Spontaneous excision of the O-polysaccharide wbkA glycosyltranferase gene is a cause of dissociation of smooth to rough Brucella colonies.

The brucellae are Gram-negative pathogens that cause brucellosis, a zoonosis of worldwide importance. The genus Brucella includes smooth and rough species that differ in that they carry smooth and rough lipopolysaccharides, respectively. Brucella abortus, B. melitensis, and B. suis are typical smooth species. However, these smooth brucellae dissociate into rough mutants devoid of the lipopolysaccharide O-polysaccharide, a major antigen and a virulence determinant encoded in regions wbo (included in genomic island-2) and wbk. We demonstrate here the occurrence of spontaneous recombination events in those three Brucella species leading to the deletion of a 5.5-kb fragment carrying the wbkA glycosyltranferase gene and to the appearance of rough mutants. Analysis of the recombination intermediates suggested homologous recombination between the ISBm1 insertion sequences flanking wbkA as the mechanism generating the deletion. Excision of wbkA was reduced but not abrogated in a recA-deficient mutant, showing the existence of both RecA-dependent and -independent processes. Although the involvement of the ISBm1 copies flanking wbkA suggested a transpositional event, the predicted transpositional joint could not be detected. This absence of detectable transposition was consistent with the presence of polymorphism in the inverted repeats of one of the ISBm1 copies. The spontaneous excision of wbkA represents a novel dissociation mechanism of smooth brucellae that adds to the previously described excision of genomic island-2. This ISBm1-mediated wbkA excision and the different %GC levels of the excised fragment and of other wbk genes suggest that the Brucella wbk locus is the result of at least two horizontal acquisition events.

European 1: a globally important clonal complex of Mycobacterium bovis.

We have identified a globally important clonal complex of Mycobacterium bovis by deletion analysis of over one thousand strains from over 30 countries. We initially show that over 99% of the strains of M. bovis, the cause of bovine tuberculosis, isolated from cattle in the Republic of Ireland and the UK are closely related and are members of a single clonal complex marked by the deletion of chromosomal region RDEu1 and we named this clonal complex European 1 (Eu1). Eu1 strains were present at less than 14% of French, Portuguese and Spanish isolates of M. bovis but are rare in other mainland European countries and Iran. However, strains of the Eu1 clonal complex were found at high frequency in former trading partners of the UK (USA, South Africa, New Zealand, Australia and Canada). The Americas, with the exception of Brazil, are dominated by the Eu1 clonal complex which was at high frequency in Argentina, Chile, Ecuador and Mexico as well as North America. Eu1 was rare or absent in the African countries surveyed except South Africa. A small sample of strains from Taiwan were non-Eu1 but, surprisingly, isolates from Korea and Kazakhstan were members of the Eu1 clonal complex. The simplest explanation for much of the current distribution of the Eu1 clonal complex is that it was spread in infected cattle, such as Herefords, from the UK to former trading partners, although there is evidence of secondary dispersion since. This is the first identification of a globally dispersed clonal complex M. bovis and indicates that much of the current global distribution of this important veterinary pathogen has resulted from relatively recent International trade in cattle.

Genomic island 2 is an unstable genetic element contributing to Brucella lipopolysaccharide spontaneous smooth-to-rough dissociation.

Brucella is a Gram-negative bacterium that causes a worldwide-distributed zoonosis. The genus includes smooth (S) and rough (R) species that differ in the presence or absence, respectively, of the O-polysaccharide of lipopolysaccharide. In S brucellae, the O-polysaccharide is a critical diagnostic antigen and a virulence determinant. However, S brucellae spontaneously dissociate into R forms, a problem in antigen and S vaccine production. Spontaneous R mutants of Brucella abortus, Brucella melitensis, and Brucella suis carried the chromosomal scar corresponding to genomic island 2 (GI-2) excision, an event causing the loss of the wboA and wboB O-polysaccharide genes, and the predicted excised circular intermediate was identified in B. abortus, B. melitensis, and B. suis cultures. Moreover, disruption of a putative phage integrase gene in B. abortus GI-2 caused a reduction in O-polysaccharide loss rates under conditions promoting S-R dissociation. However, spontaneous R mutants not carrying the GI-2 scar were also detected. These results demonstrate that the phage integrase-related GI-2 excision is a cause of S-R brucella dissociation and that other undescribed mechanisms must also be involved. In the R Brucella species, previous works have shown that Brucella ovis but not Brucella canis lacks GI-2, and a chromosomal scar identical to those in R mutants was observed. These results suggest that the phage integrase-promoted GI-2 excision played a role in B. ovis speciation and are consistent with other evidence, suggesting that this species and B. canis have emerged as two independent lineages.