Reply to “Genetic Contexts of bla_NDM-1”

REPLY

We read with interest the discussion of the genetic context of bla_NDM-1 by Sally Partridge and John Iredell in the foregoing letter. Our initial paper (4) describing NDM-1 as a chimera sought only to understand the origin of the bla_NDM-1 gene and offer some explanations of the success of this gene fusion and was not meant to describe further genetic changes subsequent to the gene fusion event. However, since the claim is that the gene fusion is more complicated than was described in our recent paper, we would like to comment on the aforementioned letter.

First, Partridge and Iredell suggest that bla_NDM-1 has been constructed by “fusion to the aphA6/ISAba125 promoter region in Acinetobacter.” This is not correct, since the first 20 bp of the structural genes of both bla_NDM-1 and aphA6 are identical. This means that the fusion has not happened in the promoter region but actually within the aphA6 structural gene itself, with the first 19 to 20 bp of the aphA6 gene being fused onto the 5′ end of a previous metallo-β-lactamase (MBL) gene, forming bla_NDM-1.

Second, Partridge and Iredell suggest the added complication of the bla_NDM-1 environment found in the Acinetobacter lwoffii plasmid pNDM-BJ01 (1) being derived by transposition. This is claimed because it appears to be part of an ISAba125-mediated composite transposon that has inserted downstream from a copy of aphA6 in this plasmid, and the evidence for this is a direct 3-bp repeat (GTT) found at the left-hand and right-hand ends of the upstream and downstream copies, respectively, of ISAba125. This may be the case, but it is clear that the construction of bla_NDM-1 was a prior event and not linked to this putative subsequent transposition event.

The upstream complete copy of the aphA6 gene found in the Acinetobacter lwoffii plasmids pNDM-BJ01/2 and also in the incAC plasmids isolated in Canada (JF503991, JF714412) is unusual in that it is found adjacent to an as-yet-unnamed insertion sequence rather than in an ISAba125 composite transposon. However, it is clear that bla_NDM-1 is formed by fusion with aphA6 when found as part of an ISAba125 composite transposon (4). Therefore, we can suggest the following sequence of events.

(i) bla_NDM-1 is formed by the fusion of a prior MBL gene with an aphA6/ISAba125 composite transposon.

(ii) A second copy of ISAba125 transposes by chance downstream of NDM-1, bracketing the bla_NDM-1 locus and forming a new composite ISAba125 transposon (Tn125) (3) which can now move by transposition and is found on several Acinetobacter baumannii chromosomes (2, 3) and has also likely led to the structures found in pNDM-BJ01/2 by transposition adjacent to aphA6.

(iii) New insertion sequences (IS903, ISEc33, IS1, ISKpn14, etc.) transpose into positions in front of bla_NDM-1 and truncate the upstream ISAba125 element. These new common sequences found upstream of bla_NDM-1, along with other common sequences associated with bla_NDM-1, now give the gene access to many new plasmid and chromosome positions by homologous recombination based on repeated sequences. Thus, the rapid emergence of NDM-1 is a combination of movement by transposition as well as movement by homologous recombination.