【作者】 董海燕；

【导师】 张建中； 万康林；

【作者基本信息】 中国疾病预防控制中心 ， 病原生物学， 2011， 博士

【摘要】 结核分枝杆菌的基因分型对研究结核病的暴发流行、传播模式、基因多态性及遗传学关系等具有重要作用。以PCR为基础的间隔区寡核苷酸分型(spacer oligonucleotide typing, Spoligotyping)和分枝杆菌散在重复单元-可变数目串联重复序列(mycobacterial interspersed repetitive units—variable number tandem repeat, MIRU-VNTR)是研究结核分枝杆菌基因分型常用的技术方法,已广泛应用于世界各地的结核病流行病学研究中。结核分枝杆菌的基因分型研究结果显示全世界的结核病的流行主要由几种结核分枝杆菌家族引起,并且不同的基因家族各具有独特的分子特征、地区性分布和致病性。因此研究我国结核分枝杆菌的基因多态性对于结核病的预防控制具有重要意义。为了解中国结核分枝杆菌的基因多态性、地区性分布以及主要的流行菌株特征,本研究从全国13个省、市、自治区收集2346株结核分枝杆菌菌株,采用Spoligotyping基因分型方法进行分析,并将2346株结核分枝杆菌的Spoligotyping指纹图谱与国际数据库SITVIT2进行比对。研究发现2346株结核菌株呈278种Spoligotyping基因型。其中2153株结核菌株可分为85个基因簇,成簇率为88.15%,而其余的193株结核菌株表现为独特的基因型。将所有菌株的基因分型结果与SITVIT2数据库进行比对分析,发现118种基因型在数据库中已存在,而其余160种基因型为新的基因型。通过结核分枝杆菌的基因家族分析,发现2346株结核分枝杆菌中包含有北京家族、CAS家族、Haarlem家族、T家族、EAI家族、MANU2家族、S家族、U家族和x家族10个家族的菌株。其中北京家族菌株最多,共1738株(包括典型的北京家族菌株和非典型的北京家族菌株),占所分析菌株的74.08%,为主要的流行菌株。进一步分析北京家族菌株在13个省的分布发现,北京地区北京家族菌株的比例最高,为92.59%,而福建省北京家族菌株的比例最低,为54.50%。北京周边等北方省市北京家族菌株的比例高于南方的一些省市,提示地域与环境因素可能与北京家族菌株的流行具有一定的关系。分析1738株北京家族菌株的基因多态性,发现仅呈现24种基因型,且DR区非常保守；北京家族菌株在我国的数量很多,而基因多态性却很低,提示北京家族菌株在我国的传播速度快。本研究中共涉及12株CAS家族菌株,这些菌株只在新疆和西藏发现,并且均分离自少数民族患者(3株分离自藏族患者,9株分离自维吾尔族患者)。提示不同的地域环境和民族构成可能与结核病的发病具有一定的关系。由于新疆和西藏均与印度接壤,而CAS家族菌株是印度的主要流行菌株,所以这些CAS家族菌株也有可能是通过旅游、贸易及移民由印度传入。本研究进一步对2346株结核菌株TbD1区是否缺失进行了分析,发现在7株EAI家族菌株中存在TbD1区,为古典型菌株,而在其它菌株中TbD1区均缺失,为现代型菌株。这些古典型EAI菌株只在福建发现,提示这些菌株可能具有地方性特点,也提示我国是一个相对现代的结核病流行的地区。西藏是中国结核病疫情最严重的地区之一,并且西藏是以藏族为主体的民族自治区,但是关于西藏藏族结核病患者结核分枝杆菌的基因多态性却了解不多。本研究对西藏藏族结核病患者中分离收集的577株结核分枝杆菌,用Spoligotyping和24位点MIRU-VNTR两种方法进行分析。目的是了解中国西藏藏族结核病患者中结核分枝杆菌的基因型特征、主要的流行菌株及相关影响因素等,并评价两种分型方法的分辨力,分析欧盟推荐的24个VNTR位点的分辨力并评价这些位点是否适合于西藏藏族结核病患者的分子流行病学研究。根据Spoligotyping分型结果,577株结核分枝杆菌呈23种基因型。其中563株结核分枝杆菌((97.57%,563/577)表现出12种已知的基因型并分属为4个基因家族：北京家族、T家族、CAS家族和MANU2家族；其它14株结核分枝杆菌表现出11种新的基因型。研究显示北京家族菌株(90.47%)为西藏藏族结核病患者的主要流行菌株。577株结核分枝杆菌的成簇性分析显示563株结核分枝杆菌可分为9个基因簇,每个基因簇的大小不同,菌株数量从2株到512株不等,成簇率为96.01%。本研究的高成簇率主要与Spoligotyping方法对ST1型菌株(512株,占所分析菌株的88.73%)的分辨率低有关。年龄、性别、卡介苗接种和治疗史等因素与北京家族菌株的相关性分析发现,北京家族菌株的流行与这些因素均无明显相关性(p>0.05)根据24位点MIRU-VNTR分型结果,577株结核菌呈347种基因型,299株结核分枝杆菌可分为69个基因簇,而其余的278株结核菌表现为独特的基因型。比较MIRU-VNTR和Spoligotyping方法的分辨力发现,MIRU-VNTR分型方法的分辨力明显高于Spoligotyping分型方法(Hunter-Gaston指数分别为0.992和0.221)。在24个MIRU-VNTR位点中,MIRU31、Qubllb、Qub26、Qub4156c、Mtub21、MIRU20和MIRU26七个位点的分辨力较高,而其余的17个位点的分辨力较低。577株结核分枝杆菌在该位点的重复次数全部为1,这表明西藏藏族结核病患者感染的是现代型的结核分枝杆菌。将MIRU-VNTR分型结果与MIRU-VNTRplus数据库中的菌株进行比对分析,577株结核菌分属于4个基因家族：北京家族、T家族、CAS家族和LAM家族。大部分菌株的基因家族分析结果与Spoligotyping结果相一致。但是有1株结核分枝杆菌的Spoligotyping分析结果显示为MANU2家族,而MIRU-VNTR结果显示为北京家族菌株；PCR扩增分析显示本研究中出现的MIRU-VNTR分型结果与Spoligotyping分型结果不符的菌株,可能是北京家族菌株与非北京家族菌株混合感染造成的。总之,本研究对中国多个省市的结核分枝杆菌的基因多态性、主要流行菌株进行了较全面的分析,并且第一次对藏族结核病患者感染结核分枝杆菌的基因多态性进行了分析。此外,本研究还评价了24位点MIRU-VNTR分型方法,尽管该方法的分辨率比Spoligotyping方法高,但是研究认为24个MIRU-VNTR位点中只有7个位点适合北京家族菌株的基因分型,而其余17个VNTR位点的分辨力较差。该24位点的MIRU-VNTR分型方法并不适合北京家族菌株流行地区的分子流行病学研究。由于不同地区结核分枝杆菌的基因多态性和流行的菌株并不相同,因此在对不同地区的结核分枝杆菌进行分子流行病学研究时,要考虑不同的VNTR位点组合,必要时要加入新的VNTR位点以增加基因分型的分辨力。更多还原

【Abstract】 The genotyping of Mycobacterium tuberculosis (M. tuberculosis) strains is important for TB control because it allows for the detection of suspected outbreaks and the tracing of transmission chains. It is also important to monitor species diversity as well as identify genetic relationship. Spacer oligonucleotide typing (Spoligotyping) and variable-number tandem repeats of mycobacterial interspersed repetitive units (MIRU-VNTR) are the common methods in the molecular epidemiological research in M. tuberculosis, and have been widely used in epidemiological research of tuberculosis around the world. The genotyping results of M. tuberculosis showed that the prevalence of tuberculosis in the world mainly caused by several special M. tuberculosis genetic lineage (family) strains. Each family strains have unique molecular characteristics, regional distribution and pathogenicity. So the genotyping of M. tuberculosis play an important role in the tuberculosis prevention and control.In order to understand the molecular epidemical characters of M. tuberculosis, such as genetic diversity, regional distribution and major epidemic strains, et al, a total of 2346 M. tuberculosis isolates from 13 provinces in China were genotyped by Spoligotyping. Two hundred seventy-eight spoligotypes were identified:2153 isolates were grouped into 85 clusters, and the remaining 193 isolates were orphans. Comparison with the SITVIT2 database revealed that 118 spoligotypes had shared international type numbers in the database and the other 160 spoligotypes were novel.Family assignment revealed that the most frequent strains were Beijing family strains (1738/2356,74.08%), including typical Beijing family strains and atypical Beijing family strains. In addition, the strains belonged to other family, such as T family, CAS family, EAI family, Haarlem family, S family, MANU2 family, X family, LAM family and U family, were also found in China.The percentage of Beijing family strains varied greatly among the 13 provinces (54.50～92.59%). The highest prevalence was found in Beijing and its surrounding areas, while a lower prevalence was found in Southern China. The geographical distribution of Beijing family strains provides evidence for the recent clonal expansion of this particular family strain. The low diversity and the highly conserved DR region of the highly prevalent Beijing family strains in this study might indicate that the Beijing family is spreading rapidly in China.One interesting finding in this study is that 12 strains tested belong to the CAS family, these strains were only found in Tibet and Xinjiang. Among the 12 CAS family strains,3 and 9 strains originated from patients of Tibetan and Uyghur ethnicity, respectively. This finding suggests that CAS family might have biogeographical specificity. Moreover, Tibet and Xinjiang share geographic borders with India where the CAS family is dominant. It has also been suggested that this family strains might be transported by trade, tourism, or migration from India.The TbDl analysis employed in this study showed that the majority of isolates (99.97%) belonged to the TbD1-/modern strains, whereas only seven strains (0.03%) were TbDl+/EAI isolates. These seven TbDl+/EAI strains were only found in Fujian province where located in Southeastern China. The M. tuberculosis strains that contained the TbD1 region were designated as ancestral M. tuberculosis strains, while the M. tuberculosis strains missing the TbD1 region were designated as modern M. tuberculosis strains. The predominance of modern M. tuberculosis strains in China supports the hypothesis that China has a relatively modern endemic focus of TB. Since the ancestral TbD1+/EAI strains were only found in Fujian province, this suggests that they are endemic there.Tuberculosis (TB) is one of the major public health problems in Tibet. However, information about lineages of M. tuberculosis isolates from Tibetan patients is unknown. In order to detect the genetic diversity of M. tuberculosis isolates from Tibetan patients in Tibet, compare techniques of MIRU-VNTR verus Spoligotyping for study of genetic pattern of M. tuberculosis strains, and evaluate the discriminatory power of each locus. Five hundred sevety-seven isolates from Tibetan TB patients throughout Tibet were genotyped by Spoligotyping and 24-locus MIRU-VNTR methods in this study.Based on Spoligotyping results, a total of 23 different spoligotypes were identified among 577 M. tuberculosis isolates. The majority (97.57%,563/577) of the isolates belonged to four known families:Beijing family, T family, CAS family and MANU2 family. While the other 14 isolates belonged to 11 novel spoligotypes. Classical Spoligotyping confirmed that Beijing family is highly prevalent in Tibet (90.47% in whole sample). A high rate of clustering (96.01%, nine clusters,2 to 512 strains/cluster) was observed upon Spoligotyping; most of it was due to the lower discriminatory power of this method for the ST1 genotype (n=512; 88.73% of the isolates). In this study, there is no relationship between Beijing family strains and age, sex, BCG vaccination and treatment history, respectively (p> 0.05)Using 24-locus MIRU-VNTR to genotype 577 M. tuberculosis isolates, three hundred forty-seven distinct MIRU-VNTR patterns were detected. Two hundred ninety-nine strains were grouped into 1 of 69 different MIRU-VNTR clusters and 278 isolates were unique. The Hunter-Gaston discriminatory index of the method was 0.991, indicating the higher power of discrimination of MIRU-VNTR than Spoligotyping (the Hunter-Gaston discriminatory index was 0.221). The discriminatory power of 7 loci (MIRU31, Qub11b2, Qub26, Qub4156c, Mtub21, MIRU20 and MIRU26) were moderately or highly discrimination accroding to their alllelic diversity, while the other 17 loci have poor discrimination. Especilly, the number repeats of these 577 isolates in MIRU24 are all 1, suggesting that these isolates are all modern M. tuberculosis strains.By direct comparison with previously typed MIRU-VNTR patterns and by genetic relationship analyses, we could identify and clearly define four epidemic groups of M. tuberculosis:Beijing family, T family, CAS family and LAM family. The family designation of most strains corresponded to the results as confirmed by Spoligotyping. However,1 isolates with a Spoligotyping profile corresponding to MANU2 was clustered with Beijing strains. This confusing finding was due to mixed infection by PCR analysis.In conclusion, the M. tuberculosis isolates from Tibetan patients in Tibet in this study exhibit low genetic polymorphism. Beijing family strains predominant and spread rapidly in Tibet. Majority of VNTR loci in this 24-locus MIRU-VNTR method is unfit for TB genityping in areas where Beijing family strains are dominant. If appropriate loci are added to the MIRU-VNTR analysis, MIRU-VNTR genotyping could be a valuable tool for strain typing and epidemiological research of M. tuberculosis. With this approach a more clear understanding about genetic pattern of M. tuberculosis can be achieved.更多还原