Use of a Mycobacterium tuberculosis H37Rv bacterial artificial chromosome library for genome mapping, sequencing, and comparative genomics

Abstract

The bacterial artificial chromosome (BAC) cloning system is capable of stably propagating large, complex DNA inserts in Escherichia coli. As part of the Mycobacterium tuberculosis H37Rv genome sequencing project, a BAC library was constructed in the pBeloBAC11 vector and used for genome mapping, confirmation of sequence assembly, and sequencing. The library contains about 5,000 BAC clones, with inserts ranging in size from 25 to 104 kb, representing theoretically a 70-fold coverage of the M. tuberculosis genome (4.4 Mb). A total of 840 sequences from the T7 and SP6 termini of 420 BACs were determined and compared to those of a partial genomic database. These sequences showed excellent correlation between the estimated sizes and positions of the BAC clones and the sizes and positions of previously sequenced cosmids and the resulting contigs. Many BAC clones represent linking clones between sequenced cosmids, allowing full coverage of the H37Rv chromosome, and they are now being shotgun sequenced in the framework of the H37Rv sequencing project. Also, no chimeric, deleted, or rearranged BAC clones were detected, which was of major importance for the correct mapping and assembly of the H37Rv sequence. The minimal overlapping set contains 68 unique BAC clones and spans the whole H37Rv chromosome with the exception of a single gap of approximately 150 kb. As a postgenomic application, the canonical BAC set was used in a comparative study to reveal chromosomal polymorphisms between M. tuberculosis, M. bovis, and M. bovis BCG Pasteur, and a novel 12.7-kb segment present in M. tuberculosis but absent from M. bovis and M. bovis BCG was characterized. This region contains a set of genes whose products show low similarity to proteins involved in polysaccharide biosynthesis. The H37Rv BAC library therefore provides us with a powerful tool both for the generation and confirmation of sequence data as well as for comparative genomics and other postgenomic applications. It represents a major resource for present and future M. tuberculosis research projects.

FIG. 1

PCR screening for unique BAC clones with specific primers for two selected genomic regions of the H37Rv chromosome, using 21 pools representating 2,016 BACs (A) and sets of 20 subpools from selected positive pools (B).

FIG. 2

PFGE gel of DraI-cleaved BAC clones used for estimating the insert sizes of BACs.

FIG. 3

Minimal overlapping BAC map of M. tuberculosis H37Rv superimposed on the integrated physical and genetic map established by Philipp et al. (18). Y and I numbers show pYUB328 (2) and pYUB412 (16) cosmids which were shotgun sequenced during the H37Rv genome sequencing project. Y cosmids marked with asterisks were shown in the integrated physical and genetic map (18). Rv numbers show the positions of representative BAC clones relative to sequenced Y and I clones. Rv numbers in boxes represent BACs which were shotgun sequenced at the Sanger Centre.

FIG. 4

Ethidium bromide-stained gel (A) and corresponding Southern blot (B) of EcoRI- and PvuII-digested Rv58 DNA hybridized with ³²P-labeled genomic DNA preparations from M. tuberculosis H37Rv, M. bovis ATCC 19210, and M. bovis BCG Pasteur.

FIG. 5

Organization of the ORFs in the 12.7-kb genomic region present in M. tuberculosis H37Rv but not present in M. bovis ATCC 19210 and M. bovis BCG Pasteur. Arrows show the direction of transcription of the putative genes. Positions of EcoRI and PvuII restriction sites are shown. Vertical dashes represent stop codons. The 11 ORFs correspond to the ORFs MTCY277.28 to MTCY277.38 (accession no. Z79701 in the EMBL nucleotide sequence data library). The junction sequences flanking the polymorphic region are shown.