Update on HCV-specific immunity

Abstract

Purpose of review

The goal of this manuscript is to review key recent findings related to the immunopathogenesis of HCV infection, especially in regards to T lymphocytes. It aims to complement other reviews in this issue on the roles of host genetics (IL-28B), acute HCV infection (when disease outcome is determined) and other factors that may influence fibrosis progression (microbial translocation). The main focus is on specific immunity and T cells in the context of success and failure to control viral infection.

Recent findings

This review focuses in on two areas of intense interest in the recent literature: (1) the relationship between the human leukocyte antigen (HLA), class I restricted T cell responses and the evolution of the virus and (2) the role of inhibitory markers on T cells in the immunopathogenesis of HCV. When appropriate, we compare findings from studies of HIV-specific immunity.

Summary

From examining the virus and the mutational changes associated with T cell responses and from analyzing the markers on T cells, there have been numerous advances in the understanding of immune evasion mechanisms employed by HCV.

Introduction

Since the discovery of the HCV virus as the cause of non-A, non-B hepatitis and the finding that a minority can clear the virus while the majority cannot, there has been an intensive search for the “key” to successful immunity. From the observations that CD4 and CD8 T cells are somehow critical for a successful control of virus, the conclusion is that induction of a brisk T-cell response may be the key.(1) However, the deeper that one examines this strategy, the more it becomes clear that the virus possesses tremendous diversity that hampers specific immunity, and utilizes numerous mechanisms of evading the immune system, including induction of T cell exhaustion. Thus our more recent understanding has been shaped by examinations focused on (1) the host-virus interaction, in part by analyzing the relationship between adaptive immunity and viral evolution; and (2) the characteristics of specific T-cells, especially regarding inhibitory markers that are relevant for control of other viral infections. Understanding of the immunopathogenesis of HCV will be further shaped by integration of these stimulatory and inhibitory signals, with continued hopes of eventual breakthroughs in designing immunotherapeutic strategies.(2)

Two extremely wily viruses and the host immune response

HCV co-infection is often encountered in those with HIV-1, as about a third of all HIV-positive patients worldwide are coinfected with HCV. The majority of individuals infected with either virus progress to the chronic phase of infection, characterized by constant high-level viral production and eventual disease. A certain percentage of infected individuals are able to control the virus via mechanisms that involve interplay between various arms of the immune response. When examining the peripheral blood of those who cleared HCV, the presence of long-lived specific memory CD4 and CD8 T cells is characteristic.(3) This is consistent with a model where HCV-specific responses may be similar in breadth and magnitude upon their induction but with viral escape or subsequent exhaustion in the face of continued antigen exposure, T cell responses decline if chronically infected.(4) Furthermore, when these memory responses are depleted (eg, from antibody-mediated depletion in chimpanzees) the consequences include loss of the protective immunity that is normally observed in those with prior clearance.(5, 6) In contrast, HIV-1-specific T cell interferon gamma (IFN-g) responses measured years after initial infection do not correlate with control and often broaden over time.(7) A given HIV-1-infected individual with high viral load and low CD4 counts due to disease progression may harbor a significant proportion of the CD8 T cells specific to HIV-1(8) while HCV-specific responses during the chronic phase are relatively much smaller in magnitude, especially when CD4 T cell counts are lower.(9–12) These responses may be preserved if antiretrovirals are initiated at higher CD4 counts(13) and may possibly be restored on therapy as highlighted by a recent longitudinal study.(14)

Lessons learned from immunogenetic studies

Recent genome-wide association studies examined predictors of successful antiviral treatment during the chronic phase of infection and all found a variety of single nucleotide polymorphisms (SNPs) upstream of the gene that encodes interleukin-28B (IL-28B).(15–17) These SNPs are also strongly linked to spontaneous clearance of HCV(18) suggesting shared mechanisms between interferon-induced clearance in the chronic phase and spontaneous control in the acute phase of HCV. Since a parallel report in this issue focuses on the role of IL-28B and because a clear link to T-cell immunity has yet to be described, we do not delve into further detail regarding these seminal observations.

Other clues have emerged from candidate gene studies examining spontaneous clearance of HCV as an outcome. Supporting the notion that adaptive immune responses are important in clearance, both class I and class II genes that define the specificity of T cell responses within the human leukocyte antigen (HLA) are known to be determinants of outcome. For class I, Thio et al. found in a cohort of individuals from the United States infected with various strains that HLA-Cw*04 was associated with chronic infection, while HLA-B*57 and HLA-A*11 were each associated with clearance.(19) In contrast, McKiernan et al. found while examining the outcomes and viruses of women infected with a single-source of HCV in Ireland that HLA-B*27 was the allele with the strongest associations with clearance, others being HLA-A*03 and HLA-Cw*01.(20) A more recent U.S. based study examining women with mixed HCV strains found HLA-B*57 and HLA-Cw*01 to be associated with viral clearance.(21) Our own examination of a cohort of 346 individuals found HLA-B*57 to be the class I gene most associated with spontaneous control of HCV.(22)

Two of the strongest or most consistent signals from the above studies, HLA-B*27 and HLA-B*57, are not only associated with clearance of HCV but also with control of HIV-1.(23) A recent genome-wide association study examining control of HIV-1 found that the SNPs most associated with clearance with the HLA class I loci shared a commonality of particular amino acids within the binding groove of the protein, suggesting that the HLA-peptide interaction is key to the mechanism underlying these genetic associations.(24) These shared associations of HLA-B*27 and HLA-B*57 imply a shared mechanism of control between two variable viruses, likely involving specific T-cells and this HLA-peptide interaction.(22, 25, 26)

Class I HLA mediates a substantial proportion of the evolution of both HIV-1 and HCV on an amino acid level. Mutation of the virus and associated loss of T cell function have been shown both by longitudinal sequencing during the acute stage of disease(27–30) and by cross-sectional analysis of viruses by correlation of mutational changes with HLA class I.(31, 32) In recent months, there have been several studies published examining this concept where mutational changes are enriched in either described or predicted epitopes in patients with the corresponding HLA type(33) and immunological escape was shown using functional assays.(34, 35) In certain cases, these changes may interfere with drug efficacy by inducing mutations associated with resistance to novel inhibitors of HCV.(36, 37) Mutational changes may also confer fitness costs to the virus by targeting key structural and/or functional areas of the protein, requiring compensatory changes to maintain the virus’ replicative fitness.(35, 38) Over time HLA-mediated mutations may have become more common in a population, resulting in loss of protective effect of certain alleles as suggested from studies of HIV-1.(39) Thus, the greater diversity of genotypes of HCV relative to clades of HIV may affect HLA-mediated protection on a genotype(40) or even subtype level.(22) As genotypes and subtypes vary by region, these studies help to explain differences in specific HLA associations from various cohorts.(19–22)

The role of class II HLA in determining the outcome of HCV has been established for some time; a recent review comprehensively examines these associations.(41) These findings highlight the critical role of CD4 T cells in the clearance of HCV and yet little is known as to why certain HLA restrictions of CD4 T cell responses would correlate with better outcomes. Reasons for the limited information include: (1) fewer clues from HIV-1, where class I associations dominate class II,(24) (2) the tools used to examine these responses (namely tetramers) have been more limited, and (3) there is little evidence that these specificities mediate significant evolution of the virus.(42, 43) Nonetheless, understanding both successful and failing CD4 T cell responses is a critical area of future study as it is potentially more relevant than CD8 T cells especially if the favorable effects are less dependent on viral strain as compared to class I responses.

Many of the referenced studies have examined viral isolates in cross-sectional fashion and only the predominating sequences of HCV. As we continue to learn from studies of viral evolution and immune responses, further lessons will be gleaned from longitudinal studies of acute infection and examining viral diversity and evolution by harnessing even more sensitive techniques.(44)

T cells are both inhibited and activated during chronic HIV-1 and HCV

T cell responses to foreign antigens are initiated when T cells are primed by antigen presenting cells (APC); the recognition of the antigen in the context of MHC on the cell surface by TCR represents the first signal that triggers T cell activation and leads to activation and differentiation of CD4 and CD8 T cells.(45) A second signal is needed to promote T cell survival, cytokine-mediated clonal expansion and progression to maturation.(46) Positive and negative signals exchanged upon encounter of APC and T cells are needed first in order to promote efficient responses to foreign antigens while successively limiting them to avoid additional immune damage or autoimmunity. Co-inhibitory molecules and transduction of their signals modulate the process of T cell activation by limiting T cell proliferation and survival. Co-inhibitory molecules typically belong to the Ig superfamily and include many molecules as PD-1, CTLA-4, and BTLA.(47–49)

In recent years, several molecules have been discovered that govern the exhaustion of T cells during chronic viral infections.(50, 51) T cell exhaustion is a hierarchical process initially characterized in the LCMV model in mice that consists of a progressive loss of different functions of T cells leading to the establishment of chronic infection. Exhausted T cells can be deleted as a result of expression of apoptotic factors and the inability to respond to cytokines. Exhausted CD8 T cells in the LCMV model in mice express high levels of different markers as PD1, 2B4, LAG-3, and CD160, each exerting inhibitory capacities.(52) The same findings have been extended to several human viral infections with persistence of antigen, such as HIV-1 and HCV. The role of exhausted T cells is assuming more importance and the blockade of inhibitory receptors could represent a novel treatment strategy also in the context of chronic viral infections.

The role of exhaustion in CD4 T cells is less clear, but the phenomenon has been observed in different viral infections.(53–58) As mentioned above, a major limitation for the description of CD4 T cells in chronic infections (especially HCV infection) is their relative paucity or absence once chronic infection is established, preventing non-functional analysis via tetramer staining.

Here we review the main inhibitory receptors described in the mouse model that have also been investigated in human chronic viral infections and in particular in the context of HCV and HIV-1 infection.

PD-1

Programmed Death-1 (PD-1) is the best characterized inhibitory receptor associated with exhaustion, is a member of the CD28 family of immune modulators, and is upregulated on CD4 and CD8 T cells upon activation.(59) PD-1 regulates T cell dysfunction during LCMV in mice and PD1 high T cells exhibit an intense exhausted gene signature. PD-1 binds to its ligands programmed death ligand-1 (PD-L1) and programmed death ligand-2 (PD-L2) and ligation of PD-1 results in dephosphorylation of signaling molecules downstream of the TCR, dampening T cell sensitivity to antigenic stimulation.(60)

The expression of PD-1 on T cells has been extensively studied in human viral infections. In HIV-1 infection, the expression levels of PD-1 have been correlated with disease progression.(61) HIV-specific CD4 T cells co-express PD-1 and CTLA-4 and their coexpression leads to the establishment of even more dysfunctional specific T cells.(62) PD-1 expression on T cells correlates with the level of viral antigens, particularly during acute viral infections. Our data suggest that these high levels of PD-1 during acute HCV infection may not correlate with the outcome of infection.(63) Interestingly, the blockade of PD-1/PD-L1 interaction by anti PD-L1 specific antibodies increases virus-specific T cell proliferation to both HCV and HIV-1.(63–67) Similar effects on proliferation and cytokine release have been observed in vitro in chronic hepatitis B.(68) These studies have been extended to HCV-specific CD4 T cells in chronic HCV infection, where PD-1 is associated with HCV-specific CD4 T cell exhaustion and PD-L1/2 neutralization restores CD4 T cell function.(69)

However, a recent study in healthy individuals found that PD-1 high CD8 T cells do not always exhibit other typical inhibitory markers of exhaustion observed in antigen-specific cells nor exhibit altered cytokine secretion.(60) The study concludes that PD-1 high CD8 T cells in healthy humans are effector memory cells, reflecting a heterogeneous set of antigen-experienced cells. This interesting study also indicates that in healthy individuals the expression of high levels of PD-1 reflects a state of immune activation and not necessarily of exhaustion. Thus, in human viral infections the expression of PD-1 and other inhibitory markers could be induced by the presence of the virus and could reflect a state of activation and not necessarily be the signature of an immune failure.

2B4

CD244 or 2B4 was also described recently as an inhibitory molecule in exhausted CD8 T cells in LCMV(50) and is predominantly expressed on NK cells, memory CD8 T cells, basophils and monocytes.(70–73) On CD8 T cells, 2B4 interacts with CD48 as a high affinity ligand to mediate both activatory and inhibitory signals.(74) 2B4 has been described as inhibitory receptor in the context of chronic hepatitis B on the basis of: a higher expression on HBV-specific CD8 T cells, a higher expression on liver-derived CD8 T cells and the association of viral clearance with decrease of 2B4 levels.(75) 2B4 expression on T cells is increased in HIV-1 infection and associated with disease progression,(76) and manipulation of this pathway enhanced proliferation of HIV-specific CD8 T cells.(77) In a recent study, HCV-specific CD8 T cells showed increased expression of 2B4 both in acute and chronic HCV infection, and 2B4 cross-linking led to both inhibition and activation of HCV-specific CD8 T cells while 2B4 stimulation may counteract enhanced proliferation of HCV specific CD8 T cells induced by PD-1 blockade. This study suggests an important role of 2B4 in the network of costimulatory/inhibitory molecules that regulate CD8 T cell function in acute and chronic HCV infection.(78)

LAG3

Lymphocyte activation gene 3 (LAG-3 or CD223) is a CD4 homologue that plays a down–regulatory role on T-cell responses. LAG-3 is an activation induced TCR receptor that binds MHC class II with high affinity.(79–81). LAG-3 is expressed on B lymphocytes, NK cells, γδT cells and activated and regulatory CD4 and CD8 T cells.(82, 83) LAG-3 cross-linking on activated human T cells induces T-cell functional unresponsiveness and inhibition of TCR-induced calcium fluxes.(84) LAG-3 has been also associated with immune exhaustion of CD8 T cells in LCMV infection.(52) Recent studies identified an up-regulation of the expression of LAG-3, among other markers, in T cells of humans with rapid progression of HIV infection as well as in rhesus macaques with SIV.(85) Thus far, in the context of chronic HCV infection, LAG3 does not seem to be upregulated on HCV-specific CD8 T cells in contrast to the increased expression of other inhibitory markers such as 2B4, CD160 and KLRG1 in association with PD-1.(86)

CD160

Glycosylphosphatidylinositol (GPI)-anchored CD160 protein is expressed mainly on CD8 T cells, natural killer (NK) T cells, and NK cells and a small percentage of CD4 T cells.(87–90) Exhausted CD8 T cells during chronic infections exhibit a specific increase of the expression of CD160 mRNA compared to naïve or memory CD8 T cells.(50) It has been recently discovered that CD160 functions also as an inhibitor of CD4 T cell activation, separately from CD8 activation, by interacting with its ligand the HVEM (herpes virus entry mediator).(91) The cross-linking of CD160 with mAb to CD160 strongly inhibits CD4 T cell proliferation and cytokine production(91) and profoundly inhibits anti-CD3/anti-CD28 induced CD4 and CD8 T cell activation.(92) Recently in the context of HIV-specific CD8 T cells, co-expression of CD160 with 2B4 on HIV-Gag specific CD8 T cells was much higher compared to HIV-Gag specific CD4 T cells, suggesting differential expression of inhibitory molecules that are associated with exhaustion between CD4 and CD8 T cells.(93)

TIM-3

T cell immunoglobulin and mucin domain-containing molecule 3 (TIM-3) is yet another marker described as an important factor in the development of chronic infections, including HCV. It was initially identified on terminally differentiated Th1 cells in mice(94) and recently described in human viral infections where cells that express high levels of TIM-3 display a dysfunctional phenotype and reduced cytokine production.(58, 95, 96)

McMahan and colleagues recently published the most comprehensive study of TIM-3 in the context of HCV infection.(97) Dual TIM-3/PD-1 expressing CD8 T cells during acute infection were associated with the development of persistence, are central memory in phenotype and are enriched in the liver of HCV infected patients. Interestingly, the blockade of PD-1 or TIM-3 enhances the proliferation of HCV-specific CD8 T cells while cytotoxicity is predominantly increased only by TIM-3 blockade. This study suggests a novel role for TIM-3 in identifying a subset of particularly exhausted HCV-specific PD-1+ T cells and identifying also a differential involvement of PD-1 and TIM-3 in mediating functional exhaustion.

Modulating inhibitory and activating T cell signals for HCV infection

A balance between costimulatory and inhibitory signals governs the activation and the function of T cells and thereby affects the outcome of chronic infections. The B7 ligands CD80 and CD86 can be expressed on activated T cells and their expression could be important in the interaction between T cells.(98) A recent investigation of the role of B7 ligand expression in patients with acute HCV infection found higher levels of CD86 during acute HCV infection linked to effective TCR stimulation with sufficient IL-2 signaling. The analysis of liver-infiltrating lymphocytes from patients with chronic HCV infection showed that intrahepatic CD8 T cells maintain the expression of activation markers like CD69, CD38 and HLA-DR but not CD86. This study improves the understanding of the deficit of the T cell stimulation at the site of HCV infection, as the early loss of supportive cytokine stimulation due to the lack of sufficient T cell stimulation could be a key mechanism that leads to chronic infections. CD86 expression during acute infection may require sufficient IL-2 signaling and its early loss favors the development of chronic HCV infection due to the persistent negative signals via inhibitory receptors as PD1.(99)

T regulatory cells (Tregs) may play a key role in containing the activity of effector T cells. A recent study observed a significant increase of Tregs in co-infected individuals compared to monoinfected individuals but without a correlation between levels of Tregs and liver fibrosis staging.(100) Another study investigating the role of Tregs in the liver compartment in chronic HCV infection showed an accumulation of CD4 Tregs infiltrating the hepatic lymphoid aggregates, portal tracts and parenchymatous lobules or nodules; thus Tregs in the liver could directly interact with cytotoxic T cells and thereby decrease proinflammatory secretions.(101)

Other authors evaluating the role of Tregs in HCV infection found that early interleukin-10 (IL-10) predominant responses are present in patients who progress to chronic infections.(102) The authors suggest a central role for IL-10 in the suppression of antiviral immunity via the inhibition of cytokine secretion by effector T cells, as lower IFN-g and IL-2 production accompanied these high IL-10 responses. IL-10 could interfere with viral clearance by driving naïve CD4 T cells to become Tregs, thus promoting viral persistence.(103)

The failure of T cells to control infection may depend on the co-expression of multiple inhibitory molecules, the density of surface expression, the presence or absence of costimulatory molecules, the coregulation of multiple negative regulatory pathways and interactions with Tregs. The detailed function and downstream cellular pathways of these inhibitory molecules are still not completely known in human chronic infections. Thus, the next wave of studies should address the significance of co-expression of different inhibitory molecules on HCV-specific CD4 and CD8 T cells, distinguish true inhibition from activation, and further define the function of these cells after blockade. Modulation of these stimulatory and inhibitory pathways may thus be a unique approach to future T cell based immunotherapies, balancing enhanced T cell functionality against pathogens versus induction of autoimmunity.

CONCLUSION

We have reviewed two areas of recent intense interest regarding the immunopathogenesis of the spontaneous clearance of HCV infection, namely evasion of the T cell response via mutational escape in class I-restricted epitopes and the examination of novel inhibitory receptors. Further understanding of T cells remains highly relevant due to their role in viral evolution and their potential to be modulated by regulating the balance between inhibitory and stimulatory signals.

KEY POINTS.

• T cell responses are critical for the spontaneous resolution / clearance of HCV infection

• Certain HLA class I alleles are associated with clearance and particular immune responses; the virus evolves via mutational escape

• Recently described inhibitory and stimulatory receptors regulate T cell responses and their modulation may be harnessed for novel immunotherapies

Abbreviations used in this manuscript

APC

antigen presenting cells

BTLA

B- and T-lymphocyte attenuator

CTLA-4

cytotoxic T-Lymphocyte Antigen 4

HBV

hepatitis B virus

HLA

human leukocyte antigen

γδT cells

gamma-delta T cells

ICS

intracellular cytokine staining

IL-10

interleukin-10

IL-28B

interleukin-28 beta-subunit

IFN-g

interferon-gamma

MHC

major histocompatibility complex

LCMV

lymphochoriomeningitis virus

mAb

monoclonal antibody

NK

natural killer

PD-1

programmed death-1

PD-L1

programmed death ligand-1

SIV

simian immunodeficiency virus

TCR

T cell receptor

TIM-3

T cell immunoglobulin and mucin domain-containing molecule 3

Tregs

T regulatory cells