Bacterial artificial chromosome-based comparative genomic analysis identifies Mycobacterium microti as a natural ESAT-6 deletion mutant

Abstract

Mycobacterium microti is a member of the Mycobacterium tuberculosis complex that causes tuberculosis in voles. Most strains of M. microti are harmless for humans, and some have been successfully used as live tuberculosis vaccines. In an attempt to identify putative virulence factors of the tubercle bacilli, genes that are absent from the avirulent M. microti but present in human pathogen M. tuberculosis or Mycobacterium bovis were searched for. A minimal set of 50 bacterial artificial chromosome (BAC) clones that covers almost all of the genome of M. microti OV254 was constructed, and individual BACs were compared to the corresponding BACs from M. bovis AF2122/97 and M. tuberculosis H37Rv. Comparison of pulsed-field gel-separated DNA digests of BAC clones led to the identification of 10 regions of difference (RD) between M. microti OV254 and M. tuberculosis. A 14-kb chromosomal region (RD1(mic)) that partly overlaps the RD1 deletion in the BCG vaccine strain was missing from the genomes of all nine tested M. microti strains. This region covers 13 genes, Rv3864 to Rv3876, in M. tuberculosis, including those encoding the potent ESAT-6 and CFP-10 antigens. In contrast, RD5(mic), a region that contains three phospholipase C genes (plcA to -C), was missing from only the vole isolates and was present in M. microti strains isolated from humans. Apart from RD1(mic) and RD5(mic) other M. microti-specific deleted regions have been identified (MiD1 to MiD3). Deletion of MiD1 has removed parts of the direct repeat region in M. microti and thus contributes to the characteristic spoligotype of M. microti strains.

FIG. 1.

M. microti strain OV254 BAC map (MiXXX) overlaid on those of M. tuberculosis H37Rv (RvXXX) and M. bovis AF2122/97 (MbXXX). The scale bars indicate the positions on the M. tuberculosis genome.

FIG. 2.

Differences in the region comprising kb 4340 to 4360 between the deletions in BCG RD1^bcg (A) and M. microti RD1^mic (C) relative to M. tuberculosis H37Rv (B).

FIG. 3.

Difference in the region comprising kb 3121 to 3127 between M. tuberculosis H37Rv (A) and M. microti OV254 (B). Gray boxes, DRs; black boxes, unique numbered spacer sequences; ∗, spacer sequence identical to that of spacer 58 reported by van Embden et al. (43). Note that spacers 33 to 36 and 20 to 22 are not shown because H37Rv lacks these spacers.

FIG. 4.

(A) IS6110 RFLP after PvuII digestion from various M. microti strains. (B) Spoligotypes of various M. microti and control strains.

FIG. 4.

(A) IS6110 RFLP after PvuII digestion from various M. microti strains. (B) Spoligotypes of various M. microti and control strains.

FIG. 5.

(A) AseI restriction fragment PFGE profiles of various M. microti strains; (B) hybridization with a radiolabeled esat-6 probe; (C) hybridization with a probe of the RD1^mic flanking region; (D) hybridization with a plcA probe. Lanes: 1 (second lane from left), M. bovis AF2122/97; 2, M. canetti; 3, M. bovis BCG Pasteur; 4, M. tuberculosis H37Rv; 5, M. microti OV254; 6, M. microti Myc 94-2272; 7, M. microti B3 type vole; 8, M. microti B4 type vole; 9, M. microti B2 type llama; 10, M. microti B1 type llama; 11, M. microti ATCC 35782; M, low-range PFGE marker (NEB).