The penicillin resistance of Enterococcus faecalis JH2-2r results from an overproduction of the low-affinity penicillin-binding protein PBP4 and does not involve a psr-like gene

Colette Duez1, Willy Zorzia,1, Frédéric Sapunaric1, Ana Amorosob,1, Iris Thamm1 and Jacques Coyette1

Centre d’Ingénierie des Protéines, Institut de Chimie, B6, Université de Liège, (Sart Tilman), B-4000 Liège, Belgium1

Author for correspondence: Jacques Coyette. Tel: +32 4 366 33 99. Fax: +32 4 366 33 64. e-mail: jcoyette{at}ulg.ac.be


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
A penicillin-resistant mutant, JH2-2r (MIC 75 µg ml-1), was isolated from Enterococcus faecalis JH2-2 (MIC 5 µg ml-1) by successive passages on plates containing increasing concentrations of benzylpenicillin. A comparison of the penicillin-binding protein (PBP) profiles in the two strains revealed a more intensely labelled PBP4 in JH2-2r. Because the sequences of the JH2-2 and JH2-2r pbp4 genes were strictly identical, even in their promoter regions, this intensive labelling could only be associated with an overproduction of the low-affinity PBP4. No psr gene analogous to that proposed to act as a regulator of PBP5 synthesis in Enterococcus hirae and Enterococcus faecium could be identified in the vicinity of pbp4 in E. faecalis JH2-2 and JH2-2r. However, a psr-like gene distant from pbp4 was identified. The cloning and sequencing of that psr-like gene from both E. faecalis strains indicated that they were identical. It is therefore postulated that the PBP4 overproduction in E. faecalis JH2-2r results from the modification of an as yet unidentified factor.

Keywords: ß-lactam resistance, low-affinity PBP, PBP-synthesis repressor (psr) gene

Abbreviations: HMM, high molecular mass; LMM, low molecular mass; PBP, penicillin-binding protein

The GenBank accession numbers for the sequences reported in this paper are Y17797 for the 8·4 kb segment of pDML521; AJ290435 for pbp4 of E. faecalis JH2-2; AJ276231 and AJ276232 for the psr-like gene of E. faecalis JH2-2 or JH2-2r, respectively.

a Present address: Laboratory of Human Histology, Institut A. Swaen, L3, Université de Liège, rue des Pitteurs 20, B-4020 Liège, Belgium.

b Present address: Departamento de Microbiologia y Immunologia, Facultad de Farmacia y Bioquimica, Universidad de Buenos Aires, Junin 956, 1113 Buenos Aires, Argentina.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Enterococci are normal human commensals. However, they are now considered as pathogens responsible for a wide variety of infections in humans (Murray, 1990 ; Jett et al., 1994 ; Schmit et al., 1994 ). They appeared in second position on a list of the five leading pathogens in a survey made in the USA between 1986 and 1997 (Hancock & Gilmore, 2000 ). Enterococcus faecalis caused 80–90% of enterococcal infections, while Enterococcus faecium accounted for 5–10% (Murray, 1990 ; Jett et al., 1994 ; Schmit et al., 1994 ).

Enterococci are intrinsically more resistant to ß-lactams than other streptococci, with 10- to 100-fold higher MICs (Gutmann, 1994 ). The vast majority of E. faecalis strains have penicillin MICs ranging between 1 and 4 µg ml-1. Many E. faecium strains have a MIC above 8 µg ml-1, with the majority ranging between 16 and 32 µg ml-1 (Murray, 1990 ). The resistance mechanism of enterococci to ß-lactams has mainly been studied in two closely related species: Enterococcus hirae and E. faecium. Both have an essential low-affinity penicillin-binding protein (PBP5) which is directly responsible for the phenomenon (Fontana et al., 1983 , 1994 ; Williamson et al., 1985 ; El Kharroubi et al., 1991 ; Piras et al., 1993 ; Ligozzi et al., 1996 ; Zorzi et al., 1996 ). When PBP5 is not synthesized, the cells become susceptible to ß-lactams with MICs below 0·2 µg ml-1 (Fontana et al., 1985 ). On the other hand, overproduction of PBP5 and/or reduction of its affinity in both species leads to high-level resistance (Fontana et al., 1983 ; Klare et al., 1992 ; Ligozzi et al., 1996 ; Zorzi et al., 1996 ).

In both species, synthesis of PBP5 was reported to be under the control of a repressor-encoding gene, psr (for PBP5 synthesis repressor), which is located immediately upstream from the pbp5 gene (Ligozzi et al., 1993 ; Massidda et al., 1998 ). Inactivation of psr by a point mutation or a deletion seemed to result in the full expression of pbp5 and increased resistance of the cells. In contrast, normal activity of psr in wild-type strains apparently reduced the expression of pbp5 to a low level and induced a higher susceptibility of the cells to penicillin (Ligozzi et al., 1993 ).

E. faecalis also has a low-affinity PBP (Signoretto et al., 1994 ). Analysis of clinical and laboratory resistant strains has shown that it is involved in ß-lactam resistance exactly as PBP5 is in E. hirae and E. faecium (Williamson et al., 1985 ; Fontana et al., 1994 ; Signoretto et al., 1994 ). However, information concerning that PBP, previously designated as PBP5 (Signoretto et al., 1994 ), mainly concerns its cloning, sequencing and expression in Escherichia coli.

In this study we could not identify a psr-like gene in the immediate vicinity of the E. faecalis low-affinity PBP4-encoding gene as observed in E. hirae or E. faecium. A psr-like gene was however found elsewhere in the genome, but it appeared not to be involved in the overproduction of the E. faecalis low-affinity PBP4.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains, isolation of plasmids and membranes.
Enterococcus faecalis and Escherichia coli cells were grown in Brain Heart infusion broth (BHI) and Luria–Bertani (LB) medium, respectively. Resistant clones of E. faecalis JH2-2 (benzylpenicillin MIC value: 5 µg ml-1) were selected by successive growths on BHI plates containing increasing concentrations of benzylpenicillin (PenG). One clone was randomly chosen and designated as JH2-2r (PenG MIC value: 75 µg ml-1). E. coli strains Top 10 F' and INV{alpha}F' were used to replicate the different E. coli plasmids described in Table 1. E. coli plasmids were extracted by the Wizard Plus Minipreps DNA purification system (Promega) or with the GFX Microplasmid prep kit (Amersham Pharmacia Biotech). Membranes of the enterococcal strains and E. coli K-12 RR1 were isolated as described previously (Lindström et al., 1970 ; Coyette et al., 1978 ). Membrane-bound PBPs (150 µg total protein) were labelled with [14C]PenG or [125I]PenG and detected by fluorography or a phosphorus storage K screen (FX-imager apparatus, Bio-Rad) (Laskey, 1980 ; Masson & Labia, 1983 ; Zorzi et al., 1996 ).


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains and plasmids used in this study

 
Molecular biology techniques.
Enterococcus genomic DNA was isolated as reported previously (Loureiro Dos Santos & Chopin, 1987 ) from cells grown unshaken at 37 °C in BH medium and collected at the end of the exponential phase.

PCR amplifications were made as described previously (Innis et al., 1990 ) using the Biotools DNA polymerase (Life Sciences International). The DNA fragments or PCR products were purified with the help of the GeneClean spin kit (Bio 101) or the Wizard PCR preps DNA purification system (Promega).

Hybridizations using oligonucleotides or DNA fragments labelled with digoxigenin were prepared according to the instructions of the DIG system user’s guide from Boehringer Mannheim.

DNA sequencing was performed on both strands using the T7 sequencing kit with [35S]dATP{alpha}S labelling, and the Autoread or ThermoSequenase sequencing kits with 5'-fluorescein or Cy5 primers, in which case the electrophoresis was performed on an ALF express DNA sequencer (Amersham Pharmacia Biotech). The nucleotide sequences were introduced in GELASSEMBLE (Pearson & Lipman, 1988 ), the ORFs were identified with CODONREFERENCE (Devereux et al., 1984 ) and homology searches (SWISS-PROT, PIR, Genpept) were made using FASTA or BLASTP (Altschul et al., 1990 ). The EMBL accession numbers for the different sequences are as follows: 8·4 kb segment of pDML521, Y17797; pbp4 of E. faecalis JH2-2, AJ290435; and psr-like of E. faecalis JH2-2 or JH2-2r, AJ276231 and AJ276232, respectively.

Construction of plasmids pDML523 to pDML526.
The JH2-2 and JH2-2r pbp4 genes were amplified by PCR using the oligonucleotides O16 and O17. The purified 2·2 kb fragments thus obtained were digested with BamHI and EcoRI and cloned into pMCL210 between the BamHI and EcoRI sites to yield pDML523 and pDML524, respectively.

The complete JH2-2 and JH2-2r psr-like genes were amplified by PCR using oligonucleotides O14 and O15 and cloned into the pGEM-T Easy vector to yield pDML525 and pDML526, respectively.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
PBP profile in E. faecalis
In the penicillin-resistant E. faecalis 56R (Signoretto et al., 1994 ), five high-molecular-mass (HMM) PBPs, namely PBPs 1 to 5, and a low-molecular-mass (LMM) PBP, designated PBP6, were proposed as membrane-bound PBPs. In that profile, however, PBP4 and PBP5 co-migrated as a single thick band, probably due to an overloading of membrane proteins on the gel. Another more recent study (Mainardi et al., 1998 ) proposed only four HMM PBPs, PBP1 to 4, in E. faecalis JH2-2. When we compared the PBP profiles of E. faecalis JH2-2 and JH2-2r, we also detected only four HMM PBPs and a barely visible LMM PBP (probably a DD-carboxypeptidase), designated in our study as PBP5 (Fig. 1, lanes 3–6). Generally, PBP2 appeared as a thin band in contrast with the three other HMM PBPs. Occasionally, PBP3 and PBP4 co-migrated and formed one very thick band. However, because of its low affinity, PBP4 could easily be distinguished from PBP3 by direct labelling and prelabelling fluorography experiments. Some membrane preparations seemed to show an additional PBP that migrated just below PBP4. However, freshly prepared membrane samples did not show this additional band, which could derive from another PBP by a spontaneous breakdown phenomenon analogous to that observed for PBP4* in membrane preparations of E. hirae ATCC 9790 (Coyette et al., 1980 ). Nevertheless, from the protein sequence comparisons, it is clear that the PBP, named PBP5 by Signoretto et al. (1994) corresponds to the PBP4 studied in this work.



View larger version (72K):
[in this window]
[in a new window]
 
Fig. 1. SDS-PAGE and fluorogram of the membrane-bound PBPs from E. hirae ATCC 9790, E. faecalis JH2-2, JH2-2r and Escherichia coli overproducing the E. faecalis PBP4. Conditions of electrophoresis: 7·2% (w/v) acrylamide; 16 h; gel length 24 cm. Membranes (150 µg protein) of E. hirae and E. faecalis and membranes (15 µg protein) of recombinant E. coli were labelled with 60 µM benzyl[125I]penicillin for 30 min either directly (odd-numbered lanes) or with 90 µM benzyl[125I]penicillin after previous treatment with 15 µM non-radioactive benzylpenicillin for 30 min (even-numbered lanes). Exposure conditions: 16 h with a phosphorus storage K screen (FX-imager apparatus, Bio-Rad). Lanes 1 and 2, E. hirae ATCC 9790; lanes 3 and 4, E. faecalis JH2-2; lanes 5 and 6, E. faecalis JH2-2r; lanes 7 and 8, E. coli overproducing the PBP4 of E. faecalis JH2-2; lanes 9 and 10, E. coli overproducing the PBP4 of E. faecalis JH2-2r.

 
PBP4 was the only PBP in E. faecalis JH2-2 and JH2-2r membrane preparations that reacted on a Western blot with polyclonal antibodies raised against E. hirae PBP5 (El Kharroubi et al., 1991 ) or PBP3r (Piras et al., 1990 ). As already previously reported (Signoretto et al., 1994 ), the E. faecalis PBP4 reacted with the antibodies much more weakly than E. hirae PBP5 or PBP3r (data not shown).

PenG acylation and deacylation rates of the PBPs were estimated by 5 min labelling of JH2-2 and JH2-2r membrane preparations with increasing concentrations of [14C]PenG (ranging from 1 to 500 µM). The concentrations yielding a 50% saturation (ID50) of PBP3 and PBP4 were 6·7±3 and 46±34 µM, respectively (which are equivalent to 2·4±1·1 and 16±11·8 µg ml-1; mean value of three different experiments). The second-order rate constants of acylation (k+2/K) of these PBPs calculated from these values were 340±150 and 50±40 M-1 s-1, respectively (Ghuysen et al., 1986 ). The deacylation constant k+3 values of both PBP3 and PBP4 complexes were close to 5x10-6 s-1 indicating a very slow breakdown of the complexes, which could thus be neglected for calculation of the k+2/K acylation constant. It is interesting to note that no difference could be seen in the affinity of PBP4 in JH2-2 and JH2-2r membranes. However, it was obvious that PBP4 was more intensely labelled in JH2-2r (Fig. 1, lanes 5 and 6) than in JH2-2 membrane samples (Fig. 1, lanes 3 and 4). When the amount of PBP4 was related to the amount of total membrane protein or to the amount of PBP1, used as an internal standard (lanes 3 and 5), it clearly appeared that the production of PBP4 was 3·5-fold (mean value) higher in JH2-2r than in JH2-2 while that of the other PBPs remained constant. This level of overproduction is much lower than the increase of the MIC value (15-fold factor when JH2-2r and JH2-2 are compared). However, as mentioned for E. faecium (Rice et al., 2001 ) and as we have observed with E. hirae transformants (unpublished results), there is no direct correlation between the MIC values and the quantities of low-affinity PBP overproduced in the resistant strains.

Cloning of pbp4, the PBP4-encoding gene of the resistant E. faecalis strain JH2-2r
The cloning and sequencing of pbp4 was undertaken well before the sequencing data of the E. faecalis V583 genome were released. The sequences of two degenerate oligonucleotides, O1 and O2 (Table 2), were based on the amino acid sequences of two peptides, WQKD(S,Q,K)SWG and AQSN(K,E)(E,D,N), relatively well conserved in the low-affinity PBP3r and PBP5 of E. hirae (El Kharroubi et al., 1991 ; Piras et al., 1993 ) and PBP2' of Staphylococcus aureus (Song et al., 1987 ). Both oligonucleotides were used as primers in an amplification experiment with Taq DNA polymerase and 2 µg of E. faecalis JH2-2r genomic DNA. The expected 223 bp amplified fragment was cloned into the PCR2.1 plasmid (Table 1) and sequenced. Translation of its sequence yielded the typical SDN motif conserved in penicillin-binding modules of PBPs (Goffin & Ghuysen, 1998 ).


View this table:
[in this window]
[in a new window]
 
Table 2. Oligonucleotides designed for hybridization or DNA amplifications by PCR

 
EcoRI, HindIII or NcoI genomic DNA fragments of JH2-2r were probed by Southern blotting with the digoxigenin-labelled oligonucleotide O3 derived from the sequence of the 223 bp PCR fragment (Table 2). A single positive signal was obtained in each digest at the level of an 11 kb EcoRI fragment, a 2·5 kb HindIII fragment and a >23 kb NcoI fragment. After electrophoresis, the 9–15 kb EcoRI DNA fragments were eluted and ligated to EcoRI-linearized pBR322 plasmid. Transformants of E. coli K-12 RR1 were screened by Southern blotting with the same hybridization probe as above. Two positive clones, bearing pDML521 and pDML522, respectively (Table 1), were selected out of 840 transformants. Each clone overproduced an additional membrane-bound PBP which migrated by SDS-PAGE as E. faecalis PBP4. This meant that the cloned 11 kb fragment contained all the signals required for transcription and translation of the PBP4-encoding gene.

Sequence of an 8.4 kb segment of the 11 kb EcoRI insert of pDML521
The sequencing of the 11 kb EcoRI insert of pDML521 was undertaken to explore the genic environment of pbp4 and to search for a gene which would function as a regulator of PBP4 synthesis, as proposed for psr in E. hirae (Ligozzi et al., 1993 ).

The organization of the 8456 bp segment sequenced from the 11 kb EcoRI insert of pDML521 is presented in Fig. 2. The first ORF (890 bp) seemed to be preceded by a Shine–Dalgarno sequence and to start at position 180 from the 5' end. The translation product was weakly homologous (24–28% identity and 32–43% similarity) to those of several bacterial gph genes which were considered to code for phosphoglycolate phosphatases (e.g. Synechocystis sp., Alcaligenes eutrophus, Haemophilus influenzae, Aquifex aeolicus, Borrelia burgdorferi, Escherichia coli). It was also weakly but significantly (46% similarity and 23% identity) homologous to the product of the yvoE gene of Bacillus subtilis (Kunst et al., 1997 ).



View larger version (4K):
[in this window]
[in a new window]
 
Fig. 2. Organization of the 8456 bp EcoRI–PstI fragment of the 11 kb EcoRI insert contained in pDML521. The genes are represented by arrows corresponding to the orientation of their transcription; names of genes beginning with y were given by analogy with homologous genes of B. subtilis (Kunst et al., 1997 ). The open circle symbol indicates a potential promoter; gph, phosphoglycolate phosphatase gene.

 
A putative 488 bp orf2 devoid of a Shine–Dalgarno sequence was found beginning 1256 bp downstream from the 5' end of the 11 kb insert. Its product showed no homology with any other protein available in the databases. orf3 was preceded by a Shine–Dalgarno sequence and detected on the complementary strand between positions 1273 and 2013. It coded for a putative 247 aa protein homologous (52% similarity and 31% identity) to a small protein of unknown function in the temperate lactococcal bacteriophages r1t (Van Sinderen et al., 1996 ) and {phi}LC3 (Lillehang et al., 1997 ).

Three overlapping genes were also located on the complementary strand (from positions 2036 to 5514). They were named ydjI, ydjG and ydjH by analogy with related genes of unknown functions present in the genome of B. subtilis and clustered in one transcriptional unit but in the logical alphabetical order (Kunst et al., 1997 ). The homology scores determined by the BESTFIT program between each pair of gene products were the following: ydjG, 57% similarity and 34% identity; ydjH, 53% similarity and 26% identity; ydjI, 55% similarity and 26% identity. The E. faecalis YdjI protein was also highly similar (71% identity and 78% similarity) to the Lactococcus lactis ORFE (Andersen et al., 1996 ) of unknown function. In L. lactis, orfE is adjacent to the pyr operon but transcribed in an orientation opposite to that of the genes involved in the biosynthesis of pyrimidines.

A 207 bp sequence separated the initial ATG codon of ydjI from that of the following pbp4 gene oriented in the opposite direction. This intervening region containing the putative pbp4 promoter was amplified from the genomic DNA of both the resistant JH2-2r strain and the sensitive JH2-2 strain, using oligonucleotides O4 and O5 (Table 2). After cloning and sequencing, both fragments appeared identical.

The pbp4 gene was followed by a partial ORF, named ydiC for its high similarity with the ydiC gene, of unknown function in B. subtilis (67% similarity and 40% identity on 204 amino acids) (Kunst et al., 1997 ). YdiC is also related to a hypothetical protein widely conserved in bacteria (e.g. Escherichia coli, Streptomyces coelicolor, Mycobacterium tuberculosis, Mycobacterium leprae, H. influenzae, Neisseria meningitidis, Aquifex aeolicus, Synechocystis sp.).

It is interesting to note that a gene organization identical to that of the 8·45 kb segment of JH2-2r (Fig. 2) was found in the genome of E. faecalis V583 (TIGR database).

The last part of the 11 kb EcoRI insert of pDML521, a 2·5 kb segment located downstream of the 8·45 kb segment described above, was not sequenced further as subclones appeared to be unstable.

Analysis of the JH2-2 and JH2-2r pbp4 sequences and expression in E. coli
The PBP4-encoding gene of the E. faecalis JH2-2 parental strain was completely sequenced. For this purpose, four pairs of oligonucleotides derived from the pbp4 sequence of E. faecalis JH2-2r – O6 and O7, O8 and O9, O10 and O11, O12 and O13 (Table 2) – were used to amplify overlapping fragments of the gene by PCR. The PCR products were cloned into the pGEM-T Easy vector and sequenced on both strands, using the universal and reverse primers.

Alignment of the translation products of the two pbp4 revealed that they were identical. They were also identical to the product of the pbp4 gene found in contig 6237 during the sequencing of the E. faecalis V583 genome (TIGR database). These results explained why the PBP4 affinities were identical when they were estimated above, on the membranes of the JH2-2 and JH2-2r strains and the Escherichia coli Top10 F' cells transformed with pDML523 and pDML524 (Table 1) carrying the JH2-2 and JH2-2r pbp4 PCR fragments, respectively. In this latter case, the membranes of the E. coli transformants were labelled with [125I]PenG and analysed by the Bio-Rad FX-imager apparatus after electrophoresis. They produced the same amount of PBP4 (Fig. 1, lanes 7–10). Membranes of E. hirae ATCC 9790 (Fig. 1, lanes 1 and 2), E. faecalis JH2-2 (Fig. 1, lanes 3 and 4) and JH2-2r (Fig. 1, lanes 5 and 6) were used in parallel as controls.

All these results, together with the demonstration that the 207 bp regions located upstream of pbp4 in both strains were identical, indicated that the differences seen in JH2-2 and JH2-2r in terms of penicillin MIC and PBP4 production could not be attributed to a modification in the coding region of the pbp4 gene nor in its putative promoter.

Identification of a psr gene in E. faecalis JH2-2 and JH2-2r
The sequence of the psr gene of E. hirae ATCC 9790 (Massidda et al., 1998 ; accession number U42211 in the EMBL database) was used to search for a homologous gene in the TIGR database, collecting the sequencing data of the genome of E. faecalis V583.

The BLAST (Pearson & Lipman, 1988 ) search server identified a nucleotide sequence encoding a protein in contig 6330 that had 63% similarity and 46% identity with the psr gene product of E. hirae. This sequence was considered as the psr gene of E. faecalis. The complete psr genes were amplified from the genomic DNA of E. faecalis JH2-2 and JH2-2r, cloned into pGEM-T Easy and sequenced. Comparison of the two sequences indicated that they were identical in the sensitive and resistant strains even in the putative promoter region. They were also identical to the sequence identified in contig 6330 of E. faecalis V583 (except that residue 348 is a threonine in JH2-2 or JH2-2r sequences and an alanine in the V583 sequence).

The E. faecalis psr-like gene encodes a 390-residue protein that has a 100 aa N-terminal extension without any significant homology in the protein databases and is absent in the 293-residue Psr of E. hirae (Massidda et al., 1998 ). Another peculiarity of the structure of the E. faecalis Psr-like protein concerns the presence of a hydrophobic peptide that seems to extend from positions 101 to 117 and separate the N-terminal extension from the typical Psr module. Such a structure stands in contrast with that of the other Psr-like proteins, which all appear to have a putative N-terminal hydrophobic membrane-anchoring peptide.

In the sequence of the E. faecalis Psr-like protein, one can find amino acid motifs almost identical to those conserved in the Psr and Psr-like proteins (Massidda et al., 1998 ). A pairing comparison within this family of proteins (using the BESTFIT program of the GCG package) gave the highest scores of identity between the E. faecalis Psr-like protein and the enterococcal Psr proteins (Ligozzi et al., 1993 ; Zorzi et al., 1996 ; Massidda et al., 1998 ).


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
E. faecalis shows only four HMM PBPs and one LMM PBP when membranes are labelled with radioactive penicillin. This is in agreement with the five PBP genes found in the genome of E. faecalis V583 when the sequences available in the TIGR database were analysed. Four of these genes coded for HMM PBPs (one belonging to class A and three to class B). The last one coded for a LMM PBP, presumably a DD-carboxypeptidase.

PBP4 in E. faecalis is a low-affinity PBP closely related to other enterococcal low-affinity PBPs described previously in E. hirae and E. faecium (El Kharroubi et al., 1991 ; Piras et al., 1993 ; Signoretto et al., 1994 ; Zorzi et al., 1996 ). It reacts poorly with anti-E. hirae PBP5 antibodies and it has a very low estimated PenG second-order acylation constant value (k+2/K=50 M-1 s-1) comparable to those of E. hirae PBP5 and PBP3r (k+2/K=5 to 20 M-1 s-1) (Piras et al., 1990 ; El Kharroubi et al., 1991 ). These enterococcal PBPs, PBP2' in S. aureus (Song et al., 1987 ) and PBP3 in B. subtilis (Murray et al., 1996 ) form subgroup B1 of class B PBPs (Goffin & Ghuysen, 1998 ). With the exception of PBP3 of B. subtilis, which is not yet well characterized, the members of this subgroup B1 have a low affinity for ß-lactams and are thus involved in ß-lactam resistance. They all possess a 120–135 aa polypeptide inserted between the N-terminal hydrophobic anchoring peptide and the non-penicillin-binding module (Goffin & Ghuysen, 1998 ). The exact function of that polypeptide is still unknown. It could be essential for the folding and/or the functioning of the PBP in the cell as different partial deletions produced in it abolished the ß-lactam binding activity (Mollerach et al., 1996 ). It is however certainly not required for the ß-lactam binding on the folded PBP as tryptic fragments of membrane-bound PBP5 of E. hirae are still able to bind [14C]PenG (Piras et al., 1990 ; El Kharroubi et al., 1991 ).

A penicillin-resistant clone, JH2-2r (MIC 75 µg ml-1), was derived from the wild-type strain JH2-2 (MIC 5 µg ml-1) which apparently overproduced PBP4. The pbp4 genes from both strains were cloned and sequenced in this work. According to the pbp4 gene orientation, opposite to that of the three preceding genes (Fig. 2), it would be surprising if pbp4 did not possess its own promoter. One should also keep in mind that when cloned into pMCL210 and introduced into Escherichia coli, the pbp4 genes were able to direct the synthesis of large amounts of PBP4. Enterococcal genes are known to be transcribed in E. coli cells (Courvalin, 1994 ). As both pbp4 genes are identical even in their putative promoter region, one can exclude the possibility that the overproduction of PBP4 in JH2-2r is due to a modification of the strength of the pbp4 promoter and also rule out a possible modification of the PBP4 affinity for PenG.

The sequences of pbp5 of E. faecalis 56R (Signoretto et al., 1994 ) and pbp4 of JH2-2 are identical except in four positions. There are three additional nucleotides in pbp4 which modify the protein sequence over a short distance and introduce an additional amino acid residue. The peptide Ala-272-Cys-Ala-Ile-Asn-Arg-Val-Tyr-Gly-280 in PBP5 is replaced in PBP4 by the peptide Ala-272-Ala-Ala-Glu-Leu-Ile-Gly-Tyr-Thr-Gly-281. PBP4 of JH2-2 or JH2-2r are thus longer by one residue, with 680 aa. In addition, two residues both at the N-terminal end of the non-penicillin-binding module (Asn-13 vs Lys-13 and Val-29 vs Gly-29 in PBP5 and PBP4, respectively) and one immediately following Lys-424 in the conserved active-site motif SxxK (Ile-424 vs Thr-425 in PBP5 and PBP4, respectively) are different. This last modification, a threonine changed into an isoleucine, could affect the PBP5 affinity in the laboratory resistant E. faecalis strain 56R. As information concerning the parental strain 56 is not available, it is difficult to speculate on the importance of such a modification.

The genic environment of pbp4 was also examined in an attempt to localize a psr repressor gene that could perhaps be related to the overproduction of PBP4 in E. faecalis JH2-2r. Different genes, most of unknown functions, presenting similarities with B. subtilis genes were identified. None however showed similarities with the psr genes found in E. hirae and E. faecium strains (Ligozzi et al., 1993 ; Zorzi et al., 1996 ; Massidda et al., 1998 ). These genes coded for proteins different from those determined by the genes present in the vicinity of pbp5 in E. hirae (O. Dardenne, unpublished results).

A psr-like gene was identified in the complete genome of E. faecalis V583 by homology search in the TIGR database. It is present on contig 6330 whereas pbp4 is on contig 6237. A psr-like gene was cloned and sequenced from the DNA of both E. faecalis JH2-2 and JH2-2r strains. From the analysis of the genic environments of both pbp4 and psr-like genes of E. faecalis V583, JH2-2 and JH2-2r strains, one can conclude that both genes are several kilobases away from each other. This organization is very different from that in E. hirae and E. faecium, where both genes are adjacent (Ligozzi et al., 1993 ; Zorzi et al., 1996 ; Massidda et al., 1998 ).

The Psr-like proteins include the enterococcal Psr (Ligozzi et al., 1993 ; Zorzi et al., 1996 ; Massidda et al., 1998 ), the B. subtilis putative regulators LytR (Lazarevic et al., 1992 ) and YvhJ (Soldo et al., 1996 ) as well as proteins required for capsular polysaccharide synthesis, such as CpsA of Streptococcus pneumoniae (Morona et al., 1997 ), CpsA of Streptococcus thermophilus (Griffin et al., 1996 ) or EpsA of Streptococcus thermophilus Sfi6 (Stingele et al., 1996 ).

Since the sequences of the psr-like and the pbp4 genes in JH2-2 and JH2-2r are identical, even in their putative promoter regions, one can exclude the possibility that overproduction of PBP4 in JH2-2r is due to the Psr-like protein. Another regulatory mechanism that remains to be identified and investigated should be involved. The function(s) of the Psr-like protein also remain(s) to be elucidated.


   ACKNOWLEDGEMENTS
 
We acknowledge the Institute for Genomic Research (TIGR) for making preliminary sequence data available on the Internet at http://www.tigr.org.

This work was supported by the Belgian programme on Interuniversity Poles of Attraction initiated by the Belgian State, Prime Minister’s Office, Services fédéraux des affaires scientifiques, techniques et culturelles (PAI no. P4/03). C.D. is Chercheur qualifié of the Fonds National de la Recherche Scientifique (FNRS, Brussels). F.S. was a fellow of the Fonds pour la Formation à la Recherche dans l’Industrie et dans l’Agriculture (FRIA, Brussels). A.A. was a recipient of a fellowship from the European Community, Program ALFA, project 5.0111.9.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). Basic local alignment search tool. J Mol Biol 215, 403-410.[Medline]

Andersen, P. S., Martinussen, J. & Hammer, K. (1996). Sequence analysis and identification of the pyrKDbF operon from Lactococcus lactis including a novel gene, pyrK involved in pyrimidine biosynthesis. J Bacteriol 178, 5005-5012.[Abstract]

Courvalin, P. (1994). Transfer of antibiotic resistance genes between Gram-positive and Gram-negative bacteria. Antimicrob Agents Chemother 38, 1447-1451.[Medline]

Coyette, J., Ghuysen, J. M. & Fontana, R. (1978). Solubilization and isolation of the membrane-bound DD-carboxypeptidase of Streptococcus faecalis ATCC 9790. Eur J Biochem 88, 297-305.[Abstract]

Coyette, J., Ghuysen, J. M. & Fontana, R. (1980). The penicillin-binding proteins in Streptococcus faecalis ATCC 9790. Eur J Biochem 110, 445-456.[Abstract]

Devereux, J., Haeberli, P. & Smithies, O. (1984). A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 12, 387-395.[Abstract]

El Kharroubi, A., Jacques, P., Piras, G., Van Beeumen, J., Coyette, J. & Ghuysen, J. M. (1991). The Enterococcus hirae R40 penicillin-binding protein 5 and the methicillin-resistant Staphylococcus aureus penicillin-binding 2' are similar. Biochem J 280, 463-469.[Medline]

Fontana, R., Cerini, R., Longoni, P., Grossato, A. & Canepari, P. (1983). Identification of a streptococcal penicillin-binding protein that reacts very slowly with penicillin. J Bacteriol 155, 1343-1350.[Medline]

Fontana, R., Grossato, A., Rossi, L., Cheng, Y. R. & Satta, G. (1985). Transition from resistance to hypersusceptibility to ß-lactam antibiotics associated with loss of low affinity PBP in a Streptococcus faecium mutant highly resistant to penicillin. Antimicrob Agents Chemother 26, 678-683.

Fontana, R., Aldegheri, M., Ligozzi, M., Lopez, H., Sucari, A. & Satta, G. (1994). Overproduction of a low-affinity penicillin-binding protein and high-level ampicillin resistance in Enterococcus faecium. Antimicrob Agents Chemother 38, 1980-1983.[Abstract]

Ghuysen, J. M., Frère, J. M., Leyh-Bouille, M., Nguyen-Distèche, M. & Coyette, J. (1986). Active-site serine D-alanyl-D-alanine-cleaving peptidase-catalysed acyl-transfer reactions. Procedures for studying the penicillin-binding proteins of bacterial plasma membranes. Biochem J 235, 159-165.[Medline]

Goffin, C. & Ghuysen, J. M. (1998). Multimodular penicillin-binding proteins: an enigmatic family of orthologs and paralogs. Microbiol Mol Biol Rev 62, 1079-1093.[Abstract/Free Full Text]

Griffin, A. M., Morris, V. J. & Gasson, M. J. (1996). The cpsABCDE genes involved in polysaccharide production in Streptococcus salivarius spp. thermophilus NCBF2393. Gene 183, 23-27.[Medline]

Gutmann, L. (1994). Résistance des entérocoques aux bêta-lactamines et conséquences sur les synergies. Méd Mal Infect (spécial) 24, 165–171.

Hancock, L. E. & Gilmore, M. S. (2000). Pathogenicity of enterococci. In Gram-positive Pathogens , pp. 251-288. Edited by V. A. Fischetti, R. P. Novick, J. J. Ferretti, D. A. Portnoy & J. I. Road. Washington, DC:ASM Press.

Innis, M. A., Gelfand, D. H., Sninsky, J. J. & White, T. J. (1990). PCR protocols. A Guide to Methods and Applications. New York: Academic Press.

Jett, B. D., Huycke, M. M. & Gilmore, M. S. (1994). Virulence of enterococci. Clin Microbiol Rev 7, 462-478.[Abstract]

Klare, I., Rodloff, A. C., Wagner, J., Witte, W. & Hakenbeck, R. (1992). Overproduction of a penicillin-binding protein is not the only mechanism of penicillin resistance in Enterococcus faecium. Antimicrob Agents Chemother 36, 783-787.[Abstract]

Kunst, F., Ogasawara, N., Moszer, I. & 148 other authors (1997). The complete sequence of the Gram-positive bacterium Bacillus subtilis. Nature 390, 249–256.[Medline]

Laskey, R. A. (1980). The use of intensifying screens or organic scintillators for visualizing radioactive molecules, resolved by gel electrophoresis. Methods Enzymol 65, 363-371.[Medline]

Lazarevic, V., Margot, P., Soldo, B. & Karamata, D. (1992). Sequencing and analysis of the Bacillus subtilis lytRABC divergon: a regulatory unit encompassing the structural genes of the N-acetylmuramoyl-L-alanine amidase and its modifier. J Gen Microbiol 138, 1949-1961.[Medline]

Ligozzi, M., Pittaluga, F. & Fontana, R. (1993). Identification of a genetic element (psr) which negatively controls expression of Enterococcus hirae penicillin-binding protein 5. J Bacteriol 175, 2046-2051.[Abstract]

Ligozzi, M., Pittaluga, F. & Fontana, R. (1996). Modification of the penicillin-binding protein 5 associated with high-level ampicillin resistance of Enterococcus faecium. Antimicrob Agents Chemother 40, 354-357.[Abstract]

Lillehang, D., Nes, I. F. & Birkeland, N. K. (1997). A highly efficient and stable system for site-specific integration of genes and plasmids into the phage {phi}LC3 attachment site (attB) of the Lactococcus lactis chromosome. Gene 188, 129-136.[Medline]

Lindström, E. B., Boman, H. G. & Steele, B. B. (1970). Resistance of Escherichia coli to penicillin. VI. Purification and characterization of the chromosomally mediated penicillinase present in ampA-containing strains. J Bacteriol 101, 218-231.[Medline]

Loureiro Dos Santos, A. L. & Chopin, A. (1987). Shotgun cloning in Streptococcus lactis. FEMS Microbiol Lett 42, 209-212.

Mainardi, J. L., Billot-Klein, D., Coutrot, A., Legrand, R., Schoot, B. & Gutmann, L. (1998). Resistance to cefotaxime and peptidoglycan composition in Enterococcus faecalis are influenced by exogenous sodium chloride. Microbiology 144, 2679-2685.[Abstract]

Massidda, O., Dardenne, O., Whalen, M. B., Zorzi, W., Coyette, J., Shockman, G. D. & Daneo-Moore, L. (1998). The PBP5 synthesis repressor (psr) gene of Enterococcus hirae ATCC 9790 is substantially longer than previously reported. FEMS Microbiol Lett 166, 355-360.[Medline]

Masson, J. M. & Labia, R. (1983). Synthesis of a 125I-radiolabeled penicillin for penicillin-binding protein studies. Anal Biochem 128, 164-168.[Medline]

Mollerach, M., Partoune, P., Coyette, J. & Ghuysen, J. M. (1996). Importance of the E46-D160 polypeptide segment of the non-penicillin-binding module for the stability of the low-affinity, multimodular class B penicillin-binding protein 5 of Enterococcus hirae. J Bacteriol 178, 1774-1775.[Abstract]

Morona, J. K., Morona, R. & Paton, J. C. (1997). Characterization of the locus encoding the Streptococcus pneumoniae type 19F capsular polysaccharide biosynthetic pathway. Mol Microbiol 23, 751-753.[Medline]

Murray, B. E. (1990). The life and times of the Enterococcus. Clin Microbiol Rev 3, 46-65.[Medline]

Murray, T., Popham, D. L. & Setlow, P. (1996). Identification and characterization of pbpC, the gene encoding Bacillus subtilis penicillin-binding protein 3. J Bacteriol 178, 6001-6005.[Abstract]

Nakano, Y., Yoshida, Y., Yamashita, Y. & Koga, T. (1995). Construction of a series of pACYC-derived plasmid vectors. Gene 162, 157-158.[Medline]

Pearson, W. R. & Lipman, D. J. (1988). Improved tools for biological sequence analysis. Proc Natl Acad Sci USA 85, 2444-2448.[Abstract]

Piras, G., El Kharroubi, A., Van Beeumen, J., Coeme, E., Coyette, J. & Ghuysen, J. M. (1990). Characterization of an Enterococcus hirae penicillin-binding protein 3 with low penicillin affinity. J Bacteriol 172, 6856-6862.[Medline]

Piras, G., Raze, D., El Kharroubi, A., Hastir, D., Englebert, S., Coyette, J. & Ghuysen, J. M. (1993). Cloning and sequencing of the low affinity penicillin-binding protein 3r-encoding gene of Enterococcus hirae S185: modular design and structural organization of the protein. J Bacteriol 175, 2844-2852.[Abstract]

Rice, L. B., Carias, L. L., Hutton-Thomas, R., Sifaoui, F., Gutmann, L. & Rudin, S. D. (2001). Penicillin-binding protein 5 and expression of ampicillin resistance in Enterococcus faecium. Antimicrob Agents Chemother 45, 1480-1486.[Abstract/Free Full Text]

Schmit, J. L., Leclercq, R., Scheimberg, A. & Laudaner, D. I. (1994). Approche épidémiologique et clinique des entérocoques: résultats d’une enquête. Méd Mal Infect (spécial) 24, 141–148.

Signoretto, C., Boaretti, M. & Canepari, P. (1994). Cloning, sequencing and expression in Escherichia coli of the low affinity penicillin-binding protein of Enterococcus faecalis. FEMS Microbiol Lett 123, 99-106.[Medline]

Soldo, B., Lazarevic, V., Mauël, C. & Karamata, D. (1996). Sequence of the 305°–307° region of the Bacillus subtilis chromosome. Microbiology 142, 3079-3088.[Abstract]

Song, M. D., Wachi, M., Doi, M., Ishino, F. & Matsuhashi, M. (1987). Evolution of an inducible penicillin-target protein in methicillin-resistant Staphylococcus aureus. FEBS Lett 221, 167-171.[Medline]

Stingele, F., Neeser, J. R. & Mollet, B. (1996). Identification and characterization of the eps (exopolysaccharide) gene cluster from Streptococcus thermophilus Sfi6. J Bacteriol 178, 1680-1690.[Abstract]

Van Sinderen, D., Karsens, H., Kok, J., Terpstra, P., Ruiters, M. H., Venema, G. & Nautra, A. (1996). Sequence analysis and molecular characterization of the temperate lactococcal bacteriophage r1t. Mol Microbiol 19, 1343-1355.[Medline]

Williamson, R., Le Bouguénec, C., Gutmann, L. & Horaud, T. (1985). One or two low-affinity penicillin-binding proteins may be responsible for the range of susceptibility of Enterococcus faecium to benzylpenicillin. J Gen Microbiol 131, 1933-1940.[Medline]

Williamson, R., Gutmann, L., Horaud, T., Delbos, F. & Acar, J. F. (1986). Use of penicillin-binding proteins for identification of enterococci. J Gen Microbiol 132, 1929-1937.[Medline]

Zorzi, W., Zhou, X. Y., Dardenne, O., Lamotte, J., Raze, D., Pierre, J., Gutmann, L. & Coyette, J. (1996). Structure of the low-affinity penicillin-binding protein 5, PBP5fm, in wild-type and highly penicillin-resistant strains of Enterococcus faecium. J Bacteriol 178, 4948-4957.[Abstract]

Received 19 February 2001; revised 11 May 2001; accepted 18 May 2001.