Molecular Biology of the Cell - Volume 3 - Alberts B., Bray D., Lewis J., Raff M., Roberts K., Watson J. 1994

From Cells to Multicellular Organisms
The Immune System
T Lymphocytes and Cellular Immunity

The diverse responses of T Cells are collectively referred to as Cell-mediated immune responses. Much like antibody production, these responses play a vital role in vertebrate defense against infection, particularly against certain Viruses and Fungi. Similarly to antibody-mediated responses, they exhibit a high degree of antigen Specificity.

There are, however, several key features that distinguish T cells from B cells.

1) Some T cells directly combat infection by destroying virus-infected cells, whereas the majority of T cells regulate The activity of other effector cells, such as B cells and macrophages.

2) Both effector and regulatory T cells act primarily over short distances by interacting directly with the cells they kill or whose activity they regulate. B cells, by contrast, secrete Antibodies that disperse far from their site of production.

3) Likely for this reason, T cells bind foreign Antigens only when the antigens are presented On the surface of another cell of the host Organism; the antigen is recognized in association with a special Class of cell-surface Glycoproteins called MHC molecules. These molecules are named thus because they are encoded by a cluster of genes known as the Major Histocompatibility Complex (MHC) (see Section 18.6.5). Consequently, T-cell activation occurs only upon direct contact with another host cell; this is the most crucial distinction between the ways T AND B cells recognize antigens.

Fig. 18-46. The T-cell receptor heterodimer, consisting of a and ß polypeptide chains; both chains are glycosylated (not shown). Each chain is 280 Amino Acids long; the large extracellular portion of each chain is folded into two Ig-like domains—one variable (V) and one constant (C). Based on Amino Acid Sequence analyses (deduced from cloned cDNA sequences), the antigen-binding site formed by the V_a and Vß domains is believed to closely resemble an antibody molecule's antigen-binding site in overall size and geometry. Unlike antibodies, which possess two antigen-binding sites, however, T-cell receptors have only a single such site (presumably because they are permanently tethered to The Plasma Membrane, where they can act cooperatively). The a/ß heterodimer shown here is noncovalently associated with an invariant set of Membrane Proteins known as the CD3 complex (not shown).

A typical T cell bears between 20,000 and 40,000 a/ß protein molecules on its surface.

18.6.1. T-Cell Receptors Are Antibody-Like Heterodimers [32]

Because T cells are activated exclusively through close contact with other cells, their antigen receptors exist solely in a membrane-bound form. Consequently, unlike antibodies—which are secreted (though they can also be membrane-bound)—such receptors proved exceptionally difficult to isolate, and identifying these molecules and the genes encoding them took considerably longer. The first receptor proteins were isolated in 1983, following the successful cultivation of pure clones of antigen-specific T cells (Section 18.6.11), which made it possible to obtain large quantities of T cells bearing identical receptors. Monoclonal Antibodies against these cloned cells could then be produced, and those recognizing the T-cell receptor were identified by their ability to block antigen-induced responses in the original cells while leaving the responses of other T-cell clones unaffected. These antibodies were subsequently utilized to purify the receptor molecules; the receptor was shown to consist of two polypeptide chains, a and ß, linked by Disulfide Bonds. Each of these chains shares a distinctive feature with antibodies: the presence of a variable N-terminal region and a constant C-terminal region (Fig. 18-46).

Approximately a year later, these structural similarities to antibodies served as a cornerstone for an ingenious method used to isolate the genes encoding T-cell receptors. Because T and B lymphocytes are closely related, most of their Genes are transcribed in common, meaning they largely contain the same mRNAs; however, mRNAs specific for the T-cell receptor are found exclusively in T cells. Researchers isolated the total mRNA from T-cell clones, used subtractive Hybridization to remove mRNAs shared with B cells (Section 5.6.4), and thereby obtained a small population of mRNAs unique to T cells. A cDNA library was then constructed from this mRNA (Section 5.6.3), and individual cDNA clones from this library were used to generate radioactive DNA probes. Working on the hypothesis that the Variability of the N-termini in T-cell receptor proteins (like the V regions of antibodies) is generated by DNA rearrangements, researchers employed each DNA probe to analyze corresponding genomic DNA in an attempt to detect rearrangements occurring during T-cell development. Through this approach, the Gene pools encoding the a and ß chains were ultimately localized to different Chromosomes. It was demonstrated that these gene pools, much like antibody gene pools, comprise discrete V, D, J, and C segments that are assembled via Site-Specific Recombination during T-cell development in the Thymus.

With a single exception, all the mechanisms utilized by B cells to generate antibody diversity are also employed by T cells to generate T-cell receptor diversity. Nevertheless, mechanisms involving nucleotide deletion and insertion during the joining of gene segments are thought to play a particularly critical role in T cells. Conversely, T cells apparently do not utilize the antigen-driven somatic hypermutation mechanism seen in B cells (Section 18.4.4).

The reason for this likely lies in the fact that hypermutation could potentially give rise to T cells reacting against "self" molecules. Such a problem is far less critical for B cells, as the majority of self-reactive B cells cannot be activated without the assistance of self-reactive helper T cells (Section 18.6.12).

Recently, a second class of T-cell receptors was discovered—heterodimers consisting of y and δ chains. These receptors are expressed on cell subpopulations of unelucidated function within the thymus, epidermis, and intestinal epithelium.

Both a/ß and y/δ T-cell receptors are physically associated at The Cell surface with the same set of polypeptide chains, known as the T3 (or CD3) complex. This complex is present on The surface of all mature T cells and is believed to participate in transducing signals from the antigen-activated T-cell receptor into the cell interior.

18.6.2. Different T-Lymphocyte Responses Are Mediated by Distinct Classes of These Cells [33]

T lymphocytes destroy virus-infected cells and either promote or suppress the responses of other leukocytes. These three functions are carried out by distinct classes of T lymphocytes: cytotoxic T cells, helper (or inducer) T cells, and suppressor T cells, respectively. Cytotoxic T cells, alongside B cells, act as the principal effector Cells of the immune system; helper and suppressor T cells are collectively referred to as regulatory T cells.

Among the three Major Classes of T cells, suppressor T cells are the least understood. For instance, while it is known that helper and cytotoxic T cells utilize identical receptors for antigen recognition (a/ß heterodimers), The Nature of the receptors employed by suppressor T cells remains unclear (although at least some appear to utilize a/ß heterodimers as well). One reason for this limited understanding is the extreme difficulty in establishing clones of these cells in culture, whereas obtaining antigen-specific clones of cytotoxic T cells and helper T cells is comparatively straightforward (Section 18.6.11).

Although cytotoxic T cells and helper T cells utilize antigen receptors encoded by the same gene segment pools, they cannot recognize the same MHC molecules on cell surfaces. This difference reflects the distinct functional roles of these two cell types.

18.6.3. Cytotoxic T Lymphocytes Destroy Virus-Infected Cells [34]

Viruses replicate within cells, where they remain sheltered from antibody attack; therefore, the most effective way to prevent their spread to other cells is to destroy the infected cell before virus assembly begins. This constitutes The primary function of cytotoxic T cells. Given the high destructive potential of these lymphocytes, it is crucial that their attack be restricted exclusively to infected cells. Time-lapse cinematography has shown that a cytotoxic T cell can focus its attack on only one target cell at a time, even when bound to multiple target cells simultaneously. How do cytotoxic cells direct their attacks with such precision?

Fig. 18-47. A cytotoxic T cell (the smaller of the two) and a target cell it is destroying in culture. A and B are electron micrographs; C shows immunofluorescence staining for tubulin. Cytotoxic T cells were harvested from mice immunized with target cells consisting of foreign tumor cells. The images illustrate how the T cell and target cell bind to one another (A, C); image B shows the T cell after it has killed the target cell. Note that in the T cell, but not in the target cell, the centrosome and its radiating microtubules are oriented toward the site of cell-cell contact (A and B from D. Zagury et al., Eur. J. Immunol., 5, 818-822, 1975; C from V. Geiger et al., J. Cell Biol., 95, 137-143, 1982. Copyright permission of the Rockefeller Univ. Press.)

The directing mechanism apparently relies on cytoskeletal reorganization within the cytotoxic cell triggered by specific contact with the target cell surface. Labeling a cytotoxic T cell with anti-tubulin antibodies during its interaction with an engaged target reveals that its centrosome reorients toward the point of contact with the target cell (Fig. 18-47). Furthermore, labeling the cell with antibodies against talin—a protein implicated in linking cell-surface receptors to cortical Actin filaments (Section 11.2.8)—demonstrates that talin concentrates in the cytotoxic cell cortex at the contact site. Evidence indicates that the aggregation of T-cell receptors at the contact zone induces a localized, talin-dependent accumulation of actin filaments; subsequently, a microtubule-dependent mechanism orients the centrosome and associated Golgi apparatus toward the contact site, thereby directing the cell's lethal machinery against the target cell. A similar cytoskeletal polarization can be observed during the functional Interaction of a helper T cell with a cell it "helps."

18.6.4. How Do Cytotoxic T Cells Kill Their Targets? [35]

Cytotoxic cells protect us not only against common viral infections, but also against tumors of viral origin. However, such tumors probably account for less than 20% of all human malignancies. Furthermore, available data provide no evidence that immune responses protect us against most Other types of cancer: under immunosuppressive conditions, humans and experimental animals are more susceptible to viral tumors (as well as tumors of suspected viral Etiology), but this does not hold true for spontaneous and chemically induced tumors. Nevertheless, the potential role of non-immunological mechanisms in cancer defense is of great interest. Tumor cells could be destroyed by macrophages or natural killer (NK) cells. NK cells are lymphocyte-like cells, likely of the same Lineage as K cells, which kill antibody-coated Eukaryotic cells (Section 18.2.5). Unlike K cells, however, NK cells can also spontaneously and relatively non-specifically kill many tumor and virus-infected cells in culture in the absence of antibodies. How they distinguish abnormal cells from normal ones in these instances remains unknown.

We do not yet know how cytotoxic T cells and NK cells kill their targets. Some lines of NK cells and cytotoxic T cells that can be maintained indefinitely in culture use a mechanism for this that is probably similar to the action of The Complement System. Binding to targets leads to the release by these cells of pore-forming proteins called perforins, which polymerize in the plasma membrane of the target cell to form transmembrane channels. It is thought that these channels make the membrane permeable, thereby promoting cell death. Perforins, which are homologous to the complement component C9, are stored in secretory vesicles and released via local exocytosis at the site of contact with the target cell. The secretory vesicles also contain Serine esterases, although whether they play any role in target cell killing is unknown. In electron micrographs, perforin channels in target cell membranes look remarkably similar to the channels formed by component C9, supporting the idea of a shared MECHANISM OF ACTION between cytotoxic cells and complement. However, conventional cytotoxic T lymphocytes and NK cells can also kill target cells in the absence of perforins through some mechanism whose molecular basis is unknown. One possibility is that these cells activate an intrinsic self-destruction mechanism within the target cell, leading it to commit "suicide".

Which molecules NK cells recognize on the targets they destroy remains unknown. Meanwhile, it has been established that cytotoxic T cells recognize viral molecules associated with MHC glycoproteins on the surface of virus-infected cells. However, the crucial role of MHC molecules in antigen "presentation" to T cells has only recently been elucidated.

18.6.5. MHC Molecules Determine Graft Rejection [36]

MHC molecules were known long before their normal function was understood. They were initially identified as the major target antigens in transplant reactions. Transferring tissue from an adult donor to an individual of the same species (allotransplantation) or a different species (xenotransplantation) typically leads to its rejection. Skin-grafting experiments between different strains of mice conducted in the 1950s showed that graft rejection is driven by an Immune Response against foreign antigens present on the surface of the donor cells. It was later demonstrated that T cells participate primarily in these reactions and that they are directed against genetically "foreign" variants of cell-surface glycoproteins termed histocompatibility (i.e., tissue compatibility) molecules. Among these, the major histocompatibility complex (MHC)—a family of glycoproteins encoded by genes constituting the MHC—is of paramount importance. MHC molecules are present on the surface of all higher vertebrates. They were first discovered in mice and named H-2 (histocompatibility-2) antigens. In humans, they are designated as HLA (human-leucocyte-associated) because they were originally found on leukocytes.

Three remarkable properties of MHC molecules puzzled immunologists for many years. First, these molecules occupy a completely unique position among target antigens regarding their significance in T-cell transplantation reactions. Second, an unusually large fraction of T lymphocytes can recognize foreign MHC molecules: whereas less than 0.001% of the body's T cells respond to a typical viral antigen, more than 0.1% react to a single foreign MHC antigen. Third, many of the loci encoding MHC molecules are more polymorphic than any other loci in higher vertebrates. This means that within a given species, each locus is represented by an unusually large number of alleles (alternative forms of the same gene)—there may be more than 100, and each allele occurs in the population at a relatively high frequency. For this reason, and also because each individual has seven or more loci encoding MHC molecules (see below), it is exceedingly rare to find two organisms with an identical set of MHC glycoproteins. This makes matching organ Donors and recipients for human transplantation very difficult (except for genetically identical twins).

Yet vertebrates do not need protection against invasion by foreign cells from other vertebrates. Consequently, the observed T-cell "obsession" with foreign MHC molecules and the exceptional polymorphism of these molecules were not only obstacles to organ transplantation, but also a profound riddle for immunologists. This puzzle was solved only after it became clear that MHC molecules direct T lymphocytes toward those host cells that display foreign antigens on their surfaces, such as virus-infected cells. Later we will see how this discovery helped to largely resolve the MHC enigma (Section 18.6.8).

18.6.6. There are two Main Classes of MHC Molecules [37]

There are two principal classes of MHC molecules—class I and class II—each representing a set of cell-surface glycoproteins encoded by two linked groups of genes that together comprise the major histocompatibility complex (Fig. 18-48). Glycoproteins of both classes are heterodimers with a homologous overall Structure. Their N-terminal domains appear to be designed for antigen binding and presentation to T cells.

Fig. 18-48. Diagram of the H-2 and HLA gene complexes; the arrangement of loci encoding class I MHC glycoproteins (red regions) and class II (black regions) is shown. There are Three types of class I glycoproteins (H-2K, H-2D, and H-2L in the mouse; HLA-A, HLA-B, and HLA-C in humans); each glycoprotein consists of an a chain encoded by one of the loci shown here, and a ß₂-microglobulin chain encoded by a gene on a different chromosome. In the mouse, there are Two Types of class II MHC glycoproteins—H-2A and H-2E; each consists of an a chain and a ß chain. Only three types of human class II molecules are shown—HLA-DP, HLA-DQ, and HLA-DR (each consisting of an a and a ß chain), although at least one or two other types also exist. Human DP and DR molecules are homologous to mouse H-2E, and human DQ to mouse H-2A. All of these loci are highly polymorphic, with the exception of H-2Fa and its human homologs DPa and DRa, whose polymorphism is significantly lower. The gene complex also contains many other loci whose products resemble class I MHC molecules, though their functions remain unknown.

Fig. 18-49. MHC class I (A) and class II (B) glycoprotein molecules. The a chain of the class I molecule, approximately 345 amino acid residues long, possesses three extracellular domains, a₁, a₂, and a₃, encoded by separate exons. The molecule is non-covalently associated with a smaller polypeptide chain, ß₂-microglobulin (96 amino acids), which is not encoded within the MHC. The a₃ domain and ß₂-microglobulin are homologous to immunoglobulin domains. ß₂-Microglobulin is invariant, but the a chain is extremely polymorphic—primarily within the a₁ and a₂ domains. By engineering hybrid genes containing a mixture of a₁, a₂, and a₃ exons from different alleles in a single locus, followed by transfection of cultured fibroblasts with these genes, it was demonstrated that the antigenic determinants recognized by T cells are formed through the interaction of the a₁ and a₂ domains.

Within class II MHC molecules, both chains are polymorphic (ß more so than a), mainly due to the a₁ and ß₁ domains. The a₂ and ß₂ domains are homologous to immunoglobulin domains. Transfection experiments, analogous to those described above for class I MHC molecules, show that the antigenic determinants of class II molecules recognized by T cells are formed jointly by the a₁ and ß₁ domains.

Thus, class I and class II MHC glycoproteins share striking similarities in many respects. Both possess four extracellular domains, three of which contain intrachain disulfide bonds. The two domains closest to the membrane resemble immunoglobulin domains. The other two domains interact to form a complex three-dimensional surface; as we will see later, this surface presumably binds foreign antigen and presents it to T cells. All chains, except for ß₂-microglobulin, are glycosylated (not shown in the diagram).

Each class I MHC gene encodes a single transmembrane polypeptide chain (designated as a), the majority of which is folded into three extracellular globular domains (a₁, a₂, a₃). Each a chain is non-covalently associated with a small extracellular non-glycosylated protein, ß₂-microglobulin, which is not directly anchored to the membrane and is encoded by a separate gene located on a different chromosome (Fig. 18-49, A). Both ß₂-microglobulin and the membrane-proximal a₃ domain are homologous to a single immunoglobulin domain. The two N-terminal domains of the a chain, furthest from the membrane, contain polymorphic (variable) residues that are recognized by T cells during transplantation reactions.

Class II MHC molecules are also heterodimers featuring two conserved immunoglobulin-like domains near the membrane and two polymorphic (variable) N-terminal domains positioned further away from the membrane. However, in these molecules, both chains are encoded within the MHC and both span the membrane (Fig. 18-49, B). The presence of Ig-like domains in both class I and class II glycoproteins suggests that MHC molecules and antibodies share a common evolutionary history (Section 18.6.20).

Strong evidence indicates that the polymorphic regions of MHC molecules of both classes interact with foreign antigen, and that it is precisely the complex of the MHC molecule and foreign antigen that is recognized by the T-cell receptor. Before discussing this evidence, however, we will examine the distinct roles played by class I and class II molecules in directing cytotoxic T lymphocytes and helper T cells to their respective target cells.

The principal functional difference between class I and class II MHC molecules is reflected in their tissue distribution. Class I MHC molecules are expressed on virtually all nucleated cells, whereas the distribution of class II molecules is largely restricted to cells involved in immune responses.

Fig. 18-50. Cytotoxic T cells recognize foreign viral antigens in association with class I MHC glycoproteins on the surface of any host cell, whereas helper T cells recognize foreign antigens in association with class II MHC glycoproteins on the surface of an antigen-presenting cell. In transplantation reactions, helper cells also react against foreign class II glycoproteins, while cytotoxic cells react against foreign class I glycoproteins.

The reason for this distribution may be that class I molecules are recognized by cytotoxic T cells, which must be able to interact with any body cell that becomes virus-infected, whereas class II molecules are recognized by helper T cells, which interact primarily with other cells involved in immune responses, such as B cells and antigen-presenting cells (Fig. 18-50; see also Section 18.6.10). The key Properties of the two classes of MHC glycoproteins are summarized in Table 18-2.

18.6.7. Cytotoxic T Cells Recognize Foreign Antigens Associated with Class I MHC Molecules [38]

That MHC molecules present foreign antigens to T cells was first clearly demonstrated in 1974 through experiments with cytotoxic T cells. Strain X mice were infected with virus A. Seven days later, active cytotoxic T cells were found in the spleens of these mice; in cell culture, these T cells could kill virus-infected strain X fibroblasts within a few hours. As expected, they killed fibroblasts infected with virus A only, but not those infected with virus B; thus, the cytotoxic T cells were virus-specific. Unexpectedly, however, these same T cells failed to kill fibroblasts infected with the very same virus A if the fibroblasts were derived from strain Y mice (Fig. 18-51). This meant that cytotoxic T lymphocytes recognized not just the virus alone, but also some difference between the two types of fibroblasts. By utilizing special mouse strains (so-called congenic strains) that were either genetically identical except for class I MHC loci, or genetically different except for those same loci, researchers were able to show that infected target cells could be lysed only if they expressed at least one class I MHC molecule identical to that of the originally infected mouse. This demonstrated that class I MHC glycoproteins are required for the presentation of cell-surface-associated viral antigens to cytotoxic T cells. Because an individual's T cells recognize an antigen only when it is associated with that individual's own MHC molecules, this dual recognition is frequently referred to as MHC restriction. Only 10 years later, through a series of experiments using cytotoxic T cells responding to Influenza virus, was the Chemical Nature of the viral antigens recognized by these cells finally elucidated.

18.6.8. Cytotoxic T Cells Recognize Fragments of Viral Proteins on the Surface of Virus-Infected Cells [39]

It has been known since the 1960s that T cells, unlike B cells (and antibodies), typically do not recognize antigenic determinants within a protein's tertiary structure (see Fig. 18-23, A), but rather recognize determinants on an unfolded polypeptide chain. The reasons for this became clear with the accumulation of evidence that antigens recognized by T cells are usually cleaved within the host cell before their fragments are presented on its surface (initially, this was demonstrated for how T helper cells recognize antigen—see Section 18.6.10 below). The first direct evidence for this antigen-presentation mechanism to cytotoxic T cells came from the finding that certain cytotoxic T cells activated by influenza virus specifically recognize internal viral proteins that would not be accessible in an intact virion. Subsequently, data were obtained indicating that T cells recognize fragments of internal viral proteins. Viruses are intracellular parasites, and their proteins are synthesized via the expression of viral genes within an infected cell (Section 5.5). Therefore, it can be hypothesized that certain fragments of the resulting viral proteins "leak" to the surface of the infected cell and associate with MHC molecules either on the surface or somewhere within the cell (Fig. 18-52).

Fig. 18-51. A classic experiment demonstrating that a cytotoxic T cell recognizes not only a viral antigen but also some surface component of the virus-infected cell. By repeating this experiment with target cells that differed from the infected mouse cells only in restricted Regions of the genome, it was shown that the cell-surface component recognized by the cytotoxic T cell is an MHC class I glycoprotein. The killing capacity of active cytotoxic T cells is most conveniently assessed using the radioactive isotope ⁵¹Cr, which is taken up by living cells and released only upon their death. Therefore, the standard assay for cytotoxic T cell activity involves incubating them for several hours with target cells containing ⁵¹Cr, followed by measuring The amount of ⁵¹Cr released from the killed target cells.

This viewpoint is supported by two distinct lines of experimental evidence. First, if normal fibroblasts in culture are briefly exposed to fragments of an internal influenza virus protein (nucleoprotein, NP in Fig. 18-52), these cells will be recognized and killed by cytotoxic T cells originally activated by influenza virus-infected fibroblasts, provided that both sets of fibroblasts express the same MHC class I glycoproteins. Second, if normal fibroblasts are transfected with a DNA sequence encoding a fragment of the influenza virus nucleoprotein, the transfected cells will be killed by the same cytotoxic T cells as in the first experiment. These and other experiments suggest that fragments of viral proteins can reach the cell surface and bind there to MHC class I molecules.

Table 18-2. Properties of MHC Class I and Class II Molecules

Fig. 18-52. A cytotoxic T cell kills a virus-infected cell when it recognizes viral protein fragments associated with MHC class I molecules on the infected cell's surface. In the diagrammed case, peptide fragments are derived from the influenza virus nucleoprotein (NP); for simplicity, only this internal viral protein is shown. Only a tiny fraction of the viral proteins synthesized in the target cell is degraded. How this Cleavage occurs and how the resulting peptide fragments reach the cell surface remains unknown; it is also unclear where these fragments initially associate with MHC glycoproteins.

It is easy to envision how viral Protein Cleavage might occur in infected cells, given that almost all cellular proteins are continuously degraded (Section 8.2.4). It is harder to understand how fragments of the viral nucleoprotein reach the cell surface, because viral proteins are synthesized on cytoplasmic Ribosomes and normally do not have access to the lumen of The Endoplasmic reticulum, where proteins destined for the cell surface typically begin their journey (Section 8.1.4). However, T-cell recognition likely requires very small amounts of antigen. Therefore, the accidental emergence of even a tiny fraction of nucleoprotein fragments onto the cell surface could give rise to a target cell recognizable by a cytotoxic T lymphocyte.

18.6.9. X-ray crystallography reveals the antigen-binding site of the MHC class I glycoprotein [40]

Our understanding of how MHC molecules present antigens to T cells advanced dramatically in 1987, when the three-dimensional structure of a human MHC class I glycoprotein was solved by X-ray crystallography. As shown in Fig. 18-53, A, this protein possesses a single putative antigen-binding site located at one end of the molecule. It consists of a deep groove between two long α helices contributed by the α₁ and α₂ domains; the floor of the groove is formed by eight strands of ß Structure within the same domains. The dimensions of the groove are approximately 2.5 nm wide and 10 nm long. This is large enough to accommodate a peptide of 10–20 amino acid residues, depending on how tightly its chain is folded or bent. Interestingly, the groove in the crystallized protein was not empty: it contained a small molecule of unknown origin. This is presumed to be a peptide that co-purifies and co-crystallizes with the MHC glycoprotein (Fig. 18-53, B). These findings strongly implicate the groove as the antigen-binding site and indicate that once a peptide is bound to this site, it dissociates very slowly. This is further supported by the observation that fibroblasts briefly exposed to influenza virus nucleoprotein fragments remain targets for virus-specific cytotoxic T cells for at least three days.

Fig. 18-53. A. Structure of a human MHC class I glycoprotein based on X-Ray Diffraction Analysis of crystals of the extracellular portion of the molecule. The extracellular portion was separated from the transmembrane segment by proteolytic cleavage with the enzyme Papain. Both domains closest to the plasma membrane (α₃ and ß₂-microglobulin) resemble a typical immunoglobulin domain (see Fig. 10-28, B). The two domains farthest from the membrane (α₁ and α₂) are strikingly similar to each other and form a groove at the top of the molecule—the putative antigen-binding site. MHC class II molecules are believed to have a very similar structure. B. Top view of the putative antigen-binding groove containing a small molecule (likely a peptide) that co-purified with the MHC protein. It is this region of the MHC molecule that interacts with the T-cell receptor. (From P. J. Bjorkman et al., Nature, 329, 506–512, 1987.)

Polymorphic amino acid residues of the MHC glycoprotein (i.e., those residues that vary depending on the allelic form of these molecules) are localized largely either within the groove, where they can bind antigen, or along its edges, where they are accessible for recognition by T-cell receptors. The variability of MHC class I molecules is thought to be the result of Selection favoring their ability to bind and present a vast array of different viral Peptides. Nevertheless, it remains surprising that a small number of distinct antigen-binding sites per organism for MHC class I molecules (a maximum of six in humans) can bind such a diverse assortment of viral peptides specifically recognized by T cells. Even more enigmatic in this regard are MHC class II glycoproteins, whose three-dimensional structure appears very similar to that of class I molecules. Although an individual produces only about 10 to 20 kinds of class II molecules (each with its own antigen-binding site), these molecules apparently can bind an almost unlimited variety of foreign peptides and present them to T helper cells, which play a pivotal role in nearly all immune responses.

18.6.10. T helper cells recognize foreign antigen fragments in association with MHC class II glycoproteins on the surface of antigen-presenting cells [41]

T helper cells are required by most other lymphocyte types for an optimal response to antigen. The crucial role of T helper cells in Immunity is dramatically illustrated by the devastating Acquired Immunodeficiency Syndrome (AIDS) epidemic. The disease is caused by a retrovirus (HUMAN IMMUNODEFICIENCY VIRUS, HIV) that kills T helper cells and thereby cripples the immune system, rendering the patient susceptible to infections by microorganisms that rarely infect healthy individuals. As a result, most AIDS patients die within a few years of symptom onset.

Before T helper cells can help other lymphocytes respond to an antigen, they must first be activated themselves. This activation occurs when a T helper cell recognizes a foreign antigen associated with an MHC class II glycoprotein on the surface of a specialized antigen-presenting cell. Such cells are present in most Tissues. Derived from the Bone Marrow, they form a heterogeneous population that includes dendritic cells in Lymphoid Organs, Langerhans cells in the skin, and certain types of macrophages. All of these specialized antigen-presenting cells, together with B cells (which can also present antigen to T helper cells—see below) and thymic epithelial cells (Section 18.6.17), are the principal cell types that normally express MHC class II molecules (see Table 18-2).

Various experiments demonstrate The Central Role of MHC class II molecules in presenting foreign antigens to T helper cells. For example, Antibody Binding to these molecules blocks the cells' ability to present antigen. Conversely, fibroblasts, which normally do not produce MHC class II molecules and cannot present foreign antigens to T helper cells, can be converted into efficient antigen-presenting cells by transfection with the gene encoding an MHC class II molecule.

Much like viral antigens presented to cytotoxic T cells, antigens presented to T helper cells are typically fragments of degraded foreign proteins. These peptides are thought to associate with MHC class II molecules in much the same way as viral peptides associate with MHC class I molecules (see Fig. 18-53). However, unlike a virus-infected cytotoxic T-cell target, an antigen-presenting cell does not synthesize the foreign protein. Instead, foreign proteins appear to be taken up by endocytosis and partially cleaved in the acidic environment of endosomes or endolysosomes (Section 6.5.9), after which specific fragments are returned to the cell surface—a sequence of events collectively known as antigen Processing (Fig. 18-54). If endocytosis is blocked by mild fixation of the antigen-presenting cells (e.g., with formaldehyde) or if proteolysis in endolysosomes and Lysosomes is inhibited (e.g., with chloroquine), these cells can no longer process a foreign protein and present it to T helper cells. However, cells treated in this way are still able to present the protein if it is cleaved into small peptides (10–15 amino acids long) before being added to the cells.

Fig. 18-54. Hypothetical scheme of protein antigen "processing" and presentation by antigen-presenting cells. MHC glycoproteins have been shown to undergo recycling (i.e., they circulate) through the endosomal compartment; thus, they could initially associate with peptide fragments in the endolysosomal compartment and subsequently return to the cell surface with the bound peptide (not shown in the diagram).

A striking property of the antigen-presenting cell is its ability to process and present virtually any antigen to the appropriate T helper cell. This lack of antigenic specificity indicates that antigen-presenting cells presumably take up antigens via fluid-phase rather than receptor-mediated endocytosis (Section 6.5.7). If this is true, then the majority of internalized and degraded proteins will be "self" proteins, whose peptide fragments will occupy the binding sites of many MHC class II molecules. It is believed that for T helper cell activation, it is sufficient for foreign peptides to bind to only a small fraction of the MHC molecules.

18.6.11. T helper cells stimulate T-lymphocyte proliferation by secreting interleukin-2 [42]

T helper cell activation is a complex process involving various secreted proteins called interleukins, which act as local chemical mediators. Activation apparently begins when a T cell, in some yet unknown manner, prompts an antigen-presenting cell to secrete one or more interleukins. The best-characterized of these mediators is interleukin-1 (IL-1). However, the combined action of IL-1 (and possibly other interleukins) and antigen binding does not directly stimulate T helper cell proliferation. Instead, it prompts the T cell to stimulate its own proliferation—inducing the T cell to secrete a growth factor called interleukin-2 (IL-2) and to synthesize cell-surface receptors for it. It is the binding of IL-2 to these receptors that drives T-cell proliferation. In this manner, a T helper cell can continue to proliferate via an autocrine mechanism (Section 12.1.7) after it has detached from the surface of the antigen-presenting cell (Fig. 18-55). T helper cells can also help stimulate the proliferation of any other T cells, including cytotoxic T cells, in which IL-2 receptor expression has been previously induced. But because IL-2 receptor expression strictly depends on antigen stimulation, this results in the proliferation not of all T cells, but only of those that have already encountered antigen.

Fig. 18-55. Proposed sequence of events during antigen-mediated stimulation of T helper cells, triggering their proliferation.

The binding of a T cell to an antigen on the surface of an antigen-presenting cell prompts the T-cell receptor to trigger the Inositol phospholipid signaling pathway (Section 12.3.9) (Signal 1). Consequently, the T cell somehow stimulates the antigen-presenting cell (Signal 2). The antigen-presenting cell then secretes interleukins, notably interleukin-1 (IL-1), which promote T-cell activation (Signal 3). The activated T cell synthesizes receptors for interleukin-2 (IL-2) and secretes IL-2; the binding of IL-2 to its receptors (Signal 4) stimulates Cell Growth and Division. Once the antigen is cleared, The production of both IL-2 and IL-2 receptors eventually ceases, bringing cell proliferation to a halt.

Once the conditions necessary for T-cell proliferation were identified, it became possible to establish indefinitely proliferating, antigen-specific T-cell lines in culture by continuously supplementing the medium with IL-2 and periodically stimulating the cells with antigen to maintain IL-2 receptor expression. Single cells could then be isolated from these lines to generate T-cell clones. As we have already seen, such clones have played a pivotal role in T-cell research. For example, they made it possible to isolate T-cell receptors and their genes; furthermore, they have been widely used to study the mechanisms of T-cell activation and The Role of T helper cells in stimulating the responses of other lymphocytes.

18.6.12. T helper cells are required for most B lymphocytes to respond to an antigen [43]

T helper cells are essential for B cells to produce antibodies against the vast majority of antigens. This was first demonstrated in the mid-1960s through experiments in which irradiated mice were administered thymus or bone marrow cells along with an antigen. Mice that received exclusively bone marrow cells or exclusively thymus cells were unable to mount an antibody response; however, when a mixture of both cell types was administered, large quantities of antibodies were produced. It was later shown that the thymus supplies T cells, whereas the bone marrow provides B cells (Fig. 18-56). Using a specific chromosomal marker to distinguish between the introduced T and B cells, researchers confirmed that the antibody-secreting lymphocytes are indeed B cells. It was thus concluded that T cells presumably assist B cells in reacting to antigens.

However, certain antigens—including many microbial Polysaccharides—can stimulate the proliferation and maturation of B lymphocytes independently of T cells. Such T cell-independent antigens are typically high-molecular-weight polymers with repeating, identical antigenic determinants. Multivalent binding of these antigens to membrane-bound antibody molecules (the antigen receptors on B cells) can deliver a signal strong enough to directly activate B cells. Evidence suggests that the cells responding in this manner to multivalent antigens largely constitute a distinct B-cell subpopulation that has evolved to mount defenses against microbial polysaccharides without requiring T-cell help.

Fig. 18-56. The experiment that first demonstrated that an animal likely requires both T and B cells for antibody production. At the radiation doses used, both cell types are destroyed in the mouse.

18.6.13. T helper cells help activate B cells by secreting interleukins [44]

Upon being activated by a foreign antigen on the surface of a specialized antigen-presenting cell, a corresponding T helper cell can assist in activating a B cell by binding to the exact same foreign antigen on the B cell's surface. While an antigen-presenting cell captures and presents antigens nonspecifically (see above), a B cell typically presents only the antigen it specifically recognizes. The antigen is selected via attachment to specific membrane-bound antibodies (antigen receptors) on the B-cell surface; it is then internalized via receptor-mediated endocytosis (Section 6.5.7), degraded, and re-expressed on the cell surface as peptides bound to MHC class II glycoproteins for recognition by the T helper cell. Thus, the T helper cell recognizes the very same antigen-MHC complexes on the B cell it is helping as it did on the antigen-presenting cell that initially activated it.

Specific contact between a T helper cell and a B cell induces an internal reorganization within the helper cell's Cytoplasm, causing its centrosome and Golgi apparatus to reorient toward the B cell—much like what occurs in a cytotoxic T cell upon contacting a target cell (see above, Fig. 18-47). In this case, however, the reorientation appears to enable the T helper cell to direct the secretion of interleukins onto the B-cell surface (and perhaps also to concentrate membrane-bound signaling molecules there). These interleukins include IL-4, which initiates B-cell activation; IL-5, which stimulates the proliferation of activated B cells; and IL-6, which drives the maturation of activated B cells into antibody-secreting cells. Some of these and other interleukins can also induce class switching in B cells, shifting them from producing one class of antibodies to another (Section 18.4.7). Some of the signals likely involved in the initial Activation of a B cell are illustrated in Fig. 18-57.

Fig. 18-57. At least three types of signals are involved in the Cytology/cytology/16.html">Early stages of B-cell activation. The relative importance of these signals remains unclear and may vary depending on the specific type of B cell and antigen. Signal 1 is generated upon antigen binding; its transmission appears to involve the inositol phospholipid signaling pathway (Section 12.3.9). This signal activates the B cell and may induce the expression of receptors for certain interleukins produced by T helper cells. Subsequently, the B cell internalizes and degrades the antigen (not shown in the diagram) and presents small antigen fragments in association with MHC class II molecules to a T helper cell. It remains unclear whether the T-cell binding delivers a signal to the B cell (depicted here as Signal 2) or simply serves to direct the secretion of interleukin-4 (IL-4) and other interleukins (not shown) toward the B-cell surface (Signal 3). In addition to activating the B cell, Signal 3 also stimulates the cell to produce more MHC class II glycoproteins, thereby enhancing the B cell's capacity to receive T-cell help. Once the B cell is activated, other interleukins produced by T helper cells (specifically IL-5, IL-6, and y-interferon) help drive B-cell proliferation and maturation into antibody-secreting cells (not shown).

How are signals transmitted from activated cell-surface receptors to the cell interior during the antigen- or interleukin-mediated stimulation of B or T cells? For interleukin receptors, the answer is not yet known. However, compelling evidence indicates that antigen receptors on both B AND T cells deliver signals to the cell by activating the inositol phospholipid pathway discussed in Chapter 12 (Section 12.3.9).

18.6.14. Certain T helper cells activate macrophages by secreting interferon [45]

The assistance provided by T helper cells is not restricted to lymphocytes. Those T helper cells that secrete IL-2 upon antigen stimulation also release other interleukins, such as y-interferon, which recruits and activates macrophages, enhancing their ability to phagocytose and destroy invading microorganisms. The capacity of T cells to recruit and activate macrophages is particularly critical for defense against certain pathogens capable of surviving inside unactivated macrophages. The CAUSATIVE AGENT OF tuberculosis is a prime example.

The antigen-triggered secretion of y-interferon and other macrophage-activating interleukins by T helper cells forms The basis of the standard tuberculin skin test. When tuberculin (an extract of tubercle bacilli) is injected into the skin of an individual who has been immunized against or previously infected with tuberculosis, a characteristic immune response develops. This response is triggered at the injection site by the secretion of interleukins from memory T helper cells reacting to the tuberculin. Recruited by these interleukins, macrophages and lymphocytes accumulate at the site, producing the characteristic Swelling indicative of a positive tuberculin reaction.

Another key function of y-interferon is inducing the expression of MHC class II glycoproteins on the surface of certain cells (such as endothelial cells) that do not normally express them. This equips such cells to present antigens to T helper cells. Through this mechanism, T helper cells can recruit additional antigen-presenting cells as needed.

Evidence suggests that there are at least two distinct subclasses of T helper cells. One subclass appears to assist primarily B cells and secretes IL-4 and IL-5, whereas the other assists other T cells and macrophages and releases IL-2 and y-interferon. Information regarding some of the interleukins secreted by T helper cells (or antigen-presenting cells) is summarized in Table 18-3.

18.6.15. Cell-adhesion proteins stabilize interactions between T cells and their targets [46]

The specific binding of antigen-MHC complexes on a target cell surface to a/ß antigen receptors on a T cell is frequently not strong enough on its own to drive a functional interaction between the two cells. Various cell-adhesion proteins (see Section 14.3) on T cells help stabilize these interactions by increasing the overall binding affinity between the cells. Earlier (Section 14.2.17), we discussed the role of lymphocyte function-associated antigen-1 (LFA-1) in helping T and B cells (as well as other leukocytes) adhere to other cells and to the Extracellular matrix. T cells also express the cell-surface protein CD2, which assists them in attaching to appropriate target cells by binding to a complementary glycoprotein on the target cell surface known as LFA-3.

Table 18-3. Properties of Selected Interleukins

* Interleukins are secreted Peptides and Proteins that mediate local interactions among leukocytes; however, they do not bind antigen. Interleukins released by lymphocytes are also referred to as lymphokines. The Amino acid sequences of all the proteins listed here are known. The sources, target cells, and modes of action shown in the table pertain primarily to the immune system; because most interleukins have many other sources, targets, and modes of action, a more accurate term for them is cytokines.

Among the best-characterized T-Cell Adhesion proteins are the CD4 and CD8 glycoproteins, which are expressed on the surface of T helper cells and cytotoxic T cells, respectively. Both glycoproteins possess extracellular domains homologous to immunoglobulin domains and are thought to bind to invariant regions of MHC molecules: CD4 binds to MHC class II glycoproteins, whereas CD8 binds to class I glycoproteins (Fig. 18-58).

Some of the accessory glycoproteins found on the surface of T lymphocytes are listed in Table 18-4.

18.6.16. T suppressor cells predominantly suppress T helper function [47]

Several years after the discovery that T lymphocytes assist B cells in antibody production, it became apparent that T cells can also suppress the response of B cells or other T cells to antigens. Such suppression was first demonstrated in mice: they could be rendered unresponsive (tolerant) to sheep red Blood Cells through repeated injections of large quantities of these cells. When T cells from tolerant mice were transferred to normal mice, the latter likewise became specifically tolerant to sheep red blood cell antigens. This indicates that the tolerant state in this case is mediated by the suppression of the immune response by T cells. Subsequent experiments utilizing surface antigenic markers revealed that the cells responsible for this effect represent a specialized class of T lymphocytes—the so-called T suppressor cells. However, as we will see later, not all forms of immunological tolerance are mediated by T suppressors.

Fig. 18-58. The role of two accessory receptor proteins located on the T cell surface. The CD8 glycoprotein on cytotoxic T cells appears to bind to class I MHC molecules, whereas the CD4 glycoprotein on T helpers binds to class II MHC molecules. It is believed that in both cases, binding occurs with the invariant regions of the MHC molecules. These cell-adhesion proteins help stabilize the binding of T CELL RECEPTORS to antigen-MHC complexes on the target cell surface, especially when the binding is weak. Under such conditions, antibodies directed against the CD8 and CD4 accessory receptor proteins inhibit T cell activation. Antibodies against CD4 and CD8 are widely used to distinguish between T helpers and cytotoxic T cells, respectively. The AIDS virus (HIV) infects T helpers by initially binding to CD4 molecules on the surface of these cells.

Table 18-4. Accessory glycoproteins on the surface of T cells

Fig. 18-59. Interaction between a T helper (Th) and a T suppressor (Ts) cell, in which the receptor of one cell recognizes an idiotope (an antigenic determinant associated with the antigen-binding site) on the receptor of the other cell. Alternatively, the suppressor cell may recognize an idiotope of a helper cell receptor fragment presented on the helper cell surface in association with an MHC molecule (not shown). In either case, the suppressor cell inhibits helper cell function; The Mechanism of this inhibition remains unknown.

T helpers and T suppressors appear to act in concert to control the activity of B cells and cytotoxic T cells—the primary effector cells of the immune system. T helpers act directly on these effector cells, whereas T suppressors are thought to act indirectly by dampening the function of T helpers upon which the effector cells depend, although the precise mechanism of this suppression is unclear. How do T suppressors recognize the T helpers they suppress? Given what we know about the antigen-recognition mechanisms of helper cells (Section 18.6.10), it is unlikely that sufficient foreign antigen (or antigen fragments) would be present on the T helper surface for recognition by T suppressors. Instead, it is probable that T suppressors often interact with T helpers by recognizing antigenic determinants associated with the antigen-binding sites of the T helper receptor—the so-called idiotopes (Section 18.4.8), as illustrated in Fig. 18-59.

The discovery of T suppressors raised the question of whether they participate in natural tolerance by suppressing self-reactive lymphocytes. Current evidence remains somewhat contradictory, yet it suggests that natural tolerance is primarily driven by the elimination of such lymphocytes (so-called clonal deletion) and does not rely on T suppressors. Since most B cells require T helpers to mount an antibody response, eliminating self-reactive T helpers is, in principle, sufficient to prevent B cell responses to self-macromolecules. Indeed, this strategy is utilized for many self-antigens. For instance, normal mice do not produce antibodies against their own complement component C5. However, B cell production of such antibodies can be induced if normal mice are infused with T helpers from mutant mice lacking C5 (otherwise genetically identical to normal mice). Thus, the sole reason normal mice fail to generate antibodies against this common serum protein is that they lack or have inactivated C5-recognizing T helpers.

Nevertheless, this mechanism cannot account for all self-antigens. For example, self-macromolecules capable of activating B cells independently of T cell help appear to eliminate the B cells that recognize them; self-macromolecules present at high concentrations presumably act similarly. Likewise, cytotoxic T cells capable of reacting against normal molecules on self cell surfaces must be eliminated, as cytotoxic T lymphocytes can be activated to some extent by antigen without T help (although T helpers dramatically enhance their response). It appears that T suppressors play a largely supportive role in natural tolerance, being recruited only when the primary mechanism of clonal deletion fails.

18.6.17. Developing T cells that strongly react with self MHC molecules are eliminated in the thymus [48]

As we have seen, the initial evidence that MHC glycoproteins participate in T CELL ANTIGEN recognition came from experiments showing that T cells can respond to an antigen associated with self MHC molecules, but not foreign MHC molecules—a phenomenon known as MHC restriction (Section 18.6.7). Shortly thereafter, thymus transplantation experiments demonstrated that T cells, during their maturation in the thymus, apparently "learn to see" antigens in the context of self rather than foreign MHC molecules. One such experiment was performed as follows: strain X mice were irradiated to destroy all endogenous T cells and reconstituted with fresh bone marrow as a source of new T cells. If a thymus from a strain Y mouse is then transplanted into such a mouse, strain X T cells will develop within it. It was found that in most cases, the resulting mature strain X T cells recognize foreign antigens in association with strain Y MHC glycoproteins, rather than those of strain X. The simplest interpretation of these results is that developing T cells possessing receptors capable of recognizing antigens in association with MHC molecules expressed in the thymus are selectively rescued and induced to proliferate. This hypothesis implies that such positive selection shapes cytotoxic cells to recognize class I MHC molecules and helper cells to recognize class II molecules. This is further supported by the observation that antibodies against class II MHC molecules specifically block T helper development, whereas antibodies against class I molecules block cytotoxic T cell development.

This interpretation, however, is not entirely satisfactory because it does not explain how selection occurs in the absence of the foreign antigens that the T cells will later recognize. One possibility is that survival and maturation require T cells to make weak contacts with self MHC molecules; consequently, T cells are selected for weak recognition of self MHC—an interaction insufficient by itself to activate mature T cells. Activation would occur only when the combination of a foreign antigen and a self MHC molecule generates a structure that binds tightly to the T cell receptor.

Evidence supporting positive selection for weak self-MHC recognition in the thymus is less definitive than the evidence for negative selection, which eliminates cells that bind too strongly to self MHC molecules or to self MHC complexed with other self molecules. The most compelling proof that strongly self-reactive T cells are purged in the thymus comes from genetic studies in mice. One such study stemmed from a serendipitous observation: a particular V segment encoding the variable region of the T cell receptor β chain confers upon any T cell expressing it The ability to "strongly" recognize specific class II MHC molecules (designated H-2E), regardless of the D and J regions of the β chain or the V region of the α chain. (This suggests that T cell receptor V segments may have been evolutionarily selected to encode MHC-binding receptors.) However, not all mouse strains express H-2E. In strains lacking H-2E, T cells bearing this specific β-chain V segment are readily detected among both immature and mature thymic lymphocytes, whereas in H-2E-expressing strains, they are found exclusively within the immature lymphocyte population. Evidently, T cells of this type are eliminated prior to maturation in the thymus.

Thymus transplantation experiments suggest that positive selection (which underlies MHC restriction) and negative selection (which establishes self-tolerance) are two distinct processes. Positive selection likely occurs on the surface of thymic epithelial cells, whereas negative selection probably takes place on the surface of bone marrow-derived cells migrating into the thymus. Both cell types express both class I and class II MHC molecules on their surfaces.

The molecular mechanisms governing T cell selection in the thymus remain unknown. Positive selection may operate via growth or survival signals delivered by thymic epithelium to weakly interacting T cells. In negative selection, bone marrow-derived cells may function as "rogue" antigen-presenting cells that kill rather than activate any T cell that recognizes them. Cells exhibiting such properties, termed veto cells, have been identified in vitro.

A striking feature of T cell development in the thymus is that over 95% of the cells die without ever leaving the organ. Such massive attrition is presumably the price of stringent selection during T cell maturation.

18.6.18. Certain MHC allelic variants fail to effectively present specific antigens to T cells, a trait governed by immune response (Ir) genes [49]

Unlike class I MHC genes, which were initially discovered through their influence on graft rejection, class II MHC genes were uncovered through their role in T cell immune responses to specific soluble antigens. When animals were immunized with a simple antigen, some mounted robust T cell responses while others failed to respond at all. Genetic analyses revealed that the ability to respond to a given antigen is governed by a single immune response (Ir) gene, with responses to different antigens often controlled by distinct Ir genes. The first Ir genes mapped were those controlling T helper responses to antigens, which localized to the class II MHC loci. Ir genes regulating cytotoxic cell responses were subsequently mapped to various class I MHC regions.

These observations remained highly enigmatic until the critical role of MHC glycoproteins in antigen presentation to T cells was elucidated. We can now explain them simply by postulating that individuals genetically "non-responder" to a given simple antigen (typically possessing a single antigenic determinant) lack an MHC molecule capable of binding that determinant and effectively presenting it to the cognate T cell. This hypothesis received strong support from in vitro studies demonstrating that purified class II MHC molecules from a "responder" animal bind the relevant antigenic peptide, whereas those from a "non-responder" do not. Further research established that class II molecules possess a single antigen-binding site (similar to class I MHC molecules—see Fig. 18-57) capable of accommodating a remarkably diverse array of Peptides with a moderate affinity constant ($K_a$) of approximately $10^6\text{ L/mol}$ ($\Delta G = -8.5\text{ kcal/mol}$, roughly equivalent to the formation energy of eight Hydrogen Bonds; see Section 3.1.1). The association rate is slow (roughly $10^5$-fold lower than a typical antibody-antigen reaction), and once bound, the peptide dissociates with a half-life exceeding a day. This sluggish dissociation rate may require a slow conformational shift in the MHC molecule.

In some cases of genetic unresponsiveness to specific antigens, a different mechanism appears to operate. Certain combinations of self MHC molecules and foreign peptides may closely mimic other self MHC molecules. Because T cells reacting against such combinations would be eliminated during thymic development via negative selection (Section 18.6.17), the animal will remain genetically incapable of mounting a response to these foreign peptides.

18.6.19. The joint-recognition hypothesis of MHC restriction accounts for alloreactivity and MHC polymorphism

The joint-recognition hypothesis of MHC restriction explains why such a vast fraction of T cells respond to foreign MHC molecules and consequently drive transplant rejection. T cells react so vigorously against foreign MHC glycoproteins because these molecules—either directly or complexed with other molecules on the foreign cell surface—mimic various combinations of self MHC molecules and foreign peptides. For example, certain T cell clones specific for a viral antigen presented by a self class I MHC molecule can cross-react with a foreign class I MHC molecule in the absence of the viral antigen.

The MHC restriction (joint recognition) hypothesis also helps account for the extraordinary polymorphism of MHC molecules. In the evolutionary arms race between pathogens and the vertebrate immune system, microbes tend to alter their antigens to evade association with MHC molecules. If a mutation proves effective in this regard, the new variant may spread widely, triggering an epidemic or epizootic. Under such circumstances, the few host individuals possessing a novel MHC molecule capable of binding the altered microbial antigen will gain a strong selective advantage. Moreover, individuals carrying two different alleles for each MHC locus (i.e., heterozygotes) stand a better chance of surviving the infection than those with identical alleles at a given MHC locus. Consequently, selection operates to drive and maintain a high diversity of MHC molecules within a population.

Although the MHC restriction hypothesis provided plausible Answers to many questions originally raised by organ transplantation experiments, it gave rise to a new puzzle: how can the relatively small number of distinct MHC molecules expressed by a given animal (fewer than two dozen) bind a sufficiently diverse repertoire of peptides to mount a T-cell response against virtually any protein antigen? Interactions between antigens, antibodies, and class I MHC glycoproteins have been elucidated through X-ray crystallographic studies of these molecules. Such structural investigations must now be extended to the interaction between the MHC-antigen complex and the T-cell receptor. Recombinant DNA technology should soon yield sufficient quantities of soluble T-cell receptors to make such projects feasible. Indeed, recombinant DNA studies have already demonstrated that all of these proteins—MHC molecules, T-cell receptors, and antibodies—share a long, common evolutionary history.

18.6.20. Molecules involved in immune recognition belong to an ancient "superfamily" [50]

Most glycoproteins mediating cell-cell recognition or antigen recognition in the immune system share related Structural motifs, suggesting that their encoding genes stem from a common evolutionary origin. This Ig superfamily includes antibodies, T-cell receptors, MHC glycoproteins, the cell adhesion proteins CD2, CD4, and CD8, several polypeptide chains of the T-cell receptor-associated CD3 complex, and various Fc receptors on lymphocytes and other leukocytes. All these proteins contain one or more Ig-like domains (homologous Ig units). The peptide chain of each such domain typically spans about 100 Amino Acids and is thought to fold into a characteristic sandwich-like structure composed of two antiparallel ß-sheets, usually stabilized by a conserved disulfide bond (Section 18.3.4). Many of these molecules exist as dimers or larger oligomers in which homologous Ig units from different chains interact with one another (Fig. 18-60).

Each Ig unit is normally encoded by a separate exon. It is likely that the entire gene superfamily evolved from an ancestral gene encoding a single homologous Ig unit, similar to the genes encoding Thy-1 or ß2-microglobulin (see Fig. 18-53). These proteins may have originally functioned in cell-cell interactions. Thy-1-like molecules have been isolated from squid Brain, making it probable that the corresponding precursor gene emerged before vertebrates split from their invertebrate ancestors roughly 400 million years ago. New family members apparently arose through exon and gene duplications; analogous duplications likely gave rise to the multiple gene segments encoding antibodies and T-cell receptors.

Its striking recognition capabilities make the immune system nearly unique among cellular systems, surpassed in complexity only by The Nervous system. Both systems comprise vast numbers of phenotypically diverse cells organized into intricate networks. Within such networks, individual cells engage in both positive and negative interactions, with a given cellular response propagating through the system to influence many other cells. Unlike the neuronal network, which remains relatively fixed in spatial arrangement, the cells forming the immunological network are in constant motion and interact only transiently. In the following chapter, we will examine the cells of the vertebrate nervous system, which stands out among all cellular systems for its intricate and sophisticated Organization.

Fig. 18-60. Several membrane proteins belonging to the immunoglobulin superfamily. Homologous immunoglobulin and immunoglobulin-like domains are shaded; note that the ends of each loop forming such a domain are joined by disulfide bonds. Most domains interact with homologous domains on an associated polypeptide chain. In addition, certain domains of the polymeric immunoglobulin Fc receptor polypeptide chain interact with one another (not shown); this receptor binds both dimeric (see Section 18.2.5) and polymeric IgM, hence its name. The Thy-1 glycoprotein is covalently attached to a membrane-bound glycosylated phospholipid molecule (see Section 8.6.13). Not depicted in the figure are the helper T-cell and cytotoxic T-cell coreceptors CD4 and CD8, respectively, the CD3 protein complex associated with T-cell receptors, and CD2 (see Table 18-4). All these proteins likewise contain immunoglobulin-like domains. In addition to the proteins listed above, the immunoglobulin superfamily encompasses cell surface molecules involved in non-immune cell-cell interactions, such as neural cell adhesion molecules (N-CAM, see Section 14.3.6).

Summary

There are at least three functionally distinct subclasses of T lymphocytes: (1) cytotoxic T cells, which directly destroy virus-infected cells; (2) helper T cells, which secrete a battery of local chemical mediators (interleukins) that assist B cells in mounting humoral immune responses (antibody production), stimulate the proliferation of activated T cells, and activate macrophages; and (3) suppressor T cells, which appear primarily to downregulate helper T-cell responses. Helper and suppressor T cells serve as the principal regulators of immune responses.

The T-cell receptor is an antibody-like heterodimer encoded by genes assembled from multiple gene segments during T-cell development in the thymus. T-cell activation occurs when these receptors bind foreign antigen fragments presented in association with MHC glycoproteins on the surface of other host cells. This MHC restriction mechanism ensures that T cells recognize a foreign antigen only when it is displayed by an appropriate target cell. There are two major classes of MHC molecules: (1) class I molecules, found on the surface of nearly all nucleated somatic cells, which present fragments of viral proteins to cytotoxic T cells; and (2) class II molecules, found on the surface of B cells and specialized antigen-presenting cells, which present foreign antigen fragments to helper T cells. The inability of certain allelic variants of class I and class II MHC molecules to present specific antigenic determinants to T cells helps explain why these molecules are so highly polymorphic.

References

General

Golub Е. S. Immunology: A Synthesis. Sunderland, MA; Sinauer, 1987.

Hood L. E., Weissman I. L., Wood W. В., Wilson J. H. Immunology, 2nd ed. Menlo Park CA; Benjamin-Cummings, 1984.

Male D., Champion W., Cooke A. Advanced Immunology. London. Gower, 1987. Paul WE., ed. Fundamental Immunology. New York. Raven, 1984. (Russian Translation available: Paul W., ed. Immunology, in three volumes. Moscow: Mir, 1987-1989.) Roitt I. M., Brostoff J., Male D.K. Immunology. London; Gower, 1985.

Cited

1. Gowans J.L., McGregor D.D. The immunological activities of lymphocytes. Prog. Allergy, 9, 1-78, 1965.

2. Greaves M. F., Owen J. J. Т., Raff M. С. Т and В Lymphocytes: Origins, Properties and Roles in Immune Responses. Amsterdam, Excerpta Medica, 1973.

3. Cooper M., Lawton A. The Development of the immune system. Sci. Am., 231(5) 59-72, 1974.

Owen J. J. T. Ontogenesis of lymphocytes. In: В and Т Cells in Immune Recognition (F.Loor, G.E. Roelants, eds.), pp. 21-34. New York, Wiley, 1977.

4. Möller G., ed. Functional Т Cell Subsets Defined by Monoclonal Antibodies. Immunol. Rev., Vol. 74, 1983.

Raff M.C. Cell-surface immunology. Sci. Am., 234(5), 30-39, 1976. Reinherz E.L., Schlossman S. F. The differentiation and function of human lymphocytes. Cell, 19, 821-827, 1980.

5. Ada G. L. Antigen binding cells in tolerance and immunity. Transplant. Rev., 5, 105-129, 1970.

Ada G.L., Nossal G. The clonal selection theory. Sci. Am., 257(2), 62-69, 1987.

Burnet F. M. The Clonal Selection Theory of Acquired Immunity. Nashville, TN, Vanderbilt University Press, 1959.

Wigzell H. Specific fractionation of immunocompetent cells. Transplant. Rev., 5, 76-104, 1970.

6. Pink J. R. L., Askonas B. A. Diversity of antibodies to cross-reacting nitrophenyl haptens in inbred mice. Eur. J. Immunol., 4, 426 429, 1974.

7. Butcher E. C., Weissman I. L. Lymphoid Tissues and Organs. In: Fundamental Immunology (W.F.Paul, ed.), pp. 109-127, New York, Raven, 1984. (Russian translation available: Butcher E. S., Weissman I. L. Lymphoid organs and tissues. In: Immunology (W. Paul, ed.), vol. 1, pp. 173-203. Moscow: Mir, 1987.)

Gallatin M., et al. Lymphocyte homing receptors. Cell, 44, 673 680, 1986.

Gowans J. L., Knight E. J. The route of re-Circulation of lymphocytes in the rat. Proc. R. Soc. Lond. (Biol.), 159, 257-282, 1964.

Sprent J. Migration and lifespan of lymphocytes. In: B and T Cells in Immune Recognition (F. Loor, G.E. Roelants, eds.), pp. 59-82. New York, Wiley, 1977.

Woodruff J. J., Clarke L.M., Chin Y.H, Specific cell-adhesion mechanisms determining migration pathways of recirculating lymphocytes. Ann. Rev. Immunol., 5, 201-222, 1987.

8. Greaves M. F., Owen J. J. T., Raff M. C. T and B Lymphocytes: Origins, Properties and Roles in Immune Responses, pp. 117-186. Amsterdam, Excerpta Medica, 1973.

Rajewsky K., Forster I., Cumano A. Evolutionary and somatic selection of the antibody repertoire in the mouse. Science, 238, 1088-1094, 1987.

9. Billingham R. E., Brent L., Medawar P. B. Quantitative studies on tissue transplantation immunity. III. Actively acquired tolerance. Philos. Trans.

R. Soc. Lond. (Biol.), 239, 357-414, 1956.

Harris D. E., Cairns L., Rosen F. S., Borel Y. A natural model of immunologic tolerance. Tolerance to murine C5 is mediated by T cells and antigen is required to maintain unresponsiveness. J. Exp. Med., 156, 567-584, 1982.

Lindstrom J. Immunobiology of myasthenia gravis, experimental autoimmune myasthenia gravis and Lambert-Eaton syndrome. Annu. Rev. Immunol., 3, 109-132, 1985.

Nossal G. J. V. Cellular mechanisms of immunologic tolerance. Annu. Rev. Immunol., 1, 33-62, 1983.

Owen R.D. Immunogenetic consequence of vascular anastomoses between bovine twins. Science, 102, 400-401, 1945.

10. Howard J.G., Mitchison N.A. Immunological tolerance. Prog. Allergy, 18, 43-96, 1975.

11. Davies D.R., Metzger H. Structural basis of antibody function. Annu. Rev. Immunol., 1, 87-118, 1983.

Kabat E. A. Structural Concepts in Immunology and Immunochemistry, 2nd ed. New York, Holt, Rinehart and Winston, 1976.

Nisonoff A., Hopper J. E., Spring S. B. The Antibody Molecule. New York, Academic Press, 1975.

12. Möller G., ed. Lymphocyte Immunoglobulin: Synthesis and Surface Representation. Transplant. Rev., Vol. 14, 1973.

13. Dutton R.W., Mishell R.I. Cellular events in the immune response. The in vitro response of normal Spleen cells to erythrocyte antigens. Cold Spring Harbor Symp. Quant. Biol., 32, 407-414, 1967.

Jerne N.K., et al. Plaque forming cells: methodology and theory. Transplant. Rev., 18, 130-191, 1974.

14. Edelman G.M. The Structure and function of antibodies. Sci. Am., 223(2), 34-42, 1970. Porter R. R. Structural studies of IMMUNOGLOBULINS. Science, 180, 713-716, 1973.

15. Ishizaka T., Ishizaka K. Biology of immunoglobulin E. Prog. Allergy, 19, 60-121, 1975.

Koshland M. E. The coming of age of the immunoglobulin J chain. Annu. Rev. Immunol., 3, 425-454, 1985.

Morgan E. L., Weigle W. O. Biological activities residing in the Fc region of immunoglobulin. Adv. Immunol., 40, 61-134, 1987.

Solari R., Kraehenbuhl J.-P. The Biosynthesis of secretory component and ist role in the Transepithelial Transport of IgA dimer. Immunol. Today, 6, 17-20, 1985.

Underdown B.J., Schiff J.M. Immunoglobulin A: strategic defense initiative at the mucosal surface. Annu. Rev. Immunol., 4, 389-418, 1986. Unkeless J. C., Scigliano E., Freedman V. Structure and function of human and murine receptors for IgG. Annu. Rev. Immunol., 6, 251-282, 1988.

16. Berzofsky J.A., Berkover I.J. Antigen-antibody interactions. In Fundamental Immunology (W.E.Paul, ed.), pp. 595-644. New York, Raven, 1984. (Russian translation available: Berzofsky J.; Berkover A. D. Vzaimodeystviye antigen-antitelo. In: Immunologiya (U. Pol, ed.), vol. 3, pp. 5-88. Moscow: Mir, 1989.)

17. Capra J. D., Edmundson А. В. The antibody combining site. Sci. Am., 236(1), 50-59, 1977.

Wu Т. Т., Kabat E. A. An Analysis of the sequences of the variable regions of Bence Jones proteins and myeloma light chains and their implications for antibody complementarity. J. Exp. Med., 132, 211-250, 1970.

18. Sakano H., et al. Domains and the hinge region of an immunoglobulin heavy chain are encoded in separate DNA segments. Nature, 277, 627633, 1979.

19. Alzari P. M., Lascombe M.-B., Poljak R. J. Three-dimentional Structure of antibodies. Annu. Rev. Immunol., 6, 555-580, 1988.

Capra J.D., Edmundson A.B. The antibody combining site. Sci. Am., 236(1), 50-59, 1977.

20. Dreyer W. J., Bennett J. C. The Molecular Basis of antibody formation: a paradox Proc. Natl. Acad. Sci. USA, 54, 864-869, 1965.

Hozumi N., Tonegawa S. Evidence for somatic rearrangement of immunoglobulin genes coding for Variable and constant regions. Proc. Natl. Acad. Sci. USA, 73, 3628-3632, 1976.

21. AltF. W., Blackwell Т.К., Yancopoulos G.D. Development of the primary antibody repertoire. Science, 238, 1079-1087, 1987.

Leder P. The genetics of antibody diversity. Sci. Am., 246(5), 72-83, 1982. Tonegawa S. The molecules of the immune system. Sci. Am., 253(4), 122-131, 1985.

22. Gearhart P. J., Johnson N. D., Douglas R., Hood L. IgG antibodies to phosphorylcholine exhibit more diversity than their IgM counterparts. Nature, 291, 29-34, 1981.

Griffiths G. M., Berek C., Kaartinen M., Milstein C. Somatic mutation and the maturation of Immune Response to 2-phenyl oxazolone. Nature, 312, 271-275, 1984. Möller G., ed. Role of Somatic Mutation in the Generation of Lymphocyte Diversity. Immunol. Rev., Vol. 96, 1987.

Rajewsky K., Forster I., Cumano A. Evolutionary and somatic selection of the antibody repertoire in the mouse. Science, 238, 1088-1094, 1987.

23. Alt F. W., et al. Ordered rearrangement of immunoglobulin heavy chain variable region segments. EMBO J., 3, 1209-1219, 1984.

Yancopoulos G. D., Alt F. W. Regulation of the assembly and expression of variable region genes. Annu. Rev. Immunol., 4, 339-368, 1986.

24. Early P., et al. Two mRNAs can be produced from a single immunoglobulin μ gene by alternative RNA Processing pathways. Cell, 20, 313319, 1980.

25. Cebra J.J., Komisar J.L., Schweitzer P. A. CН isotype switching during normal B-lymphocyte development. Annu. Rev. Immunol., 2, 493-548, 1984.

Shimizu A., Honjo T. Immunoglobulin class switching. Cell, 36, 801-803, 1984.

26. JerneN.K. The immune system. Sci. Am., 229(1), 52-60, 1973.

Jerne N. K. Toward a network theory of the immune system. Ann. Immunol. Inst. Pasteur (Paris), 125C, 378-389, 1974.

Möller G., ed. Idiotype Networks Immunol. Rev., Vol. 79, 1984. Rajewsky K., Takemori T. Genetics, expression and function of idiotypes. Annu. Rev. Immunol., 1, 569-608, 1983.

27. Lachmann P. J. Complement. In Clinical Aspects of Immunology, 4th ed. (P.J. Lachmann, D. Peters, eds.), pp. 18-49, Oxford U.K., Blackwell, 1982,

Müller-Eberhard H. J. MOLECULAR ORGANIZATION AND function of the complement system. Annu. Rev. Biochem., 57, 321-348, 1988.

Reid K. B. M., Porter R. R. The proteolytic activation systems of complement. Annu. Rev. Biochem., 50, 433-464, 1981.

28. Cooper N. R. The classical complement pathway: activation and requlation of the first complement component. Adv. Immunol., 37, 151-216, 1985.

29. Muller-Eberhard H.J., Schreiber R.D. Molecular biology and chemistry of The alternative pathway of complement. Adv. Immunol., 29, 1-53, 1980.

30. Müller-Eberhard H. J. The membrane attack complex of complement. Annu. Rev. Immunol., 4, 503-528, 1986.

31. Campbell R.D., Carroll M. C., Porter R.R. The Molecular Genetics of components of complement. Adv. Immunol., 38, 203-244, 1980. Lachmann P.J., Hobart M.J. Genetics of complement. Trends Genet., 1, 145-149, 1985.

Reid К. В. Activation and control of the complement system. Essays Biochem., 22, 27-68, 1986.

32. Allison J. P., Lanier L. L. The structure, function and serology of the T cell antigen receptor complex. Annu. Rev. Immunol., 5, 503-540, 1987. Clevers H., Alarcon В., Wileman Т., Terhorst C. The T cell receptor/CD3 complex: a dynamic ensemble. Annu. Rev. Immunol., 6, 629-662, 1988. Davis M.M., Bjorkman P.J. T-cell antigen receptor genes and T-cell recignition. Nature, 334, 395-402, 1988.

Hedrick S. M., Cohen D., Nielsen E. A., Davis M. M. Isolation of cDNA clones marrow cells - a model for the two-Cell Theory of immunocompetence. Transplant. Rev., 1, 92-113, 1969.

Davies A.J.S. The thymus and the Cellular Basis of immunity. Transplant. Rev.,1, 43-91, 1969.

44. Chesnut R. W., Grey H. M. Antigen presentation by В cells and its significance in T-B interactions. Adv. Immunol., 39, 51-94, 1986.

DeFranco A. L. Molecular aspects of B-lymphocyte activation. Annu. Rev. Cell Biol, 3, 143-178, 1987. Kishimoto Т., Hirano T. Molecular

regulation of В lymphocyte response. Annu. Rev. Immunol., 6, 485-512, 1988.

Möller G., ed. 1L-4 and 1L-5: Biology and Genetics. Immunol. Rev., Vol. 102, 1988.

Paul W. E., Ohara J. B-cell stimulating factor-1/Interleukin 4. Annu. Rev. Immunol, 5, 429-460, 1987. Cambier J.C., Ransom J. T. MOLECULAR MECHANISMS OF transmembrane signaling in В lymphocytes. Annu. Rev. Immunol., 5, 175200, 1987.

45. Möller G., ed. Gamma Interferon. Immunol. Rev., Vol. 97, 1987.

Mosmann T. R., Coffman R. L. Two types of mouse helper T-cell clone implications for immune regulation. Immunol. Today, 8, 223-227, 1987. Powrce F., Mason D. Phenotypic and functional heterogeneity of CD4+ T cells. Immunol. Today, 9, 274-277, 1988.

46. Littman D. R. The structure of the CD4 and CD8 genes. Annu. Rev. Immunol., 5, 561-584, 1987.

Springer T.A., Dustin M.L., Kishimoto Т.К., Marlin S.D. The lymphocyte-function associated LFA-1, CD2 and LFA-3 molecules: cell adhesion receptors of the immune system. Annu. Rev. Immunol., 5, 223-252, 1987.

47. Asherson G. L., Colizzi V., Zembala M. An Overview of T suppressor cell circuits. Annu. Rev. Immunol., 4, 37-68, 1986.

Dorf M.E., Benacerraf B. Suppressor cells and immunoregulation. Annu. Rev. Immunol., 2, 127-158, 1984.

Gershon R. K. T-cell control of antibody production. Contemp. Top. Immunobiol, 3, 1-40, 1974.

Harris D. E., Cairns L., Rosen F. S., Borel Y. A natural model of immunologic tolerance. Tolerance to murine C5 is mediated by T cells and antigen is required to maintain unresponsiveness. J. Exp. Med., 156, 567-584, 1982.

Mitchison N. A., Griffiths J. A., Oliveira D. B. G. Immunogenetics of suppression. Br. J. Clin. Pract, 40, 225-229, 1986.

48. Bevan M. In a radiation chimera, host H-2 antigens determine the immune responsiveness of donor cytotoxic cells. Nature, 269, 417-419, 1977. Kappler J. W., Roehm N., Marrack P. T cell tolerance by clonal elimination in the thymus. Cell, 49, 273-280, 1987.

Sprent J., Lo D., Gao E.-K., Ron Y. T cell selection in the thymus. Immunol. Rev., 101, 5-19, 1988.

Sprent J., Webb S. Function and specificity of T cell subsets in the mouse. Adv. Immunol., 41, 39-133, 1987. von Boehmer H. The developmental biology of T lymphocytes. Annu. Rev. Immunol., 6, 309-326, 1988.

49. Alien P.M., Babbitt B. R., Unanue E.R. T cell recognition of Lysozyme: the BIOCHEMICAL BASIS OF presentation. Immunol. Rev., 98, 171-187, 1987.

Buus S., Sette A., Grey H. M. The interaction between protein-derived immunogenic peptides and Ia. Immunol. Rev., 98, 115-142, 1987. Mengle-Gaw L., McDevitt H. 0. Genetics and expression of murine Ia antigens Annu. Rev. Immunol., 3, 367-396, 1985.

SchwartzR. H. Immune response (Ir) genes in the murine major histocompatibiliy complex. Adv. Immunol., 39, 31-201, 1986.

50. Williams A. F., Barclay A. N. The immunoglobulin superfamily - domains for cell surface recognition. Annu. Rev. Immunol., 6, 381-406, 1988.