Self-transferable plasmids determining the hemolysin and bacteriocin of Streptococcus faecalis var. zymogenes.

Strains of Streptococcus faecalis var. zymogenes, designated JH1 and JH3, produced a hemolysin and a bacteriocin. Hemolytic activity was lost from a low percentage of cells grown in broth at either 37 or 45 C. All nonhemolytic (Hly-) variants had lost bacteriocin activity (Ben-), and those from strain JH3 had also lost resistance to the bacteriocin (Bnr-). The majority of Hly-, Ben- variants from JH1 retained bacteriocin resistance (Bnrplus). Strains JH1 and JH3 contained a plasmid deoxyribonucleic acid species of molecular weight 38 times 10-6 (plasmids pJH2 and pJH3, respectively), and strain JH1 also contained a 50 times 10-6 molecular weight plasmid (pJH1) which has previously been shown to carry the genes determining resistance to the antibiotics kanamycin, neomycin, streptomycin, erythromycin, and tetracycline. Hly-, Bcn-, Bnr- variants of strain JH3 had completely lost plasmid pJH3. Hly-, Bcn-, Bnr- variants of strain JH1 had completely lost plasmid pJH2 and retained plasmid pJH1, but Hly-, Bcn-, Bnrplus variants had retained both plasmids pJH2 and pJH1. The Hlyplus, Bcnplus, Bnrplus traits from both parental strains were transferable to nonhemolytic S. faecalis strains during mixed incubation in broth at 37 C, and hemolytic recipient strains were found to have received plasmid pJH2 from strain JH1 and pJH3 from JH3. We conclude that the Hlyplus, Bnrplus traits are borne on plasmid pJH2 in strain JH1 and pJH3 in strain JH3 and that, in Hly-, Bcn-, Bnrplus variants of strain JH1, plasmic pJH2 has suffered a mutation affecting hemolysin and bacteriocin expression. We infer that the plasmids transfer by conjugation. Beta-hemolytic activity is the only property distinguishing the zymogenes variety from S. faecalis. Since we have shown that this activity is plasmid borne in strains JH1 and JH3, we endorse the view that the varietal status of zymogenes should be dropped.     

Conjugal transfer of Tn916, Tn916 delta E, and pAM beta 1 from Enterococcus faecalis to Butyrivibrio fibrisolvens strains.

Anaerobic filter matings of Butyrivibrio fibrisolvens H17c, CF3, D1, or GS113, representing different DNA relatedness groups, were done with Enterococcus faecalis CG110, which contains chromosomally inserted Tn916. Tetracycline-resistant transconjugants were obtained with each mating pair at average frequencies of 4.4 x 10(-6) (per recipient) and 5.2 x 10(-6) (per donor). The transfer frequencies of Tn916 into B. fibrisolvens varied 5- to 10-fold with mating time, strain, and growth stage. By using Southern hybridization with pAM120 as the probe, Tn916 was shown to insert at one or more separate chromosomal sites for each strain of B. fibrisolvens. Retransfer of Tn916 from B. fibrisolvens H17c or CF3 to E. faecalis OG1-X or JH 2-2 or to B. fibrisolvens D1 or GS113 could not be shown. Matings of E. faecalis RH110, which contains chromosomally inserted Tn916 delta E, with B. fibrisolvens 49, H17c, D1, CF3, GS113, or VV-1 resulted in erythromycin-resistant transconjugants at average frequencies of 5.3 x 10(-7) (per recipient) and 2.5 x 10(-7) (per donor). Tn916 delta E was shown by Southern hybridization with pAM120 to insert at one or more sites in the chromosome of each strain. B. fibrisolvens H17c was anaerobically filter mated with E. faecalis JH 2-SS, which contains pAM beta 1. Erythromycin-resistant transconjugants were obtained at frequencies of 2 x 10(-5) (per recipient) and 6 x 10(-5) (per donor). The presence of pAM beta 1 in these transconjugants could not be shown by agarose gel electrophoresis of plasmid minilysates but could be shown by Southern hybridization analysis.(ABSTRACT TRUNCATED AT 250 WORDS)     

Conjugal transfer of Tn916, Tn916 delta E, and pAM beta 1 from Enterococcus faecalis to Butyrivibrio fibrisolvens strains.

Anaerobic filter matings of Butyrivibrio fibrisolvens H17c, CF3, D1, or GS113, representing different DNA relatedness groups, were done with Enterococcus faecalis CG110, which contains chromosomally inserted Tn916. Tetracycline-resistant transconjugants were obtained with each mating pair at average frequencies of 4.4 x 10(-6) (per recipient) and 5.2 x 10(-6) (per donor). The transfer frequencies of Tn916 into B. fibrisolvens varied 5- to 10-fold with mating time, strain, and growth stage. By using Southern hybridization with pAM120 as the probe, Tn916 was shown to insert at one or more separate chromosomal sites for each strain of B. fibrisolvens. Retransfer of Tn916 from B. fibrisolvens H17c or CF3 to E. faecalis OG1-X or JH 2-2 or to B. fibrisolvens D1 or GS113 could not be shown. Matings of E. faecalis RH110, which contains chromosomally inserted Tn916 delta E, with B. fibrisolvens 49, H17c, D1, CF3, GS113, or VV-1 resulted in erythromycin-resistant transconjugants at average frequencies of 5.3 x 10(-7) (per recipient) and 2.5 x 10(-7) (per donor). Tn916 delta E was shown by Southern hybridization with pAM120 to insert at one or more sites in the chromosome of each strain. B. fibrisolvens H17c was anaerobically filter mated with E. faecalis JH 2-SS, which contains pAM beta 1. Erythromycin-resistant transconjugants were obtained at frequencies of 2 x 10(-5) (per recipient) and 6 x 10(-5) (per donor). The presence of pAM beta 1 in these transconjugants could not be shown by agarose gel electrophoresis of plasmid minilysates but could be shown by Southern hybridization analysis.(ABSTRACT TRUNCATED AT 250 WORDS)     

In vivo transfer of pAMβ1 from Lactobacillus reuteri to Enterococcus faecalis

Trials were conducted to determine the in vivo transferability of plasmid-mediated antibiotic resistance between two strains of enteric Gram-positive bacteria. Germfree mice were associated with the donor Lactobacillus reuteri DSM 20016 strain, carrying the broad host range pAMβ1 plasmid, and with the Enterococcus faecalis JH2SS recipient strain.
Analysis of faecal content of associated mice demonstrated that the in vivo transfer of this plasmid did occur and that frequencies of conjugation were affected by the presence of subtherapeutic levels of antibiotic in the diet.     

In vivo transfer of pAM beta 1 from Lactobacillus reuteri to Enterococcus faecalis

Trials were conducted to determine the in vivo transferability of plasmid-mediated antibiotic resistance between two strains of enteric Gram-positive bacteria. Germ-free mice were associated with the donor Lactobacillus reuteri DSM 20016 strain, carrying the broad host range pAM beta 1 plasmid, and with the Enterococcus faecalis JH2SS recipient strain. Analysis of faecal content of associated mice demonstrated that the in vivo transfer of this plasmid did occur and that frequencies of conjugation were affected by the presence of subtherapeutic levels of antibiotic in the diet.     

Horizontal transfer of virulence genes encoded on the Enterococcus faecalis pathogenicity island

Enterococcus faecalis, a leading cause of nosocomial antibiotic resistant infections, frequently possesses a 150 kb pathogenicity island (PAI) that carries virulence determinants. The presence of excisionase and integrase genes, conjugative functions and multiple insertion sequence elements suggests that the PAI, or segments thereof, might be capable of horizontal transfer. In this report, the transfer of the E. faecalis PAI is demonstrated and a mechanism for transfer elucidated. In filter matings, chloramphenicol resistance was observed to transfer from strain MMH594b, a clinical isolate possessing the PAI tagged with a cat marker, to OG1RF (pCGC) with a frequency of 3.2 x 10(-10) per donor. Secondary transfer from primary transconjugant TCRFB1 to strain JH2SS in filter and broth matings occurred with a frequency of 1 and 2 x 10(-1) per donor respectively. Analysis of the transconjugants demonstrated that a 27,744 bp internal PAI segment was capable of excision and circularization in the donor, and is mobilized as a cointegrate with a pTEF1-like plasmid. High-frequency transfer also occurred from TCRFB1 to JH2SS during transient colonization of the mouse gastrointestinal tract. This is the first demonstration of the horizontal transfer of PAI-encoded virulence determinants in E. faecalis and has implications for genome evolution and diversity.     

Horizontal transfer of tet(M) and erm(B) resistance plasmids from food strains of Lactobacillus plantarum to Enterococcus faecalis JH2-2 in the gastrointestinal tract of gnotobiotic rats

Two wild-type strains of Lactobacillus plantarum previously isolated from fermented dry sausages were analysed for their ability to transfer antibiotic resistance plasmids in the gastrointestinal tract. For this purpose, we used gnotobiotic rats as an in vivo model. Rats were initially inoculated with the recipient Enterococcus faecalis JH2-2 at a concentration of 10(10) CFU mL(-1). After a week, either of the two donors L. plantarum DG 522 (harbouring a tet(M)-containing plasmid of c. 40 kb) or L. plantarum DG 507 [harbouring a tet(M)-containing plasmid of c. 10 kb and an erm(B)-containing plasmid of c. 8.5 kb] was introduced at concentrations in the range of 10(8)-10(10) CFU mL(-1). Two days after donor introduction, the first transconjugants (TCs) were detected in faecal samples. The detected numbers of tet(M)-TCs were comparable for the two donors. In both cases, this number increased to c. 5 x 10(2) CFU g(-1) faeces towards the end of the experiment. For erm(B)-TCs, the number was significantly higher and increased to c. 10(3) CFU g(-1) faeces. To our knowledge, this is the first study showing in vivo transfer of wild-type antibiotic resistance plasmids from L. plantarum to E. faecalis.     

Evidence for a chromosome-borne resistance transposon (Tn916) in Streptococcus faecalis that is capable of "conjugal" transfer in the absence of a conjugative plasmid

Streptococcus faecalis strain DS16 harbors the conjugative hemolysin-bacteriocin plasmid pAD1 (35 megadaltons) and the nonconjugative R-plasmid pAD2 determining resistance to streptomycin, kanamycin, and erythromycin; a tetracycline resistance (Tetr) determinant is located on the chromosome. When strain DS16 was mated (on membrane filters) with the plasmid-free strain JH2-2, Tetr transconjugants could be obtained at a frequency of about 10(-6) per recipient. Analyses of transconjugants showed that some contained the Tetr determinant linked to pAD1. Subsequent studies showed that the Tetr determinant was located on a 10-megaldalton transposon, designated Tn916, which could insert into two hemolysin plasmids: pAM gamma 1 and pOB1. In addition, derivatives of DS16 devoid of pAD1 were capable of transferring Tetr to recipient strains. Transconjugants (plasmid-free) from such matings could subsequently act as donors in the transfer of Tetr. Both transposition and transfer were found to be rec independent.     

Insertion of Tn916 into Bacillus pumilus plasmid pMGD302 and evidence for plasmid transfer by conjugation.

As part of an effort to develop systems for genetic analysis of strains of Bacillus pumilus which are being used as a microbial hay preservative, we introduced the conjugative Enterococcus faecalis transposon Tn916 into B. pumilus ATCC 1 and two naturally occurring hay isolates of B. pumilus. B. pumilus transconjugants resistant to tetracycline were detected at a frequency of approximately 6.5 x 10(-7) per recipient after filter mating with E. faecalis CG110. Southern hybridization confirmed the insertion of Tn916 into several different sites in the B. pumilus chromosome. Transfer of Tn916 also was observed between strains of B. pumilus in filter matings, and one donor strain transferred tetracycline resistance to recipients in broth matings at high frequency (up to 3.4 x 10(-5) per recipient). Transfer from this donor strain in broth matings was DNase-resistant and was not mediated by culture filtrates. Transconjugants from these broth matings contained derivatives of a cryptic plasmid (pMGD302, approx 60 kb) from the donor strain with Tn916 inserted at various sites. The plasmids containing Tn916 insertions transferred to a B. pumilus recipient strain at frequencies of approx 5 x 10(-6) per recipient. This evidence suggests that pMGD302 can transfer by a process resembling conjugation between strains of B. pumilus.     

Isolation of a nontransmissible antibiotic resistance plasmid by transductional shortening of R factor RP1.

A plasmid segregant carrying tetracycline and carbenicillin resistance markers has been isolated from R factor RP1 by transductional shortening with phage P22. The new plasmid RP1-S2, which has a molecular weight of 23 times 10-6, has lost the transfer, phage sensitivity, and neomycin resistance functions of RP1. It combines readily with a W group plasmid, R388, to form a transmissible carbenicillin and trimethoprim resistance plasmid, RWP1.     

Isolation of plasmid deoxyribonucleic acid from two strains of Bacteroides.

Two clinical isolates of Bacteroides contained covalently closed circular deoxyribonucleic acid (DNA) as shown by sedimentation in an alkaline sucrose gradient, CsCl ethidium bromide equilibrium centrifugation, and electron microscopy. Bacteriodes fragilis N1175 contained a homogeneous species of plasmid DNA with a molecular weight of 25 x 10(6). Bacteroides ochraceus 2228 contained two distinct, covalently closed circular DNA elements. The larger cosedimented with the covalently closed circular DNA form of the R plasmid, R100, corresponding to a molecular weight of 70 x 10(6); the smaller sedimented as a 58S molecule with a calculated molecular weight of 25 x 10(6). The roles of these plasmids are unknown. Neither strain transferred antibiotic resistance to plasmid-negative Bacteroides or Escherichia coli, and neither produced bacteriocins active against other Bacteroides or sensitive indicator strains of E. coli.     

Transfer and expression of degradative and antibiotic resistance plasmids in acidophilic bacteria.

The genetic accessibility of selected acidophilic bacteria was investigated to evaluate their applicability to degrading pollutants in acidic environments. The IncP1 antibiotic resistance plasmids RP4 and pVK101 and the phenol degradation-encoding plasmid pPGH11 were transferred from neutrophilic bacteria into the extreme acidophilic eubacterium Acidiphilium cryptum at frequencies of 1.8 x 10(-2) to 9.8 x 10(-4) transconjugants per recipient cell. The IncQ antibiotic resistance plasmid pSUP106 was mobilizable to A. cryptum by triparental matings at a frequency of 10(-5) transconjugants per recipient cell. In the transconjugants, antibiotic resistances and the ability to degrade phenol were expressed. A. cryptum AC6 (pPGH11) grew with 2.5 mM phenol at a doubling time of 12 h and a yield of 0.52 g (dry cell weight) per g of phenol. A. cryptum harbored five native plasmids of 255 to 6.3 kb in size. Plasmids RP4 and pVK101 were transferred from Escherichia coli into Acidobacterium capsulatum at frequencies of 10(-3) and 2.3 x 10(-4) and to the facultative autotroph Thiobacillus acidophilus at frequencies of 1.1 x 10(-5) and 2.9 x 10(-6) transconjugants per recipient cell, respectively. Plasmid pPGH11 could not be transferred into the latter strains. T. acidophilus wild type contained six so far cryptic plasmids of 220 to 5 kb.     

Genetic, molecular, and functional analysis of Streptococcus faecalis R plasmid pJH1.

Streptococcus faecalis JH1 contains two conjugative plasmids, pJH1, an R plasmid that codes for resistance to kanamycin, streptomycin, erythromycin, and tetracycline, and pJH2, a hemolysin-bacteriocin plasmid. Strain JH1 was used as an antibiotic resistance donor in conjugation experiments with two plasmid-free S. faecalis recipient strains, JH2-2 and OG1-RF1. Plasmid pJH1 was purified from one transconjugant, DL77, and subjected to restriction endonuclease analyses. Five restriction enzymes, EcoRI, XbaI, BamHI, SalI, and XhoI, yielding 10, 9, 3, 2, and 2 fragments, respectively, were used to determine the size (80.7 kilobases) of pJH1 and to construct a restriction endonuclease map of the plasmid. Twenty-eight percent of the antibiotic-resistant transconjugants examined expressed only part of the resistance pattern (Kmr Smr Emr Tcr) associated with pJH1, that is, they were resistant to kanamycin, streptomycin, and erythromycin; to erythromycin and tetracycline; or to erythromycin or to tetracycline only. Most of these strains also produced hemolysin and bacteriocin, and several contained a hybrid plasmid consisting of pJH2 and specific segments of pJH1 DNA. Several of these hybrid plasmids, as well as a deletion derivative of pJH1 that coded for resistance to tetracycline but not to kanamycin, streptomycin, or erythromycin, were purified and used to confirm the arrangement of restriction endonuclease fragments on the pJH1 map and to locate the resistance determinants on this map.     

Transfer of the pheromone-inducible plasmid pCF10 among Enterococcus faecalis microorganisms colonizing the intestine of mini-pigs.

A new animal model, the streptomycin-treated mini-pig, was developed in order to allow colonization of defined strains of Enterococcus faecalis in numbers sufficient to study plasmid transfer. Transfer of the pheromone-inducible pCF10 plasmid between streptomycin-resistant strains of E. faecalis OG1 was investigated in the model. The plasmid encodes resistance to tetracycline. Numbers of recipient, donor, and transconjugant bacteria were monitored by selective plating of fecal samples, and transconjugants were subsequently verified by PCR. After being ingested by the mini-pigs, the recipient strain persisted in the intestine at levels between 10(6) and 10(7) CFU per g of feces throughout the experiment. The donor strain, which carried different resistance markers but was otherwise chromosomally isogenic to the recipient strain, was given to the pigs 3 weeks after the recipient strain. The donor cells were initially present in high numbers (10(6) CFU per g) in feces, but they did not persist in the intestine at detectable levels. Immediately after introduction of the donor bacteria, transconjugant cells appeared and persisted in fecal samples at levels between 10(3) and 10(4) CFU per g until the end of the experiment. These observations showed that even in the absence of selective tetracycline pressure, plasmid pCF10 was transferred from ingested E. faecalis cells to other E. faecalis organisms already present in the intestinal environment and that the plasmid subsequently persisted in the intestine.     

Transformation of Pseudomonas aeruginosa strain PAO with bacteriophage and plasmid DNA.

A procedure has been developed which allows transformation of P. aeruginosa strain PAO with plasmid and bacteriophage DNA at a frequency of 10(-6) per recipient cell. The method is similar in outline to that developed for Escherichia coli. It involves growing the recipient cells to 3-5 x 10(8) per ml in nutrient broth, washing the cells with 0.1 M MgCl2, resuspending in 0.175 M CaCl2 for 20 min, exposing to DNA for 1 h and then heat pulsing at 42 degrees C for 1 min. Some plasmid markers are expressed immediately, whereas others require time for phenotypic expression.     

In situ studies with membrane diffusion chambers of antibiotic resistance transfer in Escherichia coli.

Coliform bacteria were isolated from raw sewage and sewage effluent-receiving waters and tested for their antibiotic susceptibility patterns and their ability to transfer antibiotic resistance to Escherichia coli K-12 C600. An environmental isolate of E. coli (MA527) capable of transferring antibiotic resistance to C600 was mated, both in vitro and in situ, with an antibiotic-sensitive E. coli environmental isolate (MA728). In situ matings were conducted in modified membrane diffusion chambers, in the degritter tank at the Grant Street (Melbourne, Fla.) sewage treatment facility, and in the sewage effluent-receiving waters in Melbourne, Fla. The transfer frequencies in situ were 3.2 x 10(-5) to 1.0 x 10(-6), compared with 1.6 x 10(-4) to 4.4 x 10(-5) observed in vitro. Transfer was shown to occur in raw sewage but was not detected in the effluent-receiving waters. The presence of a 60-megadalton plasmid species in both donor and transconjugants, but not in the recipients, provided physical evidence for the transfer of antibiotic resistance in situ.     

Monitoring the conjugal transfer of plasmid RP4 in activated sludge and in situ identification of the transconjugants

A GFPmut3b-tagged derivative of broad host-range plasmid RP4 was used to monitor the conjugative transfer of the plasmid from a Pseudomonas putida donor strain to indigenous bacteria in activated sludge. Transfer frequencies were determined to be in the range of 4 x 10(-6) to 1 x 10(-5) transconjugants per recipient. In situ hybridisation with fluorescently labeled, rRNA-targeted oligonucleotides was used to phylogenetically affiliate the bacteria that had received the plasmid.     

Genetic and Biophysical Study of R Plasmids Conferring Sulfonamide Resistance in Shigella Strains Isolated in 1952 and 1956

The conjugative plasmids determining sulfonamide resistance in five Shigella strains, each isolated from a different patient, have been characterized. One S. flexneri 2a strain, isolated in 1952, harbored an fi(+) plasmid of molecular weight 53 x 10(6), which specified synthesis of F-like pili and bore determinants for sulfonamide resistance (Su) and bacteriocinogeny (Col). This plasmid was compatible with plasmids of groups F(I), F(II), I(alpha), and P. A second S. flexneri 2a strain isolated in 1952 harbored an fi(-) plasmid of molecular weight 59 x 10(6), bearing the Su determinant and compatible with all plasmids tested. This strain also harbored an fi(+) group-F(II) plasmid of molecular weight 42 x 10(6), which bore the Col determinant and specified synthesis of F-like pili. Three S. dysenteriae 2 strains isolated in 1956 carried apparently identical fi(-) plasmids of molecular weight 58 x 10(6), which bore the Su determinant, could form transconjugants in Pseudomonas but not in Proteus, and were incompatible with the P-group plasmid RP4.     

Transformation of Streptococcus sanguis Challis by plasmid deoxyribonucleic acid from Streptococcus faecalis.

Plasmid deoxyribonucleic acid (DNA) from Streptococcus faecalis, strain DS5, was transferred to the Challis strain of Streptococcus sanguis by transformation. Two antibiotic resistance markers carried by the beta plasmid from strain DS5, erythromycin and lincomycin, were transferred to S. sanguis at a maximum frequency of 1.8 x 10-5/colony-forming unit. Approximately 70% of the covalently closed circular DNA isolated from transformant cultures by dye buoyant density gradients was shown to be hybridizable to beta plasmid DNA. Two major differences were observed between the beta plasmid from S. faecalis and the plasmid isolated from transformed S. sanguis: (i) the beta plasmid from strain DS5 sedimented in velocity gradients at 43S, whereas the covalently closed circular DNA from transformed Challis sedimented at 41S, suggesting a 1.5-Mdal deletion from the beta plasmid occurred; (ii) although the 43S beta plasmid remained in the supercoiled configuration for several weeks after isolation, the 41S plasmid was rapidly converted to a linear double-stranded molecule. Attempts to transform S. sanguis with the alpha plasmid from S. faecalis, strain DS5, were unsuccessful.     

Determination of the chromosomal size of three different strains of Enterococcus faecalis and one strain of Enterococcus faecium.

Pulsed-field gel electrophoresis was used to determine the chromosomal size of three different strains of Enterococcus faecalis and one strain of Enterococcus faecium. The size determinations of OG1X, a strain of E. faecalis widely used in many laboratories for genetic studies, using Sma I, Not I, and Sfi I alone or in combination, ranged from 2,750 to 2,761 kb. Using the same enzymes as with OG1X, the size of HH-67, a plasmid-free clinical isolate of E. faecalis, was determined to be 2,170-2,288 kb and the size of JH2-2, an E. faecalis recipient strain, ranged from 2,008 to 2,135 kb. The size range generated for GE-1, a plasmid-free E. faecium strain, with the use of Sma I, Not I, and Apa I was 2,045-2,155 kb. Although OG1X differed in size from the other three enterococci, each individual enterococcal strain generated reproducible results in different experiments. However, for both E. faecalis OG1X and E. faecium GE-1, one of the enzymes used generated a considerably smaller molecular size than that generated by the other two enzymes. The discrepancy was due to visually undiscernible comigrating fragments, and serves to point out a potential source of error if fewer than two enzymes are used to size a genome. The size discrepancies were resolved by digesting individual fragments with a second enzyme. The molecular sizes of these enterococcal strains are larger than that recently reported for Campylobacter, smaller than that of Escherichia coli and Pseudomonas aeruginosa, and similar (OG1X) or smaller (JH2-2, HH67, and GE-1) than the 2,819-kb reported for Streptococcus mutans.