Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2008 Mar 1.
Published in final edited form as: Plasmid. 2006 Sep 22;57(2):131–144. doi: 10.1016/j.plasmid.2006.08.003

Development of a host-genotype-independent counterselectable marker and a high-frequency conjugative delivery system and their use in genetic analysis of Enterococcus faecalis

Christopher J Kristich 1,*, Josephine R Chandler 1, Gary M Dunny 1
PMCID: PMC1852458  NIHMSID: NIHMS19867  PMID: 16996131

Abstract

Enterococcus faecalis is a Gram-positive commensal bacterium of the gastrointestinal tract. E. faecalis is also an opportunistic pathogen that frequently exhibits resistance to available antibiotics. Despite the clinical significance of the enterococci, genetic analysis has been restricted by limitations inherent in the available genetic tools. To facilitate genetic manipulation of E. faecalis, we developed a conjugative delivery system for high-frequency introduction of cloned DNA into target strains of E. faecalis and a host-genotype-independent counterselectable marker for use in markerless genetic exchange. We used these tools to construct a collection of E. faecalis mutant strains carrying defined mutations in several genes, including ccfA, eep, gelE, sprE, and an alternative sigma factor (sigH). Furthermore, we combined these mutations in various permutations to create double mutants, triple mutants, and a quadruple mutant of E. faecalis that enabled tests of epistasis to be conducted on the pheromone biosynthesis pathway. Analysis of cCF10 pheromone production by the mutants revealed that both the ccfA2 and Δeep10 mutations are epistatic to mutations in gelE/sprE. To our knowledge, this represents the first example of epistasis analysis applied to a chromosomally encoded biosynthetic pathway in enterococci. Thus, the advanced tools for genetic manipulation of E. faecalis reported here enable efficient and sophisticated genetic analysis of these important pathogens.

Keywords: counterselection, pheS, markerless exchange, epistasis, Enterococcus faecalis, p-chloro-phenylalanine, pheromone biosynthesis

Introduction

The bacterium Enterococcus faecalis is a Gram-positive member of the intestinal microflora (Tannock and Cook, 2002). As a frequent etiological agent of nosocomial infections (Richards et al., 2000), E. faecalis is also a leading health concern. E. faecalis exhibits relatively high-level intrinsic resistance to some antibiotics and freely shares antibiotic resistance determinants with neighboring bacteria on mobile genetic elements (Kak and Chow, 2002), leading to the emergence of multi-resistant clones in the hospital setting. Thus, treatment of enterococcal infections is an increasingly difficult problem for clinicians armed with currently available therapeutic agents. Effective development of new antibiotics to combat multi-resistant strains demands an intimate understanding of enterococcal physiology and genetics. Thus, improved techniques for sophisticated genetic analysis of enterococci are urgently needed.

A procedure known as allelic exchange (Maloy et al., 1996) or markerless exchange (Pritchett et al., 2004)) is often used to replace a wild-type allele of a gene of interest with a cloned allele that has been manipulated in vitro. Markerless exchange is typically performed using a two-step integration-segregation strategy based on homologous recombination. Although an antibiotic-resistance marker enables selection to be used at the first step, inclusion of a counterselectable marker in the exchange plasmid is required to permit selection, rather than laborious screening, at the second step of the process (Maloy et al., 1996).

Counterselection strategies function by inhibiting growth of bacterial clones that carry a counterselectable marker. Common examples include: inhibition of growth in the presence of sucrose, mediated by the sacB gene product (Ried and Collmer, 1987); inhibition of growth in the presence of lipophilic chelators such as fusaric acid, mediated by gene products conferring tetracycline-resistance (Maloy and Nunn, 1981); and inhibition of growth by purine or pyrimidine analogs, mediated by phosphoribosyl transferases of the purine and pyrimidine base salvage pathways (Spring et al., 1994; Fukagawa et al., 1999; Peck et al., 2000; Fabret et al., 2002; Bitan-Banin et al., 2003; Pritchett et al., 2004).

Genetic analysis of enterococci has thus far been restricted by limitations of the available genetic tools. Several approaches have been used for targeted mutagenesis of chromosomal genes in E. faecalis, each of which suffers from significant drawbacks. For example, Campbell-type homologous recombination of a non-replicating plasmid has been used to disrupt target genes (Qin et al., 1998). However, due to low frequencies of electroporation and recombination in E. faecalis, recombinants are difficult to obtain, and any such recombinants suffer from the problems inherent in this type of mutation (for example, mutants that carry antibiotic resistance markers, mutations that are polar, and reversion to the wild-type genotype by plasmid excision). Another approach to perform allelic exchange in E. faecalis relies on the use of the temperature-sensitive replicon, pG+host, to deliver substrate DNA for recombination (Arbeloa et al., 2004), but the pG+host-based approaches have met with limited success in our hands. A variation on this theme, originally developed for use in lactococci (Leenhouts et al., 1996), can be used in E. faecalis, but this system lacks the benefits provided by a functional counterselectable marker. Recently, we described a markerless exchange system that employs the first counterselectable marker functional in E. faecalis (Kristich et al., 2005); however, that strategy suffers from the drawback that the strain targeted for mutagenesis must also carry an independent mutation (Δupp) for counterselection to be applied. Thus, a Δupp mutant must be constructed in the target strain prior to manipulation of a gene of interest, and the Δupp mutation itself may have undesirable or unknown phenotypic effects.

For each of the approaches mentioned above, a significant difficulty lies in the first step of mutant construction: getting manipulated DNA into E. faecalis cells. E. faecalis is not known to be naturally competent for transformation, and although electroporation can be used, current electroporation protocols suffer from low efficiency of DNA uptake. As a result, recombinants can be difficult to isolate after electroporation with a non-replicating plasmid (typically, 0–5 recombinants per μg of plasmid DNA). Consequently, a means to introduce DNA into target cells at high frequency would substantially enhance mutant generation capabilities in E. faecalis.

To facilitate efficient genetic manipulation of E. faecalis, we developed 2 key improvements of our previous markerless genetic exchange system that overcome the aforementioned problems. First, a conjugative delivery system for high-frequency introduction of cloned DNA into target strains of E. faecalis, based on the enterococcal pheromone-responsive conjugative plasmid pCF10, surmounts the barrier to DNA introduction. Second, a host-genotype-independent counterselectable marker for use in markerless genetic exchange, based on the E. faecalis phenylalanyl-tRNA synthetase, eliminates the requirement for a specific pre-existing mutation in the target strain in order for counterselection to be applied. The effectiveness of these new tools was demonstrated by the construction of a panel of E. faecalis mutant strains carrying defined, stable mutations in several genes. These included an alternative sigma factor (sigH), potentially involved in regulating post-exponential phase gene expression, as well as several genes encoding products known to be involved in the synthesis, processing, or degradation of enterococcal peptide pheromones (ccfA, eep, gelE, and sprE). Furthermore, we combined these mutations in various permutations to create double mutants, triple mutants, and a quadruple mutant of E. faecalis that enabled tests of epistasis to be conducted on the pheromone biosynthesis pathway. The epistasis analysis revealed that, as expected, both the ccfA2 and Δeep10 mutations are epistatic to mutations in gelE/sprE. Thus, this system enables efficient markerless genetic exchange and sophisticated genetic analysis of E. faecalis. The approach described in this report should be of general use for enterococcal researchers and may be adapted for use in other Gram-positive organisms.

Materials and Methods

Bacterial strains, growth media, and chemicals

Bacterial strains used in this study are listed in Table 1. Bacteria were stored at −80°C in Todd-Hewitt broth or brain-heart infusion (THB or BHI, respectively; prepared according to the manufacturer’s instructions) supplemented with 30% glycerol. Unless otherwise indicated, all culture media were purchased from Difco and all chemicals were purchased from Sigma (St. Louis, MO). p-Cl-phenylalanine was obtained from Sigma as a mixture of the L- and D-isomers (Cat. C 6506), added to culture media prior to autoclaving, and mixed thoroughly immediately upon completion of the autoclave cycle. Bacto Agar was used as a solidifying agent for all semi-solid media. Casamino Acids were obtained from Fisher Biotech. MM9YEG medium (a variation of the previously described M9YE medium (Dunny and Clewell, 1975)), is a semi-defined M9-based medium supplemented with 0.25% yeast extract and 0.5% glucose (but prepared without MgSO4 or CaCl2). When required for selective growth of E. faecalis, erythromycin (Em), tetracycline (Tc), and chloramphenicol (Cm) were used at 10 μg/ml; rifampicin (Rf) at 200 μg/ml; fusidic acid (Fa) at 25 μg/ml; and spectinomycin (Sp) and streptomycin (Sm) at 1000 μg/ml. When required for E. faecalis, X-Gal was added at 250 μg/ml. When required for selective growth of Escherichia coli, erythromycin was used in BHI at 100 μg/ml. Oligonucleotides were purchased from Integrated DNA Technologies (Coralville, IA). DNA sequencing was performed at the Biomedical Genomics Center, University of Minnesota.

Table 1.

Bacterial strains and plasmids used in this study.

Strain or plasmid Relevant characteristic or description Reference or Source
Strains
E. coli
XL1Blue E. coli cloning host Lab stock
DH5α E. coli cloning host Lab stock
EC1000 E. coli cloning host, provides RepA in trans (Leenhouts et al., 1996)
E. faecalis
OG1 Reference strain (Gold et al., 1975)
OG1SSp Spontaneous mutant of OG1; SmR, SpR (Dunny et al., 1978)
OG1RF Spontaneous mutant of OG1; RfR, FaR (Dunny et al., 1978)
OG1Sp Spontaneous mutant of OG1; SpR This work
CK111 OG1Sp upp4::P23repA4 This work
CK117 OG1RF ΔsigH2 This work
JRC104 OG1RF ccfA2 This work
JRC105 OG1RF Δ(gelE-sprE)10 This work
JRC106 OG1RF Δeep10 This work
JRC107 JRC104 Δ(gelE-sprE)10 This work
JRC109 JRC105 Δeep10 This work
JRC110 JRC107 Δeep10 This work
Plasmids
pSK− Cloning vector Stratagene
pAM714 cAD1-inducible conjugative plasmid (pAD1-derivative); EmR (Ike and Clewell, 1984)
pCF10 cCF10-inducible conjugative plasmid (Dunny et al., 1981)
pCF10–101 pCF10 ΔoriT2 Staddon et al., in press
pORI280 Requires RepA in trans for replication; EmR (Leenhouts et al., 1996)
pWM401 Low-copy shuttle vector; CmR (Wirth et al., 1987)
pCJK2 Source of constitutive promoter (Kristich et al., 2005)
pCJK11 upp::erm cassette in temperature-sensitive shuttle vector; CmR (Kristich et al., 2005)
pCJK12 pCJK11 derivative carrying P23repA1 This work
pCJK41 pCJK11 derivative carrying P23repA4 This work
pCJK34 pORI280 derivative carrying oriTpCF10 This work
pCJK45 pWM401 derivative carrying P-pheS* This work
pCJK47 Conjugative donor plasmid, carries oriTpCF10 and P-pheS*; pORI280 derivative; EmR This work
pCJK52 carries ΔsigH2 This work
pCJK66 ΔsigH2 allele cloned into pCJK47 This work
pJRC104 ccfA2 allele cloned into pCJK47 This work
pJRC105 Δ(gelE-sprE)10 allele cloned into pCJK47 This work
pJRC106 Δeep10 allele cloned into pCJK47 This work
Open in a new tab

Plasmid construction

Plasmids used in this study are listed in Table 1. All plasmids derived from pORI280 were propagated in E. coli strain EC1000 (Leenhouts et al., 1996), which supplies RepA in trans to allow replication of these plasmids. When isolating plasmids from EC1000, we found that cultivation of recombinant strains at 30°C produced the best results. All other plasmids were propagated in E. coli XL1Blue or DH5α.

The counterselectable P-pheS* cassette was constructed using an overlap extension PCR strategy (Fig. 1), which introduced the pheS312AG mutation and fused a constitutive promoter upstream of the E. faecalis pheS312AG allele. First, the synthetic constitutive promoter driving expression of upp in pCJK2 (Kristich et al., 2005), which was originally derived from pFW11 (Podbielski et al., 1996), was amplified by PCR. The pheS gene from E. faecalis OG1RF was amplified from chromosomal DNA as 2 fragments, using primers designed according to the sequence of the E. faecalis V583 pheS gene (EF1115), available at The Institute for Genomic Research website (http://www.tigr.org). The A312G mutation responsible for relaxed substrate specificity was encoded by the primers annealing internal to the pheS gene. These 3 amplicons were used as templates with the outside primers for a second PCR step, fusing the amplicons together and creating the intact P-pheS* cassette (Fig. 1). This cassette was cloned into the low-copy number shuttle plasmid pWM401 (Wirth et al., 1987) using primer-encoded SphI/XbaI restriction sites, creating pCJK45 (Fig. 1). The nucleotide sequence of the P-pheS* cassette in pCJK45 was determined to verify its integrity.

Fig. 1. Strategy for construction of plasmids used in this study.

Fig. 1

Open in a new tab

See Materials and Methods for details. The asterisks indicate the pheSA312G mutation. P, synthetic constitutive promoter. “sigH up” and “sigH down” represent the chromosomal segments upstream and downstream of sigH, respectively. Open arrows indicate PCR amplifications, solid arrows indicate cloning steps. Plasmids for markerless exchange of other genes were similarly constructed by cloning into pCJK47, as described in Materials and Methods. Unique restriction sites are shown. Note that the plasmids are not shown to scale.

A mobilizable and counterselectable donor plasmid (pCJK47) was constructed in two steps (Fig. 1). First, a 243-bp fragment containing the origin of transfer (oriT) from pCF10 (Staddon et al., in press) was amplified by PCR and cloned between the AclI sites of the repA− plasmid, pORI280 (Leenhouts et al., 1996), using primer-encoded AclI restriction sites, thereby creating pCJK34. Second, the P-pheS* cassette was excised from pCJK45 by digestion with SphI/BglII and cloned into similarly-digested pCJK34, creating pCJK47.

A donor plasmid used to create an in-frame deletion of E. faecalis sigH (pCJK66; Fig. 1) was constructed by first amplifying fragments of OG1RF chromosomal DNA flanking the sigH gene to create the upstream and downstream PCR amplicons. The resulting amplicons contained 995 and 949 bp of E. faecalis DNA, respectively. These amplicons included flanking sequences as well as sequences encoding the first 5 residues and the last 5 residues of the sigH ORF, respectively, which were retained in the deletion construct in an effort to avoid any unanticipated effects on the expression of adjacent genes. The 2 PCR amplicons were sequentially cloned into pSK− to create pCJK52, using the primer-encoded NotI/SpeI restriction sites for the upstream fragment and the primer-encoded SpeI/XmaI restriction sites for the downstream fragment. In the final deletion construct, the SpeI site forms the fusion point between the two flanking fragments and is in-frame with the few remaining codons of the sigH ORF, such that 2 SpeI-encoded amino acids are inserted into the 10-amino acid product that is produced from the ΔsigH2 allele. Over 94% of the sigH gene was eliminated in this construct. The ΔsigH2 allele was excised from pCJK52 by digestion with NotI/XmaI and cloned into similarly-digested pCJK47, creating pCJK66.

A donor plasmid used to create an in-frame deletion of the adjacent E. faecalis genes gelE and sprE (pJRC105; Table 1) was constructed according to the procedure described above for pCJK66, with modifications. The upstream and downstream PCR amplicons contained 585 and 586 bp of E. faecalis DNA, respectively, and included sequences encoding the first 33 residues of GelE and the last 25 residues of SprE, which were retained in the deletion construct in an effort to avoid any unanticipated effects on the expression of adjacent genes. The “internal” end of each amplicon (i.e., the ends destined to be fused together to create the deletion) encoded a BsaI restriction site, designed such that BsaI cleavage and ligation of the amplicons would seamlessly fuse the amplicons together creating an in-frame deletion (i.e., no restriction sites or other extra nucleotides are found between the two amplicons in the fusion construct) according to a previously described procedure (Derre et al., 1999). To create the deletion allele, the upstream amplicon was digested at the primer-encoded XbaI/BsaI sites, the downstream amplicon was digested at the primer-encoded NcoI/BsaI sites, and the two amplicons were simultaneously cloned into XbaI/NcoI-digested pCJK47, creating pJRC105. Over 93% of the gelE and sprE coding sequences were eliminated in our construct.

A donor plasmid used to create an in-frame deletion of E. faecalis eep (pJRC106; Table 1) was constructed according to the procedure described above for pJRC105 with slight modifications. The upstream and downstream PCR amplicons contained 838 and 1036 bp of E. faecalis DNA, respectively, and included sequences encoding the first 15 residues and the last 14 residues of Eep, which were retained in the deletion construct in an effort to avoid any unanticipated effects on the expression of adjacent genes. The amplicons were fused together and cloned into pCJK47 as described for pJRC105, except that the XbaI/NcoI restriction sites used for cloning are naturally present in the adjacent DNA sequence and therefore were not primer-encoded. Over 93% of the eep gene was eliminated in our construct.

A donor plasmid used to introduce point mutations encoding stop codons in E. faecalis ccfA (pJRC104; Table 1) was constructed by subcloning the 2.37-kb NcoI/SpeI fragment encoding the ccfA2 allele from pJRC102 (Chandler et al., 2005) into NcoI/XbaI-digested pCJK47, creating pJRC104.

Plasmids to transfer P23-repA alleles to the upp locus of the E. faecalis chromosome (pCJK12 and pCJK41) were created as follows. First, the P23-repA1 allele was amplified by PCR from the chromosome of E. coli EC1000. This allele encodes a fusion of the lactococcal P23 promoter (van der Vossen et al., 1987) with the pWVO1 repA gene, preceded by its natural ribosome-binding site (Leenhouts et al., 1991). Primer-encoded XhoI restriction sites were used to clone P23-repA1 between the 2 XhoI sites found in the temperature-sensitive plasmid pCJK11 (Kristich et al., 2005), thereby replacing the erm determinant (found in the pCJK11-encoded upp::erm cassette) with P23-repA1 (creating pCJK12). In this construct, the P23-repA1 allele separates two fragments of E. faecalis chromosomal DNA that flank the upp gene, effectively creating a gene replacement allele in which most of the upp ORF is replaced by P23-repA1. A plasmid bearing a modified ribosome-binding site upstream of repA (pCJK41, encoding RBS of AGGAGC) was created using the QuikChange mutagenesis strategy (Stratagene) with pCJK12 as the template for mutagenesis. The nucleotide sequence of the entire P23-repA allele was determined for each plasmid to verify the presence of the desired mutation.

Transformation of E. faecalis by plasmid DNA was carried out via electroporation of lysozyme-treated cells as previously described (Bae et al., 2002).

Construction of donor strains for conjugative delivery of markerless exchange plasmids

A spectinomycin-resistant derivative of E. faecalis OG1 (OG1Sp) was isolated by selecting for a spontaneous mutant of OG1 after plating on THB agar supplemented with 1000 μg/ml Sp at 37°C.

To serve as E. faecalis host strains for repA-dependent donor plasmids, we constructed a strain bearing a chromosomally-borne repA allele whose expression is driven by the lactococcal P23 promoter (strain CK111; Table 1). Our strategy was to integrate the P23-repA allele into the E. faecalis chromosome at the upp locus, thereby creating a stable gene replacement mutant with a selectable phenotype (upp mutants of E. faecalis are resistant to 5-fluorouracil; (Kristich et al., 2005)). This E. faecalis mutants was constructed using the temperature-sensitive plasmid, pCJK41. OG1Sp carrying pCJK41 was cultured overnight at 30°C in M9YE supplemented with 0.5% glucose and Cm (for maintenance of the plasmid). The cells were washed twice and diluted ten-fold in M9YE supplemented with 0.5% glucose, followed by incubation at 30°C for 1 h. The culture was shifted to the nonpermissive temperature of 42°C for 4 h, and aliquots were plated on BHI agar supplemented with 1 mM 5-fluorouracil at 42°C to select for upp mutants. Colonies that arose were screened using a previously described colony-PCR procedure (Kristich et al., 2005) to identify isolates carrying the P23-repA allele in the chromosome.

High-frequency conjugative delivery and markerless exchange to construct a ΔsigH2 mutant

The donor plasmid, pCJK66, was introduced into the conjugative donor strain, CK111(pCF10–101), by electroporation. Replication of pCJK66 as an episome in the donor strain was verified by re-isolating the recombinant plasmid using a Qiagen miniprep kit (not shown). The resulting transformant, CK111(pCF10–101, pCJK66) was cultivated overnight in BHI supplemented with Em and Sp at 37°C. The recipient strain (E. faecalis OG1RF) was cultivated overnight in BHI at 37°C. Both strains were washed twice in BHI, diluted ten-fold in fresh BHI (no antibiotics), and incubated separately at 37°C for 1 h. Donor and recipient cultures were mixed at a ratio of 1:9, followed by centrifugation of 0.8 ml of the resulting mixture. The cell pellet was resuspended in ~0.1 ml of supernatant and spread on a 100-mm BHI agar plate to permit conjugation. After ~5 h of incubation at 37°C, bacteria were recovered from the surface in 2 ml of PBS supplemented with 2 mM EDTA (to disrupt bacterial aggregates). Serial dilutions were made in PBS and enumerated on BHI agar supplemented with Sp, Tc, and Em (to enumerate donors) and BHI agar supplemented with Rf, Fa, Em, and X-Gal (to enumerate transconjugants harboring integrated copies of pCJK66). Blue transconjugant colonies arose within 24–36 h at 37°C at a frequency of approximately 6 x10−5 per donor. These recombinants exhibited a severe growth defect relative to OG1RF when patched to MM9YEG agar supplemented with 10 mM p-Cl-Phe due to the presence of the P-pheS* marker in the backbone of the integrated plasmid. Because it is formally possible that pCJK66 could integrate into the recipient chromosome by homologous recombination at the pheS locus (via the plasmid-encoded P-pheS* allele), a PCR assay was performed on chromosomal DNA obtained from colony-purified transconjugants using flanking primers external to the cloned pheS gene. Of 6 isolates tested, all yielded an amplicon corresponding in size to the expected wild-type product, indicating that plasmid integration had not occurred at the pheS locus. A similar PCR assay performed with primers external to the cloned sigH locus failed to produce a product from the 6 mutants, indicating that plasmid insertion had occurred at this locus.

To isolate secondary recombinants in which the integrated plasmid had been excised and lost, several primary recombinants were cultured in BHI lacking Em for approximately 20 generations at 37°C. Serial dilutions of the cell suspension were plated on MM9YEG agar supplemented with 10 mM p-Cl-Phe and X-Gal at 37°C. White colonies arose within ~16–20 h of incubation at a frequency of approximately 5 x 10−4 per viable CFU. Patching to Em-containing agar confirmed that the white colonies were also sensitive to Em, as expected if the integrated plasmid had been excised.

As outlined in Fig. 4, the plasmid-free recombinants could carry either the wild-type allele or the mutant allele. To identify recombinants carrying the chromosomal sigH deletion, 18 white p-Cl-PheR colonies were analyzed using a PCR assay to determine if they carried the wild-type or mutant allele at the target locus. Chromosomal DNA prepared from colony-purified isolates (with the DNEasy Tissue Kit, Qiagen) was used as the source of template DNA. Primers for these amplifications annealed to flanking DNA sequences external to the duplicated (cloned) chromosomal regions. Following electrophoresis on agarose gels, the wild-type and deleted alleles could be differentiated on the basis of the size of the amplicon, revealing 14 recombinants that retained the wild-type sigH locus and 4 recombinants that carried the ΔsigH2 deletion. One such ΔsigH2 recombinant (strain CK117) was retained for phenotypic analysis.

Fig. 4. Analysis of conjugative plasmid transfer into E. faecalis mutant strains.

Fig. 4

Open in a new tab

The ability of a panel of E. faecalis strains to serve as recipients for pheromone-responsive conjugative plasmid transfer was determined in a 2-h liquid conjugation assay for (A) the cCF10-specific plasmid pCF10, or (B) the cAD1-specific plasmid pAM714. The data are reported as the ratio of transconjugant CFU/donor CFU (TC/donor) for 2 independent experiments performed on separate days. Strains tested as recipients are: 1, OG1RF (wild-type); 2, CK117 (ΔsigH2); 3, JRC104 (ccfA2); 4, JRC105 (Δ(gelE-sprE)10); 5, JRC106 (Δeep10); 6, JRC107 (ccfA2 Δ(gelE-sprE)10); 7, JRC109 (Δeep10 Δ(gelE-sprE)10); 8, JRC110 (ccfA2 Δ(gelE-sprE)10 Δeep10).

Markerless exchange to construct other E. faecalis mutants

To construct the remaining single, double, triple, and quadruple mutants in E. faecalis, a procedure analogous to that described above for deletion of sigH was used, with appropriate combinations of donors and recipients. In some cases, the growth conditions prior to conjugation or the screening approach used to differentiate wild-type recombinants from those carrying the desired mutant alleles were slightly modified, based on the anticipated mutant phenotypes. For example, JRC104 candidates were initially screened using a conjugation assay to measure the ability of the candidates to serve as recipients for pCF10 conjugative transfer, because previous work demonstrated that ccfA2 mutants exhibited a defect in such an assay (Chandler et al., 2005). Mutants carrying the Δeep10 allele were initially screened using a conjugation assay to measure the ability of the candidates to serve as recipients for pAM714 conjugative transfer, because previous work demonstrated that an eep mutant exhibited a reduction in cAD1 pheromone production (An et al., 1999). Mutants carrying the Δ(gelE-sprE)10 allele were routinely screened on THB agar supplemented with 3% gelatin to identify clones lacking the characteristic halo indicative of GelE activity (Waters et al., 2003). Triple and quadruple mutants were constructed using an iterative approach, beginning with different combinations of conjugative donors and recipients. The presence of all mutant alleles was verified after each iteration during the construction of multiply-mutant strains. When strains exhibiting reduced pheromone production were used as recipients (ccfA2, Δeep10), the pCF10–101 conjugative machinery was induced in the donors by the addition of synthetic cCF10 pheromone (BioMedical Genomics Center, University of Minnesota) at 10 ng/ml during the 1-h growth period prior to conjugation.

Determination of cCF10 pheromone activity in culture supernatants

Strains to be tested were cultivated overnight in THB at 37°C (~15 h). The OD600 of the overnight cultures was determined for each, and equivalent amounts of biomass were harvested by centrifugation at 8,000 rpm (Beckman J2–21 centrifuge, JA20 rotor) for 10 min. The supernatant from each sample was retained and evaluated with an aggregation assay to measure the biological activity of cCF10 (Buttaro et al., 2000). Briefly, twofold serial dilutions of supernatant samples were made in THB medium in a round-bottom microtiter plate. Ten microliters of a 15-h old OG1RF(pCF10) indicator culture (cultivated in THB at 37°C) was added to each well. Samples were incubated with shaking at 37°C for 2 hours, and wells in which aggregation of the indicator strain occurred were scored. The titer is reported as the reciprocal of the highest supernatant dilution that induced aggregation of the indicator strain. In cases where no aggregation was observed at any dilution, the results are reported as 0.

Analysis of conjugative plasmid transfer

To determine the extent of conjugative transfer of pheromone-responsive plasmids pCF10 and pAM714 into plasmid-free recipients, strains OG1SSp(pCF10) or OG1SSp(pAM714) were used as conjugative donors. Overnight cultures of donors and recipients were cultivated at 37ºC in THB for 15 h and diluted ten-fold into THB. These subcultures were incubated at 37°C for 1 h, followed by mixing of donors and recipients at a ratio of 1:9. Conjugation occurred during an additional incubation at 37ºC for 2 hours before serial dilutions were prepared for enumeration on selective media. Transconjugants were enumerated on THB agar supplemented with Tc and Rf (for pCF10-containing transconjugants) or Em and Rf (for pAM714-containing transconjugants). Donors were enumerated on THB agar supplemented with Sm. Measurements are reported as the ratio of transconjugants/donor (TC/donor) and represent means from duplicate assays.

Results

A genotype-independent counterselectable marker for E. faecalis

To circumvent the problems associated with our previous counterselection system (see Introduction), we sought to develop a counterselection strategy for E. faecalis that did not require the presence of a particular mutation in the genome of the target strain in order to be functional.

In E. coli, a missense mutation in the phenylalanyl tRNA synthetase α-subunit (pheS294AG) confers relaxed substrate specificity, such that the mutant enzyme can aminoacylate tRNAPhe not only with phenylalanine but also with halogenated phenylalanine derivatives, including p-chloro-phenylalanine (p-Cl-Phe) (Kast and Hennecke, 1991; Ibba et al., 1994). Cultivation of E. coli clones expressing pheS294AG on medium amended with p-Cl-Phe results in inhibition of growth, presumably as a consequence of the production of non-functional proteins in which Phe residues have been replaced with p-Cl-Phe. Clones lacking pheS294AG are capable of normal growth in the presence of p-Cl-Phe (Kast and Hennecke, 1991; Kast, 1994). Thus, the growth properties of E. coli strains expressing the pheS294AG allele provide the basis for a counterselection strategy.

To determine if an analogous strategy was applicable to E. faecalis, we compared the sequence of pheS from E. faecalis V583 (GenBank Accession: AAO80915) with that of E. coli (GenBank Accession: NP_416229) using the BLAST-P algorithm at the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/BLAST/). Inspection of the sequence alignment revealed substantial sequence identity in the region surrounding A294 of the E. coli pheS with the corresponding region of the E. faecalis homolog, suggesting that A312 of the E. faecalis pheS was the functionally equivalent residue. We cloned pheS from E. faecalis OG1RF into a low-copy number plasmid and introduced the analogous A312G mutation, creating pCJK45 (Fig. 1). Because the natural promoter driving expression of pheS in E. faecalis had not been identified, we included a synthetic constitutive promoter (originally derived from plasmid pFW11 (Podbielski et al., 1996)) upstream of pheS312AG in the construct, creating the P-pheS* cassette (Fig. 1). Analysis of pCJK45-containing cells (Fig. 2) revealed that E. faecalis carrying pheS312AG grew normally in the absence of p-Cl-Phe but was substantially inhibited in the presence of 10 mM p-Cl-Phe, whereas the growth of E. faecalis carrying the parental plasmid was not inhibited under equivalent conditions. Thus, the P-pheS* cassette confers a dominant growth defect on otherwise wild-type E. faecalis in the presence of p-Cl-Phe, indicating that the P-pheS* cassette can form the basis of a host-genotype-independent counterselectable marker for E. faecalis.

Fig. 2. Conditional growth inhibition mediated by P-pheS* in E. faecalis.

Fig. 2

Open in a new tab

Wild-type E. faecalis OG1RF carrying either an empty shuttle vector (pWM401) or a derivative carrying the P-pheS* cassette (pCJK45) were cultivated overnight in BHI supplemented with Cm at 37°C. Bacteria were washed twice with PBS and resuspended in PBS to the original volume. After preparing ten-fold serial dilutions in PBS, aliquots (5 μl) of successive dilutions were spotted from left to right on MM9YEG agar supplemented with Cm and the indicated concentration of p-Cl-Phe. The plates were incubated at 37°C for 24 h and photographed using dark-field illumination.

The composition of the growth medium was found to influence P-pheS*-mediated growth inhibition by p-Cl-Phe. For example, cultivation of E. faecalis carrying P-pheS* on BHI, THB, or tryptic soy agar amended with p-Cl-Phe did not inhibit growth nearly as significantly as did cultivation on MM9YEG (not shown). We hypothesize that the presence of excess Phe in these relatively rich types of media (either free Phe or derived from another source, such as Phe-containing peptides) is able to effectively compete with p-Cl-Phe either during transport or as a substrate for the phenylalanyl-tRNA synthetase, thereby preventing p-Cl-Phe from being incorporated into protein at levels sufficient to inhibit growth. Consistent with this hypothesis, addition of 1% Casamino Acids to MM9YEG attenuates the growth-inhibitory effect of p-Cl-Phe (not shown).

Construction of a mobilizable plasmid and conjugative delivery system for efficient markerless exchange in E. faecalis

Enterococci are widely known for their proficiency in the use of conjugation as a means of exchanging mobile genetic elements. For example, enterococcal pheromone-responsive conjugative plasmids such as pCF10 (Dunny et al., 1981) can easily transfer by conjugation to plasmid-free recipient cells at remarkably high frequencies of 10−1 transconjugants/donor. Recently, pCF10 was found to mediate the conjugative mobilization, in trans, of heterologous shuttle plasmids carrying a cloned pCF10 origin of transfer (oriT) at similarly high frequencies (Staddon et al., in press). Furthermore, a mutant derivative of pCF10 carrying a deletion of its oriT (pCF10–101) could not transfer itself to recipient cells at detectable frequencies in liquid matings, but retained the ability to mobilize heterologous oriTpCF10-containing shuttle plasmids at high frequency (Staddon et al., in press). Therefore, we developed a system to use pCF10–101 for efficient delivery of conditionally-replicating donor plasmids to target cells of E. faecalis for markerless genetic exchange.

The donor plasmid for our system is a derivative of pORI280 (Law et al., 1995; Leenhouts et al., 1996), which requires the replication protein RepA in trans to replicate. We sequentially introduced a cloned copy of the pCF10 oriT (conferring pCF10–101-dependent mobilization) and the P-pheS* cassette (providing the counterselectable marker), thereby creating the donor plasmid pCJK47 (Fig. 1). Most of the unique restriction sites found in pORI280 remain in pCJK47 and can be used to clone target DNA into the donor plasmid. To serve as a conjugative donor strain, we constructed an E. faecalis strain carrying pCF10–101 that could support replication of pCJK47 derivatives. Our strategy was to integrate a constitutively expressed allele of repA into the E. faecalis chromosome at the upp locus (see Materials and Methods for details), creating strain CK111. Subsequently, pCF10–101 was introduced to yield the conjugative donor strain.

Construction of E. faecalis mutant strains

To evaluate the effectiveness of our improved system for efficient markerless exchange in E. faecalis, we constructed a panel of mutant strains according to the procedure described in Materials and Methods (see Fig. 3 for an overview of the process). Because E. faecalis is known to grow well under stressful conditions, we initially searched the E. faecalis V583 genome for genes encoding alternative sigma factors that might regulate adaptation to stressful conditions, such as sigH of Bacillus subtilis. Using B. subtilis sigH as a query (GenBank Accession: CAB11874), we identified TIGR locus EF0049 as a sigH homolog and constructed an in-frame deletion mutant of E. faecalis lacking this gene. We also constructed a panel of strains carrying mutations in several other genes whose products participate in enterococcal pheromone production or degradation. In particular, we constructed mutations in ccfA, which encodes the cCF10 pheromone precursor (Antiporta and Dunny, 2002); eep, whose product is required for production of the cCF10 and cAD1 pheromones (An et al., 1999) and whose product is thought to be a membrane-bound protease (Brown et al., 2000) involved in proteolytic processing of pheromone precursors; and gelE and sprE, which encode extracellular proteases known to degrade secreted pheromone (Waters et al., 2003). Previously, each of these genes had been independently inactivated by other means. However, the corresponding mutants each suffer from significant disadvantages due to the inherent limitations of the approaches used for inactivation of these genes. To circumvent the associated problems, we constructed stable, unmarked mutations that inactivated each of the genes in an isogenic, wild-type background (strains JRC104, JRC105, and JRC106).

Fig. 3. Efficient markerless genetic exchange strategy used to create a deletion of sigH in E. faecalis.

Fig. 3

Open in a new tab

See Materials and Methods for a detailed description of the procedure. The conjugative donor strain (left) contains a mutant pCF10 (pCF10–101) lacking its oriT. The donor supports replication of mobilizable repA− plasmids due to expression of a chromosomal copy of repA. Conjugative transfer functions supplied by pCF10–101 mediated the high-frequency mobilization of pCJK66 into the repA− recipient, OG1RF (right). Homologous recombination of pCJK66 occurred at a region of cloned chromosomal DNA (Step 1), generating an erythromycin-resistant recombinant that is sensitive to growth inhibition mediated by p-Cl-Phe due to constitutive expression of pheS*. The sigH deletion is encoded by the pCJK66-borne allele. In Step 2, plasmid excision by recombination occurred during nonselective growth, leaving either the wild-type allele (A) or the mutant allele (B) on the chromosome. The excised plasmid cannot replicate and was lost by segregation. Recombinants from Step 2 were selected by plating cells on MM9YEG agar supplemented with 10 mM p-Cl-Phe and X-Gal and screened by PCR to differentiate those carrying the wild-type sigH allele from those carrying the deletion. Cloned chromosomal DNA fragments flanking sigH on the upstream (up) or downstream (down) sides are indicated as white rectangles juxtaposed to the ΔsigH allele. An equivalent sequence of events can be drawn for plasmid integration at either the upstream or downstream fragment.

One important advantage of markerless exchange is the ability to efficiently construct strains carrying multiple mutations, thereby enabling sophisticated genetic analysis such as tests of epistasis. To demonstrate this, we used an iterative process to combine the ccfA2, Δeep10 and Δ(gelE-sprE)10 mutant alleles with each other in various permutations to form triple and quadruple mutant strains (see Table 2 for a complete list). Previously, strains that simultaneously carried mutations in more than 1 of these 3 loci were not available, precluding epistasis analysis of the current model for pheromone biosynthesis and processing. In summary, our markerless genetic exchange system enabled the efficient construction of a variety of stable, unmarked deletion mutants and point mutants, as well as strains carrying combinations of multiple mutations.

Table 2.

Determination of cCF10 pheromone activity in culture supernatants.

Strain Relevant genotype cCF10 activity* Expt 1 cCF10 activity* Expt 2
OG1RF Wild-type 0 0
CK117 ΔsigH2 0 0
JRC104 ccfA2 0 0
JRC105 Δ(gelE-sprE)10 16 8
JRC106 Δeep10 0 0
JRC107 ccfA2 Δ(gelE-sprE)10 0 0
JRC109 Δeep10 Δ(gelE-sprE)10 2 1
JRC110 ccfA2 Δ(gelE-sprE)10 Δeep10 0 0
Open in a new tab
*

cCF10 pheromone activity in stationary-phase culture supernatants was determined using a biological indicator assay, as described in Methods, for 2 independent experiments performed on separate days. cCF10 activity is reported as the reciprocal of the highest supernatant dilution that induced aggregation of the indicator strain. Supernatant samples that did not induce aggregation at any dilution are reported as 0.

During preliminary experiments analyzing conjugative mobilization of pCJK66 into OG1RF, we found that unselected chromosomal markers from the donor strain could sometimes be transferred into the recipient along with pCJK66. The mechanism of chromosomal co-transfer was not investigated, but this phenomenon requires that any transconjugants constructed by this approach are screened to ensure that donor-specific chromosomal markers (e.g., spectinomycin resistance, upp4::P23repA4) have not been co-transferred with the donor plasmid. In the case of pCJK66-containing transconjugant derivatives of OG1RF, we screened for sensitivity to spectinomycin and 5-FU, each of which are properties of the recipient strain but not the donor strain.

Phenotypic analyses of E. faecalis mutant strains

The properties of all mutants constructed in this study by markerless exchange technology were compared with wild-type E. faecalis OG1RF using several phenotypic assays. Growth rates of all mutants, as measured by changes in optical density at 600 nm, were indistinguishable from that of OG1RF in THB at 37°C (not shown). Cultivation of the mutants on THB agar supplemented with 3% gelatin (a substrate for the secreted GelE protease) overnight at 37°C revealed the characteristic halo indicative of GelE activity (Waters et al., 2003) for all strains that are gelE+ (OG1RF, CK117, JRC104, JRC106) but no halo for strains carrying the Δ(gelE-sprE)10 allele (JRC105, JRC107, JRC109, JRC110), as expected (not shown).

Given the known role for B. subtilis sigH in mediating the transition to stationary phase (Britton et al., 2002), we evaluated growth kinetics of our sigH2 mutant in several rich and semi-defined media, but were unable to identify a phenotypic defect. We also failed to identify a defect for the sigH2 mutant in a number of other assays, including sensitivity to a panel of antibiotics, biofilm formation in a microtiter-plate assay, adherence to collagen-coated surfaces, and ability to serve as a donor or recipient for conjugative plasmid transfer (Fig. 4).

Production of cCF10 pheromone by the panel of mutant strains was evaluated with a biological activity assay (Buttaro et al., 2000), using stationary-phase culture supernatants as the source of pheromone (Table 2). Because E. faecalis OG1RF produces pheromone-degrading extracellular proteases, cCF10 pheromone produced by OG1RF is not detectable under the conditions used for this assay. Previously, analysis of pheromone production by a mutant carrying an insertion in gelE demonstrated that cCF10 accumulation was substantially increased, revealing a key role for GelE in cCF10 degradation (Waters et al., 2003). Consistent with those results, our markerless Δ(gelE-sprE)10 mutant also exhibited enhanced cCF10 accumulation (Table 2, JRC105). Tests of epistasis conducted by introducing additional mutations in genes known to encode the pheromone or to be required for its processing (the ccfA2 or Δeep10 mutations introduced in the triple mutants JRC107 and JRC109) show that the phenotypic effects of the Δ(gelE-sprE)10 mutation is either completely or partially abrogated if pheromone biosynthesis is impaired (Table 2). That the ccfA2 and Δeep10 alleles are epistatic to the Δ(gelE-sprE)10 allele suggests that both ccfA and eep act upstream of gelE/sprE in the pathway determining the level of cCF10 activity in culture supernatants, consistent with current understanding (Chandler and Dunny, 2004). Thus, our improved genetic system allowed us to perform genetic tests of the current model of cCF10 biogenesis.

Pheromone production can also be assessed by monitoring the ability of a strain to serve as a recipient for pheromone-responsive conjugative plasmid transfer. Therefore, the mutants were evaluated as recipients for 2 distinct conjugative plasmids: the cCF10-specific plasmid pCF10 and the cAD1-specific plasmid pAM714 (Fig. 4). As expected, the ccfA2 mutant (JRC104) is defective as a recipient for pCF10 but not for pAM714, whereas the Δeep10 mutant is defective as a recipient for both plasmids. Although plasmid transfer frequencies have not previously been reported for the original eep insertion mutant, the latter phenotype is consistent with its deficiency in both cCF10 and cAD1 pheromone production (An et al., 1999). We also note that the 2 triple mutants (JRC107 and JRC109), each of which exhibited a reduction in cCF10 accumulation in culture supernatants (Table 2), also exhibit a reduction in pCF10 recipient ability relative to the otherwise isogenic double mutant (JRC105), confirming the biological relevance of the pheromone accumulation results.

Discussion

Genetic analysis of enterococci has thus far been limited by inefficient or non-existent tools for genetic manipulation. We now report the development of a substantially improved system for efficient markerless genetic exchange in E. faecalis that surmounts some of the previous limitations. By exploiting the enterococcal pheromone-responsive conjugative plasmid pCF10 for use as an efficient DNA delivery machine, we have overcome a significant barrier to genetic manipulation of E. faecalis – transferring DNA that has been manipulated in vitro back into target cells to serve as a substrate for recombination. Enterococci are not known to be naturally competent for transformation, and current electroporation protocols are inefficient, especially when introducing non-replicating plasmids that must undergo homologous recombination to be retained. However, the enterococcal pheromone-responsive plasmids have evolved a remarkably efficient ability to deliver DNA into neighboring cells, rendering these plasmids ideal tools for genetic manipulation. Furthermore, pCF10 has recently been shown to mobilize heterologous oriTpCF10-containing plasmids across species boundaries into Lactococcus lactis and Streptococcus agalactiae (Staddon et al., in press), suggesting that our conjugative delivery system may facilitate genetic manipulation of other Gram-positive bacteria for which no efficient means of DNA introduction currently exists. In such applications, we expect that the conjugative E. faecalis donor strains will require induction by exogenous cCF10 pheromone to enable efficient delivery.

During the construction of the mutants reported in this paper, we observed that chromosomal markers present in the conjugative donor can sometimes be co-transferred to recipients along with the donor plasmid. The mechanism of this co-transfer has not been investigated. However, the frequency of co-transfer of particular chromosomal alleles varies depending on the identity of the chromosomal segment cloned into pCJK47; thus we speculate that chromosomal co-transfer may result from an Hfr-type mechanism following transient integration (by homologous recombination) of the pCJK47-derivative into the chromosome of the conjugative donor strain. In any case, the extent of co-transfer has never been so great as to preclude the isolation of the desired mutant. The issue of co-transfer likely will be irrelevant if this system is used to mobilize DNA into heterologous species, where homologous recombination cannot occur.

We recently described the first counterselectable marker functional in E. faecalis (Kristich et al., 2005) and have used it to construct a number of interesting mutants. However, one drawback of the previous counterselection strategy is that the host strain must carry a mutation in the chromosomal upp gene for counterselection to be applied. The “second generation” P-pheS* counterselectable marker described in this report eliminates any such genotypic requirement, as the P-pheS* allele is dominant over the wild-type pheS allele. Therefore, counterselection can be applied in any strain, regardless of its genotype. Preliminary evidence indicates that this P-pheS* cassette can also inhibit growth of Staphylococcus aureus and Lactococcus lactis cultivated in the presence of p-Cl-Phe, suggesting that our P-pheS* counterselection strategy can be applied to facilitate markerless exchange in a variety of low-GC Gram positive bacteria for which no counterselectable marker is currently available.

In summary, we have developed a substantially improved system for efficient markerless genetic exchange in E. faecalis and used this system to rapidly construct a panel of single, double, triple, and quadruple mutant strains carrying unmarked in-frame deletions and point mutations. Of note, our system enabled tests of epistasis to be applied, for the first time, to a chromosomally encoded biosynthetic pathway - the enterococcal pheromone biosynthesis pathway. Thus, this technology enables the kind of sophisticated genetic analysis that is required to develop a thorough understanding of the biology of E. faecalis. It seems likely that these tools can also be successfully applied for genetic manipulation of other Gram-positive bacteria for which efficient genetic tools are currently lacking.

Acknowledgments

This work was supported by grants AI58134 and GM49530 from the NIH to GMD. CJK was supported by NRSA fellowship F32-AI56684 from the NIH. JRC is a predoctoral trainee supported by grant T32DE07288 from the NIH. We thank Tim Leonard for photography and technical advice, and Barry Wanner for bringing the work of Kast and colleagues on E. coli pheS to our attention.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. An FY, Sulavik MC, Clewell DB. Identification and characterization of a determinant (eep) on the Enterococcus faecalis chromosome that is involved in production of the peptide sex pheromone cAD1. J Bacteriol. 1999;181:5915–21. doi: 10.1128/jb.181.19.5915-5921.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Antiporta MH, Dunny GM. ccfA, the genetic determinant for the cCF10 peptide pheromone in Enterococcus faecalis OG1RF. J Bacteriol. 2002;184:1155–62. doi: 10.1128/jb.184.4.1155-1162.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Arbeloa A, Segal H, Hugonnet JE, Josseaume N, Dubost L, Brouard JP, Gutmann L, Mengin-Lecreulx D, Arthur M. Role of class A penicillin-binding proteins in PBP5-mediated beta-lactam resistance in Enterococcus faecalis. J Bacteriol. 2004;186:1221–8. doi: 10.1128/JB.186.5.1221-1228.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bae T, Kozlowicz B, Dunny GM. Two targets in pCF10 DNA for PrgX binding: their role in production of Qa and prgX mRNA and in regulation of pheromone-inducible conjugation. J Mol Biol. 2002;315:995–1007. doi: 10.1006/jmbi.2001.5294. [DOI] [PubMed] [Google Scholar]
  5. Bitan-Banin G, Ortenberg R, Mevarech M. Development of a gene knockout system for the halophilic archaeon Haloferax volcanii by use of the pyrE gene. J Bacteriol. 2003;185:772–8. doi: 10.1128/JB.185.3.772-778.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Britton RA, Eichenberger P, Gonzalez-Pastor JE, Fawcett P, Monson R, Losick R, Grossman AD. Genome-wide analysis of the stationary-phase sigma factor (sigma-H) regulon of Bacillus subtilis. J Bacteriol. 2002;184:4881–90. doi: 10.1128/JB.184.17.4881-4890.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Brown MS, Ye J, Rawson RB, Goldstein JL. Regulated intramembrane proteolysis: a control mechanism conserved from bacteria to humans. Cell. 2000;100:391–8. doi: 10.1016/s0092-8674(00)80675-3. [DOI] [PubMed] [Google Scholar]
  8. Buttaro BA, Antiporta MH, Dunny GM. Cell-associated pheromone peptide (cCF10) production and pheromone inhibition in Enterococcus faecalis. J Bacteriol. 2000;182:4926–33. doi: 10.1128/jb.182.17.4926-4933.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Chandler JR, Dunny GM. Enterococcal peptide sex pheromones: synthesis and control of biological activity. Peptides. 2004;25:1377–88. doi: 10.1016/j.peptides.2003.10.020. [DOI] [PubMed] [Google Scholar]
  10. Chandler JR, Hirt H, Dunny GM. A paracrine peptide sex pheromone also acts as an autocrine signal to induce plasmid transfer and virulence factor expression in vivo. Proc Natl Acad Sci USA. 2005;102:15617–22. doi: 10.1073/pnas.0505545102. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Derre I, Rapoport G, Msadek T. CtsR, a novel regulator of stress and heat shock response, controls clp and molecular chaperone gene expression in gram-positive bacteria. Mol Microbiol. 1999;31:117–31. doi: 10.1046/j.1365-2958.1999.01152.x. [DOI] [PubMed] [Google Scholar]
  12. Dunny G, Funk C, Adsit J. Direct stimulation of the transfer of antibiotic resistance by sex pheromones in Streptococcus faecalis. Plasmid. 1981;6:270–8. doi: 10.1016/0147-619x(81)90035-4. [DOI] [PubMed] [Google Scholar]
  13. Dunny GM, Brown BL, Clewell DB. Induced cell aggregation and mating in Streptococcus faecalis: evidence for a bacterial sex pheromone. Proc Natl Acad Sci USA. 1978;75:3479–83. doi: 10.1073/pnas.75.7.3479. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dunny GM, Clewell DB. Transmissible toxin (hemolysin) plasmid in Streptococcus faecalis and its mobilization of a noninfectious drug resistance plasmid. J Bacteriol. 1975;124:784–90. doi: 10.1128/jb.124.2.784-790.1975. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Fabret C, Ehrlich SD, Noirot P. A new mutation delivery system for genome-scale approaches in Bacillus subtilis. Mol Microbiol. 2002;46:25–36. doi: 10.1046/j.1365-2958.2002.03140.x. [DOI] [PubMed] [Google Scholar]
  16. Fukagawa T, Hayward N, Yang J, Azzalin C, Griffin D, Stewart AF, Brown W. The chicken HPRT gene: a counter selectable marker for the DT40 cell line. Nucleic Acids Res. 1999;27:1966–9. doi: 10.1093/nar/27.9.1966. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gold OG, Jordan HV, van Houte J. The prevalence of enterococci in the human mouth and their pathogenicity in animal models. Arch Oral Biol. 1975;20:473–7. doi: 10.1016/0003-9969(75)90236-8. [DOI] [PubMed] [Google Scholar]
  18. Ibba M, Kast P, Hennecke H. Substrate specificity is determined by amino acid binding pocket size in Escherichia coli phenylalanyl-tRNA synthetase. Biochemistry. 1994;33:7107–12. doi: 10.1021/bi00189a013. [DOI] [PubMed] [Google Scholar]
  19. Ike Y, Clewell DB. Genetic analysis of the pAD1 pheromone response in Streptococcus faecalis, using transposon Tn917 as an insertional mutagen. J Bacteriol. 1984;158:777–83. doi: 10.1128/jb.158.3.777-783.1984. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kak V, Chow JW. Acquired antibiotic resistances in enterococci. In: Gilmore MS, Clewell DB, Courvalin P, Dunny GM, Murray BE, Rice LB, editors. The Enterococci: Pathogenesis, Molecular Biology, and Antibiotic Resistance. American Society for Microbiology Press; Washington, D.C: 2002. pp. 355–383. [Google Scholar]
  21. Kast P. pKSS--a second-generation general purpose cloning vector for efficient positive selection of recombinant clones. Gene. 1994;138:109–14. doi: 10.1016/0378-1119(94)90790-0. [DOI] [PubMed] [Google Scholar]
  22. Kast P, Hennecke H. Amino acid substrate specificity of Escherichia coli phenylalanyl-tRNA synthetase altered by distinct mutations. J Mol Biol. 1991;222:99–124. doi: 10.1016/0022-2836(91)90740-w. [DOI] [PubMed] [Google Scholar]
  23. Kristich CJ, Manias DA, Dunny GM. Development of a method for markerless genetic exchange in Enterococcus faecalis and its use in construction of a srtA mutant. Appl Environ Microbiol. 2005;71:5837–5849. doi: 10.1128/AEM.71.10.5837-5849.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Law J, Buist G, Haandrikman A, Kok J, Venema G, Leenhouts K. A system to generate chromosomal mutations in Lactococcus lactis which allows fast analysis of targeted genes. J Bacteriol. 1995;177:7011–8. doi: 10.1128/jb.177.24.7011-7018.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Leenhouts K, Buist G, Bolhuis A, ten Berge A, Kiel J, Mierau I, Dabrowska M, Venema G, Kok J. A general system for generating unlabelled gene replacements in bacterial chromosomes. Mol Gen Genet. 1996;253:217–24. doi: 10.1007/s004380050315. [DOI] [PubMed] [Google Scholar]
  26. Leenhouts KJ, Tolner B, Bron S, Kok J, Venema G, Seegers JF. Nucleotide sequence and characterization of the broad-host-range lactococcal plasmid pWVO1. Plasmid. 1991;26:55–66. doi: 10.1016/0147-619x(91)90036-v. [DOI] [PubMed] [Google Scholar]
  27. Maloy SR, Nunn WD. Selection for loss of tetracycline resistance by Escherichia coli. J Bacteriol. 1981;145:1110–1. doi: 10.1128/jb.145.2.1110-1111.1981. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Maloy SR, Stewart VJ, Taylor RK. Genetic Analysis of Pathogenic Bacteria. A Laboratory Manual. Cold Spring Harbor Laboratory Press; Cold Spring Harbor, NY: 1996. [Google Scholar]
  29. Peck RF, Dassarma S, Krebs MP. Homologous gene knockout in the archaeon Halobacterium salinarum with ura3 as a counterselectable marker. Mol Microbiol. 2000;35:667–76. doi: 10.1046/j.1365-2958.2000.01739.x. [DOI] [PubMed] [Google Scholar]
  30. Podbielski A, Spellerberg B, Woischnik M, Pohl B, Lutticken R. Novel series of plasmid vectors for gene inactivation and expression analysis in group A streptococci (GAS) Gene. 1996;177:137–47. doi: 10.1016/0378-1119(96)84178-3. [DOI] [PubMed] [Google Scholar]
  31. Pritchett MA, Zhang JK, Metcalf WW. Development of a markerless genetic exchange method for Methanosarcina acetivorans C2A and its use in construction of new genetic tools for methanogenic archaea. Appl Environ Microbiol. 2004;70:1425–33. doi: 10.1128/AEM.70.3.1425-1433.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Qin X, Singh KV, Xu Y, Weinstock GM, Murray BE. Effect of disruption of a gene encoding an autolysin of Enterococcus faecalis OG1RF. Antimicrob Agents Chemother. 1998;42:2883–8. doi: 10.1128/aac.42.11.2883. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Richards MJ, Edwards JR, Culver DH, Gaynes RP. Nosocomial infections in combined medical-surgical intensive care units in the United States. Infect Control Hosp Epidemiol. 2000;21:510–5. doi: 10.1086/501795. [DOI] [PubMed] [Google Scholar]
  34. Ried JL, Collmer A. An nptI-sacB-sacR cartridge for constructing directed, unmarked mutations in gram-negative bacteria by marker exchange-eviction mutagenesis. Gene. 1987;57:239–46. doi: 10.1016/0378-1119(87)90127-2. [DOI] [PubMed] [Google Scholar]
  35. Spring KJ, Mattick JS, Don RH. Escherichia coli gpt as a positive and negative selectable marker in embryonal stem cells. Biochim Biophys Acta. 1994;1218:158–62. doi: 10.1016/0167-4781(94)90005-1. [DOI] [PubMed] [Google Scholar]
  36. Tannock GW, Cook G. Enterococci as Members of the Intestinal Microflora of Humans. In: Gilmore MS, Clewell DB, Courvalin P, Dunny GM, Murray BE, Rice LB, editors. The Enterococci: Pathogenesis, Molecular Biology, and Antibiotic Resistance. American Society for Microbiology Press; Washington, D.C: 2002. pp. 101–132. [Google Scholar]
  37. van der Vossen JM, van der Lelie D, Venema G. Isolation and characterization of Streptococcus cremoris Wg2-specific promoters. Appl Environ Microbiol. 1987;53:2452–7. doi: 10.1128/aem.53.10.2452-2457.1987. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Waters CM, Antiporta MH, Murray BE, Dunny GM. Role of the Enterococcus faecalis GelE protease in determination of cellular chain length, supernatant pheromone levels, and degradation of fibrin and misfolded surface proteins. J Bacteriol. 2003;185:3613–23. doi: 10.1128/JB.185.12.3613-3623.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Wirth R, An F, Clewell DB. Highly efficient cloning system for Streptococcus faecalis protoplast transformation, shuttle vectors, and applications. In: Ferretti JJ, Curtiss RI, editors. Streptococcal Genetics. American Society for Microbiology Press; Washington, D. C: 1987. pp. 25–27. [Google Scholar]

RESOURCES