Advertisement

Mechanisms of CD8+ T-cell failure in chronic hepatitis E virus infection

  • Janine Kemming
    Affiliations
    Department of Medicine II (Gastroenterology, Hepatology, Endocrinology and Infectious Diseases), Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany

    Faculty of Biology, University of Freiburg, Freiburg, Germany
    Search for articles by this author
  • Swantje Gundlach
    Affiliations
    Institute of Medical Microbiology, Virology and Hygiene, University Medical Center Hamburg-Eppendorf (UKE), Hamburg, Germany

    German Center for Infection Research (DZIF), Hamburg-Lübeck-Borstel-Riems, Germany
    Search for articles by this author
  • Marcus Panning
    Affiliations
    Institute of Virology, Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany
    Search for articles by this author
  • Daniela Huzly
    Affiliations
    Institute of Virology, Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany
    Search for articles by this author
  • Jiabin Huang
    Affiliations
    Institute of Medical Microbiology, Virology and Hygiene, University Medical Center Hamburg-Eppendorf (UKE), Hamburg, Germany

    German Center for Infection Research (DZIF), Hamburg-Lübeck-Borstel-Riems, Germany
    Search for articles by this author
  • Marc Lütgehetmann
    Affiliations
    Institute of Medical Microbiology, Virology and Hygiene, University Medical Center Hamburg-Eppendorf (UKE), Hamburg, Germany

    German Center for Infection Research (DZIF), Hamburg-Lübeck-Borstel-Riems, Germany
    Search for articles by this author
  • Sven Pischke
    Affiliations
    German Center for Infection Research (DZIF), Hamburg-Lübeck-Borstel-Riems, Germany

    Institute of Virology, Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany
    Search for articles by this author
  • Julian Schulze zur Wiesch
    Affiliations
    German Center for Infection Research (DZIF), Hamburg-Lübeck-Borstel-Riems, Germany

    Department of Medicine, University Medical Centre Hamburg-Eppendorf, Hamburg, Germany
    Search for articles by this author
  • Florian Emmerich
    Affiliations
    Institute for Transfusion Medicine and Gene Therapy, Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany
    Search for articles by this author
  • Sian Llewellyn-Lacey
    Affiliations
    Division of Infection and Immunity, Cardiff University School of Medicine, Cardiff, United Kingdom
    Search for articles by this author
  • David A. Price
    Affiliations
    Division of Infection and Immunity, Cardiff University School of Medicine, Cardiff, United Kingdom

    Systems Immunity Research Institute, Cardiff University School of Medicine, Cardiff, United Kingdom
    Search for articles by this author
  • Yakup Tanriver
    Affiliations
    Department of Medicine IV (Nephrology and Primary Care), Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany
    Search for articles by this author
  • Klaus Warnatz
    Affiliations
    Department of Rheumatology and Clinical Immunology, Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany
    Search for articles by this author
  • Tobias Boettler
    Affiliations
    Department of Medicine II (Gastroenterology, Hepatology, Endocrinology and Infectious Diseases), Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany
    Search for articles by this author
  • Robert Thimme
    Affiliations
    Department of Medicine II (Gastroenterology, Hepatology, Endocrinology and Infectious Diseases), Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany
    Search for articles by this author
  • Maike Hofmann
    Affiliations
    Department of Medicine II (Gastroenterology, Hepatology, Endocrinology and Infectious Diseases), Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany
    Search for articles by this author
  • Nicole Fischer
    Affiliations
    Institute of Medical Microbiology, Virology and Hygiene, University Medical Center Hamburg-Eppendorf (UKE), Hamburg, Germany

    German Center for Infection Research (DZIF), Hamburg-Lübeck-Borstel-Riems, Germany
    Search for articles by this author
  • Christoph Neumann-Haefelin
    Correspondence
    Corresponding author. Address: Department of Medicine II, Freiburg University Medical Center, Hugstetter Straße 55, 79106 Freiburg, Germany.
    Affiliations
    Department of Medicine II (Gastroenterology, Hepatology, Endocrinology and Infectious Diseases), Freiburg University Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany
    Search for articles by this author
Open AccessPublished:May 27, 2022DOI:https://doi.org/10.1016/j.jhep.2022.05.019

      Highlights

      • Persistent HEV genotype 3 infection occurs in about 50-70% of immunosuppressed patients.
      • HEV-specific CD8+ T cell responses in self-limiting HEV infection were vigorous and formed a functional memory.
      • Chronic HEV infection was associated with diminished HEV-specific CD8+ T cell responses that displayed features of exhaustion.
      • In a minority, CD8+ T-cell driven viral escape contributed to the failure of the HEV-specific CD8+ T-cell response.

      Background & Aims

      In immunosuppressed patients, persistent HEV infection is common and may lead to cirrhosis and liver failure. HEV clearance depends on an effective virus-specific CD8+ T-cell response; however, the knowledge gap around HEV-specific CD8+ T-cell epitopes has hindered analysis of the mechanisms of T-cell failure in persistent infection.

      Methods

      We comprehensively studied HEV-specific CD8+ T-cell responses in 46 patients with self-limiting (n = 34) or chronic HEV infection (n = 12), by epitope-specific expansion, functional testing, ex vivo peptide HLA class I tetramer multi-parametric staining, and viral sequence analysis.

      Results

      We identified 25 HEV-specific CD8+ T-cell epitopes restricted by 9 different HLA class I alleles. In self-limiting HEV infection, HEV-specific CD8+ T cells were vigorous, contracted after resolution of infection, and formed functional memory responses. In contrast, in chronic infection, the HEV-specific CD8+ T-cell response was diminished, declined over time, and displayed phenotypic features of exhaustion. However, improved proliferation of HEV-specific CD8+ T cells, increased interferon-γ production and evolution of a memory-like phenotype were observed upon reduction of immunosuppression and/or ribavirin treatment and were associated with viral clearance. In 1 patient, mutational viral escape in a targeted CD8+ T-cell epitope contributed to CD8+ T-cell failure.

      Conclusion

      Chronic HEV infection is associated with HEV-specific CD8+ T-cell exhaustion, indicating that T-cell exhaustion driven by persisting antigen recognition also occurs in severely immunosuppressed hosts. Functional reinvigoration of virus-specific T cells is at least partially possible when antigen is cleared. In a minority of patients, viral escape also contributes to HEV-specific CD8+ T-cell failure and thus needs to be considered in personalized immunotherapeutic approaches.

      Lay summary

      Hepatitis E virus (HEV) infection is usually cleared spontaneously (without treatment) in patients with fully functioning immune systems. In immunosuppressed patients, chronic HEV infection is common and can progress rapidly to cirrhosis and liver failure. Herein, we identified the presence of HEV-specific CD8+ T cells (a specific type of immune cell that can target HEV) in immunosuppressed patients, but we show that these cells do not function properly. This dysfunction appears to play a role in the development of chronic HEV infection in vulnerable patients.

      Graphical abstract

      Keywords

      Linked Article

      • The pivotal role of CD8+ T cells in hepatitis E virus infection
        Journal of HepatologyVol. 77Issue 4
        • Preview
          CD8+ T lymphocytes or cytotoxic T lymphocytes express a range of effector molecules that participate in adaptive host defense and immune memory against pathogens and, upon antigen-dependent stimulation, eliminate infected cells by secreting granzymes and perforins.1 Importantly, divergent outcomes of many viral infections (i.e. HCV; SARS-CoV-2) among individual patients have been linked to differences in virus-specific CD8+ T-cell responses.2,3 Optimal activation and priming of naïve CD8+ T cells requires recognition of distinct viral peptides through the binding of peptide-MHC adducts by T-cell receptors and co-stimulation by molecules of the CD28 family.
        • Full-Text
        • PDF
      See Editorial, pages 909–911

      Introduction

      Ten to twenty percent of individuals in Europe and North America are seropositive for HEV; however, only a minority of patients develop a symptomatic infection, and fulminant courses are mainly observed in the elderly and those with preexisting liver disease.
      EASL clinical practice guidelines on hepatitis E virus infection.
      ,
      • Horvatits T.
      • Schulze Zur Wiesch J.
      • Lutgehetmann M.
      • Lohse A.W.
      • Pischke S.
      The clinical perspective on hepatitis E.
      While HEV infections are self-limiting in immunocompetent hosts, infections with HEV genotype 3, the predominating HEV genotype in Europe and North America, persist in 50-70% of immunosuppressed patients.
      EASL clinical practice guidelines on hepatitis E virus infection.
      ,
      • Horvatits T.
      • Schulze Zur Wiesch J.
      • Lutgehetmann M.
      • Lohse A.W.
      • Pischke S.
      The clinical perspective on hepatitis E.
      Chronic infection is defined as HEV replication for >3 months, since no spontaneous HEV clearance was observed after this period. In patients with chronic HEV infection, reduction of immunosuppression as well as ribavirin treatment may result in viral clearance and are recommended by EASL guidelines.
      EASL clinical practice guidelines on hepatitis E virus infection.
      However, approximately 10% of patients do not respond to this treatment strategy, and there are currently no further treatment options available.
      EASL clinical practice guidelines on hepatitis E virus infection.
      In a substantial number of patients, chronic HEV infection progresses rapidly towards cirrhosis and liver failure. Thus, novel treatment options including immunotherapeutic strategies are urgently needed.
      Recent in vivo evidence from 2 animal models (chicken and rhesus macaque) demonstrates an important role of HEV-specific CD8+ T cells in HEV control.
      • Bremer W.
      • Blasczyk H.
      • Yin X.
      • Duron E.S.
      • Grakoui A.
      • Feng Z.
      • et al.
      Resolution of hepatitis E virus infection in CD8+ T cell depleted rhesus macaques.
      ,
      • Rogers E.
      • Todd S.M.
      • Pierson F.W.
      • Kenney S.P.
      • Heffron C.L.
      • Yugo D.M.
      • et al.
      CD8+ lymphocytes but not B lymphocytes are required for protection against chronic hepatitis E virus infection in chickens.
      However, in humans, there is still a paucity of data on HEV-specific CD8+ T-cell immunity.
      • Walker C.M.
      Adaptive immune responses in hepatitis A virus and hepatitis E virus infections.
      Studies using recombinant HEV capsid protein (corresponding to open reading frame [ORF] 2) as well as overlapping peptides covering some or all HEV domains (ORF1-3) showed broad and vigorous HEV-specific CD8+ T-cell responses in immunocompetent patients with acute infection that contracted upon resolution of infection.
      • Al-Ayoubi J.
      • Behrendt P.
      • Bremer B.
      • Suneetha P.V.
      • Gisa A.
      • Rinker F.
      • et al.
      Hepatitis E virus ORF 1 induces proliferative and functional T-cell responses in patients with ongoing and resolved hepatitis E.
      • Brown A.
      • Halliday J.S.
      • Swadling L.
      • Madden R.G.
      • Bendall R.
      • Hunter J.G.
      • et al.
      Characterization of the specificity, functionality, and durability of host T-cell responses against the full-length hepatitis E virus.
      • Gisa A.
      • Suneetha P.V.
      • Behrendt P.
      • Pischke S.
      • Bremer B.
      • Falk C.S.
      • et al.
      Cross-genotype-specific T-cell responses in acute hepatitis E virus (HEV) infection.
      • Suneetha P.V.
      • Pischke S.
      • Schlaphoff V.
      • Grabowski J.
      • Fytili P.
      • Gronert A.
      • et al.
      Hepatitis E virus (HEV)-specific T-cell responses are associated with control of HEV infection.
      • Wu T.
      • Zhang J.
      • Su Z.J.
      • Liu J.J.
      • Wu X.L.
      • Lin C.X.
      • et al.
      Specific cellular immune response in hepatitis E patients.
      However, immunosuppressed patients with chronic HEV infection displayed a narrow and weak HEV-specific CD8+ T-cell response mostly at or below the limit of detection. In vitro checkpoint inhibition targeting programmed cell death 1 (PD1) or cytotoxic T lymphocyte antigen-4 (CTLA-4), as well as successful antiviral treatment, partially restored HEV-specific CD8+ T-cell responses in some patients,
      • Suneetha P.V.
      • Pischke S.
      • Schlaphoff V.
      • Grabowski J.
      • Fytili P.
      • Gronert A.
      • et al.
      Hepatitis E virus (HEV)-specific T-cell responses are associated with control of HEV infection.
      ,
      • Abravanel F.
      • Barrague H.
      • Dorr G.
      • Saune K.
      • Peron J.M.
      • Alric L.
      • et al.
      Conventional and innate lymphocytes response at the acute phase of HEV infection in transplanted patients.
      however, checkpoint inhibition is not a suitable treatment strategy in transplanted patients after failed ribavirin treatment, since it may induce transplant rejection. It is thus necessary to more precisely define the mechanisms of HEV-specific CD8+ T-cell failure in persistent HEV infection. These studies have been limited, however, by the paucity of described HEV-specific CD8+ T-cell epitopes that can be analyzed, e.g. using peptide/HLA class I multimer technology. Indeed, to date, only 3 HLA-A∗02-restricted HEV-specific CD8+ T-cell epitopes have been defined and fine-mapped.
      • Soon C.F.
      • Behrendt P.
      • Todt D.
      • Manns M.P.
      • Wedemeyer H.
      • Sallberg Chen M.
      • et al.
      Defining virus-specific CD8+ TCR repertoires for therapeutic regeneration of T cells against chronic hepatitis E.
      Thus, we set out to define a broad range of HEV-specific CD8+ T-cell epitopes restricted by common HLA class I alleles in acute-resolving HEV infection; we then used these novel HEV-specific CD8+ T-cell epitopes to analyze the mechanisms of HEV-specific CD8+ T-cell failure in a cohort of immunocompromised patients with chronic HEV infection.

      Patients and methods

      Study cohort

      A total of 34 patients with self-limiting HEV infection and 12 patients with chronic HEV infection were included. All donors were recruited at the Freiburg University Medical Center, Germany. Acute HEV infection was determined by positive plasma HEV RNA, whereas resolved infection was defined by negative HEV RNA testing and the detection of anti-HEV IgG and/or IgM. Chronic HEV infection was defined as positive HEV RNA for >3 months according to EASL guidelines. Patient characteristics are displayed in Table 1. Next-generation sequencing was used for HLA-typing. Written informed consent was obtained in all cases and the study was conducted according to federal guidelines, local ethics committee regulations (Albert-Ludwigs-Universität, Freiburg, Germany; vote #: 437/18), and the Declaration of Helsinki (1975).
      Table 1Patient characteristics.
      Patient codeAge (years)
      At first diagnosis.
      SexHLA typeAnti-HEV IgG (AU/ml)
      cut-off 20 AU/ml.
      At first diagnosis.
      Anti-HEV IgM (AU/ml)
      cut-off 20 AU/ml.
      At first diagnosis.
      HEV RNA max (IU/ml)ALT max (U/L)Bilirubin max (mg/dl)Comment/comorbidityImmunosuppressionTherapy
      Patients with self-limiting infection
      aHEV-151MA∗02:01, A∗26:01, B∗08:01, B∗27:0555.614516,0312,6771.8Eosinophilic granulomatosis with polyangiitis
      aHEV-279MA∗01, A∗03, B∗08, B∗3593.715716,0313,8837.8First diagnosis of CUP (small-cell carcinoma)
      aHEV-351MA∗02, A∗29, B∗27, B∗5111714068,7021,2150.9NSCLC, infection during chemotherapy
      aHEV-469MA∗02:01, A∗11:01, B∗07:02, B∗14:02112155119,4675,2817
      aHEV-555FA∗03:01, A∗69:01, B∗18:01, B∗44:0225.22281,059,8622,2102.4Diabetes mellitus II
      aHEV-650MA:01:01, A∗02:01, B∗07:0111016833,1871,0571.1Gluten-sensitive enteropathy
      aHEV-757MA∗01:01, A∗24:02, B∗18:01, B∗35:08141254511,9761,99322.6Squamous cell carcinoma
      aHEV-820FA∗01:01, A∗03:01, B∗07:02, B∗55:01137217609,4951,6064
      aHEV-926MA∗01:01, A∗02:01, B∗08:01, B∗15:0183.51643541,6291.1
      aHEV-1071MA∗02:01, A∗30:01, B∗13:02, B∗15:0136.8140294,4222,64210.4Diabetes mellitus II
      aHEV-1154MA∗02:01, B∗40:01, B∗44:27109175889,0005,55921.7Acute on chronic liver failure (cryptogenic cirrhosis)
      aHEV-1280FA∗11:01, A∗68:02, B∗14:02, B∗35:015.801392,050,0009791Ovarial carcinoma
      aHEV-1364FA∗02:01, A∗03:01, B∗15:01, B∗35:081541573,1428250.8H/o AML, h/o allogeneic PBSCT 01/13
      aHEV-1421MA∗02, A∗33, B∗44, B∗51206137247,3143,7109.4HEV genotype 1 (India)
      aHEV-1536FA∗02:01, A∗11:01, B∗07:02, B∗40:0122.9152334,5509380.8Mamma carcinoma, infection during chemotherapy
      aHEV-1658FA∗01:01, A∗02:01, B∗08:01, B∗38:0126.5153598,0002,0394.8Allergic asthma
      aHEV-1757MA∗25:01, A∗31:01, B∗18:01, B∗27:05122164142,2231,5461.7
      aHEV-1849MA∗02:01, A∗68:01, B∗07:02, B∗40:0190.559.533,1873440.2HEV-associated neuralgic amyotrophy
      aHEV-1956MA∗02:01, A∗02:35, B∗07:02, B∗44:021401748738891.6Alcohol-related cirrhosis
      aHEV-2050MA∗02:01, B∗08:01, B∗45:011221921,8071,4103.4
      aHEV-2128FA∗02:01, A∗29:02, B∗07:02, B∗51:0115.893.5949,0001,2471.2Mild leukopenia with mildly reduced antibodies
      aHEV-2259FA∗02:01, A∗11:01, B∗18:01, B∗44:02165160ND3061.4H/o liver transplantation; documented seroconversion
      aHEV-2361MA∗02:01, B∗51:01, B∗51:XXNDND31,0001,0596.2First diagnosis of sarcoidosis
      aHEV-2465MA∗01:01, A∗02:01, B∗40:01, B∗51:01140143420,0001,24811Sarcoidosis
      rHEV-146FA∗02:01, B∗40:01, B∗44:0211653.9neg340.3Myocarditis
      rHEV-232FA∗02:01, A∗32:01, B∗40:02, B∗44:0215231.2neg440.6Adenoviral tonsilitis
      rHEV-352FA∗02:01, A∗24:02, B∗18:01, B∗41:0112424.4neg220.2Cervical disc prolapse
      rHEV-459MA∗02:01, A∗11:01, B∗44:02, B∗51:0177.99.2ND270.4Idiopathic hepatopathy
      rHEV-590FA∗02:01, A∗03:01, B∗35:01, B∗40:0125.96.7ND8101.9Drug-induced hepatopathy
      rHEV-652FA∗02:01, A∗03:01, B∗13:02, B∗15:0126946.4neg14117.8Drug-induced hepatopathy
      rHEV-760FA∗01, A∗11, B∗08, B∗3540.87.7ND160.2Chronic HBV infection
      rHEV-854FA∗02:01, A∗33:03, B∗35:01, B∗44:05829ND360.5NASH
      rHEV-952MA∗03:01, A∗31:01, B∗15:01, B∗27:051814.5ND911.1Autoimmune hepatitis
      rHEV-1062FA∗01:01, A∗24:02, B∗38:01, B∗57:01121162neg220.3H/o PBSCT for AML, h/o HBV reactivation
      Patients with chronic infections
      cHEV-124MA∗02:01, A∗03:01, B∗35:01, B∗39:01NDND511,9765343.8Liver transplantation (3x), kidney transplantationTAC/MMF/PREDRBV 11w
      cHEV-246FA∗01:01, A∗11:01, B∗08:01, B∗44:021.61.84,542,0237100.9Autologous PBSCT for B-non-Hodgkin lymphoma)RituximabRBV 16w
      cHEV-356MA∗02:01, A∗25:01, B∗15:01, B∗18:011581524,542,0233321.6Kidney transplantationTAC/MMF/PREDRBV 11w
      cHEV-463MA∗02, B∗08, B∗51NDND19,464,771810.5Kidney transplantationTAC/MMF/PREDRBV 11w
      cHEV-552FA∗01:01, A∗24:02, B∗08:01, B∗40:021521349,402,6291312.4Kidney transplantationTAC/MMF/PREDRBV 20w + 21w
      cHEV-658FA∗02:01, A∗24:02, B∗07:02, B∗40:01171282115,000450.6Kidney transplantationTAC/MMF/PREDRBV 20w
      cHEV-753MA∗02, B∗15:01, B∗441.92.4pos3400.7Lung transplantation, primary cerebral lymphoma, autologous PBSCTTAC/EVR/PREDRBV 47w
      cHEV-858MA∗02:01, A∗03:01, B∗07:02, B∗35:01NDND363,3631240.5Kidney transplantationTAC/MMF/PREDRBV 12w
      cHEV-974MA∗01:01, A∗32:01, B∗08:01, B∗15:01176868,290,0005081.5Sero-negative erosive polyarthritisAbataceptAbatacept paused
      cHEV-1065MA∗01:01, A∗03:01, B∗35:01, B∗57:01NDND6,127,5001810.6Autologous PBSCT (3x) for multiple myelomaCyA/MMF/ATGRBV 12w
      cHEV-1118MA∗02, B∗15:01negneg25,140,0006881.2Allogenic (HLA-identical) PBSCT for pre-T-ALLRituximabnone
      cHEV-1255FA∗11:01, A∗24:02, B∗15:02, B∗35:01NDND267,000840.5Kidney transplantationTAC/MMF/PREDnone
      aHEV, acute HEV; AML, acute myeloid leukemia; ATG, antithymocyte globulin; AU, arbitrary units; cHEV, chronic HEV; CUP, cancer of unknown primary; CyA, cyclosporin A; EVR, everolimus; MMF, mycophenolate mofetil; NASH, non-alcoholic steatohepatitis; ND, not done; NSCLC, non-small cell lung cancer; PBSCT, peripheral blood stem cell transplant; PRED, prednisone; RBV, ribavirin; rHEV, resolved HEV; TAC, tacrolimus; T-ALL, T-cell acute lymphoblastic leukemia.
      At first diagnosis.
      # cut-off 20 AU/ml.

      Statistics

      Statistical analysis was performed with GraphPad Prism 8 (USA). Statistical significance was assessed by Kruskal-Wallis testing including Dunn’s multiple comparisons test and Mann-Whitney testing. (∗p <0.05; ∗∗p <0.01; ∗∗∗p <0.001; ∗∗∗∗p <0.0001).
      Immunological and sequencing techniques are described in the supplementary methods section.

      Results

      Definition of virus-specific CD8+ T-cell epitopes in patients with self-limiting HEV infection

      To define HEV-specific CD8+ T-cell epitopes, we predicted 58 HEV-specific epitopes restricted by 10 common HLA class I alleles in silico (Table S1). We tested these epitope peptides using peripheral blood mononuclear cells (PBMCs) from 34 patients with self-limiting HEV infection expressing the respective HLA alleles (Table 1). Since results were similar for patients with acute-resolving (n = 24) and resolved HEV infection (n = 10), these patients were analyzed together. After antigen-specific expansion for 14 days, we could detect interferon-γ (IFNγ)-producing CD8+ T cells in response to peptide stimulation in 29/34 (86%) patients with self-limiting HEV infection (Fig. 1A; compare Fig. 1B for representative dot blots and Fig. S1A for negative controls). Single patients recognized up to 8 epitopes, while the majority of patients recognized 1-2 epitopes (Fig. 1A). Overall, 25/58 predicted epitopes were confirmed experimentally (Fig. 1C). 17/25 (68%) confirmed epitopes were located within ORF1, reflecting its overall length, while the remaining epitopes were located within ORF2 (Fig. 1D). Of note, the 78 positive responses were distributed quite equally to ORF1 and ORF2, indicating a relative immunodominance of the smaller ORF2. The confirmed epitopes were restricted by all HLA alleles included in the epitope prediction except for HLA-B∗07 (Fig. 1E). Strikingly, approx. 25% of confirmed epitopes and 50% of responses were restricted by HLA-A∗02:01, although the cohort was not pre-selected for HLA-A∗02:01 expression. IFNγ-producing HEV-specific CD8+ T cells could also be detected directly ex vivo by Elispot analysis in PBMCs from 3 donors with acute HEV infection (Fig. S1B,C).
      Figure thumbnail gr1
      Fig. 1Definition of virus-specific CD8+ T-cell epitopes in patients with self-limiting HEV infection.
      (A) Number of IFNγ responses in 34 individuals with self-limiting HEV infection after in vitro expansion. (B) Exemplary dot plots of positive IFNγ responses after re-stimulation with the indicated epitope peptides after 14 days in vitro culture. Percentages indicate IFNγ-positive CD8+ T cells of total CD8+ T cells. (C) Percentages of individuals with self-limiting HEV infection with positive IFNγ response targeting a given in silico predicted HLA-A- or HLA-B-restricted CD8+ T-cell epitope and strengths of individual responses. (D) Distribution of confirmed epitopes and positive responses targeting ORF1 or ORF2. (E) HLA restriction of confirmed epitopes and positive responses. ORF, open reading frame.

      Effector function and memory formation in self-limiting HEV infection

      To further characterize the HEV-specific CD8+ T-cell response in self-limiting HEV infection, we applied a peptide/HLA class I (pHLA-I) tetramer-based ex vivo enrichment strategy (gating strategy displayed in Fig. S2). HEV-specific CD8+ T cells targeting the 4 epitopes analyzed were readily detectable (Fig. 2A); their frequency did not vary by targeted epitope but differed between patients with acute vs. resolved HEV infection (Fig. 2B). During acute infection, the frequency of HEV-specific CD8+ T cells was similar compared to influenza infection and severe acute respiratory syndrome coronavirus type 2 (SARS-CoV-2) infection but significantly higher compared to acute HBV infection (Fig. 2C). In line with the formation of a potent effector CD8+ T-cell response, we detected high expression of the activation marker CD38, the proliferation marker Ki67, the effector molecule granzyme B (GranB) and the transcription factor Tbet (also known as TBX21) (Fig. 2D). Expression of the memory marker CD127 was still low on HEV-specific CD8+ T cells (Fig. 2D). In an HLA-A∗02:01+/B∗27:05+ individual, we longitudinally monitored the differentiation of HEV-specific CD8+ T cells from early acute infection to >4 years after viral clearance. Upon viral elimination, the A∗02/ORF2493-501 and B∗27/ORF11315-1323-specific CD8+ T-cell response contracted with frequencies declining by up to 2 logs within the first months, but then plateaued throughout follow-up (Fig. 2E+F; exemplary ex vivo tetramer stainings without enrichment can be found in Fig. S3). Multi-dimensional scaling analysis of the longitudinal B∗27/ORF11315-1323 -specific CD8+ T-cell response in this individual revealed that early time points cluster away from each other while late time points cluster together, which emphasizes the early dynamic differentiation patterns during acute infection followed by the establishment of a stable memory cell pool (Fig. 2G). Diffusion map embedding, incorporating longitudinal flow cytometry data from B∗27/ORF11315-1323-specific CD8+ T cells, revealed that CD8+ T cells from the earliest and latest time points localized at opposing ends of the trajectory (Fig. 2H). Upon resolution of acute infection, expression of CD38 and Ki67 was lost and a memory response formed as characterized by expression of CD127 and T-cell factor 1 (TCF1) (Fig. 2H+I). We also found a tendency towards lower expression of GranB and PD1 on HEV-specific CD8+ T cells of patients with resolved compared to acute infection (Fig. 2I). The memory cell pool in resolved HEV infection was largely composed of the CD45RA-/CCR7- effector memory subset with a minority of terminally differentiated effector memory cells re-expressing CD45RA and few CCR7+ central memory CD8+ T cells (Fig. 2J). Collectively, these data indicate that acute/self-limiting HEV infection is associated with a strong HEV-specific effector CD8+ response that evolves into a bona fide memory response after resolution.
      Figure thumbnail gr2
      Fig. 2Effector function and memory formation in self-limiting HEV.
      (A) Exemplary dot plots after ex vivo peptide/HLA class I tetramer enrichment using 10-20x 106 PBMCs in self-limiting HEV infection. (B) Calculated ex vivo frequency of epitope-specific CD8+ T cells in aHEV and rHEV infection. (C) Calculated ex vivo virus-specific CD8+ T cells in acute and resolved HEV, HBV, FLU, and SARS-CoV-2 infection. (D) Expression of CD38, Ki67, GranB, KLRG1, CD127, and Tbet in epitope-specific cells in acute HEV infection. (E) Longitudinal frequencies of B∗27/ORF11315-1323 -specific CD8+ T cells in an individual with acute-resolving HEV infection. (F) Exemplary dot plots of longitudinal B∗27/ORF11315-1323 -specific CD8+ T cells in an individual with acute-resolving HEV infection. (G) MDS analysis of longitudinally sampled B∗27/ORF11315-1323 -specific CD8+ T cells. (H) Diffusion map showing flow cytometry data of B∗27/ORF11315-1323 -specific CD8+ T cells in relation to dps. Protein expression levels of CD38, Ki67, CD127 and TCF1 are plotted on the diffusion map. (I) Exemplary histograms and marker expression patterns in HEV-specific CD8+ T cells in acute compared to resolved HEV infection. (J) Exemplary dot plot and overall expression of CD45RA and CCR7 in epitope-specific CD8+ T cells in resolved HEV infection. Bar charts show median with IQR. Kruskal-Wallis with rank-sum tests including Dunn’s multiple comparisons or Mann-Whitney testing were performed. ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001, ∗∗∗∗p <0.0001. aHEV, acute HEV; dps, days post symptom onset; FLU, influenza; MDS, multi-dimensional scaling; PBMCs, peripheral blood mononuclear cells; rHEV, resolved HEV.

      Functional CD8+ T-cell responses in chronic HEV infection are lost over time but can be partially reinvigorated by treatment-induced viral clearance

      We next tested the 58 epitope candidates in 12 patients with chronic HEV infection. These 12 patients included 8 patients with solid organ transplants, 3 patients with autologous or HLA-identical allogeneic peripheral blood stem cell transplants, and 1 patient treated with the CTLA-4 agonist abatacept for erosive polyarthritis (see Table 1 for detailed patient characteristics). After peptide-specific expansion for 14 days, we found IFNγ-producing HEV-specific CD8+ T cells in only 4/12 (33%) patients with chronic infection (Fig. 3A,B) compared to 29/34 (86%) in self-limiting infection (p = 0.0006). In these 4 patients, a total of 7 responses (1-2 per patient) targeted 6 epitopes. Of these 6 epitopes, we had previously identified 4 in self-limiting HEV infection (3 restricted by HLA-A∗02:01, 1 by HLA-A∗11:01). 2 novel HLA-B∗35:01 restricted epitopes only tested positive in 1 patient with chronic infection each. All epitopes were located in ORF1 except for A∗02/ORF2493-501 (Fig. 3C). Similar to self-limiting infection, the majority of responses (4/7) were restricted by HLA-A∗02:01 (Fig. 3D). Interestingly, when an IFNγ response was observed in response to peptide stimulation, it was of a comparable magnitude to those in self-limiting HEV infection (Fig. 3E). Elimination of chronic HEV infection was achieved in all patients upon lowering of immunosuppression and/or ribavirin therapy (Table 1). 11/12 patients were re-tested for IFNγ responses upon resolution of infection; 1 patient was lost to follow-up for immunological analysis. Notably, 4 patients without a pre-treatment IFNγ response developed HEV-specific responses upon resolution of infection (Fig. 3F). For example, patient cHEV-1, who was infected for >4 years, developed HEV-specific IFNγ responses against 4 epitopes upon ribavirin-associated viral clearance (Fig. 3G). When we arranged all chronically infected patients on a timeline according to the duration of infection, it became obvious that CD8+ T-cell responses were lost over time during persistent infection. However, the partial reinvigoration of these responses upon treatment-induced viral elimination seemed to be independent of the duration of infection (Fig. 3H). Of note, while the number of CD8+ T-cell responses in chronic infection was significantly diminished compared to acute infection (p = 0.0172), responses were readily detectable after treatment-associated resolution to a level observed in resolved infection (Fig. 3I; p >0.9999). Collectively, these data indicate that functional HEV-specific CD8+ T-cell responses are strongly diminished during the course of chronic HEV infection, but can be at least partially reinvigorated by treatment-induced viral elimination. In line with these observations, frequencies of HEV-specific CD8+ T cells were too low to be detected by ex vivo IFNγ Elispot analysis in 4 chronically infected patients at RNA+ and RNA- time points (data not shown), with the exception of donor cHEV-5, in whom ex vivo IFNγ responses against 2 HLA-A∗01:01 restricted epitopes could be detected only upon resolution of infection (Fig. S1D). Peak alanine aminotransferase (ALT) and viral load differed significantly between acute and chronic HEV infection (Fig. S4A), but did not correlate with each other (Fig. S4B). Of note, no correlation between peak ALT or viral load and the number of targeted CD8+ T-cell epitopes was found in acute infection (Fig. S4C/D, left panels). In chronic infection, however, peak ALT and viral load correlated negatively with the number of HEV-specific CD8+ T-cell responses (Fig. S4C/D, right panels).
      Figure thumbnail gr3
      Fig. 3Functional CD8+ T-cell responses in chronic HEV infection are lost over time but can be partially reinvigorated by treatment-induced viral clearance.
      (A) Number of IFNγ responses in 12 individuals with chronic HEV infection after in vitro expansion. (B) Exemplary dot plots of positive IFNγ responses after re-stimulation with indicated epitope peptides after 14 days in vitro culture. Percentages indicate IFNγ positive CD8+ T cells as a proportion of total CD8+ T cells. (C) Distribution of confirmed epitopes and positive responses targeting ORF1 or ORF2. (D) HLA restriction of confirmed epitopes and positive responses. (E) Magnitude of individual IFNγ responses in cHEV compared to aHEV infection. (F) Number of IFNγ responses in 11 individuals with cured chronic HEV infection after in vitro expansion. (G) Timeline depicting ribavirin treatment and sampling in patient P1. Dot plots show IFNγ responses to epitope peptides after 14 d in vitro expansion at the specified time points. Percentages indicate IFNγ positive CD8+ T cells of total CD8+ T cells. (H) IFNγ responses (HEV RNA+ and cured HEV RNA- time points) in chronic HEV patients plotted on a time line referring to the total duration of chronic infection f. Blue boxes indicate at least 1 positive response, white boxes indicate no response. (I) Number of positive IFNγ responses in aHEV, rHEV, cHEV, and tcHEV infection per donor with mean. Kruskal-Wallis with rank-sum test including Dunn’s multiple comparisons was performed. aHEV, acute HEV; cHEV, chronic HEV; ORF, open reading frame; rHEV, resolved HEV; tcHEV, treated chronic HEV.

      HEV-specific CD8+ T cells in chronic HEV infection are rare but functional and exhibit a memory-like phenotype after viral resolution

      To characterize the CD8+ T-cell response in chronic HEV infection in more detail, we used PBMCs from 9 HLA-A∗01:01+ (n = 4) or HLA-A∗02:01+ (n = 5) patients during chronic infection and after treatment-induced resolution to enrich either A∗01/ORF2389-397 or A∗02/ORF2493-501–specific CD8+ T cells ex vivo (Fig. 4A, patients with positive responses at one or both time points are displayed; sampling time points and course of infection are depicted in Fig. S5). 2/4 HLA-A∗01:01+ and 2/5 HLA-A∗02:01+ patients had detectable epitope-specific CD8+ T-cell responses during chronic infection. After resolution, an additional HLA-A∗01:01+ and an HLA-A∗02:01+ patient each developed positive responses (Fig. 4B). The ex vivo frequency of epitope-specific CD8+ T cells was not significantly different between chronic, resolved, and treated chronic HEV infection but was significantly lower compared to acute HEV infection (Fig. 4C). Using longitudinal samples of the 4 patients displaying responses during chronic and resolved infection (cHEV-2, cHEV-3, cHEV-9, cHEV-11), as well as late samples from the 2 patients who developed responses after resolution (cHEV-5, cHEV-6), we analyzed the evolution of the HEV-specific CD8+ T-cell phenotype before and after viral clearance (Fig. 4D,E). During chronic infection, epitope-specific CD8+ T cells expressed the activation marker CD38 and mostly displayed Ki67 as a marker of proliferation (Fig. 4E). However, they also expressed high levels of PD1 and low levels of CD127 and TCF1, with the majority of cells being in the CD127-PD1+ subset, indicating some characteristics of terminally exhausted CD8+ T cells (Fig. 4D,E). After resolution of chronic infection, CD8+ T cells displayed a memory-like phenotype (PD1 moderate, CD127+, TCF1+) with the majority of cells being CD127+PD1+ (Fig. 4D). They also lost expression of CD38 (Fig. 4E). Collectively, these data indicate that HEV-specific CD8+ T cells develop a terminally exhausted phenotype during chronic infection with albeit (partially) maintained proliferative capacity. They also indicate that a sufficient pool of memory-like CD8+ T cells remains to be restored after viral elimination.
      Figure thumbnail gr4
      Fig. 4HEV-specific CD8+ T cells in chronic HEV infection are rare but functional and obtain memory-like phenotype after viral resolution.
      (A) Dot plots after ex vivo tetramer enrichment using 10-20x 106 PBMCs in 6 patients with cHEV and tcHEV infection. (B) Percentage of tested patients with positive response in cHEV and tcHEV infection. (C) Frequencies of epitope-specific CD8+ T cells in aHEV, rHEV, cHEV and tcHEV infection. (D) Percentage of CD127-PD1+ and CD127+PD1+ virus-specific CD8+ T cells in 6 patients with cHEV and tcHEV infection. (E) Dot plots depicting protein expression levels on CD8+ bulk (grey) and epitope-specific CD8+ T cells (black) and expression of these markers in 6 patients with cHEV and tcHEV infection. Bar charts show median with IQR. Kruskal-Wallis with rank-sum tests including Dunn’s multiple comparisons was performed. ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001, ∗∗∗∗p <0.0001. aHEV, acute HEV; cHEV, chronic HEV; PBMCs, peripheral blood mononuclear cells; rHEV, resolved HEV; tcHEV, treated chronic HEV.

      Impact of immunosuppression on the HEV-specific CD8+ T-cell response and impact of HEV-specific CD8+ T-cell response on viral control

      To address the impact of immunosuppression on the induction of HEV-specific CD8+ T cells, T-cell functionality and viral titers, we longitudinally studied patient cHEV-9 (details in Table 1) from 7 days post-diagnosis of HEV infection for >2 years. At diagnosis, cHEV-9 presented with positive HEV IgG and IgM and high viral titers. We were able to characterize the A∗01/ORF2389-397 and A∗01/ORF2298-306-specific CD8+ T-cell response longitudinally (Fig. 5A,B). During abatacept treatment, the frequency of HEV-specific CD8+ T cells was low, ranging from 10-5-10-4. However, T-cell numbers were reinvigorated upon discontinuation of abatacept treatment. Indeed, their substantial increase in frequency directly preceded a sharp decline of viremia that resulted in viral clearance and normalization of liver parameters (Fig. 5C). Shortly after viral clearance, HEV-specific CD8+ T cells declined below the limit of detection in peripheral blood. Interestingly, on d771 post-diagnosis (d645 post-resolution), the HEV-specific CD8+ T-cell response had recovered and persisted at frequencies of approximately 10-5 (Fig. 5C). During abatacept treatment, HEV-specific CD8+ T cells recognized their cognate antigen (demonstrated by high expression of the activation marker CD38 [Fig. 5D]); however, they lacked GranB expression (a marker for effector function [Fig. 5E]). Upon discontinuation of abatacept, GranB expression was rescued and correlated with declining viral titers, further supporting a direct impact of immunosuppression on HEV-specific CD8+ T-cell immunity.
      Figure thumbnail gr5
      Fig. 5Impact of immunosuppression on the HEV-specific CD8+ T-cell response and impact of HEV-specific CD8+ T-cell response on viral control.
      (A) Dot plots show A∗01/ORF2389-397-specific CD8+ T cells after ex vivo tetramer enrichment in cHEV-9 at longitudinal time points. (B) Timeline depicting sampling dates (black arrows HEV RNA+; grey arrows HEV RNA-) at dpd, abatacept pause and duration of positive HEV RNA testing (grey bar) in cHEV-9. (C) Calculated ex vivo frequencies of epitope-specific CD8+ T cells at indicated dpd, viral load (stars and grey dotted line) and ALT (diamond and dark blue dotted line) (D-E) Expression of CD38 and GranB of epitope-specific CD8+ T cells in patient cHEV-9 at indicated longitudinal time points. (F) Timeline depicting sampling dates (black arrows HEV RNA+; red arrows HEV RNA-) at dpd, ribavirin treatment and duration of positive HEV RNA testing (grey bar) in cHEV-5. (G) Calculated ex vivo frequencies of epitope-specific CD8+ T cells at indicated dpi and viral load (grey line) in cHEV-5. cHEV, chronic HEV; dpd, days post diagnosis; dpi, days post infection.
      We were also able to longitudinally follow a kidney-transplanted patient on triple immunosuppression (cHEV-5) who relapsed after ribavirin therapy and required re-treatment with ribavirin (Fig. 5F). Importantly, this patient only displayed HEV-specific CD8+ T cells ex vivo (Fig. 5G) after sustained virological response following the second treatment course, further indicating a role of HEV-specific CD8+ T cells in sustained viral control. Collectively, the data from these 2 patients underline the dominant role of immunosuppression in impairing HEV-specific CD8+ T-cell responses, and also the important contribution of HEV-specific CD8+ T cells for sustained viral clearance.

      Viral escape contributes to HEV-specific CD8+ T-cell failure

      Finally, we addressed the role of mutational viral escape in chronic HEV infection. We thus sequenced the HEV genome via ultra-deep sequencing at early and late time points (interval of 2 months to 4 years) during chronic infection in 5 patients (Fig. 6A, Table S2). In cHEV-2, we found a mutation in epitope A∗01/ORF2389-397 that was targeted in this patient. The amino acid substitution was predicted to substantially reduce HLA-A∗01:01 binding (Fig. 6B). After expansion of PBMCs from cHEV-2 with the consensus or the variant peptide for 14 days, tetramer staining revealed that PBMCs only expanded in response to stimulation with the consensus peptide (Fig. 6C). When we re-stimulated the CD8+ T cells that had been expanded with the consensus peptide with either consensus or variant peptide, we observed downregulation and internalization of the T-cell receptor (TCR) as well as an increase of GranB expression as a correlate for T-cell activation upon re-stimulation with the consensus but not the variant peptide (Fig. 6D). Next, we compared the phenotype of ex vivo-enriched cells from cHEV-2 (variant epitope sequence) with those from cHEV-3 (consensus epitope sequence) (Fig. 6E). Cells targeting the conserved epitope in cHEV-3 displayed a PD1high phenotype with expression of CD38 as a correlate for ongoing antigen recognition, whereas cells targeting the variant epitope in cHEV-2 expressed reduced levels of PD1 and had lost CD38 expression (Fig. 6E). Finally, we performed high dimensional analysis of longitudinally enriched samples from cHEV-2 and cHEV-3 including markers of activation (Ki67, CD38, KLRG1, GranB), inhibitory receptors (PD1), memory (CD127, TCF1), and differentiation (CCR7, CD45RA, Tbet, Eomes). Multi-dimensional scaling analysis revealed that cells from cHEV-2 and cHEV-3 clustered away from each other, demonstrating their distinct differentiation. While all time points from cHEV-2 clustered together, the chronic infection time point in cHEV-3 was clearly separated from the 2 time points after resolution of infection, indicating that antigen withdrawal has an impact on the phenotype of T cells targeting conserved epitopes, but not escaped epitopes (Fig. 6F). Collectively, these data indicate that viral escape may contribute to CD8+ T-cell failure in chronic HEV infection.
      Figure thumbnail gr6
      Fig. 6Viral escape contributes to HEV-specific CD8+ T-cell failure.
      (A) Percentage of individuals with viral escape. (B) Longitudinal UDS data identifying viral escape variant L395V in epitope A∗01/ORF2389-397 in cHEV-2. (C) Tetramer staining after 14 days in vitro expansion with consensus or variant A∗01/ORF2389-397 epitope. (D) Tetramer staining after 14 days in vitro expansion with consensus A∗01/ORF2389-397 epitope and re-stimulation with consensus or variant peptide and histograms showing GranB expression of epitope-specific CD8+ T cells (black line) and CD8+ bulk T cells (grey). (E) Dot plots (grey: CD8+ bulk; black: antigen-specific) and histograms showing PD1 and CD38 expression of epitope-specific CD8+ T cells targeting a conserved (cHEV-3) or escaped (cHEV-2) epitope after ex vivo tetramer enrichment. (F) MDS analysis comparing longitudinal flow data from conserved and escaped epitopes. cHEV, chronic HEV; MDS, multi-dimensional scaling; ORF, open reading frame; UDS, ultra-deep sequencing.

      Discussion

      Herein, we set out to define more precisely the mechanisms of HEV-specific CD8+ T-cell failure in persistent HEV infection in immunocompromised patients. Due to the paucity of described and fine-mapped HEV-specific CD8+ T-cell epitopes,
      • Soon C.F.
      • Behrendt P.
      • Todt D.
      • Manns M.P.
      • Wedemeyer H.
      • Sallberg Chen M.
      • et al.
      Defining virus-specific CD8+ TCR repertoires for therapeutic regeneration of T cells against chronic hepatitis E.
      we first had to define fine-mapped HEV-specific CD8+ T-cell epitopes. Using a combined in silico and in vitro approach, we defined 25 novel HEV-specific CD8+ T-cell epitopes restricted by 9 different HLA alleles. Approximately two-thirds of identified CD8+ T-cell epitopes were located in ORF1, reflecting both the relative length of the protein as well as the relative number of epitopes predicted in comparison to ORF2 and ORF3 (Fig. 1D). However, in agreement with previous reports,
      • Brown A.
      • Halliday J.S.
      • Swadling L.
      • Madden R.G.
      • Bendall R.
      • Hunter J.G.
      • et al.
      Characterization of the specificity, functionality, and durability of host T-cell responses against the full-length hepatitis E virus.
      approximately 50% of HEV-specific CD8+ T-cell responses targeted ORF2, indicating that ORF2 is a relatively immunodominant CD8+ T-cell target. Nearly one-quarter of the identified epitopes were restricted by 2 HLA types each, HLA-A∗02 and HLA-B∗27 (Fig. 1E). This overrepresentation is partially based on our selection of epitope candidates: Based on our previous experiences, we selected 10 epitope candidates for HLA-A∗02 and 8 epitope candidates for HLA-B∗27, compared to 5 epitope candidates for the remaining HLA alleles. Even more prominently, however, approximately 50% of positive CD8+ T-cell responses were restricted by HLA-A∗02. This dominant targeting of HLA-A∗02-restricted epitopes may also be explained by the high proportion of HLA-A∗02+ patients in our cohort (25/34 patients with self-limiting infection). Importantly, our cohort was not pre-selected to contain a high number of HLA-A∗02+ patients. Thus, HLA-A∗02 may be associated with a symptomatic course of acute HEV infection; however, additional studies are required to address this interesting issue.
      Importantly, using our selection of predicted epitopes, the large majority (29/34) of patients with self-limited HEV infection displayed HEV-specific CD8+ T-cell responses, confirming previous reports on the immunogenicity of HEV.
      • Walker C.M.
      Adaptive immune responses in hepatitis A virus and hepatitis E virus infections.
      • Al-Ayoubi J.
      • Behrendt P.
      • Bremer B.
      • Suneetha P.V.
      • Gisa A.
      • Rinker F.
      • et al.
      Hepatitis E virus ORF 1 induces proliferative and functional T-cell responses in patients with ongoing and resolved hepatitis E.
      • Brown A.
      • Halliday J.S.
      • Swadling L.
      • Madden R.G.
      • Bendall R.
      • Hunter J.G.
      • et al.
      Characterization of the specificity, functionality, and durability of host T-cell responses against the full-length hepatitis E virus.
      • Gisa A.
      • Suneetha P.V.
      • Behrendt P.
      • Pischke S.
      • Bremer B.
      • Falk C.S.
      • et al.
      Cross-genotype-specific T-cell responses in acute hepatitis E virus (HEV) infection.
      • Suneetha P.V.
      • Pischke S.
      • Schlaphoff V.
      • Grabowski J.
      • Fytili P.
      • Gronert A.
      • et al.
      Hepatitis E virus (HEV)-specific T-cell responses are associated with control of HEV infection.
      • Wu T.
      • Zhang J.
      • Su Z.J.
      • Liu J.J.
      • Wu X.L.
      • Lin C.X.
      • et al.
      Specific cellular immune response in hepatitis E patients.
      Four of the 5 patients with self-limited HEV infection who did not display a response, at least to the epitopes analyzed here, had significant cancer or infectious comorbidities that may have impacted the virus-specific immune response. The only patient who lacked both a detectable T-cell response and substantial comorbidity suffered from neuralgic amyotrophy and Guillain Barré syndrome. This intriguing finding, although limited to just 1 patient, may indicate that HEV-associated Guillain Barré syndrome is mediated by an immune mechanism independent of HEV-specific CD8+ T cells. Importantly, we expanded on previous studies regarding the HEV-specific CD8+ T-cell response in acute infection by making 2 important observations on the single-epitope level: First, in the context of acute infection, HEV-specific CD8+ T cells display a high ex vivo frequency, substantially and significantly outnumbering virus-specific CD8+ T cells in acute HBV infection, and numerically even outnumbering influenza- and SARS-CoV-2-specific CD8+ T cells in acute infection (Fig. 2C). Impressively, epitope-specific CD8+ T cells comprised up to one-third of the total circulating CD8+ T-cell count in some patients. These data may also partially explain why acute HEV infection can have a fulminant course in vulnerable patients, such as elderly individuals and patients with preexisting liver disease.
      • El Costa H.
      • Gouilly J.
      • Abravanel F.
      • Bahraoui E.
      • Peron J.M.
      • Kamar N.
      • et al.
      Effector memory CD8 T cell response elicits hepatitis E virus genotype 3 pathogenesis in the elderly.
      The epitopes identified in our study may help to further address this important issue in future studies. Second, by longitudinal phenotypical analysis, we demonstrate that during acute infection, HEV-specific CD8+ T cells display a highly activated effector memory phenotype. After viral clearance, the HEV-specific CD8+ T-cell pool contracts as expected, and a typical memory phenotype develops, indicating that HEV-specific CD8+ T-cell immunity in acute-resolving infection resembles influenza-specific immunity.
      Next, we analyzed 12 immunocompromised patients with chronic HEV infection using the same set of epitope peptides. In agreement with previous reports,
      • Al-Ayoubi J.
      • Behrendt P.
      • Bremer B.
      • Suneetha P.V.
      • Gisa A.
      • Rinker F.
      • et al.
      Hepatitis E virus ORF 1 induces proliferative and functional T-cell responses in patients with ongoing and resolved hepatitis E.
      ,
      • Brown A.
      • Halliday J.S.
      • Swadling L.
      • Madden R.G.
      • Bendall R.
      • Hunter J.G.
      • et al.
      Characterization of the specificity, functionality, and durability of host T-cell responses against the full-length hepatitis E virus.
      ,
      • Suneetha P.V.
      • Pischke S.
      • Schlaphoff V.
      • Grabowski J.
      • Fytili P.
      • Gronert A.
      • et al.
      Hepatitis E virus (HEV)-specific T-cell responses are associated with control of HEV infection.
      only a minority of patients (4/12) displayed an HEV-specific CD8+ T-cell response that was narrowly focused, targeting only 1-2 epitopes (Fig. 3A,I). Using tetramer-based enrichment and thus a very sensitive method, we were able to detect HEV-specific CD8+ T cells in 4/9 patients with chronic HEV infection ex vivo (Fig. 4A,B). Of note, these HEV-specific CD8+ T cells displayed an activated and predominantly CD127-PD1+ phenotype similar to terminally exhausted CD8+ T cells in chronic HBV and HCV infection.
      • Wieland D.
      • Kemming J.
      • Schuch A.
      • Emmerich F.
      • Knolle P.
      • Neumann-Haefelin C.
      • et al.
      TCF1+ hepatitis C virus-specific CD8+ T cells are maintained after cessation of chronic antigen stimulation.
      ,
      • Schuch A.
      • Salimi Alizei E.
      • Heim K.
      • Wieland D.
      • Kiraithe M.M.
      • Kemming J.
      • et al.
      Phenotypic and functional differences of HBV core-specific versus HBV polymerase-specific CD8+ T cells in chronically HBV-infected patients with low viral load.
      Although this phenotype has been described in various chronic disease and cancer settings before, our study is the first to indicate that chronic antigen stimulation could also progress towards T-cell exhaustion under ongoing immunosuppression. Interestingly, exhausted HEV-specific CD8+ T cells were sensitive to immunosuppression and could only be reinvigorated in a substantial number of patients upon reduction of immunosuppression and/or application of ribavirin (Fig. 3G,H), leading to the maintenance of memory-like HEV-specific CD8+ T cells after viral clearance (Fig. 4D). Importantly, the effector function of exhausted CD8+ T cells was reduced but not abolished, as indicated by ongoing expression of GranB and IFNγ (Fig. 3, Fig. 4E), and they presumably still exert some degree of viral control. HEV-specific CD8+ T-cell responses declined with increasing duration of persistent HEV infection (Fig. 3H), indicating that long-term chronic HEV infection may lead to a reduced chance of partially reinvigorating functional CD8+ T-cell responses. Thus, prolonged antigenic stimulation may lead to long-lasting epigenetic scars on virus-specific CD8+ T cells, as observed in chronic HCV infection.
      • Hensel N.
      • Gu Z.
      • Sagar
      • Wieland D.
      • Jechow K.
      • Kemming J.
      • et al.
      Memory-like HCV-specific CD8+ T cells retain a molecular scar after cure of chronic HCV infection.
      • Tonnerre P.
      • Wolski D.
      • Subudhi S.
      • Aljabban J.
      • Hoogeveen R.C.
      • Damasio M.
      • et al.
      Differentiation of exhausted CD8+ T cells after termination of chronic antigen stimulation stops short of achieving functional T cell memory.
      • Yates K.B.
      • Tonnerre P.
      • Martin G.E.
      • Gerdemann U.
      • Al Abosy R.
      • Comstock D.E.
      • et al.
      Epigenetic scars of CD8+ T cell exhaustion persist after cure of chronic infection in humans.
      Thus, “standard” treatment of HEV infection (reduction of immunosuppression, ribavirin) as well as future immune-mediated treatment strategies might be the most promising in relatively recent persisting HEV infection, underlining the importance of early diagnosis and treatment.
      In a patient who developed chronic HEV infection during abatacept treatment, we could narrowly follow the kinetics of HEV-specific CD8+ T cells and viral clearance upon pausing abatacept. Importantly, this patient displayed a strong increase of HEV-specific CD8+ T cells that preceded a sharp decline in viremia, indicating a direct involvement of HEV-specific CD8+ T cells in viral clearance. This important role of HEV-specific CD8+ T cells was further supported by the longitudinal course of a patient who relapsed after ribavirin but only succeeded in viral clearance with a second course of ribavirin treatment that was associated with the occurrence of HEV-specific CD8+ T cells (Fig. 5F,G). These observations in the human setting are in agreement with recent data from the chicken as well as rhesus macaque HEV infection models that indicated an important role of HEV-specific CD8+ T cells in viral clearance.
      • Bremer W.
      • Blasczyk H.
      • Yin X.
      • Duron E.S.
      • Grakoui A.
      • Feng Z.
      • et al.
      Resolution of hepatitis E virus infection in CD8+ T cell depleted rhesus macaques.
      ,
      • Rogers E.
      • Todd S.M.
      • Pierson F.W.
      • Kenney S.P.
      • Heffron C.L.
      • Yugo D.M.
      • et al.
      CD8+ lymphocytes but not B lymphocytes are required for protection against chronic hepatitis E virus infection in chickens.
      The number of HEV-specific T-cell responses correlated negatively with viral load as well as ALT levels in chronic HEV infection (Fig. S4C,D). This interesting finding could either imply that a high HEV-specific T-cell reactivity contributes to viral control, or that high viral loads contribute to HEV-specific T-cell exhaustion. In addition, they suggest that ALT levels rather indicate non-specific (bystander) inflammation in the absence of HEV-specific immunity and are not a consequence of HEV-specific immunity. Importantly, all our data analyzed the HEV-specific CD8+ T-cell response in peripheral blood. Whether these analyses reflect the CD8+ T-cell response homing to the liver remains to be elucidated by future studies.
      Next to T-cell exhaustion, we also found evidence for mutational viral escape in chronic HEV infection, however, this observation was limited to a single epitope in 1 individual (Fig. 6), while the other 4 analyzed patients with persistent infection did not display mutations in the epitopes targeted in the respective patients or HLA-matched epitopes identified in other patients in our study.
      In sum, our study led to the identification and fine-mapping of a large set of HEV-specific CD8+ T-cell epitopes that allowed us to study the mechanisms of CD8+ T-cell failure in persistent HEV infection on a single-epitope level. These epitopes will also allow for the study of mechanisms of immunopathology and extrahepatic manifestations in fulminant HEV infection,
      • Pischke S.
      • Hartl J.
      • Pas S.D.
      • Lohse A.W.
      • Jacobs B.C.
      • Van der Eijk A.A.
      Hepatitis E virus: infection beyond the liver?.
      and they may serve as targets for therapeutic TCR-redirection of T cells against chronic HEV infection.
      • Soon C.F.
      • Behrendt P.
      • Todt D.
      • Manns M.P.
      • Wedemeyer H.
      • Sallberg Chen M.
      • et al.
      Defining virus-specific CD8+ TCR repertoires for therapeutic regeneration of T cells against chronic hepatitis E.
      While our understanding of the immunopathology of chronic HEV infection under immunosuppression is still incomplete, the main concept so far implied that immunosuppression prevents T-cell priming. However, our data suggests that an alternative concept may also apply: T-cell responses are primed even in the context of immunosuppression, however, they are dysfunctional, become exhausted and are eventually depleted during ongoing antigen recognition. Importantly, cessation of immunosuppressive medication could restore T-cell responses, while viral escape can contribute to T-cell failure. Our results also indicate that treatments aiming at restoring autologous HEV-specific CD8+ T-cell immunity, e.g. by reducing immunosuppression in combination with ribavirin, should be started early in persistent HEV infection. In later stages of chronic HEV infection, more personalized approaches such as the generation and application of TCR-redirected CD8+ T cells might become necessary for viral clearance in patients who fail standard treatment
      • Soon C.F.
      • Behrendt P.
      • Todt D.
      • Manns M.P.
      • Wedemeyer H.
      • Sallberg Chen M.
      • et al.
      Defining virus-specific CD8+ TCR repertoires for therapeutic regeneration of T cells against chronic hepatitis E.
      ; however, mutational viral escape should be excluded before this work- and cost-intensive procedure.

      Abbreviations

      ALT, alanine aminotransferase; CTLA-4, cytotoxic T lymphocyte antigen-4; GranB, granzyme B; IFN, interferon; ORF, open reading frame; PBMCs, peripheral blood mononuclear cells; PD1, programmed cell death 1; SARS-CoV-2, severe acute respiratory syndrome coronavirus type 2; TCF1, T-cell factor 1; TCR, T-cell receptor.

      Financial support

      The study was partially funded by German Research Foundation (DFG; SFB1160, project number 256073931, to R.T., M.H., C.N.H., Y.T., and K.W.; TRR179, project number 272983813, to T.B.; SFB1328, project number 335447717, to J.S.z.W.; SFB841, project number 80750187, to M.L. and J.S.z.W.) and the German Center for Infection Research (DZIF; TTU Hepatitis, to M.L., S.P., J.S.z.W., and N.F.). M.H. was supported by a Margarete von Wrangell fellowship (State of Baden-Wuerttemberg). D.A.P. was supported by a Welcome Trust Senior Investigator Award (100326/Z/12/Z). The funding bodies had no role in the decision to write or submit the manuscript.

      Authors’ contributions

      J.K. planned, performed, and analyzed experiments. S.G. performed HEV UDS sequencing experiments and interpreted sequencing data with the help of N.F. J.H. performed the bioinformatic analysis of the sequencing experiments. S.L.-L. and D.A.P. provided tetramers. F.E. performed four-digit HLA-typing by next-generation sequencing. M.P., D.H., M.L., S.P., J.S.z.W., Y.T., K.W., and T.B. recruited patients and collected clinical data. M.H. and C.N.H. designed the study and contributed to experimental design and planning. J.K., R.T., N.F., M.H., and C.N.H. interpreted data and wrote the manuscript.

      Data availability statement

      No data is deposited on public data bases. All requests for raw and analyzed data and materials will be reviewed by the corresponding author to verify if the request is subject to any confidentiality obligations. Patient-related data not included in the paper were generated as part of clinical examination and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material transfer agreement.

      Conflict of interest

      The authors declare no conflicts of interest that pertain to this work.
      Please refer to the accompanying ICMJE disclosure forms for further details.

      Acknowledgments

      We thank all patients and healthy donors for participating in the current study.

      Supplementary data

      The following are the supplementary data to this article:

      References

      1. EASL clinical practice guidelines on hepatitis E virus infection.
        J Hepatol. 2018; 68: 1256-1271
        • Horvatits T.
        • Schulze Zur Wiesch J.
        • Lutgehetmann M.
        • Lohse A.W.
        • Pischke S.
        The clinical perspective on hepatitis E.
        Viruses. 2019; 11
        • Bremer W.
        • Blasczyk H.
        • Yin X.
        • Duron E.S.
        • Grakoui A.
        • Feng Z.
        • et al.
        Resolution of hepatitis E virus infection in CD8+ T cell depleted rhesus macaques.
        J Hepatol. 2021; 75: 557-564
        • Rogers E.
        • Todd S.M.
        • Pierson F.W.
        • Kenney S.P.
        • Heffron C.L.
        • Yugo D.M.
        • et al.
        CD8+ lymphocytes but not B lymphocytes are required for protection against chronic hepatitis E virus infection in chickens.
        J Med Virol. 2019; 91: 1960-1969
        • Walker C.M.
        Adaptive immune responses in hepatitis A virus and hepatitis E virus infections.
        Cold Spring Harb Perspect Med. 2018;
        • Al-Ayoubi J.
        • Behrendt P.
        • Bremer B.
        • Suneetha P.V.
        • Gisa A.
        • Rinker F.
        • et al.
        Hepatitis E virus ORF 1 induces proliferative and functional T-cell responses in patients with ongoing and resolved hepatitis E.
        Liver Int. 2018; 38: 266-277
        • Brown A.
        • Halliday J.S.
        • Swadling L.
        • Madden R.G.
        • Bendall R.
        • Hunter J.G.
        • et al.
        Characterization of the specificity, functionality, and durability of host T-cell responses against the full-length hepatitis E virus.
        Hepatology. 2016; 64: 1934-1950
        • Gisa A.
        • Suneetha P.V.
        • Behrendt P.
        • Pischke S.
        • Bremer B.
        • Falk C.S.
        • et al.
        Cross-genotype-specific T-cell responses in acute hepatitis E virus (HEV) infection.
        J Viral Hepat. 2016; 23: 305-315
        • Suneetha P.V.
        • Pischke S.
        • Schlaphoff V.
        • Grabowski J.
        • Fytili P.
        • Gronert A.
        • et al.
        Hepatitis E virus (HEV)-specific T-cell responses are associated with control of HEV infection.
        Hepatology. 2012; 55: 695-708
        • Wu T.
        • Zhang J.
        • Su Z.J.
        • Liu J.J.
        • Wu X.L.
        • Lin C.X.
        • et al.
        Specific cellular immune response in hepatitis E patients.
        Intervirology. 2008; 51: 322-327
        • Abravanel F.
        • Barrague H.
        • Dorr G.
        • Saune K.
        • Peron J.M.
        • Alric L.
        • et al.
        Conventional and innate lymphocytes response at the acute phase of HEV infection in transplanted patients.
        J Infect. 2016; 72: 723-730
        • Soon C.F.
        • Behrendt P.
        • Todt D.
        • Manns M.P.
        • Wedemeyer H.
        • Sallberg Chen M.
        • et al.
        Defining virus-specific CD8+ TCR repertoires for therapeutic regeneration of T cells against chronic hepatitis E.
        J Hepatol. 2019; 71: 673-684
        • El Costa H.
        • Gouilly J.
        • Abravanel F.
        • Bahraoui E.
        • Peron J.M.
        • Kamar N.
        • et al.
        Effector memory CD8 T cell response elicits hepatitis E virus genotype 3 pathogenesis in the elderly.
        PLoS Pathog. 2021; 17e1009367
        • Wieland D.
        • Kemming J.
        • Schuch A.
        • Emmerich F.
        • Knolle P.
        • Neumann-Haefelin C.
        • et al.
        TCF1+ hepatitis C virus-specific CD8+ T cells are maintained after cessation of chronic antigen stimulation.
        Nat Commun. 2017; 815050
        • Schuch A.
        • Salimi Alizei E.
        • Heim K.
        • Wieland D.
        • Kiraithe M.M.
        • Kemming J.
        • et al.
        Phenotypic and functional differences of HBV core-specific versus HBV polymerase-specific CD8+ T cells in chronically HBV-infected patients with low viral load.
        Gut. 2019; 68: 905-915
        • Hensel N.
        • Gu Z.
        • Sagar
        • Wieland D.
        • Jechow K.
        • Kemming J.
        • et al.
        Memory-like HCV-specific CD8+ T cells retain a molecular scar after cure of chronic HCV infection.
        Nat Immunol. 2021; 22: 229-239
        • Tonnerre P.
        • Wolski D.
        • Subudhi S.
        • Aljabban J.
        • Hoogeveen R.C.
        • Damasio M.
        • et al.
        Differentiation of exhausted CD8+ T cells after termination of chronic antigen stimulation stops short of achieving functional T cell memory.
        Nat Immunol. 2021; 22: 1030-1041
        • Yates K.B.
        • Tonnerre P.
        • Martin G.E.
        • Gerdemann U.
        • Al Abosy R.
        • Comstock D.E.
        • et al.
        Epigenetic scars of CD8+ T cell exhaustion persist after cure of chronic infection in humans.
        Nat Immunol. 2021; 22: 1020-1029
        • Pischke S.
        • Hartl J.
        • Pas S.D.
        • Lohse A.W.
        • Jacobs B.C.
        • Van der Eijk A.A.
        Hepatitis E virus: infection beyond the liver?.
        J Hepat. 2017; 66: 1082-1095