Article Text

Download PDFPDF

T cell responses in hepatitis C: the good, the bad and the unconventional
  1. Paul Klenerman1,
  2. Robert Thimme2
  1. 1NIHR Biomedical Research Centre, Peter Medawar Building for Pathogen Research, University of Oxford, Oxford, UK
  2. 2Department of Medicine II, University Hospital of Freiburg, Germany
  1. Correspondence to Robert Thimme, Department of Medicine II, University Hospital Freiburg, Germany; robert.thimme{at}


Over recent years, it has become increasingly accepted that virus-specific CD4+ and CD8+ T cell responses play a major role in outcome and pathogenesis of hepatitis C virus (HCV) infection. Indeed, while the emergence of strong and multispecific T cell responses may correlate with spontaneous viral clearance, the virus has developed several mechanisms to avoid T cell control in the majority of acutely HCV-infected patients that subsequently develop persistent HCV infection. In this review, we will discuss the current knowledge about the role of cellular immune responses in HCV infection. Specifically, we will emphasise recent new insights into the effector functions of T cells, possible mechanisms of their failure and the host–virus interactions occurring at the site of the disease, the liver.

  • Immunology
  • gastrointestinal immune response
  • liver disease in pregnancy
  • hepatitis C
  • liver

Statistics from


Hepatitis C virus (HCV) is a single-stranded RNA virus that has infected 140–170 million people world-wide.1 Only a minority (20%–50%) of infected individuals clear the virus spontaneously, while the majority of patients develop chronic infection. In a substantial subset of patients, chronic hepatic inflammation leads to liver fibrosis, cirrhosis and ultimately hepatocellular carcinoma. As a result, HCV-induced liver cirrhosis and hepatocellular carcinoma are among the leading indications for liver transplantation.1

The outcome of infection, that is, virus clearance versus persistence, is strongly determined by the host immune response. The immune response against HCV involves innate as well as adaptive immunity. Innate immunity against HCV includes the activation of the type I interferon response (IFNα and β) and several innate immune effector cells such as NK cells. Recently, a series of genome-wide association studies have also revealed a critical locus for outcome after acute infection linked to the IL28B (interferon λ 3) gene locus.2 Importantly, HCV has developed several strategies to overcome these responses, for example, by cleavage of important components of the type I interferon activation cascade by the viral NS3/4a protease or by inactivation of several interferon-stimulated genes.3 ,4

Adaptive immunity is mediated by the humoral as well as the cellular immune system. Indeed, the majority of HCV-infected individuals develop antibodies against HCV, irrespective of the outcome of infection. Some of these antibodies have the capacity to neutralise viral particles and may thus have an important role in limiting viral spread.5 However, viral clearance can also be observed in some patients in the absence of neutralising antibodies. Overall, although innate and humoral immune responses most likely contribute significantly to the outcome of infection, cellular adaptive immunity has attracted much interest because of its clear impact on infection outcome and potential role in disease progression.

T cell responses in HCV infection

Key roles for CD4+ as well as CD8+ T cell responses in the outcome of HCV infection have been reported. At the time of clinical presentation, fully functional virus-specific CD4+ T cell responses become detectable in patients who later clear the infection.6–8 The loss of initially strong proliferative CD4+ T cell responses has been associated with recurrent viraemia even after several months of apparent viral control,9 suggesting that maintenance of CD4+ T cell help is a prerequisite for ongoing viral control. This assumption is supported by the finding that in vivo depletion of CD4+ T cells from HCV-recovered chimpanzees abrogates protective CD8+ T cell immunity upon rechallenge, leading to viral persistence.10 These results link CD4+ T cell help to CD8+ T cell-mediated effector functions that may have a key role in the elimination of HCV. This is further supported by the following findings: First, there is a strong temporal association between viral clearance and the appearance of HCV-specific CD8+ T cell responses in the peripheral blood and liver of acutely infected humans and chimpanzees.11–14 Second, in vivo depletion studies of CD8+ T cells revealed that virus-specific CD8+ T cells are key effector cells in controlling HCV replication.15 Finally, strong associations between protective HLA alleles such as HLA-B27, HLA-B57 and HLA-A3 and spontaneous viral clearance can be mechanistically linked to dominant CD8+ T cell epitopes.16–18

Success and failure of virus-specific T cell responses

Clearly, a scenario emerges in which a complex interplay between CD4+ and CD8+ T cell responses is required for successful viral elimination. Unfortunately, in the majority of patients, these responses fail and viral persistence develops. Below, we will discuss three key questions relating to the success and failure of virus-specific CD8+ T cell responses: (a) What effector functions do CD8+ T cells use to control HCV replication? (b) What are the evidence for and the mechanism behind HCV-specific CD8+ T cell exhaustion? (c) How does viral escape contribute to HCV-specific CD8+ T cell failure?

What effector functions do CD8+ T cells use to control HCV replication?

In principle, virus-specific CD8+ T cells can eliminate virus-infected hepatocytes by two different pathways: a non-cytolytic, cytokine-mediated inhibition of replication that does not require the destruction of virus-infected cells and a cytolytic pathway mediated by perforin-granzymes and/or surface death receptors such as Fas/FasL leading to the destruction of infected cells. The question of which pathway is active was recently addressed using an in vitro model based on an HCV replicon system within Huh-7 cells stably transduced with the HLA-A2 gene, thus allowing analysis of HCV RNA replication by measurement of luciferase activity and targeting of specific HCV epitopes by T cells. In this model, HCV-specific CD8+ T cells strongly inhibit viral replication through cytolytic and non-cytolytic mechanisms in a dose-dependent manner19 (figure 1). At a high effector to target ratio (eg, 1:1, a ratio that is most likely not reached in vivo), HCV replication was completely inhibited and accompanied by clear cytotoxicity as indicated by elevated alanine transaminase (ALT) levels. However, >95% viral inhibition also occurred at low effector to target ratios (eg, 1:100) at which no cytotoxicity was observed, indicating a dominant non-cytolytic effect at low T cell frequencies. Neutralisation experiments revealed that this non-cytolytic effect was primarily mediated by interferon γ (IFNγ).19

Figure 1

Antiviral effector functions of CD8+ T cells. A hepatitis C virus (HCV)-specific CD8+ T cell (blue) may encounter HCV-derived peptide antigen expressed by an infected hepatocyte (H) in the liver in the context of MHC Class I. Triggering of the CD8+ T cell may lead to death of the cell and, as indicated on the left, release of cytokines including IFNγ, which leads to non-cytolytic clearance of the hepatocyte and its neighbours (right-hand side). In vivo, non-cytolytic clearance is thought to dominate.

Interestingly, the contribution of non-cytolytic, IFNγ-mediated antiviral effects to the natural course of HCV infection is also supported by findings in acutely infected patients and chimpanzees. During the early phase of infection, HCV-specific CD8+ T cells display an impaired ability to secrete IFNγ (a so-called ‘stunned’ phenotype) and are not able to control viraemia.14 Subsequently, the HCV-specific CD8+ T cells start to produce IFNγ; this coincides with a rapid decrease in viraemia and ALT and correlates precisely with an emergence of a HCV-specific CD4+ T cell response and finally viral elimination.13 In chimpanzees, viral clearance during the acute phase can occur in the absence of elevated ALT levels with only minimal histologic evidence of liver cell injury but with detectable IFNγ mRNA in the liver,20 further supporting the presence of non-cytolytic effector mechanisms.

However, in most patients, virus-specific CD8+ T cells fail to clear the virus and viral persistence occurs. The mechanisms of CD8+ T cell failure are not completely understood, but growing evidence suggests that CD8+ T cell exhaustion and viral escape contribute primarily to this failure.

What are the evidence for and the mechanism behind HCV-specific CD8+ T cell exhaustion?

Virus-specific CD8+ T cell exhaustion was first described in the lymphocytic choriomeningitis virus (LCMV) mouse model as loss of virus-specific CD8+ T cell responses, subsequently visualised as virus-specific tetramer-positive CD8+ T cells that are impaired in their antiviral effector functions (eg, the ability to produce cytokines).21 ,22 A sequential loss of T cell function is observed during CD8+ T cell exhaustion, for example, that functions such as interleukin (IL)-2 production, proliferative capacity and ex vivo killing are lost first, followed by an impaired ability to secrete IFNγ at a final stage of CD8+ T cell exhaustion, and subsequent deletion.22

Chronic HCV infection has also been associated with impaired function of HCV-specific CD8+ T cells. Indeed, several studies have shown that HCV-specific CD8+ T cells derived from the peripheral blood or liver display a reduced ability to proliferate or secrete antiviral cytokines such as IFNγ.23–26 The mechanisms contributing to CD8+ T cell exhaustion in HCV are not completely understood but most likely include intrinsic regulatory pathways (such as signals mediated by the inhibitory receptor PD1) and extrinsic regulatory pathways (such as immunoregulatory cytokines or regulatory T cells). Several studies have shown that HCV-specific CD8+ T cells may express inhibitory receptors, such as PD-1, CTLA-4, 2B4 or TIM-3.24 ,27–31 Interestingly, during acute HCV infection, high expression of PD-1 on virus-specific CD8+ T cells has been suggested by some groups to correlate with development of viral persistence,32 ,33 although this was not confirmed in all studies, since PD-1 may also be expressed to high levels in those who go on to clear the virus.34 Exhausted HCV-specific CD8+ T cells may be prone to apoptosis, unless rescued in cell culture by cytokines such as IL-2.35 In vitro blockade of PD-1 restores proliferation and potentially other effector functions of HCV-specific CD8+ T cells.36

However, PD-1 may not be the whole story—several studies, first in the LCMV mouse model and also in hepatitis C, have suggested that interruption of a single pathway may be insufficient for immune reconstitution, although which additional pathways are critical is not yet clear.37 For example, HCV-specific CD8+ T cells derived from liver biopsies of chronically HCV-infected patients required CTLA-4 blockade in addition to PD-1 blockade to restore their function.38 A separate study, however, has shown that the functionality of PD-1+Tim-3+ co-expressing CD8+ T cells (which are enriched within the liver) can be restored exclusively by Tim-3 blockade.28 Yet another study showed that 2B4 stimulation may counteract enhanced proliferation of HCV-specific CD8+ T cells induced by PD1 blockade, indicating that different inhibitory receptors may counterregulate each other.31 While these combined results clearly suggest that CD8+ T cell exhaustion is associated with the co-expression of several inhibitory receptors, the role of the individual inhibitory pathways in vivo is likely to be influenced by the relative expression of their ligands. It is also important to note that not all HCV-specific CD8+ T cells present in chronically HCV-infected patients display an exhausted phenotype; a recent comprehensive study performed in 38 chronically HCV-infected patients found an exhausted phenotype (characterised by the co-expression of PD-1, 2B4, CD160 and KLRG1) in about half of the detectable HCV-specific CD8+ T cells that was linked with early T cell differentiation stages.30 ,39 In the other half of detectable HCV-specific CD8+ T cell responses obtained from chronically HCV-infected patients, there was, instead, evidence of viral escape (see below).

Extrinsic regulatory pathways are also involved in HCV-specific CD8 T cell exhaustion. For example, IL-10 levels are typically increased in chronic HCV infection, and IL-10-producing T cells have been found in the livers of chronically HCV-infected patients where they may contribute to the regulation of HCV-specific CD8+ T cells.40–44 Regulatory CD4+ T cells (Tregs) have also been suggested to contribute to HCV-specific CD8+ T cell dysfunction. For example, several groups have shown that they are active in chronically HCV-infected patients compared to controls and that depletion of CD25+ cells from peripheral blood mononuclear cells (PBMCs) results in increased responsiveness of the remaining HCV-specific CD8+ T cells.45–47 In some cases, these regulatory T cells may include HCV-specific populations.48 ,49 However, whether they contribute to the development of T cell exhaustion during the acute phase of infection is currently unclear.50 Finally, it is also important to note that lack of HCV-specific CD4+ T cell help, a hallmark of chronic HCV infection, may contribute to HCV-specific CD8+ T cell exhaustion, as has been discussed above.

Overall, these results suggest that CD8+ T cell exhaustion occurs in about half of the detectable HCV-specific CD8+ T cell responses, that it is linked to the co-expression of inhibitory receptors and the action of extrinsic negative pathways (figure 2) and that restoration of T cell function most likely requires the blockade of several inhibitory pathways. The development of exhaustion is likely determined by a complex interplay of immunological and virological factors including the duration, intensity and site of antigen triggering and the state of maturation of the T cell. These results also show a close reciprocal connection between CD8+ T cell exhaustion and viral escape, since ongoing antigen triggering seems to be a prerequisite for CD8+ T cell exhaustion30 ,32 ,51 (figure 3). However, it is yet not clear whether T cell exhaustion is a cause or simply a consequence of viral persistence.52

Figure 2

Regulation and exhaustion of CD8+ T cells. As in figure 1, a hepatitis C virus (HCV)-specific CD8+ T cell (blue) may be activated after encounter with HCV-derived peptide antigen expressed by an infected hepatocyte (H) in the liver in the context of MHC Class I. However, it is subject to counterregulation through an array of inhibitory receptors and the activity of regulatory populations as indicated, including IL10 secretion. Lack of CD4+ T cell help contributes to dysfunction of CD8+ T cell populations.

Figure 3

Escape versus exhaustion of antiviral CD8+ T cells. A: A hepatitis C virus (HCV)-specific CD8+ T cell (blue) will not be activated after encounter with HCV-derived peptide antigen expressed by an infected hepatocyte (H) if the peptide has mutated under immune selection. In this setting, repetitive stimulation does not occur, and induction of an exhausted phenotype is not seen. B: In around 50% of epitopes, escape does not occur. This leads to a distinct T cell phenotype associated with exhaustion and/or ultimately depletion of the specific population.

What are the factors determining viral escape?

HCV has a high replication rate but lacks a proof-reading mechanism. As a consequence, mutations are readily generated and are subject to immune selection pressure. Importantly, HCV-specific CD8+ T cell-mediated viral escape has been reported during acute and chronic HCV infection in several studies and suggested to play a key role in CD8+ T cell failure and subsequent viral persistence.53–58 These mutations can either occur at an epitope's HLA binding anchors, preventing binding of the peptide to the HLA molecule and thus abrogating epitope presentation, or occur at positions within the epitope that are responsible for T cell receptor binding.59 In addition, some mutations may occur in close proximity to CD8+ T cell epitopes, interfering with antigen processing by the proteasome.60

As mentioned above, in chronically HCV-infected patients, viral escape mutations are consistently present in around 50% of all targeted CD8+ T cell epitopes.61–63 So how can the absence of viral escape in the presence of a virus-specific CD8+ T cell response be explained? One explanation is that certain virus-specific CD8+ T cells target epitopes with high functional constraints that do not tolerate mutations, that is, there is a high viral fitness cost. This concept is supported by several findings: First, some viral escape mutations can revert to wild type after transmission of the virus to a recipient who does not carry the HLA allele restricting the respective epitope.57 ,58 Second, HCV escape mutants emerging early in infection are not necessarily stable, but may be eventually replaced with variants that achieve a balance between immune evasion and fitness for replication.64 Third, the absence or rare occurrence of specific escape mutations in vivo has been shown to be associated with high viral fitness costs in human hepatoma cell lines containing HCV replicons, that is, there is a severe disruption of HCV replication by introducing mutations in an otherwise highly conserved epitope.16 ,59 ,65 ,66 Finally, the protective role of specific HLA-alleles such as HLA-B27 and A3 has been suggested to result from targeting key epitopes in a functional site, for example, the requirement for at least two mutations to balance fitness and escape.16 ,67

It is also interesting to note that data from large viral genome sequencing studies demonstrated that viral escape mutations occur more frequently in CD8+ T cell responses restricted by HLA-B alleles compared to those restricted by HLA-A alleles.68–71 This finding probably indicates that HLA-B-restricted CD8+ T cell responses have a stronger antiviral efficacy and thus exert stronger evolutionary pressure on the virus. These results point towards a possible role of the genetic restriction in determining emergence of viral escape mutations. Other factors include TCR diversity, avidity, holes in the TCR repertoire or lack of CD4+ help.72–74 Clearly, additional studies are required to understand the complex interplay of all these factors in determining viral escape evolution.

Finally, it is interesting to note that recent studies have shown that, in contrast to CD8+ T cells, CD4+ T cell responses are only rarely linked to the emergence of viral escape mutations.75 ,76 Thus, viral mutations play a substantial role in escape from CD8+ T cells, but failure of CD4+ T cell responses must be largely explained through other mechanisms.

T cell responses in the liver during chronic infection

Despite all these accumulated insights into the principal mechanisms of CD8+ T cell function and failure in determining acute outcome, several important areas in HCV immunobiology remain open. Specifically, very little information is currently available about the host–virus interactions established in the liver, which may continue for decades. The issue here is not just whether T cells exert control over viral load but to what extent may they protect against or contribute towards tissue pathology, which includes inflammation and fibrosis. There are many components to this complex multicellular process, and it is complicated further by the compartmentalised nature of the disease as well as its extended and unpredictable course, so it is not surprising that this area is poorly understood. Here we will address three questions: (a) What fraction of the intrahepatic infiltrate is actually antiviral? (b) What are the functions of the antiviral responses within the liver? (c) Are there distinctive characteristics of the cellular infiltrate in chronic hepatitis?

What fraction of the intrahepatic infiltrate is actually antiviral?

A number of early studies in HCV used liver-derived T cells as lines and clones to evaluate the targets and function of the antiviral T cell response, so it has been known for 2 decades that specific cell populations are present at the major site of infection.77–79 However, the quantification of this has been relatively limited. Early studies using MHC Class I tetramers revealed that there was relative enrichment of HCV-specific CD8+ T cells in liver-derived lymphocyte (LDL) pools compared to peripheral blood.80 This is not surprising necessarily as the infiltrates incorporate antigen experienced T cell populations, so there will be enrichment as a result of segregation away from naïve cells. Additionally, it is known that there is relative enrichment of cells specific for other viruses (such as CMV, EBV and influenza) even in normal livers.81 In animal models, where this is much more easily studied, it is observed that effector cells82 and effector memory pools83 can localise in the liver as in other organs when it is not inflamed, although clearly the homing signals will be substantially altered in the case of tissue inflammation of whatever cause.

In the case of HCV, the relative enrichment of tetramer-positive populations is hard to quantify, as the peripheral blood frequencies are so low (typically <0.02%). In two studies, using four pooled or single tetramers, respectively, the ex vivo population frequencies in liver were 0.06%–4% and 1%–2.5% of CD8+ T cells,80 ,84 while other studies failed to detect any response using a limited tetramer pool.85 In studies using non-specifically expanded LDL populations, the frequencies for four single tetramers were in the same range—0.03%–1.2% with a total mean of around 0.5%—enriched two–threefold compared to blood.25 Functional T cell assays can be performed using overlapping peptide pools to cover greater breadth. Interestingly these give similar results, with intracellular cytokine staining (ICS) revealing populations of 0%–5% of CD8+ T cells to be HCV specific (mean 0.4%).62 Thus, even if each individual attempt to address this is limited and provides an underestimate of HCV specific cells, it seems that the majority of CD8+ T cells in the liver are not HCV specific, and indeed this number may exceed 90%.

The data on CD4+ T cells are much more limited, largely for technical reasons as the frequencies of many HCV-specific CD4+ T cell specificities are very low in chronic infection. One study using Class II tetramers for an immunodominant response, with a detection limit of 0.001%, was unable to find any Class II tetramer + cells in LDL .7 However, a more comprehensive recent study using overlapping peptide pools with CD154 upregulation as a read-out (ie, independent of cytokine function) revealed population sizes of 1%–5% (mean 3%), enriched around threefold compared to blood.86 Thus, for CD4+ T cells, although the data are much less complete, a comparable situation is likely as for CD8+ T cells, that is, the substantial majority of liver homed cells are not specific for HCV during chronic hepatitis C.

Unfortunately, what these experiments do not tell us are the dynamics of T cell recruitment, activation, recirculation and death of antigen-specific cells. Of relevance, a recent study showed that HCV-specific CD8+ T cells undergo massive apoptosis in the liver during chronic HCV infection.35 Also, although the HCV-specific CD8+ and CD4+ T cells are a relatively small fraction of the overall cellular infiltrate, the total numbers of cells in the liver tissue may be nevertheless high, as has been calculated for HBV.87

What are the functions of the antiviral response within the liver?

As discussed earlier, evidence from cellular phenotyping suggests that there is a very high level of cellular activation among all liver-derived CD8+ T cells, associated with upregulation of inhibitory molecules such as PD-1 and CTLA-4 and with lack of cellular function in some in vitro assays.24 ,38 However, despite this, several studies have identified functional cytokine secreting CD4+ and CD8+ T cells in LDL, and fully functional virus-specific clones can be established.77 ,78 Although IFNγ would be the classical Th1-type cytokine associated with antiviral responses, and indeed is readily observed, interestingly, there are data to suggest that additional IL-10 secretion among liver homing populations can be observed for virus-specific CD8+ T cells41 and CD4+ T cells.88 ,89 FoxP3+ CD4+ T cells can be readily detected in the liver ex vivo, perhaps representing as many as 25%–30% of CD4+ T cells.90 ,91 It is suggested that these are Treg populations, although their functional profile and IL-10 secretion as well as their specificity need to be further defined. FoxP3+ CD8+ T cells can also be seen after culture and may be relevant in vivo with similar functions to their CD4+ T cell counterparts, although they are probably much less numerous.92

Little data are available on Th2 cytokine secretion by intrahepatic virus-specific T cell pools in HCV, although early data suggested a Th1/Th2 switch in the periphery93: this would be of some interest given the importance of cytokines such as IL-13 (which can be made by NKT cells) in tissue fibrosis.94 The cell lineages most recently described, secreting IL-17 and also IL-22, are described further below, but virus-specific CD8+ T cells making IL-17 are enriched in LDL, adding further to the mix.

The relationship between such cytokine profiles and disease status is not yet clear—it has been suggested that increased IL-10 secretion41 and increased IFNγ88 are associated with reduced tissue inflammation. Recently, FoxP3+ CD4+ T cells have been linked to reduced disease progression,90 while a further study linked increased IFNγ/IL17 co-expression among CD8+ T cells to reduced fibrosis stage.95 All these studies lack prospective components, so it is very difficult to attribute causality. However, one simple conclusion would be that lack of functionality of any kind, depending on the measurements made, may be a common correlate of more advanced disease.

Overall, therefore, although T cell function may be relatively impaired in the liver (through a mixture of escape and exhaustion), it is likely that diverse functional antigen-specific and (mainly) non-specific populations are nevertheless active within liver tissue. These populations secrete a mixture of antiviral, proinflammatory and regulatory cytokines, but much more work is needed to define this potential and link it to long-term outcome.

Are there distinctive characteristics of the cellular infiltrate in chronic hepatitis C?

Given that the T cell infiltrate in chronic hepatitis C is largely made of non-HCV-specific T cells and this cell population is likely modulating the antiviral response as well as playing a role in tissue inflammation, it seems important to define this in greater detail. One of the well-known features of liver-derived populations in general is the expression of NK-associated molecules.96 One of these in particular, CD161, has attracted recent interest as it is associated with liver-homing and functional potential of T cell subsets.

CD161 is a C-type lectin, whose best defined ligand is another similar molecule, LLT-1.97 It is not regarded purely as a marker of activation or maturation since expression is seen on CD4+ and CD8+ T cells in cord blood and indeed within the thymus.98 ,99 CD161 is expressed on HCV- and HBV-specific CD8+ T cells at levels higher than those on non-hepatotropic viruses, in acute and chronic infection and especially on liver-derived populations.100 Overall, nearly half of CD8+ and CD4+ T cells in the liver during chronic hepatitis C are CD161+95 (and submitted); how might this occur?

CD161 expressing CD8+ T cells may be divided into two pools, CD161++ and CD161+, based on staining characteristics.95 ,101 CD161++ cells express high levels of CXCR6,95 ,100 a molecule linked to sinusoidal surveillance under normal conditions, as well as CCR5 and CCR6, which will allow for tissue recruitment during inflammation102 ,103 (figure 4). CD161+ T cells express much lower levels of these molecules but high levels of CXCR3,95 which is triggered by CXCL9 and 10, both highly expressed during hepatitis C.104 ,105 It seems very likely that these combinations of chemokine receptors contribute to the marked enrichment of such cells in chronic hepatitis.106 This would explain the non-antigen-specific nature of the infiltrates and the similar findings in other, non-viral hepatitides such as NASH.95 Most adult CD161++CD8+ T cells express an invariant TCRα chain and are defined as mucosal-associated invariant T cells.107 The ligand for this subset is not fully defined, but these cells are restricted by the Class-I-like molecule MR1.108

Figure 4

Homing of unconventional CD8+ T cells to the liver. CD8+ T cells exist in distinct phenotypes in the periphery, including CD161- (blue), CD161+ (green) and CD161++ (red). The latter possesses a unique functional and phenotypic profile. This includes the expression of chemokine receptors as indicated, notably CXCR6, which allows homing of these cells to the liver even under non-inflammatory conditions. CD161+ and CD161- CD8+ T cells will home to the liver under inflammatory conditions via distinct chemokine signals and possess more conventional antiviral and pro-inflammatory functions. In chronic hepatitis C, CD161+/++ populations account for around half of the total CD8+ T cell infiltrate, substantially enriched from blood.

Once present in the liver, what is the role of CD161-expressing T cell populations? So far, this is not yet defined, but the CD161 expression is linked to expression of the master transcription factor RORgt,109 which is required for Th17 and Tc17 development.95 ,98 ,99 Thus, CD161++CD8+ T cells can secrete IL17 and IL22, alone or in combination with IFNγ and TNFα. CD161++CD8+ T cells appear low in perforin and granzyme B.95 To what extent the cells possess cytotoxic potential is not yet known, but equivalent murine Tc17 subsets appear to be relatively weakly cytotoxic.

The array of cytokines available to the CD161-expressing subset is quite wide, although the high frequency of such cells in liver infiltrates raises the issue of the specific roles of RORgt-related cytokines IL17 and IL22 in chronic hepatitis. These cytokines, both of which have been extensively investigated in chronic inflammatory bowel disease, are driven by IL23 signals, but are only expressed by a fraction of the CD161++ subset.98 ,110 They do not appear to have antiviral activity for HCV,111 and while both cytokines play an important role in epithelial defence against bacteria,112 ,113 they may have distinct impacts on liver pathology. IL17 secretion may lead to upregulation of further proinflammatory chemokines and cytokines on a range of cell types including hepatocytes, thus promoting inflammation and further cellular recruitment.114 IL22, on the other hand, which is in the IL-10 family, appears to have a crucial hepatoprotective role in murine models through antiapoptotic and pro-proliferative effects on hepatocytes.115 ,116 Since human Th17 and Tc17 cells can additionally produce many other chemokines and cytokines, it may be difficult to dissect out the specific individual role of any one secreted factor, but targeting of this cell population or the IL23 axis may provide a novel pathway for modulation of liver inflammation during chronic hepatitis of various causes.

A number of other unconventional T cell subsets are incorporated in the hepatic infiltrate including iNKT cells and γ delta T cells, but the numerical dominance of CD161+/++ T cells in the liver and their broad cytokine armoury point to a potentially critical role in chronic inflammation during HCV infection. This may not be so well reflected in murine models since the relevant subset is very rare in the mouse108 and iNKT cells dominate, especially in the liver.102 However, even without murine models, there are very many questions to address, such as which cytokines are present in situ and how such cells are regulated and triggered. The presence of CD161++ mucosal-associated invariant T cells in liver and gut points to important overlaps between liver and luminal inflammation—and indeed relevant to other chronic tissue infiltrates—and there may be much to learn from recent studies on the genetics, molecular and immunologic basis of inflammatory bowel disease.


Recent studies using a range of techniques have shed important light on the function of antiviral T cell responses in hepatitis C and their dysfunction (table 1). There are many remaining questions as to the predictive value of such measurements and whether induction of functional T cell responses through vaccination could provide robust protection as a prophylactic vaccine or even a therapeutic option in chronic infection. The issue of the nature of the inflammatory infiltrate in chronic infection remains much less clear, and how the various T cell and allied subsets contribute to or protect against tissue pathology is an important question for the future. Whether this knowledge could also be used for therapeutic gain is also of interest, and this would have impact well beyond HCV. Indeed, HCV still has much to teach us about immunology, and even though we now have more powerful drugs in our armoury and vaccines potentially on the horizon, it will remain a potent adversary for some time to come.

Table 1

Key immunologic players in hepatitis C

Key points

  • Virus-specific CD4+ and CD8+ T cells play a central role in outcome and pathogenesis of HCV infection.

  • CD4+ T cell help is required for successful CD8+ T cell-mediated viral clearance.

  • CD8+ T cells can inhibit HCV replication by cytolytic and non-cytolytic mechanisms.

  • In chronic HCV infection, two major pathways, T cell exhaustion and viral escape, contribute to CD8+ T cell failure.

  • T cell exhaustion is defined by impaired CD8+ T cell effector functions and characterised by co-expression of several inhibitory receptors such as PD-1, 2B4 and CD160.

  • Viral escape is not a universal mechanism and is limited by fitness cost, for example, the inability to tolerate mutations within highly constrained epitopes.

  • The HCV-specific CD4+ and CD8+ T cell response is apparently only a small fraction of the overall inflammatory infiltrate in chronic hepatitis C.

  • The CD8+ T cell infiltrate in chronic hepatitis C is dominated by T cells expressing the C-type lectin CD161—a feature associated with unconventional functionality and phenotype.



  • Funding The authors are supported by the Wellcome Trust (WT091663MA), MRC, NIHR Biomedical Research Centre, Oxford, the James Martin School for 21st Century, Oxford, the NIH NIAID 1U19AI082630-01 and the Deutsche Forschungsgemeinschaft (TH719/3-1 and FOR1202).

  • Competing interests None.

  • Provenance and peer review Commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.