T Cell-Mediated Immunity

This module will help you

• see how T cells are activated by antigen presented on APC.
• understand the properties and functions of effector T cells.
• test your knowledge and immunology problem-solving skills.

T Cell Activation
Properties of Effector T Cells
Cytotoxic T Cells
Macrophage Activation by Th1 (Inflammatory T) Cells

T Cell Activation

Adaptive T cell-mediated immunity is driven by activation of T cells: cytotoxic T cells activated by endogenous antigen to kill infected cells, and helper T cells activated by exogenous antigen to stimulate macrophage killing of endosomal pathogens. Note that the pathogens targeted by cellular immunity are protected from antibody and complement binding by their intracellular locations. Adaptive humoral immunity also usually involves T cell activation to produce cytokines that stimulate B cell antibody synthesis. Naïve resting T cells are stimulated by antigen peptide presented on Class I MHC to cytotoxic CD8 T cells or Class II MHC to helper CD4 T cells, along with co-stimulatory signals from APC, to proliferate and differentiate into clones of fully activated effector T cells. Only professional APC (dendritic cells, macrophages, and B cells) have both Class I and Class II MHC and can deliver co-stimulatory signals. Activated T cells perform their effector functions when they encounter MHC-presented peptide on their target cells.

Naïve mature T cells constantly recirculate between secondary lymphoid organs and blood and lymphatic circulations. Lymphocyte trafficking depends on recognition of vascular addressins on vascular endothelial cells within the secondary lymphoid organs.  Most T cells use L-selectin to bind CD34 on lymph node and spleen HEV. A subset of T cells uses LPAM-1 to bind MadCAM-1 on mucosal endothelium. Only about one T cell in 10⁴-10⁶ is specific for a given antigen, and the probability of those rare T cells encountering properly presented antigen in the circulation is very low. Antigen activation of T cells occurs in the secondary lymphoid organs: Peyer's patches or tonsils for pathogens infecting mucosal tissues (respiratory, digestive, and urogenital tracts), draining lymph nodes for pathogens in the peripheral tissues, and the spleen for pathogens which are in the blood. Traffic through these secondary lymphoid organs takes T cells past APC which are likely to be presenting foreign antigen if any is present, since both antigen and leukocytes collect in secondary lymphoid organs. T cell "sampling" of self peptide:MHC complexes on APC when no infection is present serves as a survival signal for mature T cells once they leave the thymus.

Signaling between T cells and APC also depends on CAMs. Even if a T cell specific for a peptide were to encounter an APC with that peptide on the correct MHC, the probability of those molecules being close enough on both cell surfaces for binding to occur would be very small. Numerous CAMs on both cells allow them to adhere nonspecifically for a time so that if an antigen-specific interaction can occur, it will. As the T cell and APC adhere, movement of TCR and peptide:MHC complexes in the plane of the cell membranes brings them close enough to bind one another. Once TCR has bound peptide + MHC, T cell LFA-1 changes its conformation to bind even more tightly to the APC for the prolonged contact required for T cell activation. This contact may last several days, during which time the T cell divides and its progeny also adhere to the APC and receive activation signals.

To activate T cells, APC must deliver two distinct signals, one to TCR via peptide on MHC, and a second co-stimulatory signal. The professional APC that can provide co-stimulation are dendritic cells (DC), macrophages, and B cells. Of these cell types, DC deliver the best co-stimulation and are probably responsible for activating most naïve T cells. APC use the membrane molecule B7 (B7.1 = CD80 and B7.2= CD86 are slightly different forms of this molecule) to deliver a co-stimulatory signal through T cell membrane CD28. Signals received through CD28, when received with antigen signals, activate a T cell. Some antibodies to CD28 can provide co-stimulatory signals, while antibodies to B7 block the ability of the APC to activate T cells.

In general, the same cell must deliver the antigen signal and the co-stimulatory signal simultaneously to the T cell for activation to occur. The reason for this limitation is that not all self-reactive T cells are deleted in the thymus since some self antigens are sequester in other tissues or inside cells. If the antigen and co-stimulation signals could be delivered separately, T cells could be stimulated by self peptides by non-APC and then become fully activated to make an autoimmune response by an APC presenting a different peptide. T cells which bind antigen without co-stimulation become anergic, and cannot in the future respond to antigen even if proper co-stimulation is received. Why the cells are inactivated without being killed (although some do undergo apoptosis) is not known. One hypothesis is that anergic cells may compete with naïve cells for foreign peptide that resembles self and prevent autoimmunity.

Once a T cell becomes activated, it expresses other receptors which enhance co-stimulation. A key T cell molecule expressed early in activation is CD40 ligand (CD40L). CD40L binds APC CD40 and transmits activation signals to the T cell; T cells in mice that cannot express CD40L cannot go beyond the early stages of T cell activation. CD40L binding to APC CD40 signals the APC to express more B7 molecules to provide even more co-stimulation to the T cells; signaling to B cells and macrophages through CD40 also promotes their activation. Binding of other inducible signaling molecules 4-1BB (CD137) and ICOS (Inducible CO-Stimulator) on T cells to 4-1BB ligand (4-1BBL) and LICOS (Ligand of ICOS) on activated DC, macrophages and B cells also further activate both the T cell and the APC.

When a T cell becomes activated it expresses CTLA-4 a molecule very similar to CD28 and also able to bind B7, alongside its CD28 molecules. CTLA-4 binds B7 about 20 times more avidly than does CD28, and CTLA-4 binding to B7 sends an inactivation signal to the T cell. The immune response, therefore, depends on the "push" of antigen stimulation to activate T cells into effectors and the "pull" of negative signals that curtail T cell responses. These combinations of activation and inactivation signaling mechanisms are common in the immune system and are there to control a very powerful response that can damage the body if not restricted. Mice born with a defective CTLA-4 gene suffer from over-proliferation of lymphocytes.

Dendritic cells migrate to peripheral tissues from the bone marrow, where they are produced. They are found as immature dendritic cells under surface epithelial cells (skin and mucus membranes) and in most solid organs. When immature, DC express no B7 and little MHC. However, they are very active in binding antigen by pattern recognition and with the DEC-205 receptor, and in phagocytosing antigen the same way macrophages do. Immature DC can also take up antigen by ingesting extracellular fluid using macropinocytosis. DC are infected by many viruses, and other viruses which enter by macropinocytosis can replicate in the DC cytoplasm. Peptides of phagocytosed viruses can also be presented on Class I MHC, although it is not clear how they get from the endosomal to the cytoplasmic presentation pathway. DC can therefore activate both Tc with peptide on Class I MHC as well as Th with peptides on Class II MHC. DC present peptides from a wide variety of pathogens (including fungi and parasites), allergens, and alloantigens from transplanted organs.

Once immature DC have picked up antigen, they migrate to the peripheral lymphoid organs. There they enter the T cell areas and lose their ability to take up antigen, but express high levels of Class I and Class II MHC, B7, and the CAM DC-SIGN, making them excellent APC. They also secrete a chemokine (DC-CK) that attracts T cells. The signals that send DC to the lymphoid organs and change their phenotype are not known, although LPS signaling through Toll-like receptor TLR-4 also occurs for DC. Langerhans cells in the skin are an example of immature dendritic cell which become active APC in the lymph nodes.

Macrophages are very active phagocytes, especially of bacterial pathogens; they do not take up soluble antigens (proteins) well. They are found throughout the body as resident macrophages in tissues, and circulating monocytes can enter tissues in response to infection and become macrophages. Resting macrophages express little Class II MHC and B7, but increase expression of these molecules once they have phagocytosed bacteria due to signaling through receptors such as TLR-4. Pathogens which are phagocytosed most efficiently bind to receptors on macrophages recognizing patterns of common bacterial antigens such as LPS and carbohydrates: CD14, mannose receptor, scavenger receptor, and complement receptors. Bacterial antigens often serve as effective adjuvants for subunit (protein or peptide) vaccines because they induce expression of co-stimulatory molecules on APC. Macrophages migrate to secondary lymphoid organs from infection sites to present antigen to T cells.

Macrophages in the red pulp of the spleen and Kupffer cells in the liver continuously phagocytose dead host cells and, if they were to present those cells to T cells, might induce autoimmunity. However these macrophages express little B7 and MHC and are not stimulated to do so by binding and phagocytosing host cells. They are also not in locations frequented by T cells.

B cells are unique among APC because they use their membrane Ig to specifically bind and internalize soluble antigens. Antigen cross-linking of BCR stimulates endocytosis and antigen presentation by B cells. B cells constitutively have high levels of Class II MHC. Co-stimulatory molecule expression increases following contact with some bacterial carbohydrates and LPS, making it unlikely that B cells will activate T cells to respond to self peptides in the absence of infection. B cells enter secondary lymphoid organs by passing through the T cell areas, where they are likely to contact T cells. B cells activate T cells at lower antigen concentrations than do macrophages. They present predominantly bacterial and insect toxins and allergens. The importance of B cells in activating naïve T cells is uncertain. Because specific B cells are not initially present in high frequencies, DC are probably responsible for more T cell activation during the early stages of a primary response.

Once T cells are activated by peptide + MHC and receive co-stimulation, they begin to synthesize the a chain of IL-2 receptor (IL-2R, also called Tac after an early monoclonal antibody specific for it) and to secrete IL-2. Tac increases the affinity of the IL-2R, so that lower concentrations of IL-2 can signal the activated T cell to begin clonal proliferation and differentiation into an armed effector cell. Resting T cells express only the b and g chains of IL-2R and bind much less tightly to IL-2. The expression of more and higher affinity cytokine receptors by activated cells compared to resting cells keeps the immune response antigen-specific, even though the same cytokines are made in response to many different antigens. The response of T cells to a cytokine they secrete is called an autocrine response.

Properties of Armed Effector T Cells

After 4-5 days of rapid clonal expansion, T cells differentiate into armed effector cells that no longer need co-stimulation to perform their effector functions: cytotoxicity or cytokine secretion. T cells must, however, continue to bind peptide + MHC in order to kill or activate their target cells. Three classes of armed effector cells are specialized to deal with three different kinds of pathogens. CD8 CTL kill infected cells displaying cytosolic pathogen peptides on Class I MHC. CD4 Th1 cells activate macrophages with persistent vesicular pathogens whose peptides are displayed on Class II MHC. They also activate B cells to produce opsonizing antibodies. CD4 Th2 cells activate B cells that have used their membrane Ig to internalize specific antigen and display peptides on Class II MHC. B cell activation and effector functions are described in Humoral Immunity.

Armed effector cells have different membrane markers from naïve resting T cells. In particular they have lost L-selectin but gained VLA-4, so that they lose their homing for the lymph nodes and are more likely to enter virus-infected tissues or infection sites containing macrophages. Effector T cells also express higher levels of LFA-1 and CD2 and express CD45RO instead of CD45RA (CD45 is a tyrosine phosphatase that augments signaling through TCR and BCR complexes).

Two functionally distinct subsets of effector Th cells secrete cytokines which promote different activities. Inflammatory or Th1 CD4 T cells cells produce IL-2, IFNg, and TNFb, which activate Tc and macrophages to stimulate cellular immunity and inflammation. Th1 cells also secrete IL-3 and GM-CSF to stimulate bone marrow to produce more leukocytes and signal B cells to produce opsonizing antibodies (IgG₁ and IgG₃ in humans and IgG_2a and IgG_2b in the mouse). Helper or Th2 CD4 T cells cells activate naïve B cells to divide and secrete IgM; Th2 cells also secrete IL-4, IL-5, and IL-6, which stimulate neutralizing antibody production by B cells.

T cells are initially activated as T0 cells, which produce IL-2, IL-4 and IFNg. Factors which influence a Th0 cell to become a Th1 or Th2 are not well understood, but include the cytokines elicited by the pathogen, co-stimulatory signals, and the nature of the MHC:peptide presented to the Th cell. Th1 and Th2 cytokines have antagonistic effects on Th cells. The Th1 cytokine IFNg inhibits proliferation of Th2 cells, while the Th2 cytokine IL-10 inhibits Th1 secretion of IFNg and IL-2. The balance between Th1 and Th2 activity steers the immune response in the direction of cell-mediated or humoral immunity.

In some circumstances the immune response can become so polarized that pathogen cannot be eliminated. An example is the response to Mycobacterium leprae, which causes Hansen's disease (leprosy). M. leprae can survive and replicate in macrophage phagolysosomes. People infected with M. leprae who produce a Th1 response make minimal antibody but their cellular response keeps pathogen numbers low and their disease progresses slowly (tuberculoid leprosy). People infected with M. leprae who make primarily a Th2 response have high levels of M. leprae-specific antibodies, but the bacteria proliferate unchecked in their macrophages and disease progression is rapid (lepromatous leprosy). Th1 and Th2 cell activities were discovered because certain mice have very different susceptibilities to the protozoan parasite Leishmania major, which was discovered to be due to their making either humoral (susceptible) or cellular (resistant) responses.

CD8 Tc cells recognize antigen on Class I MHC. They require stronger co-stimulatory signals than CD4 T cells to become effector cells. CD8 T cells can be activated to peptides on Class I MHC by infected dendritic cells, which deliver a strong co-stimulatory signal. Other APCs do not produce strong enough co-stimulatory signals, so IL-2 from nearby CD4 Th cells binding peptide + Class II MHC on the same APC is required to fully activate CD8 Tc cells to become cytotoxic. Infected cells which are not professional APC express peptides on Class I MHC but cannot activate Tc cells because they cannot deliver co-stimulation.

The initial interaction between effector T cells and their targets occurs via CAMs, for example LFA-1 on the T cell and ICAM-1 on the target cell. If specific TCR-peptide-MHC binding does not occur, the effector T cell detaches and binds another potential target. If TCR binds specific peptide+MHC, the interacting molecules cluster on the opposing TCR and target cell membranes. Many individual TCR-peptide-MHC complexes form the center of the binding area, surrounded by many CAM-CAM interactions to hold the cells closely together. The T cell is triggered to release effector molecules into the close spaces between the cells: cytotoxins for CTL and cytokines for Th cells. The proximity of the cell membranes increases the effective concentration of secreted molecules and ensures that no other cells are nonspecifically stimulated to die or become activated.

Cytotoxic T Cells

Cytotoxic T cells (CTL) mediate antigen-specific, MHC-restricted cytotoxicity and are important for killing intra-cytoplasmic parasites that are not accessible to secreted antibody or to phagocytes. Examples include all viruses, rickettsias (causes of Rocky Mountain spotted fever and typhus), some obligate intracellular bacteria (Chlamydia), and some protozoan parasites which export their proteins from macrophage vesicles to the cytoplasm (Toxoplasma gondii). The only way to eliminate these pathogens is to kill their host cells.

Effector CTL bind their targets first via nonspecific adhesion molecules such as LFA-1 (CD 11a/18), then with specific TCR. TCR-peptide-MHC-CD8 binding reorients the CTL cytoskeleton to focus release of effector molecules towards the target cell. CTL induce apoptosis in their targets. Cells undergoing apoptosis undergo chromatin condensation and membrane vesicle shedding. DNA is cut into pieces in multiples of 200bp, which look like steps on a ladder (and is called a "DNA ladder") when run on a gel that separates the DNA by size. Fragmented DNA can be detected in individual cells using the TUNEL assay.

Programming the target cell to die requires only about 5 minutes contact between CTL and target, although the targets may appear viable for much longer. CTL then dissociate to bind and kill other target cells presenting the same epitope. During their maturation into effector CTL, CD8 cells synthesize cytotoxic molecules and store them in cytoplasmic granules for quick release upon target cell binding. Perforin has sequence homology with complement C9, and polymerizes to form pores in the target cell membrane through which water and salts can enter. At high perforin concentrations, this may be enough to destroy the target by osmotic lysis, but physiological levels of perforin are believed to be too low to induce osmotic lysis. Instead, perforin pores allow granzymes to enter the target cell. Granzymes are serine proteases which trigger the apoptotic cascade leading to DNA fragmentation and membrane-bound vesicle shedding. Apoptotic enzymes activated by granzymes can also destroy viruses or other cytoplasmic pathogens in the target cells so that the pathogens cannot infect nearby cells. Dead target cells are rapidly ingested by macrophages without activating macrophage expression of B7, preventing co-stimulation of any self-specific T cells which bind the self peptides on macrophage MHC.

Perforin polymerization must occur preferentially in the target cell plasma membrane, since the CTL is not lysed and one CTL can sequentially kill many infected targets. It is not understood how CTL are protected from perforin polymerization and granzyme entry, since the molecules are nonspecific and in allogeneic systems CTL can serve as target cells

A second mechanism for CTL killing was discovered in perforin knock-out mice. Binding of CTL membrane Fas ligand (FasL) to target cell Fas induces target cell apoptosis. Since activated Th1 and Th2 cells also express FasL, they may also have cytotoxic activity. Because mice with defects in Fas or FasL have a higher incidence of lymphoproliferative disease and autoimmunity, this cytotoxic mechanism may be important for regulating immune responses.

CTL also regulate immune responses by releasing IFNg, TNFa, and TNFb. IFNg inhibits viral replication, activates IL-1 and TAP expression by infected cells to promote antigen presentation, and recruits and activates macrophages as APC and effector cells. TNFa and TNFb act with IFNg to activate macrophages and to directly kill some target cells. Macrophages with surviving vesicular pathogens such as Toxoplasma gondii may also present pathogen peptide on Class I MHC because some pathogen proteins are exported to the cytoplasm. In these cases, CTL may play an important role in eliminating the parasite by killing host macrophages as well as activating other macrophages to engulf and kill the parasite. IFNg can also starve resident intracellular parasites by reducing tryptophan concentration. Natural Killer cells mediate early innate cytotoxic responses to viruses and tumor cells with perforin and granzymes, although recognition of targets is quite different.

Macrophage Activation by Th1 (Inflammatory T) Cells

Macrophages phagocytose and kill many pathogens using lysosomal enzymes. Some pathogens, however, are phagocytosed but evade killing by preventing fusion of lysosomes with the phagocytic vesicle or activation of the lysosomal enzymes by acidification of the phagolysosome. Some pathogens can escape the phagolysosome and live in the macrophage cytoplasm. In all these circumstances, the pathogens are protected from complement and antibodies; where the pathogens remain in vesicles, their peptides are also not presented efficiently to CD8 T cells.

Macrophage activation by armed effector Th1 cells is required to eliminate vesicular pathogens. Macrophage activation requires T cell binding to antigen peptide on macrophage Class II MHC, co-stimulation of the macrophage via CD40, and activation of the macrophage by IFNg . Th1 membrane CD40L binds to macrophage CD40 and signals the macrophage to express receptors for IFNg, which is synthesized de novo by the Th1 cell. Synthesis takes several hours, so macrophage activation is slower than CTL-mediated cytotoxicity. CD40L and IFNg secretion are directed towards the antigen-presenting macrophage by membrane polarization; activation is generally limited to infected macrophages. IFNg produced by CTL can activate macrophages presenting cytosolic peptides on Class I MHC. Macrophages can be sensitized to IFNg by low levels of LPS and membrane-bound TNFa or TNFb can substitute for the CD40L signal in macrophage activation. Activated Th2 cells express CD40L but instead of IFNg they secrete the macrophage-inhibitory cytokine IL-10.

IFNg activates macrophages to upregulate Class II MHC, B7, CD40, and TNF receptor expression to recruit more effector Th1 cells and to become more sensitive to CD40L and TNFa. Activated macrophages also fuse their phagosomes and lysosomes more efficiently and increase their synthesis of nitric oxide, oxygen radicals, antimicrobial peptides, and IL-12, which pushes Th cells towards the Th1 pathway.

When pathogens are present which cannot be phagocytosed, such as large parasites, activated macrophages release oxygen radicals, nitric oxide, and proteases into the extracellular fluid to kill pathogen. Excretion of these compounds also damages surrounding host tissues. Chronically infected macrophages may become resistant to activation. Granulomas, giant cells consisting of fused macrophages surrounded by activated T cells, form when intracellular pathogens cannot be eliminated. As cells in the center of the granuloma die, the dead tissue resembles cheese and is called caseous necrosis. The local inflammatory response resulting from activated Th1 cells and macrophages is called Delayed Type or Type 4 Hypersensitivity (DTH). If macrophages express Fas, they can be killed by FasL-expressing Th1 cells.

In addition to activating macrophages, Th1 cells secrete a variety of cytokines that regulate cellular immunity. IL-2 stimulates clonal proliferation of T cells. IL-3 and GM-CSF stimulate macrophage differentiation in the bone marrow. Multiple cytokines, including TNFa, TNFb, and Macrophage Chemotactic Protein (MCP), recruit macrophages to the site of infection.

Practice Quiz

a. enzymes which cut DNA into ladders to induce apoptosis.
b. present in an inactive form in resting CD8 cells.
c. secreted in large amounts by CTL to kill many target cells simultaneously.
d. structurally similar to macrophage cytokines that kill vesicular pathogens.
e. synthesized by CTL before they bind to target cells.

a. activate macrophages.
b. cut DNA in target cells into 200bp fragments.
c. directly inhibit viral replication in infected cells.
d. induce macrophages to express higher levels of MHC on their membranes.
e. starve intracellular parasites of tryptophan.

30. Macrophage activation by Th1 cells is an important immune mechanism for eliminating

a. bacteria which can resist lysosomal degradation.
b. bacteria whose capsule makes them resistant to phagocytosis.
c. enveloped viruses.
d. parasites that infect T cells.
e. viruses that infect macrophages.

31. T cells which activate macrophages do all of the following EXCEPT

a. activate macrophages at the site of infection.
b. activate only macrophages presenting specific antigen on MHC.
c. become activated effector cells in response to antigen plus co-stimulation.
d. have membrane CD4.
e. provide granzymes that the macrophages use to kill vesicular pathogens.

a. IL-1 and IL-6.
b. IL-2 and IFNg.
c. IL-4 and IL-5.
d. IL-10 and IL-12.
e. TNFa and FasL.

33. Macrophages kill pathogens using all of the following EXCEPT

a. nitric oxide.
b. oxygen radicals.
c. perforins.
d. peroxides.
e. proteolytic enzymes.

34. Th1 cells make IL-3 and GM-CSF, which

a. activate macrophages.
b. induce apoptosis in old macrophages.
c. stimulate macrophages to kill vesicular pathogens.
d. stimulate macrophage production in the bone marrow.
e. None of the above is a function of IL-3 and GM-CSF.

Problems

1. Leishmania major is a protozoan parasite that infects macrophages and monocytes, where it escapes destruction by lysosomal enzymes and escapes to replicate in vesicles. Different mouse strains respond differently to L. major infection: BALB/c mice produce primarily activated Th2 cells, while C57Bl/6 mice produce primarily activated Th1 cells. Which mouse strain would you expect to be more resistant to Leishmania infection, and why? What Leishmania-specific effector cells would be present in infected mice from each strain?