generally conform to the scheme in Figure 22–41, but they differ in the identity of the group supplying the active-site radical and in the cofactors used to generate it. The E. coli en- zyme (class I) requires oxygen to regenerate the tyrosyl radical if it is quenched, so this enzyme functions only in an aerobic environment. Class II enzymes, found in other microorganisms, have 5H11032-deoxyadenosylcobalamin (see Box 17–2) rather than a binuclear iron center. Class III enzymes have evolved to function in an anaerobic envi- ronment. E. coli contains a separate class III ribonu- cleotide reductase when grown anaerobically; this enzyme contains an iron-sulfur cluster (structurally dis- tinct from the binuclear iron center of the class I en- zyme) and requires NADPH and S-adenosylmethionine for activity. It uses nucleoside triphosphates rather than nucleoside diphosphates as substrates. A class IV ribo- nucleotide reductase, containing a binuclear manganese center, has been reported in some microorganisms. The evolution of different classes of ribonucleotide reductase for production of DNA precursors in different environ- ments reflects the importance of this reaction in nu- cleotide metabolism. Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules870 (b) SH (a) Fe 3H11001 OO OFe 3H11001 Fe 3H11001 HS XH HX SH HS R1 subunit R1 subunit R2 subunit ATP, dGTP, dTTP dATP, ATP, dATP Primary regulation Active site site Substrate specificity site Regulatory sites Substrates Allosteric effectors F O ADP, CDP, UDP, GDP e 3H11001 OH H11001O XH11001 XH (c) FIGURE 22–40 Ribonucleotide reductase. (a) Subunit structure. The functions of the two regulatory sites are explained in Figure 22–42. Each active site contains two thiols and a group (OXH) that can be converted to an active-site radical; this group is probably the OSH of Cys 439 , which functions as a thiyl radical. (b) The R2 subunits of E. coli ribonucleotide reductase (PDB ID 1PFR). The Tyr residue that acts as the tyrosyl radical is shown in red; the binu- clear iron center is orange. (c) The tyrosyl radical functions to gen- erate the active-site radical (OX H11080 ), which is used in the mechanism shown in Figure 22–41. 8885d_c22_870 2/6/04 1:10 PM Page 870 mac76 mac76:385_reb: 22.4 Biosynthesis and Degradation of Nucleotides 871 HH O NDP Thioredoxin (SH) 2 (glutaredoxin) Thioredoxin S S (glutaredoxin) NDP dNDP dNDP 1 2 5 6 3 4 SH X H CH 2 H Base O H OH H P OP S H11002 H11001 O H H SH X H CH 2 H Base O H OH H P OP HS O H 3H11032 2H11032 H SH X CH 2 H Base O H OH H P OP HS O H R2 subunit Ribonucleotide reductase R1 subunit H H S X CH 2 Base O H OH H P OP S H H H S X CH 2 Base O H OH H P OP S H S H CH 2 Base O H OH H P OP H11001 H HX S H11002 MECHANISM FIGURE 22–41 Proposed mechanism for ribonu- cleotide reductase. In the enzyme of E. coli and most eukaryotes, the active thiol groups are on the R1 subunit; the active-site radical (OX H11080 ) is on the R2 subunit and in E. coli is probably a thiyl radical of Cys 439 (see Fig. 22–40). Steps 1 through 6 are described in the text. 8885d_c22_871 2/6/04 1:10 PM Page 871 mac76 mac76:385_reb: Regulation of E. coli ribonucleotide reductase is unusual in that not only its activity but its substrate specificity is regulated by the binding of effector mol- ecules. Each R1 subunit has two types of regulatory site (Fig. 22–40). One type affects overall enzyme activity and binds either ATP, which activates the enzyme, or dATP, which inactivates it. The second type alters sub- strate specificity in response to the effector molecule— ATP, dATP, dTTP, or dGTP—that is bound there (Fig. 22–42). When ATP or dATP is bound, reduction of UDP and CDP is favored. When dTTP or dGTP is bound, reduction of GDP or ADP, respectively, is stimulated. The scheme is designed to provide a balanced pool of precursors for DNA synthesis. ATP is also a general activator for biosynthesis and ribonucleotide reduction. The presence of dATP in small amounts increases the reduction of pyrimidine nucleotides. An oversupply of the pyrimidine dNTPs is signaled by high levels of dTTP, which shifts the specificity to favor reduction of GDP. High levels of dGTP, in turn, shift the specificity to ADP reduction, and high levels of dATP shut the en- zyme down. These effectors are thought to induce sev- eral distinct enzyme conformations with altered speci- ficities. Thymidylate Is Derived from dCDP and dUMP DNA contains thymine rather than uracil, and the de novo pathway to thymine involves only deoxyribonu- cleotides. The immediate precursor of thymidylate (dTMP) is dUMP. In bacteria, the pathway to dUMP be- gins with formation of dUTP, either by deamination of dCTP or by phosphorylation of dUDP (Fig. 22–43). The dUTP is converted to dUMP by a dUTPase. The latter reaction must be efficient to keep dUTP pools low and prevent incorporation of uridylate into DNA. Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules872 dCDP dGDP dADP dTTP dCTP dGTP dATP Regulation at substrate- specificity sites Products dUDP dCDP dGDP dADP dTTP dCTP dGTP dATP CDP ADP Regulation at primary regulatory sites Products Substrates ATP (d)ATP GDP dUDPUDP FIGURE 22–42 Regulation of ribonucleotide reductase by deoxynu- cleoside triphosphates. The overall activity of the enzyme is affected by binding at the primary regulatory site (left). The substrate specificity of the enzyme is affected by the nature of the effector molecule bound at the second type of regulatory site (right). The diagram indicates in- hibition or stimulation of enzyme activity with the four different sub- strates. The pathway from dUDP to dTTP is described later (see Figs 22–43, 22–44). dCDP dCTP UDP dUDP dUTP dUMP dTMP ribonucleotide reductase nucleoside diphosphate kinase deaminase dUTPase thymidylate synthase CDP FIGURE 22–43 Biosynthesis of thymidylate (dTMP). The pathways are shown beginning with the reaction catalyzed by ribonucleotide reductase. Figure 22–44 gives details of the thymidylate synthase reaction. 8885d_c22_833-880 2/6/04 8:35 AM Page 872 mac76 mac76:385_reb: Conversion of dUMP to dTMP is catalyzed by thy- midylate synthase. A one-carbon unit at the hydroxy- methyl (OCH 2 OH) oxidation level (see Fig. 18–17) is transferred from N 5 ,N 10 -methylenetetrahydrofolate to dUMP, then reduced to a methyl group (Fig. 22–44). The reduction occurs at the expense of oxidation of tetrahydrofolate to dihydrofolate, which is unusual in tetrahydrofolate-requiring reactions. (The mechanism of this reaction is shown in Fig. 22–50.) The dihydrofo- late is reduced to tetrahydrofolate by dihydrofolate reductase—a regeneration that is essential for the many processes that require tetrahydrofolate. In plants and at least one protist, thymidylate synthase and dihy- drofolate reductase reside on a single bifunctional protein. Degradation of Purines and Pyrimidines Produces Uric Acid and Urea, Respectively Purine nucleotides are degraded by a pathway in which they lose their phosphate through the action of 5H11541- nucleotidase (Fig. 22–45). Adenylate yields adenosine, which is deaminated to inosine by adenosine deami- nase, and inosine is hydrolyzed to hypoxanthine (its purine base) and D-ribose. Hypoxanthine is oxidized successively to xanthine and then uric acid by xanthine oxidase, a flavoenzyme with an atom of molybdenum and four iron-sulfur centers in its prosthetic group. Mol- ecular oxygen is the electron acceptor in this complex reaction. 22.4 Biosynthesis and Degradation of Nucleotides 873 FIGURE 22–44 Conversion of dUMP to dTMP by thymidylate syn- thase and dihydrofolate reductase. Serine hydroxymethyltransferase is required for regeneration of the N 5 ,N 10 -methylene form of tetrahy- drofolate. In the synthesis of dTMP, all three hydrogens of the added methyl group are derived from N 5 ,N 10 -methylenetetrahydrofolate (pink and gray). O CH 2 OH N N H H H C H CH 2 dTMP O P O HN O OH O N H H H H H CH 2 dUMP O P O HN H H O O N H 2 N H N H HN CH 2 N H H O NH 2 N C HN H H H N H HN R N R thymidylate synthase N 5 ,N 10 -Methylene- tetrahydrofolate 7,8-Dihydrofolate Glycine Serine NADPH H11001 H H11001 H11001 NADP serine hydroxymethyl- transferase dihydrofolate reductase Tetrahydrofolate CH 2 N H O NH 2 N HN H N HN R PLP 8885d_c22_873 2/6/04 1:10 PM Page 873 mac76 mac76:385_reb: N H C GMP Guanosine nucleosidase AMP guanine deaminase Adenosine Inosine Hypoxanthine (keto form) xanthine oxidase Xanthine (keto form) Uric acid H 2 O Guanine nucleosidase H 2 O 2 O 5H11032-nucleotidase H 2 O P i 5H11032-nucleotidase H 2 O P i NH 3 H 2 O Ribose O HN N N N N H Ribose H 2 O H 2 O H11001 O 2 H O HO xanthine oxidase OH HO HN 4NH 4 H11001 O adenosine deaminase NH 2 H 2 O C C C C C N N H N C O H 2 O 2 H 2 O H11001 O 2 allantoicase N H OH Uric acid Allantoin C O H N C H C NH 3 O H 2 O allantoinase NH 2 NH 2 C ON H C H COO H11002 Allantoate urease COO H11002 CHO Glyoxylate NH 2 C O H 2 O CO 2 urate oxidase 2H 2 N 2H 2 O 2CO 2 Urea Excreted by: Primates, birds, reptiles, insects Most mammals Bony fishes Amphibians, cartilaginous fishes Marine invertebrates O 2 H11001 H 2 O H 2 N O HN N N N N N H N N N H HO HN O N H HN 2 1 FIGURE 22–45 Catabolism of purine nucleotides. Note that primates excrete much more nitrogen as urea via the urea cycle (Chapter 18) than as uric acid from purine degradation. Similarly, fish excrete much more nitrogen as NH 4 H11001 than as urea produced by the pathway shown here. GMP catabolism also yields uric acid as end prod- uct. GMP is first hydrolyzed to guanosine, which is then cleaved to free guanine. Guanine undergoes hydrolytic removal of its amino group to yield xanthine, which is converted to uric acid by xanthine oxidase (Fig. 22–45). Uric acid is the excreted end product of purine ca- tabolism in primates, birds, and some other animals. A healthy adult human excretes uric acid at a rate of about 0.6 g/24 h; the excreted product arises in part from in- gested purines and in part from turnover of the purine nucleotides of nucleic acids. In most mammals and many other vertebrates, uric acid is further degraded to al- lantoin by the action of urate oxidase. In other or- ganisms the pathway is further extended, as shown in Figure 22–45. The pathways for degradation of pyrimidines gen- erally lead to NH 4 H11001 production and thus to urea syn- thesis. Thymine, for example, is degraded to methyl- malonylsemialdehyde (Fig. 22–46), an intermediate of valine catabolism. It is further degraded through propionyl-CoA and methylmalonyl-CoA to succinyl- CoA (see Fig. 18–27). Genetic aberrations in human purine metabolism have been found, some with serious conse- quences. For example, adenosine deaminase (ADA) deficiency leads to severe immunodeficiency disease in which T lymphocytes and B lymphocytes do not de- velop properly. Lack of ADA leads to a 100-fold increase in the cellular concentration of dATP, a strong inhibitor of ribonucleotide reductase (Fig. 22–42). High levels Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules874 8885d_c22_833-880 2/6/04 8:35 AM Page 874 mac76 mac76:385_reb: NADPH H11001 H H11001 O HN C N H C C O NADP H11001 O HN C CH N H CH 2 C C O H H 2 N CH 3 Thymine dihydrouracil dehydrogenase Dihydrothymine dihydropyrimidinase H9252-Ureidoisobutyrate CH 3 C O NH CH 2 CH CH 3 C O O H11002 H 2 O H 3 N H11002 CH 2 CH C O O H11002 CH 3 aminotransferase H9252-Aminoisobutyrate H 2 O H9252-ureidopropionase H9251-Ketoglutarate Glutamate CH H 3 C C O H11002 O O H Methylmalonyl- semialdehyde H11001 NH 4 H11001 HCO 3 H11001 FIGURE 22–46 Catabolism of a pyrimidine. Shown here is the path- way for thymine. The methylmalonylsemialdehyde is further degraded to succinyl-CoA. of dATP produce a general deficiency of other dNTPs in T lymphocytes. The basis for B-lymphocyte toxicity is less clear. Individuals with ADA deficiency lack an ef- fective immune system and do not survive unless iso- lated in a sterile “bubble” environment. ADA deficiency is one of the first targets of human gene therapy trials (see Box 9–2). ■ Purine and Pyrimidine Bases Are Recycled by Salvage Pathways Free purine and pyrimidine bases are constantly re- leased in cells during the metabolic degradation of nu- cleotides. Free purines are in large part salvaged and reused to make nucleotides, in a pathway much simpler than the de novo synthesis of purine nucleotides de- scribed earlier. One of the primary salvage pathways consists of a single reaction catalyzed by adenosine phosphoribosyltransferase, in which free adenine reacts with PRPP to yield the corresponding adenine nucleotide: Adenine H11001 PRPP On AMP H11001 PP i Free guanine and hypoxanthine (the deamination prod- uct of adenine; Fig. 22–45) are salvaged in the same way by hypoxanthine-guanine phosphoribosyltrans- ferase. A similar salvage pathway exists for pyrimidine bases in microorganisms, and possibly in mammals. A genetic lack of hypoxanthine-guanine phos- phoribosyltransferase activity, seen almost ex- clusively in male children, results in a bizarre set of symptoms called Lesch-Nyhan syndrome. Children with this genetic disorder, which becomes manifest by the age of 2 years, are sometimes poorly coordinated and mentally retarded. In addition, they are extremely hostile and show compulsive self-destructive tenden- cies: they mutilate themselves by biting off their fingers, toes, and lips. The devastating effects of Lesch-Nyhan syndrome illustrate the importance of the salvage pathways. Hy- poxanthine and guanine arise constantly from the break- down of nucleic acids. In the absence of hypoxanthine- guanine phosphoribosyltransferase, PRPP levels rise and purines are overproduced by the de novo pathway, resulting in high levels of uric acid production and gout- like damage to tissue (see below). The brain is espe- cially dependent on the salvage pathways, and this may account for the central nervous system damage in chil- dren with Lesch-Nyhan syndrome. This syndrome is another target of early trials in gene therapy (see Box 9–2). ■ Excess Uric Acid Causes Gout Long thought, erroneously, to be due to “high liv- ing,” gout is a disease of the joints caused by an elevated concentration of uric acid in the blood and tis- sues. The joints become inflamed, painful, and arthritic, owing to the abnormal deposition of sodium urate crys- tals. The kidneys are also affected, as excess uric acid is deposited in the kidney tubules. Gout occurs pre- dominantly in males. Its precise cause is not known, but it often involves an underexcretion of urate. A genetic deficiency of one or another enzyme of purine metabo- lism may also be a factor in some cases. 22.4 Biosynthesis and Degradation of Nucleotides 875 8885d_c22_833-880 2/6/04 8:35 AM Page 875 mac76 mac76:385_reb: Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules876 CH 2 NH 3 O H N H11002 H11001 NH 2 C CH NH 3 H COO H11002 H11001 C Acivicin O C O CH 2 COO N H11001 Glutamine C N H11001 CH 2 C H Cl COO H11002 CH C O CH 2 NH 3 Azaserine FIGURE 22–48 Azaserine and acivicin, inhibitors of glutamine ami- dotransferases. These analogs of glutamine interfere in a number of amino acid and nucleotide biosynthetic pathways. Gout is effectively treated by a combination of nu- tritional and drug therapies. Foods especially rich in nu- cleotides and nucleic acids, such as liver or glandular products, are withheld from the diet. Major alleviation of the symptoms is provided by the drug allopurinol (Fig. 22–47), which inhibits xanthine oxidase, the enzyme that catalyzes the conversion of purines to uric acid. Allop- urinol is a substrate of xanthine oxidase, which converts allopurinol to oxypurinol (alloxanthine). Oxypurinol in- activates the reduced form of the enzyme by remaining tightly bound in its active site. When xanthine oxidase is inhibited, the excreted products of purine metabolism are xanthine and hypoxanthine, which are more water- soluble than uric acid and less likely to form crystalline deposits. Allopurinol was developed by Gertrude Elion and George Hitchings, who also developed acyclovir, used in treating people with AIDS, and other purine analogs used in cancer chemotherapy. ■ Many Chemotherapeutic Agents Target Enzymes in the Nucleotide Biosynthetic Pathways The growth of cancer cells is not controlled in the same way as cell growth in most normal tissues. Cancer cells have greater requirements for nucleotides as precursors of DNA and RNA, and consequently are generally more sensitive than normal cells to inhibitors of nucleotide biosynthesis. A growing array of important chemotherapeutic agents—for cancer and other dis- eases—act by inhibiting one or more enzymes in these pathways. We describe here several well-studied exam- ples that illustrate productive approaches to treatment and help us understand how these enzymes work. The first set of agents includes compounds that in- hibit glutamine amidotransferases. Recall that glutamine is a nitrogen donor in at least half a dozen separate re- actions in nucleotide biosynthesis. The binding sites for glutamine and the mechanism by which NH 4 H11001 is ex- tracted are quite similar in many of these enzymes. Most are strongly inhibited by glutamine analogs such as aza- serine and acivicin (Fig. 22–48). Azaserine, charac- terized by John Buchanan in the 1950s, was one of the first examples of a mechanism-based enzyme inactiva- tor (suicide inactivator; p. 211 and Box 22–2). Acivicin shows promise as a cancer chemotherapeutic agent. Other useful targets for pharmaceutical agents are thymidylate synthase and dihydrofolate reductase, en- zymes that provide the only cellular pathway for thymine synthesis (Fig. 22–49). One inhibitor that acts on thymidylate synthase, fluorouracil, is an important chemotherapeutic agent. Fluorouracil itself is not the enzyme inhibitor. In the cell, salvage pathways convert it to the deoxynucleoside monophosphate FdUMP, which binds to and inactivates the enzyme. Inhibition by FdUMP (Fig. 22–50) is a classic example of mecha- nism-based enzyme inactivation. Another prominent chemotherapeutic agent, methotrexate, is an inhibitor of dihydrofolate reductase. This folate analog acts as a competitive inhibitor; the enzyme binds methotrexate with about 100 times higher affinity than dihydrofolate. Aminopterin is a related compound that acts similarly. OH N HC N N H C C N Hypoxanthine (enol form) H C N C HC N C C N Allopurinol OH N H CH C H C N C N C C N Oxypurinol OH N H C HO xanthine oxidase FIGURE 22–47 Allopurinol, an inhibitor of xanthine oxidase. Hypoxanthine is the normal substrate of xanthine oxidase. Only a slight alteration in the structure of hypoxanthine (shaded pink) yields the medically effective enzyme inhibitor allopurinol. At the active site, allopurinol is converted to oxypuri- nol, a strong competitive inhibitor that remains tightly bound to the reduced form of the enzyme. Gertrude Elion (1918–1999) and George Hitchings (1905–1998) 8885d_c22_833-880 2/6/04 8:35 AM Page 876 mac76 mac76:385_reb: 22.4 Biosynthesis and Degradation of Nucleotides 877 HN N H F NH 2 CH CH 3 COO H11002 O Methotrexate CH 2 O NH O H N C H 2 CH 2 N N N N N CH 2 COO H11002 Fluorouracil (b)(a) 10 5 dUMP dTMP FdUMP Methotrexate Aminopterin Trimethoprim Serine PLP Glycine NADP H11001 thymidylate synthase dihydrofolate reductase serine hydroxymethyl- transferase O Trimethoprim CH 3 H 3 CO H 3 CO NH 2 NH 2 N N N 5 , N 10 -Methylene H 4 folate 7,8-Dihydrofolate H 4 folate NADPH H11001 H H11001 FIGURE 22–49 Thymidylate synthesis and folate metabolism as targets of chemotherapy. (a) During thymidylate synthesis, N 5 ,N 10 -methylenetetrahydrofolate is converted to 7,8-dihydrofolate; the N 5 ,N 10 -methylenetetrahydrofolate is regenerated in two steps (see Fig. 22–44). This cycle is a major target of several chemotherapeutic agents. (b) Fluorouracil and methotrexate are important chemothera- peutic agents. In cells, fluorouracil is converted to FdUMP, which inhibits thymidylate synthase. Methotrexate, a structural analog of tetrahydrofolate, inhibits dihydrofolate reductase; the shaded amino and methyl groups replace a carbonyl oxygen and a proton, respec- tively, in folate (see Fig. 22–44). Another important folate analog, aminopterin, is identical to methotrexate except that it lacks the shaded methyl group. Trimethoprim, a tight-binding inhibitor of bacterial di- hydrofolate reductase, was developed as an antibiotic. MECHANISM FIGURE 22–50 Conversion of dUMP to dTMP and its inhibition by FdUMP. The top row is the normal reaction mechanism of thymidylate synthase. The nucleophilic sulfhydryl group contributed by the enzyme in step 1 and the ring atoms of dUMP taking part in the reaction are shown in red; :B denotes an amino acid side chain that acts as a base to abstract a proton in step 3 . The hydrogens de- rived from the methylene group of N 5 ,N 10 -methylenetetrahydrofolate are shaded in gray. A novel feature of this reaction mechanism is a 1,3 hydride shift (step 3 ), which moves a hydride ion (shaded pink) from C-6 of H 4 folate to the methyl group of thymidine, resulting in the oxidation of tetrahydrofolate to dihydrofolate. It is this hydride shift that apparently does not occur when FdUMP is the substrate (bottom). Steps 1 and 2 proceed normally, but result in a stable complex— with FdUMP linked covalently to the enzyme and to tetrahydrofolate— that inactivates the enzyme. Thymidylate Synthase Mechanism O HN N R O H CH 3 FdUMP N 5 ,N 10 -Methylene H 4 folate O FHN C N RH11032 R C H O N H H C CH 2 O HN N RH11032 R C H O N H H C CH 2 O FHN C N R C H O H H C 1,3 hydride shift Enzyme H11001 dihydrofolate O HHN C N R C H O dTMP dUMP O HHN C N RH11032 R C H O N H H C CH 2 B B B B B B B 10 O HN N RH11032 R C H O N H H C CH 2 S O HHN C N CH 2 R C H O H C H H N C S S CH 2 C H N CH 2 C H N CH 2 C H N 5 CH 2 C H N CH 2 C H N CH 2 C H N C H H FC H5008 S S Dead-end covalent complex RH11032 NH CH 2 SH SH C H5008 RH11032 NH CH 2 RH11032 NH CH 2 1 2 1 2 3 8885d_c22_877 2/6/04 1:10 PM Page 877 mac76 mac76:385_reb: Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules878 Key Terms nitrogen cycle 834 nitrogen fixation 834 symbionts 834 nitrogenase complex 835 leghemoglobin 836 glutamine synthetase 838 glutamate synthase 838 glutamine amidotransferases 840 5-phosphoribosyl-1- pyrophosphate (PRPP) 842 tryptophan synthase 849 porphyrin 854 porphyria 854 bilirubin 854 phosphocreatine 857 creatine 857 glutathione (GSH) 857 auxin 859 dopamine 859 norepinephrine 859 epinephrine 859 H9253-aminobutyrate (GABA) 859 serotonin 859 histamine 859 cimetidine 859 spermine 860 spermidine 860 ornithine decarboxylase 860 de novo pathway 862 salvage pathway 862 inosinate (IMP) 866 carbamoyl phosphate synthetase II 868 aspartate transcarbamoylase 868 nucleoside monophosphate kinase 869 nucleoside diphosphate kinase 869 ribonucleotide reductase 869 thioredoxin 869 thymidylate synthase 873 dihydrofolate reductase 873 adenosine deaminase deficiency 874 Lesch-Nyhan syndrome 875 allopurinol 876 azaserine 876 acivicin 876 fluorouracil 876 methotrexate 876 aminopterin 876 Terms in bold are defined in the glossary. Further Reading Nitrogen Fixation Burris, R.H. (1995) Breaking the N–N bond. Annu. Rev. Plant Physiol. Plant Mol. Biol. 46, 1–19. Igarishi, R.Y. & Seefeldt, L.C. (2003) Nitrogen fixation: the mechanism of the Mo-dependent nitrogenase. Crit. Rev. Biochem. Mol. Biol., 38, 351–384. The medical potential of inhibitors of nucleotide biosynthesis is not limited to cancer treatment. All fast- growing cells (including bacteria and protists) are po- tential targets. Trimethoprim, an antibiotic developed by Hitchings and Elion, binds to bacterial dihydrofolate reductase nearly 100,000 times better than to the mam- malian enzyme. It is used to treat certain urinary and middle ear bacterial infections. Parasitic protists, such as the trypanosomes that cause African sleeping sickness (African trypanosomiasis), lack pathways for de novo nucleotide biosynthesis and are particularly sensitive to agents that interfere with their scavenging of nucleo- tides from the surrounding environment using salvage pathways. Allopurinol (Fig. 22–47) and a number of re- lated purine analogs have shown promise for the treat- ment of African trypanosomiasis and related afflictions. See Box 22–2 for another approach to combating African trypanosomiasis, made possible by advances in our un- derstanding of metabolism and enzyme mechanism. ■ SUMMARY 22.4 Biosynthesis and Degradation of Nucleotides ■ The purine ring system is built up step-by-step beginning with 5-phosphoribosylamine. The amino acids glutamine, glycine, and aspartate furnish all the nitrogen atoms of purines. Two ring-closure steps form the purine nucleus. ■ Pyrimidines are synthesized from carbamoyl phosphate and aspartate, and ribose 5-phosphate is then attached to yield the pyrimidine ribonucleotides. ■ Nucleoside monophosphates are converted to their triphosphates by enzymatic phosphory- lation reactions. Ribonucleotides are converted to deoxyribonucleotides by ribonucleotide re- ductase, an enzyme with novel mechanistic and regulatory characteristics. The thymine nu- cleotides are derived from dCDP and dUMP. ■ Uric acid and urea are the end products of purine and pyrimidine degradation. ■ Free purines can be salvaged and rebuilt into nucleotides. Genetic deficiencies in certain salvage enzymes cause serious disorders such as Lesch-Nyhan syndrome and ADA deficiency. ■ Accumulation of uric acid crystals in the joints, possibly caused by another genetic deficiency, results in gout. ■ Enzymes of the nucleotide biosynthetic pathways are targets for an array of chemotherapeutic agents used to treat cancer and other diseases. 8885d_c22_878 2/6/04 2:00 PM Page 878 mac76 mac76:385_reb: Chapter 22 Problems 879 Patriarca, E.J., Tate, R., & Iaccarino, M. (2002) Key role of bacterial NH 4 H11001 metabolism in rhizobium-plant symbiosis. Microbiol. Mol. Biol. Rev. 66, 203–222. A good overview of ammonia assimilation in bacterial systems and its regulation. Sinha, S.C. & Smith, J.L. (2001) The PRT protein family. Curr. Opin. Struct. Biol. 11, 733–739. Description of a protein family that includes many amidotrans- ferases, with channels for the movement of NH 3 from one active site to another. Ye, R.W. & Thomas, S.M. (2001) Microbial nitrogen cycles: physiology, genomics and applications. Curr. Opin. Microbiol. 4, 307–312. Amino Acid Biosynthesis Abeles, R.H., Frey, P.A., & Jencks, W.P. (1992) Biochemistry, Jones and Bartlett Publishers, Boston. This book includes excellent accounts of reaction mechanisms, including one-carbon metabolism and pyridoxal phosphate enzymes. Bender, D.A. (1985) Amino Acid Metabolism, 2nd edn, Wiley- Interscience, New York. Neidhardt, F.C. (ed.) (1996) Escherichia coli and Salmonella: Cellular and Molecular Biology, 2nd edn, ASM Press, Washing- ton, DC. Volume 1 of this two-volume set has 13 chapters devoted to detailed descriptions of amino acid and nucleotide biosynthesis in bacteria. The web-based version, EcoSal, is updated regularly. A valuable resource. Pan P., Woehl, E., & Dunn, M.F. (1997) Protein architecture, dynamics and allostery in tryptophan synthase channeling. Trends Biochem. Sci. 22, 22–27. Compounds Derived from Amino Acids Bredt, D.S. & Snyder, S.H. (1994) Nitric oxide: a physiologic messenger molecule. Annu. Rev. Biochem. 63, 175–195. Meister, A. & Anderson, M.E. (1983) Glutathione. Annu. Rev. Biochem. 52, 711–760. Morse, D. & Choi, A.M.K. (2002) Heme oxygenase-1—the “emerging molecule” has arrived. Am. J. Resp. Cell Mol. Biol. 27, 8–16. Rondon, M.R., Trzebiatowski, J.R., & Escalante-Semerena, J.C. (1997) Biochemistry and molecular genetics of cobalamin biosynthesis. Prog. Nucleic Acid Res. Mol. Biol. 56, 347–384. Stadtman, T.C. (1996) Selenocysteine. Annu. Rev. Biochem. 65, 83–100. Nucleotide Biosynthesis Blakley, R.L. & Benkovic, S.J. (1985) Folates and Pterins, Vol. 2: Chemistry and Biochemistry of Pterins, Wiley-Interscience, New York. Carreras, C.W. & Santi, D.V. (1995) The catalytic mechanism and structure of thymidylate synthase. Annu. Rev. Biochem. 64, 721–762. Eliasson, R., Pontis, E., Sun, X., & Reichard, P. (1994) Allosteric control of the substrate specificity of the anaerobic ribonucleotide reductase from Escherichia coli. J. Biol. Chem. 269, 26,052–26,057. Holmgren, A. (1989) Thioredoxin and glutaredoxin systems. J. Biol. Chem. 264, 13,963–13,966. Jordan, A. & Reichard P. (1998) Ribonucleotide reductases. Annu. Rev. Biochem. 67, 71–98. Kappock, T.J., Ealick, S.E., & Stubbe, J. (2000) Modular evolution of the purine biosynthetic pathway. Curr. Opin. Chem. Biol. 4, 567–572. Kornberg, A. & Baker, T.A. (1991) DNA Replication, 2nd edn, W. H. Freeman and Company, New York. This text includes a good summary of nucleotide biosynthesis. Lee, L., Kelly, R.E., Pastra-Landis, S.C., & Evans, D.R. (1985) Oligomeric structure of the multifunctional protein CAD that initiates pyrimidine biosynthesis in mammalian cells. Proc. Natl. Acad. Sci. USA 82, 6802–6806. Licht, S., Gerfen, G.J., & Stubbe, J. (1996) Thiyl radicals in ribonucleotide reductases. Science 271, 477–481. Schachman, H.K. (2000) Still looking for the ivory tower. Annu. Rev. Biochem. 69, 1–29. A lively description of research on aspartate transcarbamoylase, accompanied by delightful tales of science and politics. Stubbe, J. & Riggs-Gelasco, P. (1998) Harnessing free radicals: formation and function of the tyrosyl radical in ribonucleotide reductase. Trends Biochem. Sci. 23, 438–443. Villafranca, J.E., Howell, E.E., Voet, D.H., Strobel, M.S., Ogden, R.C., Abelson, J.N., & Kraut, J. (1983) Directed mutagenesis of dihydrofolate reductase. Science 222, 782–788. A report of structural studies on this important enzyme. Genetic Diseases Scriver, C.R., Beaudet, A.L., Valle, D., Sly, W.S., Childs, B., Kinzler, L.W., & Vogelstein, B. (eds) (2001) The Metabolic and Molecular Bases of Inherited Disease, 8th edn, McGraw-Hill Professional, New York. This four-volume set has good chapters on disorders of amino acid, porphyrin, and heme metabolism. See also the chapters on inborn errors of purine and pyrimidine metabolism. 1. ATP Consumption by Root Nodules in Legumes Bacteria residing in the root nodules of the pea plant con- sume more than 20% of the ATP produced by the plant. Sug- gest why these bacteria consume so much ATP. 2. Glutamate Dehydrogenase and Protein Synthesis The bacterium Methylophilus methylotrophus can synthe- size protein from methanol and ammonia. Recombinant DNA techniques have improved the yield of protein by introduc- ing into M. methylotrophus the glutamate dehydrogenase gene from E. coli. Why does this genetic manipulation in- crease the protein yield? Problems 8885d_c22_833-880 2/6/04 8:35 AM Page 879 mac76 mac76:385_reb: Chapter 22 Biosynthesis of Amino Acids, Nucleotides, and Related Molecules880 3. Transformation of Aspartate to Asparagine There are two routes for transforming aspartate to asparagine at the expense of ATP. Many bacteria have an asparagine synthetase that uses ammonium ion as the nitrogen donor. Mammals have an asparagine synthetase that uses glutamine as the nitrogen donor. Given that the latter requires an extra ATP (for the synthesis of glutamine), why do mammals use this route? 4. Equation for the Synthesis of Aspartate from Glu- cose Write the net equation for the synthesis of aspartate (a nonessential amino acid) from glucose, carbon dioxide, and ammonia. 5. Phenylalanine Hydroxylase Deficiency and Diet Tyrosine is normally a nonessential amino acid, but individu- als with a genetic defect in phenylalanine hydroxylase require tyrosine in their diet for normal growth. Explain. 6. Cofactors for One-Carbon Transfer Reactions Most one-carbon transfers are promoted by one of three co- factors: biotin, tetrahydrofolate, or S-adenosylmethionine (Chapter 18). S-Adenosylmethionine is generally used as a methyl group donor; the transfer potential of the methyl group in N 5 -methyltetrahydrofolate is insufficient for most biosynthetic reactions. However, one example of the use of N 5 -methyltetrahydrofolate in methyl group transfer is in me- thionine formation by the methionine synthase reaction (step 9 of Fig. 22–15); methionine is the immediate precursor of S-adenosylmethionine (see Fig. 18–18). Explain how the methyl group of S-adenosylmethionine can be derived from N 5 -methyltetrahydrofolate, even though the transfer poten- tial of the methyl group in N 5 -methyltetrahydrofolate is one- thousandth of that in S-adenosylmethionine. 7. Concerted Regulation in Amino Acid Biosynthesis The glutamine synthetase of E. coli is independently modu- lated by various products of glutamine metabolism (see Fig. 22–6). In this concerted inhibition, the extent of enzyme in- hibition is greater than the sum of the separate inhibitions caused by each product. For E. coli grown in a medium rich in histidine, what would be the advantage of concerted inhi- bition? 8. Relationship between Folic Acid Deficiency and Anemia Folic acid deficiency, believed to be the most common vitamin deficiency, causes a type of ane- mia in which hemoglobin synthesis is impaired and erythro- cytes do not mature properly. What is the metabolic rela- tionship between hemoglobin synthesis and folic acid deficiency? 9. Nucleotide Biosynthesis in Amino Acid Auxotrophic Bacteria Normal E. coli cells can synthesize all 20 com- mon amino acids, but some mutants, called amino acid aux- otrophs, are unable to synthesize a specific amino acid and require its addition to the culture medium for optimal growth. Besides their role in protein synthesis, some amino acids are also precursors for other nitrogenous cell products. Consider the three amino acid auxotrophs that are unable to synthe- size glycine, glutamine, and aspartate, respectively. For each mutant, what nitrogenous products other than proteins would the cell fail to synthesize? 10. Inhibitors of Nucleotide Biosynthesis Suggest mechanisms for the inhibition of (a) alanine racemase by L- fluoroalanine and (b) glutamine amidotransferases by aza- serine. 11. Mode of Action of Sulfa Drugs Some bacte- ria require p-aminobenzoate in the culture medium for normal growth, and their growth is severely inhibited by the addition of sulfanilamide, one of the earliest sulfa drugs. Moreover, in the presence of this drug, 5-aminoimidazole-4- carboxamide ribonucleotide (AICAR; see Fig. 22–33) accu- mulates in the culture medium. These effects are reversed by addition of excess p-aminobenzoate. (a) What is the role of p-aminobenzoate in these bacte- ria? (Hint: See Fig. 18–16). (b) Why does AICAR accumulate in the presence of sul- fanilamide? (c) Why are the inhibition and accumulation reversed by addition of excess p-aminobenzoate? 12. Pathway of Carbon in Pyrimidine Biosynthesis Predict the locations of 14 C in orotate isolated from cells grown on a small amount of uniformly labeled [ 14 C]succinate. Justify your prediction. 13. Nucleotides As Poor Sources of Energy Under starvation conditions, organisms can use proteins and amino acids as sources of energy. Deamination of amino acids pro- duces carbon skeletons that can enter the glycolytic pathway and the citric acid cycle to produce energy in the form of ATP. Nucleotides, on the other hand, are not similarly de- graded for use as energy-yielding fuels. What observations about cellular physiology support this statement? What as- pect of the structure of nucleotides makes them a relatively poor source of energy? 14. Treatment of Gout Allopurinol (see Fig. 22–47), an inhibitor of xanthine oxidase, is used to treat chronic gout. Explain the biochemical basis for this treatment. Patients treated with allopurinol sometimes de- velop xanthine stones in the kidneys, although the incidence of kidney damage is much lower than in untreated gout. Ex- plain this observation in the light of the following solubilities in urine: uric acid, 0.15 g/L; xanthine, 0.05 g/L; and hypox- anthine, 1.4 g/L. 15. Inhibition of Nucleotide Synthesis by Azaserine The diazo compound O-(2-diazoacetyl)-L-serine, known also as azaserine (see Fig. 22–48), is a powerful inhibitor of glut- amine amidotransferases. If growing cells are treated with azaserine, what intermediates of nucleotide biosynthesis would accumulate? Explain. O H5008 O SCH 2 NH 2 O O NNH 2 p-Aminobenzoate Sulfanilamide 8885d_c22_833-880 2/6/04 8:35 AM Page 880 mac76 mac76:385_reb: chapter I n Chapters 13 through 22 we have discussed metabo- lism at the level of the individual cell, emphasizing cen- tral pathways common to almost all cells, prokaryotic and eukaryotic. We have seen how metabolic processes within cells are regulated at the level of individual en- zyme reactions, by substrate availability, by allosteric mechanisms, and by phosphorylation or other covalent modifications of enzymes. To appreciate fully the significance of individual metabolic pathways and their regulation, we must view these pathways in the context of the whole organism. An essential characteristic of multicellular organisms is cell differentiation and division of labor. The specialized functions of the tissues and organs of complex organ- isms such as humans impose characteristic fuel re- quirements and patterns of metabolism. Hormonal sig- nals integrate and coordinate the metabolic activities of different tissues and optimize the allocation of fuels and precursors to each organ. In this chapter we focus on mammals, looking at the specialized metabolism of several major organs and tis- sues and the integration of metabolism in the whole or- ganism. We begin by examining the broad range of hor- mones and hormonal mechanisms, then turn to the tissue-specific functions regulated by these mecha- nisms. We discuss the distribution of nutrients to vari- ous organs—emphasizing the central role played by the liver—and the metabolic cooperation among these or- gans. To illustrate the integrative role of hormones, we describe the interplay of insulin, glucagon, and epi- nephrine in coordinating fuel metabolism in muscle, liver, and adipose tissue. The metabolic disturbances in diabetes further illustrate the importance of hormonal regulation of metabolism. Finally we discuss the long- term hormonal regulation of body mass. 23.1 Hormones: Diverse Structures for Diverse Functions Virtually every process in a complex organism is regu- lated by one or more hormones: maintenance of blood pressure, blood volume, and electrolyte balance; em- bryogenesis; sexual differentiation, development, and reproduction; hunger, eating behavior, digestion, and HORMONAL REGULATION AND INTEGRATION OF MAMMALIAN METABOLISM 23.1 Hormones: Diverse Structures for Diverse Functions 881 23.2 Tissue-Specific Metabolism: The Division of Labor 892 23.3 Hormonal Regulation of Fuel Metabolism 902 23.4 Obesity and the Regulation of Body Mass 910 We recognize that each tissue and, more generally, each cell of the organism secretes . . . special products or ferments into the blood which thereby influence all the other cells thus integrated with each other by a mechanism other than the nervous system. —Charles édouard Brown-Séquard and J. d’Arsonval, article in Comptes Rendus de la Société de Biologie, 1891 23 881 8885d_c23_881-919 3/1/04 1:26 PM Page 881 mac76 mac76:385_reb: fuel allocation—to name but a few. We examine here the methods for detecting and measuring hormones and their interaction with receptors, and consider a repre- sentative selection of hormone types. The coordination of metabolism in mammals is achieved by the neuroendocrine system. Individual cells in one tissue sense a change in the organism’s cir- cumstances and respond by secreting a chemical mes- senger that passes to another cell in the same or dif- ferent tissue, where it binds to a receptor molecule and triggers a change in this second cell. In neuronal sig- naling (Fig. 23–1a), the chemical messenger (neuro- transmitter; acetylcholine, for example) may travel only a fraction of a micrometer, across the synaptic cleft to the next neuron in a network. In hormonal signaling, the messengers—hormones—are carried in the bloodstream to neighboring cells or to distant organs and tissues; they may travel a meter or more before encountering their target cell (Fig. 23–1b). Except for this anatomic difference, these two chemical signaling mechanisms are remarkably similar. Epinephrine and norepinephrine, for example, serve as neurotransmitters in certain synapses of the brain and smooth muscle and as hor- mones that regulate fuel metabolism in liver and muscle. The following discussion of cellular signaling emphasizes hormone action, drawing on discussions of fuel metab- olism in earlier chapters, but most of the fundamental mechanisms described here also occur in neurotrans- mitter action. The Discovery and Purification of Hormones Requires a Bioassay How is a hormone discovered and isolated? First, re- searchers find that a physiological process in one tissue depends on a signal that originates in another tissue. In- sulin, for example, was first recognized as a substance that is produced in the pancreas and affects the volume and composition of urine (Box 23–1). Once a physiological ef- fect of the putative hormone is discovered, a quantitative bioassay for the hormone can be developed. In the case of insulin, the assay consisted of injecting extracts of pan- creas (a crude source of insulin) into experimental ani- mals deficient in insulin, then quantifying the resulting changes in glucose concentration in blood and urine. To isolate a hormone, the biochemist fractionates extracts containing the putative hormone, with the same tech- niques used to purify other biomolecules (solvent frac- tionation, chromatography, and electrophoresis), and then assays each fraction for hormone activity. Once the chem- ical has been purified, its composition and structure can be determined. This protocol for hormone characterization is de- ceptively simple. Hormones are extremely potent and are produced in very small amounts. Obtaining sufficient hormone to allow its chemical characterization often in- volves biochemical isolations on a heroic scale. When Roger Guillemin and Andrew Schally independently purified and characterized thyrotropin-releasing hor- mone (TRH) from the hypothalamus, Schally’s group processed about 20 tons of hypothalamus from nearly two million sheep, and Guillemin’s group extracted the Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism882 (a) Neuronal signaling (b) Endocrine signaling Target cells Nerve impulse Contraction Bloodstream Metabolic change SecretionNerve impulse FIGURE 23–1 Signaling by the neuroendocrine system. (a) In neu- ronal signaling, electrical signals (nerve impulses) originate in the cell body of a neuron and travel very rapidly over long distances to the axon tip, where neurotransmitters are released and diffuse to the tar- get cell. The target cell (another neuron, a myocyte, or a secretory cell) is only a fraction of a micrometer or a few micrometers away from the site of neurotransmitter release. (b) In the endocrine system, hor- mones are secreted into the bloodstream, which carries them through- out the body to target tissues that may be a meter or more away from the secreting cell. Both neurotransmitters and hormones interact with specific receptors on or in their target cells, triggering responses. 8885d_c23_881-919 3/1/04 1:26 PM Page 882 mac76 mac76:385_reb: 23.1 Hormones: Diverse Structures for Diverse Functions 883 BOX 23–1 BIOCHEMISTRY IN MEDICINE How Is a Hormone Discovered? The Arduous Path to Purified Insulin Millions of people with type I (insulin-dependent) di- abetes mellitus inject themselves daily with pure in- sulin to compensate for the lack of production of this critical hormone by their own pancreatic H9252 cells. Insulin injection is not a cure for diabetes, but it allows peo- ple who otherwise would have died young to lead long and productive lives. The discovery of insulin, which began with an accidental observation, illustrates the combination of serendipity and careful experimenta- tion that led to the discovery of many of the hormones. In 1889, Oskar Minkowski, a young assistant at the Medical College of Strasbourg, and Josef von Mering, at the Hoppe-Seyler Institute in Strasbourg, had a friendly disagreement about whether the pancreas, known to contain lipases, was important in fat diges- tion in dogs. To resolve the issue, they began an ex- periment on fat digestion. They surgically removed the pancreas from a dog, but before their experiment got any farther, Minkowski noticed that the dog was now producing far more urine than normal (a common symptom of untreated diabetes). Also, the dog’s urine had glucose levels far above normal (another symp- tom of diabetes). These findings suggested that lack of some pancreatic product caused diabetes. Minkowski tried unsuccessfully to prepare an ex- tract of dog pancreas that would reverse the effect of removing the pancreas—that is, would lower the uri- nary or blood glucose levels. We now know that insulin is a protein, and that the pancreas is very rich in pro- teases (trypsin and chymotrypsin), normally released directly into the small intestine to aid in digestion. These proteases doubtless degraded the insulin in the pancreatic extracts in Minkowski’s experiments. Despite considerable effort, no significant prog- ress was made in the isolation or characterization of the “antidiabetic factor” until the summer of 1921, when Frederick G. Banting, a young scientist working in the laboratory of J. J. R. MacLeod at the University of Toronto, and a student assistant, Charles Best, took up the problem. By that time, several lines of evidence pointed to a group of specialized cells in the pancreas (the islets of Langerhans; see Fig. 23–24) as the source of the antidiabetic factor, which came to be called in- sulin (from Latin insula, “island”). Taking precautions to prevent proteolysis, Bant- ing and Best (later aided by biochemist J. B. Collip) succeeded in December 1921 in preparing a purified pancreatic extract that cured the symptoms of ex- perimental diabetes in dogs. On January 25, 1922 (just one month later!), their insulin preparation was in- jected into Leonard Thompson, a 14-year-old boy se- verely ill with diabetes mellitus. Within days, the lev- els of ketone bodies and glucose in Thompson’s urine dropped dramatically; the extract saved his life. In 1923, Banting and MacLeod won the Nobel Prize for their isolation of insulin. Banting immediately an- nounced that he would share his prize with Best; MacLeod shared his with Collip. By 1923, pharmaceutical companies were supply- ing thousands of patients throughout the world with in- sulin extracted from porcine pancreas. With the devel- opment of genetic engineering techniques in the 1980s (Chapter 9), it became possible to produce unlimited quantities of human insulin by inserting the cloned hu- man gene for insulin in a microorganism, which was then cultured on an industrial scale. Some patients with diabetes are now fitted with implanted insulin pumps, which release adjustable amounts of insulin on demand to meet changing needs at meal times and during ex- ercise. There is a reasonable prospect that, in the fu- ture, transplantation of pancreatic tissue will provide diabetic patients with a source of insulin that responds as well as normal pancreas, releasing insulin into the bloodstream only when blood glucose rises. Frederick G. Banting, 1891–1941 J. J. R. MacLeod, 1876–1935 Charles Best, 1899–1978 J. B. Collip, 1892–1965 8885d_c23_881-919 3/1/04 1:26 PM Page 883 mac76 mac76:385_reb: hypothalamus from about a million pigs! TRH proved to be a simple derivative of the tripeptide Glu–His–Pro (Fig. 23–2). Once the structure of the hormone was known, it could be chemically synthesized in large quan- tities for use in physiological and biochemical studies. For their work on hypothalamic hormones, Schally and Guillemin shared the Nobel Prize in Physiology or Medicine in 1977, along with Rosalyn Yalow, who (with Solomon A. Berson) developed the extraordinarily sen- sitive radioimmunoassay (RIA) for peptide hormones and used it to study hormone action. RIA revolutionized hormone research by making possible the rapid, quan- titative, and specific measurement of hormones in minute amounts. Hormone-specific antibodies are the key to the ra- dioimmunoassay. Purified hormone, injected into rab- bits, elicits antibodies that bind to that hormone with very high affinity and specificity. When a constant amount of isolated antibody is incubated with a fixed amount of the radioactively labeled hormone, a certain fraction of the radioactive hormone binds to the anti- body (Fig. 23–3). If, in addition to the radiolabeled hor- mone, unlabeled hormone is also present, the unlabeled hormone competes with and displaces some of the la- beled hormone from its binding site on the antibody. This binding competition can be quantified by reference to a standard curve obtained with known amounts of unla- beled hormone. The degree to which labeled hormone is displaced from antibody is a measure of the amount of unlabeled hormone in a sample of blood or tissue extract. By using very highly radioactive hormone, re- searchers can make the assay sensitive to picograms of hormone. A newer variation of this technique, enzyme- linked immunosorbent assay (ELISA), is illustrated in Figure 5–28b. Hormones Act through Specific High-Affinity Cellular Receptors As we saw in Chapter 12, all hormones act through highly specific receptors in hormone-sensitive target cells, to which the hormones bind with high affinity (see Fig. 12–2). Each cell type has its own combination of hormone receptors, which define the range of its hor- mone responsiveness. Moreover, two cell types with the same type of receptor may have different intracellular targets of hormone action and thus may respond dif- ferently to the same hormone. The specificity of hor- mone action results from structural complementarity between the hormone and its receptor; this interaction is extremely selective, so structurally similar hormones can have different effects. The high affinity of the in- teraction allows cells to respond to very low concentra- tions of hormone. In the design of drugs intended to intervene in hormonal regulation, we need to know the relative specificity and affinity of the drug and the nat- ural hormone. Recall that hormone-receptor interac- tions can be quantified by Scatchard analysis (see Box 12–1), which, under favorable conditions, yields a quan- titative measure of affinity (the dissociation constant for the complex) and the number of hormone-binding sites in a preparation of receptor. Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism884 FIGURE 23–2 The structure of thyrotropin-releasing hormone (TRH). Purified (by heroic efforts) from extracts of hypothalamus, TRH proved to be a derivative of the tripeptide Glu–His–Pro. The side-chain car- boxyl group of the amino-terminal Glu forms an amide (red bond) with the residue’s H9251-amino group, creating pyroglutamate, and the car- boxyl group of the carboxyl-terminal Pro is converted to an amide (red ONH 2 ). Such modifications are common among the small pep- tide hormones. In a typical protein of M r ~50, the charges on the amino- and carboxyl-terminal groups contribute relatively little to the overall charge on the protein, but in a tripeptide these two charges dominate the properties of the peptide. Formation of the amide de- rivatives removes these charges. CO NH CH 2 CH C O NH CH C CH 2 Histidine ProlylamidePyroglutamate CH C CH 2 CH 2 HC pyroGlu–His–Pro–NH 2 N NH CH CH 2 CH 2 NH 2 C O O N Roger Guillemin Rosalyn S. Yalow Andrew V. Schally 8885d_c23_881-919 3/1/04 1:26 PM Page 884 mac76 mac76:385_reb: The locus of the encounter between hormone and receptor may be extracellular, cytosolic, or nuclear, de- pending on the hormone type. The intracellular conse- quences of hormone-receptor interaction are of at least six general types: (1) a change in membrane potential results from the opening or closing of a hormone-gated 23.1 Hormones: Diverse Structures for Diverse Functions 885 ion channel; (2) a receptor enzyme is activated by the extracellular hormone; (3) a second messenger (such as cAMP or inositol trisphosphate) generated inside the cell acts as an allosteric regulator of one or more en- zymes; (4) a receptor with no intrinsic enzyme activity recruits a soluble protein kinase in the cytosol, which passes on the signal; (5) an adhesion receptor on the cell surface interacts with molecules in the extracellu- lar matrix and conveys information to the cytoskeleton; or (6) a steroid or steroidlike molecule causes a change in the level of expression (transcription of DNA into mRNA) of one or more genes, mediated by a nuclear hormone receptor protein (see Fig. 12–2). Water-soluble peptide and amine hormones (insulin and epinephrine, for example) act extracellularly by binding to cell surface receptors that span the plasma membrane (Fig. 23–4). When the hormone binds to its extracellular domain, the receptor undergoes a confor- mational change analogous to that produced in an al- losteric enzyme by binding of an effector molecule. The conformational change triggers the downstream effects of the hormone. A single hormone molecule, in forming a hormone- receptor complex, activates a catalyst that produces many molecules of second messenger, so the receptor Antibody Radiolabeled hormone 1 Radiolabeled and unlabeled hormone 2 Radiolabeled ACTH [unbound] Unlabeled ACTH added (pg) () 0 1000100 Unknown sample 101 0.2 0.4 0.6 0.8 1.0 1.2 ratio [bound] Standard curve FIGURE 23–3 Radioimmunoassay (RIA). (a) A low concentration of radiolabeled hormone (red) is incubated with 1 a fixed amount of antibody specific for that hormone or 2 a fixed amount of antibody and various concentrations of unlabeled hormone (blue). In the latter case, unlabeled hormone competes with labeled hormone for bind- ing to the antibody; the amount of labeled hormone bound varies in- versely with the concentration of unlabeled hormone present. (b) A radioimmunoassay for adrenocorticotropic hormone (ACTH). A stan- dard curve of the ratio [bound] to [unbound radiolabeled ACTH] vs. [unlabeled ACTH added] is constructed and used to determine the amount of (unlabeled) ACTH in an unknown sample. If an aliquot con- taining an unknown quantity of unlabeled hormone gives, say, a value of 0.4 for the ratio [bound]/[unbound] (see arrow), the aliquot must contain about 20 pg of ACTH. Rec Rec Peptide or amine hormone binds to receptor on the outside of the cell; acts through receptor without entering the cell. Steroid or thyroid hormone enters the cell; hormone- receptor complex acts in the nucleus. Second messenger (e.g., cAMP) Plasma membrane Nucleus Altered transcription of specific genes Altered activity of preexisting enzyme Altered amount of newly synthesized proteins FIGURE 23–4 Two general mechanisms of hormone action. The pep- tide and amine hormones are faster acting than steroid and thyroid hormones. (a) (b) 8885d_c23_885 3/2/04 7:59 AM Page 885 mac76 mac76:385_reb: serves not only as a signal transducer but also as a sig- nal amplifier. The signal may be further amplified by a signaling cascade, a series of steps in which a catalyst activates a catalyst, resulting in very large amplifications of the original signal. A cascade of this type occurs in the regulation of glycogen synthesis and breakdown by epinephrine (see Fig. 12–16). Epinephrine activates (through its receptor) adenylyl cyclase, which produces many molecules of cAMP for each molecule of receptor- bound hormone. Cyclic AMP in turn activates cAMP- dependent protein kinase, which activates phosphory- lase kinase, which activates glycogen phosphorylase. The result is signal amplification: one epinephrine mol- ecule causes the production of many thousands of molecules of glucose 1-phosphate from glycogen. Water-insoluble hormones (steroid, retinoid, and thy- roid hormones) readily pass through the plasma mem- brane of their target cells to reach their receptor proteins in the nucleus (Fig. 23–4). With this class of hormones, the hormone-receptor complex itself carries the message; it interacts with DNA to alter the expression of specific genes, changing the enzyme complement of the cell and thereby changing cellular metabolism (see Fig. 12–40). Hormones that act through plasma membrane re- ceptors generally trigger very rapid physiological or bio- chemical responses. Just seconds after the adrenal medulla secretes epinephrine into the bloodstream, skeletal muscle responds by accelerating the breakdown of glycogen. By contrast, the thyroid hormones and the sex (steroid) hormones promote maximal responses in their target tissues only after hours or even days. These differences in response time correspond to different modes of action. In general, the fast-acting hormones lead to a change in the activity of one or more preex- isting enzymes in the cell, by allosteric mechanisms or covalent modification. The slower-acting hormones gen- erally alter gene expression, resulting in the synthesis of more or less of the regulated protein(s). Hormones Are Chemically Diverse Mammals have several classes of hormones, distin- guishable by their chemical structures and their modes of action (Table 23–1). Peptide, amine, and eicosanoid hormones act from outside the target cell via surface re- ceptors. Steroid, vitamin D, retinoid, and thyroid hor- mones enter the cell and act through nuclear receptors. Nitric oxide also enters the cell, but activates a cytoso- lic enzyme, guanylyl cyclase (see Fig. 12–10). Hormones can also be classified by the way they get from the point of their release to their target tissue. En- docrine (from the Greek endon, “within,” and krinein, “to release”) hormones are released into the blood and carried to target cells throughout the body (insulin is an example). Paracrine hormones are released into the extracellular space and diffuse to neighboring target cells (the eicosanoid hormones are of this type). Au- tocrine hormones are released by and affect the same cell, binding to receptors on the cell surface. Mammals are hardly unique in possessing hormonal signaling systems. Insects and nematode worms have highly developed systems for hormonal regulation, with fundamental mechanisms similar to those in mammals. Plants, too, use hormonal signals to coordinate the ac- tivities of their various tissues (Chapter 12). The study of hormone action is not as advanced in plants as in ani- mals, but we do know that some mechanisms are shared. To illustrate the structural diversity and range of action of mammalian hormones, we consider representative examples of each major class listed in Table 23–1. Peptide Hormones Peptide hormones may have from 3 to 200 or more amino acid residues. They include the pancreatic hormones insulin, glucagon, and somato- statin, the parathyroid hormone, calcitonin, and all the hormones of the hypothalamus and pituitary (described below). These hormones are synthesized on ribosomes in the form of longer precursor proteins (prohormones), Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism886 TABLE 23–1 Classes of Hormones Type Example Synthetic path Mode of action Peptide Insulin, glucagon Proteolytic processing of prohormone Catecholamine Epinephrine From tyrosine Plasma membrane receptors; second messengers Eicosanoid PGE 1 From arachidonate (20:4 fatty acid) Steroid Testosterone From cholesterol Vitamin D 1,25-Dihydroxycholecalciferol From cholesterol Retinoid Retinoic acid From vitamin A Thyroid Triiodothyronine (T 3 ) From Tyr in thyroglobulin Nitric oxide Nitric oxide From arginine H11001 O 2 Cytosolic receptor (guanylate cyclase) and second messenger (cGMP) Nuclear receptors; transcriptional regulation ? ? ? ? ? ? ? ? ? ? ? ? 8885d_c23_881-919 3/1/04 1:26 PM Page 886 mac76 mac76:385_reb: then packaged into secretory vesicles and proteolyti- cally cleaved to form the active peptides. Insulin is a small protein (M r 5,800) with two polypeptide chains, A and B, joined by two disulfide bonds. It is synthesized in the pancreas as an inactive single-chain precursor, preproinsulin (Fig. 23–5), with an amino-terminal “sig- nal sequence” that directs its passage into secretory vesicles. (Signal sequences are discussed in Chapter 27; see Fig. 27–33.) Proteolytic removal of the signal se- quence and formation of three disulfide bonds produces proinsulin, which is stored in secretory granules in pan- creatic H9252 cells. When elevated blood glucose triggers in- sulin secretion, proinsulin is converted to active insulin by specific proteases, which cleave two peptide bonds to form the mature insulin molecule. In some cases, prohormone proteins yield a single peptide hormone, but often several active hormones are carved out of the same prohormone. Pro-opiomelano- cortin (POMC) is a spectacular example of multiple hormones encoded by a single gene. The POMC gene encodes a large polypeptide that is progressively carved up into at least nine biologically active peptides (Fig. 23–6). The terminal residues of peptide hormones are often modified, as in TRH (Fig. 23–2). 23.1 Hormones: Diverse Structures for Diverse Functions 887 C COO H11002 C peptideSignal sequence Signal sequence A B B chainA chain Preproinsulin Proinsulin Mature insulin COO H11002 S S COO H11002 NH 3 H11001 NH 3 H11001 NH 3 H11001 H 3 N H11001 SS SS S S SS SS H11002 OOC – FIGURE 23–5 Insulin. Mature insulin is formed from its larger precursor preproinsulin by proteolytic processing. Removal of a 23 amino acid segment (the signal sequence) at the amino terminus of preproinsulin and formation of three disulfide bonds produces proinsulin. Further proteolytic cuts remove the C peptide from proinsulin to produce mature insulin, composed of A and B chains. The amino acid sequence of bovine insulin is shown in Figure 3–24. Pro-opiomelanocortin (POMC) gene Signal peptide 5H11032 3H11032 mRNA DNA -Lipotropin -MSH Met-enkephalin CLIP-MSH -Lipotropin COO H11002 H 3 N ACTH-MSH H11001 H9253 H9251 H9253 H9252 H9252 -EndorphinH9252 FIGURE 23–6 Proteolytic processing of the pro- opiomelanocortin (POMC) precursor. The initial gene product of the POMC gene is a long polypeptide that undergoes cleavage by a series of specific proteases to produce ACTH, H9252- and H9253-lipotropin, H9251-, H9252-, and H9253-MSH (melanocyte-stimulating hormone), CLIP (corticotropin- like intermediary peptide), H9252-endorphin, and Met-enkephalin. The points of cleavage are paired basic residues, Arg–Lys, Lys–Arg, or Lys–Lys. 8885d_c23_881-919 3/1/04 1:26 PM Page 887 mac76 mac76:385_reb: The concentration of peptide hormones within se- cretory granules is so high that the vesicle contents are virtually crystalline; when the contents are released by exocytosis, a large amount of hormone is released sud- denly. The capillaries that serve peptide-producing en- docrine glands are fenestrated (and thus permeable to peptides), so the hormone molecules readily enter the bloodstream for transport to target cells elsewhere. As noted earlier, all peptide hormones act by binding to re- ceptors in the plasma membrane. They cause the gen- eration of a second messenger in the cytosol, which changes the activity of an intracellular enzyme, thereby altering the cell’s metabolism. Catecholamine Hormones The water-soluble compounds epinephrine (adrenaline) and norepinephrine (nor- adrenaline) are catecholamines, named for the struc- turally related compound catechol. They are synthesized from tyrosine. Catecholamines produced in the brain and in other neural tissues function as neurotransmitters, but epinephrine and norepinephrine are also hormones, syn- thesized and secreted by the adrenal glands. Like the peptide hormones, catecholamines are highly concen- trated within secretory vesicles and released by exocy- tosis, and they act through surface receptors to gener- ate intracellular second messengers. They mediate a wide variety of physiological responses to acute stress (see Table 23–6). Eicosanoids The eicosanoid hormones (prostaglandins, thromboxanes, and leukotrienes) are derived from the 20-carbon polyunsaturated fatty acid arachidonate. Unlike the hormones described above, they are not synthesized in advance and stored; they are produced, when needed, from arachidonate enzymatically released from membrane phospholipids by phospholipase A 2 (see Fig. 10–18). The enzymes of the pathway leading to prostaglandins and thromboxanes (see Fig. 21–15) are very widely distributed in mammalian tissues; most cells Phospholipids Arachidonate (20:4) LeukotrienesProstaglandins Thromboxanes Tyrosine L-DOPA Dopamine Norepinephrine Epinephrine can produce these signals, and cells of many tissues can respond to them through specific plasma membrane re- ceptors. The eicosanoid hormones are paracrine hor- mones, secreted into the interstitial fluid (not primarily into the blood) and acting on nearby cells. Prostaglandins promote the contraction of smooth muscle, including that of the intestine and uterus (and can therefore be used medically to in- duce labor). They also mediate pain and inflammation in all tissues. Many antiinflammatory drugs act by in- hibiting steps in the prostaglandin synthetic pathway (see Box 21–2). Thromboxanes regulate platelet func- tion and therefore blood clotting. Leukotrienes LTC 4 and LTD 4 act through plasma membrane receptors to stimulate contraction of smooth muscle in the intestine, pulmonary airways, and trachea. They are mediators of the severe immune response called anaphylaxis. ■ Steroid Hormones The steroid hormones (adrenocorti- cal hormones and sex hormones) are synthesized from cholesterol in several endocrine tissues. They travel to their target cells through the blood- stream, bound to carrier proteins. More than 50 corti- costeroid hormones are produced in the adrenal cortex by reactions that remove the side chain from the D ring of cholesterol and introduce oxygen to form keto and hydroxyl groups. Many of these reactions involve cy- tochrome P-450 enzymes (see Box 21–1). The steroid hormones are of two general types. Glucocorticoids (such as cortisol) primarily affect the metabolism of car- bohydrates; mineralocorticoids (such as aldosterone) regulate the concentrations of electrolytes in the blood. Androgens (testosterone) and estrogens (such as estra- diol; see Fig. 10–19) are synthesized in the testes and ovaries. Their synthesis also involves cytochrome P-450 enzymes that cleave the side chain of cholesterol and introduce oxygen atoms. These hormones affect sexual development, sexual behavior, and a variety of other re- productive and nonreproductive functions. All steroid hormones act through nuclear recep- tors to change the level of expression of specific genes (p. 465). Recent evidence indicates that they also have more rapid effects, mediated by receptors localized in the plasma membrane. Cholesterol Progesterone Cortisol (glucocorticoid) Aldosterone (mineralocorticoid) Testosterone Estradiol (sex hormones) Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism888 8885d_c23_881-919 3/1/04 1:26 PM Page 888 mac76 mac76:385_reb: Vitamin D Hormone Calcitriol (1,25-dihydroxycholecal- ciferol) is produced from vitamin D by enzyme- catalyzed hydroxylation in the liver and kidneys (see Fig. 10–20a). Vitamin D is obtained in the diet or by photolysis of 7- dehydrocholesterol in skin exposed to sunlight. Cal- citriol works in concert with parathyroid hormone in Ca 2H11001 homeostasis, regulating [Ca 2H11001 ] in the blood and the balance between Ca 2H11001 deposition and Ca 2H11001 mobi- lization from bone. Acting through nuclear receptors, calcitriol activates the synthesis of an intestinal Ca 2H11001 - binding protein essential for uptake of dietary Ca 2H11001 . In- adequate dietary vitamin D or defects in the biosynthe- sis of calcitriol result in serious diseases such as rickets, in which bones are weak and malformed (see Fig. 10–20b). Retinoid Hormones Retinoids are potent hormones that regulate the growth, survival, and differentiation of cells via nuclear retinoid receptors. The prohormone retinol is synthesized from vitamin A, primarily in liver (see Fig. 10–21), and many tissues convert retinol to the hormone retinoic acid (RA). All tissues are retinoid targets, as all cell types have at least one form of nuclear retinoid receptor. In adults, the most significant targets include cornea, skin, epithelia of the lungs and trachea, and the immune system. RA reg- ulates the synthesis of proteins essential for growth or differentiation. Excessive vitamin A can cause birth de- fects, and pregnant women are advised not to use the retinoid creams that have been developed for treatment of severe acne. Thyroid Hormones The thyroid hormones T 4 (thyroxine) and T 3 (triiodothyronine) are synthesized from the pre- cursor protein thyroglobulin (M r 660,000). Up to 20 Tyr residues in thyroglobulin are enzymatically iodinated b-Carotene Vitamin A 1 (retinol) Retinoic acid 7-Dehydrocholesterol UV light Vitamin D 3 (cholecalciferol) 25-Hydroxycholecalciferol 1,25-Dihydroxycholecalciferol in the thyroid gland, then two iodotyrosine residues con- dense to form the precursor to thyroxine. When needed, thyroxine is released by proteolysis. Condensation of monoiodotyrosine with diiodotyrosine produces T 3 , which is also an active hormone released by proteolysis. The thyroid hormones act through nuclear receptors to stimulate energy-yielding metabolism, especially in liver and muscle, by increasing the expression of genes en- coding key catabolic enzymes. Nitric Oxide (NO) Nitric oxide is a relatively stable free radical synthesized from molecular oxygen and the guanidino nitrogen of arginine (see Fig. 22–31) in a re- action catalyzed by NO synthase. Arginine H11001 1 1 ? 2 NADPH H11001 2O 2 ?→ NO H11001 citrulline H11001 2H 2 O H11001 1 1 ? 2 NADP H11001 This enzyme is found in many tissues and cell types: neurons, macrophages, hepatocytes, myocytes of smooth muscle, endothelial cells of the blood vessels, and epithelial cells of the kidney. NO acts near its point of release, entering the target cell and activating the cy- tosolic enzyme guanylyl cyclase, which catalyzes the for- mation of the second messenger cGMP (see Fig. 12–10). Hormone Release Is Regulated by a Hierarchy of Neuronal and Hormonal Signals The changing levels of specific hormones regulate spe- cific cellular processes, but what regulates the level of each hormone? The brief answer is that the central nerv- ous system receives input from many internal and ex- ternal sensors—signals about danger, hunger, dietary in- take, blood composition and pressure, for example—and orchestrates the production of appropriate hormonal signals by the endocrine tissues. For a more complete answer, we must look at the hormone-producing sys- tems of the human body and some of their functional interrelationships. Figure 23–7 shows the anatomic location of the ma- jor endocrine glands in humans, and Figure 23–8 rep- resents the “chain of command” in the hormonal sig- naling hierarchy. The hypothalamus, a small region of the brain (Fig. 23–9), is the coordination center of the Thyroglobulin–Tyr proteolysis Thyroglobulin–Tyr–I (iodinated Tyr residues) Thyroxine (T 4 ), triiodothyronine (T 3 ) 23.1 Hormones: Diverse Structures for Diverse Functions 889 8885d_c23_889 3/2/04 7:59 AM Page 889 mac76 mac76:385_reb: endocrine system; it receives and integrates messages from the central nervous system. In response to these messages, the hypothalamus produces regulatory hor- mones (releasing factors) that pass directly to the nearby pituitary gland, through special blood vessels and neurons that connect the two glands (Fig. 23–9b). The pituitary gland has two functionally distinct parts. The posterior pituitary contains the axonal endings of many neurons that originate in the hypothalamus. These neurons produce the short peptide hormones oxytocin and vasopressin (Fig. 23–10), which then move down the axon to the nerve endings in the pituitary, where they are stored in secretory granules to await the signal for their release. The anterior pituitary responds to hypothalamic hormones carried in the blood, producing tropic hor- mones, or tropins (from the Greek tropos, “turn”). These relatively long polypeptides activate the next rank of endocrine glands (Fig. 23–8), which includes the adrenal cortex, thyroid gland, ovaries, and testes. These glands in turn secrete their specific hormones, which are carried in the bloodstream to the receptors of cells in the target tissues. For example, corticotropin-releasing hormone from the hypothalamus stimulates the anterior pituitary to release ACTH, which travels to the zona fas- ciculata of the adrenal cortex and triggers the release of Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism890 Hypothalamus Pituitary Thyroid Adrenals Pancreas Kidneys Ovaries (female) Testes (male) Parathyroids (behind the thyroid) Adipose tissue FIGURE 23–7 The major endocrine glands. The glands are shaded dark pink. Sensory input from environment Central nervous system Hypothalamus Anterior pituitary Neuroendocrine origins of signals First targets Ultimate targets Many tissues Reproductive organs Thyroxine (T 4 ), triiodo- thyronine (T 3 ) Progesterone, estradiol Testosterone ThyroidSecond targets Adrenal cortex Ovaries/testes Thyrotropin M r 28,000 Corticotropin (ACTH) M r 4,500 Follicle- stimulating hormone M r 24,000 Luteinizing hormone M r 20,500 Muscles, liver Cortisol, corticosterone, aldosterone Liver, bone Mammary glands Smooth muscle, mammary glands Arterioles, kidney Liver, muscles Liver, muscles, heart Somatotropin (growth hormone) M r 21,500 Prolactin M r 22,000 Oxytocin M r 1,007 Blood glucose level Posterior pituitary Insulin, glucagon, somatostatin Epinephrine Adrenal medulla Islet cells of pancreas Vasopressin (antidiuretic hormone) M r 1,040 Hypothalamic hormones (releasing factors) FIGURE 23–8 The major endocrine systems and their target tissues. Signals originating in the central nervous system (top) pass via a series of relays to the ultimate target tissues (bottom). In addition to the systems shown, the thymus, pineal gland, and groups of cells in the gastrointestinal tract also secrete hormones. Dashed lines represent neuronal connections. 8885d_c23_881-919 3/1/04 1:26 PM Page 890 mac76 mac76:385_reb: cortisol. Cortisol, the ultimate hormone in this cascade, acts through its receptor in many types of target cells to alter their metabolism. In hepatocytes, one effect of cortisol is to increase the rate of gluconeogenesis. Hormonal cascades such as those responsible for the release of cortisol and epinephrine result in large ampli- fications of the initial signal and allow exquisite fine- tuning of the output of the ultimate hormone (Fig. 23–11). At each level in the cascade, a small signal elicits a larger response. The initial electrical signal to the hypothalamus results in the release of a few nanograms of corticotropin-releasing hormone, which elicits the release of a few micrograms of corticotropin. Corticotropin acts on the adrenal cortex to cause the re- lease of milligrams of cortisol, for an overall amplifica- tion of at least a millionfold. At each level of a hormonal cascade, feedback in- hibition of earlier steps in the cascade is possible; an unnecessarily elevated level of the ultimate hormone or of one of the intermediate hormones inhibits the re- lease of earlier hormones in the cascade. These feed- back mechanisms accomplish the same end as those that limit the output of a biosynthetic pathway (compare Fig. 23–11 with Fig. 6–28): a product is synthesized (or released) only until the necessary concentration is reached. 23.1 Hormones: Diverse Structures for Diverse Functions 891 FIGURE 23–10 Two hormones of the posterior pituitary gland. The carboxyl-terminal residues are glycinamide, ONHOCH 2 OCONH 2 (as noted in Fig. 23–2, amidation of the carboxyl terminus is common in short peptide hormones). These two hormones, identical in all but two residues (shaded), have very different biological effects. Oxytocin acts on the smooth muscles of the uterus and mammary gland, caus- ing uterine contractions during labor and promoting milk release dur- ing lactation. Vasopressin (also called antidiuretic hormone) increases water reabsorption in the kidney and promotes the constriction of blood vessels, thereby increasing blood pressure. (a) (b) Posterior pituitary Hypothalamus Arteries Release of posterior pituitary hormones (vasopressin, oxytocin) Release of hypothalamic factors into arterial blood Capillary network Release of anterior pituitary hormones (tropins) Nerve axons Hypothalamus Anterior pituitary Afferent nerve signals to hypothalamus Veins carry hormones to systemic blood Anterior pituitary Posterior pituitary Cys Tyr Ile Gln Asn Pro Leu Gly CO Cys S S NH 2 Cys Tyr Phe Gln Asn Pro Arg Gly CO NH 3 Cys S S H11001 Human vasopressin (antidiuretic hormone) Human oxytocin NH 2 NH 3 H11001 FIGURE 23–9 Neuroendocrine origins of hormone signals. (a) Location of the hypothalamus and pituitary gland. (b) Details of the hypothalamus- pituitary system. Signals from connecting neurons stimulate the hypothalamus to secrete releasing factors into a blood vessel that carries the hormones directly to a capillary network in the anterior pituitary. In response to each hypothalamic releasing factor, the anterior pituitary releases the appropriate hormone into the general circulation. Posterior pituitary hormones are synthe- sized in neurons arising in the hypothalamus, transported along axons to nerve endings in the posterior pituitary, and stored there until released into the blood in response to a neuronal signal. (a) 8885d_c23_891 3/2/04 8:00 AM Page 891 mac76 mac76:385_reb: ■ Hormonal cascades, in which catalysts activate catalysts, amplify the initial stimulus by several orders of magnitude, often in a very short time (seconds). ■ Nerve impulses stimulate the hypothalamus to send specific hormones to the pituitary gland, thus stimulating (or inhibiting) the release of tropic hormones. The anterior pituitary hormones in turn stimulate other endocrine glands (thyroid, adrenals, pancreas) to secrete their characteristic hormones, which in turn stimulate specific target tissues. ■ Peptide, amine, and eicosanoid hormones act outside the target cell on specific receptors in the plasma membrane, altering the level of an intracellular second messenger. ■ Steroid, vitamin D, retinoid, and thyroid hormones enter target cells and alter gene expression by interacting with specific nuclear receptors. 23.2 Tissue-Specific Metabolism: The Division of Labor Each tissue of the human body has a specialized func- tion, reflected in its anatomy and metabolic activity (Fig. 23–12). Skeletal muscle allows directed motion; adipose tissue stores and releases energy in the form of fats, which serve as fuel throughout the body; the brain pumps ions across plasma membranes to produce elec- trical signals. The liver plays a central processing and distributing role in metabolism and furnishes all other organs and tissues with an appropriate mix of nutrients via the bloodstream. The functional centrality of the liver is indicated by the common reference to all other tissues and organs as “extrahepatic” or “peripheral.” We therefore begin our discussion of the division of meta- bolic labor by considering the transformations of carbo- hydrates, amino acids, and fats in the mammalian liver. This is followed by brief descriptions of the primary metabolic functions of adipose tissue, muscle, brain, and the medium that interconnects all others: the blood. Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism892 Cortisol (mg) Adrenocorticotropic hormone (ACTH) (H9262g) Central nervous system Hypothalamus Anterior pituitary Adrenal gland Corticotropin-releasing hormone (CRH) (ng) Infection HemorrhageFear Pain Hypoglycemia Liver AdiposeMuscle FIGURE 23–11 Cascade of hormone release following central nerv- ous system input to the hypothalamus. In each endocrine tissue along the pathway, a stimulus from the level above is received, amplified, and transduced into the release of the next hormone in the cascade. The cascade is sensitive to regulation at several levels through feed- back inhibition by the ultimate hormone. The product therefore regu- lates its own production, as in feedback inhibition of biosynthetic path- ways within a single cell. SUMMARY 23.1 Hormones: Diverse Structures for Diverse Functions ■ Hormones are chemical messengers secreted by certain tissues into the blood or interstitial fluid, serving to regulate the activity of other cells or tissues. ■ Radioimmunoassay (RIA) and ELISA are two very sensitive techniques for detecting and quantifying hormones. 8885d_c23_881-919 3/1/04 1:26 PM Page 892 mac76 mac76:385_reb: The Liver Processes and Distributes Nutrients During digestion in mammals, the three main classes of nutrients (carbohydrates, proteins, and fats) undergo enzymatic hydrolysis into their simple constituents. This breakdown is necessary because the epithelial cells lin- ing the intestinal lumen absorb only relatively small mol- ecules. Many of the fatty acids and monoacylglycerols released by digestion of fats in the intestine are re- assembled within these epithelial cells into triacylglyc- erols (TAGs). After being absorbed, most sugars and amino acids and some TAGs travel in the bloodstream to the liver; the remaining TAGs enter adipose tissue via the lym- phatic system. The portal vein is a direct route from the digestive organs to the liver, and liver therefore has first access to ingested nutrients. The liver has two main cell types. Kupffer cells are phagocytes, important in im- mune function. Hepatocytes, of primary interest here, transform dietary nutrients into the fuels and precur- sors required by other tissues and export them via the blood. The kinds and amounts of nutrients supplied to the liver vary with several factors, including the diet and the time between meals. The demand of extrahepatic tissues for fuels and precursors varies among organs and with the level of activity and overall nutritional state of the individual. To meet these changing circumstances, the liver has remarkable metabolic flexibility. For example, when the diet is rich in protein, hepatocytes supply themselves with high levels of enzymes for amino acid catabolism and gluconeogenesis. Within hours after a shift to a high- carbohydrate diet, the levels of these enzymes begin to drop and the hepatocytes increase their synthesis of en- zymes essential to carbohydrate metabolism and fat syn- thesis. Liver enzymes turn over (are synthesized and degraded) at five to ten times the rate of enzyme turnover in other tissues, such as muscle. Extrahepatic 23.2 Tissue-Specific Metabolism: The Division of Labor 893 Secretes insulin and glucagon in response to changes in blood glucose concentration. Pancreas Brain Transports ions to maintain membrane potential; integrates inputs from body and surroundings; sends signals to other organs. Adipose tissue Synthesizes, stores, and mobilizes triacylglycerols. Uses ATP to do mechanical work. Carries lipids from intestine to liver. Skeletal muscle Lymphatic system Small intestine Absorbs nutrients from the diet, moves them into blood or lymphatic system. Liver Processes fats, carbohydrates, proteins from diet; synthesizes and distributes lipids, ketone bodies, and glucose for other tissues; converts excess nitrogen to urea. Carries nutrients from intestine to liver. Portal vein FIGURE 23–12 Specialized metabolic functions of mammalian tissues. 8885d_c23_881-919 3/1/04 1:26 PM Page 893 mac76 mac76:385_reb: tissues also can adjust their metabolism to prevailing conditions, but none is as adaptable as the liver, and none is so central to the organism’s overall metabolism. What follows is a survey of the possible fates of sugars, amino acids, and lipids that enter the liver from the bloodstream. To help you recall the metabolic transfor- mations discussed here, Table 23–2 shows the major pathways and processes to which we refer and indicates by figure number where each pathway is presented in detail. Here, we present summaries of the pathways, re- ferring to the step numbers in Figures 23–13 to 23–15. Sugars The glucose transporter in hepatocytes (GLUT2) is so effective that the concentration of glucose within a hepatocyte is essentially the same as that in the blood. Glucose entering hepatocytes is phosphorylated by hex- okinase IV (glucokinase) to yield glucose 6-phosphate. Glucokinase has a much higher K m for glucose (10 mM) than do the hexokinase isozymes in other cells (p. 578) and, unlike these other isozymes, it is not inhibited by its product, glucose 6-phosphate. The presence of glu- cokinase allows hepatocytes to continue phosphory- lating glucose when the glucose concentration rises well above levels that would overwhelm other hexoki- nases. The high K m of glucokinase also ensures that the phosphorylation of glucose in hepatocytes is min- imal when the glucose concentration is low, prevent- ing the liver from consuming glucose as fuel via gly- colysis. This spares glucose for other tissues. Fructose, galactose, and mannose, all absorbed from the small intestine, are also converted to glucose 6-phosphate by enzymatic pathways examined in Chapter 14. Glucose 6-phosphate is at the crossroads of carbohydrate me- tabolism in the liver. It may take any of several major metabolic routes (Fig. 23–13), depending on the cur- rent metabolic needs of the organism. By the action of various allosterically regulated enzymes, and through hormonal regulation of enzyme synthesis and activity, the liver directs the flow of glucose into one or more of these pathways. Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism894 TABLE 23–2 Pathways of Carbohydrate, Amino Acid, and Fat Metabolism Illustrated in Earlier Chapters Pathway Figure reference Citric acid cycle: acetyl-CoA ?→ 2CO 2 16–7 Oxidative phosphorylation: ATP synthesis 19–17 Carbohydrate catabolism Glycogenolysis: glycogen ?→ glucose 1-phosphate ?→ blood glucose 15–3; 15–4 Hexose entry into glycolysis: fructose, mannose, galactose ?→ glucose 6-phosphate 14–9 Glycolysis: glucose ?→ pyruvate 14–2 Pyruvate dehydrogenase reaction: pyruvate ?→ acetyl-CoA 16–2 Lactic acid fermentation: glucose ?→ lactate H11001 2ATP 14–3 Pentose phosphate pathway: glucose 6-phosphate ?→ pentose phosphates + NADPH 14–21 Carbohydrate anabolism Gluconeogenesis: citric acid cycle intermediates ?→ glucose 14–16 Glucose-alanine cycle: glucose ?→ pyruvate ?→ alanine ?→ glucose 18–9 Glycogen synthesis: glucose 6-phosphate ?→ glucose 1-phosphate ?→ glycogen 15–8 Amino acid and nucleotide metabolism Amino acid degradation: amino acids ?→ acetyl-CoA, citric acid cycle intermediates 18–15 Amino acid synthesis 22–9 Urea cycle: NH 3 ?→ urea 18–10 Glucose-alanine cycle: alanine ?→ glucose 18–9 Nucleotide synthesis: amino acids ?→ purines, pyrimidines 22–33; 22–36 Hormone and neurotransmitter synthesis 22–29 Fat catabolism H9252 Oxidation of fatty acids: fatty acid ?→ acetyl-CoA 17–8 Oxidation of ketone bodies: H9252-hydroxybutyrate ?→ acetyl-CoA ?→ CO 2 citric acid cycle 17–19 Fat anabolism Fatty acid synthesis: acetyl-CoA ?→ fatty acids 21–5 Triacylglycerol synthesis: acetyl-CoA ?→ fatty acids ?→ triacylglycerol 21–18; 21–19 Ketone body formation: acetyl-CoA ?→ acetoacetate, H9252-hydroxybutyrate 17–18 Cholesterol and cholesteryl ester synthesis: acetyl-CoA ?→ cholesterol ?→ cholesteryl esters 21–33 to 21–37 Phospholipid synthesis: fatty acids ?→ phospholipids 21–17; 21–23 to 21–28 8885d_c23_881-919 3/1/04 1:26 PM Page 894 mac76 mac76:385_reb: 1 Glucose 6-phosphate is dephosphorylated by glucose 6-phosphatase to yield free glucose (p. 547), which is exported to replenish blood glucose. Export is the predominant pathway when glucose 6-phosphate is in limited supply, because the blood glucose concentra- tion must be kept sufficiently high (4 mM) to provide adequate energy for the brain and other tissues. 2 Glu- cose 6-phosphate not immediately needed to form blood glucose is converted to liver glycogen, or it has one of several other fates. Following glucose 6-phosphate breakdown by glycolysis and decarboxylation of the pyruvate (by the pyruvate dehydrogenase reaction), 3 the acetyl-CoA so formed can be oxidized for energy production by the citric acid cycle, with ensuing elec- tron transfer and oxidative phosphorylation yielding ATP. (Normally, however, fatty acids are the preferred fuel for energy production in hepatocytes.) 4 Acetyl- CoA can also serve as the precursor of fatty acids, which are incorporated into TAGs and phospholipids, and cho- lesterol. Much of the lipid synthesized in the liver is 23.2 Tissue-Specific Metabolism: The Division of Labor 895 transported to other tissues by blood lipoproteins. 5 Finally, glucose 6-phosphate can enter the pentose phosphate pathway, yielding both reducing power (NADPH), needed for the biosynthesis of fatty acids and cholesterol, and D-ribose 5-phosphate, a precursor for nucleotide biosynthesis. NADPH is also an essential co- factor in the detoxification and elimination of many drugs and other xenobiotics metabolized in the liver. Amino Acids Amino acids that enter the liver follow sev- eral important metabolic routes (Fig. 23–14). 1 They are precursors for protein synthesis, a process discussed in Chapter 27. The liver constantly renews its own pro- teins, which have a relatively high turnover rate (aver- age half-life of only a few days), and is also the site of biosynthesis of most plasma proteins. 2 Alternatively, amino acids pass in the bloodstream to other organs, to be used in the synthesis of tissue proteins. 3 Other amino acids are precursors in the biosynthesis of nu- cleotides, hormones, and other nitrogenous compounds in the liver and other tissues. 4a Amino acids not needed as biosynthetic precur- sors are transaminated or deaminated and degraded to yield pyruvate and citric acid cycle intermediates, with various fates; 4b the ammonia released is converted to the excretory product urea. 5 Pyruvate can be con- verted to glucose and glycogen via gluconeogenesis or 6 it can be converted to acetyl-CoA, which has several possible fates. 7 It can be oxidized via the citric acid cycle and 8 oxidative phosphorylation to produce ATP, or 9 converted to lipids for storage. 10 Citric acid cy- cle intermediates can be siphoned off into glucose syn- thesis by gluconeogenesis. The liver also metabolizes amino acids that arrive in- termittently from other tissues. The blood is adequately supplied with glucose just after the digestion and ab- sorption of dietary carbohydrate or, between meals, by the conversion of liver glycogen to blood glucose. During the interval between meals, especially if prolonged, some muscle protein is degraded to amino acids. These amino acids donate their amino groups (by transamination) to pyruvate, the product of glycolysis, to yield alanine, which 11 is transported to the liver and deaminated. Hepato- cytes convert the resulting pyruvate to blood glucose (via gluconeogenesis 5 ), and the ammonia to urea for ex- cretion. One benefit of this glucose-alanine cycle (see Fig. 18–9) is the smoothing out of fluctuations in blood glu- cose between meals. The amino acid deficit incurred in the muscles is made up after the next meal by incoming dietary amino acids. Lipids The fatty acid components of the lipids entering hepatocytes also have several different fates (Fig. 23–15). 1 Some are converted to liver lipids. 2 Under most circumstances, fatty acids are the primary oxida- tive fuel in the liver. Free fatty acids may be activated FIGURE 23–13 Metabolic pathways for glucose 6-phosphate in the liver. Here and in Figures 23–14 and 23–15, anabolic pathways are shown leading upward, catabolic pathways leading downward, and distribution to other organs horizontally. The numbered processes in each figure are described in the text. Cholesterol Fatty acids citric acid cycle Acetyl-CoA e H11002 O 2 H 2 O ADP H11001 P i oxidative phosphorylation 3 ATP 3 4 glycolysis Pyruvate Blood glucose Ribose 5- phosphate NADPH Glucose 6- phosphate 1 Liver glycogen Hepatocyte 2 Nucleotides pentose phosphate pathway 5 Triacylglycerols, phospholipids CO 2 8885d_c23_881-919 3/1/04 1:26 PM Page 895 mac76 mac76:385_reb: and oxidized to yield acetyl-CoA and NADH. 3 and 4 The acetyl-CoA is further oxidized via the citric acid cy- cle, and the oxidations in the cycle drive the synthesis of ATP by oxidative phosphorylation. 5 Excess acetyl- CoA released by oxidation of fatty acids and not required by the liver is converted to the ketone bodies, acetoac- etate and H9252-hydroxybutyrate; these circulate in the blood to other tissues, to be used as fuel for the citric acid cy- cle. Ketone bodies may be regarded as a transport form of acetyl groups. They can supply a significant fraction of the energy in some extrahepatic tissues—up to one- third in the heart, and as much as 60% to 70% in the brain during prolonged fasting. 6 Some of the acetyl- CoA derived from fatty acids (and from glucose) is used for the biosynthesis of cholesterol, which is required for membrane synthesis. Cholesterol is also the precursor of all steroid hormones and of the bile salts, which are es- sential for the digestion and absorption of lipids. The final two metabolic fates of lipids involve spe- cialized mechanisms for the transport of insoluble lipids in the blood. 7 Fatty acids are converted to the phos- pholipids and TAGs of plasma lipoproteins, which carry lipids to adipose tissue for storage as TAGs. 8 Some free fatty acids become bound to serum albumin and are carried to the heart and skeletal muscles, which absorb and oxidize free fatty acids as a major fuel. Serum al- bumin is the most abundant plasma protein; one mole- cule of serum albumin can carry up to 10 molecules of free fatty acid to the tissues where the fatty acids are released and consumed. The liver thus serves as the body’s distribution cen- ter, exporting nutrients in the correct proportions to other organs, smoothing out fluctuations in metabolism caused by intermittent food intake, and processing ex- cess amino groups into urea and other products to be disposed of by the kidneys. Certain nutrients are stored in the liver, including Fe ions and vitamin A. The liver also detoxifies foreign organic compounds, such as drugs, food additives, preservatives, and other possibly harmful agents with no food value. Detoxification often involves the cytochrome P-450–dependent hydroxyla- tion of relatively insoluble organic compounds, making them sufficiently soluble for further breakdown and ex- cretion (see Box 21–1). Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism896 citric acid cycle Fatty acids Lipids Acetyl-CoA O 2 H 2 O ADP H11001 P i oxidative phosphorylation ATP 7 6 9 Pyruvate Amino acids in blood Amino acids 2 Liver proteins 1 11 CO 2 Plasma proteins Tissue proteins Glucose Glycogen 10 4a Glucose Glycogen in muscle 5 gluconeogenesis NH 3 Urea Alanine 4b urea cycle Nucleotides, hormones, porphyrins 8 3 e H11002 Hepatocyte Amino acids in muscle FIGURE 23–14 Metabolism of amino acids in the liver. 8885d_c23_881-919 3/1/04 1:26 PM Page 896 mac76 mac76:385_reb: Adipose Tissue Stores and Supplies Fatty Acids Adipose tissue, which consists of adipocytes (fat cells) (Fig. 23–16), is amorphous and widely distributed in the body: under the skin, around the deep blood vessels, and in the abdominal cavity. It typically makes up about 15% of the mass of a young adult human, with approx- imately 65% of this mass in the form of triacylglycerols. Adipocytes are metabolically very active, responding quickly to hormonal stimuli in a metabolic interplay with the liver, skeletal muscles, and the heart. Like other cell types, adipocytes have an active gly- colytic metabolism, use the citric acid cycle to oxidize pyruvate and fatty acids, and carry out oxidative phos- phorylation. During periods of high carbohydrate intake, adipose tissue can convert glucose (via pyruvate and acetyl-CoA) to fatty acids, convert the fatty acids to TAGs, and store them as large fat globules—although, in humans, much of the fatty acid synthesis occurs in hepatocytes. Adipocytes store TAGs arriving from the liver (carried in the blood as VLDLs; see Fig. 21–40a) and from the intestinal tract (carried in chylomicrons), particularly after meals rich in fat. When fuel demand rises, lipases in adipocytes hy- drolyze stored TAGs to release free fatty acids, which can travel in the bloodstream to skeletal muscles and the heart. The release of fatty acids from adipocytes is greatly accelerated by epinephrine, which stimulates the cAMP-dependent phosphorylation of perilipin; this gives triacylglycerol lipase access to TAGs in the lipid droplet. The hormone-sensitive lipase is also stimulated by phosphorylation, but this is not the main cause of in- creased lipolysis (see Fig. 17–3). Insulin counterbal- ances this effect of epinephrine, decreasing the activity of triacylglycerol lipase. 23.2 Tissue-Specific Metabolism: The Division of Labor 897 Liver lipids Fatty acids Plasma lipoproteins Free fatty acids in blood Ketone bodies in blood Bile salts Steroid hormones Cholesterol b oxidation citric acid cycle Acetyl-CoA O 2 H 2 O ADP H11001 P i oxidative phosphorylation ATP 3 6 8 5 7 1 2 CO 2 4 e H11002 Hepatocyte NADH FIGURE 23–15 Metabolism of fatty acids in the liver. 10 H9262m FIGURE 23–16 Scanning electron micrograph of human adipocytes. In fat tissues, capillaries and collagen fibers form a supporting net- work around spherical adipocytes. Almost the entire volume of these metabolically active cells is taken up by fat droplets. 8885d_c23_881-919 3/1/04 1:26 PM Page 897 mac76 mac76:385_reb: The breakdown and synthesis of TAGs in adipose tissues constitute a substrate cycle; up to 70% of the fatty acids released by triacylglycerol lipase are reester- ified in adipocytes, re-forming TAGs. Recall from Chap- ter 15 that such substrate cycles allow fine regulation of the rate and direction of flow of intermediates through a bidirectional pathway. In adipose tissue, glyc- erol liberated by triacylglycerol lipase cannot be reused in the synthesis of TAGs, because adipocytes lack the enzyme glycerol kinase. Instead, the glycerol phosphate required for TAG synthesis is made from pyruvate by glyceroneogenesis, involving the cytosolic enzyme PEP carboxykinase (see Fig. 21–22). This enzyme is one tar- get of the drugs (thiazolidinediones) used in the treat- ment of type II diabetes, raising the possibility that de- fective regulation of cytosolic PEP carboxykinase in fat tissue may be a causative factor in type II diabetes. Human infants, and many hibernating animals, have adipose tissue called brown fat, which is specialized to generate heat rather than ATP during oxidation of fatty acids. Adult humans have very little brown fat tissue. Muscles Use ATP for Mechanical Work Metabolism in the cells of skeletal muscle—the my- ocytes—is specialized to generate ATP as the immedi- ate source of energy for contraction. Moreover, skeletal muscle is adapted to do its mechanical work in an intermittent fashion, on demand. Sometimes skeletal muscles must work at their maximum capacity for a short time, as in a 100 m sprint; at other times more prolonged work is required, as in running a marathon or extended physical labor. There are two general classes of muscle tissue, which differ in physiological role and fuel utilization. Slow-twitch muscle, also called red muscle, provides relatively low tension but is highly resistant to fatigue. It produces ATP by the relatively slow but steady process of oxidative phosphorylation. Red muscle is very rich in mitochondria and is served by very dense net- works of blood vessels, which bring the oxygen essen- tial to ATP production. It is the cytochromes in mito- chondria and the hemoglobin in blood that give the tissue its characteristic red color. Fast-twitch muscle, or white muscle, has fewer mitochondria than red mus- cle and is less well supplied with blood vessels, but it can develop greater tension, and do so faster. White muscle is quicker to fatigue, because when active, it uses ATP faster than it can replace it. There is a genetic com- ponent to the proportion of red and white muscle in any individual; with training, the endurance of fast-twitch muscle can be improved. Skeletal muscle can use free fatty acids, ketone bod- ies, or glucose as fuel, depending on the degree of mus- cular activity (Fig. 23–17). In resting muscle, the pri- mary fuels are free fatty acids from adipose tissue and ketone bodies from the liver. These are oxidized and de- graded to yield acetyl-CoA, which enters the citric acid cycle for oxidation to CO 2 . The ensuing transfer of elec- trons to O 2 provides the energy for ATP synthesis by oxidative phosphorylation. Moderately active muscle uses blood glucose in addition to fatty acids and ketone bodies. The glucose is phosphorylated, then degraded by glycolysis to pyruvate, which is converted to acetyl- CoA and oxidized via the citric acid cycle and oxidative phosphorylation. In maximally active fast-twitch muscles, the de- mand for ATP is so great that the blood flow cannot pro- vide O 2 and fuels fast enough to supply sufficient ATP by aerobic respiration alone. Under these conditions, stored muscle glycogen is broken down to lactate by fer- mentation (p. 523). Each glucose unit degraded yields three ATP, because phosphorolysis of glycogen produces glucose 6-phosphate, sparing the ATP normally con- sumed in the hexokinase reaction. Lactic acid fermen- tation thus responds to an increased need for ATP more quickly than does oxidative phosphorylation, supple- menting basal ATP production that results from aerobic oxidation of other fuels via the citric acid cycle and res- piratory chain. The use of blood glucose and muscle glycogen as fuels for muscular activity is greatly en- hanced by the secretion of epinephrine, which stimu- lates both the release of glucose from liver glycogen and the breakdown of glycogen in muscle tissue. The relatively small amount of glycogen in skeletal muscle (about 1% of its total weight) limits the amount of glycolytic energy available during all-out exertion. Moreover, the accumulation of lactate and consequent decrease in pH in maximally active muscles reduces their efficiency. Skeletal muscle, however, contains another Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism898 ATP Creatine ADP H11001 P i CO 2 Lactate Muscle contraction Bursts of heavy activity Muscle glycogen Bursts of heavy activity Fatty acids, ketone bodies, blood glucose Light activity or rest Phosphocreatine FIGURE 23–17 Energy sources for muscle contraction. Different fu- els are used for ATP synthesis during bursts of heavy activity and dur- ing light activity or rest. Phosphocreatine can rapidly supply ATP. 8885d_c23_881-919 3/1/04 1:26 PM Page 898 mac76 mac76:385_reb: source of ATP, in the form of phosphocreatine (10 to 30 mM), which can rapidly regenerate ATP from ADP by the creatine kinase reaction: During periods of active contraction and glycolysis, this reaction proceeds predominantly in the direction of ATP synthesis; during recovery from exertion, the same enzyme resynthesizes phosphocreatine from creatine at the expense of ATP. After a period of intense muscular activity, the in- dividual continues breathing heavily for some time, us- ing much of the extra O 2 for oxidative phosphorylation in the liver. The ATP produced is used for gluconeoge- nesis from lactate that has been carried in the blood from the muscles. The glucose thus formed returns to the muscles to replenish their glycogen, completing the Cori cycle (Fig. 23–18; see also Box 15–1). Heart muscle differs from skeletal muscle in that it is continuously active in a regular rhythm of contraction and relaxation, and it has a completely aer- obic metabolism at all times. Mitochondria are much more abundant in heart muscle than in skeletal muscle, making up almost half the volume of the cells (Fig. 23–19). The heart uses as its fuel mainly free fatty acids, but also some glucose and ketone bodies taken up from the blood; these fuels are oxidized via the citric acid cy- cle and oxidative phosphorylation to generate ATP. Like skeletal muscle, heart muscle does not store lipids or glycogen in large amounts. It does have small amounts of reserve energy in the form of phosphocreatine, enough for a few seconds of contraction. Because the heart is normally aerobic and obtains its energy from oxidative phosphorylation, the failure of O 2 to reach a portion of the heart muscle when the blood vessels are blocked by lipid deposits (atherosclerosis) or blood clots (coronary thrombosis) can cause that region of the heart muscle to die. This is what happens in myocardial infarction, more commonly known as a heart attack. ■ 23.2 Tissue-Specific Metabolism: The Division of Labor 899 NH 2 N Creatine COO H11002 C H11001 NH 2 CH 3 CH 2 ATP H11001 Phosphocreatine CH 2 C H11002 O H P O N H11001 NH 2 CH 3 N O H11002 COO H11002 H11001 ADP during activity during recovery Muscle: ATP produced by glycolysis for rapid contraction. Lactate Glycogen Lactate Glucose ATP ATP Blood lactate Blood glucose Liver: ATP used in synthesis of glucose (gluconeogenesis) during recovery. FIGURE 23–18 Metabolic cooperation between skeletal muscle and the liver. Extremely active muscles use glycogen as energy source, generating lactate via glycolysis. During recovery, some of this lactate is transported to the liver and converted to glucose via gluconeogen- esis. This glucose is released to the blood and returned to the muscles to replenish their glycogen stores. The overall pathway (glucose → lactate → glucose) constitutes the Cori cycle. FIGURE 23–19 Electron micrograph of heart muscle. In the profuse mitochondria of heart tissue, pyruvate, fatty acids, and ketone bodies are oxidized to drive ATP synthesis. This steady aerobic metabolism allows the human heart to pump blood at a rate of nearly 6 L/min, or about 350 L/hr—or 200 H11003 10 6 L over 70 years. 1 mH9262 8885d_c23_881-919 3/1/04 1:26 PM Page 899 mac76 mac76:385_reb: The Brain Uses Energy for Transmission of Electrical Impulses The metabolism of the brain is remarkable in several re- spects. The neurons of the adult mammalian brain nor- mally use only glucose as fuel (Fig. 23–20). (Astrocytes, the other major cell type in the brain, can oxidize fatty acids.) The brain has a very active respiratory metabo- lism (Fig. 23–21); it uses O 2 at a fairly constant rate, ac- counting for almost 20% of the total O 2 consumed by the body at rest. Because the brain contains very little glycogen, it is constantly dependent on incoming glu- cose from the blood. Should blood glucose fall signifi- cantly below a critical level for even a short time, severe and sometimes irreversible changes in brain function may result. Although the neurons of the brain cannot directly use free fatty acids or lipids from the blood as fuels, they can, when necessary, use H9252-hydroxybutyrate (a ketone body), which is formed from fatty acids in the liver. The capacity of the brain to oxidize H9252-hydroxybutyrate via acetyl-CoA becomes important during prolonged fasting or starvation, after liver glycogen has been depleted, be- cause it allows the brain to use body fat as an energy source. This spares muscle proteins—until they become the brain’s ultimate source of glucose (via gluconeogen- esis in the liver) during severe starvation. Neurons oxidize glucose by glycolysis and the cit- ric acid cycle, and the flow of electrons from these ox- idations through the respiratory chain provides almost all the ATP used by these cells. Energy is required to create and maintain an electrical potential across the neuronal plasma membrane. The membrane contains an electrogenic ATP-driven antiporter, the Na H11001 K H11001 ATPase, which simultaneously pumps 2 K H11001 ions into and 3 Na H11001 ions out of the neuron (see Fig. 11–37). The resulting transmembrane potential changes transiently as an elec- trical signal (action potential) sweeps from one end of a neuron to the other (see Fig. 12–5). Action potentials are the chief mechanism of information transfer in the nervous system, so a depletion of ATP in neurons has disastrous effects on all activities coordinated by neu- ronal signaling. Blood Carries Oxygen, Metabolites, and Hormones Blood mediates the metabolic interactions among all tis- sues. It transports nutrients from the small intestine to the liver, and from the liver and adipose tissue to other organs; it also transports waste products from the tis- sues to the kidneys for excretion. Oxygen moves in the bloodstream from the lungs to the tissues, and CO 2 gen- erated by tissue respiration returns via the bloodstream to the lungs for exhalation. Blood also carries hormonal signals from one tissue to another. In its role as signal carrier, the circulatory system resembles the nervous system; both regulate and integrate the activities of dif- ferent organs. Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism900 CO 2 ADP + P i Ketone bodies Electrogenic transport by Na + K + ATPase Starvation ATP Normal diet Glucose FIGURE 23–20 Energy sources in the brain vary with nutritional state. The ketone body used by the brain is H9252-hydroxybutyrate. (a) (b) 12.00 2.00 mg/100 g /min FIGURE 23–21 Glucose metabolism in the brain. The technique of positron emission tomography (PET) scanning shows metabolic activ- ity in specific regions of the brain. PET scans allow visualization of isotopically labeled glucose in precisely localized regions of the brain of a living person, in real time. A positron-emitting glucose analog (2- [ 18 F]-fluoro-2-deoxy-D-glucose) is injected into the bloodstream; a few seconds later, a PET scan shows how much of the glucose has been taken up by each region of the brain—a measure of metabolic activ- ity. Shown here are PET scans of front-to-back cross sections of the brain at three levels, from the top (at the left) downward (to the right). The scans compare glucose metabolism (in mg/100 g/min) when the experimental subject (a) is rested and (b) has been deprived of sleep for 48 hours. 8885d_c23_881-919 3/1/04 1:26 PM Page 900 mac76 mac76:385_reb: The average adult human has 5 to 6 L of blood. Al- most half of this volume is occupied by three types of blood cells (Fig. 23–22): erythrocytes (red cells), filled with hemoglobin and specialized for carrying O 2 and CO 2 ; much smaller numbers of leukocytes (white cells) of several types (including lymphocytes, also found in lymphatic tissue), which are central to the immune sys- tem that defends against infections; and platelets, which help to mediate blood clotting. The liquid portion is the blood plasma, which is 90% water and 10% solutes. Dissolved or suspended in the plasma is a large variety of proteins, lipoproteins, nutrients, metabolites, waste products, inorganic ions, and hormones. More than 70% of the plasma solids are plasma proteins (Fig. 23–22), primarily immunoglobulins (circulating antibodies), serum albumin, apolipoproteins involved in the transport of lipids, transferrin (for iron transport), and blood-clotting proteins such as fibrinogen and prothrombin. The ions and low molecular weight solutes in blood plasma are not fixed components but are in constant flux between blood and various tissues. Dietary uptake of the inorganic ions that are the predominant elec- trolytes of blood and cytosol (Na H11001 , K H11001 , and Ca 2H11001 ) is, in general, counterbalanced by their excretion in the urine. For many blood components, something near a dynamic steady state is achieved; the concentration of the com- ponent changes little, although a continuous flux occurs between the digestive tract, blood, and urine. The plasma levels of Na H11001 , K H11001 , and Ca 2H11001 remain close to 140, 5, and 2.5 mM, respectively, with little change in re- sponse to dietary intake. Any significant departure from these values can result in serious illness or death. The kidneys play an especially important role in maintaining ion balance by selectively filtering waste products and excess ions out of the blood while preventing the loss of essential nutrients and ions. The concentration of glucose in the plasma is also subject to tight regulation. We have noted the constant requirement of the brain for glucose and the role of the liver in maintaining blood glucose in the nor- mal range of 60 to 90 mg/100 mL. When blood glucose in a human drops to 40 mg/100 mL (the hypoglycemic condition), the person experiences discomfort and men- tal confusion (Fig. 23–23); further reductions lead to coma, convulsions, and in extreme hypoglycemia, death. 23.2 Tissue-Specific Metabolism: The Division of Labor 901 0 10 20 30 40 50 60 70 80 90 100 Blood glucose (mg/100 mL) Normal range Subtle neurological signs; hunger Release of glucagon, epinephrine, cortisol Sweating, trembling Lethargy Convulsions, coma Permanent brain damage (if prolonged) Death Blood plasmaCells Inorganic components (10%) NaCl, bicarbonate, phosphate, CaCl 2 , MgCl 2 , KCl, Na 2 SO 4 Organic metabolites and waste products (20%) glucose, amino acids, lactate, pyruvate, ketone bodies, citrate, urea, uric acid Plasma proteins (70%) Major plasma proteins: serum albumin, very-low-density lipoproteins (VLDL), low-density lipoproteins (LDL), high-density lipoproteins (HDL), immunoglobulins (hundreds of kinds), fibrinogen, prothrombin, many specialized transport proteins such as transferrin H 2 OErythrocytes Leukocytes Platelets Plasma solutes FIGURE 23–22 The composition of blood. Whole blood can be sep- arated into blood plasma and cells by centrifugation. About 10% of blood plasma is solutes, of which about 10% consists of inorganic salts, 20% small organic molecules, and 70% plasma proteins. The major dissolved components are listed. Blood contains many other substances, often in trace amounts. These include other metabolites, enzymes, hormones, vitamins, trace elements, and bile pigments. Mea- surements of the concentrations of components in blood plasma are important in the diagnosis and treatment of many diseases. FIGURE 23–23 Physiological effects of low blood glucose in humans. Blood glucose levels of 40 mg/100 mL and below constitute severe hypoglycemia. 8885d_c23_881-919 3/1/04 1:26 PM Page 901 mac76 mac76:385_reb: Maintaining the normal concentration of glucose in the blood is therefore a very high priority of the organism, and a variety of regulatory mechanisms have evolved to achieve that end. Among the most important regu- lators of blood glucose are the hormones insulin, glucagon, and epinephrine, as discussed in Section 23.3. ■ SUMMARY 23.2 Tissue-Specific Metabolism: The Division of Labor ■ In mammals there is a division of metabolic labor among specialized tissues and organs. The liver is the central distributing and processing organ for nutrients. Sugars and amino acids produced in digestion cross the intestinal epithelium and enter the blood, which carries them to the liver. Some triacylglycerols derived from ingested lipids also make their way to the liver, where the constituent fatty acids are used in a variety of processes. ■ Glucose 6-phosphate is the key intermediate in carbohydrate metabolism. It may be polymerized into glycogen, dephosphorylated to blood glucose, or converted to fatty acids via acetyl- CoA. It may undergo oxidation by glycolysis, the citric acid cycle, and respiratory chain to yield ATP, or enter the pentose phosphate pathway to yield pentoses and NADPH. ■ Amino acids are used to synthesize liver and plasma proteins, or their carbon skeletons are converted to glucose and glycogen by gluconeogenesis; the ammonia formed by deamination is converted to urea. ■ The liver converts fatty acids to triacylglycerols, phospholipids, or cholesterol and its esters, for transport as plasma lipoproteins to adipose tissue for storage. Fatty acids can also be oxidized to yield ATP or to form ketone bodies, which are circulated to other tissues. ■ Skeletal muscle is specialized to produce and use ATP for mechanical work. During strenuous muscular activity, glycogen is the ultimate fuel, supplying ATP through lactic acid fermentation. During recovery, the lactate is reconverted (through gluconeogenesis) to glycogen and glucose in the liver. Phosphocreatine is an immediate source of ATP during active contraction. ■ Heart muscle obtains nearly all its ATP from oxidative phosphorylation. ■ The neurons of the brain use only glucose and H9252-hydroxybutyrate as fuels, the latter being important during fasting or starvation. The brain uses most of its ATP for the active transport of Na H11001 and K H11001 and maintenance of the electrical potential across the neuronal membrane. ■ The blood carries nutrients, waste products, and hormonal signals among the organs. 23.3 Hormonal Regulation of Fuel Metabolism The minute-by-minute adjustments that keep the blood glucose level near 4.5 mM involve the combined actions of insulin, glucagon, epinephrine, and cortisol on meta- bolic processes in many body tissues, but especially in liver, muscle, and adipose tissue. Insulin signals these tissues that blood glucose is higher than necessary; as a result, cells take up excess glucose from the blood and convert it to the storage compounds glycogen and tria- cylglycerol. Glucagon signals that blood glucose is too low, and tissues respond by producing glucose through glycogen breakdown and (in liver) gluconeogenesis and by oxidizing fats to reduce the use of glucose. Epi- nephrine is released into the blood to prepare the mus- cles, lungs, and heart for a burst of activity. Cortisol me- diates the body’s response to longer-term stresses. We discuss these hormonal regulations in the context of three normal metabolic states—well-fed, fasted, and starving—and look at the metabolic consequences of di- abetes mellitus, which results from derangements in the signaling pathways that control glucose metabolism. The Pancreas Secretes Insulin or Glucagon in Response to Changes in Blood Glucose When glucose enters the bloodstream from the intes- tine after a carbohydrate-rich meal, the resulting in- crease in blood glucose causes increased secretion of insulin (and decreased secretion of glucagon). Insulin release by the pancreas is largely regulated by the level of glucose in the blood supplied to the pancreas. The peptide hormones insulin, glucagon, and somatostatin are produced by clusters of specialized pancreatic cells, the islets of Langerhans (Fig. 23–24). Each cell type of the islets produces a single hormone: H9251 cells produce glucagon; H9252 cells, insulin; and H9254 cells, somatostatin. When blood glucose rises, GLUT2 transporters carry glucose into the H9252 cells, where it is immediately con- verted to glucose 6-phosphate by hexokinase IV (gluco- kinase) and enters glycolysis (Fig. 23–25). The in- creased rate of glucose catabolism raises [ATP], causing the closing of ATP-gated K H11001 channels in the plasma membrane. Reduced efflux of K H11001 depolarizes the mem- brane, thereby opening voltage-sensitive Ca 2H11001 channels in the plasma membrane. The resulting influx of Ca 2H11001 Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism902 8885d_c23_881-919 3/1/04 1:26 PM Page 902 mac76 mac76:385_reb: triggers the release of insulin by exocytosis. Stimuli from the parasympathetic and sympathetic nervous systems also stimulate and inhibit insulin release, respectively. A simple feedback loop limits hormone release: insulin lowers blood glucose by stimulating glucose uptake by the tissues; the reduced blood glucose is detected by the H9252 cell as a diminished flux through the hexokinase reaction; this slows or stops the release of insulin. This feedback regulation holds blood glucose concentration nearly constant despite large fluctuations in dietary intake. 23.3 Hormonal Regulation of Fuel Metabolism 903 H9251 cell (glucagon) H9252 cell (insulin) H9254 cell (somatostatin) Pancreas Blood vessels FIGURE 23–24 The endocrine system of the pancreas. In addition to the exocrine cells (see Fig. 18–3b), which secrete digestive enzymes in the form of zymogens, the pancreas contains endocrine tissue, the islets of Langerhans. The islets contain H9251, H9252, and H9254 cells (also known as A, B, and D cells, respectively), each cell type secreting a specific polypeptide hormone. Glucose Glucose 6-phosphate Glucose transporter GLUT2 Extracellular space Pancreatic b cell hexokinase IV (glucokinase) oxidative phosphorylation glycolysis citric acid cycle [Ca 2+ ] Ca 2+ [ATP] ATP-gated K + channel + + + + + + – – – – – – depolarization Voltage-dependent Ca 2+ channel Insulin granules Insulin secretion Nucleus K + K + V m Glucose 2 4 3 5 1 FIGURE 23–25 Glucose regulation of insulin secretion by pancreatic H9252 cells. When the blood glucose level is high, active metabolism of glucose in the H9252 cell raises intracellular [ATP], which leads to closing of K H11001 channels in the plasma membrane, depolarizing the membrane. In response to the change in membrane potential, voltage-gated Ca 2H11001 channels in the plasma membrane open, allowing Ca 2H11001 to flow into the cell; this raises the cytosolic [Ca 2H11001 ] enough to trigger insulin release by exocytosis. 8885d_c23_881-919 3/1/04 1:26 PM Page 903 mac76 mac76:385_reb: Insulin Counters High Blood Glucose Insulin stimulates glucose uptake by muscle and adipose tissue (Table 23–3), where the glucose is converted to glucose 6-phosphate. In the liver, insulin also activates glycogen synthase and inactivates glycogen phosphory- lase, so that much of the glucose 6-phosphate is chan- neled into glycogen. Insulin also stimulates the storage of excess fuel as fat (Fig. 23–26). In the liver, insulin activates both the oxidation of glucose 6-phosphate to pyruvate via gly- colysis and the oxidation of pyruvate to acetyl-CoA. If not oxidized further for energy production, this acetyl- CoA is used for fatty acid synthesis in the liver, and the fatty acids are exported as the TAGs of plasma lipopro- teins (VLDLs) to the adipose tissue. Insulin stimulates TAG synthesis in adipocytes, from fatty acids released from the VLDL triacylglycerols. These fatty acids are ul- timately derived from the excess glucose taken up from the blood by the liver. In summary, the effect of insulin is to favor the conversion of excess blood glucose to two storage forms: glycogen (in the liver and muscle) and triacylglycerols (in adipose tissue) (Table 23–3). Glucagon Counters Low Blood Glucose Several hours after the intake of dietary carbohydrate, blood glucose levels fall slightly because of the ongoing oxidation of glucose by the brain and other tissues. Low- ered blood glucose triggers secretion of glucagon and decreases insulin release (Fig. 23–27). Glucagon causes an increase in blood glucose con- centration in several ways (Table 23–4). Like epineph- rine, it stimulates the net breakdown of liver glycogen Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism904 FIGURE 23–26 The well-fed state: the lipogenic liver. Immediately after a calorie-rich meal, glucose, fatty acids, and amino acids enter the liver. Insulin released in response to the high blood glucose concentration stimulates glucose uptake by the tissues. Some glucose is exported to the brain for its energetic needs, and some to fat and muscle tissue. In the liver, excess glucose is oxidized to acetyl-CoA, which is used to synthesize fatty acids for export as triacylglycerols in Pancreas Intestine Glucose Pyruvate ATP Acetyl- CoA TAG Amino acids Urea a-Keto acids Protein synthesis Glycogen Lymphatic system VLDL TAG Adipose tissue Muscle Fatty acids ATP CO 2 ATP CO 2 CO 2 Liver Brain to brain, adipose, muscle Glucose Amino acids Fats Glucose Glucose Insulin Insulin TAG VLDLs to fat and muscle tissue. The NADPH necessary for lipid synthesis is obtained by oxidation of glucose in the pentose phosphate pathway. Excess amino acids are converted to pyruvate and acetyl- CoA, which are also used for lipid synthesis. Dietary fats move via the lymphatic system, as chylomicrons, from the intestine to muscle and fat tissues. 8885d_c23_881-919 3/1/04 1:26 PM Page 904 mac76 mac76:385_reb: by activating glycogen phosphorylase and inactivating glycogen synthase; both effects are the result of phos- phorylation of the regulated enzymes, triggered by cAMP. Glucagon inhibits glucose breakdown by glycol- ysis in the liver and stimulates glucose synthesis by gluconeogenesis. Both effects result from lowering the concentration of fructose 2,6-bisphosphate, an allosteric inhibitor of the gluconeogenic enzyme fructose 1,6-bis- phosphatase (FBPase-1) and an activator of phospho- fructokinase-1. Recall that [fructose 2,6-bisphosphate] is ultimately controlled by a cAMP-dependent protein phosphorylation reaction (see Fig. 15–23). Glucagon also inhibits the glycolytic enzyme pyruvate kinase (by promoting its cAMP-dependent phosphorylation), thus blocking the conversion of phosphoenolpyruvate to pyruvate and preventing oxidation of pyruvate via the citric acid cycle. The resulting accumulation of phos- phoenolpyruvate favors gluconeogenesis. This effect is augmented by glucagon’s stimulation of the synthesis of the gluconeogenic enzyme PEP carboxykinase. By stim- ulating glycogen breakdown, preventing glycolysis, and promoting gluconeogenesis in hepatocytes, glucagon 23.3 Hormonal Regulation of Fuel Metabolism 905 Metabolic effect Target enzyme ↑ Glucose uptake (muscle, adipose) ↑ Glucose transporter (GLUT4) ↑ Glucose uptake (liver) ↑ Glucokinase (increased expression) ↑ Glycogen synthesis (liver, muscle) ↑ Glycogen synthase ↓ Glycogen breakdown (liver, muscle) ↓ Glycogen phosphorylase ↑ Glycolysis, acetyl-CoA production (liver, muscle) ↑ PFK-1 (by ↑ PFK-2) ↑ Pyruvate dehydrogenase complex ↑ Fatty acid synthesis (liver) ↑ Acetyl-CoA carboxylase ↑ Triacylglycerol synthesis (adipose tissue) ↑ Lipoprotein lipase TABLE 23–3 Effects of Insulin on Blood Glucose: Uptake of Glucose by Cells and Storage as Triacylglycerols and Glycogen Pancreas Amino acids TAG Adipose tissue Muscle Liver Brain Glucagon ATP Ketone bodies CO 2 CO 2 Fatty acids Fatty acids ATP Protein Ketone bodies ATP Proteins CO 2 CO 2 Ketone bodies Glycogen Pyruvate gluconeogenesis Glucose Glucose Glycerol Glucose 6-phosphate FIGURE 23–27 The fasting state: the gluco- genic liver. After some hours without a meal, the liver becomes the principal source of glucose for the brain. Liver glycogen is broken down, and the glucose 1-phosphate produced is converted to glucose 6-phos- phate, then to free glucose, which is released into the bloodstream. Amino acids from the degradation of proteins and glycerol from the breakdown of TAGs in adipose tissue are used for gluconeogenesis. The liver uses fatty acids as its principal fuel, and excess acetyl-CoA is converted to ketone bodies for export to other tissues for fuel; the brain is especially dependent on this fuel when glucose is in short supply. 8885d_c23_881-919 3/1/04 1:26 PM Page 905 mac76 mac76:385_reb: enables the liver to export glucose, restoring blood glu- cose to its normal level. Although its primary target is the liver, glucagon (like epinephrine) also affects adipose tissue, activating TAG breakdown by causing cAMP-dependent phosphor- ylation of perilipin and triacylglycerol lipase. The acti- vated lipase liberates free fatty acids, which are ex- ported to the liver and other tissues as fuel, sparing glucose for the brain. The net effect of glucagon is there- fore to stimulate glucose synthesis and release by the liver and to mobilize fatty acids from adipose tissue, to be used instead of glucose as fuel for tissues other than the brain (Table 23–4). All these effects of glucagon are mediated by cAMP-dependent protein phosphorylation. During Fasting and Starvation, Metabolism Shifts to Provide Fuel for the Brain The fuel reserves of a healthy adult human are of three types: glycogen stored in the liver and, in relatively small quantities, in muscles; large quantities of triacylglycerols in adipose tissues; and tissue proteins, which can be de- graded when necessary to provide fuel (Table 23–5). In the first few hours after a meal, the blood glucose level is diminished slightly, and tissues receive glucose released from liver glycogen. There is little or no syn- thesis of lipids. By 24 hours after a meal, blood glucose has fallen further, insulin secretion has slowed, and glu- cagon secretion has increased. These hormonal signals Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism906 Caloric equivalent Estimated survival Type of fuel Weight (kg) (thousands of kcal (kJ)) (months) * Normal-weight, 70 kg man Triacylglycerols (adipose tissue) 15 141 (589) Proteins (mainly muscle) 6 24 (100) Glycogen (muscle, liver) 0.225 0.90 (3.8) Circulating fuels (glucose, fatty acids, 0.023 0.10 (0.42) triacylglycerols, etc.) Total 166 (694) 3 Obese, 140 kg man Triacylglycerols (adipose tissue) 80 752 (3,140) Proteins (mainly muscle) 8 32 (134) Glycogen (muscle, liver) 0.23 0.92 (3.8) Circulating fuels 0.025 0.11 (0.46) Total 785 (3,280) 14 TABLE 23–5 Available Metabolic Fuels in a Normal-Weight 70 kg Man and in an Obese 140 kg Man at the Beginning of a Fast TABLE 23–4 Effects of Glucagon on Blood Glucose: Production and Release of Glucose by the Liver Metabolic effect Effect on glucose metabolism Target enzyme ↑ Glycogen breakdown (liver) Glycogen ?→ glucose ↑ Glycogen phosphorylase ↓ Glycogen synthesis (liver) Less glucose stored as glycogen ↓ Glycogen synthase ↓ Glycolysis (liver) Less glucose used as fuel in liver ↓ PFK-1 ↑ Gluconeogenesis (liver) Amino acids ↑ FBPase-2 Glycerol ?→ glucose ↓ Pyruvate kinase Oxaloacetate ↑ PEP carboxykinase ↑ Fatty acid mobilization (adipose tissue) Less glucose used as fuel by liver, muscle ↑ Triacylglycerol lipase Perilipin phosphorylation ↑ Ketogenesis Provides alternative to glucose as ↑ Acetyl-CoA carboxylase energy source for brain * Survival time is calculated on the assumption of a basal energy expenditure of 1,800 kcal/day. ? ? ? ? ? 8885d_c23_881-919 3/1/04 1:26 PM Page 906 mac76 mac76:385_reb: mobilize triacylglycerols, which now become the primary fuel for muscle and liver. Figure 23–28 shows the re- sponses to prolonged fasting. 1 To provide glucose for the brain, the liver degrades certain proteins—those most expendable in an organism not ingesting food. Their nonessential amino acids are transaminated or deaminated (Chapter 18), and 2 the extra amino groups are converted to urea, which is exported via the bloodstream to the kidney and excreted. Also in the liver, 3 the carbon skeletons of gluco- genic amino acids are converted to pyruvate or inter- mediates of the citric acid cycle. 4 These intermedi- ates, as well as the glycerol 5 derived from triacyl- glycerols in adipose tissue, provide the starting materi- als for gluconeogenesis in the liver, yielding glucose for the brain. Eventually the use of citric acid cycle inter- mediates for gluconeogenesis depletes oxaloacetate, in- hibiting entry of acetyl-CoA into the citric acid cycle. 6 Acetyl-CoA produced by fatty acid oxidation now accu- mulates, favoring 7 the formation of acetoacetyl-CoA and ketone bodies in the liver. After a few days of fast- ing, the levels of ketone bodies in the blood rise (Fig. 23.3 Hormonal Regulation of Fuel Metabolism 907 Acetyl-CoA Protein degradation yields glucogenic amino acids. 1 Citric acid cycle intermediates are diverted to gluconeogenesis. 3 Acetyl-CoA accumulation favors ketone body synthesis. 7 Lack of oxaloacetate prevents acetyl-CoA entry into the citric acid cycle; acetyl-CoA accumulates. 6 Hepatocyte (2H11003) P i Oxaloacetate Protein Amino acids NH 3 Glucose 6-phosphate Phosphoenol- pyruvate Citrate Ketone bodies are exported via the bloodstream to the brain, which uses them as fuel. 8 Ketone bodies Acetoacetyl-CoA Fatty acids (imported from adipose tissue) are oxidized as fuel, producing acetyl-CoA. 5 Glucose is exported to the brain via the bloodstream. 4 Glucose Urea is exported to the kidney and excreted in urine. 2 Urea Fatty acids FIGURE 23–28 Fuel metabolism in the liver during prolonged fast- ing or in uncontrolled diabetes mellitus. After depletion of stored car- bohydrates, 1 to 4 proteins become an important source of glu- cose, produced from glucogenic amino acids by gluconeogenesis. 5 to 8 Fatty acids imported from adipose tissue are converted to ke- tone bodies for export to the brain. Broken arrows represent reactions with reduced flux under these conditions. The steps are further de- scribed in the text. 8885d_c23_881-919 3/1/04 1:26 PM Page 907 mac76 mac76:385_reb: 23–29) as these fuels are exported from the liver to the heart, skeletal muscle, and brain, which use them in- stead of glucose ( 8 ). Acetyl-CoA is a critical regulator of the fate of pyru- vate; it allosterically inhibits pyruvate dehydrogenase and stimulates pyruvate carboxylase (see Fig. 15–20). In these ways acetyl-CoA prevents it own further pro- duction from pyruvate while stimulating the conversion of pyruvate to oxaloacetate, the first step in gluconeo- genesis. Triacylglycerols stored in the adipose tissue of a normal-weight adult could provide enough fuel to main- tain a basal rate of metabolism for about three months; a very obese adult has enough stored fuel to endure a fast of more than a year (Table 23–5). When fat reserves are gone, the degradation of essential proteins begins; this leads to loss of heart and liver function, and even- tually death. Stored fat can provide adequate energy (calories) during a fast or rigid diet, but vitamins and minerals must be provided, and sufficient dietary gluco- genic amino acids are needed to replace those being used for gluconeogenesis. Rations for those on a weight- reduction diet are therefore commonly fortified with vi- tamins, minerals, and amino acids or proteins. Epinephrine Signals Impending Activity When an animal is confronted with a stressful situation that requires increased activity—fighting or fleeing, in the extreme case—neuronal signals from the brain trig- ger the release of epinephrine and norepinephrine from the adrenal medulla. Both hormones dilate the respira- tory passages to facilitate the uptake of O 2 , increase the rate and strength of the heartbeat, and raise the blood pressure, thereby promoting the flow of O 2 and fuels to the tissues (Table 23–6). Epinephrine acts primarily on muscle, adipose, and liver tissues. It activates glycogen phosphorylase and in- activates glycogen synthase by cAMP-dependent phos- phorylation of the enzymes, thus stimulating the con- version of liver glycogen to blood glucose, the fuel for anaerobic muscular work. Epinephrine also promotes the anaerobic breakdown of muscle glycogen by lactic acid fermentation, stimulating glycolytic ATP formation. The stimulation of glycolysis is accomplished by raising the concentration of fructose 2,6-bisphosphate, a potent allosteric activator of the key glycolytic enzyme phos- phofructokinase-1 (see Figs 15–22, 15–23). Epinephrine Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism908 Glucose Fatty acids Ketone bodies 0 8 7 6 5 4 3 2 1 Plasma concentration (m M ) 24 Days of starvation 68 FIGURE 23–29 Concentrations of fatty acids, glucose, and ketone bodies in the plasma during the first week of starvation. Despite the hormonal mechanisms for maintaining the level of glucose in the blood, it begins to diminish after two days of fasting. The level of ke- tone bodies, almost immeasurable before the fast, rises dramatically after 2 to 4 days of fasting. These water-soluble ketones, acetoacetate and H9252-hydroxybutyrate, supplement glucose as an energy source dur- ing a long fast. Fatty acids cannot serve as a fuel for the brain; they do not cross the blood-brain barrier. TABLE 23–6 Physiological and Metabolic Effects of Epinephrine: Preparation for Action Immediate effect Overall effect Physiological ↑ Heart rate ↑ Blood pressure Increase delivery of O 2 to tissues (muscle) ↑ Dilation of respiratory passages Metabolic ↑ Glycogen breakdown (muscle, liver) ↓ Glycogen synthesis (muscle, liver) Increase production of glucose for fuel ↑ Gluconeogenesis (liver) ↑ Glycolysis (muscle) Increases ATP production in muscle ↑ Fatty acid mobilization (adipose tissue) Increases availability of fatty acids as fuel ↑ Glucagon secretion ↓ Insulin secretion Reinforce metabolic effects of epinephrine ? ? ? ? ? ? ? ? ? ? ? ? ? 8885d_c23_881-919 3/1/04 1:26 PM Page 908 mac76 mac76:385_reb: CH 3 C O CH 2 COO H11002 H11001 H 2 O CH 3 C O CH 3 H11001HCO 3 H11002 Acetoacetate Acetone also stimulates fat mobilization in adipose tissue, acti- vating (by cAMP-dependent phosphorylation) both peri- lipin and triacylglycerol lipase (see Fig. 17–3). Finally, epinephrine stimulates glucagon secretion and inhibits insulin secretion, reinforcing its effect of mobilizing fuels and inhibiting fuel storage. Cortisol Signals Stress, Including Low Blood Glucose A variety of stressors (anxiety, fear, pain, hemorrhage, infections, low blood glucose, starvation) stimulate re- lease of the corticosteroid hormone cortisol from the adrenal cortex. Cortisol acts on muscle, liver, and adi- pose tissue to supply the organism with fuel to with- stand the stress. Cortisol is a relatively slow-acting hor- mone that alters metabolism by changing the kinds and amounts of certain enzymes synthesized in its target cell, rather than by regulating the activity of existing en- zyme molecules. In adipose tissue, cortisol leads to an increase in the release of fatty acids from stored TAGs. The fatty acids are exported to serve as fuel for other tissues, and the glycerol is used for gluconeogenesis in the liver. Corti- sol stimulates the breakdown of muscle proteins and the export of amino acids to the liver, where they serve as precursors for gluconeogenesis. In the liver, cortisol promotes gluconeogenesis by stimulating synthesis of the key enzyme PEP carboxykinase (see Fig. 14–17b); glucagon has the same effect, whereas insulin has the opposite effect. Glucose produced in this way is stored in the liver as glycogen or exported immediately to tis- sues that need glucose for fuel. The net effect of these metabolic changes is to restore blood glucose to its nor- mal level and to increase glycogen stores, ready to sup- port the fight-or-flight response commonly associated with stress. The effects of cortisol therefore counter- balance those of insulin. Diabetes Mellitus Arises from Defects in Insulin Production or Action Diabetes mellitus, caused by a deficiency in the secretion or action of insulin, is a relatively common disease: nearly 6% of the United States popu- lation shows some degree of abnormality in glucose metabolism that is indicative of diabetes or a tendency toward the condition. There are two major clinical classes of diabetes mellitus: type I diabetes, or insulin- dependent diabetes mellitus (IDDM), and type II dia- betes, or non-insulin-dependent diabetes mellitus (NIDDM), also called insulin-resistant diabetes. In type I diabetes, the disease begins early in life and quickly becomes severe. This disease responds to insulin injection, because the metabolic defect stems from a paucity of pancreatic H9252 cells and a consequent inability to produce sufficient insulin. IDDM requires insulin ther- apy and careful, lifelong control of the balance between dietary intake and insulin dose. Characteristic symptoms of type I (and type II) diabetes are excessive thirst and frequent urination (polyuria), leading to the intake of large volumes of water (polydipsia) (“diabetes mellitus” means “excessive excretion of sweet urine”). These symptoms are due to the excretion of large amounts of glucose in the urine, a condition known as glucosuria. Type II diabetes is slow to develop (typically in older, obese individuals), and the symptoms are milder and often go unrecognized at first. This is really a group of diseases in which the regulatory activity of insulin is defective: insulin is produced, but some feature of the insulin-response system is defective. These individuals are insulin-resistant. The connection between type II di- abetes and obesity (discussed below) is an active area of research. Individuals with either type of diabetes are unable to take up glucose efficiently from the blood; recall that insulin triggers the movement of GLUT4 glucose trans- porters to the plasma membrane of muscle and adipose tissue (see Fig. 12–8). Another characteristic metabolic change in diabetes is excessive but incomplete oxida- tion of fatty acids in the liver. The acetyl-CoA produced by H9252 oxidation cannot be completely oxidized by the cit- ric acid cycle, because the high [NADH]/[NAD H11001 ] ratio produced by H9252 oxidation inhibits the cycle (recall that three steps convert NAD H11001 to NADH). Accumulation of acetyl-CoA leads to overproduction of the ketone bod- ies acetoacetate and H9252-hydroxybutyrate, which cannot be used by extrahepatic tissues as fast as they are made in the liver. In addition to H9252-hydroxybutyrate and ace- toacetate, the blood of diabetics also contains acetone, which results from the spontaneous decarboxylation of acetoacetate: Acetone is volatile and is exhaled, and in uncon- trolled diabetes, the breath has a characteristic odor sometimes mistaken for ethanol. A diabetic individual who is experiencing mental confusion due to high blood glucose is occasionally misdiagnosed as intoxicated, an error that can be fatal. The overproduction of ketone bodies, called ketosis, results in greatly increased con- centrations of ketone bodies in the blood (ketonemia) and urine (ketonuria). The ketone bodies are carboxylic acids, which ion- ize, releasing protons. In uncontrolled diabetes this acid production can overwhelm the capacity of the blood’s bicarbonate buffering system and produce a lowering of blood pH called acidosis or, in combination with ketosis, ketoacidosis, a potentially life-threatening condition. Biochemical measurements on blood and urine samples are essential in the diagnosis and treatment of diabetes. A sensitive diagnostic criterion is provided by 23.3 Hormonal Regulation of Fuel Metabolism 909 8885d_c23_881-919 3/1/04 1:26 PM Page 909 mac76 mac76:385_reb: the glucose-tolerance test. The patient fasts over- night, then drinks a test dose of 100 g of glucose dis- solved in a glass of water. The blood glucose concen- tration is measured before the test dose and at 30 min intervals for several hours thereafter. A healthy indi- vidual assimilates the glucose readily, the blood glucose rising to no more than about 9 or 10 mM; little or no glu- cose appears in the urine. Diabetic individuals assimi- late the test dose of glucose poorly; their blood glucose level far exceeds the kidney threshold (about 10 mM), causing glucose to appear in the urine. ■ SUMMARY 23.3 Hormonal Regulation of Fuel Metabolism ■ The concentration of glucose in blood is hormonally regulated. Fluctuations in blood glucose level (normally 60 to 90 mg/100 mL, or about 4.5 mM) due to dietary intake or vigorous exercise are counterbalanced by a variety of hormonally triggered changes in the metabolism of several organs. ■ High blood glucose elicits the release of insulin, which speeds the uptake of glucose by tissues and favors the storage of fuels as glycogen and triacylglycerols, while inhibiting fatty acid mobilization in adipose tissue. ■ Low blood glucose triggers release of glucagon, which stimulates glucose release from liver glycogen and shifts fuel metabolism in liver and muscle to fatty acid oxidation, sparing glucose for use by the brain. In prolonged fasting, triacylglycerols become the principal fuel; the liver converts the fatty acids to ketone bodies for export to other tissues, including the brain. ■ Epinephrine prepares the body for increased activity by mobilizing blood glucose from glycogen and other precursors. ■ Cortisol, released in response to a variety of stressors (including low blood glucose), stimu- lates gluconeogenesis from amino acids and glycerol in the liver, thus raising blood glucose and counterbalancing the effects of insulin. ■ In diabetes, insulin is either not produced or not recognized by the tissues, and the uptake of blood glucose is compromised. When blood glucose levels are high, glucose is excreted. Tissues then depend on fatty acids for fuel (producing ketone bodies) and degrade cellular proteins to provide glucogenic amino acids for glucose synthesis. Uncontrolled diabetes is characterized by high glucose levels in the blood and urine and the production and excretion of ketone bodies. 23.4 Obesity and the Regulation of Body Mass In the United States population, 30% of adults are obese and another 35% are overweight. (Obe- sity is defined in terms of body mass index (BMI): BMI H11005 weight in kg/(height in m) 2 . A BMI below 25 is con- sidered normal; 25 to 30 is overweight, and greater than 30, obese.) Obesity is life-threatening. It significantly in- creases the chances of developing type II diabetes as well as heart attack, stroke, and cancers of the colon, breast, prostate, and endometrium. Consequently, there is great interest in understanding how body mass and the storage of fats in adipose tissue are regulated. ■ To a first approximation, obesity is the result of tak- ing in more calories in the diet than are expended by the body’s energy-consuming activities. The body can deal with an excess of dietary calories in three ways: (1) convert excess fuel to fat and store it in adipose tis- sue, (2) burn excess fuel by extra exercise, and (3) “waste” fuel by diverting it to heat production (ther- mogenesis) in uncoupled mitochondria. In mammals, a complex set of hormonal and neuronal signals act to keep fuel intake and energy expenditure in balance, so as to hold the amount of adipose tissue at a suitable level. Dealing effectively with obesity requires under- standing these various checks and balances under nor- mal conditions, and how these homeostatic mechanisms fail in obesity. The Lipostat Theory Predicts the Feedback Regulation of Adipose Tissue The lipostat theory postulates a mechanism that in- hibits eating behavior and increases energy consump- tion whenever body weight exceeds a certain value (the set point); the inhibition is relieved when body weight drops below the set point (Fig. 23–30). This theory pre- dicts that a feedback signal originating in adipose tissue influences the brain centers that control eating behav- ior and activity (metabolic and motor). The first such factor, leptin, was discovered in 1994, and several oth- ers are now known. Leptin (Greek leptos, “thin”) is a small protein (167 amino acids) that is produced in adipocytes and moves through the blood to the brain, where it acts on recep- tors in the hypothalamus to curtail appetite. Leptin was first identified as the product of a gene designated OB (obese) in laboratory mice. Mice with two defective copies of this gene (ob/ob genotype; lowercase letters signify a mutant form of the gene) show the behavior and physiology of animals in a constant state of starva- tion: their serum cortisol levels are elevated; they are unable to stay warm, they grow abnormally, do not re- produce, and exhibit unrestrained appetite. As a con- sequence of the last effect, they become severely obese, Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism910 8885d_c23_881-919 3/1/04 1:26 PM Page 910 mac76 mac76:385_reb: weighing as much as three times more than normal mice (Fig. 23–31). They also have metabolic disturbances very similar to those of diabetic animals, and they are insulin-resistant. When leptin is injected into ob/ob mice, they lose weight and increase their locomotor ac- tivity and thermogenesis. A second mouse gene, designated DB (diabetic), has also been found to have a role in appetite regula- tion. Mice with two defective copies (db/db) are obese and diabetic. The DB gene encodes the leptin recep- tor. When the leptin receptor is defective, the signaling function of leptin is lost. The leptin receptor is expressed primarily in regions of the brain known to regulate feeding behavior— neurons of the arcuate nucleus of the hypothalamus (Fig. 23–32a). Leptin carries the message that fat re- serves are sufficient, and it promotes a reduction in fuel intake and increased expenditure of energy. Leptin- receptor interaction in the hypothalamus alters the re- lease of neuronal signals to the region of the brain that affects appetite. Leptin also stimulates the sympathetic nervous system, increasing blood pressure, heart rate, and thermogenesis by uncoupling electron transfer from ATP synthesis in the mitochondria of adipocytes (Fig. 23–32b). Recall that thermogenin, also called uncou- pling protein (UCP), forms a channel in the inner mi- tochondrial membrane that allows protons to reenter the mitochondrial matrix without passing through the ATP synthase complex (see Fig. 19–30). This permits continual oxidation of fuel (fatty acids in an adipocyte) without ATP synthesis, dissipating energy as heat and consuming dietary calories or stored fats in potentially very large amounts. 23.4 Obesity and the Regulation of Body Mass 911 FIGURE 23–30 Set-point model for maintaining constant mass. When the mass of adipose tissue increases, released leptin inhibits feeding and fat synthesis and stimulates oxidation of fatty acids. When the mass of adipose tissue decreases, a lowered leptin production fa- vors a greater food intake and less fatty acid oxidation. Food fat synthesis fat b oxidation Energy, heat Adipose tissue FIGURE 23–31 Obesity caused by defective leptin production. Both these mice, which are the same age, have defects in the OB gene. The mouse on the right was provided with purified leptin by daily injec- tion, and weighs 35 g. The mouse on the left got no leptin, conse- quently ate more food and was less active, and weighs 67 g. Nucleus Increased expression of UCP gene TAG Lipid droplet Perilipin Adenylyl cyclase Adipocyte cAMP H + G protein Protein kinase A Uncoupling protein (UCP) Fatty acids oxidation H9252 H9252 Hormone- sensitive lipase Norepinephrine Heat 3 Adrenergic receptor H9252 Hypothalamus Ventromedial nucleus Paraventricular nucleus Arcuate nucleus Posterior pituitary Anterior pituitary Leptin (via blood) Adipose tissue Neuronal signal via sympathetic neuron (a) (b) FIGURE 23–32 Hypothalamic regulation of food intake and energy expenditure. (a) Anatomy of the hypothalamus. (b) Interactions be- tween the hypothalamus and an adipocyte, described later in text. 8885d_c23_881-919 3/1/04 1:26 PM Page 911 mac76 mac76:385_reb: Leptin Stimulates Production of Anorexigenic Peptide Hormones Two types of neurons in the arcuate nucleus control fuel intake and metabolism (Fig. 23–33). The orexigenic (appetite-stimulating) neurons stimulate eating by pro- ducing and releasing neuropeptide Y (NPY), which causes the next neuron in the circuit to send the signal to the brain, Eat! The blood level of NPY rises during starvation, and is elevated in both ob/ob and db/db mice. The high NPY concentration presumably underlies the obesity of these mice, who eat voraciously. The anorexigenic (appetite-suppressing) neurons in the arcuate nucleus produce H9251-melanocyte-stimulating hormone (H9251-MSH), formed from its polypeptide pre- cursor pro-opiomelanocortin (POMC; Fig. 23–6). Release of H9251-MSH causes the next neuron in the circuit to send the signal to the brain, Stop eating! The amount of leptin released by adipose tissue de- pends on both the number and the size of adipocytes. When weight loss decreases the mass of lipid tissue, lep- tin levels in the blood decrease, the production of NPY is diminished, and the processes in adipose tissue shown in Figure 23–32 are reversed. Uncoupling is diminished, Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism912 Ghrelin Stomach Colon Adipose tissue Pancreas Leptin Insulin PYY 3-36 Neuron Muscle, adipose tissue, liver Arcuate nucleus “Eat more; metabolize less” “Eat less; metabolize more” Anorexigenic (a-MSH) PYY 3-36 receptor Melanocortin receptor Melanocortin receptor NPY receptor Leptin receptor or insulin receptor Ghrelin receptor Food intake Energy expenditure Orexigenic (NPY) FIGURE 23–33 Hormones that control eating. In the arcuate nucleus, two sets of neurosecretory cells receive hormonal input and relay neu- ronal signals to the cells of muscle, adipose tissue, and liver. Leptin and insulin are released from adipose tissue and pancreas, respec- tively, in proportion to the mass of body fat. The two hormones act on anorexigenic neurosecretory cells (red) to trigger release of H9251-MSH; this produces neuronal signals to eat less and metabolize more fuel. Leptin and insulin also act on orexigenic neurosecretory cells (green) to inhibit the release of NPY, reducing the “eat” signal sent to the tis- sues. As described later in the text, the gastric hormone ghrelin stim- ulates appetite by activating the NPY-expressing cells; PYY 3–36 , re- leased from the colon, inhibits these neurons and thereby decreases appetite. Each of the two types of neurosecretory cells inhibits hor- mone production by the other, so any stimulus that activates orexi- genic cells inactivates anorexigenic cells, and vice versa. This strength- ens the effect of stimulatory inputs. 8885d_c23_881-919 3/1/04 1:26 PM Page 912 mac76 mac76:385_reb: slowing thermogenesis and saving fuel, and fat mobi- lization slows in response to reduced signaling by cAMP. Consumption of more food combined with more efficient utilization of fuel results in replenishment of the fat re- serve in adipose, bringing the system back into balance. Leptin Triggers a Signaling Cascade That Regulates Gene Expression The leptin signal is transduced by a mechanism also used by receptors for interferon and growth factors, the JAK-STAT system (Fig. 23–34; see Fig. 12–9). The lep- tin receptor, which has a single transmembrane seg- ment, dimerizes when leptin binds to the extracellular domain of two monomers. Both monomers are phos- phorylated on a Tyr residue of the intracellular domain by a Janus kinase (JAK). The P –Tyr residues be- come docking sites for three proteins that are signal transducers and activators of transcription (STATs 3, 5, and 6, sometimes called fat-STATS). The docked STATs are then phosphorylated on Tyr residues by the same JAK. After phosphorylation, the STATs dimerize then move to the nucleus, where they bind to specific DNA sequences and stimulate the expression of target genes, including the gene for POMC, from which H9251-MSH is produced. The increased catabolism and thermogenesis trig- gered by leptin are due in part to increased synthesis of the mitochondrial uncoupling protein UCP in adipocytes. Leptin stimulates the synthesis of this un- coupling protein by altering synaptic transmissions from neurons in the arcuate nucleus to adipose and other tis- sues. In these tissues, leptin causes increased release of norepinephrine, which acts through H9252 3 -adrenergic re- ceptors to stimulate transcription of the gene for UCP. The resulting uncoupling of electron transfer from ox- idative phosphorylation consumes fat and is thermo- genic (Fig. 23–32). Might human obesity be the result of insufficient leptin production, and therefore treatable by the injec- tion of leptin? Blood levels of leptin are in fact usually much higher in obese animals (including humans) than in animals of normal body mass (except, of course, in ob/ob animals, which cannot make leptin). Some down- stream element in the leptin response system must be defective in obese individuals, and the elevation in lep- tin is the result of an (unsuccessful) attempt to over- come the leptin resistance. In those very rare humans with extreme obesity who have a defective leptin gene (OB), leptin injection does result in dramatic weight loss. In the vast majority of obese individuals, however, the OB gene is intact. In clinical trials, the injection of leptin did not have the weight-reducing effect observed in obese ob/ob mice. Clearly, most cases of human obe- sity involve one or more factors in addition to leptin. The Leptin System May Have Evolved to Regulate the Starvation Response Although much of the initial interest in leptin resulted from its possible role in preventing obesity, the leptin system probably evolved to adjust an animal’s activity and metabolism during periods of fasting and starvation, not to restrict weight. The reduction in leptin level trig- gered by nutritional deficiency reverses the thermo- genic processes illustrated in Figure 23–32, allowing fuel conservation. Leptin activates AMP-dependent pro- tein kinase (AMPK), which regulates many aspects of fuel metabolism. Leptin also triggers decreased pro- duction of thyroid hormone (slowing basal metabolism), decreased production of sex hormones (preventing reproduction), and increased production of glucocorti- coids (mobilizing the body’s fuel-generating resources). By minimizing energy expenditures and maximizing the use of endogenous reserves of energy, these leptin- mediated responses may allow an animal to survive periods of severe nutritional deprivation. 23.4 Obesity and the Regulation of Body Mass 913 Cytosol Plasma membrane Leptin receptor monomer Nucleus Leptin Leptin P JAK STAT P STAT P STAT P STAT DNA Neuropeptide (POMC, etc.) mRNA P STAT STAT JAK JAK JAK P FIGURE 23–34 The JAK-STAT mechanism of leptin signal transduc- tion in the hypothalamus. Leptin binding induces dimerization of the leptin receptor, followed by phosphorylation of Tyr residues of the receptor, catalyzed by Janus kinase (JAK). STATs bound to the phos- phorylated leptin receptor through their SH2 domains are now phos- phorylated on Tyr residues by a separate activity of JAK. The STATs dimerize, binding each other’s P –Tyr residues, and enter the nucleus. Here, they bind specific regulatory regions in the DNA and alter the expression of certain genes. The products of these genes ultimately in- fluence the organism’s feeding behavior and energy expenditure. 8885d_c23_881-919 3/1/04 1:26 PM Page 913 mac76 mac76:385_reb: Insulin Acts in the Arcuate Nucleus to Regulate Eating and Energy Conservation Insulin secretion reflects both the size of fat reserves (adiposity) and the current energy balance (blood glu- cose level). Insulin acts on insulin receptors in the hy- pothalamus to inhibit eating (Fig. 23–33). Insulin re- ceptors in the orexigenic neurons of the arcuate nucleus inhibit the release of NPY, and insulin receptors in the anorexigenic neurons stimulate H9251-MSH production, thereby decreasing fuel intake and increasing thermo- genesis. By mechanisms discussed in Section 23.3, in- sulin also signals muscle, liver, and adipose tissues to increase catabolic reactions, including fat oxidation, which results in weight loss. Leptin makes the cells of liver and muscle more sen- sitive to insulin. One hypothesis to explain this effect suggests cross-talk between the protein tyrosine kinases activated by leptin and those activated by insulin (Fig. 23–35); common second messengers in the two signal- ing pathways allow leptin to trigger some of the same downstream events that are triggered by insulin, through insulin receptor substrate-2 (IRS-2) and phos- phoinositide 3-kinase (PI-3K) (Chapter 12). Adiponectin Acts through AMPK Adiponectin is a peptide hormone (224 amino acids) produced almost exclusively in adipose tissue. It circu- lates in the blood and powerfully affects the metabolism of fatty acids and carbohydrates in liver and muscle. Adiponectin increases the uptake of fatty acids from the blood by myocytes and the rate at which fatty acids un- dergo H9252 oxidation in the muscle. It also blocks fatty acid synthesis and gluconeogenesis in hepatocytes, and it stimulates glucose uptake and catabolism in muscle and liver (Fig. 23–36). These effects of adiponectin occur indirectly, through activation of the key regulatory en- zyme AMPK by increased cytosolic [AMP]. Increased [AMP] also results from ATP consumption during in- tense muscular activity, but it can be brought about by adiponectin through other, unknown mechanisms. When activated, AMPK phosphorylates a number of tar- get proteins critical to the metabolism of fatty acids and carbohydrates, with profound effects on the metabolism of the whole animal. One enzyme regulated by AMPK is acetyl-CoA car- boxylase, which produces malonyl-CoA, the first inter- mediate committed to fatty acid synthesis. Malonyl-CoA is a powerful inhibitor of the enzyme carnitine acyl- transferase I, which starts the process of H9252 oxidation by transporting fatty acids into the mitochondrion (see Fig. 17–6). By phosphorylating and inactivating acetyl-CoA carboxylase, AMPK inhibits fatty acid synthesis while relieving the inhibition (by malonyl-CoA) of H9252 oxidation (Fig. 23–37). Mice with defective adiponectin genes are less sen- sitive to insulin than those with normal adiponectin, and they show poor glucose tolerance; ingestion of dietary carbohydrate causes a long-lasting rise in their blood glucose. These metabolic defects resemble those of Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism914 Leptin receptor Insulin receptor Plasma membrane Cytosol P P P P Leptin JAKJAK IRS-2 PI-3K Inhibition of food intake FIGURE 23–35 A possible mechanism for cross-talk between recep- tors for insulin and leptin. The insulin receptor has intrinsic Tyr ki- nase activity (see Fig. 12–6), and the leptin receptor, when occupied by its ligand, is phosphorylated by a soluble Tyr kinase (JAK). One possible explanation for the observed interaction between leptin and insulin is that both may phosphorylate the same substrate—in the case shown here, insulin receptor substrate-2 (IRS-2). When phosphory- lated, IRS-2 activates PI-3K, which has downstream consequences that include inhibition of food intake. IRS-2 serves here as an integrator of the input from two receptors. AMPK Muscle Fatty acid uptake Glucose uptake b Oxidation Liver Glycolysis Fatty acid synthesis Gluconeogenesis Adiponectin FIGURE 23–36 Effects of adiponectin on muscle and adipose tissue. By interacting with its receptors on the surface of myocytes and he- patocytes, adiponectin activates their AMPK. The activated kinase phosphorylates key metabolic enzymes (see Fig. 23–37, for example), shifting metabolism toward oxidation of fatty acids and away from lipid and glucose synthesis. 8885d_c23_881-919 3/1/04 1:26 PM Page 914 mac76 mac76:385_reb: humans with type II diabetes, who also are insulin- insensitive and clear glucose from the blood only slowly. Indeed, individuals with obesity or type II dia- betes have lower blood adiponectin levels than nondia- betic controls. Moreover, the drugs used in treatment of type II diabetes—the thiazolidinediones, such as rosigli- tazone (Avandia) and pioglitazone (Actos) (p. 807)— increase the expression of adiponectin mRNA in adipose tissue and increase blood adiponectin levels in experi- mental animals; they also activate AMPK. It appears that adiponectin, acting through AMPK, modulates the sen- sitivity of cells and tissues to insulin. Perhaps this hor- mone will prove to be one of the links between type II diabetes and its most important predisposing factor, obesity. Three factors improve the health of individuals with type II diabetes: regular exercise, use of thiazolidine- diones, and dietary restriction. We have seen that ex- ercise activates AMPK, as does adiponectin, and that thiazolidinediones increase the concentration of adiponectin in plasma, increasing insulin sensitivity. Di- etary restriction may act by regulating the expression of genes that encode proteins involved in fatty acid ox- idation and in energy expenditure via thermogenesis. Diet Regulates the Expression of Genes Central to Maintaining Body Mass Proteins in a family of ligand-activated transcription fac- tors, the peroxisome proliferator-activated recep- tors (PPARs), respond to changes in dietary lipid by altering the expression of genes involved in fat and car- bohydrate metabolism. These transcription factors were first recognized for their roles in peroxisome synthesis—thus their name. Their normal ligands are fatty acids or fatty acid derivatives, but they can also bind synthetic agonists and can be activated in the lab- oratory by genetic manipulation. PPARH9251, PPARH9254, and PPARH9253 are members of the nuclear receptor superfam- ily. They act in the nucleus by forming heterodimers with another nuclear receptor, RXR (retinoid X recep- tor), binding to regulatory regions of DNA near the genes under their control and changing the rate of tran- scription of those genes (Fig. 23–38). PPARH9253, expressed primarily in liver and adipose tissue, is involved in turning on genes necessary to the differentiation of fibroblasts into adipocytes and genes that encode proteins required for lipid synthesis and storage in adipocytes. PPARH9253 is activated by drugs of the thiazolidinedione class, which are used to treat type II diabetes. PPARH9251 in hepatocytes turns on the genes necessary for H9252 oxidation of fatty acids and formation of ketone bodies during fasting. PPARH9254 is a key regulator of fat oxidation, which acts by sensing changes in dietary lipid. It acts in liver and muscle, stimulating the transcription of at least nine genes encoding proteins for H9252 oxidation and for energy dissipation through uncoupling of mitochondria. Normal mice overfed on high-fat diets accumulate massive amounts of both brown and white fat, and fat droplets accumulate in the liver. But when the same overfeeding experiment is done with mice that have a genetically 23.4 Obesity and the Regulation of Body Mass 915 [AMP] Acetyl-CoA + CO 2 Malonyl-CoA Fatty acid synthesis Cytosol Mitochondrion Fatty acid Fatty acid Acetyl-CoA b oxidation Inactive Carnitine acyltransferase I AMPK AMPK ACC ACC P FIGURE 23–37 Regulation of fatty acid synthesis and H9252 oxidation by AMPK action on acetyl-CoA carboxylase. When activated by elevated 5H11541-AMP, AMPK phosphorylates a Thr residue on acetyl- CoA carboxylase (ACC), inactivating it. This prevents the synthesis of malonyl- CoA, the first intermediate in fatty acid synthesis, and reduction in [malonyl-CoA] relieves the inhibition of carnitine acyl- transferase I, allowing fatty acids to enter the mitochondrial matrix to undergo H9252 oxidation. 8885d_c23_881-919 3/1/04 1:26 PM Page 915 mac76 mac76:385_reb: altered, always active PPARH9254, this fat accumulation is prevented. In mice with a nonfunctioning leptin recep- tor (db/db), activated PPARH9254 prevents the development of obesity that would otherwise occur (see Fig. 23–31). By stimulating fatty acid breakdown in uncoupled mi- tochondria, PPARH9254 causes fat depletion, weight loss, and thermogenesis. Seen in this light, thermogenesis is both a means of keeping warm and a defense against obesity. Clearly, PPARH9254 is a potential target for drugs to treat obesity. Short-Term Eating Behavior Is Set by Ghrelin and PYY 3–36 Ghrelin is a peptide hormone (28 amino acids) pro- duced in cells lining the stomach. It was originally rec- ognized as the stimulus for the release of growth hor- mone (ghre is the Proto-Indo-European root of “grow”), then subsequently shown to be a powerful appetite stimulant that works on a shorter time scale (between meals) than leptin and insulin. Ghrelin receptors are lo- cated in the pituitary gland (presumably mediating growth hormone release) and in the hypothalamus (af- fecting appetite), as well as in heart muscle and adipose tissue. The concentration of ghrelin in the blood varies strikingly between meals, peaking just before a meal and dropping sharply just after the meal (Fig. 23–39). In- jection of ghrelin into humans produces immediate sen- sations of intense hunger. Individuals with Prader-Willi syndrome, whose blood levels of ghrelin are exception- ally high, have an uncontrollable appetite, leading to ex- treme obesity that often results in death before the age of 30. PYY 3–36 is a peptide hormone (34 amino acids) se- creted by endocrine cells in the lining of the small in- testine and colon in response to food entering from the stomach. The level of PYY 3–36 in the blood rises after a meal and remains high for some hours. It is carried in the blood to the arcuate nucleus, where it acts on orex- igenic neurons, inhibiting NPY release and reducing hunger (Fig. 23–33). Humans injected with PYY 3–36 feel little hunger and eat less than normal amounts for about 12 hours. This interlocking system of neuroendocrine controls of food intake and metabolism presumably evolved to protect against starvation and to eliminate counterpro- ductive accumulation of fat (extreme obesity). The dif- ficulty most people face in trying to lose weight testi- fies to the remarkable effectiveness of these controls. Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism916 Insulin (pmol/L) Grehlin (pg/mL) Breakfast 800 700 600 500 400 300 200 600 500 400 300 200 100 0 DinnerLunch (a) (b) 6 a.m. 12 noon 6 p.m. Time of day 12 midnight Thiazolidinedione PPARg PPARg RXR Cytosol Plasma membrane Nuclear envelope DNA Regulated gene Response element (DNA sequence that regulates transcription in response to PPARg ligand T) RXR T T T T FIGURE 23–38 Mode of action of PPARs. PPARs, when bound to their cognate ligand, form heterodimers with the nuclear receptor RXR. The dimer binds specific regions of DNA, response elements, stimulating transcription of genes in those regions. FIGURE 23–39 Variations in ghrelin and insulin relative to meal times. (a) Plasma levels of ghrelin rise sharply just before the normal time for meals (7 a.m. breakfast, 12 noon lunch, 5:30 p.m. dinner) and drop precipitously just after meals, paralleling the subjective feel- ings of hunger. (b) Insulin levels rise immediately after each meal, in response to the increase in blood glucose concentration. 8885d_c23_881-919 3/1/04 1:26 PM Page 916 mac76 mac76:385_reb: Chapter 23 Further Reading 917 Key Terms neuroendocrine system 882 radioimmunoassay (RIA) 884 Scatchard analysis 884 endocrine glands 886 paracrine 886 autocrine 886 insulin 887 epinephrine 888 norepinephrine 888 catecholamines 888 eicosanoid hormones 888 steroid hormones 888 vitamin D hormone 889 retinoid hormones 889 thyroid hormones 889 nitric oxide (NO) 889 NO synthase 889 hypothalamus 889 posterior pituitary 890 anterior pituitary 890 tropic hormone 890 tropin 890 hepatocyte 893 adipocyte 897 myocyte 898 erythrocyte 901 leukocyte 901 lymphocyte 901 platelets 901 blood plasma 901 plasma proteins 901 cortisol 909 diabetes mellitus 909 type I diabetes 909 type II diabetes 909 glucosuria 909 ketosis 909 acidosis 909 ketoacidosis 909 glucose-tolerance test 910 thermogenesis 910 leptin 910 thermogenin (uncoupling protein) 911 orexigenic 912 neuropeptide Y (NPY) 912 anorexigenic 912 H9251-melanocyte-stimulating hormone (H9251-MSH) 912 Janus kinase (JAK) 913 STAT (signal transducer and activator of transcription) 913 thermogenic 913 AMP-dependent protein kinase (AMPK) 913 adiponectin 914 PPAR (peroxisome proliferator-activated receptor) 915 ghrelin 916 PYY 3–36 916 Terms in bold are defined in the glossary. SUMMARY 23.4 Obesity and the Regulation of Body Mass ■ Obesity is increasingly common in the developed countries and predisposes people toward several life-threatening conditions. ■ Adipose tissue produces leptin, a hormone that regulates feeding behavior and energy expendi- ture so as to maintain adequate reserves of fat. Leptin production and release increase with the number and size of adipocytes. ■ Leptin acts on receptors in the arcuate nucleus of the hypothalamus, causing the release of anorexigenic peptides, including H9251-MSH, that act in the brain to inhibit eating. Leptin also stimulates sympathetic nervous system action on adipocytes, leading to uncoupling of mitochondrial oxidative phosphorylation, with consequent thermogenesis. ■ The signal-transduction mechanism for leptin involves phosphorylation of the JAK-STAT system. On phosphorylation by JAK, STATs can bind to regulatory regions in nuclear DNA and alter the expression of genes for the proteins that set the level of metabolic activity and determine feeding behavior. Insulin acts on receptors in the arcuate nucleus, with results similar to those caused by leptin. ■ The hormone adiponectin stimulates fatty acid uptake and oxidation and inhibits fatty acid synthesis. Its actions are mediated by AMPK. ■ Ghrelin, a hormone produced in the stomach, acts on orexigenic neurons in the arcuate nucleus to produce hunger before a meal. PYY 3–36 , a peptide hormone of the intestine, acts at the same site to lessen hunger after a meal. Further Reading General Background and History Crapo, L. (1985) Hormones: The Messengers of Life, W. H. Freeman and Company, New York. Short, entertaining account of the history and results of hormone research. Litwack, G. & Norman, A.W. (1997) Hormones, 2nd edn, Academic Press, Orlando, FL. An excellent, authoritative, well-illustrated, clear introduction to hormone structure, synthesis, and action. Yalow, R.S. (1978) Radioimmunoassay: a probe for the fine structure of biologic systems. Science 200, 1236–1245. History of the development of radioimmunoassays; the author’s Nobel lecture. Hormone Structure and Function Litwack, G. & Schmidt, R.J. (2001) Biochemistry of hormones I: polypeptide hormones. In Textbook of Biochemistry with Clinical Correlations, 5th edn (Devlin, T.M., ed.), pp. 905–958. John Wiley & Sons, Inc., New York. 8885d_c23_881-919 3/1/04 1:26 PM Page 917 mac76 mac76:385_reb: Chapter 23 Hormonal Regulation and Integration of Mammalian Metabolism918 1. ATP and Phosphocreatine as Sources of Energy for Muscle During muscle contraction, the concentration of phosphocreatine in skeletal muscle drops while the concen- tration of ATP remains fairly constant. However, in a classic experiment, Robert Davies found that if he first treated mus- cle with 1-fluoro-2,4-dinitrobenzene (p. 97), the concentra- tion of ATP declined rapidly while the concentration of phosphocreatine remained unchanged during a series of con- tractions. Suggest an explanation. 2. Metabolism of Glutamate in the Brain Brain tissue takes up glutamate from the blood, transforms it into gluta- mine, then releases it into the blood. What is accomplished by this metabolic conversion? How does it take place? The amount of glutamine produced in the brain can actually ex- ceed the amount of glutamate entering from the blood. How does this extra glutamine arise? (Hint: You may want to re- view amino acid catabolism in Chapter 18; recall that NH H11001 4 is very toxic to the brain.) Problems Litwack, G. & Schmidt, R.J. (2001) Biochemistry of hormones II: steroid hormones. In Textbook of Biochemistry with Clinical Correlations, 5th edn (Devlin, T.M., ed.), pp. 959–989. John Wiley & Sons, Inc., New York. Tissue-Specific Metabolism Arias, I.M., Boyer, J.L., Chisari, F.V., Fausto, N., Schachter, D., & Shafritz, D.A. (2001) The Liver: Biology and Pathobiol- ogy, 4th edn, Lippincott, Williams & Wilkins, Philadelphia. Advanced text; includes chapters on the metabolism of carbohydrates, fats, and proteins in the liver. Nosadini, R., Avogaro, A., Doria, A., Fioretto, P., Trevisan, R., & Morocutti, A. (1989) Ketone body metabolism: a physiolog- ical and clinical overview. Diabetes Metab. Rev. 5, 299–319. Randle, P.J. (1995) Metabolic fuel selection: general integration at the whole-body level. Proc. Nutr. Soc. 54, 317–327. Hormonal Regulation of Fuel Metabolism Attie, A.D. & Raines, R.T. (1995) Analysis of receptor-ligand interactions. J. Chem. Ed. 72, 119–123. Carling, D. (2004) The AMP-activated protein kinase cascade— a unifying system for energy control. Trends Biochem. Sci. 29, [in press]. Elia, M. (1995) General integration and regulation of metabolism at the organ level. Proc. Nutr. Soc. 54, 213–234. Holloszy, J.O. & Kohrt, W.M. (1996) Regulation of carbohydrate and fat metabolism during and after exercise. Annu. Rev. Nutr. 16, 121–138. Pilkis, S.J. & Claus, T.H. (1991) Hepatic gluconeogenesis/ glycolysis: regulation and structure/function relationships of substrate cycle enzymes. Annu. Rev. Nutr. 11, 465–515. Advanced review. Snyder, S.H. (1985) The molecular basis of communication between cells. Sci. Am. 253 (October), 132–141. Introductory discussion of the human endocrine system. Zammit, V.A. (1996) Role of insulin in hepatic fatty acid partitioning: emerging concepts. Biochem. J. 314, 1–14. Control of Body Mass Auwerx, J. & Staels, B. (1998) Leptin. Lancet 351, 737–742. Brief overview of the leptin system and JAK-STAT signal transductions. Elmquist, J.K., Maratos-Flier, E., Saper, C.B., & Flier, J.S. (1998) Unraveling the central nervous system pathways underlying responses to leptin. Nat. Neurosci. 1, 445–450. Short, excellent review. Flier, J.S. & Maratos-Flier, E. (1998) Obesity and the hypothalamus: novel peptides for new pathways. Cell 92, 437–440. Freake, H.C. (1998) Uncoupling proteins: beyond brown adipose tissue. Nutr. Rev. 56, 185–189. Review of the structure, function, and role of uncoupling proteins. Friedman, J.M. (2002) The function of leptin in nutrition, weight, and physiology. Nutr. Rev. 60, S1–S14. Intermediate-level review of all aspects of leptin biology. Inui, A. (1999) Feeding and body-weight regulation by hypothalamic neuropeptides—mediation of the actions of leptin. Trends Neurosci. 22, 62–67. Short, intermediate-level review of the leptin system. Jequier, E. & Tappy, L. (1999) Regulation of body weight in humans. Physiol. Rev. 79, 451–480. Detailed review of the role of leptin in body-weight regulation, the control of food intake, and the roles of white and brown adipose tissues in energy expenditure. Korner, J. & Aronne, L.J. (2003) The emerging science of body weight regulation and its impact on obesity treatment. J. Clin. Invest. 111, 565–570. Marx, J. (2003) Cellular warriors at the battle of the bulge. Science 299, 846–849. Short review of biochemistry of weight control, and introduction to several papers in the same issue of Science on obesity in humans. Unger, R.H. (2003) The physiology of cellular liporegulation. Annu. Rev. Physiol. 65, 333–347. Advanced review. Wang, Y.-X., Lee, C.-H., Tiep, S., Yu, R.T., Ham, J., Kang, H., & Evans, R.M. (2003) Peroxisome-proliferator-activated receptor activates fat metabolism to prevent obesity. Cell 113, 159–170. Describes the role of PPARH9254. Wood, S.C., Seeley, R.J., Porte, D., Jr., & Schwartz, M.W. (1998) Signals that regulate food intake and energy homeostasis. Science 280, 1378–1383. Review of the roles of leptin, insulin, and other neuropeptides in the regulation of feeding and catabolism. 8885d_c23_881-919 3/1/04 1:26 PM Page 918 mac76 mac76:385_reb: Chapter 23 Problems 919 3. Absence of Glycerol Kinase in Adipose Tissue Glycerol 3-phosphate is required for the biosynthesis of tria- cylglycerols. Adipocytes, specialized for the synthesis and degra- dation of triacylglycerols, cannot use glycerol directly, be- cause they lack glycerol kinase, which catalyzes the reaction Glycerol H11001 ATP 88n glycerol 3-phosphate H11001 ADP How does adipose tissue obtain the glycerol 3-phosphate necessary for triacylglycerol synthesis? 4. Oxygen Consumption during Exercise A sedentary adult consumes about 0.05 L of O 2 in 10 seconds. A sprinter, running a 100 m race, consumes about 1 L of O 2 in 10 sec- onds. After finishing the race, the sprinter continues to breathe at an elevated (but declining) rate for some minutes, consuming an extra 4 L of O 2 above the amount consumed by the sedentary individual. (a) Why does the need for O 2 increase dramatically dur- ing the sprint? (b) Why does the demand for O 2 remain high after the sprint is completed? 5. Thiamine Deficiency and Brain Function Individu- als with thiamine deficiency show some characteristic neu- rological signs and symptoms, including loss of reflexes, anx- iety, and mental confusion. Why might thiamine deficiency be manifested by changes in brain function? 6. Potency of Hormones Under normal conditions, the human adrenal medulla secretes epinephrine (C 9 H 13 NO 3 ) at a rate sufficient to maintain a concentration of 10 H1100210 M in cir- culating blood. To appreciate what that concentration means, calculate the diameter of a round swimming pool, with a wa- ter depth of 2.0 m, that would be needed to dissolve 1.0 g (about 1 teaspoon) of epinephrine to a concentration equal to that in blood. 7. Regulation of Hormone Levels in the Blood The half-life of most hormones in the blood is relatively short. For example, when radioactively labeled insulin is injected into an animal, half of the labeled hormone disappears from the blood within 30 min. (a) What is the importance of the relatively rapid inac- tivation of circulating hormones? (b) In view of this rapid inactivation, how is the level of circulating hormone kept constant under normal conditions? (c) In what ways can the organism make rapid changes in the level of a circulating hormone? 8. Water-Soluble versus Lipid-Soluble Hormones On the basis of their physical properties, hormones fall into one of two categories: those that are very soluble in water but rel- atively insoluble in lipids (e.g., epinephrine) and those that are relatively insoluble in water but highly soluble in lipids (e.g., steroid hormones). In their role as regulators of cellu- lar activity, most water-soluble hormones do not enter their target cells. The lipid-soluble hormones, by contrast, do en- ter their target cells and ultimately act in the nucleus. What is the correlation between solubility, the location of recep- tors, and the mode of action of these two classes of hormones? 9. Metabolic Differences between Muscle and Liver in a “Fight or Flight” Situation During a “fight or flight” situation, the release of epinephrine promotes glycogen breakdown in the liver, heart, and skeletal muscle. The end product of glycogen breakdown in the liver is glucose; the end product in skeletal muscle is pyruvate. (a) What is the reason for the different products of glycogen breakdown in the two tissues? (b) What is the advantage to an organism that must fight or flee of these specific glycogen breakdown routes? 10. Excessive Amounts of Insulin Secretion: Hyperin- sulinism Certain malignant tumors of the pancreas cause excessive production of insulin by the H9252 cells. Affected indi- viduals exhibit shaking and trembling, weakness and fatigue, sweating, and hunger. (a) What is the effect of hyperinsulinism on the metab- olism of carbohydrates, amino acids, and lipids by the liver? (b) What are the causes of the observed symptoms? Suggest why this condition, if prolonged, leads to brain damage. 11. Thermogenesis Caused by Thyroid Hormones Thyroid hormones are intimately involved in regulating the basal metabolic rate. Liver tissue of animals given excess thy- roxine shows an increased rate of O 2 consumption and in- creased heat output (thermogenesis), but the ATP concen- tration in the tissue is normal. Different explanations have been offered for the thermogenic effect of thyroxine. One is that excess thryroxine causes uncoupling of oxidative phos- phorylation in mitochondria. How could such an effect ac- count for the observations? Another explanation suggests that the thermogenesis is due to an increased rate of ATP uti- lization by the thyroxine-stimulated tissue. Is this a reason- able explanation? Why? 12. Function of Prohormones What are the possible ad- vantages in the synthesis of hormones as prohormones? 13. Sources of Glucose during Starvation The typical human adult uses about 160 g of glucose per day, 120 g of which is used by the brain. The available reserve of glucose (~20 g of circulating glucose and ~190 g of glycogen) is ad- equate for about one day. After the reserve has been depleted during starvation, how would the body obtain more glucose? 14. Parabiotic ob/ob mice By careful surgery, researchers can connect the circulatory systems of two mice so that the same blood circulates through both animals. In these parabi- otic mice, products released into the blood by one animal reach the other animal via the shared circulation. Both animals are free to eat independently. If an ob/ob mouse (both copies of the OB gene are defective) and a normal OB/OB mouse (two good copies of the OB gene) were made parabiotic, what would happen to the weight of each mouse? 8885d_c23_881-919 3/1/04 1:26 PM Page 919 mac76 mac76:385_reb: The bacterial RNA polymerase 8885d_c24_920 2/11/04 3:10 PM Page 920 mac76 mac76:385_reb: