Integration of a transposon Tn1-encoded inhibitor-resistant beta-lactamase gene, bla(TEM-67) from Proteus mirabilis, into the Escherichia coli chromosome

Abstract

Proteus mirabilis NEL-1 was isolated from a urine sample of a patient hospitalized in a long-term care facility. Strain NEL-1 produced a beta-lactamase with a pI of 5.2 conferring resistance to amoxicillin and amoxicillin-clavulanic acid. Sequencing of a PCR amplicon by using TEM-specific primers revealed a novel bla(TEM) gene, bla(TEM-67). TEM-67 was an IRT-1-like TEM derivative related to TEM-65 (Lys39, Cys244) with an additional Leu21Ile amino acid substitution in the leader peptide. The biochemical properties of TEM-67 were equivalent to those described for TEM-65. Analysis of sequences surrounding bla(TEM-67) revealed that it was located on a transposon, Tn1, which itself was located on a 48-kb non-self-transferable plasmid, pANG-1. Electroporation of plasmid pANG-1 into Escherichia coli DH10B resulted in the integration of bla(TEM-67) into the chromosome, whereas it remained episomal in the P. mirabilis CIP103181 reference strain. Further characterization of pANG-1 revealed the presence of two identical sequences on both sides of Tn1 that contained an IS26 insertion sequence followed by a novel colicin gene, colZ, which had 20% amino acid identity with other colicin genes. The characterization of this novel TEM derivative provides further evidence for the large diversity of plasmid-encoded beta-lactamases produced in P. mirabilis and for their spread to other enterobacterial species through transposable-element-mediated events.

FIG. 1.

Plasmid pANG-1 and recombinant plasmids pMZ-1 to -3. (A) Restriction analysis of the natural plasmid pANG-1 (panel I) and Southern hybridization of that gel with a bla_TEM intragenic probe (panel II). Lanes 1, undigested plasmid pANG-1; lanes 2, HincII-restricted plasmid; lanes 3, EcoRV-restricted plasmid; lanes 4, HindIII-restricted plasmid; lanes 5, HindIII-EcoRV-restricted plasmid; lanes 6, EcoRV-HincII-restricted plasmid. (B) Schematic map of plasmid pANG-1 encoding TEM-67 from P. mirabilis NEL-1. E and H, EcoRV and HindIII cleavage sites, respectively. The numbers within the bars representing the enzyme cleavage sites indicate the sizes of the fragments in kilobase pairs. EcoRV-HindIII double digestion is indicated with dotted lines (C) Schematic map of recombinant plasmids pMZ-1, pMZ-2, and pMZ-3. The thin line represents the cloned inserts from pANG-1, while the dotted lines indicate the pK19 or pPCRscriptCam vector sequence. The open boxes represent the genes, and the arrows indicate their translational orientation. (D) Structure of the colZ promoter region. The conserved regions (−35, −10, and +1) for RNA polymerase binding are represented, as is the ribosome binding site (RBS).

FIG. 2.

Analysis of TEM-67 purification by sodium dodecyl sulfate-polyacrylamide gel electrophoresis with Coomassie blue staining. Lane 1, crude β-lactamase extract; lane 2, protein extract after first Q-Sepharose column; lane 3, purified protein. Molecular size markers were run in lane M, and their sizes are indicated on the right.

FIG. 3.

Pulsed-field gel electrophoresis and Southern hybridization. (A) Macrorestriction profiles of P. mirabilis NEL-1, E. coli MZ1 to -9, and E. coli DH10B isolates obtained after digestion with XbaI and SfiI. (B) Southern hybridization of the macrorestricted DNA with an internal probe for bla_TEM-67. Lanes P, P. mirabilis NEL-1; lanes 1 to 9, E. coli MZ1 to -9, respectively; lanes E, E. coli DH10B. Molecular size markers were run in lanes M, and their sizes are indicated on the left.

FIG.4.

Southern hybridization, with an internal probe for bla_TEM-67, of HindIII-restricted whole cell DNAs of P. mirabilis MZ1 to -3 (lanes 1 to 3, respectively), P. mirabilis CIP103181 (lane 4), P. mirabilis NEL-1 (lane 5), E. coli DH10B (lane 6), and E. coli MZ1 to -9 (lanes 7 to 15, respectively). Molecular size markers are indicated on the left.

FIG. 5.

Amino acid sequence alignment of the C-terminal end of ColZ with those of five known colicins: ColA, ColE1, ColK, and ColB from E. coli (23, 30, 34, 38) and ColU from Shigella boydii (39). Asterisks indicate conserved amino acids within the colicin proteins. Underlined amino acid sequence represent the 10 known alpha-helices as determined for ColA (27). Helices 8 and 9 are shown in grey.