The proliferation and differentiation of naive T cells require signals provided by molecules on APCs, called costimulators, in addition to antigen-induced signals (Fig. 1). The requirement for costimulatory signals was first suggested by the experimental finding that T-cell antigen-receptor signaling alone (e.g., induced by anti-CD3 antibodies that cross-link TCR-CD3 complexes, mimicking antigens) resulted in lower responses than those seen with antigens presented by activated APCs. This result indicated that APCs express molecules that work together with antigen for inducing T-cell activation. These molecules were called costimulators, and the second signal for T-cell activation was called costimulation, the first signal being an antigen. In the absence of costimulation, T cells that encounter antigens may enter a state of prolonged inability to functionally respond or die.
Fig1. Functions of costimulators in T-cell activation. (A) The resting antigen-presenting cell (APC) (typically dendritic cells [DCs] presenting self antigens) expresses few or no costimulators and fails to activate naive T cells. Antigen recognition without costimulation may make T cells unresponsive (tolerant) or lead to death of the T cells; we will discuss this phenomenon in Chapter 15. (B) Microbes and cytokines produced during innate immune responses activate APCs to express costimulators, such as B7 molecules. The APCs (usually presenting microbial antigens) then become capable of activating naive T cells. Activated APCs also produce cytokines such as interleukin-12 (IL-12), which stimulate the differentiation of naive T cells into a subset of effector cells (not shown).
The B7:CD28 Family of Costimulators
The best characterized costimulatory pathway in T-cell activation involves the T-cell surface receptor CD28, which binds the costimulatory molecules B7-1 (CD80) and B7-2 (CD86) expressed on the surface of activated APCs. (The term costimulator is variably used for the receptor that delivers the activating signal, e.g., CD28, or the ligands that provide the stimulus, e.g., B7.) CD28 was discovered when stimulatory (agonistic) antibodies against human T-cell surface molecules were screened for their ability to enhance T-cell responses when added together with an activating anti-CD3 antibody. This was soon followed by the identification of the ligands for CD28, shown to be two homologous proteins, named B7-1 (CD80) and B7-2 (CD86), often collectively called B7. The role of CD28 and B7 in T-cell activation has been established by the T-cell immune deficiency caused by knockout of genes encoding these proteins in mice and by the ability of agents that bind to and block B7 molecules to inhibit T-cell responses in experimental animals and in humans. The development of therapeutic agents based on these principles is described later.
B7-1 and B7-2 are structurally similar integral membrane single-chain glycoproteins, each with two extracellular immunoglobulin (Ig)-like domains. CD28 is a disulfide-linked homodimer, each subunit of which has a single extracellular Ig domain. Its cytoplasmic portion contains several tyrosine and proline residues that are involved in binding of signaling proteins and in the delivery of activating signals (discussed later). CD28 is expressed on the vast majority of CD4+ T cells, but as people age, as many as 40% of blood CD8+ T cells, mainly effector and memory cells, lose CD28 expression. This loss of CD28 is believed to be the result of chronic antigen stimulation (e.g., in chronic viral infections) and may be a way that T cells limit their own activation by persistent antigens.
The expression of B7 costimulators is increased by microbial products and the innate immune responses to infections, which ensures that T lymphocytes are activated when needed (i.e., upon encountering microbes). The B7 molecules are expressed mainly on APCs, including DCs, macrophages, and B lymphocytes. They are expressed at low levels on resting APCs and are induced by various stimuli, including microbial products that engage Toll-like receptors and cytokines such as interferon-γ (IFN-γ) produced during innate immune reactions to microbes. The induction of costimulators by microbes and by the cytokines of innate immunity promotes T-cell responses to microbial antigens. This illustrates an important role of innate immune responses in enhancing adaptive immunity. In addition, activated CD4+ T cells themselves enhance the expression of B7 costimulators on the APCs by a pathway dependent on CD40, described later, and this positive feedback loop serves to amplify T-cell responses. Of all potential APCs, mature DCs express the highest levels of costimulators and this is one reason why they are the most potent activators of naive T cells.
In Chapter 6, we mentioned the essential role of adjuvants in inducing primary T-cell responses to protein antigens such as vaccines. Many adjuvants are products of microbes or mimic molecules produced by microbes and necrotic cells and thus elicit innate immune responses. One of the major functions of adjuvants in T-cell activation is to stimulate the expression of B7 costimulators on APCs.
Unactivated, or resting, APCs in normal tissues are capable of presenting self antigens to naive T cells, but because these tissue APCs express only low levels of costimulators, potentially self-reactive T cells that see the self antigens are not activated and may be rendered permanently unresponsive. Regulatory T cells, which are important for tolerance to self antigens, are also dependent on B7:CD28-mediated costimulation for their generation and maintenance. It is possible that the low levels of B7 costimulators that are constitutively expressed by resting APCs function together with the self antigens that are displayed by these APCs to maintain regulatory T cells.
CD28 signals work in cooperation with antigen recognition to promote the survival, proliferation, and differentiation of the antigen-specific T cells. Costimulatory signaling via CD28 amplifies signaling pathways that are also induced downstream of the TCR and may trigger additional signals that cooperate with TCR-induced signals (Fig. 2). PI3-kinase is recruited to the cytoplasmic tail of CD28, and activated. PIP3 generated by activated PI3-kinase recruits the kinase AKT, facilitating its activation. AKT, in turn, alters cellular metabolism and promotes cell survival. CD28 can also contribute to the activation of the JNK mitogen-activated protein (MAP) kinase via the RAC small G protein and can amplify the activation of nuclear factor κB (NF-κB). The net results of these signaling pathways in T cells are the increased expression of anti-apoptotic proteins, such as BCL-2 and BCL-XL , which promote cell survival; increased metabolic activity; enhanced proliferation; production of cytokines such as IL-2; and differentiation of the naive T cells into effector and memory cells.
Fig2. Mechanisms of T-cell costimulation by CD28. CD28 engagement by B7 induces several signals, some of which enhance T-cell receptor (TCR) signals and others that function together with TCR signals to stimulate the expression of survival proteins, cytokines, and cytokine receptors. This results in cell proliferation and differentiation toward effector and memory cells (see Chapters 10 and 11). DC, Dendritic cell; IL-2, interleukin-2.
Previously activated effector and memory T cells are less dependent on costimulation by the B7:CD28 pathway than are naive cells. This property of effector and memory cells enables them to respond to antigens presented by various APCs that may reside in nonlymphoid tissues and may express no or low levels of B7. For instance, the differentiation of CD8+ T cells into effector CTLs requires costimulation, but effector CTLs can respond to and kill other cells that do not express costimulators.
Other receptors homologous to CD28 and their ligands homologous to B7 have been identified, and these proteins regulate T-cell responses both positively and negatively (Fig. 3). ICOS (inducible costimulator, CD278), which is related to CD28, is expressed mainly on activated T cells. Its ligand, called ICOS-L (CD275), is expressed on DCs, B cells, and other cell populations. ICOS plays an essential role in T cell–dependent antibody responses, particularly in the germinal center reaction. It is required for the development and activation of T follicular helper cells, which are essential for the formation of germinal centers and for the generation of B cells that produce high-affinity antibodies. The inhibitory receptors of the CD28 family are discussed later.
Fig3. The major members of the B7 and CD28 families. The known B7 family ligands are expressed on antigen-presenting cells (APCs) (dendritic cells [DCs], macrophages, and B cells) and CD28 family receptors are expressed mainly on T cells. Different CD28 family members stimulate or inhibit different stages and types of T-cell responses. The functions of cytotoxic T-lymphocyte antigen–4 (CTLA-4) and programmed cell death protein–1 (PD-1) are discussed in Chapter 15, and the role of inducible costimulator (ICOS) in the generation and function of T follicular helper cells is discussed in Chapter 12. Other widely distributed molecules with limited homology to B7, such as B7-H3 and B7-H4, have been identified, but their physiologic roles are not yet established. Other inhibitory receptors have also been identified, such as BTLA, TIM-3, and TIGIT, but these are not homologous to CD28 and are not shown. Tfh, T follicular helper.
Other Costimulatory Pathways
Numerous other T-cell surface molecules have been shown to deliver costimulatory signals in vitro, but their physiologic role in promoting T-cell activation is less clear than that of the CD28 family. Several putative costimulatory receptors belong to the large tumor necrosis factor receptor (TNFR) superfamily, and their ligands are members of the TNF family. Many of these receptors are expressed on effector T cells and regulatory T cells and have been shown to stimulate or to inhibit immune responses under various experimental conditions. OX40 (CD134) is a TNFR family member, expressed on activated CD4+ and CD8+ T cells, that functions to maintain cell survival and sustained responses. Its ligand, OX40L, is expressed on activated APCs. 4-1BB (CD137) and CD27 are two other TNFR superfamily molecules that are expressed on activated and memory T cells as well as regulatory T cells; their roles in regulating immune responses are not well defined.
The interaction of CD40L on T cells with CD40 on APCs enhances T-cell responses by activating the APCs. CD40 ligand (CD40L) is a TNF superfamily membrane protein that is expressed primarily on activated T cells, and CD40 is a member of the TNFR superfamily expressed on B cells, macrophages, and DCs. The functions of CD40 in activating macrophages in cell-mediated immunity and activating B cells in humoral immune responses are described in Chapters 10 and 12, respectively. Activated helper T cells express CD40L, which engages CD40 on the APCs and activates the APCs, making them more potent by enhancing their expression of B7 molecules and cytokines such as IL-12 that promote T-cell differentiation (Fig. 4). This phenomenon is sometimes called licensing, meaning that activated T cells license APCs to become more powerful stimulators of immune responses. Thus, the CD40 pathway indirectly amplifies T-cell responses by inducing costimulators on APCs, but CD40L does not by itself function as a costimulator for T cells.
Fig4. Role of CD40 in T-cell activation. Antigen recognition by T cells induces the expression of CD40 ligand (CD40L) on the activated T cells. CD40L engages CD40 on antigen-presenting cells (APCs) and may stimulate the expression of more B7 molecules and the secretion of cytokines that activate T cells. Thus, CD40L on the T cells makes the APCs better at promoting and amplifying T-cell activation. DC, Dendritic cell.
Therapeutic Targeting of Costimulators
Based on the understanding of costimulatory pathways, therapeutic agents have been developed for controlling injurious immune responses by inhibiting costimulation, called costimulatory blockade (Fig. 5). CTLA-4-Ig, a fusion protein consisting of the extracellular domain of CTLA-4 (discussed later) and the Fc portion of human IgG, binds to B7-1 and B7-2 and blocks the B7:CD28 interaction. The reason for using the extracellular domain of CTLA-4 rather than of CD28 to bind to and block B7 molecules is that CTLA-4 has a higher affinity for B7s than does CD28. Attachment of the Fc portion of IgG increases the in vivo half-life of the protein. CTLA-4-Ig is an approved therapy for rheumatoid arthritis and transplant rejection. Inhibitors of the CD40L:CD40 pathway are in clinical trials for transplant rejection and autoimmune diseases.
Fig5. The mechanism of therapeutic costimulatory blockade. (A) The normal T-cell response induced by antigen recognition and costimulation mediated by B7-CD28. (B) A fusion protein consisting of the extra cellular portion of cytotoxic T-lymphocyte antigen–4 (CTLA-4) and the Fc tail of an immunoglobulin G (IgG) molecule is used to bind to and block B7 molecules, thus preventing their interaction with the activating receptor CD28 and inhibiting T-cell activation. DC, Dendritic cell.
CD40 also plays an important role in T-dependent activation of B cells and macrophages, and patients with mutations in CD40L or CD40 genes have serious immune deficiencies.
Regulation of T-Cell Responses by Inhibitory Receptors
The outcome of antigen recognition by T cells is determined by a balance between engagement of activating and inhibitory receptors at the time of TCR recognition of the antigen. Although many inhibitory receptors have been described, the two whose physiologic role in self-tolerance is best defined are members of the CD28 family, called CTLA-4 and PD-1. These receptors have been called coinhibitors, to contrast them with costimulators. Studies of these inhibitory receptors have led to new therapeutic approaches for manipulating immune responses.
CTLA-4. CTLA-4 (cytotoxic T-lymphocyte antigen–4, so named because of how it was discovered, also called CD152) is homologous to CD28 (see Fig. 3) and, like CD28, it binds to B7 molecules.
CTLA-4 functions as a competitive inhibitor of CD28 and reduces the availability of B7 for the CD28 receptor (Fig. 6). CTLA-4 is expressed constitutively at high levels on Tregs and transiently on recently activated T cells, and it prevents the activation of responding T cells. It has an unusual mechanism of action. CTLA-4 on one T cell (e.g., a Treg) can inhibit responses of other T cells. Recall that CD28 and CTLA-4 recognize the same ligands, B7-1 (CD80) and B7-2 (CD86) (see Fig. 3). CTLA-4 has a 10- to 20-fold higher affinity for B7 than does CD28. The cytoplasmic tail of CTLA-4 does not appear to have any signaling function; instead, it contains a motif that connects it to clathrin, a protein involved in receptor-mediated endo cytosis. Thus, CTLA-4 is an endocytic receptor. It binds to B7 molecules on APCs, removes them, and promotes their ingestion into the T cells, a process that has been called transendocytosis (in which one cell endocytoses a protein from another cell). CTLA-4 also binds and blocks access to B7 molecules on APCs. Therefore, when CTLA-4 is expressed on either Tregs or activated T cells, it out-competes CD28 and reduces the amount of B7 available on the APCs to provide costimulation via CD28. The competitive inhibition and removal of B7 are especially important when B7 levels on APCs are low (particularly on resting APCs displaying self antigens and tumor antigens). Thus, CTLA-4 is efficient at inhibiting responses to these types of anti gens. When B7 levels increase, for example, after exposure to microbes, there is relatively more engagement of the low-affinity receptor CD28, allowing the generation of effective immune responses.
Fig6. Mechanism of action of cytotoxic T-lymphocyte antigen–4 (CTLA-4). CTLA-4 on regulatory or activated T cells binds to B7 molecules on antigen-presenting cells (APCs) and removes these molecules from the surface of the APCs, making the B7 costimulators unavailable to CD28 and blocking T-cell activation. This action of CTLA-4 is able to suppress immune responses best when B7 levels are low, enabling CTLA-4 to out-compete the lower-affinity receptor CD28. Treg, Regulatory T cell.
Because CTLA-4 limits the initial, costimulation-dependent activation of T cells in secondary lymphoid organs, mutating or blocking this receptor leads to severely dysregulated immune responses. This is illustrated by the finding that knockout mice lacking CTLA-4 and people with a loss-of-function mutation in the CTLA4 gene develop inflammatory lesions containing activated T cells and macrophages affecting multiple organs, with an as yet unexplained antibody deficiency in some patients. This is an example of a primary immune regulatory disease. Mutations of even one allele of CTLA4 cause this lymphoproliferative disease. It is thought that haploinsufficiency reduces the level of expressed CTLA-4 enough that it cannot compete effectively with CD28. Mutations affecting a protein called LRBA that is involved in the endocytosis and recycling of CTLA-4 cause a similar systemic inflammatory dis ease, further emphasizing the critical role of CTLA-4 trafficking in its inhibitory function. Because these diseases are caused by excessive B7-CD28–mediated costimulation resulting from failure of CTLA-4–mediated competition, they can be treated by blocking B7 molecules with the drug CTLA-4-Ig (see Fig.5).
PD-1. Another inhibitory receptor of the CD28 family is PD-1 (programmed cell death protein–1, so called because it was mistakenly thought to be involved in programmed cell death, also called CD279). It is expressed on activated T cells (and a few other cell types) and recognizes two ligands, called PD-L1 and PD-L2; PD-L1 is expressed on APCs and many other tissue cells, and PD-L2 is expressed mainly on APCs.
PD-1 inhibits signals from the TCR complex and CD28 and thus inhibits T-cell activation. Engagement of PD-1 by either of its ligands leads to the phosphorylation of an immunoreceptor tyrosine-based inhibitory motif (ITIM) and a switch motif (ITSM) in the cytoplasmic tail. These motifs bind the tyrosine phosphatase SHP2, which removes phosphates from various tyrosine-phosphorylated substrates (Fig. 7). Thus, PD-1 counteracts kinase-dependent signals from the TCR-coreceptor complex and from CD28 and other costimulatory receptors, resulting in inactivation of the T cells. Rare cases of PD-1 deficiency develop pulmonary autoimmunity. Both alleles of PD1 have to be mutated to develop this disease, as is typical of signaling molecules. PD-1 expression on T cells increases with antigen stimulation, so it is especially important for controlling responses to prolonged antigen exposure, as with self antigens, tumors, and chronic infections.
Fig7. Mechanism of action of programmed cell death protein–1 (PD-1). Engagement of its ligands, most often PD-L1, by PD-1 on T cells recruits a receptor-associated phosphatase, SHP2, to an immunoreceptor tyrosine-based inhibitory motif (ITIM) or immunoreceptor tyro sine-based switch motif (ITSM) in the cytoplasmic tail of PD-1, which removes phosphates from other proteins and inhibits kinase-dependent signals from CD28 and the T-cell receptor (TCR) complex. Note that only some signals from the TCR complex and CD28 are shown. APC, Antigen-presenting cell; ITAM, immunoreceptor tyrosine-based activation motif.
Although both CTLA-4 and PD-1 control immune responses, their roles may be complementary and nonoverlapping (Table 1). For example, PD-1 appears to be most important for terminating the responses of effector T cells, especially CD8+ cells, in peripheral tissues, whereas CTLA-4, as discussed previously, limits the initial activation of T cells in secondary lymphoid organs. Competitive inhibition of CD28 by CTLA-4 is a major mechanism that regulatory T cells use to prevent responses to self antigens. PD-1 limits responses to chronic antigen stimulation, for example, against microbes, especially viruses, and thus limits immunopathology associated with these infections. PD-1 also prevents responses to self antigens. Tumors have evolved to exploit this pathway to prevent their own destruction.
Table1. Actions and Functions of CTLA-4 and PD-1
Antibodies that block the CTLA-4 and PD-1 inhibitory receptors are approved for the immunotherapy of tumors. They work by preventing CTLA-4 or PD-1 from binding their ligands, thereby reducing inhibition and thus enhancing T-cell activation and enabling the cancer-bearing individual to mount more effective antitumor immune responses. Because these inhibitory receptors impose checkpoints on immune responses, blocking them therapeutically to enhance immune responses is called checkpoint blockade. As one might predict from the role of these inhibitory receptors in maintaining self-tolerance, blocking them for cancer immunotherapy induces autoimmune reactions in many patients. Conversely, antibodies that activate PD-1 (so-called agonists) are being tested to inhibit immune responses in autoimmune diseases.
Several other inhibitory receptors have been identified, including some belonging to the TNF receptor family and others to the T-cell immunoglobulin and mucin (TIM) family. The significance and biologic roles of these receptors are not as well established as the mechanisms of action and functions of CTLA-4 and PD-1. There is great interest in defining the roles of these receptors in the regulation of immune responses and in targeting these molecules therapeutically, especially for cancer immunotherapy.
