The Molecular Epidemiology Of Tuberculosis Based On Variable Number Tandem Repeat Typing

Objectives: To study the polymophism of Variable Number Tandem Repeats (VNTR)at 15 loci of Mycobacterium tuberculosis (MTB) isolated from Tianjin and theapplication of VNTR genotyping in the study of molecular epidemiology oftuberculosis.Methods: 100 isolates of MTB were selected from Tianjin Tuberculosis ControlCenters. The clinical isolates of MTB used in this study were obtained from June2005 to April 2006. The VNTR loci were amplified by PCR which used extractedgenomic DNA of MTB isolates as templates. The number of copies of tandem repeatsof twelve MIRU loci and five ETR loci was determined by Quantity One software.PAUP software,beta 4.0 version,was used to perform phylogenetic analysis byUPGMA. TreeView was used to print out the dendrogram generated by PAUPanalysis. Some PCR products were subjected to DNA sequencing.The discriminatorypower of VNTR typing of MIRU and ETR was assessed by Hunter-Gaston index(HGDI) and The allelic diversity of each VNTR locus was calculated. Questionnairewas designed base on genotyping results and conventional epidemiological data fromclusted patients. SPSS version 11.5 was used for univariate analysis.Results:1. Basic demographic description: Of the 100 patients, 66％were male.The medianage at presentation was 36 years(range from 15 to 82 years).81 of the 100 patientswere living in urban area of Tianjin. 19％had a past history of TB. Sixty-eightpatients had a scar present.2. The most common ETR-VNTR patterns were 42435, which showed thecharacteristics of Beijing family.3. VNTR typing using 5 ETR loci yielded 23 distinct patterns, 17 of which wereunique and 9 of which were clustered. The biggest size of clusters contained 58isolates. MIRU typing discriminated 45 distinct patterns in 100 MTB isolates, 64 isolats fell into 9 clusters and 36 of which were unique. The biggest size of clustercontained 34 isolates. The maximal discrimination was apparently achieved bycombining ETR and MIRU, resulting in 52 distinct patterns, 59 isolates were groupedinto 11 clusters, 41 isolates presented unique pattern. The biggest size of clusters had29 isolates. Univariate analysis showed that there are no significance between patientsin clusters and nonclusted.4. MIRU-26 and 31 were highly discriminative (h-0.54 and 0.45), MIRU-10, 40 andETR-A were moderately discriminative (0.3＜h＜0.4), and loci 39,27,4,16,23, 2,20, 24,ETR-B, ETR-C were poorly discriminative (h＜0.3). Calculation of the discriminativeindex was performed using the HGDI, the discriminative power of ETR, MIRU andthe combination of these two panels are 0.650,0.862,and 0.909 respectitively.5. All strains were divided into three main groups, groupsⅠ,Ⅱ,Ⅲ. 89 isolatesbelonged toⅢgroup, 6 and 5 isolates went into groupⅠandⅡrespectively.6. Epidemiological link through contact tracing was investigated. We found that thereare two classmates in the same cluster.Conclusions: Our study showed that the MIRU-10, 26, 31, 40 and ETR-A displayedhigh polyporphism among the Tianjin MTB isolates, Others showed no polyporphism.The abilities of MIRU and ETR loci for discriminiating Beijing family of MTB werelimited and more VNTR loci with high polyporphism should be used. Our results alsosuggested that MTB Beijing family is prevalent in Tianjin areas. Genotyping based onVNTR is a very promising tool for studying the molecular epidemiology of MTB.