Rakesh K. Bakshi, Maureen A. Cox and Allan J. Zajac
Department of Microbiology, University of Alabama at Birmingham, Birmingham, AL, USA
CTL; Cytotoxic T lymphocytes; Killer T cell; Tc
Cytotoxic T lymphocytes are specialized subsets of differentiated T cells which have the functional capability to kill target cells expressing nonself antigens. They are important components of the adaptive immune response and help control intra-cellular pathogens, as well as contain the development of tumors.
The Significance of Cytotoxic T Cells
Cytotoxic T lymphocytes CTL are a cornerstone of the adaptive immune response Cox and Zajac 2010; Zhang and Bevan 2011. Due to their exquisite ability to detect and destroy cells presenting foreign antigens, they play a vital role in the initial control of many intracel-lular pathogens as well as help confer long-lived immunological protection against subsequent reinfections. Even in situations when the infection is not completely eradicated, such as during persistent chronic or latent infections, CTL function to dampen the initial burst of pathogen replication and then operate to either hold the infection at a steady state level or contain and limit the secondary spread of pathogens as they reactivate from latency. CTL responses also participate in tumor immunosurveillance and can limit the outgrowth of malignancies. CTL responses are not only induced by natural infections or as malignancies develop but are also generated in response to vaccines. Thus, CTL responses are versatile, durable, and vital for host defense.
The search and destroy functions of CTL are governed by the ability to co-recognize and respond to short peptide fragments expressed on the surface of the target cell in non-covalent association with either major histocompatibility complex MHC class I or MHC class II molecules Fig. 1. CD8 T cells, which form the most prevalent population of CTL, usually
Recognize peptide fragments of 8-9 residues in length bound to MHC class I molecules Neefjes et al. 2011. These antigenic peptides are most often derived from self and nonself e. g., virus-derived proteins which have been synthesized within the target cell and then subjected to proteolytic degradation. The antigen-processing machinery within the cell ensures that samples of these endogenous peptides associate with nascent MHC class I molecules and then localize to the cell surface for presentation to patrolling T cells. An alternative cross presentation pathway can also operate to enable peptides from exogenous proteins, which have been taken up by the target cells, to be presented by MHC class I molecules. Cytotoxic CD4 T cells are less common than CD8 CTL and their antigen-recognition properties are distinct. CD4 T cells recognize MHC class II complexes bound to peptide fragments which are generally slightly longer than those which associate with MHC class I molecules. Additionally, the peptides presented by MHC class II molecules are generally derived from exogenous antigens which have been endocytosed by specialized antigen-presenting cells APCs and then degraded. The resulting panels of peptides then assemble with MHC class II molecules and are presented at the cell surface. Notably, whereas MHC class I complexes are expressed by almost all cell types, the distribution of MHC class II molecules is largely confined to cells of the immune system, which limits the protective efficacy of CD4 CTL responses.
Which MHC-peptide complexes an individual T cell is capable of binding and responding to is determined by the precise sequence of the T cell`s unique T cell receptor TCR Bridgeman et al. 2012. The coding sequence, and thus exact structure, of the TCR is determined by a process of genetic recombination that favors diversification within the receptor regions that are most important for MHC-peptide interactions. This generates a tremendously diverse repertoire of TCRs. Nevertheless, TCRs are clonally expressed and thus an individual T cell will express a unique TCR that dictates its specific antigen and MHC recognition profile. At the population level, the lower limit of the number of T cells with nonidentical TCRs has been estimated to be 25 x 10 . The actual repertoire is likely to be considerably greater, perhaps several thousandfold higher, because of additional sequence variation within the a - and Я-chains which form the TCR complex Robins et al. 2009. This colossal inventory provides the host with a T cell pool with the potential power to detect and respond to the myriad of foreign antigens that it may encounter during its lifetime. Nevertheless, although the T cell repertoire is huge, the number of distinct naive clones which recognize any given MHC-peptide complex is small. Estimates from murine studies have determined that there are perhaps several hundred individual T cells which express appropriate TCRs that allow them to respond to a specific peptide epitope.
Pathogens and other antigens, such as those associated with tumor development, are typically comprised of multiple epitopes which can potentially be recognized by distinct T cell clones. Nevertheless, not all epitopes are equally recognized and the magnitude, longevity, and efficacy of responses to individual epitopes varies, even if they are encoded by the same pathogen Frelinger 200 . The hierarchical pattern of responses that emerge is determined by many factors including the properties of the antigen such as its stability, level of expression, and which cell types are presenting the antigen, as well as intrinsic T cell factors such as the avidity of the T cell for the MHC-peptide complex and the frequencies as well as activation states of the T cells which can respond to the specific antigen. The epitope hierarchy can influence the ability of CTL to control infections, as responses to individual epitopes can differ in their protective potential, and the pattern of immunodominance may change over time, during persistent infections, during recall responses, or as tumors evolve. Moreover, certain T cells may exhibit a degree of cross-reactivity which may confer beneficial additional immuno-logical protection or cause detrimental immunopathology.
Priming the CTL Response
CTL responses are induced following an antigen-driven activation process which typically results in massive T cell expansion, their differentiation and acquisition of effector functions, and the disbursement of the cells throughout the host Fig. 2a Cox and Zajac 2010; Sheridan and Lefrancois 2011. The overall response is typically comprised of CD4 helper T cells, CD8 CTL, which form the major CTL population, and under certain conditions a smaller number of CD4 CTL. Pronounced CD8 T cell responses have been detected during many infections including influenza virus, vaccinia virus, Epstein-Barr virus EBV, yellow-fever virus, lymphocytic choriomeningitis virus LCMV, and early following human immunodeficiency virus HIV infection. Moreover, it is becoming increasingly well appreciated that the vast majority of CD8 T cells detected at the peak of the response are pathogen specific Zhang and Bevan 2011.
Usually the T cell responses are triggered in secondary lymphoid organs - the spleen and lymph nodes - as naive T cells encounter professional APCs and recognize MHC-peptide complexes. This interaction between the T cells TCR and its cognate-presented antigen is the first signal that begins the subsequent clonal expansion and cellular differentiation; however, this initial interaction alone is insufficient to sustain the response. During infections, APCs also become activated as part of the host`s innate defense mechanism, leading to the upregulation of many surface receptors including the costimulatory molecules CD80 and CD8 . In order to prevent aberrant immune responses to innocuous or self-antigens, T cells must also receive costimulation via CD28 interacting with
CD80 or CD8 on the APC. This second signal
Promotes T cell survival and a burst of interleukin IL-2 production. In addition, other costimulatory molecules, including many members of the tumor necrosis factor TNF receptor superfamily, can also influence the ensuing response. Infections with pathogens also elicit the production of other immunological warning signs including proinflammatory cytokines such as IL-12, antiviral factors such as type I interferons, and alarmins such as IL-33. All of these elements can serve as important third signals for driving the full activation of the T cell response and amplifying effector activities Curtsinger and Mescher 2010. As the small number of naive T cells detect antigen signal 1,
Cytotoxic T Lymphocytes, Fig. 2 The phases of the CTL response: a Acute antigenic exposure, in conjunction with costimulatory signals and inflammatory cues, can drive massive proliferation and differentiation of CD8 T cells. This process ensures that an expanded pool of effector T cells form which operate to clear the inducing antigen. As the antigen is cleared, a contraction phase ensues during which most of the highly functional, terminally differentiated, T cells succumb to apoptosis. At the conclusion of the contraction phase, a pool of memory CD8 T cells remain and are represented at a greater frequency than their naive counterparts. These memory populations can be stably maintained for years following priming and function to confer long-lived immunological protection. b Aberrant T cell responses develop under conditions of persistent antigenic stimulation, such as during chronic viral infections. Typically the burst size of the response is smaller than that observed following acute stimulation. Although the responding cells develop certain effector traits and can attain a highly activated terminally differentiated phenotype, a spectrum of functional defects can arise. As effector activities are gradually extinguished, the antigen-specific T cells may not be stably maintained
Engage costimulatory molecules signal 2, and sense certain soluble factors signal 3, they are driven into cell cycle, proliferate vigorously, and form a highly differentiated effector pool.
During the proliferative phase of the response, the properties of the responding T cells profoundly change. The expression of adhesion molecules, as well as cytokine and chemokine receptors, shifts allowing the cells to migrate from the sites of priming in secondary lymphoid organs to other locations where they may encounter infected cells Cox and Zajac 2010; Sheridan and Lefrancois 2011. Metabolic changes also occur which are necessary for the rapid and substantial cell division that the T cell will undergo. Importantly, the responding cells differentiate into effector cells with the ability to rapidly produce numerous cytokines as well as kill target cells. These effector activities are triggered as T cells detect and bind to their cognate MHC-peptide complexes on target cells and can be initiated by the newly differentiated effector cell without the requirement for costimulation or third signals. The CTL population is, however, functionally heterogeneous with certain T cells more polyfunctional and capable of eliciting a broad range of effector activities, whereas other constituents of the effector pool may possess more limited functional abilities
Makedonas and Betts 200 . Most of the CD8 T cells that are initially
Generated become terminally differentiated effector CTL as the response develops. If the inducing antigen is controlled, then the vast majority of these terminally differentiated T cells undergo apoptosis. Ideally the response does not completely collapse. Instead, as homeostasis is restored, a population of memory T cells remain present Fig. 2a. Memory T cells can be maintained at remarkably stable levels and have been detected for decades following smallpox vaccination. Maintaining this memory population contributes to immunological protection against reexposure to antigen or reinfection with the pathogen. Although there are far fewer antigen-specific T cells present during the memory phase than at the peak of the response, the frequency of these cells is higher than at the onset of the priming event. Moreover, by comparison with their naive counterparts, memory T cells are generally tuned to mount faster and more efficacious responses upon antigenic activation.
Memory T cells are distinct from their naiive and effector counterparts, and they are also diverse Lanzavecchia and Sallusto 2005. Asubsetofmemorycells which express lower levels of the adhesion molecule CD 2L and the chemokine receptor CCR7 are referred to as effector memory cells. They are preferentially located in non-lymphoid tissues where they are immediately available to rapidly elicit effector activities, including cytokine production and cytotoxicity, in response to reexposure to antigen at these sites. They are sensitive to MHC-antigen activation but do not require professional APC or other signals in order to function. These tissue resident cells are often retained at the site of the primary infection, such as in the lung following influenza infection or the skin and dorsal root ganglion following herpes simplex virus infection, but can be more systemi-cally distributed Sheridan and Lefrancois 2011. A population of memory T cells, termed central memory T cells TCM, express higher levels of CD 2L and CCR7 and are primarily present in secondary lymphoid organs. In some ways, TCM display a less differentiated pheno-type than effector memory T cells and are more prone to proliferate upon secondary stimulation. Although TCM may be less efficient at deploying immediate effector activities, they can attain these traits as they begin to divide in response to antigenic activation.
The induction of robust effector responses and the development of T cell memory are usually associated with antigens that are cleared from the host, which occurs, for example, following vaccination or during acute infections. A spectrum of functional and phenotypic differences in T cell responses have been observed under conditions of antigen persistence which develop as a result of the failure to eradicate the infection or tumor Fig. 2b Yi et al. 2010. The production of cytokines including interferon IFN-g, TNF-a, and IL-2, as well as cytotoxic effector molecules such as perforin and granzymes, may be diminished or abolished as the T cells responding to these persistent stimuli fail to attain or lose polyfunctional traits and succumb to exhaustion. Differences in prolifera-tive capacity and dependency on homeostatic cytokines such as IL-7 and IL-15 have also been observed, and severe functional exhaustion can culminate in the physical loss of antiviral CD8 T cells. Comparative analysis of CD8 T cell responses to viral infections which result in different levels of antigenic exposure, such as influenza virus, cytomegalovirus, EBV, hepatitis C virus, and HIV, indicates that antiviral CD8 T cells may adopt different preferred phenotypic and functional set points. This is influenced by the level and repetitiveness of the antigenic stimuli, as well as by the balance of activatory and suppressive cytokines, and also by the presence of helper and regulatory CD4 T cells.
Transcriptional Control of CTL Differentiation
The profound phenotypic and functional changes that the responding T cells undergo during the priming phase reflect modifications in gene expression which are regulated by transcription factors Cox et al. 2011; Kallies 2008. Inflammatory cytokines such as IL-12 and IFN-g induce expression of the transcription factor T-bet tbx21, Which promotes the differentiation of
Type I helper CD4 T cells as well as CD8 CTL
Responses. High levels of T-bet drive the terminal differentiation of effector CTL, which have potent cytolytic capacity but limited proliferative abilities. Notably, if CD4 T cell help is insufficient, then T-bet levels are high, which promotes effector CD8 T cell development but curbs the formation of effective memory responses. The related transcription factors FOXO1 and FOXO3 also regulate effector T cells and in the case of FOXO1, this is due to inhibition of T-bet which restricts the differentiation and expansion of the response. Lower amounts of T-bet, however, permit the formation of memory T cells, which can be maintained over time and give rise to secondary effector cells during recall responses. In mice, T-bet deficiency results in reduced effector CTL and favors the development of long-lived memory T cells. Nevertheless, some effector phenotype cells are still generated, and some cytotoxic potential is still retained as a result of compensation by another transcription factor, eomesodermin Eomes. Eomes expression is most commonly associated with memory T cell formation, and this transcription factor is repressed by the inflammatory conditions that induce T-bet. Deletion of Eomes alone does not prevent development of an effector CTL pool; however, these cells are poorly maintained as memory cells and appear incapable of responding to secondary challenge. If both T-bet and Eomes are deleted from CD8 T cells, however, the CTL response is aberrant and infection control compromised.
In addition to T-bet and Eomes, the transcription factor B-lymphocyte-induced maturation protein Blimp-1 prdm1 Is also critical for CTL development. The absence of Blimp-1 results in poor expression of the cytolytic molecules perforin and granzyme B and poor control of viral infections. One function of Blimp-1 may be to induce greater levels of T-bet and reduce Eomes expression, thereby favoring effectorCTL development. In Blimp-1-deficient T cells, Tbx21 Expression is lower, and Eomes Expression is significantly higher at the mRNA level. Similar to T-bet, the level of Blimp-1 expressed in a cell is critical to determining the functional capacity of CD8 T cells. Therefore, controlling the amount of Blimp-1 expressed within a responding CTL molds that cell`s ultimate fate and function.
The related transcription factors Id2 and Id3 support T cell survival and promote memory formation. Interestingly, Blimp-1 suppresses Id3, which likely helps to cement the development of terminally differentiated effector T cells at the expense of memory formation. Moreover, Bcl - , a transcription factor associated with memory T cells, can also repress Blimp-1 expression. Overexpression of Bcl- drives expansion of memory T cells, and the effector cells generated under these conditions display reduced killing and granzyme B expression. As Blimp-1 and Bcl - are mutual repressors, this increase in Bcl - expression may be preventing the Blimp-1-driven differentiation of the effector response that ordinarily occurs following T cell activation.
Antigenic activation is the principle driver of the T cell response, but the levels of cytokines, which flux during the course of many infections and immune responses, also regulate the T cell transcriptional network Cox et al. 2011. As discussed above, T-bet levels are determined in part by inflammatory cytokines such as IL-12. IL-2 signaling in CD8 T cells strongly induce the expression of Eomes and Blimp-1. Moreover, signaling through the IL-10 and IL-21 cytokine receptors activates the transcription factor STAT-3 which influences the expression of Blimp-1 and Bcl - . Hence, the composition of the cytokine milieu affects the phenotypic and functional attributes attained by the responding T cells.
CTL Effector Mechanisms
Since CTL recognize and respond to MHC-peptide complexes expressed upon the cell surface, cell-cell contact is necessary for the elaboration of their effector activities. Thus, by comparison with antibody responses, CTL cannot confer sterilizing immunity and prevent infections by neutralizing the cell-free form of the pathogens present in circulation or other body fluids. CTL responses do, however, provide an important cellular mechanism for eliminating pathogen-infected cells or other cell types expressing nonself or tumor-associated antigens. For lytic infections, the rapidity of CTL response is critical as swift action by these cells limits the release of the pathogen and eliminates the infected cell before full replicative potential is reached. For non-lytic infections, which can parasitize the target cell without causing its destruction, CTL function to destroy these factories of pathogen production.
CTL functions are commonly associated with terminally differentiated effector cells which possess cytolytic functions and can also express a variety of cytokines and chemokines Makedonas and Betts 200 . The secretion of effector cytokines and chemokines is initiated following contact of the T cell with a target cell and in the case of IFN-g requires new protein synthesis. The production of cytokines such as IFN-g and TNF-a by T cells can mediate the clearance of certain infections including hepatitis B virus without causing target cell death Guidotti and Chisari 2001. Conversely, in the case of influenza infection, cytolytic effector functions can be elaborated against pulmonary epithelial cells without activating cytokine production. Such multifunctional capabilities provide a degree of flexibility and functional breadth necessary to combat diverse intracellular pathogens and tumors.
The best studied and most common mechanism of CTL-induced cell death is perforin - and granzyme-mediated cytotoxicity Fig. 3a. As CTL gain their effector capabilities, they develop cytolytic granules which contain the pore-forming protein perforin and a variety of granzymes. As these effector CTL recognize target cells via their TCR, an immunological synapse forms between the engaged cells Jenkins and Griffiths 2010. At the time of target cell engagement, preexisting perforin is the first to be released as marked cytoskeletal rearrangements occur within the CTL and the lytic granules migrate along microtubules toward the immunological synapse where they are discharged. This activation process can also stimulate the production of new perforin which replenishes the CTL and sustains effector capabilities.
Upon release perforin polymerizes in the target cell membrane permitting entry of granzymes which deliver the "kiss of death" to the target cell. In addition, at high concentrations, granzymes can access target cells in a perforin-independent manner. Although CTL express a variety of granzymes, the roles of granzyme B in the lytic process are most well defined Ewen et al. 2012. Granzyme B has many molecular targets including BH3-interacting domain death agonist BID, procaspase-3, and inhibitor of caspase-activated DNase ICAD. Granzyme B cleaves BID into its truncated form, tBID and procaspase-3 into activated caspase-3. tBID translocates into the outer mitochondrial membrane and promotes the oligomerization of Bax and Bak, leading to membrane disruption and the release of cyto-chrome C into the cytosol. The cytochrome C release activates caspase-9, which further activates caspase-3, ultimately resulting in the cleavage of ICAD, which can also be directly acted upon by granzyme B, into caspase-activated DNase CAD. Once activated, CAD then enters the nucleus causing DNA fragmentation and apoptosis.
The mechanisms of action for other granzyme molecules A, H, K, and M are less well understood Ewen et al. 2012. Notably, patients with a variety of inflammatory conditions express detectable levels of soluble granzymes, which may have non-cytotoxic biological roles. It has been shown that granzyme A can elicit the production of inflammatory cytokines.
Cytotoxic T Lymphocytes, Fig. 3 CTL killing mechanisms. a Perforin/granzyme-mediated cytotoxicity occurs as CTL engage a target cell via their TCR, relaying a signal that releases prestored cytotoxic granules containing perforin and granzymes. Perforin forms a pore in the target cell membrane which enables the entry of granzymes. Granzyme B functions as a key inducer of apoptosis, resulting in target cell destruction. b CTL can also cause target cell apoptosis by activating the death receptor pathway via FasL-Fas interactions or by the production of TNF-a. Engagement of Fas on the target cell by FasL, which is expressed on certain populations of CTL, launches a signaling cascade via formation of the DISC complex which ultimately results in apoptosis. TNF-a, which is produced by certain CTL subsets, also causes death receptor-mediated apoptosis following binding to its receptor TNFR1 and recruiting TRADD, leading to the formation of complex II. Signaling through complex II activates downstream executioner caspases leading to cell death
In addition, granzyme A and its family member granzyme K are able to induce cell death independently of caspase activity and the release of cytochrome C from mitochondria. Granzymes A and K are also both capable of inducing the production of reactive oxygen species which can cause cell death. In addition to production of reactive oxygen species, granzyme K can also activate mitochondrial apoptosis pathways, as measured by cytochrome C release.
Throughout the killing process, the CTL must avoid autolysis and self-destruction by their own lytic granules. In order to preserve their own integrity, CTL encode the protease inhibitor-9 in humans or its murine ortholog, serine protease inhibitor - , which functions to stabilize the granules within the CTL and also protects against granzyme B-mediated apoptosis Ashton-Rickardt 2010. In addition, at the time of granule deployment, CTL release cathepsin-B which prevents perforin from attacking the effector T cell.
Death Receptor-Mediated Apoptosis
Although perforin/granzyme-dependent cytotox-icity is a principal effector mechanism, CTL can also mediate death receptor apoptosis via the Fas ligand FasL-Fas or TNF receptor pathways Fig. 3b Smyth and Trapani 1998. The FasL-Fas pathway is initiated, as activated CTL upregulate FasL on their cell surface, usually following engagement with a target cell. The interaction between FasL on the CTL and Fas on the surface of the target cell promotes the formation of the death-inducing signaling complex DISC, which is comprised of Fas-associated death domain FADD, procaspase-8, and procaspase-10. Signaling through DISC results in cleavage of these procaspases, releasing the active caspases, which in turn activate caspase-3. Caspase-3, as an executioner caspase, then cleaves ICAD, activating CAD, thus inducing DNA fragmentation and cell death. Caspase-8 also cleaves BID into tBID, which oligomerizes Bax and Bak, causing the release of cytochrome C from the mitochondria. This release of cytochrome C activates caspase-9 and elicits a downstream signaling pathway which also culminates in cell death.
TNF-a can be secreted by CTL upon antigenic stimulation and is also a mediator of cell death. TNF-a binding to TNF receptor 1 TNFR1 on the target cell results in trimerization of the receptor complex and recruitment of TNFR-associated death domain TRADD. FADD is also recruited forming complex II, which leads to caspase activation and induction of cell death through DISC, paralleling Fas-mediated apoptosis. TNF-a-induced cell death can also be mediated by membrane-bound TNF-a, independently of the secreted cytokine.
CTL-mediated cell death is usually a tightly regulated and exquisitely antigen-dependent process. Nevertheless, CTL can sometimes cause apoptosis in bystander cells that are not directly recognized by the effector cell. This can result from FasL present on the CTL surface engaging Fas-expressing target cells in the absence of TCR signaling. Similarly, cells expressing TNFR1 may be susceptible to bystander killing caused by membrane-bound or secreted TNF.
Evading the CTL Response
The power of CTL responses in combating pathogens is well illustrated by the various strategies which have evolved to counteract the actions of the effector response Keckler 2007. Many pathogens encode specific molecules which interfere with antigen-processing and presentation pathways, thereby circumventing T cell recognition. In addition, an effective CTL response can exert pressure on the pathogen or tumor to select for viable genetic variants with escape mutations within antigenic epitopes. Such alterations can render the existing population of CTL ineffective if they cannot recognize or efficiently respond to the variant pathogen or tumor. Pathogens have also adopted immune evasion strategies which directly interfere with the CTL-mediated killing process. For example, certain viruses encode serpins that block caspase activation, which is critical for the cytolytic process. These molecules include one of the most potent inhibitors of caspase-mediated cell death, CrmA, encoded by cowpox virus, as well as related serpins encoded by vaccinia virus, myxoma virus, and ectromelia virus. Granzyme B is directly inhibited by CrmA, and this death pathway is also targeted by the herpes simplex virus glycoprotein J and the ade-novirus-encoded L4-100K protein. Death receptor signaling is blocked by vFLIP and related molecules encoded by herpesviruses and poxvi-ruses, which disrupt the formation and function of the DISC, impeding the induction of apoptosis. Fas-mediated cell death is blocked by the adeno-virus E3 region-encoded receptor internalization and degradation complex, as well as by the myx-oma virus LAP protein which reduces surface expression of this death receptor. Soluble TNF decoy receptors encoded by poxviruses also directly block the action of this cytokine.
Cytotoxic CD4 T cells
CD4 T cells are most commonly associated with helper functions which promote B cell and CD8 T cell responses and with regulatory properties which can suppress the development of pathogenic immune responses. CD4 T cells can, however, also adopt a cytotoxic phenotype, and cytolytic CD4 T cell responses have been detected following several infections including influenza virus, West Nile virus, hepatitis C virus, human cytomegalovirus, HIV, EBV, and LCMV Marshall and Swain 2011. The effector mechanisms used by CD4 CTL mirror those used by conventional CD8 CTL. Like their CD8+ counterparts, CD4 CTL have multifunctional properties and can express cytokines including TNF-a and IFN-g as well as cytolytic effector molecules including perforin, granzymes, FasL, and, in humans, granulysin. CD4 CTL are detected in both lymphoid and non-lymphoid organs and may display a terminally differentiated phenotype, similar to
The highly cytolytic effector CD8 CTL,
CCR7-, CD27-, CD28-, CD62L-, CD45RO+, and may express the senescence marker CD57.
CD4 CTL recognize peptide-MHC class II complexes which are primarily expressed by professional APC, but under inflammatory conditions expression can be induced on certain other cell types including fibroblasts and endo-thelial cells. Therefore, by comparison with CD8 T cells, which recognize broadly expressed MHC class I complexes, the more limited tissue distribution of MHC class II molecules likely limits the efficacy of the CD4 CTL response. Nevertheless, CD4 CTL may be important during EBV infection or B cell lymphocytic leukemia, as they are able to recognize and kill infected or malignant MHC II-expressing B cells. In addition, early cytolytic CD4 responses are associated with better control of HIV-1 infection and under certain circumstances CD4 CTL can also have a pathogenic role.
Although CTL play a vital protective role, by clearing infections, eliminating tumors, or limiting the levels of pathogen replication, there can be detrimental consequences to these responses. Since CTL can kill target cells and produce inflammatory cytokines, they can cause tissue damage. Such immunopathology is classically described following infection of mice with
The non-cytopathic virus LCMV. CD8 CTL
Responses are essential for the clearance of this infection. This effector response can also cause the death of the infected animal, an outcome which depends upon the route of infection, the age of the mouse, the levels of virus replication, and the tempo of the T cell response. During hepatitis B and C virus infections, virus-specific T cells infiltrate the liver, causing destruction of hepatocytes, leading to liver cirrhosis. The extent of liver injury coincides with the number of virus-specific CTL present in the liver, and pathology can be mollified by depleting hepatitis-specific CTL at the peak of the infection. This immunopathology is attributed to the cytopathic activity of the CTL, although secretion of effector cytokines and chemokines can recruit other inflammatory cells into the liver, enhancing the magnitude of tissue damage. Interestingly, during many chronic infections such as viral hepatitis or HIV, CTL lose function as they succumb to exhaustion Yi et al. 2010. This rebalancing of the effector response may limit the ability to control the infection but also may serve as a strategy to curtail immunopathology.