Chapter 76: Functions of Tlymphocytes: T-Cell Receptors for Antigen

T-CELL RECEPTOR HETERODIMERS

The structural basis of T-cell recognition of antigen has been known since the 1990s, when a plethora of studies demonstrated that the proteins of the T-cell antigen receptor are structurally related to immunoglobulin molecules.¹ The T-cell receptor is formed by a heterodimer, that is, two disulfide-bond-linked polypeptides that are expressed on the cell surface and are associated with a collection of coreceptor accessory and invariant CD3 proteins. In contrast to immunoglobulins, the T-cell receptor is not secreted and also remains membrane bound throughout the activation process. In the majority of T cells, the T-cell receptor heterodimer consists of an α and a β chain, but a small subset of T cells expresses a γδ heterodimer. Following the rule of allelic exclusion, each individual T cell expresses a single α and a single β chain (or a single γ or δ chain, respectively), and can either be αβ or γδ. Each chain is composed of a variable region, consisting of a hydrophobic leader sequence of 18 to 29 amino acids and an aminoterminal domain of 102 to 119 amino acids, and a constant region with a carboxyl-terminal region segment of 87 to 113 amino acids. The variable region is responsible for the variation in the primary structure among different T-cell receptor polypeptides and represents the antigen-binding site, while the constant region is invariant among chains of the same class. Similar to other surface-membrane receptors, each chain is followed by a small connecting peptide, a transmembrane region of 20 to 24 amino acids, and a small cytoplasmic region of 5 to 12 residues at the carboxyl terminus anchoring the polypeptide in the cell membrane. The T-cell receptor chains fold into tertiary structures that are very similar to that of the light and heavy chains of the immunoglobulin molecule. Overall, the structural similarities between the T-cell receptor and immunoglobulins place the genes encoding these receptor proteins in the so-called immunoglobulin supergene family.

αβ Heterodimers

More than 90 percent of mature T cells express an αβ heterodimer, making this the major class of T-cell receptor. Without glycan side chains, each α or β polypeptide has a respective size of only 27 kDa or 32 kDa. However, within minutes after being translated into protein, both chains are glycosylated and assembled into a heterodimer composed of a single acidic α glycoprotein of 39 to 46 kDa linked to a more basic 40- to 44-kDa β-glycoprotein via a disulfide bond between the constant regions of the two chains (Fig. 76–1).

γδ Heterodimers

Less than 10 percent of blood T cells and thymocytes exclusively express a γδ heterodimer. Within the γδ T-cell population, specific γδ T-cell subsets can be defined by their T-cell receptor gene element usage, and their functionally distinct responses to infections with certain organisms such as Listeria monocytogenes.² Alternatively, γδ T cells can be grouped according to their tissue location. In secondary lymphoid tissues, only 1 to 5 percent of the CD3-positive T cells express γδ receptors. Murine studies demonstrate that many epithelial tissues, such as the epidermis, the intestine, the lung, and the uterus, are enriched for γδ-expressing T cells, indicating that γδ T cells are involved in the surveillance of body barriers.³

The amino acid sequence of the γ chain resembles the T-cell receptor β chain, whereas the amino acid sequence of the δ chain resembles the α chain. Like the homologous αβ heterodimer, the γδ heterodimer is also associated with the CD3 complex and is capable of initiating T-cell activation upon binding of specific ligands. Despite the similarities in chain structure and size, however, the tertiary structure of variable regions of γδ T-cell receptors has a closer resemblance to immunoglobulin variable regions than to the variable regions of αβ T-cell receptors.

T-CELL RECEPTOR HETERODIMER GENE REARRANGEMENT

Similar to immunoglobulin genes, each chain of the T-cell receptors is encoded by distinct genetic elements that rearrange during development generating a diverse T-cell repertoire (Fig. 76–2).⁴ Located at band q35 on the long arm of chromosome 7, the β-chain complex has two closely linked genes, each capable of encoding the β-chain constant region. Each constant region gene is associated with a cluster of functional Jβ-gene segments and a single Dβ segment. The functional gene encoding the variable region of the β chain is constructed from the rearrangement of any of approximately 50 variable region gene segments to either one of the two Dβ regions and one of 13 Jβ regions.

The α-chain complex is located at band q11.2 on the long arm of chromosome 14 and thus is linked to the immunoglobulin heavy-chain complex. The α-chain gene complex consists of one constant region gene and at least 50 different variable region gene segments. The functional gene encoding the α-chain variable region is derived from the juxtaposition of any one of the variable region gene segments with one of the many Jα segments through rearrangement that generally involves the deletion of the intervening DNA.

The organization of the γ and δ genes is similar to that of the α and β genes, but some significant differences exist. First, the gene complex encoding the δ genes is located entirely within the α-chain gene complex between the Vα and Jα gene segments. Consequently, any rearrangement of the α-chain genes inactivates the genes encoding the δ chain. Second, there are fewer V gene segments in the γ and δ gene complexes than at either the T-cell receptor α or β gene loci. The γ-gene complex on band p15 on the short arm of chromosome 7, for example, has only approximately 12 Vγ gene segments, two virtually identical Jγ segments, and two constant region gene segments. Moreover, there are only approximately four Vδ gene segments, three Dδ gene segments, three Jδ gene segments, and a single constant region gene in the δ gene complex. Consequently, most of the variability in the γ and δ chains is found in the junctional region formed during the process of γδ T-cell receptor gene rearrangement. The amino acids encoded by this region form the center of the T-cell receptor-binding site.

The identification of TCR gene rearrangements is also widely used for diagnostic and therapeutic purposes: In suspected lymphoproliferative disorders, PCR-based clonality testing has been standardized and guidelines and consensus reporting systems have been established.⁵ Moreover, the molecular analysis for clonal T-cell receptor gene rearrangements can be used to detect minimal residual disease in patients treated for clonal T-cell disorders, such as T-cell acute lymphoblastic leukemia (T-ALL), and guide clinical decision making and provide prognostic information.^6,7

ANTIGEN PRESENTATION TO T-CELL RECEPTOR HETERODIMERS

Despite their similarities in structure, there are important differences in the way that T-cell receptors and immunoglobulins recognize antigen: immunoglobulins can bind antigens directly, while T-cell receptors generally require that peptide antigens are bound to a molecule of the major histocompatibility complex (MHC) on the surface of another cell.^8,9 MHC molecules, also known as histocompatibility antigens, are highly polymorphic glycoproteins, and have immunoglobulin-like structures themselves. There are two basic classes of MHC molecules: Class I MHC molecules generally bind and present intracellular proteins, that is, peptides that are derived from proteins synthesized and degraded in the cytoplasm of the cell. Class II MHC molecules generally bind peptides that are derived from exogenous proteins and degraded in intracellular vesicles. The human histocompatibility antigens HLA-A, HLA-B, and HLA-C are class I molecules, whereas HLA-D antigens DP, DQ, and DR are examples of class II molecules.

MHC class I molecules bind peptides that are usually 8 to 10 amino acids long. Their binding is stabilized by contacts in the free amino and carboxyl termini of the peptide and the peptide-binding groove of all MHC class I molecules. In addition, the peptide-binding groove is closed at both ends. The corresponding binding groove on class II molecules is open at either end, and peptides that bind to MHC class II molecules are generally at least 13 amino acids long. For both class I and class II molecules, the peptide binding groove is located in the central cleft between the two α helices of the MHC molecule. Steric factors, hydrogen bonding, and hydrophobic interactions between the peptide and the MHC molecule stabilize the peptide within this cleft and generate a tertiary structure that is further modified by amino acid residues of both the MHC and the peptide antigen.

MHC molecules are encoded by a family of MHC genes, which are located on chromosome 6 in humans (Chap. 137). MHC genes can be divided into three subgroups, with antigen presentation being encoded by class I (i.e., HLA-A, -B, -C) and class II (i.e., HLA-DP, DQ, DR) genes. Each of these gene loci exists in different alleles, and both maternal and paternal alleles are expressed concomitantly. The particular combination of class I and class II MHC alleles found on an individual chromosome is known as the MHC haplotype. Therefore, the number of different MHC molecules is greatly increased not only by MHC gene polymorphism, but also by codominant expression of MHC gene products. The resulting differences within MHC molecules are primarily found in the amino acids lining the clefts that hold the peptide antigen, allowing the MHC molecules encoded by each allele to bind a distinctive array of different peptides.

Structural studies show that the T-cell receptor can recognize both the MHC-bound peptide and the polymorphic amino acid residues themselves that surround the peptide-binding pocket.¹⁰ The specificity of a T-cell receptor (TCR) is defined both by the peptide it recognizes and by the MHC molecule binding it, and TCRs are therefore likely to engage a composite peptide/MHC receptor.^11,12 Additional TCR–MHC binding parameters, such as ligands per cell or dissociation time defined by the affinity of antigen, also have an effect on subsequent T-cell activation, but their exact relationship remains largely controversial.¹³

Some T cells, however, do not recognize MHC-bound peptides. Instead they recognize antigens that are presented by MHC class I-like molecules encoded by genes that map outside the MHC region. One such family of molecules is called CD1. Despite structural similarities with MHC class I molecules, CD1 molecules present larger extracellular peptides in a MHC class II-like fashion, and are also able to present glycolipids to T cells. Glycolipids are components of mycobacterial membranes; in cells infected with mycobacteria, the CD1 molecules are therefore able to bind and present membrane components such as lipoarabinomannan or mycolic acid. T cells that recognize these complexes play an important role in the immune response to Mycobacterium tuberculosis.^14,15

As discussed above, structural studies show that γδ TCRs assume a different tertiary structure than αβ TCRs. As a consequence, γδ TCRs are able to recognize a wider variety of ligands, such as bacterial phosphoantigens, nonclassical MHC-I molecules and unprocessed proteins, which distinguishes them from the great majority of αβ T cells.¹⁶ Other γδ receptors can recognize determinants presented by CD1 molecules.¹⁷ A number of studies have described increased numbers of γδ T cells in a variety of infectious and autoimmune diseases. Therefore, it is speculated that γδ T cells link innate and adaptive immune responses under infectious and inflammatory conditions. In addition, it is hypothesized that γδ T cells might play a role in novel T-cell-based immunotherapy strategies.^18,19

GENERATION OF T-CELL RECEPTOR DIVERSITY

TCR diversity is achieved by several mechanisms, some of which are the same as those that generate diversity among immunoglobulin molecules (Chap. 75). The joining of different V (variable), D (diversity), and J (joining) elements to produce a complete V gene, the presence of uncorrected errors made during the recombination of these genetic elements, and the combinatorial diversity afforded by the random pairing of two chains encoded by separated gene complexes all function to enhance the diversity of the T-cell antigen receptor repertoire. An important difference between T cells and B cells, however, is that B cells are capable of undergoing somatic hypermutation (Chap. 75). This process requires expression of activation-induced deaminase (AID) along with other enzymes expressed primarily by B cells within the germinal center of secondary lymphoid tissue during the immune response to antigen (Chaps. 6 and 75).

TCRs do not undergo somatic mutation, probably because of the central role they play in directing host immune defenses: During differentiation, immature αβ T-cell precursors pass through the thymus, where they are educated to distinguish self from non-self by cell-surface proteins of the MHC (Chaps. 6 and 74). Because the ligand for the αβ TCR is processed antigen presented by MHC molecules, close interaction with the MHC might be lost if the variable region of the TCR were allowed to diverge significantly from the inherited germline repertoire. Furthermore, somatic mutation of expressed TCR variable region genes may lead to constitutive T-cell activation to processed self-antigen presented by self-MHC molecules, which could lead to a breakdown in tolerance to self-antigens and autoimmunity.

COMPOSITION OF THE T-CELL RECEPTOR COMPLEX

The CD3 complex of polypeptides and CD247, also known as the zeta chain (ζ chain) of the TCR, are closely associated with and required for the surface expression of the polypeptide heterodimer.²⁰ Unlike the TCR heterodimers, these polypeptides are invariant and are found on all T cells that express αβ or γδ heterodimers. The CD3 polypeptides are designated CD3γ, CD3δ, and CD3ε. The CD3ε chain couples with either the CD3γ or the CD3δ chain to generate heterodimers that each form a tight association with the αβ (or γδ) receptor heterodimer on the T-cell surface (see Fig. 76–1). Each CD3 polypeptide has a negatively charged amino acid in the central portion of the hydrophobic transmembrane region that stabilizes the CD3 complex with the two chains of the TCR. The ζ chain (CD247), on the other hand, forms a disulfide-like homodimer that primarily associates with the two TCR chains and only weakly associates with the CD3 complex. As such, it cannot be coimmunoprecipitated easily with antibodies to the CD3 polypeptides. In addition, the ζ chain lacks a significant extracellular domain (see Fig. 76–1).

MOLECULAR FEATURES OF THE T-CELL RECEPTOR COMPLEX

The genes encoding CD3γ, CD3δ, or CD3ε chains are clustered on the long arm of chromosome 11 in band q23. CD3γ has a 16-kDa polypeptide backbone that is heavily glycosylated to assume a final molecular mass of 25 to 28 kDa. CD3δ and CD3ε are each 20 kDa in molecular mass. The CD3δ is a glycoprotein consisting of 30 percent carbohydrate. In contrast, CD3ε is not glycosylated. CD3δ and CD3γ are highly homologous at both the protein and nucleic acid sequence level. The nucleic acid sequence of each predicts CD3δ and CD3γ to have typical signal peptides, respective hydrophilic extracellular domains of 79 to 89 amino acids, hydrophobic transmembrane regions of 27 amino acids, and hydrophilic intracellular domains of 44 to 55 amino acids. CD3ε is similar, with a 22-residue signal peptide, an extracellular domain of 104 amino acids, a transmembrane domain, and a comparatively long intracellular domain of 81 amino acids. Each CD3 polypeptide has one immunoglobulin-like domain in its extracellular domain that is defined by an intrachain disulfide bond (see Fig. 76–1), indicating that these polypeptides are members of the immunoglobulin superfamily. However, unlike the αβ or γδ chains of the TCR, there is no variability in the extracellular domains of the CD3 proteins, indicating that these molecules do not contribute to the specificity of antigen recognition.

The ζ chain has no sequence or structural homology to the other three CD3 chains. It is a nonglycosylated protein of 16-kDa molecular mass that is encoded by a gene found on chromosome 1. The ζ chain has only a very short extracellular domain of 6 to 9 amino acids, a transmembrane domain of 21 amino acids, and a long intracellular domain of 113 amino acids.

The cytoplasmic domains of all the CD3 polypeptides and the ζ chain each contain sequences termed immunoreceptor tyrosine-based activation motifs (ITAMs). Each ITAM contains two copies of the sequence tyrosine-X-X-leucine separated by six to eight amino acid residues, in which X represents an unspecified amino acid. The cytoplasmic domains of each CD3 polypeptide contain one ITAM, whereas each ζ chain contains three ITAMs (see Fig. 76–1). These sequences allow the CD3 proteins to associate with cytosolic protein tyrosine kinases following TCR ligation, thus transducing a signal to the interior of the T cell. The cytoplasmic domains of CD3ε and CD3ζ are particularly important in this regard.

SIGNAL TRANSDUCTION VIA THE T-CELL RECEPTOR COMPLEX

Although the major components of the TCR signaling machinery have been identified and mapped, key components continue to be discovered, adding to the complex network of TCR signal transduction. Signal transduction from the TCR heterodimer to intracellular proteins is mediated by the CD3 polypeptides and the ζ chain. Their spatial organization and the mechanisms by which this is regulated are, however, still poorly understood.²¹ Upon binding to a specific ligand, the TCR αβ (or γδ) heterodimer undergoes steric changes that result in the phosphorylation of the ITAMs of the ζ chain and each of the CD3 polypeptides (see Fig. 76–1). When tyrosine residues in the ITAMs become phosphorylated, they can act as docking sites for adapter proteins or tyrosine kinases, such as the zeta-associated protein of 70 kDa (ZAP-70), which possesses a Src homology 2 (SH2) domain and a Src homology 3 (SH3) domain. Following ligation of the TCR, Src family protein tyrosine kinases (e.g., Lck) are recruited and become activated. This differentially phosphorylates the ITAMs of the accessory molecules in the TCR complex.²² ZAP-70 is recruited to the phosphorylated ITAMs of the ζ chain via its SH2 and SH3 domains, and subsequently becomes activated. Activated ZAP-70 phosphorylates tyrosine residues in the intracytoplasmic segments of surface CD6 and membrane-anchored adapter protein called linker of activation of T cells (LAT), both of which constitute a device for the amplification and diversification of signals that are responsible for most of the responses that result from engagement of the TCR.²³

The network of proteins that interact with the LAT signalosome includes SLP-76, phospholipase Cγ₁ (PLC-γ₁) and GRAP2. Activated PLC-γ₁ mediates hydrolysis of phosphatidylinositol-(4,5)-bisphosphate at the cell membrane, which generates the second messengers inositol-(1,4,5)-trisphosphate and polyunsaturated diacyglycerols, leading to a rapid increase in cytosolic free calcium and activation of the θ isoform of protein kinase C (PKC). Cytosolic free calcium binds to calmodulin, an ubiquitous calcium-dependent regulatory protein. The calcium–calmodulin complex activates the cytoplasmic phosphatase calcineurin, which, in turn, catalyzes the removal of an inhibitory phosphate group on the nuclear factor of activated T cells (NFATs) that retains NFAT proteins in the cytoplasm. Removal of the phosphates from NFAT1 and NFAT2 by activated calcineurin allows these transcription factors to translocate into the nucleus, where they enhance transcription of several activation-induced genes, including those encoding interleukin (IL)-2, IL-4, and tumor necrosis factor.²⁴ The importance of this pathway in T-cell activation is underscored by the strong immunosuppressive activity of calcineurin inhibitors such as cyclosporine and FK506, which are commonly administered in the clinic to treat autoimmune diseases and to prevent graft rejection.

In parallel, diacylglycerol in the plasma membrane recruits RasGRP1, a guanine nucleotide-exchange factor (GEF), which acts on Ras to activate extracellular receptor-activated kinase 1 and 2 (ERK1/2). Activated ERK phosphorylates Elk, which, in turn, stimulates transcription of Fos, a component of the activation protein-1 (AP-1) factor that is a necessary component of the transcription-factor complex required for expression of IL-2 and other critical T-cell proteins. LAT-bound SLP-76 also interacts with Nck and with Vav1 to promote reorganization of the actin cytoskeleton, and with FYB to increase the binding of the integrin CD58 (LFA-1) to its ligand intercellular adhesion molecule (ICAM)-1. DYN2, a member of the dynamin superfamily of large guanosine triphosphatases (GTPases), is also recruited by phosphorylated LAT molecules and participates in the generation of filamentous actin. Dedicator of cytokinesis 2 (DOCK2), another GEF, also activates GTPases of the Rac family, which, in turn, activate another mitogen-activated protein (MAP) kinase called p38, and initiate a parallel enzymatic cascade resulting in the activation of yet another MAP kinase called c-Jun N-terminal kinase (JNK), otherwise known as stress-activated protein kinase (SAPK). Activated JNK phosphorylates c-Jun, the second component of the AP-1 transcription factor required for IL-2 transcription. The guanosine triphosphate (GTP)-bound form of Rac also induces cytoskeletal reorganization, thereby facilitating the clustering of the TCR complex, accessory molecules, and other accessory proteins at the site of contact between the T cell and the antigen-presenting cell (APC).

STRUCTURE OF CD4 AND CD8

CD4 and CD8 are glycoproteins that share structural features of other immunoglobulin superfamily receptor molecules. CD8 has two isoforms with different expression patterns and presumably different functions, and is expressed as a CD8α/CD8β heterodimer or as a CD8α/CD8α homodimer.²⁵ These chains are encoded by genes that are linked closely to the immunoglobulin κ light-chain locus at band p12 on the short arm of chromosome 2. The protein sequence of the aminoterminal domains of each CD8 chain shares greater than 28 percent homology with κ light-chain variable regions. Therefore, these domains are called the variable-region-like (V-like) domains. Following this V-like domain, the CD8 molecule has a short region rich in prolines, threonines, and serines that resembles the immunoglobulin hinge region. This region also contains sites for O-linked glycosylation. A hydrophobic transmembrane region anchors the hinge-like region. The CD8 molecule has a 25-amino-acid cytoplasmic tail consisting of highly basic residues. Two cysteines within the V-like domain form a disulfide bridge that stabilizes the immunoglobulin-like fold. An additional cysteine residue is located each within the V-like domain, the hinge region, the transmembrane segment, and the cytoplasmic domain. These cysteines form intermolecular disulfide bridges between two CD8 molecules, thereby stabilizing the CD8α/CD8β heterodimers or CD8α/CD8α homodimers on the T-cell surface. The cell surface CD8 heterodimer shares structural geometry with the heterodimers formed by the pairing of immunoglobulin light and heavy chains.

CD4, on the other hand, is expressed as a monomer on the surface of a subset of peripheral T cells, mononuclear phagocytes and some blood-derived dendritic cells. It is a 55-kDa monomeric glycoprotein that is encoded by a gene that maps to the short arm of chromosome 12. It consists of 5 external domains, a stretch of hydrophobic transmembrane residues, and a highly basic cytoplasmic tail of 38 residues. Similar to CD8, the aminoterminal domain of CD4 also has extensive homology to immunoglobulin light-chain variable regions. However, following this immunoglobulin-like domain is a domain of 270 amino acids that bears little resemblance to other proteins of the immunoglobulin superfamily.

The cytoplasmic regions of CD4 and CD8 are conserved among vertebrates, suggesting that they are essential for the function of these molecules. The cytoplasmic region of CD4 contains five serines and threonines, one or more of which is phosphorylated by PKC upon activation of T cells by phorbol esters or exposure to antigen. Subsequent to phosphorylation, the CD4 glycoprotein is internalized concomitant with T-cell activation. Similarly, the CD8 protein also possesses a highly charged and conserved cytoplasmic domain that may be involved in transmembrane signal transduction.

FUNCTION OF CD4 AND CD8

In addition to MHC antigen presentation, TCRs generally require activation via CD4 and CD8 as coreceptors. Imaging studies and affinity measurements have demonstrated that CD4 and CD8 molecules associate on the plasma membrane with components of the TCR and contribute to antigen recognition and stabilization of TCR–MHC interactions.²⁶ The adhesion between the CD3/TCR complex and the MHC glycoproteins expressed by an APC or target cell is more than 100-fold enhanced by CD8 or CD4, probably by focusing MHC molecules of the APC or target cell onto the T-cell surface, allowing for specific recognition of “processed” antigen that is cradled within the MHC glycoproteins. However, CD4 and CD8 differ in their expression patterns (see section “Helper and Cytotoxic T Cells and CD4+ T Cell Subsets” below) and MHC-binding specificities: CD8 binds to the nonpolymorphic α₃ domain of the HLA class I molecule (HLA-A, -B, or -C), whereas CD4 binds to the nonpolymorphic β₂ domain of HLA class II molecules (HLA-DP, -DQ, and -DR).²⁷ Therefore, T cells expressing CD4 or CD8 generally recognize antigens presented by class II or class I MHC glycoproteins, respectively. This selectivity is underscored by studies on knockout mice that lack expression of either of these accessory molecules. Mice lacking CD4 or CD8 fail to develop class II-restricted or class I-restricted T cells, respectively, indicating that these coreceptors play essential roles in the maturation of T cells in the thymus. A similar defect is observed in patients with the bare lymphocyte syndrome who have a congenital immune deficiency caused by genetic defects in their capacity to make MHC class II molecules.²⁸ Although patients have normal numbers of B cells and T cells, they have markedly reduced numbers of CD4+ T cells, thus accounting in part for their profound immune deficiency.

In addition to serving as coreceptors, CD4 or CD8 molecules enhance antigen responsiveness by transducing a signal either directly or in concert with the CD3/TCR complex. This is mediated through their interaction with the SRC family tyrosine kinase Lck.²⁹ Lck is noncovalently associated with the cytoplasmic tails of CD4 and/or CD8. When a T cell recognizes a peptide presented by an appropriate MHC antigen, the interaction of CD4 or CD8 with the MHC molecule brings Lck close to the TCR complex. Lck then phosphorylates the tyrosine residues in the ITAMs of CD3 polypeptides and the ζ chain, thereby initiating the receptor signaling required for T-cell activation.

Finally, CD4 also is a cellular coreceptor for HIV.³⁰ Binding of CD4 along with chemokine receptors such as CCR5 or CXCR4 facilitates entry of the virus into host T cells and stimulates them in an antigen-driven immune response.³¹ Targeting HIV entry/fusion by specific monoclonal antibodies and/or inhibitors is, therefore, an important therapeutic approach in HIV.³² Additionally, CD4 is also of prognostic relevance, as disease progression correlates with depletion of blood T cells that express CD4 (Chap. 81).

PRECURSOR THYMOCYTES

T lymphocytes develop in the marrow from a common lymphoid progenitor that also gives rise to B lymphocytes. While B-lymphocyte precursors remain in the marrow, T-cell precursors migrate to the thymus, where they undergo distinct maturation steps and immunologic education. This is accompanied by characteristic TCR gene and surface expression changes of the CD3 complex, CD4, and CD8. At the early stage, thymocytes are double-negative and express neither CD4 nor CD8. This is a highly heterogeneous population, which includes γδ T cells, αβ T cells that also express the NK1.1 receptor commonly found on natural killer (NK) cells, and immature thymocytes that do not yet express a complete TCR molecule, but are thought to be precursors to the αβ lineage. The latter start to express CD8 and CD4 and enter the double-positive stage, where they undergo positive/negative selection events and CD4/CD8 cell fate choice. This results in the generation of mature thymocytes and peripheral T cells that express either CD4 or CD8, but not both (Chap. 74).³³

HELPER AND CYTOTOXIC T CELLS

The mutually exclusive expression of CD4 or CD8 on mature T cells defines two major blood T-cell subsets: blood T cells that express CD8 normally constitute 25 to 35 percent of the peripheral T-cell population. They recognize antigens presented by MHC class I molecules and differentiate into cytotoxic CD8 T cells. Their main function is lysis of the target cell bearing the surface antigens for which a cytotoxic T cell is specific. Within this subset, there are a range of phenotypes defined both by function and expression of markers with an immunoregulatory function. Blood T cells that solely express the CD4 surface antigen are designated helper T cells. They normally comprise approximately 65 percent of blood T cells. Generally, their function is the production of lymphokines upon activation by foreign antigens presented by MHC class II molecules, regulating and/or assisting in the active immune response. Helper T cells can differentiate into several subtypes, each secreting different cytokines to facilitate a different type of immune response.

CD4+ T-CELL SUBSETS

Th1 and Th2 Cells

Based on distinct cytokine patterns upon activation, mature CD4+ T cells may be divided into subsets–the first of which identified were named T-helper type 1 (Th1) and T-helper type 2 (Th2).³⁴ In general, Th1 cells are a major source of interferon-γ (IFN-γ), and also the major T-cell population involved in activating macrophages and clearing intracellular pathogens. Th2 cells are a major source of IL-4 and are important for the generation of immunoglobulin (Ig) E, the production of eosinophils, and the immune defense against infections by parasites. Both subsets are produced from a noncommitted population of precursor T cells. The process by which commitment develops is called polarization.

In addition to IFN-γ, Th1 cells produce lymphotoxin β, IL-2, and IL-12, whereas Th2 cells also produce IL-5, IL-13, and IL-25. Human Th1 and Th2 cells can also differ in the array of surface antigens or cytokine receptors they express. Th1 cells preferentially express CD26, membrane IFN-γ, the chemokine receptors CCR1 (CD191), CCR2 (CD192), CCR5 (CD195), CXCR3 (CD183), and CXCR6 (CD186), and the receptor for IL-12 (IL-12R or CD212). Higher levels of the lymphocyte activation gene 3 (LAG-3 or CD223), a ligand for MHC class II antigens that is structurally related to CD4, have also been described. Th2 cells preferentially express CD62L, the α chain of the IL-4 receptor (IL-4Rα), the α chain of the IL-33 receptor (IL-33Rα), CD30, and the chemokine receptors CCR3 (CD193), CCR4 (CD194), CCR8 (CDw198), and, to some extent, CXCR4 (CD184).^35,36,37 Distinctive expression levels of these cytokine and chemokine receptors, along with distinctive binding activities for various endothelial selectins, most likely account for the differences in the response to cytokines and tissue-specific migration of these helper T-cell subsets.

The cytokines produced by each subset also stimulate polarization of additional T cells to the same subset, while inhibiting the polarization of the other subset. In naïve CD4+ T cells, the Th1 cytokine IFN-γ induces or activates the signal transducer and activator of transcription (STAT) 1, STAT4, and the T-box transcription factor T-BET, while simultaneously modulating IL-2 and Th2 cytokines, resulting in an attenuation of Th2 cell development (Fig. 76–3).³⁸ Several other studies demonstrate that T-BET also physically interacts with other transcription factors important for alternative T helper cell developmental decisions to functionally repress the opposing subtype specific gene expression programs and promote Th1 development.³⁹ Similarly, IL-4 activates or enhances the expression STAT5, STAT6, and GATA3, transcription factors that play important roles in Th2 cell development (Fig. 76–3).⁴⁰ Another Th2 cytokine, IL-10, inhibits Th1 cell activation, thereby limiting the production of Th1-type cytokines. Because of these self-amplifying and mutually excluding feedback loops, an immune response becomes increasingly polarized once it develops along a Th1 or Th2 pathway, particularly upon protracted stimulation by chronic infection or prolonged exposure to environmental antigens. Other factors that drive polarization are chemokines (explaining the differential expression of chemokine receptors on Th1/Th2 cells), eicosanoids, oxygen free radicals, various inflammatory mediators, and direct cell-to-cell interaction with APCs.

Functionally, Th1 cells predominantly drive cellular immunity to fight viruses and other intracellular pathogens, eliminate cancerous cells, and stimulate delayed-type hypersensitivity (DTH) skin reactions. This is mostly achieved by stimulating macrophage Fc receptor expression, phagocytosis, and antigen presentation, enhancing the capacity of macrophages to kill intracellular pathogens. Th2 cells drive humoral immunity and upregulate antibody production to fight extracellular organisms. They initiate the antibody response to antigen by activating naïve antigen-specific B cells to produce IgM antibodies, subsequently stimulate the production of switched immunoglobulin isotypes, including IgA, IgE, and neutralize and/or weakly opsonize subtypes of IgG (Chap. 75), probably via IL-4 as a B-cell stimulatory/growth factor. In addition to stimulating the production of IgE antibodies, the cytokines made by Th2 cells induce differentiation of mast cells and eosinophils.³⁴

Several studies have indicated that Th2 polarization and accumulation at inflammatory sites is likely to trigger the hypersensitivity reaction in allergic diseases.⁴¹ On the other hand, these responses are protective against metazoan parasite infections such as helminths: Th2 responses are host protective while extracellular parasites migrate through the body, and reduce the number of parasites either through direct killing in the tissues or expulsion from the intestines.⁴² Other studies demonstrate that eosinophilia and elevated IgE that accompany infection with Schistosoma mansoni are caused by the induction of Th2-type cells in the immune response to parasite ova.⁴³ The Th2 polarization in these diseases/conditions is most likely a result of minimal IL-4 secretion during initial activation. If an antigen is present at high concentrations but does not trigger acute inflammation and attendant production of IL-12, the local concentration of IL-4 increases over time and induces a Th2 polarization of cells. On the other hand, pathogens that induce acute inflammation and/or engage toll-like receptors on accessory cells and macrophages can promote production of IFN-γ and IL-12, thereby stimulating development of the immune response along the Th1 pathway.³⁸ Immune responses restricted to Th1 cells, for example, are observed in patients with leprosy who have developed cellular immunity to Mycobacterium leprae⁴⁴ or M. tuberculosis,⁴⁵ in patients with Yersinia enterocolitica⁴⁶ or arthritis triggered by infection with Borrelia burgdorferi.⁴⁷

However, it is probably overly simplistic to view the Th1 pathway as being the more aggressive of the two, generating acute organ-specific autoimmune diseases and inflammations, while the Th2 pathway predisposes to atopic diseases and systemic autoimmune disease. For example, Helicobacter pylori-associated peptic ulcer can be regarded as a Th1-driven immunopathologic response to some H. pylori antigens, while deregulated and exhaustive H. pylori-induced T-cell–dependent B-cell activation can support the onset of low-grade B-cell lymphoma.⁴⁸ In addition, many chronic inflammatory and autoimmune conditions such as rheumatoid arthritis (RA), type 1 diabetes, and multiple sclerosis (MS) are mixed Th1/Th2 conditions, and a clear Th1/Th2 bias has also not been identified yet for many types of cancer.³⁴

CD4+CD25+ Regulatory T Cells

Another type of CD4 cells that suppresses rather than provides helper activity are regulatory T (T_REG) cells. T_REGs possess potent suppressive capacity and can exert diverse suppressive mechanisms allowing them to influence a very broad range of cell populations in a variety of anatomical locations and disease scenarios, both by direct cell–cell contact and the secretion of cytokines.⁴⁹ The cardinal phenotypic features of CD4⁺ T_REG cells include their constitutive expression of the transcription factor forkhead box P3 (FOXP3), their cell surface expression of CD25 (the low-affinity receptor for IL-2 and their cell surface and cytoplasmic expression of the coinhibitory receptor cytotoxic T-lymphocyte antigen 4 (CTLA-4 or CD152).⁵⁰

FOXP3 is an essential transcription factor required to manifest the T_REG cell phenotype,⁵¹ but it does not function alone and requires the expression of additional transcription factors to define the T_REG cell phenotype and to establish its characteristic transcriptional programme.⁵² Patients with the immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX syndrome) are found to have germline mutations in the gene encoding FOXP3, which maps to the long arm of the X chromosome at Xp11.23 (Chap. 80).⁵³ T_REG developmental deficiency or dysfunction is a hallmark of IPEX, leading to severe, multiorgan, autoimmune phenomena. Patients typically have autoimmune skin conditions, such as bullous pemphigoid or alopecia universalis, and autoimmune endocrinopathies similar to those seen in patients with the autoimmune polyendocrine candidiasis ectodermal dystrophy syndrome (APECED syndrome), which is associated with genetic defects in the autoimmune regulator (AIRE) gene responsible for the generation of T-cell tolerance in the thymus (Chaps. 6 and 80).^54,55 The IPEX syndrome demonstrates the importance of T_REG cells in maintaining tolerance to self-antigens and in preventing runaway immune responses to environmental antigens that might evolve into cross-reactive autoimmunity.

Additional key factors and signals required for T_REG cell development and survival include IL-2, transforming growth factor-β (TGFβ) and co-stimulatory molecules.⁵⁶ It is also becoming clear that unique epigenetic changes that are partly induced by TCR signalling are typical of the T_REG cell lineage, and that they can be used to differentiate T_REG subpopulations such as thymus-derived T_REG (tT_REG) cells and peripherally derived T_REG (pT_REG) cells.^57,58 These terms have replaced “natural FOXP3+ T_REG cells” and “induced or adaptive T_REG cells” to more accurately describe the anatomical location of their differentiation. tT_REG cells differentiate from CD4+CD8^– T cells that have undergone positive and negative selection to self-antigens presented in the thymus (Chap. 6) and are thought to play a role in maintaining tolerance to self-antigens. pT_REG cells, on the other hand, differentiate upon antigen encounter under certain conditions and during normal homeostasis of the gut.⁵⁹ As such, pT_REGs are thought to play an important role in the development and maintenance of mucosal immune tolerance and in the control of severe chronic allergic inflammation and “altered” self-antigens of inflamed tissues or neoplastic cells. They are also required to minimize tissue damage in inflammatory settings such as viral infection⁶⁰ or mediate tolerance to allografts.⁶¹ tT_REG and pT_REG cells have subtle differences in the methylation status of conserved noncoding sequence 2 (CNS2; also known as the T_REG cell-specific demethylated region [TSDR]) in the FOXP3 locus, potentially influencing the stability of the cells under inflammatory or pathogenic conditions.⁶²

T_REG cells can specifically suppress immune responses via contact-dependent and cytokine-mediated mechanisms.⁶³ Upon activation through their TCR, T_REGs can (1) produce antiinflammatory cytokines (e.g., IL-10, TGF-β, or IL-35), (2) reduce the availability of IL-2 via absorption by the CD25 receptor, (3) lyse other immune effector cells via granzyme secretion or CD95-CD95L–mediated cell killing, (4) modulate the activation state and/or function of APCs and other immune effector cells, and/or (5) release suppressor factors, such as galectin-1⁶⁴ (a β-galactoside–binding protein that can bind and inhibit the function of many glycoproteins, including CD7, CD43, and CD45) and fibrinogen-like protein 2 (FGL2; a member of the fibrinogen family that mediates it suppressive effect through binding to low affinity Fcγ receptors expressed on APCs).⁶⁵ Consequently, T_REG cells can act directly against specific target antigens, while activating suppressive functions of other types of immune effector cells such as CD4+ and CD8+ cells.

Th17 T Cells

Naïve CD4+ T cells also can differentiate into Th17 T cells that play an important role in immune responses to certain extracellular pathogens and fungi.⁶⁶ These cells produce IL-17 (sometimes referred to as IL-17A) and a closely related cytokine, IL-17F, which can form biologically active homodimers or heterodimers and induce substantial tissue reactions because of the broad distribution of the IL-17 and IL-22 receptors. Principal cytokines involved in the differentiation of naïve blood CD4+ T cells into Th17 cells are IL-23 and IL-1β (see Fig. 76–3), but a combination of TCR stimulation and the cytokines TGF-β and IL-6 are also required.⁶⁷ Prostaglandins, most notably prostaglandin E₂, can synergize with IL-23 and IL-1β to drive differentiation of CD4+ T cells into Th17 cells.⁶⁸ These cytokines and factors can induce activation and/or expression of transcription factors that are distinct from those used by Th1 or Th2 cells, including the retinoic acid-related orphan receptor γ (RORγt) and STAT3 (see Fig. 76–3),^69,70 which, in turn, can induce expression of IL-17 and IL-17F.^71,72 However, for full commitment of precursors to the Th17 lineage, RORγt and STAT3 must act in cooperation with other transcription factors, including RORα, interferon regulatory factor 4 (IRF4), and runt-related transcription factor 1 (RUNX1). Th17 cells also express high levels of the IL-23R, CCR4, CCR6, CXCR4, CD161, and multiple CD49 integrins, but not CCR2, CCR5, or CCR7.^37,73,74 In contrast to Th1 or Th2 cells, Th17 cells do not elaborate IFN-γ or IL-4, both of which can inhibit expression of IL-17.⁷⁵

Th17 cells play a central role in inflammation and defense against intestinal bacteria, extracellular pathogens, and fungal infections, mostly via activation of neutrophils. Common pathogens that induce mainly Th17 responses include Gram-positive Propionibacterium acnes, gram-negative Citrobacter, Klebsiella pneumoniae, bacteroides and Borrelia species, and fungi such as Candida albicans.⁷⁶ Th17 cells are abundantly found in the intestinal lamina propria, where they are induced and stimulated by commensal bacteria, maintain epithelial integrity, and clear extracellular pathogens.⁷⁷

Th17 cells are the principal producers of IL-17 in response to specific immune stimulation, but NK and natural killer T (NKT) cells are also able to produce IL-17. IL-17 is a proinflammatory cytokine that has pleiotropic effects on multiple target cells, resulting in enhanced antigen presentation, antibody production, macrophage activation, cellular extravasation, and neutrophil migration.⁷⁸ In addition to IL-17 and IL-17F, Th17 cells elaborate other proinflammatory factors, including chemokines (e.g., CXCL8 [IL-8] and CCL20), cytokines (e.g., IL-6, tumor necrosis factor-α, IL-21, and IL-22), growth factors (e.g., granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor), acute phase proteins (e.g., C-reactive protein), and antimicrobial peptides and mucins.⁷⁶

The importance of Th17 cells in the defense against certain microorganisms is reflected in the rare primary immunodeficiency disorder called autosomal dominant hyper-IgE syndrome, in which a mutation in STAT3 abrogates Th17 differentiation (Chap. 80).^79,80 Patients lack Th17 cells and have an increased susceptibility to infection with the various species of Staphylococcus or Candida. Furthermore, the loss of intestinal commensal bacteria that are essential for the induction of Th17 cells through the use of antibiotics can cause depletion in intestinal Th17 cells, and might account in part for the increased incidence of gastrointestinal infections with C. albicans or Clostridium difficile observed in patients subjected to long-term, broad-spectrum antibiotic therapy.⁸¹ Because of their capacity to enhance inflammation in an antigen-specific manner, Th17 cells also have been implicated in the development and/or propagation of several autoimmune disease, such as rheumatoid arthritis,⁸² systemic lupus erythematosus (SLE),⁸³ MS,⁸⁴ inflammatory bowel disease,⁸⁵ glucocorticoid-resistant asthma,⁸⁶ and psoriasis.⁸⁷

T_FH Cells

T follicular helper cells (T_FH cells) constitute another subset of CD4+ T cells that regulates the development of antigen-specific B-cell immunity in the germinal center of secondary lymphoid follicles. T_FH cells express the CXCR5 chemokine receptor, allowing them to home to the CXCL13-rich B-cell zones of lymphoid follicles where they engage antigen-specific B cells in cognate intercellular interactions, and cell-surface proteins such as programmed cell death-1 (PD-1), inducible T-cell costimulator (iCOS), B- and T-lymphocyte attenuator (BTLA), and CD40L, which allow them to form stable contacts with antigen-primed B cells.⁸⁸ Such interactions play a critical role in B-cell differentiation into plasma cells or memory B cells in response to antigenic stimulation. In addition, T_FH cells secrete cytokines such as IL-4, IFN-γ, IL-10, and/or IL-21, which partly overlap with cytokines characteristic of other T-effector cells, helping to modify the differentiation fate of B lymphocytes (Chap. 75).

T_FH cells also express the cytoplasmic adaptor protein signal lymphocyte activation molecule (SLAM)-associated protein (SAP), required for lymphocyte interactions, and the transcription factor B-cell lymphoma 6 (BCL-6), required for T_FH cell differentiation. Several lines of evidence suggest that T_FH differentiation is a multistage process, but that dendritic cells (DCs) are crucial for CD4+ T cell priming and initial acquisition of T_FH cell characteristics, including the induction of BCL-6 expression.⁸⁹ The flexibility and plasticity of T_FH cells, which is mostly mediated by chromatin modifications, is underlined by the expression of a constellation of transcription factors, including BCL-6, BATF, STAT3, IRF4, c-Maf, and GATA-3, many of which are expressed by other Th effector subsets (see Fig. 76–3).

Th9 Cells

Another helper T-cell subset that has recently emerged are Th9 cells. In contrast to Th17 cells and T_REGs, they are induced by the combination of TGF-β and IL-4 and regulated by the transcription factors PU.1, STAT6, IRF4, and GATA-3.⁹⁰ They primarily produce IL-9, IL-10, and IL-21.⁹¹ Functionally, Th9 cells appear to be effector rather than regulatory, and they have been implicated in the development of allergic reactions, particularly in the lungs.⁹² Recent data indicates that Th9 cells in vivo might be implicated in tumor immunity in melanoma.^93,94

MEMORY T CELLS

Following a successful immune response to antigen including the exposure to and recognition of antigen, expansion of T-cell subsets and exertion of effector function, both naïve CD4+ and CD8+ T-cells can develop into long-lived memory T cells that provide enhanced protection from re-exposure to the same or a related pathogen.^95,96 Memory T-cell populations maintain the ability to both survive independently of cognate antigen⁹⁷ and self-renew in response to external homeostatic signals such as IL-15 and IL-7, hence maintaining themselves at stable levels for many years.⁹⁸ In addition, they have less-stringent requirements for activation and an enhanced capacity for lymphokine production upon rechallenge with the same antigen, and require a lower level of costimulatory factors.

Memory CD4+ or CD8+ T lymphocytes can also be distinguished from naïve cells based on their surface phenotypes of CD45 isoforms, rate of cycling, and migration. CD45, also known as leukocyte common antigen or T200, consists of a family of membrane glycoproteins, ranging from 180 to 220 kDa, that are expressed on all leukocytes.⁹⁹ Each member is the product of a single complex gene on chromosome 1 that contains 34 exons. Exons 3 through 7 may be spliced differently at the RNA transcript level to generate several distinct messenger RNA and protein products. The deduced amino acid sequences of these protein products have extracellular domains ranging from 391 to 552 amino acids, a transmembrane region, and a highly conserved cytoplasmic domain of 705 amino acids. This large cytoplasmic domain contains an intrinsic tyrosine phosphatase activity that is important in the regulation of various activation pathways involving tyrosine kinase activity, such as those involved in signal transduction via the TCR for antigen.¹⁰⁰ Different isoforms of CD45, designated as CD45R, have distinct expression patterns during lymphocyte ontogeny and activation. Therefore, cell subsets can be readily characterized by flow cytometry approaches using specific monoclonal antibodies. Naïve CD4+ T cells express CD45RA, whereas memory CD4+ T cells and CD8+ T cells express CD45RO. CD45RB can also be useful for distinguishing memory T cells. Within the CD4+ memory T-cell population, for example, there is an increase of helper activity associated with the shift from a CD45RB^bright to a CD45RB^dim phenotype.¹⁰¹

In addition, the distinct expression of chemokine receptors (CCRs) or homing molecules can be used to characterize T-cell subsets: relative to naïve T cells, memory T cells express lower levels of L-selectin (CD62L) and higher levels of CD29 and CD44.¹⁰² More recent studies demonstrate that memory cells can be further subdivided into CD44+CD62L+CCR7+ central memory T cells (T_CM cells) and CD44+CD62L−CCR7− effector memory T cells (T_EM cells).¹⁰³ Because of their constitutive expression of CCR7 and CD62L, T_CM cells home to secondary lymphoid organs, where they have little or no immediate effector function, but show greater sensitivity to antigenic stimulation in comparison to naïve T cells, are less dependent on costimulation, and upregulate CD40L to a greater extent. However, upon recognition of their cognate antigen, they have a high proliferative potential and can rapidly differentiate into large numbers of effector cells. T_EM cells, in contrast, have higher migratory potential and display immediate effector function. Therefore, T_CM cells are predominantly found in the CD4 lineage and are enriched in lymph nodes and tonsils, whereas T_EM cells are more frequent in the CD8 compartment in lung, liver, and intestines. Accordingly, CD8+ T_EM cells carry large amounts of perforin, and both CD4+ and CD8+ T_EM cells can produce IFN-γ, IL-4, and IL-5 within hours after following antigenic stimulation.

It has long been controversial whether memory T cells arise during the contraction phase and develop directly from effector cells, or whether they diverge early during an immune response, and arise in parallel with short-lived effector cells. Recent studies, however, have provided evidence for an early delineation of the effector versus memory T-cell fates regulated through specific transcription factors and cytokines, such as IL-7 receptor α-chain (IL-7R) expression, IL-2, IL-12, and T-BET, EOMES, and BLIMP-1.^104,105

IMMUNE MODULATORY MOLECULES

CD28

CD28 is a 44-kDa disulfide-linked homodimer that is expressed on most resting T cells and plasma cells. Mature thymocytes have higher levels of CD28 than immature cells. Among human peripheral T cells, more than 90 percent of CD4+ T cells and approximately 50 percent of CD8 T cells express CD28. In general, activation of T cells induces enhanced expression of CD28, but ligation of CD28 leads to its transient downregulation.¹⁰⁶ CD28 is another member of the immunoglobulin superfamily and binds to both CD80 and CD86 using a highly conserved motif (MYPPPY) in a loop that resembles the third complementarity-determining region of immunoglobulin molecules. CD28 binds to CD80 with relatively low affinity (dissociation constant [Kd] = 4 μM) and dissociates very rapidly (K_off = 1.6 s⁻¹),¹⁰⁷ and its binding to CD86 may be even weaker.¹⁰⁸

CD28 functions as a major costimulatory molecules that is essential for T-cell activation, probably in the context of TCR/CD28 microcluster formation.¹⁰⁹ Ligation of CD28 by CD80 or CD86 or by anti-CD28 antibodies activates distinct signaling pathways that function together with the signals induced by ligation of the TCR to allow for T-cell activation and proliferation.¹¹⁰ Following coligation of CD28, the Src kinases Lck and Fyn may phosphorylate a tyrosine within an ITAM found in the cytoplasmic domain of CD28, allowing the latter to bind and to activate phosphatidylinositide 3-kinase via its SH2 domains. CD28 signaling also facilitates GTP/guanosine diphosphate (GDP) exchange on Ras, resulting in activation of the MAP kinase pathway, activation of Akt kinase, and activation of the adapter protein Vav and the associated Rac pathway. These signals enhance the transcription of IL-2 and the stability of IL-2 transcripts, thereby stimulating T-cell proliferation.¹¹¹ Although mice lacking CD28 can mount effective T-cell responses, they are defective in T-cell–dependent antibody responses, suggesting that CD28 is necessary for T-cell–B-cell interactions and the proficient generation of antibody responses to antigen.¹¹²

The requirement for the same cell to present both the specific antigen and the costimulatory signal is crucial to prevent destructive autoimmune responses to self-tissues. This restricts the initiation of T-cell responses to APCs that express both the peptide antigen in the context of self-MHC molecules and the ligands for CD28, namely, CD80 and CD86. This is particularly important as not all self-reactive T cells undergo deletion in the thymus because not all self-peptides are presented in the thymus (Chap. 6). This is especially true for specialized tissues that express proteins that are never expressed in the thymus. If simultaneous ligation of the TCR and CD28 was not required, then T cells that recognize the self-peptide expressed by the MHC of such specialized tissues could become activated, leading to autoimmune rejection of the specialized tissue. Instead, ligation of the TCR in the absence of CD28 ligation leads to a state of hyporesponsiveness or anergy, in which the T cell expressing that receptor becomes refractory to activation.¹¹³ Anergized T-cell clones produce only negligible amounts of IL-2, which is crucial for clonal expansion following T-cell activation. Interestingly, clonal T-cell anergy was found not to be a terminal fate, as the addition of exogenous IL-2 during restimulation could reverse the phenotype.¹¹⁴ These characteristics are not limited to T cells, as anergic B cells also demonstrate some of the hallmark features, such as reduced proliferation and effector function. Anergy is an important basis for development of peripheral tolerance for self-antigens that are not expressed in the thymus (Chaps. 6 and 74).

CTLA-4 (CD152)

CTLA-4 (CD152) is another receptor for CD80 and CD86. It is a 50-kDa disulfide-linked homodimer that shares 31 percent identity with CD28. The gene encoding this receptor is closely linked with that encoding CD28 on the long arm of chromosome 2 at 2q33–q34. However, in contrast to the constitutive expression of CD28, T cells—with the exception of T_REGs—express CD152 only upon activation. Expression of CD152 peaks at approximately 24 hours after activation and then subsides by 72 hours but is always approximately 30- to 50-fold lower than that of CD28. CD28 ligation is particularly effective in inducing CD152.

CD152 binds to both CD80 and CD86 using the same highly conserved motif (MYPPPY) used by CD28, which, like CD28, is also in a loop that resembles the third complementarity-determining region of immunoglobulin molecules. However, despite its lower expression levels, CD152 binds to CD80 and CD86 approximately 20 times more avidly than CD28, with a Kd of 0.4 and 2.2 μM, respectively.^107,108 In contrast to CD28, ligation of CD152 transmits a negative signal to T-cell activation.¹¹⁰ Instead of an ITAM, CD152 possesses an immunoreceptor tyrosine inhibitory motif (ITIM) in its cytoplasmic domain. Ligation of CD152 induces tyrosine phosphorylation of the ITIM, which, in turn, recruits the tyrosine phosphatase SHP-2 that can deactivate the phosphorylated ITAMs of the ζ chains of the TCR complex. Mice made genetically deficient in CD152 develop a fatal lymphoproliferative disorder that is characterized by massive cell activation and infiltration into tissues, indicating that CD152 serves as an important brake on unregulated T-cell activation.¹¹⁵ Mice that are genetically modified to express only the extracellular domain of CD152 allowing the binding of CD80 or CD86, however, are protected from organ infiltration by T cells, suggesting that modulation of CD28 signals by competitive sequestration of its ligands can regulate tissue infiltration by autoreactive T cells.¹¹⁶

Moreover, anti-CD152 monoclonal antibodies that block the interaction of CD152 with CD80 and CD86 can enhance T-cell responses in vitro and in vivo. This has prompted their evaluation as immune-enhancing agents in models of autoimmune disease and in transplantation models,¹¹⁷ and more recently in clinical vaccine studies and trials in combination with the blockade of other immunomodulatory molecules.^118,119

Other Members of the CD28 Receptor Family

Homology-based cloning strategies have identified other proteins that are structurally related to CD28/CTLA-4 or its ligands CD80/CD86. These proteins are categorized as members of the CD28 or CD80 (B7) families, respectively, which belong to the immunoglobulin superfamily, a structurally and functionally highly heterogeneous family of T-cell cosignaling molecules.¹²⁰

Two other important members of the CD28 family are iCOS (CD278) and PDCD1 (CD279, PD-1). Whereas CD278 is found primarily on activated T cells, CD279 can be found on activated T cells, B cells, some myeloid cells, and NK cells. CD278 and CD279, respectively, bind to the iCOS-ligand (iCOS-L or CD275) and the PD-ligands PD-L1 (CD274) or PD-L2 (CD273).¹²¹ CD273, CD274, and CD275 belong to the CD80 (B7) family of surface molecules and are found or can be induced on B cells, APCs, and other nonhematopoietic tissues. CD278 primarily functions as a costimulatory molecule for cells bearing CD275.¹²² CD279 plays a negative regulatory role on activated T cells and modifies T cell/APC contact durations by inhibiting TCR stop signals.¹²¹ Like CD152, CD279 possesses an ITIM motif in its cytoplasmic tail that, upon phosphorylation, can recruit the tyrosine phosphatase SHP-2.¹²³ In this regard, CD279 may play a role similar to that of CD152, helping to brake cellular activation when bound to its ligands CD273 or CD274.

In general, the molecular mechanisms of T-cell costimulation and coinhibition are viewed as an evolving concept, as receptors and ligands exhibit great diversity in their expression, structure and function, and this is likely to be dependent on the context of anatomical location and the existence of underlying healthy and pathological immune responses.¹²⁰

T-CELL ADHESION MOLECULES

In addition to CD3/TCR molecules and CD4 or CD8, several other surface proteins are required for efficient T-cell antigen recognition.¹²⁴ Some of these surface proteins are termed adhesion molecules, as they facilitate the adhesion of the T cell to its appropriate APC or target cell (Fig. 76–4). Their main function is to permit the T-cell antigen receptor complex to interact better with the MHC glycoproteins of the other cell, allowing for efficient T-cell antigen recognition and activation. Each member of this group of accessory molecules has distinctive affinities for the surface molecules expressed by the APC or target cell, which is reflected by differential expression patterns on specific T-cell subsets. Adhesion and antigen response are largely considered the final step of the multi-stage T-cell trafficking process that involves migration, adhesion, and signaling.

Lymphocyte Function-Associated Glycoproteins

The lymphocyte function-associated (LFA) molecules are an important family of glycoproteins that facilitate efficient cell–cell adhesion.¹²⁴ The family consists of the three major surface molecules LFA-1, LFA-2 (CD2), and LFA-3 (CD58). LFA-1 belongs to a family of three related glycoproteins LFA-1, MAC-1 (CD11b/CD18), and gp150,95 (CD11c/CD18). These proteins also are called “integrins” because they are hypothesized to coordinate the binding of cells to other cell types and to extracellular proteins. Each protein consists of a distinct α subunit noncovalently associated with the β₂ subunit glycoprotein of 95 kDa, designated as CD18. Because they share a common β₂ subunit, these molecules also are referred to as the β₂ integrins. The α subunit of LFA-1, designated CD11a, is a 180-kDa glycoprotein. Coupled together with the common β₂ subunit, this 180-kDa molecule is expressed on more than one-third of all marrow cells, all T cells, all B cells, and all NK cells. The α subunit of MAC-1 is a glycoprotein of 170 kDa, designated CD11b. MAC-1 is expressed on NK cells, monocytes, macrophages, granulocytes, and small subpopulations of T and B cells. The α subunit of p150,95, designated CD11c, is a 150-kDa glycoprotein that is not expressed by T lymphocytes.

The shared β₂ subunit (CD18) has extensive sequence homology to the β₃ subunit (CD61) of the platelet adhesion receptor glycoprotein IIb/IIIa (CD41/CD61) and the β₁ subunit (CD29) of a family of related adhesion proteins, termed very-late-activation (VLA) antigens. Many of these receptors function in cell–cell interactions and recognize their ligands at sites that contain the amino acid sequence Arg-Gly-Asp. In addition, the α subunit provides some selectivity. LFA-1, because of its α subunit, binds best to cell-surface ligands called ICAMs, namely, ICAM-1 (CD54), ICAM-2 (CD102), and ICAM-3 (CD50). CD54 and CD102 are expressed on endothelial cells as well as APCs. The binding of LFA-1 on lymphocytes to these molecules allows lymphocytes to migrate through blood vessel walls. CD50 is expressed only on leukocytes, including T cells, and is thought to play an important role in the adhesion of T cells with LFA-1 expressed on APCs (see Fig. 76–4). The LFA glycoproteins are essential for proper T-cell function and host immunity, and integrin gene defects leading to leukocyte adhesion deficiencies result in severe immune deficiencies. In humans, three LADs (LAD I to III) have been described.¹²⁵ Although clinically distinct, they exhibit several common features including recurrent bacterial infections and leukocytosis.

LFA-1, CD2, and CD50 binding partners are CD54, CD102, LFA-1, and CD58 on the APC (see Fig. 76–4). Binding prolongs the time the T cell is exposed to antigen, and allows to sample large numbers of MHC molecules on the plasma membrane of the APC for the presence of specific peptide antigen. When a naïve T cell recognizes its specific peptide in the context of the MHC, signaling through the TCR induces a conformational change in LFA-1 that greatly increases its affinity for CD54 and CD102. This stabilizes the association between the antigen-specific T cell and the APC. This association can last for several days during which time the naïve T cell proliferates, forming daughter cells that also adhere to the APC and that differentiate into armed effector T cells.

Very-Late-Activation Antigens

VLAs received this terminology because the first identified VLA molecules, namely VLA-1 and VLA-2, initially were found on T cells only weeks after repetitive stimulation in vitro.¹²⁶ VLA molecules are β₁ integrins and share a common β₁ unit (CD29) that is paired with any one of six different α chains (α₁ to α₆), designated CD49a to CD49f. CD49a, CD49b, CD49c, CD49d, CD49e, and CD49f form molecules called VLA-1, VLA-2, VLA-3, VLA-4, VLA-5, and VLA-6, respectively, when paired with CD29. Despite their nomenclature, some of these VLA molecules, most notably VLA-4, are also expressed constitutively by some T cells and are rapidly induced on others. VLA-4 plays an important role in facilitating the attachment of cells that bear this molecule to the endothelium through its binding to vascular cell adhesion molecule-1 (VCAM-1), designated CD106. CD106 can be upregulated by various proinflammatory cytokines, allowing VLA-4 to play an important role in facilitating the homing of T cells to endothelium at sites of inflammation.

CD2

CD2 is a glycoprotein of approximately 50 kDa found on all T lymphocytes, large granular lymphocytes, and thymocytes.¹²⁷ CD2 facilitates cell–cell adhesion by binding to CD58, a 55- to 70-kDa surface glycoprotein that is expressed on erythrocytes and leukocytes as well as on endothelial, epithelial, and connective tissue cells (see Fig. 76–4).Monoclonal antibodies that bind CD2 may inhibit a variety of T-lymphocyte functions, including antigen-specific T-lymphocyte proliferative responses to lectins, alloantigens, and soluble antigens. Anti-CD2 inhibits cytotoxic T-lymphocyte–mediated cell killing by binding to the T cell rather than to the target, which generally does not express CD2. On the other hand, antibodies directed against CD58 inhibit cytotoxic T-lymphocyte–mediated cell killing by binding to CD58 on the target cell, thus blocking interaction of CD2 with CD58.

T-cell antigen recognition following the principles described above takes place within a contact zone termed the immunological synapse, which organizes the involved membrane proteins at the interface between the T cell and the APC. Immune synapse formation involves polymerization of F-actin and polarization of the cytoskeleton, resulting in the redistribution of TCRs, costimulation, and accessory molecules into the region beneath the T-cell: APC contact site.¹²⁸ These molecules are segregated into central (cSMAC), peripheral (pSMAC) and distal (dSMAC) supramolecular activation clusters (SMACs). The cSMAC contains a concentration of TCR:CD3:peptide:MHC complexes, costimulatory molecules such as CD28, and signaling molecules such as PKCθ, while the pSMAC is enriched with adhesion molecules such as LFA-1. The dSMAC contains large glycoproteins such as CD45.¹²⁹ In addition to priming of T-cell responses, the immunological synapse is also important for effector functions.¹³⁰ For example, the pSMAC has been suggested to polarize cytotoxic granules released by cytotoxic CD8 T cells toward the target cells, therefore preventing the leakage of cytolytic granule contents to bystander cells, and the cSMAC is implicated as a site for receptor internalization and degradation.