N-Glycosylation Mutations within Hepatitis B Virus Surface Major Hydrophilic Region Contribute Mostly to Immune Escape
De-Min Yu a,
Xin-Hua Li a, f,
Vannary Mom b,
Zhong-Hua Lu d,
Xiang-Wei Liao a,
Yue Han a,
Christian Pichoud b,
Qi-Ming Gong a,
Dong-hua Zhang a,
Yan Zhang e,
Paul Deny b,
Fabien Zoulim b, c,
Xin- XinZhang a, Corresponding author contact information, E-mail the corresponding author, E-mail the corresponding author
a Department of Infectious Disease, Institute of Infectious and Respiratory Diseases, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, People’s Republic of China
b INSERM, U1052, 151 cours Albert Thomas, 69424 Lyon cedex 03, France
c Lyon University and Hospices Civils de Lyon, Lyon, France
d Wu Xi Hospital of Infectious Diseases, People’s Republic of China
e Ministry of Education Key Laboratory of Systems Biomedicine, Shanghai Center for Systems Biomedicine (SCSB), Shanghai Jiao Tong University, People’s Republic of China
f The Third Affiliated Hospital, Sun Yat-sen University, Guangzhou, People’s Republic of China
Abstract
Background & aims
HBV immune escape represents a challenge to prevention, diagnosis, and treatment of hepatitis B. Here, we analyzed the molecular and clinical characteristics of HBV immune escape mutants in a Chinese cohort of chronically infected patients.
Methods
Two hundred sixteen patients with HBsAg and anti-HBs were studied, with one hundred eighty-two HBV carriers without anti-HBs as a control group. Recombinant HBsAg bearing the most frequent N-glycosylation mutations were expressed in CHO and Huh7 cells. After confirming N-glycosylation at the most frequent sites (129 and 131), together with inserted mutations, functional analysis were performed to study antigenicity and secretion capacity.
Results
One hundred twenty-three patients had the wild-type HBs gene sequence, 93 patients (43%) had mutants in the major hydrophilic region (MHR), and 47 of the 93 patients had additional N-glycosylation mutations, which were transmitted horizontally to at least 2 patients, one of whom was efficiently vaccinated. The frequency of N-glycosylation mutation in the case group was much higher than that of the control group (47/216 Vs 1/182).Compared with wild-type HBsAg, HBsAg mutants reacted weakly with Anti-HBs using a chemiluminescent microparticle enzyme immunoassay. Native gel analysis of secreted virion in supernatants of transfected Huh7 cells indicated that mutants had better virion enveloping and secretion capacity than wild-type HBV.
Conclusion
Our results suggest that specific HBsAg MHR N-glycosylation mutations are implicated in HBV immune escape in a high endemic area. 作者: StephenW 时间: 2013-12-10 08:15
内乙型肝炎病毒的N-糖基化突变表面主要亲水区晴有助于免疫逃逸
德闵迂 a,
辛华丽 a,F ,
Vannary Momb ,
忠华路德,
翔伟Liao a ,
岳花,
基督教Pichoud b ,
齐明贡 a,
冬华Zhang a ,
严Zhang e ,
保罗Deny b ,
法比安斯基Zoulimb , C,
鑫XinZhang a ,通讯作者的联系信息,电子邮件通讯作者,电子邮件通讯作者
Patients who recover from acute hepatitis B clear surface antigen (HBsAg) from serum and anti-HBs, becomes detectable as hallmark of HBV immune control. However, anti-HBs appears weeks after HBsAg clearance and neutralizing antibodies are unlikely to contribute to early viral control but to prevent viral spread from cells that remain infected after hepatitis resolution [1]. This notion is supported by the evidence that chimpanzees resolving hepatitis B are protected from viral rechallenge [2]. Similarly, in non infected individuals an adequate anti-HBs production (>10mUI/ml) indicate a protective immunity after vaccination [1]. Several evidences suggest that anti-HBs is present in chronic HBsAg carriers also, but not detected because hidden in circulating immune-complexes [3]. Thus the simultaneous detection of HBsAg/anti-HBs (HBsAg/anti-HBs-codetection) represents an atypical serologic profile puzzling virologists and clinicians since the late '70ies [3-5]. Initially, it was explained by heterologous antibodies against subtypes different from the infecting HBV subtype as HBsAg subtypes d/y infections could associate with y/d anti-HBs subtypes [6]. The hypothesis of primary infection with 2 subtypes followed by clearance of only one
of them was contradicted by the animal model where humoral response against the a determinant of HBsAg prevented the re-infection with different HBV subtypes [3]. In addition, ay anti-HBs was found in HBsAg carriers from Japan, where this subtype is uncommon [7]. Therefore, the hypothesis of a mixed infection was dismissed in favour of clonal selection of antibody diversity[3]. The high similarity between 2 HBV serotypes (i.e HBsAg /ad and HBsAg/ay) could induce the clonal expansion not only of high-titre, high-affinity anti-a and anti-d, but also lower-titre, low affinity heterotypic anti-y [3-4]. Nevertheless, the question remains why only a minority of HBsAg carriers exhibit this condition. After the introduction of molecular biology techniques the characterization of viral quasispecies in clinical settings of HBV infections despite vaccination or HBIg treatment showed that HBsAg/anti-HBs-co-detection was associated with viral strains
carrying HBsAg mutations that escaped the anti-HBs neutralization [8-9]. These findings raise the question whether genetic heterogeneity of HBV and particularly of the S gene could be present in other HBsAg/anti-HBs carriers also.
Serum HBsAg results from one open reading frame providing 3 carboxy-terminal colinear proteins of different length (small, medium and large, SHBs, MHBs, LHBs). In addition to the S gene MHBs contains pre-S2 sequence (55 aa) and LHBs preS1 (108/119 aa depending on subtype) and preS2 sequences. PreS1 and PreS2 and the aa 100-169 domain of SHBs (Major Hydrophylic Region, MHR) are exposed on viral particles and highly immunogenic. The studies of HBsAg variants focused mainly on SHBs, because it is expressed at the highest levels, predominates in both virions and subviral particles and contains within MHR the “a determinant” (aa 124–147), main target of neutralizing antibodies. Two major loops are present in the “a” determinant (the 1st at aa 124-137; the 2nd at aa 139-147), defined by multiple potential disulfide bridges, between aa 139/147 (or 149) and 121/124; conserved cysteines at 124, 137, 139 and 149 are essential for conformation and antigenicity [9]. MHR is the principal anti-HBs binding site after natural infection, immunisation and HBIg prophylaxis, although areas up and downstream may be also
important in neutralisation. Initially characterized vaccine and HBIg escape mutants showed
glycine to arginine substitutions at position 145 (G145R); later other mutations between 120 and 208 were described in patients with active or passive immune-prophylaxis or CHB [10].
Overall, in both Western and Asian cohorts a significantly higher S gene variability, particularly of MHR with higher number of aa substitutions was found in HBsAg/anti-HBs positive patients than in controls [11-16]. Two French studies showed in the same patients a significantly higher number of aa changes in the "s" gene (aa substitution/100 aa: 4.55 vs 1.66), even more evident in MHR (5.71-4.68 vs 1.43-2.07) and "a" determinant (9.52-11.00 vs 2.43-2.48). Aa substitutions were significantly more frequent at positions 120,126, 129,130, 144 and 165; G145R/T was the most frequent [11-12]. These findings suggested that emergence of "a" determinant variants could favour viral escape from immune neutralization. Most of Asian studies confirmed the European findings except for a higher prevalence of aa changes in the 1st loop (aa 124-137) of "a" determinant than in the 2nd (aa137-145) [13-15]. When HBV genotype C infected individuals were analysed the classic G145R substitution was significantly less frequent, with prevalent mutations at aa 126, 129, 130, 131,133,134 confirming a major influence of genotype on selection of viral
quasispecies [13-15]. However, a potential role of immune suppression and antiviral therapy in European patients can not be rule out. Interestingly, some studies showed a significant variability of S gene at its CTL epitope (aa 87-95) that might contribute to HBV neutralization failure [13].
The major mechanism hypothesised for HBsAg/anti-HBs-co-detection remained the change in primary sequence of "a" determinant that alters its antigenic structure making the anti-HBs response less effective. Nevertheless, the real impact of many mutations on HBsAg antigenicity has yet to be determined. Yu D-M et al addressed this issue investigating the biological features of 3 MHR mutants, identified among 216 Chinese HBsAg/Anti-HBs carriers [16]. Authors confirmed a significantly higher heterogeneity of "s" gene in HBsAg/anti-HBs patients as compared to 182 controls, particularly within MHR (mutation prevalence 43.1% vs 15.5%). Interestingly, 22% (47 of 216) of HBsAg/Anti-HBs carriers harbored new N-glycosylation sites in the MHR as compared to only 1 of 182 (0.5%) controls. The 8 potential N-glycosylation sites were all located within or before the 1st loop of HBsAg (aa substitutions 113,114,116,123,129,130,131, with an additional
"N-X-T/S" motif by insertion between 114-115). They confirmed clustered mutations in 129,130 and 131 and selected mutations in 129 and 131, together with the 3 aa insertions functionally studied. They showed firstly by PNGaseF treatment and site direct mutagenesis, that these mutations produced new N-Glycosilation sites, then by deglycosylation experiments that the anti-HBs binding capacity of the 3 mutated proteins was several times lower. Finally,transcomplentation experiments with replication competent/secretion defective HBV plasmid and mutated HBV-env constructs showed an increase secretion of virions. These findings contribute to a better understanding of the underlying biological mechanisms of HBsAg/anti-HBs-co-detection and confirm a major role of the interplay between virus and immune system. On the one hand the HBV replication pathway leads to mutations in the S gene prompting the selection of variants with better fitness once the specific immune response has developed [16]. On the other hand when both HBsAg and anti-HBs are detected in highly viremic, HBeAg positive, immune tolerant carriersthe alternative possibility that the heterogeneity of the antibodies production could overcome the variability of the virus still holds true, as suggested by Zhang et al [17]. However, this is a classical
chicken and egg question since a single nucleotide change is enough to induce aa substitutions at position 120, 126 and 160 shifting subtype d to y, i to t and w to r [5].
In addition, Yu D-M et al demonstrate the infectivity of 129 mutant in one responder to HBV
vaccine suggesting that individuals harbouring these mutated HBV strains could represent a risky reservoir for the spread of HBV infection, even in vaccinated individuals [16]. However, such an event appear to occur only in specific clinical settings such as immune-compromised individuals who deserve a special attention and careful monitoring. Another case was a HBsAg/HBeAg positive immunetolerant patient, with initially undetectable anti-HBs, who turned positive at the time of transition to immune elimination phase, when the 131 mutant became detectable. According to Authors the mutant was horizontally transmitted at that time, but, alternatively, we may speculate that the variant was selected and emerged when the specific adaptive anti-HBV immune response was mounted [16]. This finding further supports the view that immune selection plays a pivotal role
and suggests that the prevalence of HBsAg/anti-HBs-co-detection should increase in CHB
patients. Actually, its reported prevalence is extremely variable, ranging from 2.5% to over 30% [5,11-17]; the changes of diagnostic criteria overtime could explain at least in part such variability, but patient heterogeneity seems a major cause. Accordingly, the HBsAg/anti-HBs prevalence is lower (2.5-5%) in HBsAg carriers identified by population screening, but increases up to 30% in CHB patients. Shields at al showed that 63% of CHB patients were HBsAg/anti-HBs positive as compared to 14% of asymptomatic carriers [18]. Furthermore, liver disease was shown to progress more frequently to HCC in Korean HBsAg/anti-HBs positive CHB patients who had also a higher prevalence of pre-S mutations (42.4% vs 20.4%) [19-20]. Yu et al found a higher prevalence of N-glycosylation mutation in HCC patients as compared to controls (53% vs 0%, p<0.001).
Overall these findings suggest that HBsAg/anti-HBs-co-detection is linked to a long lasting immune clearance phase that favours the selection of highly variable viral quasispecies not only in the S gene, but also in regions of the genome with relevant oncogenic potential.
In conclusion the possible clinical implications of HBsAg/anti-HBs-co-detection are different
according to the immune competence of the chronic HBsAg carrier as indicated in table n.1.
However, not all studies confirm the correlation between HBsAg/anti-HBs-co-detection and
severity of liver disease, thus its clinical significance remains to be better characterized in larger prospective cohort studies. 作者: StephenW 时间: 2013-12-10 08:33