Lactamase from Erwinia persicina - Antimicrobial Agents and ...

4 downloads 345 Views 120KB Size Report
Feb 4, 2002 - encoded a clavulanic-acid-inhibited Ambler class A ß-lactamase, .... Numbering is according to Ambler et al. ..... ratory Standards, Wayne, Pa.
ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Nov. 2002, p. 3401–3405 0066-4804/02/$04.00⫹0 DOI: 10.1128/AAC.46.11.3401–3405.2002 Copyright © 2002, American Society for Microbiology. All Rights Reserved.

Vol. 46, No. 11

Identification of a Chromosome-Borne Expanded-Spectrum Class A ␤-Lactamase from Erwinia persicina Sophie Vimont, Laurent Poirel, Thierry Naas, and Patrice Nordmann* Service de Bacte´riologie-Virologie, Ho ˆpital de Biceˆtre, Assistance Publique/Ho ˆpitaux de Paris, Faculte´ de Me´decine Paris-Sud, 94275 Le Kremlin-Biceˆtre, France Received 4 February 2002/Returned for modification 20 May 2002/Accepted 9 August 2002

From whole-cell DNA of an enterobacterial Erwinia persicina reference strain that displayed a penicillinaserelated antibiotic-resistant phenotype, a ␤-lactamase gene was cloned and expressed in Escherichia coli. It encoded a clavulanic-acid-inhibited Ambler class A ␤-lactamase, ERP-1, with a pI value of 8.1 and a relative molecular mass of ca. 28 kDa. ERP-1 shared 45 to 50% amino acid identity with the most closely related enzymes, the chromosomally encoded enzymes from Citrobacter koseri, Kluyvera ascorbata, Kluyvera cryocrescens, Klebsiella oxytoca, Proteus vulgaris, Proteus penneri, Rahnella aquatilis, Serratia fonticola, Yersinia enterocolitica, and the plasmid-mediated enzymes CTX-M-8 and CTX-M-9. The substrate profile of the noninducible ERP-1 was similar to that of these ␤-lactamases. ERP-1 is the first extended-spectrum ␤-lactamase from an enterobacterial species that is plant associated and plant pathogenic. ated extended-spectrum ␤-lactamases (ESBLs) of the CTX-M group remains unknown, whereas they are spreading worldwide. The origin of CTX-M-2 has been identified as the chromosomally encoded ␤-lactamase from Kluyvera ascorbata (C. Humeniuk et al., unpublished data [GenBank accession no. CAB59824]), whereas that of CTX-M-8 is the chromosomally encoded ␤-lactamase from K. georgiana (L. Poirel, P. Ka¨mpfer, and P. Nordmann, unpublished data). We report here the characterization of a ␤-lactamase from E. persicina (representative of Erwinia species stricto sensu) that was found to be ␤-lactamase positive according to results of preliminary nitrocefin testing. Its substrate profile included some extended-spectrum cephalosporins that could not have been suspected on the sole analysis of the ␤-lactam resistance phenotype of the strain. Sequence analysis and kinetic parameters revealed that this plant-associated pathogen produced a class A ESBL that is distantly related to known ␤-lactamases.

The genus Erwinia is classified in the Enterobacteriaceae. Most members of this genus characteristically cause diseases of plants, vegetables, and fruits (31). The phylogenetic position of the genus Erwinia, along with other members of the Enterobacteriaceae associated with plants, have been explored by using 16S ribosomal DNA sequence comparisons (10, 11). Four phylogenetic groups, each representing a branch of a lineage, have been proposed (31). One group comprises Erwinia persicina (E. persicinus), E. amylovora, E. rhapontici, E. psidii, E. pyrifoliae, E. mallotivora, and E. tracheiphila. The second group consists of E. cacticida, E. carotovora subspecies, E. chrysanthemi, and E. cypripedii. This latter group was moved to the genus Pectobacterium primarily because it produces pectolytic enzymes for pathogenesis (21). The third group is comprised of E. alni, E. nigrifluens, E. rubrifaciens, E. paradisiaca, E. quercicina, and E. salicis. It has been suggested that this group be moved to the genus Brenneria. A fourth group includes members that have been reclassified into the genus Pantoea (21). Detailed antibiotic resistance patterns of Erwinia species is known only for seven E. persicina strains (23). Reports of Erwinia isolates as human pathogens are limited since most of the reported Erwinia isolates in clinical microbiology have been now assigned to the Pantoea group. The phytopathogen E. persicina (11) has been isolated from human urinary tract infections (23). E. chrysanthemi may cause death of human gastrointestinal cells in culture and expresses virulence determinants similar to those of other well-known human enterobacterial pathogens (9). We have searched for ␤-lactamase producers among Erwinia reference strains for two main reasons. First, some Erwinia species, like Streptomyces sp., have been reported to produce carbapenems and thus could concomitantly produce ␤-lactamases (20). Second, the origin of many class A plasmid-medi-

MATERIALS AND METHODS Bacterial strains. Representative reference strains of three groups of Erwinia sp. were used: E. persicina 105199T, E. mallotivora 105167T, E. psidii 105200T, E. rhapontici 105202T, E. amylovora 8282T, Pectobacterium cactidica 105191T, Brenneria alni 104916T, B. quercina 105201T, and B. rubrifaciens 105203T. These strains were from the Institut Pasteur (Paris, France) strain collection. They were tested for ␤-lactamase production with nitrocefin-containing disks (bioMe´rieux, ´ toile, France). Escherichia coli DH10B (Life Technologies, Eragny, Marcy l’E France) was used as the recipient strain in electroporation and protein expression experiments (27). Antimicrobial agents and MIC determinations. The antimicrobial agents were obtained in the form of standard laboratory powders and were used immediately after their solubilization. The agents and their sources have been described elsewhere (27). Antibiotic disks (Sanofi Diagnostics Pasteur, Marnes-la-Coquette, France) were used for routine antibiograms (www.sfm.asso.fr). MICs were determined by an agar dilution technique on Mueller-Hinton agar (Sanofi Diagnostics Pasteur) with an inoculum of 104 CFU per spot (27). The plates were incubated at 35°C for 18 h. The MICs of ␤-lactams were determined alone or in combination with a fixed concentration of clavulanic acid (2 ␮g/ml) or tazobactam (4 ␮g/ml). Interpretative criteria were those of the NCCLS (22). Cloning experiments and analysis of recombinant plasmids. Whole-cell DNA of E. persicina 105199T was extracted as described previously (8). All enzymes used in cloning experiments were from Amersham Pharmacia Biotech (Orsay, France). Sau3AI-restricted whole-cell DNA of E. persicina 105199T was ligated

* Corresponding author. Mailing address: Service de Bacte´riologieVirologie, Ho ˆpital de Biceˆtre, 78 rue du Ge´ne´ral Leclerc, 94275 Le Kremlin-Biceˆtre Cedex, France. Phone: 33-1-45-21-36-32. Fax: 33-145-21-63-40. E-mail: [email protected]. 3401

3402

VIMONT ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

FIG. 1. Alignment of the ERP-1 amino acid sequence with those of the most closely related enzymes: CTX-M-9 from E. coli (28), CTX-M-8 from Enterobacter cloacae (5), KLUA-1 from K. ascorbata (GenBank accession no. CAB59824), and YENT from Yersinia enterocolitica (30). Numbering is according to Ambler et al. (1). Dashes represent identical amino acid residues. The vertical arrow is the cleavage site of the leader peptide of the mature ␤-lactamase ERP-1. Structural elements characteristic of class A ␤-lactamases and of serine ␤-lactamases are shaded. The amino acids of the omega loop are underlined.

into the BamHI-site of pBK-CMV phagemid (27). Recombinant plasmids were transformed by electroporation (Gene Pulser II; Bio-Rad, Ivry-sur-Seine, France) into E. coli DH10B electrocompetent cells (Life Technologies). Antibiotic-resistant colonies were selected onto Trypticase soy (TS) agar plates containing amoxicillin (50 ␮g/ml) and kanamycin (30 ␮g/ml). Recombinant plasmid DNA was obtained from 100 ml of TS broth cultures grown overnight in the presence of amoxicillin (100 ␮g/ml) at 37°C. Plasmid DNAs were extracted and purified with Qiagen plasmid DNA Maxi kit (Qiagen, Courtaboeuf, France). Plasmid content and hybridization experiments. Extraction of plasmid DNA from E. persicina 105199T was attempted as previously reported (4, 8). Southern hybridizations were performed as described previously (29) with whole-cell DNA of E. persicina 105199T and with BamHI-restricted whole-cell DNAs of the studied reference Erwinia and related bacteria. An enhanced chemiluminescence nonradioactive labeling and detection kit (Amersham Pharmacia Biotech, Orsay, France) was used (27) with a PCR-obtained 739-bp internal fragment of blaERP-1 as the probe (ERP-1A, 5⬘-ACACCACTGAACGTATTTGC-3⬘; ERP-1B, 5⬘-TG CTGGGT AAAATAGATGGC-3⬘). ␤-Lactamase purification and biochemical parameters. A culture of E. coli DH10B(pSV-1) was grown overnight at 37°C in 4 liters of TS broth containing amoxicillin (100 ␮g/ml). The ␤-lactamase extract was obtained after sonification, as described previously (26). It was dialyzed overnight against 20 mM Tris-HCl buffer (pH 8.0). It was loaded onto a preequilibrated Q-Sepharose column (Amersham Pharmacia Biotech) in 20 mM Tris-HCl buffer (pH 8.0). The ␤-lactamase activities, as determined qualitatively for each fraction by using nitrocefin

hydrolysis (Oxoid, Dardilly, France), were recovered in the flowthrough and dialyzed overnight against 50 mM sodium phosphate buffer (pH 7.0). The ␤-lactamase was then loaded onto an S-Sepharose column preequilibrated with the same buffer and eluted with a linear NaCl gradient (0 to 1 M). The fractions containing the highest ␤-lactamase activity were pooled and dialyzed overnight against 50 mM phosphate buffer (pH 7.0) prior to a 10-fold concentration (Vivaspin 10,000 MW; Sartorius, Go ¨ttingen, Germany). The protein content was measured by using the Bio-Rad DC protein assay, and the specific activities of the crude extract and of the purified ␤-lactamase were compared. The specific activities of crude ␤-lactamase extract and purified enzyme were determined as previously reported (27) with 100 ␮mol of cephalothin as the substrate. One unit of enzyme activity was defined as the activity that hydrolyzed 1 ␮mol of cephalothin per min. The purity of the enzymes was estimated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) (29). To determine the cleavage site of the mature protein of the identified ␤-lactamase, the purified enzyme was submitted to an Edman sequence analysis (12) (Laboratory for Protein Microsequencing, Institut Pasteur, Paris, France). Purified enzyme and marker proteins were subjected to SDS-PAGE (29). Proteins were then electrotransferred onto a polyvinylidene difluoride membrane (Immobilon-P; Millipore, Guyancourt, France) by using the Mini-Protean II transfer cell (8 by 7.3 cm; Bio-Rad) in 50 mM Tris-borate buffer (pH 8.7) at room temperature (3.5 V/cm, overnight). The membrane was then rinsed in distilled water and stained with a solution made of 0.1% Coomassie brillant blue R-250 in methanol and water (50:40 [vol/vol]). The protein band was then excised with a razor blade and allowed to air dry. The amino-terminal sequence of the mature

␤-LACTAMASE FROM ERWINIA PERSICINA

VOL. 46, 2002

3403

TABLE 1. Percentage of amino acid identity of ␤-lactamase ERP-1 and the most closely related class A ␤-lactamasesa ␤-Lactamase

ERP-1 CTX-M8 CTX-M9 CDIV KLUA-1 KLUC-1 SHV-1 OXY-1 HUGA CUMA SFO-1 RAHN-1 YENT

% Amino acid identity ERP-1

CTX-M-8

CTX-M-9

CDIV

KLUA-1

KLUC-1

SHV-1

OXY-1

HUGA

CUMA

SFO-1

RAHN-1

YENT



50 –

50 84 –

49 71 69 –

49 83 80 67 –

48 76 79 68 77 –

39 37 37 34 36 35 –

48 73 71 70 74 72 35 –

45 61 59 65 63 62 34 64 –

45 61 59 64 63 62 34 64 94 –

46 72 72 68 75 74 35 73 66 65 –

49 72 69 64 74 71 34 69 62 61 75 –

50 54 55 53 57 56 34 55 51 51 54 54 –

a ERP-1, E. persicina (this study); CDIV, C. koseri (formerly C. diversus) (26); KLUA-1, K. ascorbata (GenBank accession no. 272038); KLUC-1, K. cryocrescens (8); Klebsiella pneumoniae (SHV-1 [3]; OXY-1, Klebsiella oxytoca (2); HUGA, P. penneri (17); CUMA, P. vulgaris (25); SFO-1, S. fonticola (24); RAHN-1, R. aquatilis (4); YENT, Y. enterocolitica (30). CTX-M-8 and CTX-M-9 are plasmid-encoded class A enzymes (5, 28). –, 100% identity.

␤-lactamases was determined with an automated Edman sequencer on a 473A model gas-phase sequencer (Applied Biosystems). Purified enzyme and ␤-lactamase extract from 100-ml culture of E. persicina were subjected to analytical isoelectric focusing, as previously described (27). The purified ␤-lactamase ERP-1 was used for kinetic measurements performed at 30°C in 100 mM sodium phosphate (pH 7.0). The rates of hydrolysis were determined with a spectrophotometer Ultrospec 2000 (Amersham Pharmacia Biotech). The wavelengths and absorption coefficients of ␤-lactams have been previously referenced (27). Km and kcat values were determined by analyzing the ␤-lactam hydrolysis under initial rate conditions by using the Eadie-Hoffstee linearization of the Michaelis-Menten equation. In the cases of low Km values, Ki values were determined with cephalothin as the substrate. Various concentrations of clavulanic acid, tazobactam, cefoxitin, and imipenem were preincubated with the enzyme for 3 min at 30°C before we tested the rate of cephalothin (100 ␮M) hydrolysis. The 50% inhibitory concentrations (IC50) of these inhibitors were determined as the concentration of inhibitor that inhibited hydrolytic activity by 50%. Induction studies were performed with cultures of E. persicina 105199T with 0.5 to 2 ␮g of cefoxitin per ml or 0.06 to 0.24 ␮g of imipenem per ml as ␤-lactam inducers (27) and 100 ␮mol of cephalothin as the substrate. DNA sequencing and protein analysis. Both strands of the cloned DNA fragment of recombinant plasmid pSV-1 were sequenced with an Applied Biosystems sequencer (ABI 377). The nucleotide sequence and the deduced protein sequence were analyzed with software available at the National Center for Biotechnology Information website (www. ncbi.nlm.nih.gov). Multiple protein sequence alignments were carried out with the program CLUSTALW (www .biomed.pasteur.fr). Nucleotide sequence accession number. The nucleotide sequence of blaERP-1 has been assigned to the GenBank nucleotide database under the accession number no. AY077733.

RESULTS AND DISCUSSION Cloning and sequence analysis of a ␤-lactamase gene from E. persicina. Five E. coli DH10B recombinant clones were obtained after selection on kanamycin- and amoxicillin-containing TS agar plates. A recombinant plasmid, pSV-1, that had the shortest insert was retained for further analysis. Part of its 3.9-kb DNA insert was sequenced an open reading frame (ORF) of 882 bp was identified among the sequenced 1,222 bp (data not shown). The G⫹C content of this ORF was 61.3%, a level slightly higher than the G⫹C ratio of Erwinia genes (50 to 58%). Within the deduced protein of this ORF (293 amino acids), designated ERP-1, characteristic elements of Ambler class A and serine ␤-lactamases were identified (Fig. 1) (1, 16).

␤-Lactamase ERP-1 shared only 50% amino acid identity with the most closely related plasmid-mediated ␤-lactamases, CTX-M-8 and CTX-M-9 (5, 28) and, similarly, low identity with chromosomally encoded class A enzymes (Table 1). As found for example in K. cryocrescens and Rahnella aquatilis (4, 8), no LysR-type regulator gene was identified in the nucleotide sequence upstream of blaERP-1, whereas regulatory genes are found upstream of the chromosomally encoded ␤-lactamase genes of Citrobacter koseri, P. penneri, P. vulgaris, and Serratia fonticola (14, 15, 17, 19). Plasmid detection from cultures of E. persicina 105199T failed. A positive hybridization signal was detected at the chromosomal DNA migration position of E. persicina 105199T, suggesting a chromosomal origin of blaERP-1 (data not shown). BamHI-restricted DNAs of Erwinia, Pectobacterium, and Brenneria strains were screened by the Southern technique with the same probe. No hybridization signal was obtained, thus excluding the presence of blaERP-1-like genes in strains that were nitrocefin negative, except for E. rhapontici 105202T. Susceptibility testing. E. persicina 105199T was resistant to amoxicillin and of intermediate susceptibility to ticarcillin, cephalothin, and cefsulodin (Table 2). It was susceptible to the other ␤-lactam antibiotics tested. Its overall susceptibility to ␤-lactams was higher than that reported for seven other E. persicina strains that showed a decreased susceptibility to piperacillin, cefuroxime, ceftriaxone, ceftazidime, and cefotaxime (23). The ␤-lactam resistance phenotype of E. persicina 105199T suggested the presence of a narrow-spectrum penicillinase rather than that of an ESBL. However, once cloned in pBK-CMV (pSV-1) and expressed in E. coli DH10B, ERP-1 conferred also resistance or intermediate susceptibility to piperacillin and cefuroxime. MICs of cefotaxime, ceftriaxone, cefpirome, ceftazidime, cefepime, and aztreonam were increased but still in the range of susceptibility (Table 2). Addition of clavulanic acid and tazobactam strongly lowered the MICs of ␤-lactams (Table 2). These results indicated that ERP-1 is an ESBL that weakly accounts for the ␤-lactam resistance profile observed in E. persicina strain. The discrepancy between the ␤-lactam resistance profiles observed in E. persicina 1051999T and in E. coli(pSV-1) may be due to

3404

VIMONT ET AL.

ANTIMICROB. AGENTS CHEMOTHER.

TABLE 2. MICs of ␤-lactams for E. persicina reference strain 105199T, E. coli DH10B(pSV-1), and E. coli DH10B reference strain MIC (␮g/ml) ␤-Lactam(s)b

Amoxicillin Amoxicillin ⫹ CLA Ticarcillin Ticarcillin ⫹ CLA Piperacillin Piperacillin ⫹ TZB Cephalothin Cefsulodin Cefuroxime Cefotaxime Cefotaxime ⫹ CLA Cefotaxime ⫹ TZB Ceftazidime Ceftazidime ⫹ CLA Ceftazidime ⫹ TZB Ceftriaxone Cefepime Cefpirome Cefoxitin Moxalactam Aztreonam Imipenem

a

E. persicina 105199T

E. coli DH10B(pSV-1)

E. coli DH10B

256 0.5 64 0.5 4 0.25 16 32 0.12 0.06 ⬍0.06 ⬍0.06 ⬍0.06 ⬍0.06 ⬍0.06 0.06 ⬍0.06 ⬍0.06 1 0.06 0.12 0.06

⬎512 8 ⬎512 64 512 2 ⬎512 ⬎512 128 1 0.06 0.06 0.25 0.06 0.06 4 1 1 2 0.25 4 0.12

2 2 1 1 1 1 2 16 2 ⬍0.06 ⬍0.06 ⬍0.06 ⬍0.06 ⬍0.06 ⬍0.06 ⬍0.06 ⬍0.06 ⬍0.06 2 0.12 0.06 0.06

a E. persicina 105199T and E. coli DH10B(pSV-1) produced ␤-lactamase ERP-1. b CLA, clavulanic acid at a fixed concentration of 2 ␮g/ml; TZB, tazobactam at a fixed concentration of 4 ␮g/ml.

the chromosmal versus the plasmid location of blaERP-1 or to leakage of the periplasmic enzyme into the growth medium as reported for E. amylovora (7). The ␤-lactam resistance pattern conferred by ERP-1, once its gene is expressed in E. coli, resembled that of the cloned chromosomally encoded ␤-lactamase genes of other enterobacterial species such as C. koseri, K. cryocrescens, K. ascorbata, P. penneri, P. vulgaris, P. penneri, R. aquatilis, and S. fonticola. Biochemical analysis of ␤-lactamase ERP-1. The specific activity of the purified ␤-lactamase ERP-1, measured with 100 ␮mol of cephalothin as the substrate, was 502 U mg of protein⫺1 with a 95-fold purification factor. Its purity was estimated to be ⬎95% by SDS-PAGE analysis (data not shown). ERP-1 had a robust hydrolysis against benzylpenicillin, amoxicillin, ticarcillin, piperacillin, cephalothin, cefuroxime, and ceftriaxone (Table 3). A significant hydrolytic activity was also observed against cefotaxime, whereas no activity was detectable against ceftazidime. The activity against cefuroxime was shared by the chromosomally encoded ␤-lactamases of C. diversus, K. cryocrescens, P. vulgaris, P. penneri, R. aquatilis, and S. fonticola (2, 4, 8, 17–19, 24, 25). Inhibition studies showed that the IC50 values of clavulanic acid, tazobactam, and sulbactam were low: 60, 40, and 500 nM, respectively. Based on its kinetic parameters, ␤-lactamase ERP-1 belongs to the group 2be ␤-lactamases of the BushMedeiros-Jacoby-classification (6). Isoelectric focusing analysis showed that cultures of E. persicina 105199T and E. coli DH10B(pSV-1) gave a single and identical ␤-lactamase with a pI value of 8.1. The cleavage site of the leader peptide was calculated to be between two alanine

TABLE 3. Kinetic parameters of purified ␤-lactamase ERP-1a Substrate

Benzylpenicillin Amoxicillin Ticarcillin Piperacillin Cephalothin Cefuroxime Cefoxitin Cefotaxime Ceftazidime Ceftriaxone Cefpirome Cefepime Aztreonam Imipenem

kcat (s⫺1) Km (␮M)

17 12 2 10 47 4 ⬍0.01 5 ⬍0.1 11 ⬍0.01 ⬍0.01 2 ⬍0.01

12c 28 4c 11c 44 43 NDd 89 ND 52 ND ND 87 35

kcat/Km (s⫺1 䡠 mM⫺1)

Relative kcat/Kmb

1,400 450 500 900 1,100 90 ND 60 ND 210 ND ND 23 ⬍0.001

100 32 35 64 73 7 ND 4 ND 15 ND ND 1 ND

a Data are the means of three independent experiments. Standard deviations were within 15%. b kcat/Km values were expressed relative to that of benzylpenicillin, which was set at 100. c In cases with low Km values, Ki values were determined with cephalothin as the substrate. d ND, not determinable.

residues (between the VFA and AG motifs [Fig. 1]). This cleavage site position was confirmed by Edman sequence analysis, which determined the N-terminal end of mature ␤-lactamase ERP-1 to be the motif AGDSLQ. The relative molecular mass of ERP-1 determined with the purified enzyme was ca. 28 kDa (data not shown). Induction studies failed to detect expression induction of ␤-lactamase with cultures of E. persicina 105199T. This results was consistent with the absence of a LysR-type regulator gene located upstream of blaERP-1. Thus, this chromosomally encoded class A ␤-lactamase was not inducible, as opposed to the naturally occurring class A enzymes of C. koseri, P. penneri, P. vulgaris, and S. fonticola. Conclusion. Comparison of EPR-1 sequence to those of other class A ESBLs identified several amino acid residues that may be involved in its extended spectrum of hydrolysis (18). Indeed, the serine residue at position 237 is also found in the chromosomally encoded ␤-lactamases of K. ascorbata, K. cryocrescens, P. penneri, P. vulgaris, R. aquatilis, and S. fonticola. ␤-Lactamase ERP-1, like the CTX-type enzyme Toho-1 (13), has glycine residues in the strand B3 (positions 228, 232, and 238) that may increase its flexibility or that allows ERP-1 to bind extended-spectrum cephalosporins. Additionally, the omega loop of ERP-1 (involved in the catalytic site of class A enzymes) is different from that of related enzymes (Fig. 1) since Thr160 may form an hydrogen bond with Thr180 as in non-ESBL enzymes (13, 18). The present study identified a novel, chromosomally encoded ESBL from an enterobacterial species and provides further support for the view that ESBLs may be present in enterobacterial species even though they may not be suspected based upon analysis of the ␤-lactam resistance profile. Although ERP-1 is not closely related to known plasmid-mediated ESBLs, it may be the progenitor of enzymes yet to be identified. This is the first identication of an ESBL from an enterobacterial species that is a natural contaminant and pathogen of fruits, plants, and vegetables. This result may

␤-LACTAMASE FROM ERWINIA PERSICINA

VOL. 46, 2002

provide an interesting contribution to the current debate on health risks of genetically engineered plants as a reservoir for the spread of antibiotic resistance genes.

16.

ACKNOWLEDGMENTS This work was financed by a grant from the Ministe`re de l’Education Nationale et de la Recherche (grant UPRES-EA), Universite´ Paris XI, Paris, France. We thank C. Bizet for the Erwinia reference strains. REFERENCES 1. Ambler, R. P., A. F. W. Coulson, J.-M. Fre`re, J. M. Ghuysen, B. Joris, M. Forsman, R. C. Le´vesque, G. Tiraby, and S. G. Waley. 1991. A standard numbering scheme for the class A ␤-lactamases. Biochem. J. 276:269–270. 2. Arakawa, Y., M. Ohta, N. Kido, M. Mori, H. Ito, T. Komatsu, Y. Fujii, and N. Kato. 1989. Chromosomal ␤-lactamase of Klebsiella oxytoca, a new class A enzyme that hydrolyzes broad-spectrum ␤-lactam antibiotics. Antimicrob. Agents Chemother. 33:63–70. 3. Barthe´le´my, M., J. Pe´duzzi, and R. Labia. 1988. Complete amino acid sequence of p453-plasmid-mediated PIT-2 ␤-lactamase (SHV-1). Biochem. J. 251:73–79. 4. Bellais, S., L. Poirel, N. Fortineau, J. W. Decousser, and P. Nordmann. 2001. Biochemical-genetic characterization of the chromosomally encoded extended-spectrum class A ␤-lactamase from Rahnella aquatilis. Antimicrob. Agents Chemother. 45:2965–2968. 5. Bonnet, R., J. L. M. Sampaio, R. Labia, C. De Champs, D. Sirot, C. Chanal, and J. Sirot. 2000. A novel CTX-M ␤-lactamase (CTX-M-8) in cefotaximeresistant Enterobacteriaceae isolated in Brazil. Antimicrob. Agents Chemother. 44:1936–1942. 6. Bush, K., A. A. Medeiros, and G. A. Jacoby. 1995. A functional classification scheme for ␤-lactamases and its correlation with molecular structure. Antimicrob. Agents Chemother. 39:1211–1233. 7. Chatterjee, A. K., R. F. Buss, and M. P. Starr. 1977. Unusual susceptibility of Erwinia amylovora to antibacterial agents in relation to the barrier function of its cells envelope. Antimicrob. Agents Chemother. 11:897–905. 8. Decousser, J. W., L. Poirel, and P. Nordmann. 2001. Characterization of a chromosomally encoded extended-spectrum class A ␤-lactamase from Kluyvera cryocrescens. Antimicrob. Agents Chemother. 45:3595–3598. 9. Duarte´, X., C. T. Anderson, M. Grimson, R. D. Barabote, R. E. Strauss, L. S. Gollahon, and M. J. D. San Francisco. 2001. Erwinia chrysanthemi strains cause death of human gastrointestinal cells in culture and express an intiminlike protein. FEMS Microbiol. Lett. 190:81–86. 10. Hao, M. V., D. J. Brenner, A. G. Steigerwalt, Y. Kosako, and K. Komagata. 1990. Erwinia persicinus, a new species isolates from plants. Int. J. Syst. Bacteriol. 40:379–383. 11. Hauben, L., E. R. B. Moore, L. Vauterin, M. Steenackers, J. Mergaert, L. Verdonck, and J. Swings. 1998. Phylogenetic position of phytopathogens within the Enterobacteriaceae. Syst. Appl. Microbiol. 21:384–397. 12. Hewick, R. M., M. W. Hunkapiller, D. Le Hoo, and W. J. Dreyer. 1981. A gas-liquid solid phase peptide and protein sequenator. J. Biol. Chem. 256: 7990–7997. 13. Ibuka, A., A. Taguchi, M. Ishiguro, S. Fushinobu, Y. Ishii, S. Kamitori, K. Okuyama, K. Yamaguchi, M. Konno, and H. Matsuzawa. 1999. Crystal structure of the E166A mutant of extended-spectrum ␤-lactamase Toho-1 at ˚ resolution. J. Mol. Biol. 285:2079–2087. 1.8 A 14. Ishiguro, K., and K. Sugimoto. 1996. Purification and characterization of the Proteus vulgaris BlaA protein, the activator of the ␤-lactamase gene. J. Biochem. 120:98–103. 15. Jones, M. E., and P. M. Bennett. 1995. Inducible expression of the chromo-

17.

18. 19. 20. 21.

22.

23. 24.

25.

26.

27.

28.

29. 30.

31.

3405

somal cdiA from Citrobacter diversus NF85, encoding an Ambler class A ␤-lactamase, is under similar genetic control to the chromosomal ampC, encoding an Ambler class C enzyme from Citrobacter freundii OS60. Microb. Drug Resist. 1:285–291. Joris, B., P. Ledent, O. Dideberg, E. Fonze, J. Lamotte-Brasseur, J. A. Kelly, J. M. Ghuysen, and J.-M. Fre`re. 1991. Comparison of the sequences of class A ␤-lactamases and of the secondary structure elements of penicillin-recognizing proteins. Antimicrob. Agents Chemother. 35:2294–2301. Liassine, N., S. Madec, B. Ninet, C. Metral, M. Fouchereau-Peron, R. Labia, and R. Auckenthaler. 2002. Postneurosurgical meningitis due to Proteus penneri with selection of a ceftriaxone-resistant isolate: analysis of chromosomal class A ␤-lactamase HugA and its LysR-type regulatory protein HugR. Antimicrob. Agents Chemother. 46:216–219. Matagne, A., J. Lamotte-Brasseur, and J.-M. Fre`re. 1998. Catalytic properties of class A ␤-lactamases: efficiency and diversity. Biochem. J. 330:581– 598. Matsumoto, Y., and M. Inoue. 1999. Characterization of SFO-1, a plasmidmediated inducible class A ␤-lactamase from Enterobacter cloacae. Antimicrob. Agents Chemother. 43:307–313. McGowan, S. J., B. W. Bycroft, and G. P. C. Salmond. 1998. Bacterial production of carbapenems and clavams: evolution of ␤-lactam antiobiotic pathways. Trends Microbiol. 6:203–208. Mergaert J., L. Verdonck, and K. Kersters. 1993. Transfer of Erwinia ananas (synonym, Erwinia uredovora) and Erwinia stewartii to the genus Pantoea emend. as Pantoea ananas (Serrano 1928) comb. nov. and Pantoea stewartii (Smith 1898) comb. nov. resepectively, and description of Pantoea stewartii sp. indologenes subsp. nov. Int. J. Syst. Bacteriol. 43:162–173. National Committee for Clinical Laboratory Standards. 2000. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically, 5th ed. Approved standard M7-A4. National Committee for Clinical Laboratory Standards, Wayne, Pa. O’Hara, C. M., A. G. Steigerwalt, B. C. Hill, J. M. Miller, and D. J. Brenner. 1998. First report of a human isolate of Erwinia persicinus. J. Clin. Microbiol. 36:248–250. Pe´duzzi, J., S. Farzaneh, A. Reynaud, M. Barthe´le´my, and R. Labia. 1997. Characterization and amino acid sequence analysis of a new oxyimino cephalosporin-hydrolyzing class A ␤-lactamase from Serratia fonticola CUV. Biochim. Biophys. Acta 1341:58–70. Pe´duzzi, J., A. Reynaud, P. Baron, M. Barthe´le´my, and R. Labia. 1994. Chromosomally encoded cephalosporin-hydrolyzing ␤-lactamase of Proteus vulgaris RO104 belongs to Ambler’s class A. Biochim. Biophys. Acta 1207: 31–39. Perilli, M., N. Franceschini, B. Segatore, G. Amicosante, A. Oratore, C. Duez, B. Joris, and J.-M. Fre`re. 1991. Cloning and nucleotide sequencing of the gene encoding the ␤-lactamase from Citrobacter diversus. FEMS Microbiol. Lett. 67:79–84. Poirel, L., M. Guibert, D. Girlich, T. Naas, and P. Nordmann. 1999. Cloning, sequence analyses, expression and distribution of ampC-ampR from Morganella morganii clinical isolates. Antimicrob. Agents Chemother. 43:769– 776. Sabate, M., R. Tarrago, F. Navarro, E. Miro, C. Verges, J. Barbe, and G. Prats. 2000. Cloning and sequence of the gene encoding a novel cefotaximehydrolyzing ␤-lactamase (CTX-M-9) from Escherichia coli in Spain. Antimicrob. Agents Chemother. 44:1970–1973. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Seoane, A., and J. M. Garcia-Lobo. 1991. Nucleotide sequence of a new class A ␤-lactamase gene from the chromosome of Yersinia enterocolitica: implications for the evolution of class A ␤-lactamases. Mol. Gen. Genet. 228:215– 220. Starr, M. P., and A. K. Chatterjee. 1972. The genus Erwinia: enterobacteria pathogenic to plants and animals. Annu. Rev. Microbiol. 26:389–426.