【作者】 苏齐鉴；

【导师】 梁浩；

【作者基本信息】 广西医科大学 ， 流行病与卫生统计学， 2009， 博士

【摘要】 第一部分:广西HIV-1耐药性流行情况及其关联因素研究目的:调查广西HIV-1耐药性流行情况,并探讨可能与耐药性突变有关联的因素。方法:在广西柳州、南宁两市对304名HIV感染者/AIDS患者进行问卷调查,收集可能与HIV-1耐药性有关联的资料,并采集研究对象外周静脉血,提取病毒RNA,用巢式逆转录聚合酶链反应(nested RT-PCR)扩增编码病毒蛋白酶1～99位氨基酸和逆转录酶1～285位氨基酸的核酸序列,通过测序获得DNA序列。将序列提交到斯坦福大学HIV耐药性数据库,获得耐药性突变位点及耐药水平有关信息。应用MEGA 4软件构建系统进化树,分析样本序列的亚型。将已接受抗病毒治疗的患者按不同特征分组,用卡方检验或Fisher确切概率法比较不同对比组的耐药率,分析研究对象特征与耐药性的关联性。结果:获得219份DNA序列,其中86份来自接受抗病毒治疗的患者,133份来自未治疗者。经系统进化树分析,鉴定出218份样本的亚型,包括CRF01_AE 176份,CRF07_BC 26份,CRF08_BC 13份,B亚型2份,C亚型1份。蛋白酶抑制剂(PIs)、核苷类逆转录酶抑制剂(NRTIs)和非核苷类逆转录酶抑制剂(NNRTIs)相关突变的发生率分别为6.9%(14/219)、30.14%(66/219)和14.16%(31/219)。CRF07_BC发生PIs相关突变的频率高于CRF01_AE,CRF08_BC发生NRTIs相关突变的频率高于CRF01_AE和CRF07_BC。PIs次要突变A71V/T主要出现于CRF07_BC,NRTIs次要突变T69S主要出现于CRF08_BC。在219例患者中,至少对1种药物低度以上耐药者18例,总耐药率8.22%(18/219),其中治疗患者耐药率为17.44%(15/86),未治疗者耐药率为2.26%(3/133)。在已经产生耐药的患者中,对NRTIs和NNRTIs的高度耐药分别占77.78%和72.22%,交叉耐药的比例均高达100%,二重耐药的比例也达77.78%。耐药与治疗时间,服药依从性和职业有关,与性别、年龄、文化程度、民族、病毒亚型、传播途径、治疗方案等因素无关。治疗1～2年与治疗时间≥2年耐药率的差异无显著性意义,而治疗时间1年以上者,耐药率高于治疗<1年者(OR=8.089,P<0.030)。服药依从性为50～90%的患者,耐药率高于服药依从性≥90%者(OR=8.591,P=0.030);农民患者的耐药率高于其它职业和无业人员(OR=4.223,P=0.017);农民患者服药依从性低于其它职业和无业人员(P=0.007)。结论:广西HIV-1耐药率处于较低的水平,但是对NRTIs和NNRTIs耐药的患者多为高中度耐药、交叉耐药和二重耐药。抗病毒治疗时间、服药依从性和职业因素与耐药有关,抗病毒治疗1年以上、服药依从性低于90%、职业为农民,耐药率相对较高。农民耐药率较高可能与服药依从性较低有关,提供方便的取药渠道,加强教育以提高农民患者服药依从性可能有助于降低农民的耐药率。感染不同亚型的患者耐药率没有差异,但是不同亚型耐药模式有所不同,A71V/T主要出现于CRF07_BC,而T69S主要出现于CRF08_BC。第二部分:CRF01_AE、CRF07_BC和CRF08_BC主要耐药性突变的基因屏障研究目的:通过计算CRF01_AE、CRF07_BC和CRF08_BC发生PIs、NRTIs和NNRTIs主要耐药性突变的基因屏障,从理论上比较不同亚型发生特定耐药性突变的可能性大小,分析不同亚型的耐药模式是否有差异。方法:从本研究第1部分获得pol区DNA序列中,选取全部未治疗患者的序列133株,经过质量筛选和基因重组分析,保留测序质量好、重组断点与参考序列比较无显著偏移的序列,检查密码子简并性,只保留少于2个简并碱基的密码子。通过查询斯坦福大学的HIV drug resistance database数据库,确定PIs主要耐药性突变21个、NRTIs主要耐药性突变16个和NNRTIs主要耐药性突变18个,对照密码子表确定耐药性密码子及其对应的野生型密码子种类。统计每个野生型密码子突变为耐药密码子所需的碱基转换和颠换的数量,根据碱基转换(ts)和基因颠换(tv)发生概率之比约为2.5: 1的规律,将每个碱基转换赋值为1,颠换赋值为2.5,计算每个突变位点所有赋值之和,即为某个序列发生该耐药性突变的基因屏障值。用Kruskal-Wallis test法比较3种不同亚型发生某耐药性突变的基因屏障均值,当差异有显著性意义时,用Nemenyi法进行两两比较。结果:蛋白酶编码区的大多数野生型密码子呈高度保守,只有L10和N88的野生型密码子在3种亚型间有较大差异,但是由于发生L10V/I和N88D/S突变时所需的碱基替换种类和数量均相同,因而3种亚型的基因屏障没有差异(P=1.000)。NRTIs相关突变位点T/S69、K70、L74、V75、V118、L210、T215和K219的野生型密码子使用存在程度不同的偏性,其中T/S69、V118和L210的偏性可对基因屏障产生影响,不同亚型T/S69D、V118I和L210W基因屏障的差异有显著性意义(P<0.001),其中,CRF01_AE和CRF07_BC发生T/S69D突变的基因屏障均高于CRF08_BC(P<0.01),而发生V118I和L210W的基因屏障均低于CRF08_BC(P<0.01)。在NNRTIs相关突变中,K103、V106、V108和V/I179的野生型密码子变异较大,其中V106M和V/I179F突变的基因屏障存在差异,经检验,只有V106M的差异有显著性意义(P<0.001),CRF07_BC发生V106M突变的基因屏障低于CRF01_AE和CRF08_BC(P<0.01),而CRF01_AE与CRF08_BC之间的差异无显著性意义。结论:3种亚型发生大部分主要耐药性突变的基因屏障没有差异,只有L210W、V118I、V106M和T/S69D有差异,提示了在相同的药物选择压力下,CRF01_AE和CRF07_BC比CRF08_BC更容易发生V118I和L210W突变;而CRF08_BC比CRF01_AE和CRF07_BC更容易发生T/S69D突变;CRF07_BC比CRF08_BC和CRF01_AE更容易发生V106M突变;相对于CRF08_BC,CRF07_BC发生V118I、L210W和V106M突变均比较容易。第三部分:广西HIV-1主要流行株亚型快速鉴定方法的建立目的:建立一种简便快捷的多重巢式PCR方法,用于鉴定CRF01_AE、CRF07_BC、CRF08_BC、B和C亚型毒株。方法:根据多重巢式PCR原理,针对HIV-1的gag基因区设计亚型特异性引物作内侧引物,以HIV-1 M组通用引物Gag F2/Gag e2作外侧引物。抽取已知HIV-1亚型的阳性样本58份,未知亚型的HIV阳性样本14份,另取已知为CRF01_AE、CRF07_BC、CRF08_BC、B、C亚型的样本各1份作阳性对照,以HIV阴性样本10份作阴性对照。从全血中提取人基因组DNA,用外侧引物进行第1轮PCR扩增待测样本DNA,然后以第1轮PCR产物为模板,用CRF01_AE、CRF_BC、B和C亚型特异性引物在同一反应体系中进行第2轮第1次PCR,扩增产物经2.5%琼脂糖凝胶电泳,根据各亚型目的条带大小来判断亚型。当判断结果为CRF_BC时,以第1轮PCR产物为模板,用另1套专用于鉴别CRF07_BC和CRF08_BC的引物进行第2轮第2次PCR,扩增产物进行2.5%琼脂糖凝胶电泳,根据目的条带的位置来鉴别CRF07_BC和CRF08_BC。为了验证鉴定结果的可靠性,用HIV-1 M组通用引物306/Cn-gag扩增第1轮PCR产物,回收约670bp的目的片段,通过基因测序和系统进化树分析以确定样本的亚型。分别计算两种方法的灵敏度,并以基因测序法为参照,计算多重巢式PCR法的特异度。然后从72份待测样本中,随机抽取CRF01_AE和CRF07_BC样本各10份,CRF08_BC样本5份,按相同的反应体系与反应条件重复实验3次,计算检测结果的重复性。结果:多重巢式PCR能扩增72份待测样本中的66份,包括CRF01_AE 40份,CRF07_BC 14份,CRF08_BC 7份,B亚型5份。10份HIV阴性对照样本均无扩增。306/Cn-gag同样能扩增66份样本,通过基因测序获得64份样本的DNA序列,其余2份样本测序失败,系统进化树分析鉴定出CRF01_AE 39份,CRF07_BC 14份,CRF08_BC 6份,B亚型5份。多重巢式PCR法的灵敏度为91.7%(66/72),基因测序法的灵敏度为88.9%(64/72),经Fisher确切概率法检验,两种方法灵敏度的差异无显著性意义(P=0.492)。多重巢式PCR法与基因测序法对64份样本的鉴定结果完全一致,据此计算出多重巢式PCR法的特异度为100%。3次重复检测均能正确鉴定出CRF01_AE、CRF07_BC和CRF08_BC的全部样本,因此3种亚型鉴定结果的重复性均为100%。结论:本研究建立的多重巢式PCR法可准确鉴定CRF01_AE、CRF07_BC、CRF08_BC和B亚型毒株。该法具有较高的灵敏度、特异度和重复性,操作简便快速、费用低,适用于广西以及周边地区对HIV-1毒株进行亚型鉴定。更多还原

【Abstract】 Part One:Prevalence of HIV-1 drug resistance and factors related to drug resistance.Objective To investigate the prevalence of HIV-1 drug resistance in Guangxi, and explore the factors that may be related to HIV-1 drug resistance. Methods A survey was performed among 304 HIV-positive and AIDS subjects through questionnaires to collect informations on HIV-1 drug resistance. Peripheral venous blood samples were collected from all subjects. RNA were extracted from plasma, and nested RT-PCR were employed to amplify nucleic acid fragments encoding protease ( amino acids1-99) and reverse transcriptase (amino acids 1-285), and the amplified products were subjected to DNA sequencing. Sequences obtained were then analyzed with Stanford HIV drug resistance algorithm for informations about mutations and levels of drug resistance. Viral subtypes were determined by phylogenetic analysis with MEGA 4 software. ART-treated subjects were devided into groups with different characteristics, and then resistance rates were compared among these groups with Chi-Square test or Fisher’s Exact test to observe the correlations between these characteristics and drug resistance. Results 219 pol sequences were obtained from 86 ART-treated subjects and 133 ART-naive subjects. Phylogenetic analysis indicated that subtypes could be determined for 218 samples, in which 176 were designated as subtype CRF01_AE, 26 as CRF07_BC, 13 as CRF08_BC, 2 as B, and 1 as C. One sample was undetermined with phylogenetic analysis. Mutations associated with resistance to PIs, NRTIs, and NNRTIs were present in 6.9% (14/219), 30.14% (66/219), 14.16% (31/219) of samples. CRF07_BC strains had a significantly higher prevalence of PI-related mutations than CRF01_AE strains, however, CRF08_BC had a higher prevalence of NRTI-related mutations than CRF01_AE and CRF07_BC. PI minor mutation A71V/T was more common in CRF08_BC than in CRF01_AE, and NRTI minor mutation T69S occured mainly in CRF08_BC, but rare in CRF01_AE and CRF07_BC. 18 out of 219 samples (8.22%) were found to harboured mutations that confered low to high levels of resistance to at least one antiretroviral drug (ARV), Of which 15 were from ART-treated subjects and 3 from ART-naive subjects, giving a resistance rate of 17.44% (15/86) for ART-treated subjects and 2.66% (3/133) for ART-naive subjects, respectively. Among the subjects harboured resistance-related mutations, the percentage of high-level resistance to NRTIs or NNRTIs was 77.78% and 72.22%, respectively, and the percentages of cross-resistance were both 100% within each drug class and 77.78% between two drug classes. Time to initial ART, Adherence to ART, and employment were related to resistance rate, other factors such as age, sex, education, nationality, viral subtype, transmission route, and regimen were not related to resistance risk. Resistance risk was not significant different between the group experienced 1 to 2 year ART treatment and the group experienced≥2 year ART treatment, however, a significant higher resistance risk was observed among subjects exposed to ART treatment exceeding 1 year compared to those exposed to treatment less than 1 year(OR=8.089,P<0.030). Resistance rate was higher in the group with 50-90% adherence compared to the group with≥90% adherence(OR=8.591, P=0.030). Farmer has a higher resistance risk than other employment workers and those unemployed (OR=4.223, P=0.017), and dose adherence was lower in farmer than in other employment workers and those unemployed (P=0.007). Conclusions The prevalence of HIV-1 drug resistance in Guangxi is at a relatively low level compared to many other provinces of China, however, high level resistance, cross resistance within or between drug classes are common in patients harboured mutations related to NRTIs and NNRTIs resistance. Therapy time, adherence to ART, and employment is related to resistance. Over 1 year of ART treatment, with low than 90% adherence, and farm employment may related to higher rate of drug resistance. Higher rate of drug resistance of farmers is related to lower adherence to ART regimens. It may be possible to reduce rate of drug resistance of farmers by providing a more convenient access to ART and education to promote adherence to ART regimens. The rates of drug resistance are not different among patients infected with different subtypes of HIV-1, but patterns of drug resistance may vary according to subtypes of the virus. A71V/T occur more frequently in CRF07_BC strains, however, T69S mainly in CRF08_BC strains.Part Two: Study on Genetic barrier to primary resistance mutations in subtypes CRF01_AE, CRF07_BC, and CRF08_BC.Objective To compare the genetic barriers to development of primary mutations among CRF01_AE, CRF07_BC, and CRF08_BC isolates, and discuss theoretically the genetic reasons underlying varying patterns and probabilities of resistance related mutations in different subtypes. Methods Selected pol sequences from all 133 ART naive subjects according to the result of Part One of the present study. Quality of sequences were checked and then subjected to recombinant analysis before inclusion to assure the quality of sequences and simgle subtype in pol region. Sequences with recombinant breakpoints that were not obviously different from reference sequences and codons with less than two ambiguities per codons were included. 21 primary PIs mutations, 16 primary NRTIs mutations, and 18 primary NNRTIs mutions were included according to Staford HIV drug resistance database. Drug resistance codons and their corresponding wild-type codons were determined using codon table, and then nucleotide transitions (ts) and transversions (tv) were count for each primary mutation. According to the phenomena that transitions occurr on average 2.5 more frequent than transversions, each transition was scored as 1, and each transversion scored as 2.5, and then, genetic barrier wich was taken as the sum of the scores was calculated for a particular drug resistance substitution. The mean genetic barriers for particular substitutions were compared among the three subtypes by Kruskal-Wallis test. When significance was found, the Nemenyi test was used to compare each pair of subtypes. Results Most wild-type codons were conserved among subtypes at positions where major PIs mutations may occur, only at position L10 and N88 codon usages varied between subtypes, but did not make an impact on genetic barriers because the types and quantities of nucleotide substitutions for particular resistance related mutations were the same in all three subtypes. At positions related to NRTIs resistance, different extent of variation in wild-type codon preferences were found among subtypes at position T/S69, K70, L74, V75, V118, L210, T215, and K219, and codon usage preferences at position T/S69, V118, and L210 had an impact on genetic barriers to resistance (P<0.001). CRF01_AE and CRF07_BC had higher genetic barriers for T/S69D substitution than CRF08_BC (P<0.01), and had lower genetic barriers for V118I and L210W substitution than CRF08_BC (P<0.01). At NNRTIs resistance-related positions, wild-type codon usage varied at position K103, V106, V108, and V/I179, and none but the synonymous differences at position V106 made an impact on genetic barrier for the V106M mutation (P<0.001). CRF07_BC had a lower genetic barrier than CRF01_AE and CRF08_BC (P<0.001), and no significance was seen between CRF01_AE and CRF08_BC. Conclusions The genetic barriers for evolution to most primary mutations are similar among CRF01_AE, CRF07_BC, and CRF08_BC, however, the genetic barriers vary among subtypes for L210W, V118I, V106M, and T/S69D mutations. These various genetic barriers for different subtypes suggest that under the same ARV selection, development of V118I and L210W may be easier for subtypes CRF01_AE and CRF07_BC than for CRF08_BC, but contrarily, CRF08_BC may be easier to develop T/S69D substitution than CRF01_AE and CRF07_BC. The V106M substitution may be more common for CRF07_BC strains that for CRF08_BC or CRF01_AE strains. Development of V118I, L210W and V106M may be more common for CRF07_BC strains that for CRF08_BC strains.Part Three: Development of a rapid subtyping assay for HIV-1 strains circulating in Guangxi.Objective To develop a nested multiplex PCR assay for easy determination of HIV-1 subtype CRF01_AE, CRF07_BC, CRF08_BC, B, and C circulating in Guangxi. Method Subtype-specific primers (inner primers) were designed for subtypes CRF01_AE, CRF07_BC, CRF08_BC, B, and C based on their gag sequences. Primer Gag F2 (forward) and Gag e2 (reverse) universal for HIV-1 M group strains were used as outer primers. 72 samples including 58 HIV-positive samples of known subtype, 14 HIV-positive samples of unknown subtype were included in this study. In addition, 5 positive control samples were used as posittive control, and subtype CRF01_AE, CRF07_BC, CRF08_BC, B and C had 1 sample each, and 10 samples from HIV-uninfected subjects were used as negative control. DNAs were extracted from whole blood and used as templates for first-round PCR with the primer pair Gag F2/Gag e2. Second-round PCR was performed by using the first-round PCR products as templates with subtype-specific primers for subtypes CRF01_AE, CRF_BC, B, and C. All these subtype-specific primers were simultaneously contained in one reaction mixture. PCR products were analyzed by electrophoresis on a 2.5% agarose gel, and subtype determination was made based on PCR product sizes. When CRF_BC were determined, in order to distinguish between CRF07_BC and CRF08_BC, an additional PCR reaction was performed to amplified the first-round PCR products, and followed by an analysis of electrophoresis on a 2.5% agarose gel to determine if the subtype was CRF07_BC or CRF08_BC. To check the accuracy of the subtyping results by nested multiplex PCR, all 72 samples to be subtyped were amplified by using the first-round PCR products as templates with primer pair 306/Cn-gag, and the 670-bp products were purified and analyzed by DNA sequenceing and phylogenetic analysis. The subtyping results by nested multiplex PCR were compared with that by sequence-based phylogenetic analysis, and the sensitivity and specificity of nested multiplex PCR were evaluated. 10 samples belonging to CRF01_AE, 10 samples belonging to CRF07_BC, and 5 samples belonging to CRF08_BC were subtyped three times by nested multiplex PCR, and repeatabilities were computed for detection of the three subtypes. Results 66 out of the 72 HIV-positive samples could be amplified by nested multiplex PCR, and all the 10 HIV-negative samples were negative. Of the 66 samples amplified, 40 were designated as CRF01_AE, 14 as CRF07_BC, 7 as CRF08_BC, 2 as subtype B, and 1 as subtype C. 306/Cn-gag could also amplified the same 66 samples amplified by nested multiplex PCR, and 64 sequences were obtained by sequencing, the remaining 2 samples failed sequencing. The sensitivity was 91.7% (66/72) for nested multiplex PCR and 88.9% (64/72) for sequence-based phylogenetic analysis. There were no significance between two methods with respect to the sensitivity (Fisher exact test, P=0.492). Since the same results were given to the 64 samples by the two methods, the specificity of nested multiplex PCR was 100% compared to sequence-based phylogenetic analysis. All samples belonging to CRF01_AE, CRF07_BC, and CRF08_BC in the repeatability assay could be correctly determined in three detection, and repeatabilities were computed to be 100% for detection of these three subtypes by nested multiplex PCR. Conclusions We have developed a nested multiplex PCR assay for accurate determination of HIV-1 subtypes CRF01_AE, CRF07_BC, CRF08_BC,and B circulating in Guangxi, China. This assay has high sensitivity, specificity, and repeatability, and is a simple, rapid, and low-cost alternative to sequence-based phylogenetic analysis for subtyping of HIV-1 strains circulating in Guangxi and the surrounding areas.更多还原