Influence of soil variables on in situ plasmid transfer from Escherichia coli to Rhizobium fredii

A model system was established to determine whether intergeneric plasmid transfer occurs in soil and how various soil variables affect the rate of plasmid transfer. The donor bacterium, Escherichia coli HB101 carrying plasmid pBLK1-2 (pRK2073::Tn5), and the recipient bacterium, Rhizobium fredii USDA 201, were inoculated into a sterile Adelphia fine-sandy-loam soil. Transconjugants were enumerated by direct plating on antibiotic-amended HM [N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid; 2-(N-morpholino) ethanesulfonic acid] salts medium. Randomly chosen transconjugants were verified by serological typing and Southern hybridization with a Tn5 gene probe. The maximum transfer frequency was observed after 5 days of incubation (1.8 x 10(-4) per recipient). The influences of clay (0 to 50% addition), organic matter (0 to 15% addition), soil pH (4.3 to 7.25), soil moisture (2 to 40%), and soil incubation temperature (5 to 40 degrees C) on plasmid transfer were examined. Maximum transfer frequencies were noted at a clay addition of 15%, an organic matter addition of 5%, a soil pH of 7.25, a soil moisture content of 8%, and a soil incubation temperature of 28 degrees C. These results indicate that intergeneric plasmid transfer may occur in soil and that soil variables may significantly affect the rate of transfer.     

Plasmid transfer in natural soil: a case by case study with nitrogen-fixing Enterobacter

Abstract In strains of nitrogen-fixing Enterobacter agglomerans , isolated from the rhizosphere of cereals, the nif genes are located on large plasmids. Plasmid pEA9 (200 kb) is self-transmissible between closely related strains. To collect data on possible uncontrolled gene spread, for planned releases of such bacteria, plasmid pEA9 was labelled with transposons (Tn 1725 and Tn 5 ) and used in mating experiments between homologous Enterobacter strains with soil as substrate. The soil was from a plot into which an actual release was being planned. In the majority of experiments it was not sterilized.
Survival and plasmid transfer is described, as are variations in temperature, time, moisture, pH and soil packing. Further experiments were with or without added energy sources, and with or without plant roots. Under standard conditions (22°C, pH 5.2, 15.5% moisture, loose soil, 2 × 10⁷ inoculated donor and recipient cells each per g soil, 3 days incubation) sterilized soil gave low rates of plasmid transfer (10⁻⁶ per donor) but non-sterilized soil gave none. Adding Luria broth or sucrose to non-sterilized soil elicited strong cell propagation, together with plasmid transfer (optimum after incubation for 1 day: 10⁻⁴ exconjugants per donor). No transfer could be registered in the presence of wheat seedling roots for periods up to 5 weeks.     

Influence of Environmental Factors on Plasmid Transfer in Soil Microcosms

A model system was established to determine whether plasmid transfer occurs in soil and how various environmental conditions and cellular energy states affect the rate of plasmid transfer. Different donor and recipient bacteria were inoculated into sterile sandy lutitic soil microcosms. Dispersion studies were performed with a multipoint inoculator sampler. Transconjugant cells were enumerated by direct plating on antibiotic-amended LB medium. The influences of soil moisture (6.7 to 60%), incubation temperature (4° to 44°C) and pH (5.3 to 9.2) on cell dispersal and on plasmid transfer were examined. Maximum transfer frequencies were observed at: 20% of moisture content, pH between 7 and 8, and 30°C. These results indicate that plasmid transfer may occur in soil and that environmental conditions may significantly affect the rate of transfer.     

Delivery system for creation of one-step in vivo lac gene fusions in Pseudomonas spp. involved in biological control

The suicide plasmid pVA838 carrying the operon fusion transposon Tn5-lac was used as a delivery system to introduce Tn5-lac into Pseudomonas sp. strain M114. Random, in vivo lac gene fusions were successfully isolated in a one-step conjugation approach with this vector system. Tn5-lac-containing exconjugants were recovered at a frequency of approximately 10(-7) per recipient. However, when the mating temperature was increased from the normal growth temperature (28 degrees C) to 34 degrees C, the frequency was increased to approximately 10(-4) per recipient. A number of in vivo lac gene fusions were isolated and characterized in strain M114, a potentially important bacterium for biological control purposes.     

Delivery system for creation of one-step in vivo lac gene fusions in Pseudomonas spp. involved in biological control.

The suicide plasmid pVA838 carrying the operon fusion transposon Tn5-lac was used as a delivery system to introduce Tn5-lac into Pseudomonas sp. strain M114. Random, in vivo lac gene fusions were successfully isolated in a one-step conjugation approach with this vector system. Tn5-lac-containing exconjugants were recovered at a frequency of approximately 10(-7) per recipient. However, when the mating temperature was increased from the normal growth temperature (28 degrees C) to 34 degrees C, the frequency was increased to approximately 10(-4) per recipient. A number of in vivo lac gene fusions were isolated and characterized in strain M114, a potentially important bacterium for biological control purposes.     

Conjugational transfer of recombinant DNA in cultures and in soils: host range of Pseudomonas putida TOL plasmids.

Recombinant TOL plasmid pWWO-EB62 allows Pseudomonas putida to grow on p-ethylbenzoate. This plasmid can be transferred to other microorganisms, and its catabolic functions for the metabolism of alkylbenzoates are expressed in a limited number of gram-negative bacteria, including members of pseudomonad rRNA group I and Escherichia coli. Transfer of the recombinant plasmid to Erwinia chrysanthemi was observed, but transconjugants failed to grow on alkylbenzoates because they lost catabolic functions. Pseudomonads belonging to rRNA groups II, III, and IV, Acinetobacter calcoaceticus, and Alcaligenes sp. could not act as recipients for TOL, either because the plasmid was not transferred or because it was not stably maintained. The frequency of transfer of pWWO-EB62 from P. putida as a donor to pseudomonads belonging to rRNA group I was on the order of 1 to 10(-2) transconjugant per recipient, while the frequency of intergeneric transfer ranged from 10(-3) to 10(-7) transconjugant per recipient. The profile of potential hosts was conserved when the donor bacterium was Escherichia coli or Erwinia chrysanthemi instead of P. putida. No intergeneric gene transfer of the recombinant TOL plasmid was observed in soils; however, intraspecies transfer did take place. Intraspecies transfer of TOL in soils was affected by the type of soil used, the initial inoculum size, and the presence of chemicals that could affect the survival of the donor or recipient bacteria.     

Conjugational transfer of recombinant DNA in cultures and in soils: host range of Pseudomonas putida TOL plasmids.

Recombinant TOL plasmid pWWO-EB62 allows Pseudomonas putida to grow on p-ethylbenzoate. This plasmid can be transferred to other microorganisms, and its catabolic functions for the metabolism of alkylbenzoates are expressed in a limited number of gram-negative bacteria, including members of pseudomonad rRNA group I and Escherichia coli. Transfer of the recombinant plasmid to Erwinia chrysanthemi was observed, but transconjugants failed to grow on alkylbenzoates because they lost catabolic functions. Pseudomonads belonging to rRNA groups II, III, and IV, Acinetobacter calcoaceticus, and Alcaligenes sp. could not act as recipients for TOL, either because the plasmid was not transferred or because it was not stably maintained. The frequency of transfer of pWWO-EB62 from P. putida as a donor to pseudomonads belonging to rRNA group I was on the order of 1 to 10(-2) transconjugant per recipient, while the frequency of intergeneric transfer ranged from 10(-3) to 10(-7) transconjugant per recipient. The profile of potential hosts was conserved when the donor bacterium was Escherichia coli or Erwinia chrysanthemi instead of P. putida. No intergeneric gene transfer of the recombinant TOL plasmid was observed in soils; however, intraspecies transfer did take place. Intraspecies transfer of TOL in soils was affected by the type of soil used, the initial inoculum size, and the presence of chemicals that could affect the survival of the donor or recipient bacteria.     

Gene escape model: transfer of heavy metal resistance genes from Escherichia coli to Alcaligenes eutrophus on agar plates and in soil samples

Conjugal transfer from Escherichia coli to Alcaligenes eutrophus of the A. eutrophus genes coding for plasmid-borne resistance to cadmium, cobalt, and zinc (czc genes) was investigated on agar plates and in soil samples. This czc fragment is not expressed in the donor strain, E. coli, but it is expressed in the recipient strain, A. eutrophus. Hence, expression of heavy metal resistance by cells plated on a medium containing heavy metals represents escape of the czc genes. The two plasmids into which this DNA fragment has been cloned previously and which were used in these experiments are the nonconjugative, mobilizable plasmid pDN705 and the nonconjugative, nonmobilizable plasmid pMOL149. In plate matings at 28 to 30 degrees C, the direct mobilization of pDN705 occurred at a frequency of 2.4 x 10(-2) per recipient, and the mobilization of the same plasmid by means of the IncP1 conjugative plasmids RP4 or pULB113 (present either in a third cell [triparental cross] or in the recipient strain itself [retromobilization]) occurred at average frequencies of 8 x 10(-4) and 2 x 10(-5) per recipient, respectively. The czc genes cloned into the Tra- Mob- plasmid pMOL149 were transferred at a frequency of 10(-7) to 10(-8) and only by means of plasmid pULB113. The direct mobilization of pDN705 was further investigated in sandy, sandy-loam, and clay soils. In sterile soils, transfer frequencies at 20 degrees C were highest in the sandy-loam soil (10(-5) per recipient) and were enhanced in all soils by the addition of easily metabolizable nutrients.(ABSTRACT TRUNCATED AT 250 WORDS)     

Gene escape model: transfer of heavy metal resistance genes from Escherichia coli to Alcaligenes eutrophus on agar plates and in soil samples.

Conjugal transfer from Escherichia coli to Alcaligenes eutrophus of the A. eutrophus genes coding for plasmid-borne resistance to cadmium, cobalt, and zinc (czc genes) was investigated on agar plates and in soil samples. This czc fragment is not expressed in the donor strain, E. coli, but it is expressed in the recipient strain, A. eutrophus. Hence, expression of heavy metal resistance by cells plated on a medium containing heavy metals represents escape of the czc genes. The two plasmids into which this DNA fragment has been cloned previously and which were used in these experiments are the nonconjugative, mobilizable plasmid pDN705 and the nonconjugative, nonmobilizable plasmid pMOL149. In plate matings at 28 to 30 degrees C, the direct mobilization of pDN705 occurred at a frequency of 2.4 x 10(-2) per recipient, and the mobilization of the same plasmid by means of the IncP1 conjugative plasmids RP4 or pULB113 (present either in a third cell [triparental cross] or in the recipient strain itself [retromobilization]) occurred at average frequencies of 8 x 10(-4) and 2 x 10(-5) per recipient, respectively. The czc genes cloned into the Tra- Mob- plasmid pMOL149 were transferred at a frequency of 10(-7) to 10(-8) and only by means of plasmid pULB113. The direct mobilization of pDN705 was further investigated in sandy, sandy-loam, and clay soils. In sterile soils, transfer frequencies at 20 degrees C were highest in the sandy-loam soil (10(-5) per recipient) and were enhanced in all soils by the addition of easily metabolizable nutrients.(ABSTRACT TRUNCATED AT 250 WORDS)     

Isolation of Tn916-like conjugal elements from swine lot effluent

Isolates of Enterococcus faecalis obtained from a swine farrowing house outflow were examined for genetic elements similar to Tn916. Of the enterococci isolated, 71% were resistant to tetracycline. Among the tetracycline-resistant enterococci isolated from the outflow samples, approximately 34% were able to transfer the tetracycline resistance phenotype to Bacillus thuringiensis in cross-genus matings. The frequencies of transfer for 10 random isolates were comparable to those for transfer of Tn916 from E. faecalis to B. thuringiensis. In addition, these elements were shown to mobilize plasmid pC194 between Bacillus species, as did Tn916. Southern blot and polymerase chain reaction (PCR) analysis showed these elements share extensive structural homology with Tn916. The selected conjugal elements were capable of transfer to a Bacillus recipient in a soil environment. When the swine waste was introduced into the soil, the tetracycline resistant fecal enterococci levels rose from essentially undetectable levels to approximately 4 x 10(4) and remained at this level for 4 weeks. After six months, including one winter, levels had decreased to 5 x 10(3).     

Gene transfer in magnetic bacteria: transposon mutagenesis and cloning of genomic DNA fragments required for magnetosome synthesis.

Broad-host-range IncP and IncQ plasmids have been transferred to the aerobic magnetic bacterium Aquaspirillum sp. strain AMB-1. Conjugal matings with Escherichia coli S17-1 allowed high-frequency transfer of the RK2 derivative pRK415 (4.5 x 10(-3) transconjugant per recipient cell) and the RSF1010 derivative pKT230 (3.0 x 10(-3) transconjugant per recipient). These plasmids successfully formed autonomous replicons in transconjugants and could be isolated and transformed back into E. coli, illustrating their potential as shuttle vectors. A mobilizable plasmid containing transposon Tn5 was transferred to Aquaspirillum sp. strain AMB-1 and also to the obligately microaerophilic magnetic bacterium Aquaspirillum magnetotacticum MS-1. Five nonmagnetic kanamycin-resistant mutants of Aquaspirillum sp. strain AMB-1 in which Tn5 was shown to be integrated into the chromosome were obtained. Different genomic fragments containing the mutagenized regions were cloned into E. coli. Two genomic fragments were restriction mapped, and the site of Tn5 insertion was determined. They were shown to be identical, although derived from independent transposon insertions. One of these clones was found to hybridize strongly to regions of the A. magnetotacticum MS-1 chromosome. This is the first report of gene transfer in a magnetic bacterium.     

Mobilization and transfer of Azospirillum lipoferum plasmid by the Tn5-Mob transposon into a plasmid-free Agrobacterium tumefaciens strain

Azospirillum lipoferum 4B harbors five cryptic plasmids. Several suicide plasmids were used to transfer Tn5-Mob to A. lipoferum 4B. Tn5-Mob insertion mutations of this strain could be obtained at frequencies of 10(-8)-10(-7) per recipient cell. One hundred Tn5-Mob A. lipoferum 4B mutants were used in bacterial matings with a plasmid-free Agrobacterium tumefaciens recipient strain. This is the first report of mobilization, transfer, and replication of an Azospirillum plasmid in Agrobacterium tumefaciens. One transconjugant was found which had lost an indigenous plasmid.     

Plasmid dynamics in a model soil column

Continuous-flow column reactors were used to study the dynamics of plasmid exchange in a structured, thermodynamically open system containing either Enterobacter cloacae or Pseudomonas cepacia , both carrying the transmissible recombinant plasmid R388::Tn1721. Plasmid transfer rates were higher in vermiculite and sterile soil columns supplied with nutrient solution than those in sterile and non-sterile soil columns without input of nutrient solution. For both species, donor and recipient strains took about 5 days to reach their maximum densities in effluents from the columns supplied with nutrient solution. After about 8 day s the donor and transconjugant populations of P. cepacia in the effluent solution decreased exponentially, whereas E. cloacae donor, recipient and transconjugant strains maintained steady-state concentrations. The difference between plasmid stability in the two species may have significant consequences in terms of releasing plasmid-bearing genetically modified microorganisms into the natural environment. The plasmid is persistent in E. cloacae in non-sterile soil even though its transfer to the marked recipient in non-sterile soil was minimal.     

Efficiency of the transfer of a pSAM2-derivative plasmid between two strains of Streptomyces lividans in conditions ranging from agar slants to non-sterile soil microcosms

Abstract: The aim of this work was to determine the efficiency of the conjugative plasmid pTS130 to transfer in various environmental conditions between two strains of Streptomyces lividans . This plasmid is a derivative of the conjugative and integrative plasmid pSAM2 isolated originally from Streptomyces ambofaciens and capable of transfer to a large range of bacteria. Our results demonstrate the high frequency of the conjugation mechanism since more than 60% of the recipient cells developed on agar slants harbored the plasmid pTS130 (as evidenced by Southern hybridization with a pSAM2 derivative plasmid probe). When donor and recipient strains were inoculated into sterile and non-sterile soil microcosms, transconjugants were detected after two days of incubation in both cases. However, the number of donor, recipient and transconjugant cells were established at a lower level in the non-sterile soil than in the sterile soil experiments. Moreover, nutrient amendment of the sterile soil was found to increase the population levels of parental strains and transfer frequencies both significantly and simultaneously. On the other hand, modifying water potential of the soil microcosms did not result in affecting the establishment of the Streptomyces lividans cells or the transfer rate.     

Construction, transfer and properties of a novel temperature-sensitive integrable plasmid for genomic analysis of Staphyiococcus aureus

As an alternative approach to genetic transfer and analysis, a novel integrable plasmid system was developed that should prove useful for mapping and cloning various genes in Staphylococcus aureus and other Gram-positive bacteria. The use of a restriction-deficient recipient strain and an improved protocol for protoplast plasmid transformation facilitated direct cloning of a recombinant plasmid (pPQ126) in S. aureus NCTC 8325-4. Plasmid pPQ126 (13.6 kb) is a novel, temperature-sensitive integrable plasmid containing genes encoding resistance to erythromycin and chloramphenicol (from plasmid pTV1ts), and resistance to gentamicin (from transposon Tn4001). When introduced into an appropriate recipient strain at the permissive temperature (30 degrees C), pPQ126 replicates autonomously. Integration of pPQ126 is directed into homologous chromosomal target sequences (chromosomal insertions of Tn551 or Tn4001) by growing a population of cells containing autonomous pPQ126 in the presence of gentamicin, erythromycin, and chloramphenicol at 39 degrees C (nonpermissive temperature). Elevated temperature both selects for and maintains pPQ126 as an integrated replicon. Integration of pPQ126 occurs at significantly reduced frequency in a recombination-deficient host, and does not occur in the absence of host chromosomal homology. Integrated pPQ126 excises from the chromosome under permissive conditions (30 degrees C), and excision results in derivatives of pPQ126 that harbour DNA of chromosomal origin.     

Broad-host-range plasmid pRK340 delivers Tn5 into the Legionella pneumophila chromosome.

Transposon Tn5 was introduced into Legionella pneumophila on plasmid pRK340, which is temperature sensitive for plasmid maintenance. The presence of plasmid DNA was confirmed by agarose gel electrophoresis and by conjugal transfer of the plasmid to Escherichia coli. Tn5 insertions were obtained by culturing L. pneumophila at the nonpermissive temperature (43 degrees C) on buffered charcoal-yeast extract agar containing kanamycin. Of the 260 kanamycin-resistant colonies picked, 220 failed to conjugate pRK340 to E. coli. Plasmid DNA was not visualized from eight randomly picked Tn5-containing strains, and Southern hybridizations indicated that Tn5, but not pRK340, inserted into multiple sites in the Legionella chromosome. In addition, the streptomycin resistance determinant on Tn5 was expressed in L. pneumophila.     

Persistence of Free Plasmid DNA in Soil Monitored by Various Methods, Including a Transformation Assay

The persistence and stability of free plasmid pUC8-ISP DNA introduced into 10-g samples of various soils and kept at 23°C were monitored over a period of 60 days. The soils were sampled at a plant science farm and included a loamy sand soil (no. 1), a clay soil (no. 2), and a silty clay soil (no. 3). Four different methods allowed monitoring of (i) the production of acid-soluble radioactive material from [³H]thymidine-labeled plasmid DNA, (ii) the decrease of hybridizing nucleotide sequences in slot blot analysis, (iii) the loss of plasmid integrity measured by Southern hybridization, and (iv) the decay of the biological activity as determined by transformation of Ca²⁺-treated Escherichia coli cells with the DNA extracted from soil. Acid-soluble material was not produced within the first 24 h but then increased to 45% (soil no. 1), 27% (soil no. 2), and 77% (soil no. 3) until the end of incubation. A quite parallel loss of material giving a slot blot hybridization signal was observed. Southern hybridization indicated that after 1 h in the soils, plasmid DNA was mostly in the form of circular and full-length linear molecules but that, depending on the soil type, after 2 to 5 days full-length plasmid molecules were hardly detectable. The transforming activity of plasmid DNA reextracted from the soils followed inactivation curves over 2 to 4 orders of magnitude and dropped below the detection limit after 10 days. The inactivation was slower in soil no. 2 (28.2-h half-life time of the transforming activity of a plasmid molecule) than in soils no. 3 (15.1 h) and no. 1 (9.1 h). The studies provide data on the persistence of free DNA molecules in natural bacterial soil habitats. The data suggest that plasmid DNA may persist long enough to be available for uptake by competent recipient cells in situ.     

Transfer-defective and tetracycline-sensitive mutants of the incompatibility group HI plasmid R27 generated by insertion of transposon 7

Transposon Tn7 insertion was used to obtain either transfer-defective (Tra-) or tetracycline-sensitive (Tc-) mutants of the HI incompatibility group (IncHI) plasmid R27. The 600 apparent R27::Tn7 derivatives fell into three classes: Tra-, Tc-, and Tra- Tc-. Mutants of R27 defective in the thermosensitive mode of transfer characteristic of IncH plasmids were obtained with transfer frequencies of less than 1 X 10(-8) transconjugants per recipient after 18 hr at 26 degrees C. These mutants, which were generated at a frequency of 1 per 100 insertions, were nonleaky and nonrevertible. Tc- mutants of R27, generated at a frequency of 0.5 per 100 insertions, were also nonrevertible. Loss of tetracycline resistance was associated with an increased frequency of transfer (average 3.6 X 10(-3) transconjugants per donor per hour at 30 degrees C) compared with transfer of the wild-type R27 plasmid (1.6 X 10(-8) per donor per hour). Tn7 insertions which generated Tc- or Tra- mutants of R27 had no effect on entry exclusion of other H group plasmids. The molecular weights of Tra- and Tc- R27::Tn7 derivatives were approximately 120.5 MDa, corresponding to the sum of R27 (112 MDa) and Tn7 (8.5 MDa). A third class of Tn7 insertion derivatives (Tra- Tc-) was obtained; however, strains expressing this phenotype were plasmid free, and appeared to have Tn7 integrated at a chromosomal site. Restriction digestion with XbaI and subsequent hybridization with ColE1::Tn7 were used to compare R27::Tn7 derivatives and to locate Tn7 insertion sites. Loss of tetracycline resistance was associated with Tn7 insertion into a 24-kb XbaI fragment of R27. Although loss of plasmid transfer in several R27::Tn7 derivatives was accompanied by insertion of Tn7 into a 14-kb XbaI fragment of the plasmid, these mutants had also undergone a small increase in the size of the 24-kb XbaI fragment of R27.     

Mobilization of Haemophilus influenzae chromosomal markers by an Escherichia coli F'' factor.

Filter matings between E. coli K-12 strains carrying an F'::Tn5,Tn9 factor with H. influenzae Rd strains gave rise to kanamycin-chloramphenicol-resistant H. influenzae strains at a frequency of approximately 10(-6). Transfer of the F' factor to H. influenzae was verified by expression of unselected markers in H. influenzae (lac+ or cotransfer of the nonselected antibiotic resistance), physical presence of a high-molecular-weight plasmid in recipient H. influenzae cells, and detection by Southern hybridization analysis of DNA sequences specific for the F' factor replication and partition functions in recipient H. influenzae cells. H. influenzae (F' Tn5,Tn9) strains were capable of transferring kanamycin and chloramphenicol resistances to other H. influenzae strains and were capable of mobilizing H. influenzae chromosomal markers at a low frequency. Insertion of a Tn5 element in the H. influenzae genome near the novobiocin resistance gene increased the frequency of transfer of novobiocin resistance about 30-fold. Transfer of other chromosomal markers also increased, although to a lesser extent, and ordered transfer of chromosomal markers could be demonstrated. Gene transfer was insensitive to DNase I, and transfer of chromosomal (but not F' factor) markers was dependent on the H. influenzae rec-1 and rec-2 gene functions.     

Transfer of the conjugal tetracycline resistance transposon Tn916 from Streptococcus faecalis to Staphylococcus aureus and identification of some insertion sites in the staphylococcal chromosome.

The Streptococcus faecalis pheromone-dependent conjugative plasmid pAD1::Tn916 and the membrane filter-dependent conjugative plasmid pPD5::Tn916 were used to introduce Tn916 into Staphylococcus aureus by intergeneric protoplast fusions and intergeneric membrane-filter matings. In recombinants obtained by protoplast fusion where no plasmid DNA could be detected, tetracycline resistance resulted from transposition of Tn916 from pAD1 to the S. aureus chromosome. Transformation analyses showed that S. aureus Tn916 chromosomal insertions occurred near pig, ilv, uraA, tyrB, fus, ala, and the trp operon. DNA hybridization analyses of EcoRI- and HindIII-digested chromosomal DNAs confirmed the diversity of chromosomal sites involved and demonstrated that the inserts were Tn916 insertions rather than integrations of all or part of pAD1::Tn916. Both pAD1::Tn916 and pPD5::Tn916 were transferred to S. aureus by membrane-filter matings. These plasmids remained intact and expressed tetracycline resistance in S. aureus. S. aureus strains carrying pAD1::Tn916, but not a chromosomal insert of Tn916, and any one of several conjugal gentamicin-resistance plasmids lost their ability to serve as conjugal donors of the gentamicin-resistance plasmids.