【作者】 承海；

【导师】 徐德忠； 闫永平； 门可； 张景霞； 苏海霞；

【作者基本信息】 第四军医大学 ， 流行病与卫生统计学， 2009， 博士

【摘要】 目的和意义乙型肝炎(hepatitis B,HB)是严重威胁人类健康的重要传染性疾病,也是长期以来倍受全球关注的公共健康问题之一。我国是乙型肝炎高度流行区,其中宫内传播是主要传播途径之一。携带乙肝表面抗原(hepatitis B surface antigen,HBsAg)的孕妇其胎儿宫内感染发生率大约为5%-15%,这也是造成流行地区存在大量慢性HBV感染者的主要原因之一。目前认为,HBV感染的发生、发展和临床转归是一个在环境因素影响下,宿主因素和病毒因素相互制约、相互影响的过程。以既往本课题组在HBV宫内感染领域长期研究为基础,本研究分别从宿主和病毒角度出发,以宿主人类白细胞抗原(human leukocyte antigen,HLA)等位基因多态性和HBV基因组异质性为研究内容,通过比较发生宫内感染的母子与携带HBsAg但未发生宫内感染的母亲及其新生儿在上述两方面的差异,探寻可能与HBV宫内感染相关的HLA-DRB1等位基因和病毒基因组特异性突变,为进一步阐明HBV宫内感染机制和制定有效的防治策略提供科学的研究依据。方法1.以陕西省妇幼保健院(1994年7月-12月和2003年1月-2004年10月)和山西省太原市传染病医院(2003年8月-2006年4月)为研究现场,以研究期间连续入院分娩的HBsAg阳性产妇及其新生儿为研究对象,以母亲HBsAg阳性,新生儿于出生后24小时内、接受主(被)动免疫接种前,股静脉血清标本HBsAg阳性为宫内感染判定标准。从两地区分别收集12对(西安)和11对(太原)发生宫内感染的母子纳入宫内感染组;采用1︰2(病例︰对照)配比法,以来自于同一地区,母亲分娩时外周静脉血HBV血清标志物(HBsAg和HBeAg)滴度在两组间无显著性差异,两组新生儿同期出生、民族和性别一致为配比因素,在同期收集的研究对象中选择未发生宫内感染的46对母子纳入对照组。2.采用序列特异性引物PCR法(SSP-PCR)鉴定研究对象HLA-DRB1等位基因型,采用直接计数法计算和比较宫内感染组和对照组新生儿各等位基因频率的差异。3.将上述太原地区收集的8例HBV-DNA阳性(≥103copies/ml)的感染组新生儿及其母亲作为感染组研究对象;采用1︰1(病例︰对照)配比法,依前述配比因素从对照组中选取8例HBsAg和HBV-DNA阳性母亲作为对照。应用克隆-测序法,提取研究对象血清标本中HBV-DNA,采用高保真PCR扩增病毒2.5kb基因组片段,与T载体连接后应用蓝白斑法筛选,挑取阳性克隆进行测序分析。4.采用本课题组既往研究中应用的参考株序列,使用MegAlign软件分析HBV基因组2.5kb片段,判定病毒基因型和血清亚型,比较宫内感染组和对照组之间病毒基因组突变分布及频率的差异。5.从Genbank数据库中重新选择16株参考株序列(分别代表A-F基因型),使用RDP3beta软件对全部克隆序列及参考株进行重组分析。排除重组序列、无义突变和缺失后,使用ClustalX1.81软件对序列进行多重对位排列,使用MEGA3.1软件对排列结果进行编辑,使用Phylip软件采用邻接法构建系统发育树,使用PAML4软件包中CODEML程序,采用基于密码子替代模型的最大似然法对各组HBV蛋白质编码序列进行适应性进化分析,检测各组序列阳性选择位点。结果1.采用SSP-PCR方法鉴定宿主HLA-DRB1基因座13个等位基因。分析结果显示,在总体水平上13个等位基因频率在感染组新生儿和对照组新生儿间均未见统计学差异。但在感染组中,母亲为HBeAg阴性的新生儿携带DRB1*08和09等位基因频率分别为21.43%和35.71%,高于对照组DRB1*08和09等位基因频率(3.57%和7.14%),其中DRB1*09基因频率在两组间存在显著性差异(P=0.031,OR=7.22,95%CI为1.19-43.98)。分析结果同样显示,HBeAg阴性的感染组新生儿DRB1*08和09的基因频率(22.22%和27.78%)均显著高于对照组新生儿(P为0.038和0.026,OR为4.97和4.67,95%CI为1.19-20.79和1.29-16.93)。此外,HBeAg阴性的感染组新生儿携带DRB1*08和09等位基因的频率也高于HBeAg阳性的感染组新生儿(均为3.57%),其中DRB1*09频率在两类感染新生儿之间的差异具有统计学意义(P=0.028,OR=10.39,95%CI为1.10-98.20)。2.从24例研究对象的血清标本中共获得89株HBV克隆,每例研究对象平均3.7±0.9株(3至6株/人)。与各基因型参考序列比较后可见,89株克隆片段核苷酸序列与M12906(C基因型)平均相似率为97.88%,显著高于其它基因型参考序列(P<0.001),表明全部克隆均属C基因型。依据病毒克隆HBsAg蛋白序列第122位氨基酸残基为K,第160位为R,判定全部病毒克隆为adr血清亚型。3.以C基因型参考株(M12906)为标准进行突变位点分析,结果显示,C启动子区核苷酸序列中,对照组C1637G、G1658A、G1719T、A1727G、A1762T和G1764A突变频率显著高于感染组母子(P<0.05),此六个位点均位于核转录因子结合位点内,而G1742T突变频率在前者显著低于后者(P<0.01)。前C/C区共有13处位点突变频率在三组间存在显著性差异(P<0.05),但突变类型均为同义突变。前S/S区有显著意义的19处突变中,C3116T、C3175T、C96T、G162A、T473C和C705T等6处为错义突变,其中对照组病毒C3116T和C705T突变频率显著高于感染组(P<0.001),C3175T、C96T和T473C突变仅见于宫内感染组母子携带的病毒,且在感染新生儿携带的病毒中该类位点突变频率显著高于其母亲,G162A在各组病毒间突变频率也存在显著性差异,且在感染母亲组中突变频率显著高于对照组(P=0.0068)。此6处错义突变分别造成外膜蛋白前S1区A90V和P110S,前S2区P36L,S区S3N、C107R和A184V氨基酸替代,除S区S3N外,其余5处错义突变均位于外膜蛋白T细胞和(或)B细胞表位内。此外,T473C和T499A突变也造成聚合酶L461S和Y470N氨基酸替代。4.经RDP3beta软件分析,89株病毒克隆和16株参考序列均未发现重组信号,所有病毒克隆无重组株。采用CODEML法进行阳性选择位点分析可见,各组序列前C/C编码区和前S编码区选择模型(M2、M3和M8)分析结果均无显著意义;而在S区和聚合酶编码区,选择模型均显著优于中性模型(P<0.05)。优先模型M8(经验贝叶斯法)在三组病毒外膜主蛋白序列中均检出阳性选择位点(ω值分别为C组4.13、M组3.27、N组2.52),且该类位点多分布于外膜主蛋白主要亲水区。与C组相比,M组和N组病毒序列存在更多的阳性选择位点,尤其是氨基酸位点68T和126I仅在N组病毒序列中检出。聚合酶/逆转录酶编码区(与外膜蛋白编码区重叠部分)同样检出阳性选择位点,但与外膜主蛋白分析结果相比,其阳性选择位点在各组的分布呈现相反模式。结论1.母亲为HBeAg阴性的宫内感染新生儿携带HLA-DRB1*08和09等位基因频率均远高于对照组新生儿,而在HBeAg阴性的感染组新生儿中DRB1*08和09等位基因频率也均显著高于对照组新生儿及HBeAg阳性的感染组新生儿,此结果提示宿主HBeAg阴性时,DRB1*08和09等位基因可能与宫内传播相关,并可能为危险因素。2.包括A1762T/G1764A等在内的CP区6处核转录因子结合位点内突变在宫内感染组突变频率显著低于对照组,提示发生A1762T/G1764A等突变的病毒株不易在母胎间传播,即CP区保守的重要功能区可能通过调控病毒基因组复制和病毒抗原表达等相关机制,利于病毒宫内感染。这是关于CP区特异性突变,尤其是A1762T/G1764A双突变与HBV宫内传播间关系在国际上的首次明确报道。HBeAg/HBcAg抗原在宫内感染过程中也呈现一定的保守性。3.外膜蛋白中具有显著意义的错义突变多位于免疫表位内,表明外膜蛋白免疫原性的异质性可能与宫内感染相关。其中C3175T/C96T/T473C突变仅存在于宫内感染组母子携带的病毒株中,且在感染新生儿携带病毒中突变频率显著高于其母亲携带的病毒,表明出现此类突变的病毒准种易于在宫内环境中传播,即此类位点由于其突变使宿主易于发生HBV宫内感染。与之相反,C3116T和C705T等免疫表位内突变可能不利于HBV宫内感染的发生。本课题组在前期研究中曾发现上述两类位点,而此5个突变位点为同类研究中首次发现,是对本课题组既往研究成果的进一步补充。4.适应性进化分析结果同样显示HBeAg/HBcAg氨基酸序列在宫内感染过程中具有保守性。外膜主蛋白中所检出的阳性选择位点多分布于亲水区,提示宫内感染过程中存在的选择压力可能主要来源于宿主的免疫反应。宫内感染母子携带的病毒具有更多的阳性选择位点,且部分免疫表位内阳性选择位点具有组间特异性,提示宫内感染过程中,在宿主免疫压力的作用下,宫内感染母子所携带的病毒株外膜蛋白发生更为广泛的适应性变异,这种免疫表位多样性可能利于病毒在母胎间的传播,支持上述外膜蛋白免疫表位异质性可能与HBV宫内传播密切相关的结论。这也是首次运用生物信息学技术和方法对HBV基因组异质性参与宫内感染的可能机制进行的初步探讨。由此,本研究应用分子流行病学方法,从HBV基因组突变频率和适应性进化分析两方面论证了HBV基因组异质性和宫内感染的关系,且取得一致的结果。更多还原

【Abstract】 AIMHepatitis B (HB) is one of the most severe infectious diseases, which remains a serious global public health problem. In China, one of the major reasons for the high prevalence of hepatitis B virus (HBV) infection is that mothers transmit the HBV to their neonates during the intrauterine period. It is estimated that HBV infection occurs in approximately 5% to 15% of neonates born to mothers positive for hepatitis B surface antigen (HBsAg), which is an important cause for a large number of persistent HBsAg positive carriers in areas with a high HBV prevalence, too. At present, studies have shown that some factors including environmental factors, viral factors, host immune factors and genetic factors may be associated with HBV infection. Based on our previous studies on risk factors and mechanisms associated with HBV intrauterine infection, in the present study, HBsAg positive mothers with their neonates infected with HBV via intrauterine transmission were recruited as cases, and HBsAg positive mothers with their neonates, who had no HBV transmission, were selected as controls. The possible association between HLA-DRB1 polymorphism and disease susceptibility as well as resistance was investigated by comparing the frequencies of HLA genotypes between two groups of neonates. And the possible association between the HBV genomic heterogeneity and intrauterine infection were investigated by comparing the quasi species isolated from two groups of hosts. These investigations might be helpful to comprehensively understand the detailed mechanisms of HBV intrauterine infection and provide powerful support for the valid preventive measures.METHODS1. Subjects of this study had been recruited consecutively for our previous retrospective studies carried out in the Maternal and Neonatal Health Hospital of ShaanXi Province and the Infectious Disease Hospital of Taiyuan, Shanxi Province. HBV intrauterine infection was defined as the situation where the neonate was positive of HBsAg in serum specimens within 24 hours after birth and before the immunoprophylaxis, whose mother was HBsAg carrier but father was not. Twelve neonates (Shaanxi) as well as eleven neonates (Shanxi) who met this criterion with their mothers were classified into intrauterine infection group. Twenty-four (Shaanxi) and twenty-two (Shanxi) neonates who were not infected with HBV were collected from the same cohort and classified as controls (control group). Each case was compared with two controls based on factors which were associated with the pregnant women as well as the neonates. These factors included the HBV marker levels in the maternal serum after delivery immediately, similar age and same gender of the neonates.2. HLA-DRB1 alleles of neonates were typed using sequence specific prime-polymerase chain reaction technique (SSP-PCR). The frequencies of HLA-DRB1 alleles were calculated by direct counting and compared between two groups by the Fisher’s exact test.3. In the further study, eight HBV infected neonates with HBV-DNA positive and their mothers from Shanxi province were collected, as well as eight mothers from control group based on same factors. Quasi species of HBV were isolated from the maternal and neonatal serum specimens by using high fidelity PCR and cloning-sequencing techniques.4. Based on the reference sequences of genotypes A to H, the genotypes and the serotypes of clones were identified by analysis of the viral sequences using MegAlign software. And the difference in the proportion of clones with specific mutation was analyzed among the three groups by Fisher’s exact test.5. For the phylogenetic analysis, sixteen reference strains of genotypes A to F, which were reported to be representative of human HBV genotypes, were retrieved from Genbank database. All clone sequences with reference sequences were screened for recombination by the recombination detection methods with default parameters implemented in RDP3beta software. After the clones with recombination, deletion and premature stop codons having been excluded, viral sequences were aligned with reference strains by using the ClusatlX1.81 program and edited by MEGA3.1 software package. Phylogenetic tree was reconstructed by the neighbor-joining method (NJ method) using the programs in the Phylip software package. The selective pressure was analyzed by codon-based maximum likelihood method (CODEML) included in the PAML4 software package.RESULTS1. Thirteen alleles of HLA-DRB1 were identified in this study. There were no significant differences of allele frequencies between two groups. While the mothers were HBsAg carriers with HBeAg negative, the allele frequencies of DRB1*08 and 09 were higher in infection group (21.43% and 35.71%) than control group (3.57% and 7.14%). Especially, there was a significant difference of DRB1*09 frequency between two groups (P=0.031, OR=7.22, 95%CI 1.19-43.98). It was also observed that the allele frequencies of DRB1*08 and 09 in infection group of HBeAg-negative neonates (22.22% and 27.78%) were higher than those in infection group of HBeAg-positive neonates, and frequency of DRB1*09 was significantly different (P=0.028, OR=10.39, 95%CI 1.10-98.20). The allele frequencies of DRB1*08 and 09 in infection group of HBeAg-negative neonates were also significantly higher than those in non-infection group (P=0.038 and 0.026, OR=4.97 and 4.67, 95%CI 1.19-20.79 and 1.29-16.93).2. A total of 89 HBV clones were obtained from 24 hosts. The 2.5 kb nucleotide sequences of all clones were most similar to the genome of genotype C (reference strain M12906) with the mean similarity rate of 97.88%, which was highest among those of genotypes A to H (P<0.001). This result indicated that all clones belonged to genotype C. All clones belonged to serotype adr because their aa122 of HBsAg was K and aa160 was R.3. There were seven single nucleotide substitutions in the core promoter with significant difference in the proportions of mutations among three groups (i.e. C1637G, G1658A, G1719T, A1727G, G1742T, A1762T and G1764A). Except G1742T, the others were located in binding sites of transcription factors with significantly higher proportions in control group than infection group (P<0.05). Thirteen mutations in pre-C/C were detected with significantly different proportions among three groups, but all of them were synonymous and resulted in no amino acid substitutions in HBeAg/HBcAg. Nineteen point-mutations were detected in pre-S/S region with significant difference in the proportions of mutations among three groups, six of which were non-synonymous. Mutations of C3116T and C705T, which were detected in all clones isolated from group C, emerged with lower proportions in infection group. Mutations of C3175T, C96T and T473C were only detected in clones isolated from infection group and emerged more frequently in infected neonates. Except G162A, amino acid residues coded by the other five mutations were located in T and/or B cell epitopes of HBsAg (i.e. A90V and P110S in preS1, P36L in preS2, C107R and A184V in S). Among these point-mutations with statistic significances in the pre-S/S, only T473C and T499A were non-synonymous and resulted in substitutions of residues L461S and Y470N in the polymerase domain, respectively.4. Based on the analysis results, there were no recombination strains in all clones and reference strains. The CODEML analysis of pre-C/C and pre-S reading frames coding HBeAg/HBcAg and pre-S domain showed that there were no significant evidences for positive selection in both regions in this study. But in major S protein and polymerase protein, the selection models of M2, M3 and M8 significantly favored over the neutral models of M0, M1 and M7 (P<0.05). Furtherly, the positively selected sites were detected in the major S protein by M8 model with the Bayes Empirical Bayes (BEB) analysis and mainly distributed in the two major hydrophilic regions withωvalues 4.13 of group C, 3.27 of group M and 2.52 of group N. Compared with the analysis result of group C, more positively selected sites were detected in groups M and N. Particularly, residues 68T and 126I were detected only in group N. The positively selected sites were also identified in P gene reading frame overlapping with S gene among three groups, but showed a different pattern in the S antigenCONCLUSIONS1. While the mothers were HBsAg carriers with HBeAg negative, the allele frequencies of DRB1*08 and 09 were significantly higher in infection group than control group. It was also observed that the allele frequencies of DRB1*08 and 09 in HBeAg-negative neonates of infection group were higher than those in HBeAg-positive neonates of infection group, as well as significantly higher than those in non-infection group. These results indicated that HLA-DRB1*08 and 09 might be risk factors for intrauterine infection while the hosts were HBeAg negative.2. Including A1762T/G1764A, there were six point-mutations located in binding sites of transcription factors with significantly higher proportions in control group than infection group. It indicated a disadvantage for virus with specific mutations, such as A1762T/G1764A, to transmit from the mother to the fetus. It also revealed that conservation of the major functional regions in CP might be propitious to HBV intrauterine infection. To the best of our knowledge, this was the first report involving the definite association between the heterogeneity of CP and HBV intrauterine infection. Our results also indicated the conservation of HBeAg/HBcAg in intrauterine HBV transmission.3. The analysis results suggested that epitope variants of viral surface protein might contribute to viral intrauterine transmission by affecting immunogenicity of HBsAg. C3175T/C96T/T473C were detected only in the infection groups and emerged more frequently in infected neonates, which indicated that these mutations in epitopes might be tightly associated with HBV intrauterine infection with advantage for virus to transmit from the mother to the fetus. Our findings also indicated that two mutations of C3116T and C705T might be disadvantageous to intrauterine HBV transmission, which provided the supplement to our previous studies.4. The result of phylogenetic analysis also indicated that the conservation of HBeAg/HBcAg might be beneficial to HBV maternal-fetal transmission. The positively selected sites of major surface protein were located in hydrophilic regions which suggested that the selective pressure coming from host’s immune reaction might act dominantly on the surface protein during the HBV transmission from the mother to the fetus. It also supported the opinion that varieties in immune epitopes of HBsAg might contribute to intrauterine HBV infection.更多还原