免疫学（英文版）：Chapter 23

■ Experimental Animal Models ■ Cell-Culture Systems ■ Protein Biochemistry ■ Recombinant DNA Technology ■ Analysis of DNA Regulatory Sequences ■ Gene Transfer into Mammalian Cells ■ Microarrays—An Approach for Analyzing Patterns of Gene Expression Addition of Expression Profile of Diffuse Large B-cell Lymphoma. Experimental Systems ??????????? ??????? ?? ??????? ????? ??? used to unravel the complex cellular interactions of the immune response. In vivo systems, which involve the whole animal, provide the most natural experi- mental conditions. However, in vivo systems have a myriad of unknown and uncontrollable cellular interactions that add ambiguity to the interpretation of data. At the other extreme are in vitro systems, in which defined populations of lymphocytes are studied under controlled and conse- quently repeatable conditions; in vitro systems can be sim- plified to the extent that individual cellular interactions can be studied effectively. Yet they have their own limitations, the most notable of which is their artificiality. For example, pro- viding antigen to purified B cells in vitro does not stimulate maximal antibody production unless T cells are present. Therefore a study of antibody production in an artificial in vitro system that lacks T cells could lead to the incorrect con- clusion that B cells do not synthesize high levels of antibod- ies. One must ask whether a cellular response observed in vitro reflects reality or is a product of the unique conditions of the in vitro system itself. This chapter describes some of the experimental systems routinely used to study the immune system. It also covers some recombinant DNA techniques that have revolution- ized the study of the immune system in the past decade or so. Other chapters also cover experimental systems and tech- niques in detail. Table 23-1 lists them and directs the reader to the appropriate location for a description. Experimental Animal Models The study of the immune system in vertebrates requires suit- able animal models. The choice of an animal depends on its suitability for attaining a particular research goal. If large amounts of antiserum are sought, a rabbit, goat, sheep, or horse might be an appropriate experimental animal. If the goal is development of a protective vaccine, the animal cho- sen must be susceptible to the infectious agent so that the efficacy of the vaccine can be assessed. Mice or rabbits can be used for vaccine development if they are susceptible to the pathogen. But if growth of the infectious agent is limited to humans and primates, vaccine development may require the use of monkeys, chimpanzees, or baboons. For most basic research in immunology, mice have been the experimental animal of choice. They are easy to handle, are genetically well characterized, and have a rapid breeding cycle. The immune system of the mouse has been character- ized more extensively than that of any other species. The value of basic research in the mouse system is highlighted by the enormous impact this research has had on clinical inter- vention in human disease. Inbred Strains Can Reduce Experimental Variation To control experimental variation caused by differences in the genetic backgrounds of experimental animals, immu- nologists often work with inbred strains—that is, genetically identical animals produced by inbreeding. The rapid breed- ing cycle of mice makes them particularly well suited for the production of inbred strains, which are developed by re- peated inbreeding between brother and sister littermates. In this way the heterozygosity of alleles that is normally found in randomly outbred mice is replaced by homozygosity at all chapter 23 ART TK ART TK E loci. Repeated inbreeding for 20 generations usually yields an inbred strain whose progeny are homozygous at more than 98% of all loci. More than 150 different inbred strains of mice are available, each designated by a series of letters and/ or numbers (Table 23-2). Most strains can be purchased by immunologists from such suppliers as the Jackson Labora- tory in Bar Harbor, Maine. Inbred strains have also been pro- duced in rats, guinea pigs, hamsters, rabbits, and domestic fowl. Because inbred strains of animals are genetically identi- cal (syngeneic) within that strain, their immune responses can be studied in the absence of variables introduced by indi- vidual genetic differences—an invaluable property. With inbred strains, lymphocyte subpopulations isolated from one animal can be injected into another animal of the same strain without eliciting a rejection reaction. This type of experi- mental system permitted immunologists to first demonstrate that lymphocytes from an antigen-primed animal could trans- fer immunity to an unprimed syngeneic recipient. Adoptive-Transfer Systems Permit the in Vivo Examination of Isolated Cell Populations In some experiments, it is important to eliminate the im- mune responsiveness of the syngeneic host so that the re- sponse of only the transferred lymphocytes can be studied in isolation. This can be accomplished by a technique called adoptive transfer: first, the syngeneic host is exposed to x-rays that kill its lymphocytes; then the donor immune cells are introduced. Subjecting a mouse to high doses of x-rays (650–750 rads) can kill 99.99% of its lymphocytes, after which the activities of lymphocytes transplanted from the spleen of a syngeneic donor can be studied without inter- ference from host lymphocytes. If the host’s hematopoietic cells might influence an adoptive-transfer experiment, then higher x-ray levels (900–1000 rads) are used to eliminate the entire hematopoietic system. Mice irradiated with such doses will die unless reconstituted with bone marrow from a syn- geneic donor. The adoptive-transfer system has enabled immunologists to study the development of injected lymphoid stem cells in various organs of the recipient, and have facilitated the study of various populations of lymphocytes and of the cellular in- teractions required to generate an immune response. Such ex- periments, for instance, first enabled immunologists to show that a T helper cell is necessary for B-cell activation in the humoral response. In these experiments, adoptive transfer of purified B cells or purified T cells did not produce antibody in the irradiated host. Only when both cell populations were transferred was antibody produced in response to antigen. SCID Mice and SCID-Human Mice Are a Valuable Animal Model for Immunodeficiency An autosomal recessive mutation resulting in severe com- bined immunodeficiency disease (SCID) developed sponta- neously in a strain of mice called CB-17. These CB-17 SCID mice fail to develop mature T and B cells and consequently are severely compromised immunologically. This defect is due to a failure in V(D)J recombination. SCID mice must be housed in a sterile (germ-free) environment, because they cannot fight off microorganisms of even low pathogenicity. The absence of functional T and B cells enables these mice to accept foreign cells and grafts from other strains of mice or even from other species. Apart from their lack of functional T and B cells, SCID mice appear to be normal in all respects. When normal bone- marrow cells are injected into SCID mice, normal T and B cells develop, and the mice are cured of their immunodeficiency. This finding has made SCID mice a valuable model system for the study of immunodeficiency and the process of differentia- tion of bone-marrow stem cells into mature T or B cells. Interest in SCID mice mushroomed when it was found that they could be used to study the human immune system. In this system, portions of human fetal liver, adult thymus, 526 PART IV The Immune System in Health and Disease TABLE 23-1 Immunological methods described in other chapters Method Location Bone-marrow transplantation Ch. 2 Clinical Focus Preparation of immunotoxins Fig. 4-22 Genetic engineering of Fig. 5-20 and chimeric mouse-human Ch 5 Clinical Focus monoclonal antibodies Determination of antibody affinity Fig. 6.2 by equilibrium dialysis Precipitation reactions Fig. 6.4 Immunodiffusion and Figs. 6.5 and 6.6 immunoelectrophoresis Hemagglutination Fig. 6.7 Radioimmunoassay (RIA) Fig. 6.9 ELISA assays Fig. 6.10 ELISPOT assay Fig. 6.11 Western blotting Fig. 6.12 Immunoprecipitation Fig. 6.13 Immunofluorescence Fig. 6.14 Flow cytometry Fig. 6.15 Production of congenic mice Fig. 7-3 Mixed lymphocyte reaction (MLR) Fig. 14-16 Cell-mediated lympholysis (CML) Fig. 14-17 Production of vaccinia vector vaccine Fig. 18-5 Production of multivalent Fig. 18-7 subunit vaccines HLA typing Fig. 21-4 and adult lymph nodes are implanted into SCID mice (Fig- ure 23-1). Because the mice lack mature T and B cells of their own, they do not reject the transplanted human tissue. The implanted human fetal liver contains immature lymphocytes (stem cells), which migrate to the implanted human tissues, where they mature into T and B cells, producing a SCID- human mouse. Because the human lymphocytes are exposed to mouse antigens while they are still immature, they later recognize mouse cells as self and do not mount an immuno- logic response against the mouse host. Experimental Systems CHAPTER 23 527 TABLE 23-2 Some inbred mouse strains commonly used in immunology Strain Common substrains Characteristics A A/He High incidence of mammary tumors in some substrains A/J A/WySn AKR AKR/J High incidence of leukemia AKR/N AKR/Cum Thy 1.2 allele in AKR/Cum, and Thy 1.1 allele in other substrains (Thy gene encodes a T-cell surface protein) BALB/c BALB/cj Sensitivity to radiation BALB/c AnN Used in hybridoma technology BALB/cBy Many myeloma cell lines were generated in these mice CBA CBA/J Gene (rd) causing retinal degeneration in CBA/J CBA/H CBA/N Gene (xid) causing X-linked immunodeficiency in CBA/N C3HC3H/He Gene (rd) causing retinal degeneration C3H/HeJ High incidence of mammary tumors in many substrains (these carry a C3H/HeN mammary-tumor virus that is passed via maternal milk to offspring) C57BL/6 C57BL/6J High incidence of hepatomas after irradiation C57BL/6By High complement activity C57BL/6N C57BL/10 C57BL/10JVery close relationship to C57BL/6 but differences in at least two loci C57BL/10ScSn C57BL/10NFrequent partner in preparation of congenic mice C57BR C57BR/cdj High frequency of pituitary and liver tumors Very resistant to x-irradiation C57LC57L/J Susceptibility to experimental autoimmune encephalomyelitis (EAE) C57L/N High frequency of pituitary and reticular cell tumors C58 C58/J High incidence of leukemia C58/LwN DBA/1 DBA/1J High incidence of mammary tumors DBA/1N DBA/2 DBA/2J Low immune response to some antigens DBA/2N Low response to pneumococcal polysaccharide type III HRS HRS/J Hairless (hr) gene, usually in heterozygous state NZB NZB/BINJ High incidence of autoimmune hemolytic anemia and lupus-like nephritis NZB/N Autoimmune disease similar to systemic lupus erythematosus (SLE) in F 1 progeny from crosses with NZW NZW NZW/N SLE-type autoimmune disease in F 1 progeny from crosses with NZB P P/J High incidence of leukemia SJL SJL/J High level of aggression and severe fighting to the point of death, especially in males Tendency to develop certain autoimmune diseases, most susceptible to EAE SWR SWR/J Tendency to develop several autoimmune diseases, especially EAE 129 129/J High incidence of spontaneous teratocarcinoma 129/SvJ SOURCE: Adapted from Federation of American Societies for Experimental Biology, 1979, Biological Handbooks, Vol. III: Inbred and Genetically Defined Strains of Laboratory Animals. The beauty of the SCID-human mouse is that it enables one to study human lymphocytes within an animal model. This valuable system has proved useful in research on the development of various lymphoid cells and also as an impor- tant animal model in AIDS research, since mouse lympho- cytes cannot be infected with HIV, whereas the lymphocytes of a SCID-human mouse are readily infected. Cell-Culture Systems The complexity of the cellular interactions that generate an immune response has led immunologists to rely heavily on various types of in vitro cell-culture systems. A variety of cells can be cultured, including primary lymphoid cells, cloned lymphoid cell lines, and hybrid cells. Primary Lymphoid Cell Cultures Primary lymphoid cell cultures can be obtained by isolating lymphocytes directly from blood or lymph or from various lymphoid organs by tissue dispersion. The lymphocytes can then be grown in a chemically defined basal medium (con- taining saline, sugars, amino acids, vitamins, trace elements, and other nutrients) to which various serum supplements are added. For some experiments, serum-free culture condi- tions are employed. Because in vitro culture techniques re- quire from 10- to 100-fold fewer lymphocytes than do typical in vivo techniques, they have enabled immunologists to assess the functional properties of minor subpopulations of lymphocytes. It was by means of cell-culture techniques, for example, that immunologists were first able to define the functional differences between CD4 + T helper cells and CD8 + T cytotoxic cells. Cell-culture techniques have also been used to identify numerous cytokines involved in the activation, growth, and differentiation of various cells involved in the immune re- sponse. Early experiments showed that media conditioned, or modified, by the growth of various lymphocytes or antigen- presenting cells would support the growth of other lymphoid cells. Conditioned media contain the secreted products from actively growing cells. Many of the individual cytokines that characterized various conditioned media have subsequently been identified and purified, and in many cases the genes that encode them have been cloned. These cytokines, which play a central role in the activation and regulation of the immune response, are described in Chapter 12 and elsewhere. Cloned Lymphoid Cell Lines A primary lymphoid cell culture comprises a heterogeneous group of cells that can be propagated only for a limited time. This heterogeneity can complicate the interpretation of ex- perimental results. To avoid these problems, immunologists use cloned lymphoid cell lines and hybrid cells. Normal mammalian cells generally have a finite life span in culture; that is, after a number of population doublings char- acteristic of the species and cell type, the cells stop dividing. In contrast, tumor cells or normal cells that have undergone transformation induced by chemical carcinogens or viruses can be propagated indefinitely in tissue culture; thus, they are said to be immortal. Such cells are referred to as cell lines. The first cell line—the mouse fibroblast L cell—was de- rived in the 1940s from cultured mouse subcutaneous con- nective tissue by exposing the cultured cells to a chemical carcinogen, methylcholanthrene, over a 4-month period. In the 1950s, another important cell line, the HeLa cell, was de- rived by culturing human cervical cancer cells. Since these early studies, hundreds of cell lines have been established, each consisting of a population of genetically identical (syngeneic) cells that can be grown indefinitely in culture. Table 23-3 lists some of the cell lines used in immunologic research and briefly describes their properties. Some were derived from spontaneously occurring tumors of lympho- cytes, macrophages, or other accessory cells involved in the im- mune response. In other cases, the cell line was induced by transformation of normal lymphoid cells with viruses such as Abelson’s murine leukemia virus (A-MLV), simian virus 40 528 PART IV The Immune System in Health and Disease SCID mouse Transplant human thymus and lymph-node tissue under kidney capsule Inject with human fetal liver cells (stem cells) Stem cells migrate to the human thymus Human thymus releases mature human T cells into circulation SCID–human mouse FIGURE 23-1 Production of SCID-human mouse. This system permits study of human lymphocytes within an animal model. In this example, human T-cells are transferred to SCID mouse, but B-cells also can be transferred by the use of bone-marrow precursors. (SV40), Epstein-Barr virus (EBV), or human T-cell leukemia virus type 1(HTLV-1). Lymphoid cell lines differ from primary lymphoid cell cultures in several important ways: They survive indefinitely in tissue culture, show various abnormal growth properties, and often have an abnormal number of chromosomes. Cells with more or less than the normal diploid number of chro- mosomes for a species are said to be aneuploid. The big advantage of cloned lymphoid cell lines is that they can be grown for extended periods in tissue culture, enabling im- munologists to obtain large numbers of homogeneous cells in culture. Until the late 1970s, immunologists did not succeed in maintaining normal T cells in tissue culture for extended periods. In 1978, a serendipitous finding led to the observa- tion that conditioned medium containing a T-cell growth factor was required. The essential component of the condi- tioned medium turned out to be interleukin 2 (IL-2). By cul- turing normal T lymphocytes with antigen in the presence of IL-2, clones of antigen-specific T lymphocytes could be iso- lated. These individual clones could be propagated and stud- ied in culture and even frozen for storage. After thawing, the clones continued to grow and express their original antigen- specific functions. Development of cloned lymphoid cell lines has enabled immunologists to study a number of events that previously could not be examined. For example, research on the molec- ular events involved in activation of naive lymphocytes by antigen was hampered by the low frequency of naive B and T cells specific for a particular antigen; in a heterogeneous population of lymphocytes, the molecular changes occurring in one responding cell could not be detected against a back- ground of 10 3 –10 6 nonresponding cells. Cloned T- and B-cell lines with known antigenic specificity have provided immu- nologists with large homogeneous cell populations in which to study the events involved in antigen recognition. Similarly, the genetic changes corresponding to different maturational stages can be studied in cell lines that appear to be “frozen” at different stages of differentiation. Cell lines have also been useful in studying the soluble factors produced by lymphoid cells. Some cell lines secrete large quantities of various cyto- kines; other lines express membrane receptors for particular cytokines. These cell lines have been used by immunologists to purify various cytokines and their receptors and eventu- ally to clone their genes. With the advantages of lymphoid cell lines come a number of limitations. Variants arise spontaneously in the course of prolonged culture, necessitating frequent subcloning to limit the cellular heterogeneity that can develop. If variants are selected in subcloning, it is possible that two subclones derived from the same parent clone may represent different subpopu- lations. Moreover, any cell line derived from tumor cells or transformed cells may have unknown genetic contributions characteristic of the tumor or of the transformed state; thus, researchers must be cautious when extrapolating results ob- tained with cell lines to the normal situation in vivo. Neverthe- less, transformed cell lines have made a major contribution to the study of the immune response, and many molecular events discovered in experiments with transformed cell lines have been shown to take place in normal lymphocytes. Hybrid Lymphoid Cell Lines In somatic-cell hybridization, immunologists fuse normal B or T lymphocytes with tumor cells, obtaining hybrid cells, or Experimental Systems CHAPTER 23 529 TABLE 23-3 Cell lines commonly used in immunologic research Cell line Description L-929 Mouse fibroblast cell line; often used in DNA transfection studies and to assay tumor necrosis factor (TNF) SP2/0 Nonsecreting mouse myeloma; often used as a fusion partner for hybridoma secretion P3X63-Ag8.653 Nonsecreting mouse myeloma; often used as a fusion partner for hybridoma secretion MPC 11 Mouse IgG2b-secreting myeloma P3X63-Ag8 Mouse IgG1-secreting myeloma MOPC 315 Mouse IgA-secreting myeloma J558 Mouse IgA-secreting myeloma 7OZ/3 Mouse pre–B-cell lymphoma; used to study early events in B-cell differentiation BCL 1 Mouse B-cell leukemia lymphoma that expresses membrane IgM and IgD and can be activated with mitogen to secrete IgM CTLL-2 Mouse T-cell line whose growth is dependent on IL-2; often used to assay IL-2 production Jurkat Human T-cell leukemia that secretes IL-2 DO11.10 Mouse T-cell hybridoma with specificity for ovalbumin PU 5-1.8 Mouse monocyte-macrophage line P338 D1 Mouse monocyte-macrophage line that secretes high levels of IL-1 WEHI 265.1 Mouse monocyte line P815 Mouse mastocytoma cells; often used as target to assess killing by cytotoxic T lymphocytes (CTLs) YAC-1 Mouse lymphoma cells; often used as target for NK cells HL-60 Human myeloid-leukemia cell line COS-1 African green monkey kidney cells transformed by SV40; often used in DNA transfection studies heterokaryons, containing nuclei from both parent cells. Ran- dom loss of some chromosomes and subsequent cell prolifer- ation yield a clone of cells that contain a single nucleus with chromosomes from each of the fused cells; such a clone is called a hybridoma. Historically, cell fusion was promoted with Sendai virus, but now it is generally done with polyethylene glycol. Normal antigen-primed B cells can be fused with cancerous plasma cells, called myeloma cells (Figure 23-2). The hybridoma thus formed continues to express the antibody genes of the normal B lymphocyte but is capable of unlimited growth, a characteristic of the myeloma cell. B-cell hybridomas that secrete antibody with a single antigenic specificity, called monoclonal antibody, in reference to its derivation from a single clone, have revolutionized not only immunology but biomedical research as well as the clinical laboratory. Chapter 4 describes the production and uses of monoclonal antibod- ies in detail (see Figures 4-21). T-cell hybridomas can also be obtained by fusing T lym- phocytes with cancerous T-cell lymphomas. Again, the result- ing hybridoma continues to express the genes of the normal T cell but acquires the immortal-growth properties of the can- cerous T lymphoma cell. Immunologists have generated a number of stable hybridoma cell lines representing T-helper and T-cytotoxic lineages. Protein Biochemistry The structures and functions of many important molecules of the immune system have been determined with the tech- niques of protein biochemistry, and many of these tech- niques are in constant service in experimental immunology. For example, fluorescent and radioactive labels allow immu- nologists to localize and visualize molecular activities, and the ability to determine such biochemical characteristics of a protein as its size, shape, and three-dimensional structure has provided essential information for understanding the func- tions of immunologically important molecules. Radiolabeling Techniques Allow Sensitive Detection of Antigens or Antibodies Radioactive labels on antigen or antibody are extremely sen- sitive markers for detection and quantification. There are a number of ways to introduce radioactive isotopes into pro- teins or peptides. For example, tyrosine residues may be labeled with radioiodine by chemical or enzymatic proce- dures. These reactions attach an iodine atom to the phenol ring of the tyrosine molecule. One of the enzymatic iodina- tion techniques, which uses lactoperoxidase, can label pro- teins on the plasma membrane of a live cell without labeling proteins in the cytoplasm, allowing the study of cell-surface proteins without isolating them from other cell constituents. 530 PART IV The Immune System in Health and Disease Polyethylene glycol Chromosomes Normal T or B cell (dies after 7–10 days in culture) Cancerous T or B cell (grows continuously in culture) Heterokaryon Nucleus of cancer cell Nucleus of normal lymphocyte Random chromosomal loss Hybridoma (expresses some normal B-cell or T-cell genes but grows indefinitely like a cancer cell) B-cell hybridoma T-cell hybridoma Monoclonal antibody Interleukin 2 (IL-2) FIGURE 23-2 Production of B-cell and T-cell hybridomas by somatic-cell hybridization. The resulting hybridomas express some of the genes of the original normal B or T cell but also exhibit the immortal-growth properties of the tumor cell. This procedure is used to produce B-cell hybridomas that secrete monoclonal antibody and T-cell hybridomas that secrete various growth factors. A general radiolabeling of cell proteins may be carried out by growing the cells in a medium that contains one or more radiolabeled amino acids. The amino acids selected for this application are those most resistant to metabolic modification during cell growth so that the radioactive label will appear in the cell protein rather than in all cell constituents. Leucine marked with 14 C or 3 H, and cysteine or methionine labeled with 35 S, are the most commonly used amino acids for meta- bolic labeling of proteins. Table 23-4 lists some properties of the radioisotopes used in immunologic research. Biotin Labels Facilitate Detection of Small Amounts of Proteins In some instances direct labeling of proteins, especially with enzymes or other large molecules, as described in Chapter 6, may cause denaturation and loss of activity. A convenient la- beling system has been developed which may be used in con- junction with the ELISA and ELISPOT assays described in Chapter 6. This labeling technique exploits the high affinity of the reaction between the vitamin biotin and avidin, a large molecule that may be labeled with radioactive isotopes, with fluorescent molecules, or with enzymes. Biotin is a small molecule (mol. wt. 244) that can be coupled to an antibody (or to any protein molecule) by a gentle chemical reaction that causes no loss of antibody activity. After the biotin-coupled antibody has reacted in the assay system, the labeled avidin is introduced and binding is measured by detecting the label on the avidin molecule (Figure 23-3). The reaction between bio- tin and avidin is highly specific and of such high affinity that the bond between the two molecules under most assay condi- tions is virtually irreversible. Gel Electrophoresis Separates Proteins by Size and Charge When subjected to an electric field in an electrophoresis chamber, a charged molecule will move toward the oppo- sitely charged electrode. The rate at which a charged mole- cule moves in a stable field (its electrophoretic mobility) depends upon two factors specific to the molecule: one is the sign and magnitude of its net electrical charge, and the other is its size and shape. All other factors being equal, if mole- cules are of equal size the one with higher net charge will move faster in an applied electrical field due to the molecular seiving properties of the solid medium. It also follows that small molecules will move faster than large ones of the same net charge. Although there are exceptions in which the shape of a molecule may increase or decrease its frictional drag and cause atypical migration behavior, these general principles underlie all electrophoretic separations. Most electrophoretic separations are not conducted in free solution but rather in a stable supporting medium, such as a gel. The most popular in reseach laboratories is a poly- merized and crosslinked form of acrylamide. Separation on polyacrylamide gels, commonly referred to as polyacrylamide Experimental Systems CHAPTER 23 531 TABLE 23-4 Radioisotopes commonly used in immunology laboratories Isotope Half-life Radiation type* Autoradiography ? 125 I 60.0 da H9253 + 131 I 6.8 da H9253 + 51 Cr 27.8 da H9253 – 32 P 14.3 da H9252 + 35 S 87.4 da H9252 + 14 C 57.30 yrs H9252 + 3 H 12.35 yrs H9252 – * H9253 (gamma) radiation may be detected in a solid scintillation counter. H9252 (beta) radiation is detected in a liquid scintillation counter by its ability to convert energy to photons of light in a solution containing phosphorescent compounds. ? Radiation may also be detected by exposure to x-ray film. 35 S and 14 C must be placed in direct contact with film for detection. 3 H cannot be detected by normal autoradiographic techniques. Labeled avidin Ag Ab Biotin active ester Biotinylated Ab Avidin bound to biotinylated Ab FIGURE 23-3 Labeling of antibody with biotin. An antibody prepara- tion is mixed with a biotin ester, which reacts with the antibody. The biotin-labeled antibody can be used to detect antigens on a solid sub- strate such as the well of a microtiter plate. After washing away unbound antibody, the bound antibody can be detected with labeled avidin. The avidin can be radioactively labeled or linked to an enzyme that catalyzes a color reaction, as in ELISA procedures (see Figure 6-10). gel electrophoresis (PAGE), may be used for analysis of pro- teins or nucleic acids (Figure 23-4a). In one common application, the electrophoresis of pro- teins through a polyacrylamide gel is carried out in the pres- ence of the detergent sodium dodecyl sulfate (SDS). This method, known as SDS-PAGE, provides a relatively simple and highly effective means of separating mixtures of proteins on the basis of size. SDS is a negatively charged detergent that binds to protein in amounts proportional to the length of the protein. This binding destroys the characteristic tertiary and secondary structure of the protein, transforming it into a negatively charged rod. A protein binds so many negatively charged SDS molecules that its own intrinsic charge becomes insignificant by comparison with the net charge of the SDS molecules. Therefore, treatment of a mixture of proteins with SDS transforms them into a collection of rods whose electric charges are proportional to their molecular weights. This has two extremely useful consequences. First, it is possible to sep- arate the components of a mixture of proteins according to molecular weight. Second, because the electrophoretic mobil- ity, or distance traveled by a species during SDS-PAGE, is in- versely proportional to the logarithm of its molecular weight, that distance is a measure of its molecular weight. The gel is stained with a dye that reacts with protein to visualize the locations of the proteins. The migration distance of a protein in question is then compared with a plot of the distances migrated by a set of standard proteins (Figure 23-4b). Another electrophoretic technique, isoelectric focusing (IEF), separates proteins solely on the basis of their charge. This method is based on the fact that a molecule will move in an electric field as long as it has a net positive or negative charge; molecules that bear equal numbers of positive and negative charges and therefore have a net charge of zero will not move. At most pH values, proteins (which characteristi- cally bear a number of both positive and negative charges) have either a net negative or a net positive charge. However, 532 PART IV The Immune System in Health and Disease Apparent mass (kd) 70 10 20 30 40 50 60 1.00.2 0.4 0.6 0.8 Relative mobility Anode Cathode Sample wells Sample Buffer Gel Plastic frame ? + + Top Mass (kd) Stable pH gradient Bottom 200 100 68 43 36 29 17 12 (a) (c) (b) Buffer Molecules migrate to position at which their net charge is zero ? ? ?? ? ? + ? ++ ? ?? + pH 7.0 6.0 5.0 ? 7.0 6.0 5.0 ? ++ + ?? ?? ++ + ? Direction of electrophoresis FIGURE 23-4 Gel electrophoresis. (a) A standard PAGE apparatus with cathode at the top and anode at the bottom. Samples are loaded on the top of the gel in sample wells and electrophoresis is accom- plished by running a current from the cathode to the anode. (b) The electrophoretic mobility, or distance traveled by a species during SDS- PAGE, is inversely proportional to the log of its molecular weight. The molecular weight of a protein is readily determined by the log of its migration distance with a standard curve that plots the migration dis- tances of the set of standard proteins against the logs of their molecu- lar weights. (c) Isoelectric focusing, or IEF, separates proteins solely by charge. Proteins are placed on a stable pH gradient and subjected to electrophoresis. Each protein migrates to its isoelectric point, the point at which its net charge is zero. [Part (b) after K. Weber and M. Osborn, 1975, The Proteins, 3rd ed., vol. 1, p. 179. Academic Press.] for each protein there is a particular pH, called its isoelectric point (pI), at which that protein has equal numbers of posi- tive and negative charges. Isoelectric focusing makes use of a gel containing substances, called carrier ampholytes, that ar- range themselves into a continuous pH gradient when sub- jected to an electric field. When a mixture of proteins is ap- plied to such a gel and subjected to electrophoresis, each protein moves until it reaches that point in the gradient where the pH of the gel is equal to its isoelectric point. It then stops moving because it has a net charge of zero. Isoelectric focusing is an extremely gentle and effective way of separat- ing different proteins (Figure 23-4c). A method known as two-dimensional gel electrophoresis (2D gel electrophoresis) combines the advantages of SDS- PAGE and isoelectric focusing in one of the most sensitive and discriminating ways of analyzing a mixture of proteins. In this method, one first subjects the mixture to isoelectric focusing on an IEF tube gel, which separates the molecules on the basis of their isoelectric points without regard to mol- ecular weight. This is the first dimension. In the next step, one places the IEF gel lengthwise across the top of an SDS- polyacrylamide slab (that is, in place of the sample wells in Figure 23-4a) and runs SDS-PAGE. Preparatory to this step, all proteins have been reacted with SDS and therefore mi- grate out of the IEF gel and through the SDS-PAGE slab ac- cording to their molecular weights. This is the second dimen- sion. The position of the proteins in the resulting 2D gel can be visualized in a number of ways. In the least sensitive the gel is stained with a protein-binding dye (such as Coomassie blue). If the proteins have been radiolabeled, the more sensi- tive method of autoradiography can be used. Alternatively, silver staining is a method of great sensitivity that takes ad- vantage of the capacity of proteins to reduce silver ions to an easily visualized deposit of metallic silver. Finally, immuno- blotting—blotting of proteins onto a membrane and detec- tion with antibody (see Figure 6-13)—can be used as a way of locating the position of specific proteins on 2D gels if an ap- propriate antibody is available. Figure 23-5 shows an autora- diograph of a two-dimensional gel of labeled proteins from murine thymocytes. X-Ray Crystallography Provides Structural Information A great deal of information about the structure of cells, parts of cells, and even molecules has been obtained by light micro- scopy. The microscope uses a lens to focus radiation to form an image after it has passed through a specimen. However, a practical limitation of light microscopy is the limit of resolu- tion. Radiation of a given wavelength cannot resolve struc- tural features less than about 1/2 its wavelength. Since the shortest wavelength of visible light is around 400 nm, even the very best light microscopes have a theoretical limit of res- olution of no less than 200 nm. Because of the much shorter wavelength (0.004 nm) of the electron at the voltages normally used in the electron microscope, the theoretical limit of resolution of the electron microscope is about 0.002 nm. If it were possible to build an instrument that could actually approach this limit, the elec- tron microscope could readily be used to determine the detailed atomic arrangement of biological molecules, since the constituent atoms are separated by distances of 0.1 nm to 0.2 nm. In practice, aberrations inherent in the operation of the magnetic lenses that are used to image the electron beam limit the resolution to about 0.1 nm (1?). This practical limit can be reached in the examination of certain specimens, par- ticularly metals. Other considerations, however, such as specimen preparation and contrast, limit the resolution for biological materials to about 2 nm (20 ?). To determine the arrangement of a molecule’s atoms, then, we must turn to x-rays, a form of electromagnetic radiation that is readily generated in wavelengths on the order of size of interatomic distances. Even though there are no microscopes with lenses that can focus x-rays into images, x-ray crystallography can reveal molecular structure at an extraordinary level of detail. X-ray crystallography is based on the analysis of the diffrac- tion pattern produced by the scattering of an x-ray beam as it passes though a crystal. The degree to which a particular atom scatters x-rays depends upon its size. Atoms such as carbon, oxygen, or nitrogen, scatter x-rays more than do hydrogen atoms, and larger atoms, such as iron, iodide, or mercury give intense scattering. X-rays are a form of electromagnetic waves; Experimental Systems CHAPTER 23 533 Acidic Basic FIGURE 23-5 Two-dimensional gel electrophoresis of 35 S-methionine labeled total cell proteins from murine thymocytes. These proteins were first subjected to isoelectric focusing (direction of migration indi- cated by red arrow) and then the focused proteins were separated by SDS-PAGE (direction of migration indicated by blue arrow). The gel was exposed to x-ray film to detect the labeled proteins. [Courtesy of B. A. Osborne.] as the scattered waves overlap, they alternately interfere with and reinforce each other. An appropriately placed detector records a pattern of spots (the diffraction pattern) whose dis- tribution and intensities are determined by the structure of the diffracting crystal. This relationship between crystal structure and diffraction pattern is the basis of x-ray crystallographic analysis. Here is an overview of the procedures used: OBTAIN CRYSTALS OF THE PROTEIN OF INTEREST. To those who have not experienced the frustrations of crystallizing proteins, this may seem a trivial and incidental step of an otherwise highly sophisticated process. It is not. There is great variation from protein to protein in the conditions required to produce crystals that are of a size and geometrical formation appro- priate for x-ray diffraction analysis. For example, myoglobin formed crystals over the course of several days at pH 7 in a 3M solution of ammonium sulfate, but 1.5 M ammonium sulfate at pH 4 worked well for a human IgG1. There is no set formula that can be applied, and those who are consistently successful are persistent, determined, and, like great chefs, have a knack for making just the right “sauce.” SELECTION AND MOUNTING. Crystal specimens must be at least 0.1 mm in the smallest dimension and rarely exceed a few millimeters in any dimension. Once chosen, a crystal is har- vested into a capillary tube along with the solution from which the crystal was grown (the “mother liquor”). This keeps the crys- tal from drying and maintains its solvent content, an important consideration for maintaining the internal order of the speci- men. The capillary is then mounted in the diffraction apparatus. GENERATING AND RECORDING A DIFFRACTION PATTERN. The precisely positioned crystal is then irradiated with x-rays of a known wavelength produced by accelerating electrons against the copper target of an x-ray tube. When the x-ray beam strikes the crystal, some of it goes straight through and some is scat- tered; sensitive detectors record the position and intensity of the scattered beam as a pattern of spots (Figure 23-6a,b). INTERPRETING THE DIFFRACTION PATTERN. The core of diffraction analysis is the mathematical deduction of the detailed structure that would produce the diffraction pattern observed. One must calculate to what extent the waves scat- tered by each atom have combined to reinforce or cancel each other to produce the net intensity observed for each spot in the array. A difficulty arises in the interpretation of complex diffraction patterns because the waves differ with respect to phase, the timing of the period between maxima and min- ima. Since the pattern observed is the net result of the inter- action of many waves, information about phase is critical to calculating the distribution of electron densities that is re- sponsible. The solution of this “phase problem” looms as a major obstacle to the derivation of a high-resolution struc- ture of any complex molecule. The problem is solved by derivatizing the protein—mod- ifying it by adding heavy atoms, such as mercury, and then obtaining crystals that have the same geometry as (are iso- 534 PART IV The Immune System in Health and Disease X-ray source X-ray beam Crystal Detector (e.g., film) Diffracted beams (a) (b) Tyr 100H Gly 97 Gly 96 Asp 101 Tyr 102 Tyr 100I Ala 100J Met 100K Trp 103 (c) FIGURE 23-6 X-ray crystallography. (a) Schematic diagram of an x-ray crystallographic experiment in which an x-ray beam bombards the crystal and diffracted rays are detected. (b) Section of x-ray dif- fraction pattern of a crystal of murine IgG2a. (c) Section from the electron-density map of murine IgG2a. [Part (a) from L. Stryer, 1995, Biochemistry, 4th ed.; parts (b) and (c) courtesy of A. McPherson.] morphous with) those of the underivatized protein. The dif- fraction pattern of the isomorphous crystal is obtained and compared with that of the native protein. Usually, armed with a knowledge of the diffraction patterns of two or more isomorphous heavy-atom derivatives, the phases for the na- tive protein can be calculated by reference to the characteris- tic diffraction patterns generated by heavy-atom landmarks. The phases established, it is possible to move on to a calcula- tion of the distribution of electron density. This is accom- plished by Fourier synthesis, a mathematical treatment par- ticularly suited to the analysis of periodic phenomena such as those involving waves. In this case, it is used to compute the distribution of electron density along the x, y, and z axes within a unit cell of the crystal. The deduced electron density can then be visualized on a computer (Figure 23-6c). DERIVATION OF THE STRUCTURE. The resolution of a model depends upon a number of factors. First of all, the ultimate resolution possible is set by the quality of the crystal and the internal order of the crystal. Even the highest-quality crystals have a degree of internal disorder that establishes a limit of resolution of about 2 ?. Second, a factor of para- mount importance is the number of intensities fed into the Fourier synthesis. A relatively small number of spots may produce a low-resolution (6 ?) image that traces the course of the polypeptide chain but provides little additional struc- tural information. On the other hand, the processing of data provided by tens of thousands of spots allows the tracing of very detailed electron-density maps. Provided one knows the amino-acid sequence of the protein, such maps can guide the construction of high-resolution, three-dimensional models. Amino-acid sequence data is necessary because it can be dif- ficult, and in some cases impossible, to unambiguously dis- tinguish among some amino-acid side chains on even the most detailed electron-density maps. Since 1960, when the first detailed structures of proteins were deduced, the structures of many thousands of proteins have been solved. These range from small and (relatively) sim- ple proteins such as lysozyme, consisting of a single polypep- tide chain, to poliovirus, an 8,500,000 dalton, stunningly complex nucleoprotein made up of RNA encased by multiple copies of four different polypeptide subunits. Of particular importance to immunologists are the large number of immu- nologically relevant molecules for which detailed crystal struc- tures are now available. These include many immunoglob- ulins, most of the major and minor proteins involved in the MHC and T-cell–receptor complexes, and many other impor- tant immunological macromolecules, with new structures and structural variants appearing every month. Recombinant DNA Technology The various techniques called recombinant DNA technology have had an impact on every area of immunologic research. Genes can be cloned, DNA can be sequenced, and recombi- nant proteins can be produced, supplying immunologists with defined components for study of the structure and function of the immune system at the molecular level. This section briefly describes some of the recombinant DNA techniques com- monly employed in immunologic research; examples of their use have been presented throughout the book. Restriction Enzymes Cleave DNA at Precise Sequences A variety of bacteria produce enzymes, called restriction endonucleases, that degrade foreign DNA (e.g., bacteriophage DNA) but spare the bacterial-cell DNA, which contains methylated residues. The discovery of these bacterial enzymes in the 1970s opened the way to a major technological ad- vance in the field of molecular biology. Before the discovery of restriction endonucleases, double-stranded DNA (dsDNA) could be cut only with DNases. These enzymes do not recog- nize defined sites and therefore randomly cleave DNA into a variable series of small fragments, which are impossible to sort by size or sequence. In contrast, restriction endonucleases rec- ognize and cleave DNA at specific sites, called restriction sites, which are short double-stranded segments of specific se- quence containing four to eight nucleotides (Table 23-5). Experimental Systems CHAPTER 23 535 TABLE 23-5 Some restriction enzymes and their recognition sequences Sequence* 5H11032n 3H11032 Microorganism source Abbreviation 3H11032n 5H11032 Bacillus amyloliquefaciens H BamHI G G A T C C C C T A G G Escherichia coli RY 13 EcoRI G A A T T C C T T A A G Haemophilus aegyptius HaeIII G G C C C C G G Haemophilus influenzae Rd HindIII A A G C T T T T C G A A Haemophilus parainfluenzae HpaIG T T A A C C A A T T G Nocardia otitidis-caviarum NotIG CGGCCGC C GCCGGCG Providencia stuartii 164 PstIG T G C A G G A C G T C Staphylococcus aureus 3A Sau3AG A T C CT A G *Blue lines indicate locations of single-strand cuts within the restriction site. Enzymes that make off-center cuts produce fragments with short single- stranded extensions at their ends. SOURCE: New England Biolabs, http://www.neb.com. A restriction endonuclease cuts both DNA strands at a spe- cific point within its restriction site. Some enzymes, such as HpaI, cut on the central axis and thus generate blunt-ended fragments. Other enzymes, such as EcoRI, cut the DNA at staggered points in the recognition site. In this case, the end of each cleaved fragment is a short segment of single-stranded DNA, called a sticky end. When two different DNA molecules are cut with the same restriction enzyme that makes staggered cuts, the sticky ends of the fragments are complementary; under appropriate conditions, fragments from the two mole- cules can be joined by base pairing to generate a recombinant DNA molecule. Several hundred different restriction endonu- cleases have been isolated and many are available commer- cially, allowing researchers to purchase enzymes that cut DNA at defined restriction sites. Cloning of DNA Sequences The development of DNA-cloning technology in the 1970s provided a means of amplifying a given DNA fragment to such an extent that unlimited amounts of identical DNA frag- ments (cloned DNA) could be produced. Cloning Vectors Are Useful to Replicate Defined Sequences of DNA In DNA cloning, a given DNA fragment is inserted into an autonomously replicating DNA molecule, called a cloning vector, so that the inserted DNA is replicated with the vector. A number of different viruses have been used as vectors, including bacterial viruses, insect viruses, and mammalian retroviruses. A common bacterial virus used as a vector is bacteriophage H9261.Ifa gene is inserted into bacteriophage H9261 and the resulting recombinant H9261 phage is used to infect E. coli, the inserted gene will be expressed by the bacteria. Retroviruses, which can infect virtually any type of mam- malian cell, are a common vector used to clone DNA in mammalian cells. Retroviruses are RNA viruses that contain reverse transcriptase, an enzyme that catalyzes conversion of the viral RNA genome into DNA. The viral DNA then inte- grates into the host chromosomal DNA, where it is retained as a provirus, replicating along with the host chromosomal DNA at each cell division. When a retrovirus is used as a vec- tor in research, most of the retroviral genes are removed so that the vector cannot produce viral particles; the retroviral genes that are left include a strong promoter region, located at the 5H11032 end of the viral genome, in a sequence called the long terminal repeat (LTR). If a gene is inserted into such a retro- viral vector and the vector is then used to infect mammalian cells, expression of the gene will be under the control of the retroviral promoter region. Plasmids are another common type of cloning vector. A plasmid is a small, circular, extrachromosomal DNA mole- cule that can replicate independently in a host cell; the most common host used in DNA cloning is E. coli. In general, the DNA to be cloned is inserted into a plasmid that contains an antibiotic-resistance gene. After the recombinant plasmid is incubated with bacterial cells, the cells containing the recom- binant plasmid can be selected by their ability to grow in the presence of the antibiotic. Another type of vector that is often used for cloning is called a cosmid vector. This type of vector is a plasmid that has been genetically engineered to contain the COS sites of H9261-phage DNA, a drug-resistance gene, and a replication ori- gin. COS sites are DNA sequences that allow any DNA up to 50 kb in length to be packaged into the H9261-phage head. Cloning of cDNA and Genomic DNA Allows the Isolation of Defined Sequences Messenger RNA (mRNA) isolated from cells can be tran- scribed into complementary DNA (cDNA) with the enzyme reverse transcriptase. The cDNA can be cloned by inserting it into a plasmid vector carrying a gene that confers resistance to an antibiotic, such as ampicillin. The resulting recombi- nant plasmid DNA is subsequently transferred into specially treated E. coli cells by one of several techniques; the transfer process is called transfection. If the foreign DNA is incor- porated into the host cell and expressed, the cell is said to be transformed. When the cells are cultured on agar plates containing ampicillin, only transformed cells containing the ampicillin-resistance gene will survive and grow (Figure 23-7). A collection of DNA sequences within plasmid vectors repre- senting all the mRNA sequences derived from a cell or tissue is called a cDNA library. A cDNA library differs from a genomic library (see Figure 23-8) by virtue of the fact that it contains only the sequences derived from mRNA, the sequences that represent expressed genes. Genomic cloning, cloning of the entire genome of an ani- mal, requires specialized vectors. E. coli plasmid vectors are impractical for cloning of all the genomic DNA fragments that constitute a large genome because of the low efficiency of E. coli transformation and the small number of transformed colonies that can be detected on a typical petri dish. Instead, cloning vectors derived from bacteriophage H9261 are used to clone genomic DNA fragments obtained by cleaving chromo- somal DNA with restriction enzymes (Figure 23-8). Bacterio- phage H9261 DNA is 48.5 kb long and contains a central section of about 15 kb that is not necessary for H9261 replication in E. coli and can therefore be replaced with foreign genomic DNA. As long as the recombinant DNA does not exceed the length of the original H9261-phage DNA by more than 5%, it can be pack- aged into the H9261-phage head and propagated in E. coli. This means that somewhat more than 1.5 H11003 10 4 base pairs can be cloned in one particle of H9261 phage. A collection of H9261 clones that includes all the DNA sequences of a given species is called a genomic library. It has been calculated that about 1 million different recombinant H9261-phage particles would be needed to form a genomic DNA library representing an entire mam- malian genome, which contains about 3 H11003 10 9 base pairs. Often the 20–25 kb stretch of DNA that can be cloned in bacteriophage H9261 is not long enough to include the regulatory 536 PART IV The Immune System in Health and Disease Experimental Systems CHAPTER 23 537 Plasmid vector DNA fragment to be cloned Ampicillin-resistance gene Enzymatically insert DNA into plasmid vector Recombinant plasmid Mix E. coli cells with plasmids in presence of CaCl 2 Culture on nutrient agar plates containing ampicillin Bacterial chromosome Transformed E. coli cell survives Cells that do not take up plasmid die on ampicillin plates Independent plasmid replication Cell multiplication Colony of cells each containing copies of the same recombinant plasmid + Human DNA (~3 × 10 9 bp) Recombinant λ DNA Package into λ head Anneal and ligate Cut with BamHI Replaceable region Partially digest with Sau3A 49 kb Bacteriophage λ DNA 20 kb fragment with sticky ends λ vector arms with sticky ends FIGURE 23-7 cDNA cloning using a plasmid vector. A plasmid containing a replication origin and an ampicillin-resistance gene is cut with a restriction endonuclease that produces blunt ends. After addition of a poly-C tail to the 3H11032 ends of the cDNA and of a comple- mentary poly-G tail to the 3H11032 ends of the cut plasmid, the two DNAs are mixed, annealed, and joined by DNA ligase, forming the recom- binant plasmid. Uptake of the recombinant plasmid into E. coli cells is stimulated by high concentrations of CaCl 2 . Transformation occurs with a low frequency, but the transformed cells can be selected in the presence of ampicillin. [Adapted from H. Lodish et al., 1995, Molecu- lar Cell Biology, 3rd ed. Scientific American Books.] FIGURE 23-8 Genomic DNA cloning using bacteriophage H9261 as the vector. Genomic DNA is partly digested with Sau3A, producing fragments with sticky ends. The central 15-kb region of the H9261-phage DNA is cut out with BamHI and discarded. These two restriction en- zymes produce complementary sticky ends, so the genomic and DNA fragments can be annealed and ligated. After the resulting re- combinant DNA is packaged into a H9261-phage head, it can be propa- gated in E. coli. sequences that lie outside the 5H11032 and 3H11032 ends of the direct cod- ing sequences of a gene. As noted already, larger genomic DNA fragments—between 30 and 50 kb in length—can be cloned in a cosmid vector. A recombinant cosmid vector, although not a fully functional bacteriophage, can infect E. coli and replicate as a plasmid, generating a cosmid library. Recently, a larger E. coli virus, called bacteriophage P1, has been used to pack- age DNA fragments up to 100 kb long. Even larger DNA frag- ments, greater than a megabase (1000 kb) in length, can be cloned in yeast artificial chromosomes (YACs), which are lin- ear DNA segments that can replicate in yeast cells (Table 23-6). The BAC, or bacterial artifical chromosome, is another useful vector. BACs can accept pieces of DNA up to 100–300 kb in length. Although YACs accept larger inserts of foreign DNA, BACs are much easier to propagate and are the vector of choice for many large-scale cloning efforts. Selection of DNA Clones Once a cDNA or genomic DNA library has been prepared, it can be screened to identify a particular DNA fragment by a technique called in situ hybridization. The cloned bacterial colonies, yeast colonies, or phage plaques containing the re- combinant DNA are transferred onto nitrocellulose or nylon filters by replica plating (Figure 23-9). The filter is then treated with NaOH, which both lyses the bacteria and dena- tures the DNA, allowing single-stranded DNA (ssDNA) to bind to the filter. The filter with bound DNA is then incu- bated with a radioactive probe specific for the gene of inter- est. The probe will hybridize with DNA in the colonies or plaques on the filter that contain the sought-after gene, and they can be identified by autoradiography. The position of the positive colonies or plaques on the filter shows where the corresponding clones can be found on the original agar plate. Various radioactive probes can be used to screen a library. In some cases, radiolabeled mRNA or cDNA serves as the probe. If the protein encoded by the gene of interest has been purified and partly sequenced, it is possible to work backward from the amino-acid sequence to determine the probable nu- cleotide sequence of the corresponding gene. A known se- quence of five or six amino-acid residues is all that is needed to synthesize radiolabeled oligonucleotide probes with which to screen a cDNA or genomic library for a particular gene. To cope with the degeneracy of the genetic code, peptide seg- ments containing amino acids encoded by a limited number of codons are usually chosen. Oligonucleotides representing 538 PART IV The Immune System in Health and Disease TABLE 23-6 Vectors and maximum length of DNA that they can carry Maximum length of Vector type cloned DNA (kb) Plasmid 20 Bacteriophage H9261 25 Cosmid 45 Bacteriophage P1 100 Bacterial artificial chromosome (BAC) 100–300 Yeast artificial chromosome (YAC) >1000 Incubate Wash and expose filter to photographic film Radioactively labeled DNA probe Treat filter with NaOH to lyse bacteria or phage and to denature DNA Replication of colonies on special filter Petri dish with colonies of bacteria containing recombinant plasmids or plaques from recombinant λ phage ssDNA bound to filter Colonies containing gene of interest Position of desired colonies detected by autoradiography FIGURE 23-9 Selection of specific clones from a cDNA or ge- nomic DNA library by in situ hybridization. A nitrocellulose or nylon filter is placed against the plate to pick up the bacterial colonies or phage plaques containing the cloned genes. After the filter is placed in a NaOH solution and heated, the denatured ssDNA becomes fixed to the filter. A radioactive probe specific for the gene of interest is incubated with the filter. The position of the colonies or plaques containing the desired gene is revealed by autoradiography. all possible codons for the peptide are then synthesized and used as probes to screen the DNA library. Southern Blotting Detects DNA of a Given Sequence DNA fragments generated by restriction-endonuclease cleav- age can be separated on the basis of length by agarose gel electrophoresis. The shorter a fragment is, the faster it moves in the gel. An elegant technique developed by E. M. Southern can be used to identify any band containing fragments with a given gene sequence (Figure 23-10). In this technique, called Southern blotting, DNA is cut with restriction enzymes and the fragments are separated according to size by electro- phoresis on an agarose gel. Then the gel is soaked in NaOH to denature the dsDNA, and the resulting ssDNA fragments are transferred onto a nitrocellulose or nylon filter by capillary action. After transfer, the filter is incubated with an appropri- ate radiolabeled probe specific for the gene of interest. The probe hybridizes with the ssDNA fragment containing the gene of interest, and the position of the band containing these hybridized fragments is determined by autoradiogra- phy. Southern-blot analysis played a critical role in unravel- ing the mechanism by which diversity of antibodies and T-cell receptors is generated (see Figures 5-2 and 9-2). Northern Blotting Detects mRNA Northern blotting (named for its similarity to Southern blotting) is used to detect the presence of specific mRNA molecules. In this procedure the mRNA is first denatured to ensure that it is in an unfolded, linear form. The mRNA mol- ecules are then separated according to size by electrophoresis and transferred to a nitrocellulose filter, to which the mRNAs will adhere. The filter is then incubated with a labeled DNA probe and subjected to autoradiography. Northern-blot analy- sis is often used to determine how much of a specific mRNA is expressed in cells under different conditions. Increased lev- els of mRNA will bind proportionally more of the labeled DNA probe. Polymerase Chain Reaction Amplifies Small Quantities of DNA The polymerase chain reaction (PCR) is a powerful tech- nique for amplifying specific DNA sequences even when they are present at extremely low levels in a complex mixture (Fig- ure 23-11). The procedure requires that the DNA sequences that flank the desired DNA sequence be known, so that short oligonucleotide primers can be synthesized. The DNA mix- ture is denatured into single strands by a brief heat treat- ment. The DNA is then cooled in the presence of an excess of the oligonucleotide primers, which hybridize with the comple- mentary ssDNA. A temperature-resistant DNA polymerase is then added, together with the four deoxyribonucleotide tri- phosphates, and each strand is copied. The newly synthesized DNA duplex is separated by heating and the cycle is repeated. In each cycle there is a doubling of the desired DNA se- quence; in only 25 cycles the desired DNA sequence can be amplified about a million-fold. The DNA amplified by the PCR can be further character- ized by Southern blotting, restriction-enzyme mapping, and Experimental Systems CHAPTER 23 539 Gel electrophoresis Hybridize with labeled DNA or RNA probe DNA Cleave with restriction enzymes Blotting: capillary action transfers DNA from gel to filter Filter Autoradiography Alkaline solution Paper towels Gel Filter DNA fragmentsDNA fragments DNA of interest FIGURE 23-10 The Southern-blot technique for detecting specific sequences in DNA fragments. The DNA fragments produced by restriction-enzyme cleavage are separated by size by agarose gel elec- trophoresis. The agarose gel is overlaid with a nitrocellulose or nylon filter and a thick stack of paper towels. The gel is then placed in an al- kaline salt solution, which denatures the DNA. As the paper towels soak up the moisture, the solution is drawn through the gel into the fil- ter, transferring each ssDNA band to the filter. This process is called blotting. After heating, the filter is incubated with a radiolabeled probe specific for the sequence of interest; DNA fragments that hybridize with the probe are detected by autoradiography. [Adapted from J. Darnell et al., 1990, Molecular Cell Biology, 2nd ed., Scientific American Books.] direct DNA sequencing. The PCR technique has enabled im- munologists to amplify genes encoding proteins that are important in the immune response, such as MHC molecules, the T-cell receptor, and immunoglobulins. Analysis of DNA Regulatory Sequences The transcriptional activity of genes is regulated by promoter and enhancer sequences. These sequences are cis-acting, mean- ing that they regulate only genes on the same DNA molecule. The promoter sequence lies upstream from the gene it regu- lates and includes a TATA box, where the general transcrip- tion machinery, including RNA polymerase II, binds and begins transcription. The enhancer sequence confers a high rate of transcription on the promoter. Unlike the promoter, which always lies upstream from the gene it controls, the enhancer element can be located anywhere with respect to the gene (5H11032 of the promoter, 3H11032 of the gene, or even in an in- tron of the gene). The activity of enhancer and promoter sequences is con- trolled by transcription factors, which are DNA-binding pro- teins. These proteins bind to specific nucleotide sequences within promoters and enhancers and act either to enhance or suppress their activity. Enhancer and promoter sequences and their respective DNA-binding proteins have been identi- fied by a variety of techniques, including DNA footprinting, gel-shift analysis, and the CAT assay. 540 PART IV The Immune System in Health and Disease Region to be amplified Primer 1 Primer 2 3' 5' 5' 3' 3' + 3' 3'5' 3' 5' 5' 5' Add excess primers 1 and 2 Heat to 95° to melt strands Cool to 60° to anneal primers Heat to 95° to melt strands Add dNTPs and Taq polymerase to extend primers Cool to 60° to anneal primers Cycle 1 Cycle 2 Add dNTPs and Taq polymerase to extend primers Repeat annealing and extension steps After multiple cycles, desired region is amplified: FIGURE 23-11 The polymerase chain reaction (PCR). DNA is de- natured into single strands by a brief heat treatment and is then cooled in the presence of an excess of oligonucleotide primers com- plementary to the DNA sequences flanking the desired DNA seg- ment. A heat-resistant DNA polymerase is used to copy the DNA from the 3H11032 ends of the primers. Because all of the reaction compo- nents are heat stable, the heating and cooling cycle can be repeated many times, resulting in alternate DNA melting and synthesis, and rapid amplification of a given sequence. [Adapted from H. Lodish et al., 1995, Molecular Cell Biology, 3rd ed., Scientific American Books.] DNA Footprinting Identifies the Sites Where Proteins Bind DNA The binding sites for DNA-binding proteins on enhancers and promoters can be identified by a technique called DNA footprinting (Figure 23-12a). In this technique, a cloned DNA fragment containing a putative enhancer or promoter se- quence is first radiolabeled at the 5H11032 end with 32 P. T he labeled DNA is then divided into two fractions: one fraction is in- cubated with a nuclear extract containing a DNA-binding protein; the other DNA fraction is not incubated with the extract. Both DNA samples are then digested with a nuclease or a chemical that makes random cuts in the phosphodiester bonds of the DNA, and the strands are separated. The result- ing DNA fragments are run on a gel to separate fragments of different sizes. In the absence of DNA-binding proteins, a complete ladder of bands is obtained on the electrophoretic gel. When a protein that binds to a site on the DNA fragment is present, it covers some of the nucleotides, protecting that stretch of the DNA from digestion. The electrophoretic pat- tern of such protected DNA will contain blank regions (or footprints). Each footprint represents the site within an en- hancer or promoter that binds a particular DNA-binding protein. Experimental Systems CHAPTER 23 541 Protected from cleavage GC Sp1 binds to GC box DNase I 32 P 32 P 32 P 32 P 32 P 32 P 32 P No protein added Sp1 protein End-label with 32 P Restriction fragment or oligonucleotide X-ray film Electrophoresis Autoradiography Footprint Mix GC GC Nuclear extract Restriction fragment or oligonucleotide Nuclear extract Specific binding by Sp1 DNA without added proteins DNA-protein complex Free DNA fragment End-label with 32 P X-ray film Electrophoresis Autoradiography Mix GC (a) DNA footprinting (b) Gel-shift analysis GC FIGURE 23-12 Identification of DNA sequences that bind protein by DNA-footprinting and gel-shift analysis. (a) In the footprinting tech- nique, labeled DNA fragments containing a putative promoter or en- hancer sequence are incubated in the presence and absence of a DNA-binding protein (e.g., Sp1 protein, which binds to a “GC box,” a GC-rich region of DNA). After the samples are treated with DNase and the strands separated, the resulting fragments are electrophoresed; the gel then is subjected to autoradiography. A blank region (footprint) in the gel pattern indicates that protein has bound to the DNA. (b) In gel- shift analysis, a labeled DNA fragment is incubated with a cellular extract containing transcription factors. The electrophoretic mobility of the DNA-protein complex is slower than that of free DNA fragments. [Adapted from J. D. Watson et al., 1992, Recombinant DNA, 2nd ed., W. H. Freeman and Company.] Gel-Shift Analysis Identifies DNA-Protein Complexes When a protein binds to a DNA fragment, forming a DNA- protein complex, the electrophoretic mobility of the DNA fragment in a gel is reduced, producing a shift in the position of the band containing that fragment. This phenomenon is the basis of gel-shift analysis. In this technique, radioactively labeled cloned DNA containing an enhancer or a promoter se- quence is incubated with a nuclear extract containing a DNA- binding protein (Figure 23-12b). The DNA-protein complex is then electrophoresed and its electrophoretic mobility is com- pared with that of the cloned DNA alone. A shift in the mobil- ity indicates that a protein is bound to the DNA, retarding its migration on the electrophoretic gel. CAT Assays Measure Transcriptional Activity One way to assess promoter activity is to engineer and clone a DNA sequence containing a reporter gene attached to the promoter that is being assessed. When this sequence, or con- struct, is introduced into eukaryotic cells, transcription will be initiated from the promoter if it is active, and the reporter gene will be transcribed and its protein product synthesized. Measuring the amount of this protein produced is thus a way to determine the activity of the promoter. Most reporter genes are chosen because they encode pro- teins that can be easily measured, such as the enzyme chloram- phenicol acetyltransferase (CAT), which transfers the acetyl group from acetyl-CoA to the antibiotic chloramphenicol (Figure 23-13). The more active the promoter, the more CAT will be produced within the transfected cell. By introducing mutations into promoter sequences and then assaying for pro- moter activity with the corresponding reporter gene, con- served sequence motifs have been identified within promoters. Another reporter gene, the firefly luciferase gene, is also conve- nient and easy to use. Luciferase activity is analyzed by the emission of light, which is detected by a luminometer. Gene Transfer into Mammalian Cells A variety of genes involved in the immune response have been isolated and cloned by use of recombinant DNA tech- niques. The expression and regulation of these genes has been studied by introducing them into cultured mammalian cells and, more recently, into the germ line of animals. Cloned Genes Transferred into Cultured Cells Allow in Vitro Analysis of Gene Function Diverse techniques have been developed for transfecting genes into cells. A common technique involves the use of a retrovirus in which a viral structural gene has been replaced with the cloned gene to be transfected. The altered retrovirus is then used as a vector for introducing the cloned gene into cultured cells. Because of the properties of retroviruses, the recombi- nant DNA integrates into the cellular genome with a high fre- quency. In an alternative method, the cloned gene of interest is complexed with calcium phosphate. The calcium-phosphate– DNA complex is slowly precipitated onto the cells and the DNA is taken up by a small percentage of them. In another transfection method, called electroporation, an electric cur- rent creates pores in cell membranes through which the cloned DNA is taken up. In both of these latter methods, the trans- 542 PART IV The Immune System in Health and Disease Promoter CAT gene Transfect into cells mRNA CAT enzyme Lyse cells Add [ 14 C] chloramphenicol and acetyl-CoA Inactive promoter Active promoter Incubate at 37° Acetylated chloramphenicol Chloramphenicol Autoradiogram of thin-layer chromatogram FIGURE 23-13 CAT assay for assessing functional activity of a pro- moter sequence. In this assay, a DNA construct consisting of the pro- moter of interest and the reporter gene encoding chloramphenicol acetyltransferase (CAT) is introduced (transfected) into eukaryotic cells. If the promoter is active, the CAT gene will be transcribed and the CAT enzyme will be produced within the transfected cell. The presence of the enzyme can easily be detected by lysing the cell and incubating the cell lysate with [ 14 C] chloramphenicol and acetyl-CoA. If present, the CAT en- zyme will transfer the acetyl group from acetyl-CoA to the chloram- phenicol, forming acetylated chloramphenicol, which can be easily detected by thin-layer chromatography. [Adapted from J. D. Watson et al., 1992, Recombinant DNA, 2nd ed., W. H. Freeman and Company.] fected DNA integrates, apparently at random sites, into the DNA of a small percentage of treated cells. Generally, the cloned DNA being transfected is engineered to contain a selectable marker gene, such as one that confers resistance to neomycin. After transfection, the cells are cul- tured in the presence of neomycin. Because only the trans- fected cells are able to grow, the small number of transfected cells in the total cell population can be identified and selected. Transfection of cloned genes into cells has proved to be highly effective in immunologic research. By transfecting genes involved with the immune response into cells that lack those genes, the product of a specific gene can be studied apart from interacting proteins encoded by other genes. For example, transfection of MHC genes, under the control of appropriate promoters, into a mouse fibroblast cell line (L929, or simply L, cells) has enabled immunologists to study the role of MHC molecules in antigen presentation to T cells (Figure 23-14). Transfection of the gene that encodes the T-cell receptor has provided information about the antigen-MHC specificity of the T-cell receptor. Cloned Genes Tranferred into Mouse Embryos Allow in Vivo Analysis of Gene Function Development of techniques to introduce cloned foreign genes (called transgenes) into mouse embryos has permitted im- munologists to study the effects of immune-system genes in vivo.Ifthe introduced gene integrates stably into the germ-line cells, it will be transmitted to the offspring. Two techniques for producing transgenic mice are described in this section; one of these has been used to produce knockout mice, which cannot express a particular gene product (Table 23-7). Transgenic Mice Aid in the Analysis of Gene Function The first step in producing transgenic mice is injection of for- eign cloned DNA into a fertilized egg. In this technically de- manding process, fertilized mouse eggs are held under suction at the end of a pipet and the transgene is microinjected into one of the pronuclei with a fine needle. The transgene inte- grates into the chromosomal DNA of the pronucleus and is passed on to the daughter cells of eggs that survive the pro- cess. The eggs then are implanted in the oviduct of “pseudo- pregnant” females, and transgenic pups are born after 19 or Experimental Systems CHAPTER 23 543 Class II MHC α chain DNA Class II MHC β chain DNA Transfection Fibroblasts Selection with G418 and mycophenolic acid Class II MHC neo gpt FIGURE 23-14 Transfection of the genes encoding the class II MHC H9251 chain and H9252 chain into mouse fibroblast L cells, which do not nor- mally produce these proteins. Two constructs containing one of the MHC genes and a selectable gene were engineered: the H9251-chain gene with the guanine phosphoribosyl transferase gene (gpt), which confers resistance to the drug G418, and the H9252-chain gene with a neomycin gene (neo), which confers resistance to mycophenolic acid. After trans- fection, the cells are placed in medium containing both G418 and my- cophenolic acid. Only those fibroblasts containing both the neo and gpt genes (and consequently the genes encoding the class II MHC H9251 and H9252 chains) will survive this selection. These fibroblasts will express both class II MHC chains on their membranes. Some tumor-associated antigens under examination as potential targets for mono Characteristic Transgenic mice Knockout mice Cells receiving DNA Zygote Embryonic stem (ES) cells DNA constructs used Natural gene or cDNA Mutated gene Means of delivery Microinjeciton into zygote and Transfer of ES cells to blastocyst implantation into foster mother and implantation into foster mother Outcome Gain of a gene Loss of gene TABLE 23-7 Comparison of transgenic and knockout mice 20 days of gestation (Figure 23-15). In general the efficiency of this procedure is low, with only one or two transgenic mice produced for every 100 fertilized egg collected. With transgenic mice, immunologists have been able to study the expression of a given gene in a living animal. Al- though all the cells in a transgenic animal contain the trans- gene, differences in the expression of the transgene in differ- ent tissues has shed light on mechanisms of tissue-specific gene expression. By constructing a transgene with a particular promoter, researchers can control the expression of a given transgene. For example, the metallothionein promoter is acti- vated by zinc. Transgenic mice carrying a transgene linked to a metallothionein promoter express the transgene only if zinc is added to their water supply. Other promoters are functional only in certain tissues; the insulin promoter, for in- stance, promotes transcription only in pancreatic cells. Trans- genic mice carrying a transgene linked to the insulin pro- moter, therefore, will express the transgene in the pancreas but not in other tissues. Because a transgene is integrated into the chromosomal DNA within the one-celled mouse embryo, it will be inte- grated into both somatic cells and germ-line cells. The result- ing transgenic mice thus can transmit the transgene to their offspring as a Mendelian trait. In this way, it has been possi- ble to produce lines of transgenic mice in which every mem- ber of a line contains the same transgene. A variety of such transgenic lines are currently available and are widely used in immunologic research. Included among these are lines carry- ing transgenes that encode immunoglobulin, T-cell receptor, class I and class II MHC molecules, various foreign antigens, and a number of cytokines. Several lines carrying oncogenes as transgenes also have been produced. Gene-Targeted Knockout Mice Assess the Contribution of a Particular Gene One of the limitations with transgenic mice is that the trans- gene is integrated randomly within the genome. This means that some transgenes insert in regions of DNA that are not transcriptionally active, and hence the gene is not expressed. To circumvent this limitation, researchers have developed a technique in which a desired gene is targeted to specific sites within the germ line of a mouse. The primary use of this tech- nique has been to replace a normal gene with a mutant allele or a disrupted form of the gene, thus knocking out the gene’s function. Transgenic mice that carry such a disrupted gene, called knockout mice, have been extremely helpful to immu- nologists trying to understand how the removal of a particular gene product affects the immune system. Various knockout mice are being used in immunologic research, including mice that lack particular cytokines or MHC molecules. Production of gene-targeted knockout mice involves the following steps: ■ Isolation and culturing of embryonic stem (ES) cells from the inner cell mass of a mouse blastocyst 544 PART IV The Immune System in Health and Disease Collect fertilized eggs Offspring Implant injected eggs into oviduct of pseudo-pregnant female Pseudo-pregnant female Test for presence of transgene Breed transgenics About 10–30% of offspring contain transgene × Inject cloned DNA into one of the pronuclei × FIGURE 23-15 General procedure for producing transgenic mice. Fertilized eggs are collected from a pregnant female mouse. Cloned DNA (referred to as the transgene) is microinjected into one of the pronuclei of a fertilized egg. The eggs are then implanted into the oviduct of pseudopregnant foster mothers (obtained by mating nor- mal females with a sterile male). The transgene will be incorporated into the chromosomal DNA of about 10%–30% of the offspring and will be expressed in all of their somatic cells. If a tissue-specific pro- moter is linked to a transgene, then tissue-specific expression of the transgene will result. ■ Introduction of a mutant or disrupted gene into the cultured ES cells and selection of homologous recombinant cells in which the gene of interest has been knocked out (i.e., replaced by a nonfunctional form of the gene) ■ Injection of homologous recombinant ES cells into a recipient mouse blastocyst and surgical implantation of the blastocyst into a pseudo-pregnant mouse ■ Mating of chimeric offspring heterozygous for the disrupted gene to produce homozygous knockout mice The ES cells used in this procedure are obtained by culturing the inner cell mass of a mouse blastocyst on a feeder layer of fibroblasts or in the presence of leukemia-inhibitory factor. Under these conditions, the stem cells grow but remain pluri- potent and capable of later differentiating in a variety of di- rections, generating distinct cellular lineages (e.g., germ cells, myocardium, blood vessels, myoblasts, nerve cells). One of the advantages of ES cells is the ease with which they can be genetically manipulated. Cloned DNA containing a desired gene can be introduced into ES cells in culture by various transfection techniques. The introduced DNA will be in- serted by recombination into the chromosomal DNA of a small number of ES cells. The insertion constructs introduced into ES cells contain three genes: the target gene of interest and two selection genes, such as neo R ,which confers neomycin resistance, and the thymidine kinase gene from herpes simplex virus (tk HSV ), which confers sensitivity to gancyclovir, a cytotoxic nucleotide analog (Figure 23-16a). The construct often is engineered with the target-gene sequence disrupted by the neo R gene and with the tk HSV gene at one end, beyond the sequence of the target gene. Most constructs will insert at random by nonhomolo- gous recombination rather than by gene-targeted insertion through homologous recombination. As illustrated in Figure 23-16b, a two-step selection scheme is used to obtain those ES cells that have undergone homologous recombination, where- by the disrupted gene replaces the target gene. The ES cells obtained by this procedure are heterozygous for the knockout mutation in the target gene. These cells are clonally expanded in cell culture and then injected into a Experimental Systems CHAPTER 23 545 (b) Selection of ES cell carrying knockout gene Recombinants with random insertion Recombinants with gene-targeted insertion Nonrecombinant cells Treat with neomycin (nonrecombinant ES cells die) Treat with gancyclovir (nonhomologous ES recombinant cells die) Homologous ES recombinants with targeted disruption in gene X survive (a) Formation of recombinant ES cells Blastocyst Blastocyst ES cells Target gene insertion construct Introduce into cultured ES cells Homologous recombination Nonhomologous recombination ES cell DNA Gene-targeted insertion Random insertion neo R tk HSV neo R neo R tk HVS FIGURE 23-16 Formation and selection of mouse recombinant ES cells in which a particular target gene is disrupted. (a) In the engineered insertion construct, the target gene is disrupted with the neo R gene, and the thymidine kinase tk HSV gene is located outside the target gene. The construct is transfected into cultured ES cells. If homologous recombi- nation occurs, only the target gene and the neo R gene will be inserted into the chromosomal DNA of the ES cells. If nonhomologous recom- bination occurs, all three genes will be inserted. Recombination occurs in only about 1% of the cells, with nonhomologous recombination much more frequent than homologous recombination. (b) Selection with the neomycin-like drug G418 will kill any nonrecombinant ES cells because they lack the neo R gene. Selection with gancyclovir will kill the nonhomologous recombinants carrying the tk HSV gene, which confers sensitivity to gancyclovir. Only the homologous ES recombinants will survive this selection scheme. [Adapted from H. Lodish et al., 1995, Molecular Cell Biology, 3rd ed., Scientific American Books.] mouse blastocyst, which subsequently is implanted into a pseudo-pregnant female. The transgenic offspring that de- velop are chimeric, composed of cells derived from the geneti- cally altered ES cells and cells derived from normal cells of the host blastocyst. When the germ-line cells are derived from the genetically altered ES cells, the genetic alteration can be passed on to the offspring. If the recombinant ES cells are homozy- gous for black coat color (or other visible marker) and they are injected into a blastocyst homozygous for white coat color, then the chimeric progeny that carry the heterozygous knock- out mutation in their germ line can be easily identified (Figure 23-17). When these are mated with each other, some of the off- spring will be homozygous for the knockout mutation. “Knock-In” Technology Allows the Replacement of an Endogenous Gene In addition to deleting a gene of choice, it also is possible to replace the endogenous gene with a mutated form of that gene. As in the strategy for knocking out a gene, DNA con- structs that carry mutations in a particular gene can be ex- changed for the endogenous gene. It also is possible to re- place all of an endogenous gene with a DNA sequence of choice. In a recent report, for example, the CD4 gene was replaced with the one for H9252-galactosidase. In these experi- ments, the CD4 promoter was left intact to drive the expres- sion of H9252-galactosidase, which catalyzes the color change of certain reporter chemicals to blue. Because the CD4 pro- moter drove the expression of H9252-galactosidase, only those thymic cells destined to express CD4 turned blue in the pres- ence of the reporter chemicals. Data from these experiments were useful in tracing CD4/CD8 lineage commitment in developing T cells. Inducible Gene Targeting, the Cre/Lox System, Targets Gene Deletion In addition to the deletion of genes by gene targeting, recent experimental strategies have been developed that allow the specific deletion of a gene of interest in precisely the tissue of choice. These technologies rely on the use of site-specific recombinases from bacteria or yeast. The most commonly used recombinase is Cre, isolated from bacteriophage P1. Cre recognizes a specific 34-bp site in DNA known as loxP and, upon recognition, catalyzes a recombination. Therefore, DNA sequences that are flanked by loxP are recognized by Cre and 546 PART IV The Immune System in Health and Disease × Mate chimeric mice to homozygous white mice Surgically transfer embryo into pseudopregnant mouse Inject ES cells into blastocoel cavity of early embryo. ES cells are heterozygous for knockout mutation in gene X and homozygous for black coat color; embryo is homozygous for white coat color Chimeric progeny have black-and-white coats. White areas are derived from recipient blastocoel cells, black areas from ES cells Black progeny develop from germ-line cells derived from ES cells and are heterozygous for disrupted gene X FIGURE 23-17 General procedure for producing homozygous knockout mice. ES cells homozygous for a marker gene (e.g., black coat color) and heterozygous for a dis- rupted target gene (see Figure 23-18) are in- jected into an early embryo homozygous for an alternate marker (e.g., white coat color). The chimeric transgenic offspring, which have black-and-white coats, then are mated with homozygous white mice. The all-black progeny from this mating have ES-derived cells in their germ line, which are heterozy- gous for the disrupted target gene. Mating of these mice with each other produces ani- mals homozygous for the disrupted target gene, that is, knockout mice. [Adapted from M. R. Capecchi, 1989, Trends Genet. 5:70.] the recombinational event results in the deletion of the inter- vening DNA sequences. In other words, animals that ubiqui- tously express Cre recombinase will delete all loxP-flanked sequences. The real innovation of this technique is that ex- pression of the Cre recombinase gene can be controlled by the use of a tissue specific promoter. This allows tissue-specific expression of the recombinase protein and thus tissue-specific deletion of DNA flanked by loxP. For example, one could ex- press Cre in B cells using the immunoglobulin promoter, and this would result in the targeted deletion of loxP-flanked DNA sequences only in B cells. This technology is particularly useful when the targeted deletion of a particular gene is lethal. For example the DNA polymerase H9252 gene is required for embryonic development. In experiments designed to test the Cre/lox system, scientists flanked the mouse DNA polymerase H9252 gene with loxP and mated these mice with mice carrying a Cre transgene under the control of a T-cell promoter (Figure 23-18a). The results of this mating are offspring that express the Cre recombinase specifically in T cells. Using such mice, the scientists were able to examine the effects of deleting the enzyme DNA polymerase H9252 specifically in T cells. The effects of the deletion of this gene could not be examined in a conventional gene-targeting ex- periment, because deletion of DNA polymerase H9252 throughout the animal would be lethal. However, with the Cre/lox system, it now is possible to examine the effects of the deletion of this gene in a specific tissue of the immune system. The Cre/lox system also can be used to turn on gene ex- pression in a particular tissue. Just as the lack of a particular gene may be lethal during embryonic development, the ex- pression of a gene can be toxic. To examine tissue-specific expression of such a gene, it is possible to insert a translational stop sequence flanked by loxP into an intron at the beginning of the gene (Figure 23-18b). Using a tissue-specific promoter driving Cre expression, the stop sequence may be deleted in the tissue of choice and the expression of the potentially toxic gene examined in this tissue. These modifications of gene- targeting technology have been very useful in determining the effects of particular genes in cells and tissues of the immune system. Experimental Systems CHAPTER 23 547 Collect fertilized eggs Offspring Implant injected eggs into oviduct of pseudo-pregnant female Pseudo-pregnant female Test for presence of transgene Breed transgenics About 10–30% of offspring contain transgene × Inject cloned DNA into one of the pronuclei × FIGURE 23-18 Gene targeting with Cre/loxP (a) Conditional dele- tion by Cre recombinase. The targeted DNA polymerase H9252 gene is modified by flanking the gene with loxP sites (for simplicity, only one al- lele is shown). Mice are generated from ES cells by standard proce- dures. Mating of the loxP-modified-mice with a Cre transgenic will generate double transgenic mice in which the loxP-flanked DNA poly- merase H9252 gene will be deleted in the tissue where Cre is expressed. In this example, Cre is expressed in thymus tissue (striped) so that dele- tion of the loxP-flanked gene occurs only in the thymus (white) of the double transgenic. Other tissues and organs still express the loxP- flanked gene (orange). (b) Activation of gene expression using Cre/lox. A loxP-flanked translational STOP cassette is inserted between the pro- moter and the potentially toxic gene, and mice are generated from ES cells using standard procedures. These mice are mated to a transgenic line carrying the Cre gene driven by a tissue-specific promoter. In this example, Cre is expressed in the thymus, so that mating results in ex- pression of the toxic gene (blue) solely in the thymus. Using this strategy, it is possible to determine the effects of expression of the potentially toxic gene in a tissue-specific fashion. [Adapted from B. Sauer, 1998, Methods 14:381.] Microarrays—An Approach for Analyzing Patterns of Gene Expression In the past few years, a new approach has emerged designed to assess differences in gene expression between various cell types or the same cells treated in different fashions. This tech- nology, referred to as microarray technology or gene profiling, has the ability to rapidly and reliably scan large numbers of different mRNAs. The principle is simple and is derived from what we already know about RNA and DNA hybridization. mRNA is isolated from a given sample. Then, when cDNA synthesis is initiated the first strand of the cDNA is labeled with the tag. This forms the pool of target sequences. The next step is to hybridize the labeled cDNA to a microarray. There are many microarrays commercially avail- able, which fall mainly into two classes; those composed of cDNA, and those composed of oligonucleotides. Microarrays of cDNAs are, as the name suggests, a collection of cDNA that have been arranged, or arrayed, on a solid substrate in defined locations. The substrate varies but usually is a nylon membrane or a glass slide. If a very small amount of cDNA is used, the spots of cDNA arrayed on the substrate can be as small as 100–300 H9262m in size; it is relatively simple to array as many as 30,000 cDNAs on a single microscope slide (Figure 23-19a). The actual process of arraying the cDNA is usually ac- complished using robotics. The cDNAs are most frequently obtained from available cDNA libraries and, in some cases, are PCR products amplified from the cDNA library using primers specific for certain known genes. The oligonucleotide arrays are usually a collection of oli- gos 20–25 nucleotides long (Figure 23-19b). The advantage of this type of array is that one only needs sequences of genes of interest. No cDNA library is needed. However, the cost of assembling such an array is high, since the oligos have to be made and then spotted onto the filter or glass slide. Another problem with this approach is, depending upon the length of the oligo, there can be a degree of non-specific hybridization that hinders the final analysis of the data. This problem can be avoided by making longer oligos—which further increases the cost. For these reasons, oligo arrays are used most often by large pharmaceutical or biotechnology companies. Although the source of the targets used for both cDNA and oligo arrays are cDNA, the preparation of the target dif- fers depending upon the microarray. The target preparation for cDNA arrays involves labeling the cDNA with different fluorescent dyes such as Cy3 and Cy5 (Figure 28-19a). Cy3 and Cy5 are cyanine-based dyes that are easily conjugated to nucleic acids and are highly stable and emit less background fluorescence than conventional fluorescent dyes. Suppose you wish to compare two different cell types, or one cell type in two different states of activation. cDNA from one popula- tion is prepared using mRNA as a template. First strand syn- thesis of the mRNA is performed using one nucleotide con- jugated to Cy3. Then, using mRNA from the second cell pop- ulation, cDNA is prepared using a nucleotide conjugated to Cy5. These two populations of cDNA, one marked with Cy3 and the other with Cy5, are hybridized to the microarray. If one of the targets hybridizes to a cDNA on the array, a green (Cy3) or red (Cy5) fluorescence emission is detected. If both hybridize to the cDNA, yellow fluorescence is detected (the combination of the red and green emissions from both dyes). The arrays are analyzed by scanning the array at two different wavelengths to distinguish between the Cy3 and Cy5 signals. Once scanned at two wavelengths, the signals are compared and the signal intensity of each dye is determined and com- pared. The results are presented as a ratio between the two samples. In the case of oligo-based microarrays, the usual approach is to label the target cDNA with a biotin-labeled nucleotide during first-strand synthesis of the mRNA. The biotin- labeled cDNA is hybridized to the oligo array and detected by the use of the fluorescent strepavidin (Figure 28-19b). The procedure is then repeated with cDNA from the other cell type and another microarray is used. The resultant microar- rays are analyzed by either phosphoimaging or fluorescent- based scanning. This is most commonly accomplished using specialized scanners developed for scanning microarrays. The difference between this procedure and the cDNA- based array described above is that two microarrays are used. This is possible since the method for producing the oligo- based microarrays is more precise and it is possible to ensure that the same oligo will be present in precisely the same posi- tion on two separate microarrays. This is not possible with the technology used to prepare cDNA microarrays. There- fore, both targets must be hybridized to the same array to derive an accurate comparison. There is an advantage to us- ing two microarrays. Quantitation of expression levels is eas- ier when using one labeled target per microarray. When two targets are hybridized to the same array, it is always necessary to “subtract” the fluorescence of one target from the other before it is possible to obtain quantitative data. Since only one target is hybridized to a single oligo microarray, subtrac- tion is not necessary. The application of microarray technology to immunology is apparent. One could easily ask what is the difference between T cells and B cells. Or what is the difference between an acti- vated T cell and a resting T cell? The list of possible compar- isons is immense. To begin to answer some of the interesting immunology questions, Louis Staudt and co-workers at the NIH have developed an array they term “Lymphochip.” The Lymphochip is an array that consists of more than 10,000 human genes and is enriched in genes expressed in lymphoid cells. It also includes genes from normal as well as transformed lymphocytes. This particular microarray has provided a great deal of useful information, including a profile of T cells com- pared to B cells, plasma cells compared to germinal center B cells, and gene expression pattens induced by various signaling pathways. The Lymphochip and other clinical applications of microarrays are described in the Clinical Focus box. 548 PART IV The Immune System in Health and Disease SUMMARY ■ Inbred mouse strains allow immunologists to work rou- tinely with syngeneic, or genetically identical, animals. With these strains, aspects of the immune response can be studied uncomplicated by unknown variables that could be introduced by genetic differences between animals. ■ In adoptive-transfer experiments, lymphocytes are trans- ferred from one mouse to a syngeneic recipient mouse that has been exposed to a sublethal (or potentially lethal) dose of x-rays. The irradiation inactivates the immune cells of the recipient, so that one can study the response of only the transferred cells. ■ With in vitro cell-culture systems, populations of lympho- cytes can be studied under precisely defined conditions. Such systems include primary cultures of lymphoid cells, cloned lymphoid cell lines, and hybrid lymphoid cell lines. Unlike primary cultures, cell lines are immortal and homogeneous. Experimental Systems CHAPTER 23 549 Isolate mRNA Synthesize labeled cDNA targets, denature Spot onto substrate Spot onto substrate DNA microarray Amplify by PCR Hybridize to microarray Measure ratio of label array 1/ array 2 Measure ratio Cy5/Cy3 Hybridize to microarray Target analysis Microarray preparation Cy3 Cy5 Isolate mRNA cDNA synthesis Add label Biotin Array 1 Array 2 Selected cDNA Oligo microarray Synthesize oligos FPO FPO FPO (a) (b) FIGURE 23-19 DNA microarray analysis using cDNA microarrays (a) or high-density oligonucleotide microarrays (b). As described in the text, microarray analysis relies on the isolation of RNA from the tissues or cells to be analyzed, the conversion of RNA into cDNA, and the subsequent labeling of DNA during target preparation. The labeled target sequences are hybridized to either a cDNA microarray (a) or an oligo microarray (b). ■ Biochemical techniques provide tools for labeling impor- tant proteins of the immune system. Labeling antibodies with molecules such as biotin and avidin allows accurate determination of the level of antibody response. Gel elec- trophoresis is a convenient tool for separating and deter- mining the molecular weight of a protein. ■ The ability to identify, clone, and sequence immune-sys- tem genes using recombinant DNA techniques has revolu- tionized the study of all aspects of the immune response. Both cDNA, which is prepared by transcribing mRNA with reverse transcriptase, and genomic DNA can be cloned. Generally, cDNA is cloned using a plasmid vector; the re- 550 PART IV The Immune System in Health and Disease RNA from 38 samples of acute leukemia, labeled the RNA with biotin, and hybrid- ized the biotinylated RNA to commercial high-density microarrays that contained oligonucleotides corresponding to some 6817 human genes. Whenever the biotin- labeled RNA recognized a homologous oligonucleotide, hybridization occurred. Analysis revealed a group of 50 genes that were highly associated with either AML or ALL when compared with con- trol samples. These 50 genes were then used to sample nucleic acid from 34 in- dependent leukemias as well as samples from 24 presumed-normal human bone- marrow or blood samples. The result? A set of markers that clearly classified a tumor as ALL or AML. The results of the microarray analysis further suggested that the treatments for AML and ALL can be targeted more pre- cisely. For example, an AML expressing genes x, y, and z might respond to one treatment modality better than an AML that expresses a, b, and c. Several phar- maceutical companies have established research groups to evaluate different treatments for tumors based on the tu- mor’s microarray profile. This designer- approach to oncology is expected to pro- duce much more effective treatments of individual tumors, and ultimately, en- hanced survival rates. Microarray analysis is likely to be very useful in the diagnosis of tumors of the immune system. Most notably, a labora- tory at the National Institutes of Health (NIH) has developed a specialized DNA microarray containing more than 10,000 human cDNAs that are enriched for genes expressed in lymphocytes. Some of these cDNAs are from genes of known function, others are unknown cDNAs derived from normal or malignantly transformed lymphocyte cDNA libraries. This specialized array is called the “Lymphochip” because the lymphocyte cDNAs are arrayed on a silicon wafer. The group at NIH asked whether they could use the Lymphochip to divide the B-cell leukemia known as diffuse large B-cell lymphoma (DLBCL) into subgroups, an important question because this type of lymphoma has a highly variable clinical course, with some patients responding well to treatment while others respond poorly. Earlier attempts to define sub- groups within this group had been unsuccessful. A definition of subgroups within DLBCL could be useful in design- ing more effective treatments. Using the Lymphochip, the group at NCI identified two genotypically distinct subgroups of DLBCL. One group was comprised of tumors expressing genes characteristic of germinal-center B cells and was called “germinal-center–B-like DLBCL (see Fig- ure). The other group more resembled activated B cells and was termed “acti- vated B-like DLBCL.” Significantly, pa- tients with germinal-center–B-like DLBCL had a higher survival rate than those with activated B-like DLBCL. Normally all patients with DLBCL receive multi-agent chemotherapy. Patients who do not respond well to chemotherapy are then considered for bone-marrow transplanta- tion. The data obtained from this study suggests that patients with activated B-like DLBCL will not respond as well to chemotherapy and may be better served It is almost impos- sible to distinguish visually between B and T cells without molecular analysis. Similarly, it can be quite difficult to dis- tinguish one tumor from another. Two of the best-known acute leukemias are AML, which arises from a myeloid pre- cursor (hence the name, acute myeloid leukemia) and ALL (acute lymphoid leukemia), which arises from lymphoid precursors. Both leukemias are derived from hematopoietic stem cells, but the prognosis and treatment for the two dis- eases are quite different. Until recently, the two diseases could be diagnosed with some degree of confidence using a combination of surface phenotyping, karotypic analysis, and histochemical analysis, but no single test was conclu- sive; reliable diagnosis depended upon the expertise of the clinician. The difference between an ALL diag- nosis and an AML diagnosis can mean the difference between life and death. ALL responds best to corticosteroids and chemotherapeutics such as vincristine and methotrexate. AML is usually treated with daunorubicin and cytarabine. The cure rates are dramatically diminished if the less appropriate treatment is deliv- ered due to misdiagnosis. In 1999, a breakthrough in diagnosis of these two leukemias was achieved us- ing microarray technology. Todd Golub, Eric Lander, and their colleagues isolated CLINICAL FOCUS Microarray Analysis as a Diagnostic Tool for Human Diseases combinant DNA containing the gene to be cloned is prop- agated in E. coli cells. Genomic DNA can be cloned within a bacteriophage vector or a cosmid vector, both of which are propagated in E. coli. Even larger genomic DNA frag- ments can be cloned within bacteriophage P1 vectors, which can replicate in E. coli, or yeast artificial chromo- somes, which can replicate in yeast cells. Polymerase chain reaction (PCR) is a convenient tool for amplifying small quantities of DNA. ■ Transcription of genes is regulated by promoter and enhancer sequences; the activity of these sequences is Experimental Systems CHAPTER 23 551 vides us with a unique opportunity to examine differences between any dis- tinct populations of cells. One can com- pare which genes are expressed in com- mon or differentially in a na?ve T cell and a memory T cell. What is the difference between a normal T cell and a T cell dy- ing by apoptosis? Comparisons like these will be a rich source of insight into differences in cell populations. The key to using this valuable information will be the development of tools to analyze the vast quantities of data that can be obtained from this new approach. by bone-marrow transplantation shortly after diagnosis. As a direct result of this work, ongoing clinical trials are evaluat- ing how best to treat patients with acti- vated B-like DLBCL. Gene profiling is not restricted to diagnosis of cancer. This technology pro- Pr oba bility Genes DLBCL Biopsies Overall Survival (years) 042610812 GC B-like Activated B-like p = 0.01 1.0 0.5 0.0 GC B-Like DLBCL Activated B-like DLBCL (a) (b) Diffuse large B-cell lymphoma (DLBCL) is at least two distinct dis- eases. (a) Shown are differences in gene expression between sam- ples taken from patients with either germinal center B-like DLBCL (left, orange) or activated B-like DLBCL (right, blue). Relative expres- sion of the 100 genes (y axis) that discriminate most significantly be- tween the two DLBCL types is depicted over a 16-fold range using the graded color scale at bottom. Note the strikingly different gene expression profiles of the two diseases. (b) Plot of overall DLBCL pa- tient survival following chemotherapy. Gene expression profiles of tumor-biopsy samples allow the assignment of patients to the cor- rect prognostic categories and may aid in the treatment of this com- plex disease. [Adapted from L. M. Staudt, 2002. Gene expression pro- filing of lymphoid malignancies. Annu. Rev. Med. 53:303-318.] controlled by DNA-binding proteins. Footprinting and gel-shift analysis can be used to identify DNA-binding pro- teins and their binding sites within the promoter or enhancer sequence. Promoter activity can be assessed by the CAT assay. ■ Cloned genes can be transfected (transferred) into cul- tured cells by several methods. Commonly, immune-sys- tem genes are transfected into cells that do not normally express the gene of interest. Cloned genes also can be in- corporated into the germ-line cells of mouse embryos, yielding transgenic mice, which can transmit the incorpo- rated transgene to their offspring. Expression of a chosen gene can then be studied in a living animal. Knockout mice are transgenics in which a particular target gene has been replaced by a nonfunctional form of the gene, so the gene product is not expressed. The Cre/lox system provides a mechanism that allows tissue-specific expression or dele- tion of a particular gene. ■ Microarrays are a powerful approach for the examination of tissue-specific gene expression and comparison of gene expression in different cells. It has already begun to revolu- tionize the study of gene regulation and gene expression. References Alizadeh A. A., et al. 2000. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000 403:503-11. Bell, J. 1989. The polymerase chain reaction. Immunol. Today 10:351. Betz, U. A. K., et al. 1996. Bypass of lethality with mosaic mice generated by Cre-loxP-mediated recombination. Current Biol- ogy 6:1307. Camper, S. A. 1987. Research applications of transgenic mice. Biotechniques 5:638. Capecchi, M. R. 1989. Altering the genome by homologous recombination. Science 244:1288. Denis, K. A., and O. N. Witte. 1989. Long-term lymphoid cul- tures in the study of B cell differentiation. In Immunoglobulin Genes. Academic Press, p. 45. Depamphilis, M. L., et al. 1988. Microinjecting DNA into mouse ova to study DNA replication and gene expression and to pro- duce transgenic animals. Biotechniques 6(7):622. Koller, B. H., and O. Smithies. 1992. Altering genes in animals by gene targeting. Annu. Rev.Immunol. 10:705. McCune, J. M., et al. 1988. The SCID-Hu mouse; murine model for analysis of human hematolymphoid differentiation and function. Science 241:1632. Meinl, E., et al. 1995. Immortalization of human T cells by her- pesvirus saimiri. Immunol. Today 16:55. Melton, D. W. 1994. Gene targeting in the mouse. BioEssays 16:633. Sauer, B. 1998. Inducible gene targeting in mice using the Cre/lox system. Methods 14:381. Schlessinger, D. 1990. Yeast artificial chromosomes: tools for mapping and analysis of complex genomes. Trends Genet. 6(8):254. Sharpe, A. H. 1995. Analysis of lymphocyte costimulation in vivo using transgenic and knockout mice. Curr.Opin. Immunol. 7:389. Shaffer A. L., A. Rosenwald, E. M. Hurt, J. M. Giltnane, L. T. Lam, O. K. Pickeral, and L. M. Staudt. 2001. Signatures of the immune response. Immunity 15:375-85. Schulze A., and J. Downward. 2001. Navigating gene expression using microarrays—a technology review. Nat Cell Biol. 3:E190-5. USEFUL WEB SITES http://www.biomednet.com/db/mkmd Access to all known knockouts in mice, updated regularly. http://www.jax.org/ Home page for The Jackson Laboratory, the major repository of inbred mice in the world. http://www.neb.com/ Home page for New England Biolabs, a molecular biology company. Useful information concerning restriction enzymes is found at this site, under Technical Resources. http://www.public.iastate.edu/~pedro/research_tools.html A very useful site for molecular biology, containing links to many informative sites. Updated regularly. Study Questions CLINICAL FOCUS QUESTION How has microarray technology changed disease diagnosis and how is it likely to influence treat- ment of diseases in the future? 1. Explain why the following statements are false. a. The amino-acid sequence of a protein can be determined from the nucleotide sequence of a genomic clone encod- ing the protein. b. Transgenic mice can be prepared by microinjection of DNA into a somatic-cell nucleus. c. Primary lymphoid cultures can be propagated indefi- nitely and are useful in studies of specific subpopulations of lymphocytes. 2. Fill in the blanks in the following statements with the most appropriate terms: a. In inbred mouse strains, all or nearly all genetic loci are ;such strains are said to be . 552 PART IV The Immune System in Health and Disease b. SCID mice have a genetic defect that prevents develop- ment of functional and cells. c. B-cell hybridomas are formed by fusion of with .They are capable of growth and are used to produce . d. A normal lymphoid cell that undergoes can give rise to a cell line, which has an life span. 3. The gene diagrammed below contains one leader (L), three exons (E), and three introns (I). Illustrate the primary transcript, mRNA, and the protein product that could be generated from such a gene. 4. The term transfection refers to which of the following? a. Synthesis of mRNA from a DNA template b. Synthesis of protein based on an mRNA sequence c. Introduction of foreign DNA into a cell d. The process by which a normal cell becomes malignant e. Transfer of a signal from outside a cell to inside a cell 5. Which of the following are required to carry out the PCR? a. Short oligonucleotide primers b. Thermostable DNA polymerase c. Antibodies directed against the encoded protein d. A method for heating and cooling the reaction mixture periodically e. All of the above 6. Why is it necessary to include a selectable marker gene in transfection experiments? 7. What would be the result if a transgene were injected into one cell of a four-cell mouse zygote rather than into a fertilized mouse egg before it divides? 8. A circular plasmid was cleaved with EcoRI, producing a 5.4- kb band on a gel. A 5.4-kb band was also observed when the plas- mid was cleaved with HindIII. Cleaving the plasmid with both enzymes simultaneously resulted in a single band 2.7 kb in size. Draw a diagram of this plasmid showing the relative location of its restriction sites. Explain your reasoning. 9. DNA footprinting is a suitable technique for identifying which of the following? a. Particular mRNAs in a mixture b. Particular tRNAs in a mixture c. Introns within a gene d. Protein-binding sites within DNA e. Specific DNA sites at which restriction endonucleases cleave the nucleotide chain 10. Explain briefly how you might go about cloning a gene for interleukin 2 (IL-2). Assume that you have available a mono- clonal antibody specific for IL-2. 11. You have a sample of a mouse DNA-binding protein and of the mRNA that encodes it. Assuming you have a mouse genomic library available, briefly describe how you could select a clone carrying a DNA fragment that contains the gene that encodes the binding protein. 12. What are the major differences between transgenic mice and knockout mice and in the procedures for producing them? 13. How does a knock-in mouse differ from a knockout mouse? 14. How does the Cre/lox technology enhance knockout and knock-in strategies? 15. For each term related to recombinant DNA technology (a–i), select the most appropriate description (1–10) listed below. Each description may be used once, more than once, or not at all. Terms a. Yeast artificial chromosome b. Restriction endonuclease c. cDNA d. COS sites e. Retrovirus f. Plasmid g. cDNA library h. Sticky ends i. Genomic library Descriptions (1) Cleaves mRNA at specific sites. (2) Cleaves double-stranded DNA at specific sites. (3) Circular genetic element that can replicate in E. coli cells. (4) Used to clone DNA in mammalian cells. (5) Formed from action of reverse transcriptase. (6) Collection of DNA sequences within plasmid vectors representing all of the mRNA sequences derived from a cell. (7) Produced by action of certain DNA-cleaving enzymes. (8) Used to clone very large DNA sequences. (9) Used to introduce larger-than-normal DNA fragments in H9261-phage vectors. (10) Collection of H9261 clones that includes all the DNA sequences of a given species. Experimental Systems CHAPTER 23 553