Potential conflict of interest: Nothing to report.
Supported by the SFB900 “Chronic Infections: Microbial Persistence and its Control” of the German Research Foundation (DFG; project A5; reference number: 19222095) and by the Integrated Research and Treatment Center Transplantation (IFB-Tx; project 37; reference number: 01EO0802) funded by the Federal Ministry of Education and Research (BMBF).
Hepatitis E virus (HEV) infection is usually self-limited but may lead to acute hepatitis and rarely to fulminant hepatic failure. Persistent HEV infections have recently been described in organ transplant recipients receiving immunosuppressive medications, suggesting that HEV is controlled by adaptive immune responses. However, only few studies have investigated HEV-specific T-cell responses and immune correlates for the failure to clear HEV infection have not been established so far. We investigated T-cell responses against HEV in 38 subjects including anti-HEV-positive (exposed, n = 9) and anti-HEV-negative (n = 10) healthy controls, 12 anti-HEV-positive but HEV RNA-negative organ transplant recipients, and seven transplant recipients with chronic hepatitis E. Proliferation as well as cytokine production of CD4+ and CD8+ T cells was studied after stimulation with overlapping peptides spanning all proteins encoded by HEV-open reading frame (ORF)2 and HEV-ORF3. We show that (1) strong and multispecific HEV-specific T-cell responses are present in exposed healthy controls, and to a lesser extent also in recovered patients after transplantation; (2) that these responses are absent in patients with chronic hepatitis E but become detectable after viral clearance; and (3) that HEV-specific T-cell responses can be restored in vitro by blocking the PD-1 or CTLA-4 pathways. However, a combination of PD-1 and CTLA-4 blockade had no synergistic effects. We conclude that chronic hepatitis E is associated with impaired HEV-specific T-cell responses and suggest that enhancing adaptive cellular immunity against HEV might prevent persistent HEV infections. (HEPATOLOGY 2012)
The hepatitis E virus (HEV), a nonenveloped, single-stranded RNA virus, is the causative agent of acute hepatitis E. 1 Acute hepatitis E may rarely progress to fulminant hepatic failure, which more often occurs in pregnant women especially from developing countries 2 and in patients with pre-existing chronic liver diseases. 3 At least five different HEV genotypes have been described, with four of them being able to infect humans. HEV genotype 3 has frequently been associated with zoonotic infections, 4, 5 whereas HEV genotypes 1 and 2 appear to primarily infect humans. We recently confirmed the anthropo-zoonotic capacity of HEV genotype 3 by experimentally infecting pigs with a serum sample of a chronic HEV-infected patient. 6
HEV infection represents a particular problem for immunocompromised individuals, as these patients can develop persistent HEV infection. Cases of chronic hepatitis E were reported in solid organ transplant recipients, 6–10 patients with HIV infection, 11, 12 and individuals suffering from Non-Hodgkin's lymphoma. 13 In most cases, chronic HEV was reported in liver or kidney transplanted patients with a prevalence rate of 1%-2% in low endemic areas and higher prevalence in south-west France. 6–8 We identified chronic HEV infection also in heart transplant recipients. 14 Factors associated with the development of chronic HEV infection may include distinct immunosuppressive regimens such as therapy with tacrolimus. 15 Overall, chronic HEV infection is now considered as a significant clinical problem in solid organ transplant recipients associated with considerable morbidity and mortality. Clinical data suggest that immune responses are important to control the infection.
Strong and multispecific CD4+ and CD8+ T-cell responses have been shown to be of importance for the control of both hepatitis B virus (HBV) and hepatitis C virus (HCV) infections. 16–24 However, few studies investigated T-cell immunity in HEV infection. Some groups have analyzed HEV-specific cellular immune responses by screening potential T-cell epitopes in the open reading frame (ORF)2 and 3 regions of HEV describing HEV-specific lymphoproliferative responses in patients with acute hepatitis E. 25, 26 In addition, pregnant women with acute HEV infection showed a Th2-biased T-cell response 27 and a stronger reactivity against HEV ORF2 and ORF3 proteins was associated with a milder course of disease in acute and fulminant hepatitis E. 28 However, these studies had limitations, as usually only one functional readout was applied and CD4+ and CD8+ T-cell responses were not distinguished. Moreover, and importantly, no study until now has addressed the role of T-cell responses in resolving and chronic HEV infection. Thus, we here aimed to study cellular immune responses in different patient groups including organ transplant recipients with chronic and resolved hepatitis E. We show that (1) strong and multispecific HEV-specific T-cell responses are present in exposed healthy controls and to a lesser extent also in recovered patients after transplantation; (2) that these responses are absent in patients with chronic hepatitis E but become detectable after viral clearance; and (3) that HEV-specific T-cell responses can be restored in vitro by blocking the PD-1 or CTLA-4 pathways. However, a combination of PD-1 and CTLA-4 blockade was not synergistic.
Informed consent in writing was obtained from each patient included in this study. The study protocol conformed to the ethical guidelines of the Institutional Review Committee. Immune responses against HEV were studied in a total of 38 subjects including 19 healthy immunocompetent individuals and 19 immunocompromised organ recipients. The immunocompetent group included anti-HEV-positive (“exposed”) subjects (n = 9; median age 32; range 26-66) and anti-HEV-negative (“no exposure”) individuals (n = 10; median age 30; range 25-37). The immunocompromised group included transplanted patients who developed chronic hepatitis E (n = 7; median age 49; range 34-66) and transplanted patients who resolved HEV infection (n = 12; median age 53; range 27-69). Out of these 12 patients, three subjects acquired HEV infection post transplantation, one subject before transplantation, and no information on the time of HEV acquisition was available for the remaining patients. Presence of antibodies (IgG) against HEV was tested by using a commercially available enzyme-linked immunosorbent assay kit (Abbott Laboratories, North Chicago, IL) according to the manufacturer's instructions. All study subjects were negative for hepatitis B surface antigen (HBsAg) and anti-HCV except one transplant recipient with resolved HEV infection (LTxR2), being anti-HCV positive with serum HCV RNA levels of 1 Mio IU/mL. The presence of HEV RNA was confirmed by both qualitative and quantitative real-time polymerase chain reaction (PCR).
T-Cell Proliferation and Cytokine Levels.
All assays were performed as previously described. 29, 30 For details, please see Supporting Information. Briefly, T-cell proliferation was studied with the carboxyfluorescein diacetate succinimidyl ester (CFSE) method after stimulating peripheral blood mononuclear cells (PBMCs) in vitro with HEV overlapping peptide pools spanning ORF2 and ORF3 regions of the genome. Cytometric bead array (CBA) and intracellular cytokine stainings (ICS) were performed to measure cytokine levels.
Blocking PD-1 and CTLA-4 Pathways.
To determine the role of PD-1 and CTLA-4 on T-cell responses during chronic hepatitis E, cells were cultured in vitro after CFSE labeling in the presence of HEV peptide pools and by adding antihuman PDL-1 (eBioscience, San Diego, CA) and CTLA-4 (BD PharMingen, Becton Dickinson, Heidelberg, Germany) antibodies separately or in combination at a concentration of 5 μg/mL along with peptide pools. Fluorescence-activated cell sorting (FACS) stainings were performed at day 7 using CD4-PE and CD8-PE Cy7 antibodies.
The Mann-Whitney U test was applied for univariate comparison of independent continuous variables and Fisher's exact t test for discrete variables using Statistica 9.0 software (Statsoft, Tulsa, OK). P < 0.05 was considered significant.
Characteristics of Study Subjects.
A total of 38 subjects were studied including 19 organ transplant recipients and 19 immunocompetent healthy individuals. Patients with chronic hepatitis E had higher alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels at baseline as compared with resolved subjects (148 U*L−1 versus 23 U*L−1, P < 0.05; 87 U*L−1 versus 27 U*L−1, P < 0.05, respectively). Genotyping was performed in six of the seven patients with chronic HEV, which revealed HEV genotype 3 in all patients (Table 1). Individual characteristics of organ transplant recipients are shown in Table 2. Three of the seven patients with chronic infection cleared HEV after reduction of immunosuppression and three other patients became HEV RNA-negative during treatment with ribavirin. One of the four ribavirin-treated patients did not clear the virus. One transplant recipient with resolved hepatitis E (KTxR1) was treated with ribavirin during acute HEV infection. In this patient (KTxR1) T-cell responses were studied very early after acute hepatitis E. Anti-HEV IgG titers were higher in patients with chronic hepatitis E than in transplanted patients with resolved hepatitis E or seropositive healthy subjects (Supporting Information Fig. 1).
Table 1. Baseline Characteristics
Seropositive Healthy (n=9)
Seronegative Healthy (n=10)
Genotyping was only possible in six patients (all genotype 3).
Frequency and Strength of HEV-Specific T-Cell Proliferative Responses.
HEV-specific T-cell proliferative responses were investigated in all study subjects after stimulating PBMCs in vitro with HEV overlapping peptide pools for 7 days. Representative FACS plots are shown in Fig. 1A,B. Although strong and multispecific HEV-specific proliferative responses were found in most healthy seropositive subjects (7/9), T-cell responses were weaker in transplanted patients (Fig. 1C, Table 3). However, HEV peptide pools were also recognized by the majority of the anti-HEV-positive/HEV-RNA-negative transplanted patients (8/12) without a distinct dominant pattern across the different peptide pools. The strongest responses were detected in patient KTxR1, who was tested shortly after acute hepatitis E. In contrast, only two of seven organ transplant recipients with chronic hepatitis E had detectable HEV-specific CD4+ responses and only one patient showed HEV-specific CD8+ T-cell responses. In addition, the strength (average sum of stimulation index/patient) and breadth (number of recognized pools/patient) of HEV-specific proliferative responses were much lower in viremic patients as compared with both groups of HEV-recovered subjects (Table 3). No HEV-specific proliferative responses were detectable in seronegative healthy subjects. Thus, these data demonstrate a clear correlation between recovery from HEV infection and detectability of HEV-specific T-cell responses in the peripheral blood, even in patients receiving immunosuppressive medications.
Table 3. Summary of Proliferative T-cell Responses
Seropositive Healthy (n=9)
Seronegative Healthy (n=10)
Breadth, magnitude, and strength of HEV-specific T-cell responses in the different study groups.
High levels of interferon-gamma (IFN-γ) responses were observed in subjects with resolved hepatitis E (transplant or healthy seropositive) to most of the peptide pools, whereas IFN-γ production was not observed in any post-transplant patient with chronic hepatitis E (Fig. 2A). In contrast to IFN-γ levels, interleukin (IL)-10 production was found only in HEV RNA-positive patients (Fig. 2B). IL-17 production was detected in all groups with no obvious differences (Fig. 2C). In addition, intracellular cytokine staining for IFN-γ, tumor necrosis factor (TNF), and macrophage inflammatory protein (MIP)-1β was performed in a total of 23 subjects. Strong and significant IFN-γ levels were observed in both CD4+ and CD8+ T-cells of seropositive healthy subjects in response to most of the peptide pools. This was in contrast to transplanted patients with chronic or resolved HEV infection where intracellular IFN-γ responses were much weaker (Fig. 3A,D). HEV-specific TNF- and MIP-1β secretion of CD8+ T-cells is shown in Fig. 3B,C and did not reveal clear differences between the different groups of patients.
Longitudinal Investigation of HEV-Specific T-Cell Proliferative Responses.
We also had the chance to study proliferative T-cell responses longitudinally in transplanted patients with chronic HEV infection before and after HEV clearance. As indicated above, CD4+ and CD8+ T-cell responses were undetectable in five and six of seven chronic hepatitis E patients respectively at baseline (Fig. 1c). These weak HEV-specific T-cell responses could be confirmed in three subjects who were tested at a second independent timepoint when the subjects were still HEV-RNA positive (LTxC2; HTxC6; KTxC7). During further follow-up, five patients cleared HEV RNA: two of them by reducing immunosuppressive medication (LTxC1 and KTxC7) and three during treatment with ribavirin (HTxC3, HTxC4, and HTxC5). Of note, multispecific CD4+ and CD8+ T-cell responses against all different HEV peptide pools became detectable rapidly (within 4 weeks) after viral clearance in four of the five patients (Fig. 4). In patient LTxC1 HEV-specific T-cell responses appeared only 8 weeks after viral clearance. The absence of HEV-specific T-cell responses during HEV viremia but appearance shortly after clearance could indicate a possible role of T-cell responses in the control of HEV infection.
Restoration of HEV-Specific T-Cell Responses by Blocking PDL-1 or CTLA-4.
HEV-specific T-cell responses were weak or undetectable in the peripheral blood during persistent HEV infection. Thus, the next question we addressed was if these weak T-cell responses could be restored by in vitro blockade of the coinhibitory receptors PD-1 and CTLA-4.
Expression levels of these molecules was studied ex vivo in patients with chronic HEV infection, and their expression was detectable on both CD4+ and CD8+ T cells in all the patients included in the study (Supporting Information Fig. 2).
Restoration of HEV-specific CD4+ T-cell responses was observed in 4/5 patients by PD-1 blocking, whereas adding anti-CTLA-4 increased HEV-specific CD4+ proliferative responses in only one subject (Fig. 5). Of note, the combination of PDL-1 and CTLA-4 antibodies did not further enhance CD4+ T-cell proliferation in most subjects. In contrast, blocking both PD-1 and CTLA-4 pathways together seemed to be even counterproductive in subjects LTxC2, HTxC6, and KTxC7, as the increased proliferation induced by PD-1 blockade alone was diminished by the combination. Also, for HEV-specific CD8+ T-cell responses the effects of blocking coinhibitory receptors in vitro was diverse and varied between patients (Fig. 5). Two subjects responded to adding PDL-1 antibodies to the culture, whereas patient KTxC7 showed an increased proliferation of CD8+ T-cells by blocking CTLA-4 only. Again, the combination of blocking PD-1 and CTLA-4 pathways showed synergistic effects in only one individual (LTxC2) (Fig. 5). Thus, HEV-specific T-cell responses could be restored in vitro by blocking coinhibitory receptors to some extent in all patients. However, there was a considerable interindividual variability of the distinct effects of anti-PDL-1 and anti-CTLA-4 antibodies, including intraindividual differences between CD4+ and CD8+ T-cells.
Chronic hepatitis E has been recognized as an increasing clinical problem in immunocompromised patients since several groups across Europe and North America reported cases of progressive severe liver disease associated with HEV infection. 7, 10, 15 Defining immune correlates for the failure to clear HEV infection could therefore be of importance, in particular as therapeutic options for chronic hepatitis E are still limited. 8, 15 Even though ribavirin has recently proven some efficacy against HEV, 31, 32 the potential side effects of ribavirin treatment may limit its use in some groups of organ transplant recipients.
The present study is the first investigating HEV-specific T-cell responses in patients with persistent HEV infection. In line with our expectations and with data obtained in other chronic viral infections, 33 both HEV-specific CD4+ and CD8+ T-cell responses were rather weak in individuals with persistent HEV viremia, whereas strong and multispecific HEV-specific T-cell responses were detectable in most but not all seropositive individuals. Even though the differences were evident, results were not uniform in all subjects. Importantly, T-cells not only proliferated in response to stimulation with HEV peptides but also produced INF-γ, which is believed to be one of the key cytokines to suppress replication of viruses. We applied an unbiased HLA-independent technique to study T-cell responses using overlapping peptide pools spanning two of the three HEV ORFs. 29 HEV-ORF2 and -ORF3 are relatively conserved and thus we expected to detect most of the T-cell responses present in these patients. However, it has to be considered that the chronically infected patients were infected with HEV genotype 3, whereas the peptides used were derived from genotype 1. Still, HEV-specific responses could be restored in vitro in these patients carrying HEV genotype 3 using genotype 1-derived peptides when antibodies were added to block coinhibitory pathways. Moreover T-cell responses became detectable directly ex vivo once the patients had cleared HEV. Thus, these data suggest that T-cell epitopes are largely conserved across HEV genotypes. This would be in agreement with clinical findings from the recent large scale phase III vaccine trial demonstrating that an HEV genotype-1 derived vaccine protected from HEV genotype 4 infections in China. 34
Previous studies investigated HEV-specific T-cell responses during acute hepatitis E and in HEV-resolved subjects. 25, 26 Our findings are in line with these earlier data confirming T-cell responses in recovered individuals. Our results indicate that the memory T-cell responses against HEV were much stronger than the T-cell responses detectable during or after acute hepatitis B or C, 29, 35, 36 even though most seropositive healthy control subjects had most likely been exposed several decades ago. Of note, HEV-specific T-cell responses were also detectable in the majority of HEV-recovered organ transplant recipients receiving immunosuppressive medications. However, the strength of T-cell responses was much weaker in recovered patients after transplantation and, not unexpectedly, heart-transplanted patients who received immunosuppression with three different classes of drugs showed generally the weakest T-cell responses. In one organ transplant recipient (KTxR1), we were able to study T-cell responses very early after acute HEV infection. This patient indeed showed the strongest T-cell responses among all HEV RNA-negative immunosuppressed patients included in this study, further supporting a potential role of HEV-specific T-cell responses to control HEV infection.
The pattern of HEV-specific cytokine responses was very well in line with general immunological concepts on the regulation of T-cell responses in viral infections. 37 IFN-γ and MIP-1β secretion indicating a Th1-type response were mainly found in seropositive HEV RNA-negative controls, whereas IL-10 indicating a Th2-type response was only found in patients with chronic hepatitis E who did not show HEV-specific T-cell proliferation. Moreover, weak CD4+ and CD8+ proliferative responses in HEV viremic subjects could be restored in part in vitro by blocking the PD-1 or CTLA-4 pathways. High levels of PD-1 expression on both CD4+ and CD8+ T-cells have been reported in different chronic viral infections in mice 38 and humans including HIV, 39, 40 HBV, 41, 42 and HCV. 43–45 Some studies found that blocking the PD-1 pathway can restore in part the function not only of CD8+ but also of CD4+ T-cells. 39, 46 PD-1 was indeed expressed on both CD4+ and CD8+ T cells in patients with chronic hepatitis E. However, it is important to note that blocking PD-1 alone was not able to recover T-cell functionality in all patients but blockade of another inhibitory molecule, CTLA-4, led to increased T-cell proliferation in “PD-1-resistant” patients. Interestingly, the combination of anti-PDL-1 and anti-CTLA-4 antibodies had no synergistic effects but frequently diminished the positive effects of PD-1 or CTLA-4 blocking. This observation is well in line with our recent findings in patients with chronic hepatitis C where we also found that targeting two different costimulatory or coinhibitory receptors had no synergistic but rather counteractive effects. 30 Similarly, nonredundant roles for the CTLA-4 and PD-1 pathways have been described in driving T-cell exhaustion in chronic hepatitis B. 47 Overall, these data indicate an individual “private” costimulatory receptor usage of T-cells during viral infections. We even observed interindividual differences between HEV-specific CD4+ and CD8+ T-cells as, for example, PD-1 blocking induced a robust restoration of CD4+ T-cell proliferation in patient KTxC7, whereas anti-CTLA-4 was required to increase expansion of CD8+ T-cells in the same patient (Fig. 5). We suggest that the concept of private individual receptor usage should be considered when therapeutic attempts are made to target molecules such as PD-1 or CTLA-4.
Future studies should aim to investigate in more detail possible correlations between HEV-specific T-cell responses and clinical disease activity or the outcome of therapeutic interventions. We suggest that patients with detectable T-cell responses may not necessarily require antiviral treatment but could be observed for spontaneous viral clearance before interferon alpha or ribavirin treatment is initiated.
In conclusion, chronic hepatitis E is associated with impaired HEV-specific T-cell responses which can be restored in vitro by blockade of coinhibitory receptors. We suggest that enhancing adaptive cellular immunity against HEV might prevent persistent HEV infections.