Interactions with antigen-presenting cells direct T cell activation

The way in which a T cell first encounters antigen largely dictates how it will react subsequently. A wide spectrum of cells can present antigen, depending on how and where the antigen is first encountered by cells of the immune system. In a lymphoid organ, the three main types of APC are:

Fig. 8.2 Localization of antigen-presenting cells (APCs) in lymph nodes

A lymph node represented schematically showing afferent and efferent lymphatics, follicles, the outer cortical B cell area, and the paracortical T cell area. Different APCs predominate in these areas and selectively take up different types of antigen, which then persist on the surface of the cells for variable periods. Polysaccharides are preferentially taken up by marginal zone macrophages and may persist for months or years, whereas antigens on recirculating macrophages in the medulla may last for only a few days or weeks. The recirculating ‘veiled’ cells (Langerhans’ cells and dermal dendritic cells), which originally come from the skin, change their morphology to become interdigitating dendritic cells within the lymph node. Both these cells and the follicular dendritic cells have long processes, which are in intimate contact with lymphocytes.

Activation of naive T cells on first encounter with antigen on the surface of an APC is called priming, to distinguish it from the responses of effector T cells to antigen on the surface of their target cells and the responses of primed memory T cells.

Dendritic cells are crucial for priming T cells

Dendritic cells which are found in abundance in the T cell areas of lymph nodes and spleen, are the most effective cells for the initial activation of naive T cells. They pick up antigens in peripheral tissues, then migrate to lymph nodes, where they express high levels of adhesion and costimulatory molecules, as well as MHC class II molecules, which allow them to interact with CD4⁺ TH cells.

Once they have migrated, DCs stop synthesizing MHC class II molecules, but maintain high levels of MHC class II molecules containing peptides from antigens derived from the tissue where they originated. Interdigitating DCs are believed to be the major APCs involved in primary immune responses because they induce T cell proliferation more effectively than any other APC.

The majority of dendritic cells enter lymph nodes via afferent lymphatics. Originally it was thought that these cells were mostly derived from Langerhans’ cells in the skin, but it now appears that a substantial proportion of the early migrating DCs in afferent lymph are dermal dendritic cells (they do not express langerin (CD207) a marker of Langerhans’ cells). Moreover DCs derived from Langerhans’ cells tend to localize in the paracortex of the lymph node, whereas dermal DCs remain near lymphoid follicles. Dendritic cells arriving from the periphery of the body transport antigen to the lymph node and process it for presentation to T cells. A minor proportion of the DCs in lymph nodes, arrive from the blood across the HEV, using the same route as T cells and B cells (see Fig. 6.15), however these cells have not acquired antigen in the periphery and they can only acquire it from lymph or transfer from other cells. As they mature, dendritic cells express CCR7 which allows them to localize to the lymphoid tissues. There is also some evidence that DCs from skin and the gut have distinctive chemokine receptors, which allow them to selectively recirculate to their own lymphoid organs. As they mature, DCs also increase expression of key costimulatory molecules, including CD40, CD80 and CD86 (B7-1 and B7-2).

Macrophages and B cells present antigen to primed T cells

Macrophages and B cells are less effective than DCs at antigen presentation to naive T cells, partly because they only express appropriate costimulatory molecules upon infection or contact with microbial products. Although they migrate to lymph nodes, the numbers of macrophages in afferent lymph is relatively few by comparison with DCs and this too limits their effectiveness in activating naive T cells. Macrophages:

B cells can:

If antigen concentrations are very low, B cells with high-affinity antigen receptors (IgM or IgD) are the most effective APC because other APCs simply cannot capture enough antigen. Therefore, for secondary responses, when the number of antigen-specific B cells is high, B cells may be a major APC.

The properties and functions of some APCs are summarized in Figures 8.3 and 8.4.

Fig. 8.3 Antigen presentation

Mononuclear phagocytes (1), B cells (2), and DCs (3) can all present antigen to MHC class II restricted TH cells. Macrophages take up bacteria or particulate antigen via non-specific receptors or as immune complexes, process it, and return fragments to the cell surface in association with MHC class II molecules. Activated B cells can take up antigen via their surface immunoglobulin and present it to T cells associated with their MHC class II molecules. DCs constitutively express MHC class II molecules and take up antigen by pinocytosis.

Fig. 8.4 Antigen-presenting cells

Many APCs are unable to phagocytose antigen, but can take it up in other ways, such as by pinocytosis. Endothelial cells (not normally considered to be APCs) that have been induced to express MHC class II molecules by interferon-γ (IFNγ) are also capable of acting as APCs, as are some epithelial cells. Another example is the thyroid follicular cell, which acts as an APC in the pathogenesis of Graves’ autoimmune thyroiditis.

Antigens are partially degraded before binding to MHC molecules

The processing of antigens to generate peptides that can bind to MHC class II molecules occurs in intracellular organelles. Phagosomes containing endocytosed proteins fuse with lysosomes where a number of proteases are involved in breaking down the proteins to smaller fragments. The proteases include:

Alkaline agents such as chloroquine or ammonium chloride diminish the activity of proteases in the phagolysosomes and therefore interfere with antigen processing.

In laboratory studies, the requirement for internal degradation of antigen by APCs can be circumvented by the use of synthetic peptides. This ability to use synthesized peptides of known sequences has enabled researchers to readily identify epitopes recognized by T cells with different specificities.

The relative importance of different amino acids within a defined epitope can also be investigated by amino acid replacements at different sites to identify residues that bind to the MHC molecule and those that bind to the TCR.

Proteasomes are cytoplasmic organelles that degrade cytoplasmic proteins

Although the assembly of MHC class I molecules occurs in the ER of the cell, peptides destined to be presented by MHC class I molecules are generated from cytosolic proteins. The initial step in this process involves an organelle called the proteasome – a multi-protein complex which forms a barrel-like structure (Fig. 8.6).

Fig. 8.6 Generation of immunoproteasomes by replacement of active subunits

The 20 S proteasome, shown in cartoon form, is composed of four stacked disks, two identical outer disks of α subunits, and two similar inner disks comprised of β subunits. Each disk has seven different subunits. Peptides enter into the body of the proteasome for cleavage into peptides. Only three of the β subunits are active. In normal proteasomes, these are called MB1, delta, and Z (1). Interferon-γ treatment of cells results in replacement of these three subunits by the two MHC-encoded proteins, PSMB8 and PSMB9, as well as a third inducible protein, PSMB10 (2). These subunits are shown adjacent to each other here, whereas they are actually in separate parts of the β ring and some would be hidden at the back of the structure shown.

Proteasomes provide the major proteolytic activity of the cytosol. They have a range of different endopeptidase activities and they degrade denatured or ubiquitinated proteins to peptides of about 5–15 amino acids (ubiquitin is a protein that tags other proteins for degradation).

Two genes, PSMB8 and PSMB9 located in the class II region of the MHC (Fig. 8.7), encode proteasome components that subtly modify the range of peptides produced by proteasomes. The expression of these genes is induced by interferon-γ (IFNγ). The proteins displace constitutive subunits of the proteasome and along with a third inducible proteasome component (PSMB10 encoded on a different chromosome) influence processing of peptides by creating a wider range of peptide fragments suitable for binding MHC class I molecules. Additional subunits associate with the ends of core (20 S) proteasomes and may influence antigen processing. These include interferon-inducible PA28 (proteasome-activator-28) molecules as well as a complex of proteins that result in a larger 26 S particle.

Fig. 8.7 MHC genes involved in antigen processing and presentation

Genes encoding the two subunits of the peptide transporter (TAP) and two components (PSMB8 and 9) of the multisubunit proteasome (see Fig. 8.6) are located in the murine and human class II regions. The Tapasin gene is located just centromeric of the MHC.

Proteasomes may not be the only proteases involved in producing peptides for presentation by MHC class I molecules. There is evidence for the involvement of enzymes, such as the giant tripeptidyl aminopeptidase II (TPPII) complex.

Transporters move peptides to the ER

The products of two genes, TAP1 and TAP2, that map in the MHC (see Fig. 8.7), function as a heterodimeric transporter that translocates peptides into the lumen of the ER. TAP is a member of the large ATP-binding cassette (ABC) family of transporters localized in the ER membrane. Microsomes from cells lacking TAP1 or TAP2 could not take up peptide in experiments in vitro. Using a similar system it was shown that the most efficient transport occurred with peptide substrates of 8–15 amino acids. Although this size is close to the length preference of MHC class I molecule binding sites, it suggests that some additional trimming may be required by enzymes in the lumen of the ER, particularly ERAAP (ER-associated aminopeptidase).

A multi-component complex loads peptides onto MHC class-I molecules

MHC class-I α-chains are initially associated with the chaperone, calnexin. Once released from calnexin they bind to β2-microglobulin to form a complete MHC class-I molecule, which is incorporated into a peptide-loading complex associated with the TAP transporters (Fig. 8.8). MHC class I molecule complexes lacking peptide are unstable, ensuring that only functionally useful complexes are available for interaction with TCRs.

Fig. 8.8 Peptide loading onto MHC class-I molecules

MHC class-I alpha chains (α) are initially held in the endoplasmic reticulum associated with the chaperone protein calnexin. After combining with β2-microglobulin, to form a complete MHC class-I molecule, they are released from calnexin and join a peptide loading complex, consisting of tapasin (Tps), ER-protein-p57 and calreticulin (CRT). Tapasin also associates with the TAP transporters (I and II). Proteins are degraded in the cytosol by proteasomes to produce polypeptides which are transported into the ER and loaded onto the MHC class-I molecule. The peptides may be trimmed by ER-associated aminopeptidases (ERAAP). The MHC class-I molecule with a bound peptide is finally released from the peptide loading complex, to be transported to the plasma membrane.

Antigen processing affects which peptides are presented

Originally it was thought that the MHC haplotype of an individual largely controlled which sets of antigenic peptides would be presented to T cells. We now know that antigen processing is at least as important: The availability of peptides to load onto MHC class-I molecules in the ER is dependent on:

Each one of these factors also depend on the amino-acid sequence of the original protein (Fig. 8.9), and to some extent on genetic variations in molecules involved in antigen processing (Fig. 8.10). All of these considerations are important in developing vaccines, where the aim is to identify an immunodominant region of a pathogen to stimulate T cells; but it is not enough for a peptide to bind to MHC molecules, it must also be processed properly if it is to be immunogenic.

Fig 8.9 Production of antigenic peptides

Production of antigenic peptides by the proteasome and their transport to the ER varies depending on the amino acid sequence of the epitope. The upper diagram shows the amino acid sequence of an immunodominant epitope of HIV matrix protein p17, and 5 mutants that often emerge during HIV infection. The sequence is shown in single amino acid code, and the variants indicate single, double or triple mutations in the wild-type sequence (wt). The lower diagram shows how efficiently these different mutants are digested by proteasomes and transported to the ER by binding to TAP transporters. The bars show the proportion of peptides generated which are presented by different MHC molecules (A11, B8, etc.). For example the KFV mutant produces peptides (red, recognized by HLA-A29, -A30 and -B44) that have a high affinity for the transporters and are transported well to the ER. Conversely the KFV mutant generates some HLA-11-binding peptides (dark blue) but they are not well transported, compared with peptides of the wild-type sequence.

(Based on data from Tenzer et al., 2009, Nature Immunology, 10, 636–646.)

Fig. 8.10 Linkage of TAP and class I alleles

Different MHC class I molecules in rats can accommodate peptides (blue bars) with either a positive charge at the C terminus (+) or a neutral amino acid (o). Similarly, TAP molecules (orange) come in two forms, which differ in the types of peptide they preferentially transport into the ER. Most rat strains have the appropriate TAP allele on the same haplotype as the class I gene that it serves best.

Cross-presentation can occur if exogenous antigen is presented on class-I molecules

Cross-presentation is a phenomenon which may occur when exogenous peptides are presented by MHC class I molecules. This is an exception to the principle that endogenously synthesized proteins are the source of peptides for MHC class I, and its mechanism and physiological importance is debated. However it could allow antigen-presenting cells to present viral antigens on their own MHC class-I molecules and activate TC cells, even if they had not themselves been infected.

Some class I-like molecules can present limited sets of antigens

In addition to the standard MHC class I molecules (class Ia), a number of class I-like molecules (class Ib), encoded in the MHC or elsewhere on the genome, can present very limited sets of antigens.

Why are antigen-processing genes located in the MHC?

The finding of a cassette of antigen-processing genes such as PSMBs and TAPs in the class II region of the MHC is striking. There is some evidence, especially from studies in rats, that particular alleles of TAP are genetically linked with alleles of class I genes that are most suited to receive the kind of peptides preferentially transported by the products of that TAP allele (see Fig. 8.10). The clustering of antigen processing and presenting genes in the MHC of most vertebrate species may not be fortuitous. It may help to coordinate co-evolution of some molecules as well as facilitating exchange of sequences between loci.

Class II molecules are loaded with exogenous peptides

MHC class II molecules are produced in the ER, complexed to a polypeptide called the invariant chain (Ii) (encoded outside the MHC), which stabilizes the complex and prevents the inappropriate binding of antigen. The αβ–Ii complex is transported from the Golgi to an antigen processing compartment which appears as a multivesicular body (also called the MIIC compartment), specialized for the transport and loading of MHC class II molecules. The compartment has characteristics of both endosomes and lysosomes with an onion-skin appearance under the electron microscope, comprising multiple membrane structures (Fig. 8.11). The αβ complex spends 1–3 hours in this compartment before reaching the cell surface (Fig. 8.12).

Fig. 8.11 MHC class II molecule processing compartment

Electron micrograph of ultrathin cryosections from B cells showing a multilaminar MIIC vesicle. The bar represents 100 nm. MHC class II molecules are revealed by antibodies coupled to 10 nm gold particles and HLA-DM by large gold particles (15 nm).

(Courtesy of Dr Monique Kleijmeer.)

Fig. 8.12 Pathways of MHC class-II protein traffic

Newly synthesized MHC class II molecules in the trans-Golgi network are associated with a protein Ii which prevents loading with peptide and contains sequences that enable the MHC class II molecule to exit the ER. Class II molecules move to an acidic endosome compartment, either via the cell surface or directly, where they encounter antigen that has been internalized and partly degraded into peptides. The endosomes form multivesicular bodies (MIIC) where the Ii protein is replaced by an antigenic peptide. Tubular vesicles bud from the multivesicular bodies carrying MHC class-II molecules loaded with peptide to the cell surface. These cell surface complexes may subsequently recycle through endosomes.

Exogenous antigens reach the MIIC compartment from acidic endosomes, where they have been partly degraded by the actions of proteases and chaperone proteins. The Ii chain is cleaved by cathepsins into small fragments, one of which, termed CLIP (class II-associated invariant peptide), is located in the groove of the class II molecule until replaced by peptides destined for presentation. The exchange of CLIP for other peptides is orchestrated by HLA-DM, an MHC class-II-like chaperone protein, consisting of α and β chains both encoded within the MHC class II region, but which does not itself have a peptide binding site (Fig. 8.13). HLA-DM binds to the αβ-CLIP complex to stabilize it until it has bound a suitable antigenic peptide. In cell lines lacking HLA-DM, the class II molecules are unstable and the cells no longer process and present proteins. Their class II molecules end up at the cell surface occupied by CLIP fragments of the invariant chain.

Fig. 8.13 HLA-DM acts like a catalyst to influence binding of peptides in exchange for CLIP

An MHC class II molecule is shown loaded with the Ii chain. The Ii chain is cleaved to the CLIP fragment. Association of the complex with HLA-DM then allows CLIP to be exchanged for other peptides derived from endocytosed proteins present in MIIC vesicles.

A further MHC-encoded molecule, HLA-DO (see Fig. 8.7), which associates with DM, regulates peptide loading. Like conventional MHC class II molecules, HLA-DO is a heterodimer, consisting of the DOA and DOB chains.

Peptide loading is also affected by the amount of antigenic peptide supplied to the MIIC compartment which varies depending on the cell-type. Macrophages are more efficient at degrading antigen than dendritic cells, and are therefore less efficient at antigen presentation.

MHC-II peptide complexes recycle from the plasma membrane

Class-II/peptide complexes are released from the multivesicular bodies as tubular/vesicular structures that fuse with the plasma membrane. The complexes tend to cluster on the cell surface in lipid rafts, which probably favors formation of the immunological synapse (see below). There is a continuous recycling of complexes from the plasma membrane to endosomal compartments, and the proportion of complex that is present on the plasma membrane is regulated by ubiquitinylation, which varies between cells. For example, mature dendritic cells tend to maintain more of the class-II/peptide complexes on the cell surface than immature dendritic cells.

T cell interaction with APCS

The interactions between a T cell and an antigen presenting cell develops over time, in three phases.

The initial encounter of T cells with APCs is by non-specific binding through adhesion molecules, particularly ICAM-1 (CD54) on the APC and the integrin LFA-1 (CD11/18), present on all immune cells. Transient binding permits the T cell to interact with many APCs; T cells in vivo are highly active and a single T cell may contact up to 5000 dendritic cells in one hour. Adhesion between the cells is enhanced by the interaction of CD2 (LFA-2) on the T cell with CD58 (LFA-3) on the APC (in rodents, CD48 performs a similar function to CD58). CD2 contributes towards the initial activation signal for the T cell, but more importantly, it allows the TCR time to recognize specific MHC/peptide on the APC. The initial phase of antigen presentation may last for several hours, but in the absence of a specific interaction, the APC and T cell dissociate.

When the T cell encounters the appropriate MHC/peptide, a conformational change in LFA-1 on the T cell, signaled via the TCR, results in tighter binding to ICAM-1 and prolonged cell–cell contact. (Fig. 8.14) The joined cells can exist as a pair for up to 12 hours, and this marks the second phase of interaction. At this stage an ‘immunological synapse’ forms and the T cell may be activated (Fig. 8.15).

Fig. 8.14 Color-enhanced reconstruction of an immunological synapse

A live-cell fluorescence image, showing the peripheral zone of adhesion molecules (red) surrounding the core containing TCRs (green), is superimposed on a scanning electron micrograph of a T cell (purple) interacting with a DC (dark green).

(Courtesy of Dr Mike Dustin and Science.)

Fig. 8.15 Diagram of an immunological synapse

Molecular interactions in the immunological synapse. TCR–peptide–MHC, CD2–CD48 and CD28–CD80/86 all contribute to the formation of a synapse. The TCR assembles with MHC molecule–antigen peptide molecules and moves with the costimulatory molecules to the centre of the synapse, while the adhesion molecules LFA-1–ICAM-1 move to the periphery. The diagram shows presentation by an MHC class-I molecule, but similar structures occur in MHC class-II mediated antigen presentation.

In the third phase, the APC and T cell dissociate and the activated T cell undergoes several rounds of division and differentiation.

The immunological synapse is a highly ordered signaling structure

The interactions between APCs and T cells have been studied extensively in vitro, where the cells form a ‘bulls-eye’ structure at the point of contact (see Fig. 8.14). This ‘immunological synapse’ is thought to reflect, in idealized conditions, the events that occur within a lymph node when T cells and dendritic cells interact; in vitro high doses of antigen may be used, and it is known that the size of the synapse relates to the amount of antigen present.

In the synapse, TCRs cluster in a patch on the cell surface and are surrounded by a ring of adhesion molecules. The movement of TCRs to the center of the synapse is a dynamic process and as few as 10 cognate interactions between a TCR and MHC/peptide can initiate synapse formation and T-cell activation. The final configuration of the synapse has TCRs at the hub of the synapse ringed by ICAM-1–LFA-1 interactions. Large CD45 molecules, a phosphatase which is ‘leukocyte common antigen’, are relegated to an outer ring.

The majority of work on immunological synapses has been done using CD4⁺ TH cells. TC cells form similar structures when they recognize their targets, but with a distinct secretory domain. Lytic granules from the cytotoxic cell are shifted towards the synapse along microtubules and accumulate just below the plasma membrane, before the granule contents are released into the inter-cellular space by exocytosis.

Costimulation by B7 binding to CD28 is essential for T cell activation

Productive T cell proliferation appears to depend on the formation of a stable central cluster of TCRs interacting with MHC molecules. The affinity of the binding of a single TCR molecule to its specific MHC/peptide is not high, and the formation of the immunological synapse requires the concerted interactions of a number of additional molecules, including CD4 which enhances binding of TH cells to MHC class-II molecules and CD8 to MHC class-I molecules. These molecules increase the sensitivity of a T cell for its target antigen by ~100fold. Although signaling efficiency for CD8⁺ T cells and thymocytes relates closely to the affinity of their TCR for the MHC/peptide, this is only partly true for the CD4⁺ T cells.

The specific MHC/peptide/TCR interaction, though necessary, is not sufficient to fully activate the T cell. A second signal, referred to as costimulation, is of crucial importance for T cell activation. The most potent costimulatory molecules are B7s, which are homodimeric members of the immunoglobulin superfamily molecules; they include B7–1 (CD80) and B7–2 (CD86). They are constitutively expressed on DCs, but can be upregulated on monocytes, B cells, and other APCs, particularly when stimulated by inflammatory cytokines and by interaction of microbial products with Toll-like receptors on the APC.

The B7 co-receptors bind to CD28 and its homolog CTLA-4 (CD152), which is expressed after T cell activation. CD28 is the main costimulatory ligand expressed on naive T cells. CD28 stimulation:

Many cells in the tissues can be induced to express MHC class-II molecules and present antigenic peptides to CD4⁺ T cells, however, they are mostly ineffective in inducing T cell activation and proliferation, because they lack the necessary costimulatory molecules (Fig. 8.16).

Fig. 8.16 Dual signaling is necessary for full T cell activation

A T cell requires signals from both the TCR and CD28 for activation. In the absence of costimulatory molecules inactivation or anergy results. If CD28 is bound by B7 on the surface of a professional APC, the T cell is activated and produces IL-2 and its receptor (IL-2R). The cell divides and differentiates into an effector T cell, which no longer requires signal 2 for its effector function. However, if CTLA-4 on the T cell binds to B7, activation is inhibited.

Ligation of CTLA-4 inhibits T cell activation

CTLA-4 (CD152) is an alternative ligand for B7, with higher affinity than CD28. It is an inhibitory receptor limiting T cell activation, resulting in less IL-2 production. CTLA-4 appears to act by reducing the time for interaction between the APC and the T cell, so a stronger activation signal is needed otherwise incomplete signaling will occur. As a T cell matures, it expresses higher levels of CTLA-4 and hence requires stronger activation stimuli to continue division (Fig. 8.17).

Fig. 8.17 Role of CTLA-4 in controlling T cell activation

Before activation, T cells express CD28, which ligates B7–1 and B7–2 on APCs (e.g. dendritic cells). After activation, CTLA-4 is expressed, which is an alternative high-affinity ligand for B7. CTLA-4 ligates B7, so the T cells no longer receive an activation signal.

Intracellular signaling pathways activate transcription factors

An appropriate stimulatory signal initiates a cascade of intracellular signals, leading to the activation of transcription factors and expression of genes required for cell division. Two widely-used immunosuppressive drugs, cyclosporin and tacrolimus, interfere with the activation pathways.

TCR-binding activates tyrosine kinases

Receptors on the surface of T cells signal to the interior of the cell using signal transduction pathways that are common to many other cell types, including B cells. A key principle is the clustering of receptors upon ligand binding which leads to activation of associated tyrosine kinases, that phosphorylate tyrosine residues in the cytoplasmic tails of the clustered receptors, followed by recruitment of additional kinases and signaling molecules, in cascades.

TCR α and β chains are associated with the CD3γ, δ, and ε molecules, the ζ and η chains and the enzyme Lck (p56^lck), which is attached to the intracellular portions of CD4 or CD8 (Fig. 8.w1). (The label p56^lck signifies a lymphocyte-specific tyrosine kinase of 56 kDa.) The first steps involve tyrosine kinases of the Src family, particularly Lck and Fyn, which phosphorylate target sequences found in the TCR ζ chain termed immunoreceptor tyrosine-based activation motifs (ITAMs).

Fig. 8.w1 Intracellular signaling in T cell activation

T cell activation involves the transduction of signals from both the TCR and CD28. Clustering of surface receptors such as TCR, CD4, CD28, and CD45 results in activation of tyrosine kinases Fyn and Lck. CD4, which is associated with the TCR complex, binds to the kinase, Lck. Such kinases become activated by dephosphorylation, possibly by phosphatase domains on CD45 (leukocyte common antigen). Lck can now phosphorylate the ITAM domains on the γ chains of CD3 (see Figs 5.3 and 5.15), which allows them to associate with other kinases including Fyn and ZAP-70. Fyn activates phospholipase C (PLCγ), which leads to two pathways by the cleavage of phosphatidylinositol bisphosphate (PIP₂) into diacylglycerol (DAG) and inositol trisphosphate (IP₃). IP₃ releases Ca²⁺ from intracellular (ER) stores to activate calcium-dependent enzymes such as calcineurin. Calcineurin removes phosphate from the transcription factor NF-AT (nuclear factor of activation of T cells), which causes its translocation to the nucleus. DAG activates protein kinase C (PKC), which then activates the transcription factor NF-κB. Meanwhile, ZAP-70, Fyn, and PI-3 kinase (PI-3 K) (associated with CD28) integrate signals via kinase cascades in the cytoplasm that activate specific transcription factors. Adapter proteins are used to link the various receptors to the common intracellular signaling components. GTP-binding proteins (G proteins) are involved in activating a set of protein kinases in the MAP kinase cascade. The transcription factors translocate to the nucleus to activate genes, including ‘immediate early genes’ Fos and Jun for cell division and the promoter AP-1, which acts with NF-AT on the IL-2 gene.

Phosphorylation of the ITAMs initiates a series of steps (see Fig. 8.w1), that lead to the activation of transcription factors including NFAT, and NFκB, and their translocation to the nucleus. The IL-2 enhancer contains a binding site for a nuclear factor, NF-AT (nuclear factor of activation of T cells), which also interacts with Fos and Jun, to induce IL-2 gene expression of the IL-2 receptor (IL-2R) gene.

In summary, TCR stimulation results in activation of a variety of tyrosine kinases and downstream effectors that regulate cellular responses, resulting in the regulation of IL-2 gene expression, which is largely responsible for activating the T cell and inducing division. An understanding of the requirements for T cell division has allowed researchers to generate long-lived T cell lines, which are used in many areas of immunological research (Method box 8.1).

Method box 8.1 T cell lines

To test the response of lymphocytes to a specific antigen that is being presented, the lymphocyte stimulation test (Fig. MB8.1.1) may be used. This test measures the response of the T cells to antigen as indicated by their entering the cell cycle and incorporating precursors of DNA synthesis.

Fig. MB8.1.1 A class I MHC tetramer

In the lymphocyte stimulation test, whole blood in saline solution is first layered on Ficoll Isopaque (which has a density between, and therefore separates, white cells and red cells) and centrifuged (400 g). This separates the lymphocytes from the other cell and serum constituents. The cells are washed (to remove contaminants such as antigen) and then put into test tubes with a suspension of antigen and culture medium. Tritiated thymidine (³H-thymidine) is added 16 hours before the cells are harvested. The cells are harvested on a glass-fiber filter disk and their radioactivity measured by placing the disk in a liquid scintillation counter. A high count indicates that the lymphocytes have undergone transformation and confirms their responsiveness to the antigen. This test can also be used for cells from lymphoid tissue.

Interleukin-2 drives T cell division

T cell activation leads to the production of IL-2 and IL-2 receptors, so a T cell can act on itself and surrounding cells. In most CD4⁺ cells and some CD8⁺ T cells, there is a transient production of IL-2 for 1–2 days. During this time the interaction of IL-2 with the high-affinity IL-2R results in T cell division.

On resting T cells, the IL-2R is predominantly present as a low-affinity form consisting of two polypeptide chains, a β chain (p75) that binds IL-2 and a common γc chain that signals to the cell. When the T cell is activated, it produces an α chain (CD25), which contributes to IL-2 binding and, together with the β and γc chains, forms the high-affinity receptor (Fig. 8.18). IL-2 is internalized within 10–20 minutes and the β and γc chains are degraded in lysosomes while the α chain is recycled to the cell surface. Sustained signaling by IL-2 over several hours is needed to drive T-cell division.

Fig. 8.18 Expression of the high-affinity IL-2 receptor on T cells

The high-affinity IL-2R consists of three polypeptide chains, shown schematically. Resting T cells do not express the α chain, but after activation they may express up to 50 000 α chains per cell. Some of these associate with the β chain to form the high-affinity IL-2R. (The γc chain is a common signaling chain of several cytokine receptors.)

The transient expression of the high-affinity IL-2R for about 1 week after stimulation of the T cell, together with the induction of CTLA-4, helps limit T cell division. In the absence of positive signals, the T cells will start to die by apoptosis.

In view of the importance of IL-2 in T cell division, it was surprising that the rare patients who lack CD25 (and IL-2 receptor knockout mice) develop an immunoproliferative condition. These observations lead to an awareness that IL-2 also has a regulatory function in T cell development – regulatory T cells (Tregs) are characterized by high CD25 expression, and IL-2 is required for their generation in the thymus and maintenance in the periphery (see Chapter 11).

Other cytokines contribute to activation and division

As T cell division and immune responses are not ablated in mice lacking the IL-2 or IL-2R genes, it suggests that other cytokines may support T-cell division. The cytokine IL-15 is structurally similar to IL-2; it acts on a receptor which shares the β and γc chains of the IL-2 receptor and it causes expansion of T cells and some NK cell populations. IL-15 is made by APCs and may therefore be very important in initial T cell activation before IL-2 is produced.

Although it was originally identified as a B cell growth and differentiation factor IL-4 can also induce division of naive T cells. The relative importance of these other cytokines will vary depending on the state of T cell, and these cytokines may partly overlap the function of IL-2.

Other cytokines contribute indirectly to T cell proliferation. For example, IL-1 and IL-6 induce the expression of IL-2R on resting T cells and thus may enhance their responsiveness to IL-2.

Danger signals enhance antigen presentation

For appropriate immune responses to be generated, APCs must respond to infection, but not to high levels of harmless substances that may fluctuate in the environment. Mucosal tissue in the gut are in contact with high concentrations of harmless food antigens, while respiratory mucosa contacts many airborne antigens such as pollen, but strong immune responses against these antigens are undesirable.

APC activation is generally a response to infection, or at least the presence of substances, such as constituents of bacterial cell walls, characteristic of infection. This requirement explains the action of adjuvants derived from bacterial components, which are used to enhance immune responses in vaccines.

The concept of immune activation only in response to infection (or adjuvant as a surrogate for infection), and not to other antigens, has been promoted as the ‘danger’ hypothesis. This idea proposes that the immune system does not merely distinguish self from non-self, but responds to clues that an infection has taken place before responding strongly to antigens.

In other words, foreign substances may be innocuous or invisible to the immune system unless accompanied by danger signals, such as infection. These danger signals are provided by receptors for microbial products on APCs, such as the Toll-like receptors (TLRs, see Fig. 6.21).