Novel methicillin resistance gene mecD in clinical Macrococcus caseolyticus strains from bovine and canine sources

Abstract

Methicillin-resistant Macrococcus caseolyticus strains from bovine and canine origins were found to carry a novel mecD gene conferring resistance to all classes of β-lactams including anti-MRSA cephalosporins. Association of β-lactam resistance with mecD was demonstrated by gene expression in S. aureus and deletion of the mecD-containing island in M. caseolyticus. The mecD gene was located either on an 18,134-bp M. caseolyticus resistance island (McRI_mecD-1) or a 16,188-bp McRI_mecD-2. Both islands were integrated at the 3' end of the rpsI gene, carried the mecD operon (mecD-mecR1_m-mecI_m), and genes for an integrase of the tyrosine recombinase family and a putative virulence-associated protein (virE). Apart from the mecD operon, that shared 66% overall nucleotide identity with the mecB operon, McRI_mecD islands were unrelated to any mecB-carrying elements or staphylococcal cassette chromosome mec. Only McRI_mecD-1 that is delimitated at both ends by direct repeats was capable of circular excision. The recombined excision pattern suggests site-specific activity of the integrase and allowed identification of a putative core attachment site. Detection of rpsI-associated integrases in Bacillus and S. aureus reveals a potential for broad-host range dissemination of the novel methicillin resistance gene mecD.

Figure 1. Phylogenetic tree of mec genes encoding PBP2a.

Evolutionary analysis was performed for nucleotide sequences using the UPGMA method in MEGA7. The percentage of nucleotide (nt) and amino acid (aa) identity between mecD and other mec genes was determined by sequence alignment with Clustral OMEGA [ http://www.ebi.ac.uk/Tools/msa/clustalo/]. *) mecC1 of S. xylosus S04009 does not encode a functional PBP2a due to a frameshift mutation close to the 5′ end of the gene.

Figure 2. Divergent promoters and operator sequence in the intergenic region between mecR1_m and mecD genes of M. caseolyticus.

The −10 and −35 promoter sequences are underlined. Start codons and ribosomal binding sites (RBS) are in bold type. The operator sequence is highlighted in green and inverted repeats marked by arrows. The consensus sequence [A/G]NATTACA[A/T]NTGTA [A/G][T/G]NT (with bases acceptable for one given position between square brackets and “N” for any base) was used to identity operator sequence recognized by MecI/BlaI repressors. The sequence is shown for M. caseolyticus IMD0819 (Genbank acc. no KY013611, position 15264–15440).

Figure 3. Structures of McRI_mecD-1 and McRI_mecD-2 and flanking sequences.

Comparison was performed with sequences of M. caseolyticus strains JCSC5402 (Genbank acc. no. NC_011999, position 215180–250630), IMD0473 (Genbank acc. no KY013610), IMD0819 (Genbank acc. no KY013611), KM0211 (Genbank acc. no KY013612) and KM1352 (Genbank acc. no KY013613) using Easyfig software. Gray areas indicate regions with between 68% to 100% nucleotide sequence identity. Regions encompassing McRI_mecD-1, McRI_mecD-2 and McCI_IMD0819 are indicated by horizontal black lines and flanking direct repeats (DRs) by vertical blue lines. The open reading frames (orfs) are represented by arrows: mecD, mecR1_m and mecI_m are shown in red, orfs encoding integrase (int) or transposase (tnp) are shown in yellow, reverse transciptases (rt) in beige, orfs associated with restriction-modification in green and virulence-associated orfs in mauve; the orfs occurring in all M. caseolyticus strains are shown in black and additional strain-specific orfs in blue. The primers used in this study are indicated by small black arrowheads (a), mecD-F; (b), mecD-R; (c), truA-F; (d), s66-R; (e), int-0819-F; (f), int-0473-F; (g), araC-F; (h), orf21-R; (i), orf20-F; (j), cop-R) and cleavage sites for restriction endonucleases HindIII (H3) and HincII (H2) and EcoRI (E1) by thin vertical lines.

Figure 4

(a) Phylogenetic tree of integrases of the tyrosine recombinase family. Colored boxes group members that share homologous integration sites (rpsI, gene for 30S ribosomal protein S9; sodm, superoxide dismutase gene; bcat, gene for branched-chain amino acid aminotransferase; guaA, GMP synthetase gene; groEL, chaperonin gene; rpsR, gene for 30S ribosomal protein S18; nlpA, gene for component of ABC-type metal ion transport system; smrB, gene for SsrA-binding protein). Analysis was performed for amino acid (aa) sequences using the UPGMA method in MEGA7. Host strains and genetic elements if known are indicated. References for the used sequences can be found in Supplementary Table S3. (b) Putative core attachment sites recognized by rpsI-associated integrases. For M. caseolyticus strains, the imperfect direct repeats (DR) carrying the att consensus sequence are indicated. For all other species the att site found at the 3′ end of the rpsI gene is given. The rpsI stop codon is underlined. Positions that hold variant bases are unshaded. All species, except S. aureus N315 and M. caseolyticus KM1352, carry an integrase gene downstream of rpsI.