Molecular and biochemical heterogeneity of class B carbapenem-hydrolyzing beta-lactamases in Chryseobacterium meningosepticum

Abstract

Although the carbapenem-hydrolyzing beta-lactamase (CHbetaL) BlaB-1 is known to be in Chryseobacterium meningosepticum NCTC 10585, a second CHbetaL gene, bla(GOB-1), was cloned from another C. meningosepticum clinical isolate (PINT). The G+C content of bla(GOB-1) (36%) indicated the likely chromosomal origin of this gene. Its expression in Escherichia coli DH10B yields a mature CHbetaL with a pI of 8.7 and a relative molecular mass of 28.2 kDa. In E. coli, GOB-1 conferred resistance to narrow-spectrum cephalosporins and reduced susceptibility to ureidopenicillins, broad-spectrum cephalosporins, and carbapenems. GOB-1 had a broad-spectrum hydrolysis profile including penicillins and cephalosporins (but not aztreonam). The catalytic efficiency for meropenem was higher than for imipenem. GOB-1 had low amino acid identity with the class B CHbetaLs, sharing 18% with the closest, L-1 from Stenotrophomonas maltophilia, and only 11% with BlaB-1. Most of the conserved amino acids that may be involved in the active site of CHbetaLs (His-101, Asp-103, His-162, and His-225) were identified in GOB-1. Sequence heterogeneity was found for GOB-1-like and BlaB-1-like beta-lactamases, having 90 to 100% and 86 to 100% amino acid identity, respectively, among 10 unrelated C. meningosepticum isolates. Each isolate had a GOB-1-like and a BlaB-1-like gene. The same combination of GOB-1-like and BlaB-1-like beta-lactamases was not found in two different isolates. C. meningosepticum is a bacterial species with two types of unrelated chromosome-borne class B CHbetaLs that can be expressed in E. coli and, thus, may represent a clinical threat if spread in gram-negative aerobes.

FIG. 1

Nucleotide sequence of a 2,384-bp DNA fragment of recombinant plasmid pBS2 carrying bla_GOB-1 and the 117 bp of the 3′ end of the endo-beta-N-acetylglucosaminidase gene of C. meningosepticum PINT. The deduced amino acid sequences are given in a single-letter code. The start and stop codons of the bla_GOB-1 gene and the stop codon of the endo-beta-N-acetylglucosaminidase gene are in bold. The vertical arrow indicates the peptide leader cleavage site in E. coli as determined by N-terminal sequencing. The putative −35 and −10 sequences of the putative promoter and ribosome binding site (RBS) for bla_GOB-1 are underlined. Primers 1, 2, and 3 used to PCR amplify bla_GOB-1-like genes from other C. meningosepticum isolates are indicated by an arrow.

FIG. 2

Multiple-sequence alignment of amino acid sequence of GOB-1 from C. meningosepticum PINT isolate with those of eight class B CHβLs. Sequence comparison was performed first by aligning the proteins by using the ClustalW program. Then, adjustments were made to reduce the number of gaps and to maintain alignment of the putative active residues of the active sites. The origins of metallo-β-lactamases are as follows: CphA-1 from A. hydrophila AE036 (31), BII from B. cereus 5/B/6 (29), BlaB (BlaB-1) from C. meningosepticum CIP 6058 (NCTC 10585) (46), IND-1 from C. indologenes (5), VIM-1 from P. aeruginosa VR-143/97 (28), CcrA from B. fragilis TAL 3636 (44), IMP-1 from S. marcescens TN9106 (36), and L-1 from S. maltophilia IID1275 (57). Amino acids that were identical for at least five out of nine aligned amino acid sequences are shaded in grey. Stars refer to conserved amino acids identified by crystal structure determination as interacting in the binding to the Zn²⁺ cofactor or to the water molecule in the B. cereus 569H/9 enzyme or in CcrA (12, 20). The numbering scheme refers to the CcrA enzyme (44). Dashes indicate gaps introduced to optimize the alignment.

FIG. 3

Dendrogram obtained for nine representative CHβLs calculated with ClustalW followed by adjustments to reduce the number of gaps and to maintain alignment of the residues identified in the active sites of some CHβLs. Branch lengths are to scale and proportional to the number of amino acid changes. The percentages at the branching point (bold and underlined) refer to the number of times a particular node was found in 100 bootstrap replications (the stars indicate uncertainty of nodes with bootstrap values of less than 50%). The distance along the vertical axis has no significance. BlaB-1 (BlaB) and GOB-1 were from C. meningosepticum, IND-1 was from C. indologenes, CphA-1 was from A. hydrophila, L-1 was from S. maltophilia, BII was from B. cereus, VIM-1 and VIM-2 were from P. aeruginosa, CcrA was from B. fragilis, and IMP-1 was from S. marcescens. Percent amino acid identities to GOB-1 are indicated in parentheses.

FIG. 4

Amino acid comparison of the GOB-1-like β-lactamases from nine C. meningosepticum isolates. Dashes indicate identical amino acids, and dots indicate undetermined sequences. GOB-1 was from C. meningosepticum PINT, CIP 6057, and AB1572, GOB-2 was from C. meningosepticum HO1J100, GOB-3 was from C. meningosepticum CIP 6059, GOB-4 was from C. meningosepticum GEO, GOB-5 was from C. meningosepticum CIP 6058, GOB-6 was from C. meningosepticum AMA, and GOB-7 was from C. meningosepticum CIP 79.5. Numbering is according to the GOB-1 sequence.

FIG. 5

Amino acid comparison of the BlaB-1-like β-lactamases from 10 C. meningosepticum isolates. Dashes indicate identical amino acids. BlaB-1 was from C. meningosepticum CIP 6058 and PINT, BlaB-2 was from C. meningosepticum AMA, BlaB-3 was from C. meningosepticum GEO and CIP 6059, BlaB-4 was from C. meningosepticum CIP 6057, BlaB-5 was from C. meningosepticum CIP 79.5, BlaB-6 was from C. meningosepticum CIP 7830, BlaB-7 was from C. meningosepticum AB1572, and BlaB-8 was from C. meningosepticum H01J100. Numbering is according to the BlaB-1 sequence.