M A J O R A R T I C L E
A Common HLA–DPA1 Variant is a Major
Determinant of Hepatitis B Virus Clearance in
Ping An,1,aCheryl Winkler,1,aLi Guan,2Stephen J. O'Brien,3Zheng Zeng,4and the HBV Study Consortiumb
1Basic Research Laboratory and2Laboratory of Genomic Diversity, SAIC-Frederick, and3Laboratory of Genomic Diversity, National Cancer Institute-
Frederick, Maryland; and4Department of Infectious Diseases, Peking University First Hospital, Beijing, People's Republic of China
A recent genome-wide study showed that the single nucleotide polymorphisms (SNPs) in the HLA-DP
region were associated with chronic hepatitis B virus (HBV) infection in Japanese and Thai persons. We
tested the effects of HLA-DP SNPs for all major HBV outcomes in Han Chinese (n 5 1742): HBV resistance,
clearance, chronic infection, cirrhosis, and hepatocellular carcinoma. HLA - DPA1 rs3077 T was strongly
associated with decreased risk of chronic HBV infection (odds ratio, .62; P 5 .001), consistent with the
previous report. We showed for the first time to our knowledge that it is a predictor for HBV clearance (odds
ratio, 2.41; P ,.001). However, rs3077 was not associated with the development of cirrhosis or hepatocellular
Hepatitis B virus (HBV) infection is a significant public
health problem. The World Health Organization
estimates that 2 billion persons have been infected with
HBV and that 350 million persons live with chronic
HBV infection. Approximately 25% of persons with
chronic HBV infection will develop cirrhosis and
hepatocellular carcinoma (HCC) . HBV infection
rates are disparate among world populations with high
prevalence rates recorded in East Asia and Africa;
hepatitis B surface antigen (HBsAg) carrier rates are
reported to be 5.3%–12% in China, 8% in Thailand,
and ?10% in Africa . HBV infection in Asian pop-
ulations is maintained through mother-to-infant verti-
cal transmissionor early
Approximately 90% of infants and preschool children
with HBV infection will fail to achieve viral clearance
and develop chronic HBV infection, whereas only 5%–
10% of adult HBV infections lead to persistent infection
. A very small proportion of patients with persistent
HBV infection can spontaneously clear the virus with-
out treatment. It has been reported that, in western
countries, 1%–2% of HBV carriers become HBsAg
negative each year , whereas in populations where
HBV infection is endemic, such as HanChinese, therate
of HBsAg clearance is much lower (.05%–.1% per year)
Variable outcomes of HBV infection are likely to be
multifactorial, with environmental, viral, and host ge-
netic factors contributing to the observed variability in
HBV clearance and pathogenesis. Candidate gene asso-
ciation studies have implicated a number of genes in
HBV resolution or persistence, including HLA class I
and II alleles  and non-HLA genes (eg, MBP ).
HLA - DRB1*1302 was consistently associated with
HBV clearance in Gambian  and European American
Recently, Kamatani, et al  performed the first ge-
nome-wide association study for chronic HBV infection
with use of a 2-stage design with 786 Japanese with
chronic hepatitis B carriers and 2201 control subjects.
The group identified 11 correlated SNPs in the region of
HLA - DPA1 and HLA - DPB1 and validated 2
Received 28 July 2010; accepted 12 November 2010.
aPing An and Cheryl Winkler contributed equally to this work.
bMembers of the HBV Study Consortium are listed at the end of the text.
Potential conflicts of interest: none reported.
Reprints or correspondence: Ping An, MD, MPH, NCI-Frederick Bldg. 560,
Rm 21-19, Frederick, MD 21702 (Ping.An@nih.gov) or Zheng Zeng, MD, PhD,
Department of Infectious Diseases, Peking University First Hospital, Beijing, China
The Journal of Infectious Diseases
? The Author 2011. Published by Oxford University Press on behalf of the Infectious
Diseases Society of America. All rights reserved. For Permissions, please e-mail:
HLA-DPA1 Polymorphism and HBV Clearance
d JID 2011:203 (1 April)
independent SNPs in 3 additional Japanese and Thai pop-
ulations comprising .3000 patients and control subjects .
This study identified association of HLA - DPA1 SNPs rs3077
and HLA - DPB1 rs9277535 with chronic hepatitis infection, but
they did not determine whether the association with persistent
infection was attributable to clearance of HBV or resistance to
Because prevalence rates of HBV infection in China are ex-
tremely high, we established a population-based study to in-
vestigate host genetic factors associated with HBV infection and
pathogenesis in the Chinese Han population from northern
China . In this report, we determined the effects of HLA-DP
SNPs on HBV infection, clearance, and progression to cirrhosis
and HCC in Han Chinese .
PATIENTS AND METHODS
Multicenter Chinese HBV Cohort
The Chinese HBV Cohort enrolled participants during
2003–2007 from hospitals in cities in northern China. An
study. Local internal review board approvals from participating
hospitals and informed consent from each participant were
The case-control study comprises the full spectrum of HBV
infection status: HBV case groups include natural clearance,
chronic asymptomatic, and symptomatic HBV infection;
cirrhosis; and HCC plus a group of hypernormal control sub-
jects lacking serological evidence of previous or current HBV
infection. Case definitions are in accordance with the predefined
criteria , based on diagnosis protocol issued by the Associ-
ation ofInfectiousDiseases andParasitesDiseases of China .
A total of 1742 samples were genotyped and analyzed in this
study (Table 1). All individuals, except patients with HCC, were
at least 40 years of age at enrollment, to allow for sufficient time
of disease progression.
HBV clearance was defined for persons who had (1) tests
seropositive for antibody to HBsAg (anti-HBs) and antibody to
hepatitis B core antigen (anti-HBc) without the presence of
HBsAg or (2) anti-HBs positive and no self-reported and clinic/
hospital record of hepatitis B vaccination. All clearance partic-
ipants were at least 40 years of age at enrollment, to avoid po-
tential confounding by HBV vaccination that became available
in the mid-1980s; infant vaccination was introduced in 1992
. Hypernormal control subjects were at least 40 years of age
at enrollment and were seronegative for any HBV serological
markers and anti-HCV at enrollment. Chronic HBV infection
was defined by 2 positive test results at least 1 year apart for
HBsAg and anti-HBc and with no evidence of cirrhosis by
imaging and laboratory tests. We used the term ‘‘hypernormal’’
to distinguish from the general ‘‘normal’’ population, which has
a high prevalence of anti-HBc. All participants in this study were
negative for hepatitis A virus, hepatitis D virus, and hepatitis E
virus infection. Virological testing and criteria of cirrhosis and
HCC are presented in the Supplementary Methods.
Two SNPs, HLA - DPA1 rs3077 and HLA - DPB1 rs9277535,
were genotyped using TaqMan assays (Applied Bisosystems).
The rs3077 C/T (major/minor alleles in Asians) in this report
correspondstothe G/Aalleles, respectively,onthe reversestrand
in Kamatani et al .
We used SAS, version 9.13 (SAS Institute) for analyses. The
logistic regression model was applied to case-control compar-
isons for the different phenotypic outcomes for dominant and
To determine whether the HLA - DPA1 and HLA - DPB1 SNPs
were associated with chronic hepatitis B in Han Chinese, we
compared genotype frequencies between the chronically HBV
-infected groups (ie, chronic HBV infection, cirrhosis, and
HCC) and hypernormal control subjects lacking infection of
multiple hepatitis viruses.
HLA-DPAI rs3077 and HBV Infection
Carriage of the HLA - DPA1 rs3077 T allele was a protective
factor for chronic HBV infection (odds ratio [OR], .62; 95%
confidence interval [CI], .47–.83; adjusted P 5 .001; dominant
model); the results were similarly significant for the additive
model. Adjusting for age and sex did not change the results
(Table 2). This result replicates and extends the Kamatani et al
 findings of the rs3077 association with persistent HBV
infection in Japanese and Thai Asians to Han Chinese.
HLA-DPA1 rs3077 and HBV Clearance
We next determinedwhether HLA - DPA1 rs3077 was associated
with HBV clearance, a phenotype not explored by the genome-
Characteristics of Participants in the Hepatitis B Virus
Characteristic No.Female, %
Clearance 28730.0 50 6 9.6
46.0 6 7.0
49.9 6 9.0
51.4 6 10.4
48.7 6 7.3
Chronic hepatitis 59137.0
Liver cirrhosis 37040.0
HBV-negative control subjecta
NOTE. Abbreviations: SD, standard deviation.
aAll HBV serological markers were negative.
d JID 2011:203 (1 April)
d An et al.
Table 2.Association of HLA - DPA1 Variant rs3077 With Hepatitis B Virus (HBV) Outcomes
Cases, n (%) Controls, n (%) Dominant (T allele)a
Additive (T allele)a
TCCCTTTCCC OR (95% CI)P OR (95% CI)P
HBV chronic infection (n 5 1,218) Hypernormal individualsc(n 5 227)
113 (.09)485 (.40) 620 (.51)35 (.15) 103 (.45)89 (.39)unadj. .62 (.47–.83).0013.68 (.55–.84) .0002
adj. .62 (.47–.83).0013 .68 (.55–.84).0002
Clearance (n 5 287) HBV chronic infectiond(n 5 1,218)
41 (.14)160 (.56) 86 (.30)113 (.09) 485 (.40) 620 (.51)unadj. 2.42 (1.84–3.20)3.47 3 10210
4.65 3 10210
1.77 (1.47–2.14) 3.75 3 1029
4.31 3 1029
adj.2.41 (1.83–3.18)1.77 (1.46–2.14)
Cirrhosis (n 5 370) Chronic hepatitis B (n 5 591)
31 (.08) 143 (.39)196 (.53) 61 (.10)241 (.41) 289 (.49)unadj. .85 (.66–1.10).22 .87 (.71–1.06).17
adj. .84 (.64–1.10).2 .86 (.70–1.05).15
HCC (n 5 265)All other HBV chronic infectione(n 5 961)
23 (.09) 104 (.39) 138 (.52)92 (.10) 384 (.40)485 (.50) unadj..94 (.71–1.23) .64.94 (.77–1.16).58
adj. .93 (.70–1.23).6 .93 (.75–1.15).52
NOTE. CI, confidence interval; HCC, hepatocellular carcinoma; OR, odds ratio.
aAnalyses were done using logistic regression model, with adjustment (adj.) of age groups (,50 or >50 years) and sex, or without adjustment (unadj.).
bThe ’T’ or ’C’ allele corresponds to ’A’ or ’G’ allele, respectively, in Kamatani et al , which were referred to the reverse strand of the sequence.
cIndividuals without any HBV serological markers.
dIncludes chronic HBV infection, cirrhosis and HCC.
eIncludes chronic HBV infection and cirrhosis.
HLA-DPA1 Polymorphism and HBV Clearance
d JID 2011:203 (1 April)
wide association study . We compared a group of 287
individuals who spontaneously cleared HBV infection with
1216 individuals with chronic HBV infection. The rs3077 T
allele was strongly predictive of HBV clearance in the domi-
nant model (OR, 2.41; 95% CI, 1.83–3.18; P ,.001), with
similar effects for the additive model (OR, 1.77; adjusted P
,.001) (Table 2). Results were very similar with and without
adjustment for sex and age. We separately tested for associa-
tion with the group with only anti-HBs reflecting full reso-
lution of HBV (n 5 105) or with the group with both anti-
HBs and anti-HBc reflecting partial HBV clearance; results in
both clearance groups remained highly significant (Supple-
mentary Table S1), with a stronger effect for the complete
HLA-DPB1 rs9277535 and HBV Infection and Clearance
Similar analyses for HLA - DPB1 rs9277535 indicated a weaker
protective association (OR, .81; P 5 .036; additive model) with
chronic HBV infection and with failure of HBV clearance (OR,
1.29; P 5 .004; additive model) (Supplementary Table S2). This
SNP had stronger association in the Japanese group , raising
the possibility of different degrees of linkage disequilibrium
(LD) between Chinese and Japanese populations. Indeed,
comparison of LD structure between 2 populations with use of
55 HapMap genotyped SNPs in the HLA - DPA1/DPB1 region
in Japanese between HLA-DP1 and HLA - DPB1 (Supple-
mentary Fig. S1A and S1B); the level of LD between rs3077 in
HLA-DP1 and rs9277535 in HLA - DPB1 is weaker in Chinese
(D# 5 .75; r25 .40) than in Japanese (D# 5 .87; r25 .67). This
finding suggests that rs3077 is more likely to be either functional
or tracking causal variation, because its effects are consistent in
both the Japanese and Chinese Han populations, unlike
HLA-DPA1 rs3077 and Development of Cirrhosis and HCC
Because the impact on HBV infection and clearance is so pro-
nounced for HLA - DPA1 rs3077, we hypothesized that, in the
context of chronic HBV infection, the HLA - DPA1 locus might
influence progression to more severe outcomes of chronic HBV
infection by controlling HBV replication and/or modulating
immune inflammatory response. However, a comparison of 370
patients with cirrhosis with 590 patients with chronic HBV in-
fection without cirrhosis revealed no significant associations
(Table 2). The results were similar when patients with chronic
HBV infection were compared with patients with the more
severe phenotype of decompensated cirrhosis (data not shown).
With these sample sizes, we had 86% and 83% power to detect
an association with an OR of 1.5 with cirrhosis and HCC,
respectively. When comparing 265 patients with HCC with 961
patients with chronic HBV infection and cirrhosis , no associ-
ation of rs3077 was observed (Table 2). Consistent with HLA -
DPA1 rs3770, no associations with the more severe outcomes of
cirrhosis or HCC were detected for HLA - DPB1 rs9277535
(Supplementary Table S2).
Potential Regulatory Function of HLA-DPA1 rs3077
We explored the potential function of rs3077 throughseveral in-
silicon bioinformatic tools. The SNP rs3077 T was found to be
associated with higher expression levels of HLA - DPA1 in
peripheral blood mononuclear cells of European descent
(Supplementary Figure S2A) by a search in SNPExpress, a da-
tabase of the transcript and exon levels of gene expression as
measured by genome-wide gene expression and genotypes .
It remains to be demonstrated whether this association was also
present in Han Chinese samples because of the high level of
diversity at the HLA region. There are several mechanisms by
which HLA - DPA1 rs3077 may regulate expression levels: a
microRNA binding site is disrupted, which may affect micro-
RNA binding, leading to repressed translation or destabilized
mRNA(SupplementaryFigure S2B),orcould alter transcription
factor binding (Supplementary Figure S2C).
Class II HLA molecules, such as HLA-DP, present antigen to
CD41T helper cells. T cell helper response is crucial in HBV
clearance, as evidenced by the vigorous humoral immune
response observed in patients achieving spontaneous clearance
. In our study, we replicated the results observed in the
Kamatani et al  genome-wide association study, showing that
individuals carrying the rs3077 C allele were susceptible to HBV
infection, and we further showed, for the first time to our
knowlegde, that rs3077 C allele is also strongly associated with
failure to clear HBV infection. Furthermore, comparison of
the strength and effect size of rs3077 on HBV infection and
clearance in our study indicate that the rs3077T protective effect
is mainly attributable to more efficient viral clearance, rather
than decreased susceptibility to infection, an aspect that was not
determined in the previous study . Our study, together
with the study by Kamatani et al , indicates a major role of
HLA - DPA1 in modulation of HBV acquisition and resolution
of HBV infection.
The impact of rs3077C on the burden of HBV infection in
Asian populations is substantial, considering its high allele
frequency. The carriage of rs3077C may affect epidemic HBV
infection, because the increasing susceptibility of individuals
carrying the allele increases the number of transmitters and the
likelihood of a recipient developing a chronic infection. Other
environmental and genetic factors that contribute to HBV
infection remain to be identified.
SNP rs3077, located in the 3# untranslated region of HLA -
DPA1, may be tracking HLA - DPA1 alleles that bind relevant
HBV epitopes or tracking a variant in a regulatory element
d JID 2011:203 (1 April)
d An et al.
that affects HLA - DPA1 stability or expression. However,
there is evidence that the rs3077 C allele is associated with
lower HLA - DPA1 expression, suggesting that rs3077 risk
allele or its linked true regulatory SNP plays its role by down
regulating HLA - DPA1 expression. Lower levels of HLA -
DPA1 on target cell surfaces might be less effective in pre-
senting viral antigen to CD41T helper cells, leading to an
impaired immune response to viral invasion or to the reso-
lution of HBV infection. The reason why this SNP has no
obvious impact on disease progression to cirrhosis or HCC
remains to be explored. Further studies on HBV disease
progression, treatment response, and vaccine response are
warranted. Vaccine response to HBsAg is controlled in part
by the HLA-DR region , but the role of HLA - DPA1 in
vaccine response has not been investigated. Elucidation of the
mechanism of HLA-DP interaction with HBV will be a crit-
ical step in our understanding of immune regulation of HBV
infection, clearance, and vaccine response and may lead to
targeted epidemic control strategies.
HBV Study Consortium Participants: Department of In-
fectious Diseases, Peking University First Hospital, Beijing,
P.R.China (Zheng Zeng, Yanyan Yu, Xiaoyuan Xu, Haiying Lu);
Institute of Liver Diseases Research, Beijing Military General
Hospital, Beijing, P.R.China (Darong Hu); Beijing Ditan Hos-
pital (Rongbing Wang, Yifan Chen); Department of Surgery,
Beijing Institute of Tumor Prevention and Therapy, Beijing,
P.R.China (Cunyi Hao); Department of Infectious Diseases,
Shanxi Medical University, Taiyuan, P.R.China (Heping Zhou);
Department of Infectious Diseases, Qinhuangdao No. 3 Hos-
pital, Qinhuangdao, P.R.China (Zhonghou Han, Yuhua Wang,
Xiewen Sun); Department of Surgery, Inner Mongolia Medical
College, Hohhot, P.R.China (Lidao Bao, Xiping Zhang);
Department of Infectious Diseases, Xuzhou No. 3 Hospital,
Xuzhou, P.R.China (Dasi Guo); Department of Infectious Dis-
eases, Xinjian Medical University, Urumoqi, P.R.China (Yaoxin
Zhang); Department of Infectious Diseases, the Second Affiliate
Hospital of China Medical University, Shenyang, P.R.China
(Xiaoguang Dou); Institute of Liver Diseases Research, Peking
University Second Hospital, Beijing, P.R.China (Lai Wei);
Department of Surgery, Peking Union Medical College, Beijing,
P.R.China (Jingan Rui, Qiang Qu).
Supplementary dataareavailable athttp://jid.oxfordjournals.org
This work was supported in whole or in part with federal funds
from the National Cancer Institue, National Institutes of Health
(HHSN26120080001E), and the Intramural Research Program of the
National Institutes of Health, National Cancer Institute, Center for Cancer
We thank Dr. George Nelson, for statistical advice and comments on the
manuscript; Ms. Yu-Chun Zhou and Mr. Bailey Kessing, for technical and
database assistance; and the study participants. The content of this pub-
licationdoesnot necessarily reflectthe views or policies of the Departmentof
Health and Human Services, nor does mention of trade names, commercial
products or organizations imply endorsement by the US Government.
1. Yu MW, Hsu FC, Sheen IS, et al. Prospective study of hepatocellular
carcinoma and liver cirrhosisinasymptomatic chronichepatitis B virus
carriers. Am J Epidemiol 1997; 145:1039–47.
2. Custer B, Sullivan SD, Hazlet TK, Iloeje U, Veenstra DL, Kowdley KV.
Global epidemiology of hepatitis B virus. J Clin Gastroenterol 2004;
3. Alward WL, McMahon BJ, Hall DB, Heyward WL, Francis DP, Bender
TR. The long-term serological course of asymptomatic hepatitis B virus
carriers and the development of primary hepatocellular carcinoma.
J Infect Dis 1985; 151:604–9.
4. Chu CM, Liaw YF. HBsAg seroclearance in asymptomatic carriers of
high endemic areas: appreciably high rates during a long-term follow-
up. Hepatology 2007; 45:1187–92.
5. Yuen MF, Wong DK, Sablon E, et al. HBsAg seroclearance in chronic
hepatitis B in the Chinese: virological, histological, and clinical aspects.
Hepatology 2004; 39:1694–701.
6. Thio CL, Thomas DL, Karacki P, et al. Comprehensive analysis of class
I and class II HLA antigens and chronic hepatitis B virus infection.
J Virol 2003; 77:12083–7.
7. Thomas HC, Foster GR, Sumiya M, et al. Mutation of gene of
mannose-binding protein associated with chronic hepatitis B viral
infection. Lancet 1996; 348:1417–9.
8. Thursz MR, Kwiatkowski D, Allsopp CE, Greenwood BM, Thomas
HC, Hill AV. Association between an MHC class II allele and clearance
of hepatitis B virus in the Gambia. N Engl J Med 1995; 332:1065–9.
9. Kamatani Y, Wattanapokayakit S, Ochi H, et al. A genome-wide
association study identifies variants in the HLA-DP locus associated
with chronic hepatitis B in Asians. Nat Genet 2009; 41:591–5.
10. Zeng Z, Guan L, An P, Sun S, O’Brien SJ, Winkler CA. A population-
based study to investigate host genetic factors associated with hepatitis
B infection and pathogenesis in the Chinese population. BMC Infect
Dis 2008; 8:1.
11. Chinese medical association of infectious diseases and parasites and
Chinese medical association of hepatology. Protocol of prevention and
treatment in viral hepatitis. Chin J Hepatol 2000; 6:324–9.
12. Liang X, Bi S, Yang W, et al. Epidemiological serosurvey of hepatitis B
in China–declining HBV prevalence due to hepatitis B vaccination.
Vaccine 2009; 27:6550–7.
13. Heinzen EL, Ge D, Cronin KD, et al. Tissue-specific genetic control of
splicing: implications for the study of complex traits. PLoS Biol 2008;
14. Penna A, Del Prete G, Cavalli A, et al. Predominant T-helper 1 cytokine
profile of hepatitis B virus nucleocapsid-specific T cells in acute self-
limited hepatitis B. Hepatology 1997; 25:1022–7.
15. Hohler T, Reuss E, Evers N, et al. Differential genetic determination
of immune responsiveness to hepatitis B surface antigen and to
hepatitis A virus: a vaccination study in twins. Lancet 2002; 360:
HLA-DPA1 Polymorphism and HBV Clearance
d JID 2011:203 (1 April)