Journal of Clinical Virology 61 (2014) 40–46

CD8+ T cell responses specific for hepatitis B virus core protein inpatients with chronic hepatitis B virus infection

Wei Cao, Zhifeng Qiu, Ting Zhu, Yanling Li, Yang Han, Taisheng Li *Department of Infectious Diseases, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, 1# ShuaifuYuan, Dongcheng District, Beijing 100730, China

A R T I C L E I N F O

Article history:Received 24 March 2014Received in revised form 3 June 2014Accepted 22 June 2014

Keywords:CCR7–CD27+ T memory cellHBV core proteinInactive HBsAg carrierSpecific T cell response

A B S T R A C T

Background: Chronic hepatitis B virus (HBV) infection includes a set of heterogeneous clinical patterns,and core-protein-specific T cell response is important for virus control and disease progression, yet is notwell elucidated.Objectives: To analyze the phenotypic and functional profiles of HBV-core-protein-specific CD8+ T cells indifferent clinical patterns of chronic HBV infection.Study design: A total of 46 HBV patients were recruited and classified according to their clinical status.CD8+ T cell responses in different patterns of chronic HBV infections were tested with flow cytometryusing overlapping 15-mer peptides covering HBV core protein. Meanwhile, the CCR7/CD27 phenotypes ofthese CD8+ T cells were also determined.Results: Frequencies of gamma interferon (IFN-g) positive CD8+ T cells in inactive HBV surface antigen(HBsAg) carriers in response to the core protein peptide pools were generally stronger than those ofchronic HBV carriers and resolved individuals, especially with regards to peptide pool C13–C24.Moreover, phenotypic studies further highlighted the group of CD8+ CCR7–CD27+ T memory cells, whichshowed significantly higher levels of IFN-g secretion in inactive HBsAg carriers than those in chronichepatitis B patients, chronic HBV carriers and resolved individuals.Conclusions: Core-protein-specific T cell response plays an important role in chronic HBV infection.Inactive HBsAg carriers showed a much stronger core-protein-specific cytotoxic T cell response thanother types of chronically infected patients. CD8+ CCR7–CD27+ T memory lymphocytes may be crucial inthe immune pathogenesis of chronic HBV infection.

ã 2014 Elsevier B.V. All rights reserved.

Contents lists available at ScienceDirect

Journal of Clinical Virology

journal homepage: www.elsevier .com/ locate / jcv

1. Background

Hepatitis B virus (HBV) infection is a global health problem.Approximately 350 million people worldwide have persistent HBVinfection, which leads to increased risk of cirrhosis or hepatocel-lular carcinoma. Although mechanisms for the chronicity of HBVinfection are not well elucidated, there is consensus thatpathogenesis of HBV infection is largely immune mediated, and

Abbreviations: HBsAg, hepatitis B surface antigen; IFN-g, gamma interferon;CTL, cytotoxic T lymphocytes; HLA, human leucocyte antigen; ALT, alanineaminotransferase; (anti-) HBe, (antibodies to) hepatitis B e antigen; (anti-) HBcAg,(antibodies to) hepatitis B core antigen; anti-HBs, antibodies to hepatitis B surfaceantigen; HIV-1, human immunodeficiency virus type 1; CMV, cytomegalovirus;PBMC, peripheral blood mononuclear cell; IL-2, interleukin-2; IQR, interquartilerange; AST, aspartate transaminase; CHB, chronic hepatitis B patient; CHBC, chronicHBV carrier; IHBC, inactive HBsAg carrier; RS, resolved individual.* Corresponding author. Tel.: +86 10 69155086; fax: +86 10 69155046.E-mail address: litsh@263.net (T. Li).

http://dx.doi.org/10.1016/j.jcv.2014.06.0221386-6532/ã 2014 Elsevier B.V. All rights reserved.

that CD8+ cytotoxic T lymphocytes (CTLs) play crucial roles in viralcontrol and liver inflammation [1,2]. In chronic HBV infection,virus-specific CTL responses are rather weak and limited comparedwith the robust responses in acute infection, which may lead toviral persistence and disease progression [3,4]. However, currentstudies are more focused on the dichotomy between T cellresponses in acute and chronic infections, obscuring the diversitywithin chronic HBV infection itself. Chronic HBV infection is highlyheterogeneous with variable levels of virus replication and liverinflammation. Active chronic hepatitis B can progress into cirrhosisor carcinoma if left untreated, while HBV carriers, in the absence ofliver inflammation, could be asymptomatic for life. However, inmost cases HBV-specific T cell impairments are analyzed with afew human leucocyte antigen (HLA)-restricted HBV epitopes[5,6,7], which is limited by HLA typing and host ethnicity [8,9].Available results are controversial, and there is still no conclusionwith regard to the exact role these T cells take in chronic HBVinfection [10]. To overcome these limitations, peptide poolscovering the whole HBV genome or part of it, instead of single

W. Cao et al. / Journal of Clinical Virology 61 (2014) 40–46 41

peptides, have been used to study HBV-specific T cell functions inrecent years [11,12].

It is known that pathogen-specific CTLs may differ in majorphenotypes and functions with different viral infections [13,14].However, studies of HBV-specific T phenotypes and functions inchronic HBV infection are very incomplete. A series of surfacemolecules have been known as markers of CD8+ T cell differentia-tion. CCR7 is a chemokine receptor involved in lymphocyterecirculation to secondary lymphoid tissues [15,16]. CD27, amember of the tumor necrosis factor receptor (TNFR) family, isa co-receptor in T cell regulation [17]. Both molecules are down-regulated with T cell differentiation towards the effector phase.And the chronological expression of CCR7+ CD27+, CCR7–CD27+and CCR7–CD27- can represent the early, median to late stages ofCD8+ T memory cell differentiation. Simultaneous analysis of thephenotypes and intracellular cytokine production of CD8+ T cells inresponse to HBV core peptide pools in chronic HBV infections, willhelp to better understand the pathogenesis of HBV chronicity.

2. Objectives

The present study aims to provide initial assessment of HBVcore-protein-specific CD8+ T cell function between subtypes ofchronic HBV infections by multiparameter flow cytometry, tofurther identify the contributory factors of disease outcomes.

3. Study design

3.1. Patients

Between October 2009 and April 2011, a total of 46 patientswith chronic HBV infection were enrolled from outpatient clinic ofinfectious diseases, Peking Union Medical College Hospital. Thesepatients were divided into four groups according to their clinicaland virological assessments, the chronic hepatitis B patients(13 patients, CHBs), chronic HBV carriers (10, CHBCs), inactiveHBsAg carriers (10, IHBCs) and resolved individuals (13, RS’s).Standards of patient selection and classification were based on the2009 AASLD (American Association for the Study of Liver Diseases)guidelines on chronic hepatitis B (Table 1) [18]. All of the subjectswere negative for antibodies to hepatitis C virus, humanimmunodeficiency virus type 1 (HIV-1), and for other markersof viral or autoimmune hepatitis. None of these patients hadreceived any antiviral therapy before. All subjects provided writteninformed consent.

Table 1Baseline characteristics of patients with chronic HBV infection.

Groups Chronic hepatitis Bpatients (CHBs)

Chronic HBV carriers (CHBC

No. 13 10

Male/Female 8/5 3/7

Age 44.2 � 10.5 32.7 � 9.1

Serum viral markersHBsAg + +

HBeAg + +

Anti-HBs � �

Anti-HBe � �

Anti-HBc + +

HBV DNAlog copies ml�1

6.63 � 1.01 6.46 � 2.03

ALT level U l�1 262.6 � 81.2 Normal rangea

a Normal range: ALT 5–40 Ul�1.

3.2. Virological assessment

Serum HBV DNA levels were assessed by real-time fluorescentquantitative polymerase chain reaction method (real-time-PCR)using real-time-PCR system (ABI Prism 7500, ABI) with a detectionlimit of approximately 103 viral copies ml�1. The experimentalprocedures were performed in strict accordance with the reagentkit (Da An Gene Co., Ltd.) package insert. HBsAg, anti-HBs, total andimmunoglobulin M anti-HBc, HBeAg and anti-HBe were deter-mined by commercial enzyme-linked immunosorbent assay(ELISA).

3.3. Synthetic HBV core protein peptide pools

Core protein of Hepatitis B virus genotype C (serotype adw2)consisting of 183 residues was selected (Entrez Protein Locus:AAP06598). A panel of 35 15-mer peptides overlapped by 10residues and covering the whole core protein sequence wereconstructed (C1, C2, . . . ,C34, C35) and synthesized by the ChinesePeptide Company, Co., Ltd. (Hangzhou, China). The purity of eachpeptide was determined to be greater than 80% by high-pressureliquid chromatography analysis. These 15-mer peptides werefurther pooled in three mixtures, C1–C12, C13–C24 and C25–C35,covering the core amino acids 1–70, 61–130 and 121–183,respectively. Detailed information of the synthetic peptidesequences is provided in Appendix A.

3.4. Isolation, stimulation and staining of peripheral bloodmononuclear cells (PBMCs)

PBMCs were isolated from 10 ml fresh heparinized blood eachby Ficoll-Hypaque density gradient centrifugation, purified andresuspended at the concentration of 3.0 � 106ml�1 in RPMI1640–10% fetal calf serum (FCS). PBMC stimulation was performedwith the three synthetic core peptide pools (single peptide2.5 mg ml�1) or with PMA (25 ng ml�1) plus ionomycin (1 mg ml�1)as a positive control at 37 �C for 18–22 h, with brefeldin A(10 mg ml�1, Sigma–Aldrich, U.S.A) present for the last 4 h ofstimulation. Negative controls were also prepared asabovementioned except for the stimuli.

Post-stimulation PBMCs were washed, surface-stained withanti-CD27 fluorescein isothiocyanate (FITC), anti-CCR7 phycoery-thrin (PE)-Cy7 conjugated, and anti-CD8 allophycocyanin (APC)-Cy7 antibodies (BD Pharmigen, U.S.A), permeabilized, and fixedwith Cytofix/Cytoperm (BD Pharmigen, U.S.A) according to themanufacturer’s instructions. Then, anti-IL-2 PE and anti-IFN-g

s) Inactive HBsAg carriers (IHBCs) Resolved individuals (RS’s)

10 136/4 7/637.8 � 8.1 25.5 � 0.7

+ �� �� ++ �� +Undetectable Undetectable

Normal rangea Normal rangea

42 W. Cao et al. / Journal of Clinical Virology 61 (2014) 40–46

Alexa-647 (BD Pharmigen, U.S.A) antibodies were added and keptfor 15 min, and the cells were washed twice and analyzed by flowcytometry.

3.5. Statistical analysis

Normal variables were summarized as means and standarddeviations, and non-normal variables as medians and interquartilerange (IQR). Normal data were compared by Student t-test orone-way ANOVA adjusted for multiple comparisons, as appropri-ate. Multiple comparisons of non-normal data were carried out byKruskal–Wallis test. All tests were two-sided, and a P-value �0.05was considered significant. P < 0.009 was considered significantfor further comparisons within samples. Associations betweenvariables were assessed using Spearman’s rank correlations. Allstatistical procedures were performed using SPSS 16.0 software(SPSS Inc., Chicago, IL, USA).

4. Results

4.1. HBV-specific CD8+ T cell responses to HBV core peptide pools

Frequencies of IFN-g positive and/or interleukin-2 (IL-2)positive CD8+ T cells in response to core peptide pools weremeasured in each patient group (Fig. 1A, Table 2). The totalfrequencies of IFN-g + CD8+ T cells upon stimulation of the corepeptide pools (C1-C35) in CHBs, CHBCs, IHBCs and RS’s were 0.135%(0.038%, 0.225%), 0.023% (0.003%, 0.198%), 0.136% (0.077%, 0.180%)and 0.027% (0.011%, 0.095%), respectively. IHBCs showed strongerresponses to the core peptide pools than the other groups(P < 0.009 v.s. CHBCs and RS’s), while responses of CHBs weremoderately but not statistically elevated. With regard to eachsingle peptide pool, IHBCs held much stronger responses topeptide pool C13–C24 than CHBCs and RS’s, with the frequencies ofIFN-g + CD8+ T cells 0.068% (0.029%, 0.151%) (P < 0.009) (Fig. 2).IHBCs also displayed stronger IFN-g responses to peptide pool C1–C12 compared with CHBCs and RS’s, though not statisticallysignificant. However, no marked difference was observed inresponse to peptide pool C25–C35 between the four groups. IL-2production and dual cytokine production in response to the

Fig. 1. Intracellular cytokine analysis by flow cytometry. (A) IFN-g and IL-2 production byIFN-g and IL-2 production in four phenotypes of CD8+ T cells: CCR7+ CD27-, CCR7+ CD

peptide pools were generally very low, and not distinguishablebetween groups (Table 2, Fig. 2).

4.2. CCR7/CD27 phenotype analysis of peripheral CD8+ T cells

Frequencies of peripheral CD8+ CCR7+ T cells in CHBs, CHBCs,IHBCs and RS’s were 34.1% � 14.3%, 49.4% � 17.8%, 48.0% � 16.0%and 57.8% � 20.4%, respectively. Frequencies of peripheral CD8+CD27+ T cells were 48.2% � 16.1% for CHBs, 64.2% � 16.1% forCHBCs, 65.2% � 14.0% for IHBCs, and 76.2% � 14.7% for RS’s. CHBsshowed the lowest levels of CD27 and CCR7 expression (Fig. 3).

More specifically, frequencies of peripheral CD8+ CCR7+ CD27+T cells in CHBs, CHBCs, IHBCs and RS’s were 32.5% � 14.7%,48.2% � 18.0%, 46.5% � 15.8% and 57.1% � 20.5%, respectively. Fre-quencies of dual negative CD8+ T cells in CHBs, CHBCs, IHBCs andRS’s were 50.3% � 15.6%, 34.6% � 15.8%, 33.3% � 13.7% and23.1% � 14.6%, respectively. Again, CHBs showed a marked lowerlevel of CD8+ CCR7+ CD27+ T cells, with elevated proportion of CD8+ CCR7–CD27- T cells. No significant differences have beenobserved between the CD8+ CCR7–CD27+ phenotypes in eachgroup (Fig. 3).

4.3. Phenotypic and functional profiles of core-protein-specific CD8+ Tcells

The intracellular cytokine (IFN-g and/or IL-2) productions ofvarious CCR7/CD27 phenotypes were also compared (Fig. 1B,C).And the major differences existed in cytokine responses to peptidepool C13–C24. Frequencies of IFN-g + CD8+ CCR7–CD27+ T cells inIHBCs were 0.103% (0.063%, 0.234%), markedly higher than those inthe other three groups (P < 0.009). Frequencies of IFN-g + CD8+CCR7–CD27- T cells in IHBCs and CHBCs were 0.034% (0.008%,0.131%) and 0.033% (0%, 0.110%), respectively, both higher thanthose in RS’s (P < 0.009) (Fig. 4). No significant difference wasobserved with the C1–C12 and C25–C35 pools, in terms of theresponsive cytokine production of each phenotype in differentgroups (data not shown here).

A correlation analysis between the level of CCR7 or CD27expression and intracellular IFN-g production was also done.Surface expression of both molecules was inversely correlated with

the CD8+ T cell population; (B) The CCR7/CD27 phenotypes of CD8+ T cells; and (C)27+, CCR7–CD27-, CCR7–CD27+.

Table 2HBV-core-protein-specific CD8+ T cell responses in chronic HBV infection by cytokine production.

Cytokine production Chronic hepatitis B patients (CHBs) Chronic HBV carriers (CHBCs) Inactive HBsAg carriers (IHBCs) Resolved individuals (RS’s)

Peptide Pool C1–C12IFN-g + CD8+ T% 0.037 (0.02, 0.062) 0.012 (0, 0.027) 0.038 (0.014, 0.072) 0.012 (0.003, 0.027)IL-2 + CD8+ T % 0.015 (0, 0.042) 0 (0, 0.030) 0.012 (0, 0.017) 0.004 (0.003, 0.013)IL-2&IFN-g + CD8+ T% 0 (0, 0) 0 (0, 0) 0 (0, 0) 0 (0, 0004)

Peptide Pool C13–C24IFN-g + CD8+ T% 0.042 (0, 0.076) 0 (0, 0.032) 0.068 (0.029, 0.151)a 0.005 (0, 0.01)IL-2 + CD8+ T% 0.018 (0, 0.026) 0.023 (0.003, 0.110) 0.014(0.003, 0.062) 0.02 (0, 0.061)IL-2&IFN-g + CD8+ T% 0 (0, 0) 0 (0, 0) 0 (0, 0) 0 (0, 0.008)

Peptide Pool C25–C35IFN-g + CD8+ T% 0.025 (0, 0.061) 0 (0, 0.052) 0 (0, 0.029) 0.019 (0, 0.032)IL-2 + CD8+ T% 0.004 (0, 0.025) 0 (0, 0.007) 0.003 (0, 0.038) 0.007 (0, 0.018)IL-2&IFN-g + CD8+ T% 0 (0, 0) 0 (0, 0.006) 0 (0, 0) 0 (0, 0)

Peptide Pools C1–C35IFN-g + CD8+ T% 0.135 (0.038, 0.225) 0.023 (0.003, 0.198) 0.136 (0.077, 0.180)b 0.023 (0.011, 0.086)IL-2 + CD8+ T% 0.045 (0, 0.103) 0.036(0.009, 0.154) 0.031

(0.013, 0.122)0.043 (0.019, 0.08)

IL-2&IFN-g + CD8+ T% 0 (0, 0) 0.004 (0, 0.025) 0.004 (0,0.016) 0.008 (0.004, 0.016)

Cytokine production profiles of CD8+ T cells were measured in responses to core peptide pools as stated above, and were presented in the table as medians and IQR. IHBCshowed much stronger IFN-g response to the whole core peptide pools C1–C35 than the other groups (aP < 0.009 v.s. CHBCs and RS’s). They displayed stronger IFN-g responseto peptide pool C13–C24 than CHBCs and RS’s (bP < 0.009, Fig. 2), and also stronger response to C1–C12, though not statistically significant. Responses to C25–C35 showed nodifference between the four groups. IL-2 production or dual cytokine production were low and not distinguishable between different groups.

W. Cao et al. / Journal of Clinical Virology 61 (2014) 40–46 43

levels of intracellular IFN-g production (CCR7, r = �0.445, P< 0.001; CD27, r = �0.471, P < 0.001, figure not shown).

5. Discussion

Hypo-responsiveness of HBV-specific T cells has been consid-ered an important determinant of virus persistence in chronic HBVinfection. How they differ and affect the prognosis in chronic HBVinfection is still a mystery. In the present study, we explored core-protein-specific CD8+ T cell responses and their phenotypicprofiles in chronic HBV patients with divergent outcomes. Ofthe chronic HBV patients, IHBCs presented the strongest core-protein-specific CTL response. Further analysis showed adistinctive phenotypic T cell responsiveness, indicating CD8+

Fig. 2. HBV-specific CD8+ T cell responses (IFN-g) to peptide pools of HBV core protein.

C13–C24 in IHBCs were higher than those in CHBCs and RS’s (P < 0.009). Frequencies of IFhigher than those in CHBCs and RS’s (P < 0.009).

CCR7–CD27+ T subsets with stronger response might play animportant role in viral control and disease quiescence.

The overall core-protein-specific T cell response in chronic HBVinfection turned out to be quite weak, with a preference of IFN-gproduction over IL-2, which may indicate an altered or impairedcytokine profiles in chronic HBV infection [19–21]. Still, a generallystronger CTL response was observed in inactive HBsAg carriers, inresponse to both the peptide pool C13–C24 and the whole mixedpeptides, which indicated possible better preservation of T cellfunctions in this group. The IHBC state is determined by thepresence of the surface antigen but with an undetectable HBV DNAin PCR-based assays and repeatedly normal liver function. Onbiopsy, minimal or absence of liver disease activity could beobserved. We did not assess the pathological status of our

Frequencies of IFN-g positive CD8+ T cells in response to peptide pools C1–C12 andN-g positive CD8+ T cells in response to the whole peptide library in IHBCs were also

Fig. 3. CCR7/CD27 phenotype analysis of peripheral CD8+ T cells. Frequencies of peripheral CD8+ CCR7+ CD27+, CD8+ CCR7–CD27-, CD8+ CCR7–CD27+, CD8+ CCR7+ and CD8+CD27+ T cells in patients with chronic HBV infection were shown, with P values marked in the figure. Levels of CCR7 and CD27 expression in CHBs were much lower than thosein other patients.

Fig. 4. C13–C24 specific CD8+ T cell response in different CCR7/CD27 phenotypes. Responses of CD8+ CCR7-CD27+ T cells to peptide pool C13–C24 in IHBCs were significantlyhigher than those in other infected patients (P < 0.009). Frequencies of IFN-g positive CD8+ CCR7–CD27- T cells to the same peptide pool in IHBCs and CHBs were both higherthan those in RS’s (P < 0.009).

44 W. Cao et al. / Journal of Clinical Virology 61 (2014) 40–46

W. Cao et al. / Journal of Clinical Virology 61 (2014) 40–46 45

outpatients. However, all the recruited IHBCs completely met theclinical standards, and had been regularly followed in the clinic fordecades before being considered for this study. This non-/or very-low-replicative immune tolerant phase with cessation of HBVactivation and histological remission of liver disease could bepermanent [22,23]. Reported long-term follow-up of thesepatients suggested relative benign prognosis with very low riskof disease progression [24–26]. However, little was known abouttheir cellular immune functions. We found for the first time thatthe average level of HBV core-protein-specific CTL response inIHBCs is much higher than those in other chronic HBV patients,indicating its important role in disease control. Many studiesshowed an inverse correlation between the CTL intensity and theviraemia levels [5,12,21]. In IHBCs, the existing low level of viralantigen may present as an effective stimulus to the immunesystem, which relatively preserves the T cell function. In contrast,overload or absence of viraemia often fails to induce potent T cellresponse, just as situations in chronic carriers or resolvedindividuals, with the latter the weakest T response. Nevertheless,20–30% of IHBCs may end up with spontaneous reactivation ofhepatitis B after a long period of quiescence [25,27]. In our study,we did notice variations in response levels of inactive carriers.Further follow-up is needed to decide whether thishyporesponsiveness predicts a later disease reactivation.

Use of mixed peptide pools of HBV core protein allowsidentification of immunogenic regions in this protein regardlessof the specific HLA-I typing. Most of the previous studies in thisfield were based on HLA-A2 restricted epitopes, which were morecommon in Caucasian patients. However, a large population ofchronic HBV infection lives in Asia, with quite different HLA-Iprofiles, such as HLA-A11, -A24, -A33 and -A30 [8,9,28–30]. Ourresults indicated possible epitopes in peptide pools C1–C12 andC13–C24 (core 1–70 and core 61–130, respectively), especially core61–130, due to the stronger T cell response in these regions. Inaddition to those previously well-defined HLA-A2 and -A24restricted epitopes, there are probably epitopes restricted by othercommon Asian HLA-I molecules existing in this area. Furtherexploration of these epitopes would enhance the understanding ofimmune responses in this large HBV infected population.

Further phenotypic studies showed a marked difference inCCR7/CD27 expression of peripheral CD8+ T cells in chronic HBVinfections. In CHBs, frequencies of CD8+ CCR7–CD27- T cells werethe highest. This elevated proportion of peripheral effector T typeindicated a persistently activated immune system of CHBs, thoughhyporesponsiveness of these cells may exist. By contrast, levels ofCCR7 and CD27 expression in both CHBCs and IHBCs werecomparable with those in RS’s, demonstrating relative immunequiescence of these patients. Further correlation analysis showedthat CCR7 and CD27 expression levels were inversely correlatedwith the CTL responses, which was in consistence with previousreports [31].

Comparison of the cytokine production between differentgroups highlighted the population of CD8+ CCR7–CD27+ T cells. Inthe better responsive IHBC group, CD8+ CCR7–CD27+ T cellsshowed stronger response to peptide pool C13–C24 as well as tothe whole peptide pools. In contrast, the effector CD8+ CCR7–CD27- T responses in CHBs and IHBCs were comparable to eachother. These data suggested that CD8+ CCR7–CD27+ T cells mayplay an important part in the control of virus replication inIHBCs, further contributing to the benign outcome. Studies of CD8

+ CCR7–CD27+ T cell population in healthy subjects havedemonstrated that this population belongs to a stage betweennaïve and effector CD8+ T cells, with cytotoxic function butweaker than that of the effector type, hence the name ‘pre-effector cells’ [13,31]. However, the exact role of this population ininfections has been poorly defined. It has been noted thatphenotypic diversity in pathogen-specific CD8+ T cells graduallyevolved with disease chronicity. For example, the major respon-sive CD8+ T cell subsets in vivo were reported the early typeCCR7 � CD27+ for chronic EBV and HCV infection, CCR7–CD27+for chronic HIV infection, and CCR7–CD27- for chronic cytomeg-alovirus (CMV) infection [14]. Likewise, we assumed that adistinctive highly responsive CD8+ T subset also exists in chronicHBV infection, probably CD8+ CCR7–CD27+ T cells. However,more studies are needed with a larger population and bettercharacterization of these cells. Another interesting finding is thatCD8+ CCR7–CD27- T cells, traditionally regarded as the majorcytotoxic effector, showed no difference in responses to corepeptide pools between CHBs and IHBCs, indicating the limitedimpact of these cells on the outcomes of chronic HBV infection.

In conclusion, defective specific T cell function is associated withHBV persistence and chronicity. Clinical features of inactive HBsAgcarriers make themselves a promising study population in theperspective of T function maintenance and long-term diseasecontrol. Our findings demonstrated a stronger core-protein-specificCD8+ T cell response in inactive HBsAg carriers. The low butadequate amount of circulating antigen in inactive carriers probablyserved as effective stimulus for CTLs, thus avoiding the viraemicexhaustion of specific Tcells. And the highlighted CD8+ CCR7–CD27+T cell subsets in these patients provided more insights into thecytokine strategies for functional T recovery. Therefore, virologicalcontrol in chronic HBV infection is a necessary but not the only stepin specific T function reconstitution [9,32].

Funding

This study was funded by the National Natural ScienceFoundation of China (Grant 81071372 to L.T.), and by NationalKey Technologies R&D Program for the 12th Five-year Plan (Grant2012ZX10001003-001 to L.T.).

Conflict of interest

None declared.

Ethical Approval

Not required.

Appendix A.

Sequences and purities of the HBV core protein pools:HBV core protein (183 amino acids, genotype C (serotype adw2),

Entrez Protein Locus: AAP06598):6 mdidpykefg asvellsflp sdffpsirdl ldtasalyre alespehcsp

hhtalrqail7 cwgelmnlat wvgsnledpa srelvvsyvn vnmglkirql lwfhiscltf

gretvleylv8 121 sfgvwirtpp ayrppnapil stlpettvvr rrgrsprrrt psprrrrsqs

prrrrsqsre sqc

Peptide pool C1–C12 Peptide pool C13–C24 Peptide pool C25–C35

No. Sequence Purity (%) No. Sequence Purity(%) No. Sequence Purity (%)

1 mdidpykefgasvel 82.3 13 cwgelmnlatwvgsn 84.6 25 sfgvwirtppayrpp 81.02 ykefgasvellsflp 81.9 14 mnlatwvgsnledpa 80.1 26 irtppayrppnapil 97.53 asvellsflpsdffp 85.9 15 wvgsnledpasrelv 81.3 27 ayrppnapilstlpe 89.84 lsflpsdffpsirdl 85.9 16 ledpasrelvvsyvn 81.9 28 napilstlpettvvr 91.65 sdffpsirdlldtas 88.4 17 srelvvsyvnvnmgl 87.4 29 stlpettvvrrrgrs 83.46 sirdlldtasalyre 95.7 18 vsyvnvnmglkirql 80.4 30 ttvvrrrgrsprrrt 87.77 ldtasalyrealesp 86.5 19 vnmglkirqllwfhi 94.9 31 rrgrsprrrtpsprr 94.78 alyrealespehcsp 80.8 20 kirqllwfhiscltf 80.4 32 prrrtpsprrrrsqs 92.29 alespehcsphhtal 82.4 21 lwfhiscltf gretv 91.3 33 psprrrrsqsprrrr 89.9

10 ehcsphhtalrqail 80.7 22 scltfgretvleylv 82.8 34 rrsqsprrrrsqsre 82.811 hhtalrqailcwgel 89.4 23 gretvleylvsfgvw 83.9 35 prrrrsqsresqc 95.212 rqailcwgelmnlat 93.9 24 leylvsfgvwirtpp 88.5

46 W. Cao et al. / Journal of Clinical Virology 61 (2014) 40–46

References

[1] Bertoletti A, Gehring AJ. The immune response during hepatitis B virusinfection. J Gen Virol 2006;87:1439–49.

[2] Maini MK, Boni C, Lee CK, Larrubia JR, Reignat S, Ogg GS, et al. The role of virus-specific CD8 cells in liver damage and viral control during persistent hepatitisB virus infection. J Exp Med 2000;191:1269–80.

[3] Chisari FV, Ferrari C. Hepatitis B virus immunopathogenesis. Annu RevImmunol 1995;13:29–60.

[4] Bertoni R, Sidney J, Fowler P, Chesnut RW, Chisari FV, Sette A. Humanhistocompatibility leucocyte antigen-binding supermotifs predict broadlycross-reactive cytotoxic T lymphocyte responses in patients with acutehepatitis. J Clin Invest 1997;100:503–13.

[5] Webster GJ, Reignat S, Brown D, Ogg GS, Jones L, Seneviratne SL, et al.Longitudinal analysis of CD8+ T cells specific for structural and nonstructuralhepatitis B virus proteins in patients with chronic hepatitis B: implications forimmunotherapy. J Virol 2004;78:5707–19.

[6] Mizukoshi E, Sidney J, Livingston B, Ghany M, Hoofnagle JH, Sette A, et al.Cellular immune responses to the hepatitis B virus polymerase. J Immunol2004;173(9):5863–71.

[7] Shi TD, Wu YZ, Jia ZC, Zhou W, Zou LY. Therapeutic polypeptides based on HBVcore 18–27 epitope can induce CD8+ CTL-mediated cytotoxicity in HLA-A2+human PBMCs. World J Gastroenterol 2004;10(13):1902–6.

[8] Krausa P, Browning MJ. HLA-A2 polymorphism and immune functions. Eur JImmunogenet 1996;23:261–74.

[9] Tan AT, Loggi E, Boni C, Chia A, Gehring AJ, Sastry KS, et al. Host ethnicity andvirus genotype shape the hepatitis B virus-specific T-cell repertoire. J Virol2008;82(22):10986–97.

[10] Isogawa M, Furuichi Y, Chisari FV. Oscillating CD8+ T cell effector functionsafter antigen recognition in the liver. Immunity 2005;23(1):53–63.

[11] Chang JJ, Thompson AJ, Visvanathan K, Kent SJ, Cameron PU, Wightman F, et al.The phenotype of hepatitis B virus-specific T cells differ in the liver and bloodin chronic hepatitis B virus infection. Hepatology 2007;46(5):1332–40.

[12] Boni C, Fisicaro P, Valdatta C, Amadei B, Di Vincenzo P, Giuberti T, et al.Characterization of hepatitis B virus (HBV)-specific T-cell dysfunction inchronic HBV infection. J Virol 2007;81(8):4215–25.

[13] Romero P, Zippelius A, Kurth I, Pittet MJ, Touvrey C, Iancu EM, et al. Fourfunctionally distinct populations of human effector-memory CD8+ Tlymphocytes. J Immunol 2007;178(7):4112–9.

[14] Appay V, Dunbar PR, Callan M, Klenerman P, Gillespie GM, Papagno L, et al.Memory CD8+ T cells vary in differentiation phenotype in different persistentvirus infections. Nat Med 2002;8(4):379–85.

[15] Kim CH, Rott L, Kunkel EJ, Genovese MC, Andrew DP, Wu L, et al. Rules ofchemokine receptor association with T cell polarization in vivo. J Clin Invest2001;108(9):1331–9.

[16] Masopust D, Vezys V, Marzo AL, Lefrancoiset L. Preferential localization ofeffector memory cells in nonlymphoid tissue. Science 2001;291:2413–7.

[17] Hintzen RQ, De Jong R, Lens SM, van Lier PA. CD27: marker and mediator of Tcell activation. Immunol Today 1994;15(7):307–11.

[18] Lok A-SF, McMahon BJ. AASLD practice guidelines chronic hepatitis B: update2009. Hepatology 2009;50(3):1–36.

[19] Bertoletti A, Ferrari C. Innate and adaptive immune responses in chronichepatitis B virus infections: towards restoration of immune control of viralinfection. Gut 2012;61:1754–64.

[20] Sandra P, Shilpa C, Antonio R, Alexander E, Roger W, Nikolai VN. CD8+ T cellcontrol of hepatitis B virus replication: direct comparison between cytolyticand noncytolytic Functions. J Immunol 2010;184:287–95.

[21] Fisicaro P, Valdatta C, Massari M, Loggi E, Biasini E, Sacchelli L, et al. Antiviralintrahepatic T- cell responses can be restored by blocking programmed death-1 pathway in chronic hepatitis B. Gastroenterology 2010;138:682–93.

[22] Chu C, Liaw YF. Natural history of chronic hepatitis B virus infection: animmunopathological study. J Gastroenterol Hepatol 1997;12:S218–22.

[23] Sharma SK, Saini N, Chwla Y. Hepatitis B virus: inactive carriers. Virol J2005;82–6.

[24] Hsu YS, Chien RN, Yeh CT, Sheen IS, Chiou HY, Chu CM, et al. Long-termoutcome after spontaneous HBeAg seroconversion in patients with chronichepatitis B. Hepatology 2002;35:1522–7.

[25] De Franchis R, Meucci G, Vecchi M, Tatarella M, Colombo M, Del Ninno E, et al.The natural history of asymptomatic hepatitis B surface antigen carriers. AnnIntern Med 1993;118:191–4.

[26] Gigi E, Lalla T, Orphanou E, Sinakos E, Vrettou E, Gigi MR. Long term follow-upof a large cohort of inactive HBsAg (+)/HBeAg (–)/anti-HBe (+) carriers inGreece. J Gastrointestin Liver Dis 2007;16(1):19–22.

[27] Perillo RP. Acute flares in chronic hepatitis B: the natural and unnatural historyof an immunologically mediated liver disease. Gastroenterology2001;120:1009–22.

[28] Zhang GL, Yang Y, Feng Z, Zhang YZ, Feng ML, Ji YH, et al. Report of HLAdistribution of Chinese bone marrow bank (Shanghai). Chin J Organ Transplan2001;22(3):179–85.

[29] Wu GG, Deng ZH, Gao SQ, Cheng LH, Jin SZ, Zhou D, et al. Study of HLApolymorphism in the 6965 han bone marrow registry donors. Chin J Hematol2004;25(8):473–7.

[30] Tan AT, Koh S, Goh V, Bertoletti A. Understanding the immunopathogenesis ofchronic hepatitis B virus: an Asian prospective. J Gastroenterol Hepatol2008;23(6):833–43.

[31] Tomiyama H, Takata H, Matsuda T, Takiguchi M. Phenotypic classification ofhuman CD8+ T cells reflecting their function: inverse correlation betweenquantitative expression of CD27 and cytotoxic effector function. Eur J Immunol2004;34(4):999–1010.

[32] Boni C, Laccabue D, Lampertico P, Giuberti T, Viganò M, Schivazappa S, et al.Restored function of HBV-specific T cells after long-term effective therapy withnucleos(t)ide analogues. Gastroenterology 2012;143(4):963–73.