土壤中质粒pLV1016 在快生型大豆根瘤菌间的水平转移

在滤膜、液体培养基和土壤微宇宙3种系统中,研究了接合型质粒pLV1016 由快生型大豆根瘤菌(Rhizobiumfredii)QB1131 向R.frediilux Lux3的水平转移及pLV1016 由QB1131 向土著细菌的转移.接合培养1d后,分别计算供、受体菌的生长速率和质粒转移速率常数(γ).结果表明,相同接种浓度下,滤膜接合时γ值最高,土壤中γ值最低,γ值不受土壤是否灭菌和是否有大豆植株的影响,γ值与初始接种浓度负相关,与供、受体的生长速率正相关.在未灭菌土中检测到pLV1016 可转移到土著细菌,土著接合子分别属于根瘤菌属和假单胞菌属.     

Conjugative Transfer of Chromosomal Genes between Fluorescent Pseudomonads in the Rhizosphere of Wheat

Bacteria released in large numbers for biocontrol or bioremediation purposes might exchange genes with other microorganisms. Two model systems were designed to investigate the likelihood of such an exchange and some factors which govern the conjugative exchange of chromosomal genes between root-colonizing pseudomonads in the rhizosphere of wheat. The first model consisted of the biocontrol strain CHA0 of Pseudomonas fluorescens and transposon-facilitated recombination (Tfr). A conjugative IncP plasmid loaded with transposon Tn5, in a CHA0 derivative carrying a chromosomal Tn5 insertion, promoted chromosome transfer to auxotrophic CHA0 recipients in vitro. A chromosomal marker (pro) was transferred at a frequency of about 10(sup-6) per donor on wheat roots under gnotobiotic conditions, provided that the Tfr donor and recipient populations each contained 10(sup6) to 10(sup7) CFU per g of root. In contrast, no conjugative gene transfer was detected in soil, illustrating that the root surface stimulates conjugation. The second model system was based on the genetically well-characterized strain PAO of Pseudomonas aeruginosa and the chromosome mobilizing IncP plasmid R68.45. Although originally isolated from a human wound, strain PAO1 was found to be an excellent root colonizer, even under natural, nonsterile conditions. Matings between an auxotrophic R68.45 donor and auxotrophic recipients produced prototrophic chromosomal recombinants at 10(sup-4) to 10(sup-5) per donor on wheat roots in artificial soil under gnotobiotic conditions and at about 10(sup-6) per donor on wheat roots in natural, nonsterile soil microcosms after 2 weeks of incubation. The frequencies of chromosomal recombinants were as high as or higher than the frequencies of R68.45 transconjugants, reflecting mainly the selective growth advantage of the prototrophic recombinants over the auxotrophic parental strains in the rhizosphere. Although under field conditions the formation of chromosomal recombinants is expected to be reduced by several factors, we conclude that chromosomal genes, whether present naturally or introduced by genetic modification, may be transmissible between rhizosphere bacteria.     

Plasmid Transfer between Spatially Separated Donor and Recipient Bacteria in Earthworm-Containing Soil Microcosms

Most gene transfer studies have been performed with relatively homogeneous soil systems in the absence of soil macrobiota, including invertebrates. In this study we examined the influence of earthworm activity (burrowing, casting, and feeding) on transfer of plasmid pJP4 between spatially separated donor (Alcaligenes eutrophus) and recipient (Pseudomonas fluorescens) bacteria in nonsterile soil columns. A model system was designed such that the activity of earthworms would act to mediate cell contact and gene transfer. Three different earthworm species (Aporrectodea trapezoides, Lumbricus rubellus, and Lumbricus terrestris), representing each of the major ecological categories (endogeic, epigeic, and anecic), were evaluated. Inoculated soil microcosms, with and without added earthworms, were analyzed for donor, recipient, and transconjugant bacteria at 5-cm-depth intervals by using selective plating techniques. Transconjugants were confirmed by colony hybridization with a mer gene probe. The presence of earthworms significantly increased dispersal of the donor and recipient strains. In situ gene transfer of plasmid pJP4 from A. eutrophus to P. fluorescens was detected only in earthworm-containing microcosms, at a frequency of (symbl)10(sup2) transconjugants per g of soil. The depth of recovery was dependent on the burrowing behavior of each earthworm species; however, there was no significant difference in the total number of transconjugants among the earthworm species. Donor and recipient bacteria were recovered from earthworm feces (casts) of all three earthworm species, with numbers up to 10(sup6) and 10(sup4) bacteria per g of cast, respectively. A. trapezoides egg capsules (cocoons) formed in the inoculated soil microcosms contained up to 10(sup7) donor and 10(sup6) recipient bacteria per g of cocoon. No transconjugant bacteria, however, were recovered from these microhabitats. To our knowledge, this is the first report of gene transfer between physically isolated bacteria in nonsterile soil, using burrowing earthworms as a biological factor to facilitate cell-to-cell contact.     

Transfer of the Pea Symbiotic Plasmid pJB5JI in Nonsterile Soil

Transfer of the pea (Pisum sativum L.) symbiotic plasmid pJB5JI between strains of rhizobia was examined in sterile and nonsterile silt loam soil. Sinorhizobium fredii USDA 201 and HH003 were used as plasmid donors, and symbiotic plasmid-cured Rhizobium leguminosarum 6015 was used as the recipient. The plasmid was carried but not expressed in S. fredii strains, whereas transfer of the plasmid to R. leguminosarum 6015 rendered the recipient capable of nodulating pea plants. Confirmation of plasmid transfer was obtained by acquisition of plasmid-encoded antibiotic resistance genes, nodulation of pea plants, and plasmid profiles. Plasmid transfer in nonsterile soil occurred at frequencies of up to 10⁻⁴ per recipient and appeared to be highest at soil temperatures and soil moisture levels optimal for rhizobial growth. Conjugation frequencies were usually higher in sterile soil than in nonsterile soil. In nonsterile soil, transconjugants were recovered only with strain USDA 201 as the plasmid donor. Increasing the inoculum levels of donor and recipient strains up to 10⁹ cells g of soil⁻¹ increased the number of transconjugants; peak plasmid transfer frequencies, however, were found at the lower inoculum level of 10⁷ cells g of soil⁻¹. Plasmid transfer frequencies were raised in the presence of the pea rhizosphere or by additions of plant material. Transconjugants formed by the USDA 201(pJB5JI) × 6015 mating in soil formed effective nodules on peas.     

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.     

Frequency of horizontal gene transfer of a large catabolic plasmid (pJP4) in soil.

Limited work has been done to assess the bioremediation potential of transfer of plasmid-borne degradative genes from introduced to indigenous organisms in the environment. Here we demonstrate the transfer by conjugation of the catabolic plasmid pJP4, using a model system with donor and recipient organisms. The donor organism was Alcaligenes eutrophus JMP134 and the recipient organism was Variovorax paradoxus isolated from a toxic waste site. Plasmid pJP4 contains genes for mercury resistance and 2,4-dichlorophenoxyacetic (2,4-D) acid degradation. A transfer frequency of approximately 1/10(3) donor and recipient cells (parent cells) was observed on solid agar media, decreasing to 1/10(5) parent cells in sterile soil and finally 1/10(6) parent cells in 2,4-D-amended, nonsterile soil. Presumptive transconjugants were confirmed to be resistant to Hg, to be capable of degrading 2,4-D, and to contain a plasmid of size comparable to that of pJP4. In addition, we confirmed the transfer through PCR amplifications of the tfdB gene. Although transfer of pJP4 did occur at a high frequency in pure culture, the rate was significantly decreased by the introduction of abiotic (sterile soil) and biotic (nonsterile soil) stresses. An evaluation of the data from this model system implies that the reliance on plasmid transfer from a donor organism as a remediative strategy has limited potential.     

Detection and characterization of plasmid pJP4 transfer to indigenous soil bacteria

Prior to gene transfer experiments performed with nonsterile soil, plasmid pJP4 was introduced into a donor microorganism, Escherichia coli ATCC 15224, by plate mating with Ralstonia eutropha JMP134. Genes on this plasmid encode mercury resistance and partial 2, 4-dichlorophenoxyacetic acid (2,4-D) degradation. The E. coli donor lacks the chromosomal genes necessary for mineralization of 2,4-D, and this fact allows presumptive transconjugants obtained in gene transfer studies to be selected by plating on media containing 2,4-D as the carbon source. Use of this donor counterselection approach enabled detection of plasmid pJP4 transfer to indigenous populations in soils and under conditions where it had previously not been detected. In Madera Canyon soil, the sizes of the populations of presumptive indigenous transconjugants were 10(7) and 10(8) transconjugants g of dry soil(-1) for samples supplemented with 500 and 1,000 microg of 2,4-D g of dry soil(-1), respectively. Enterobacterial repetitive intergenic consensus PCR analysis of transconjugants resulted in diverse molecular fingerprints. Biolog analysis showed that all of the transconjugants were members of the genus Burkholderia or the genus Pseudomonas. No mercury-resistant, 2, 4-D-degrading microorganisms containing large plasmids or the tfdB gene were found in 2,4-D-amended uninoculated control microcosms. Thus, all of the 2,4-D-degrading isolates that contained a plasmid whose size was similar to the size of pJP4, contained the tfdB gene, and exhibited mercury resistance were considered transconjugants. In addition, slightly enhanced rates of 2,4-D degradation were observed at distinct times in soil that supported transconjugant populations compared to controls in which no gene transfer was detected.     

Predictive model of conjugative plasmid transfer in the rhizosphere and phyllosphere.

A computer simulation model was used to predict the dynamics of survival and conjugation of Pseudomonas cepacia (carrying the transmissible recombinant plasmid R388:Tn1721) with a nonrecombinant recipient strain in simple rhizosphere and phyllosphere microcosms. Plasmid transfer rates were derived for a mass action model, and donor and recipient survival were modeled as exponential growth and decay processes or both. Rate parameters were derived from laboratory studies in which donor and recipient strains were incubated in test tubes with a peat-vermiculite solution or on excised radish or bean leaves in petri dishes. The model predicted donor, recipient, and transconjugant populations in hourly time steps. It was tested in a microcosm planted with radish seeds and inoculated with donor and recipient strains and on leaf surfaces of radish and bean plants also growing in microcosms. Bacteria were periodically enumerated on selective media over 7 to 14 days. When donor and recipient populations were 10(6) to 10(8) CFU/g (wet weight) of plant or soil, transconjugant populations of about 10(1) to 10(4) were observed after 1 day. An initial rapid increase and a subsequent decline in numbers of transconjugants in the rhizosphere and on leaf surfaces were correctly predicted.     

Predictive model of conjugative plasmid transfer in the rhizosphere and phyllosphere

A computer simulation model was used to predict the dynamics of survival and conjugation of Pseudomonas cepacia (carrying the transmissible recombinant plasmid R388:Tn1721) with a nonrecombinant recipient strain in simple rhizosphere and phyllosphere microcosms. Plasmid transfer rates were derived for a mass action model, and donor and recipient survival were modeled as exponential growth and decay processes or both. Rate parameters were derived from laboratory studies in which donor and recipient strains were incubated in test tubes with a peat-vermiculite solution or on excised radish or bean leaves in petri dishes. The model predicted donor, recipient, and transconjugant populations in hourly time steps. It was tested in a microcosm planted with radish seeds and inoculated with donor and recipient strains and on leaf surfaces of radish and bean plants also growing in microcosms. Bacteria were periodically enumerated on selective media over 7 to 14 days. When donor and recipient populations were 10(6) to 10(8) CFU/g (wet weight) of plant or soil, transconjugant populations of about 10(1) to 10(4) were observed after 1 day. An initial rapid increase and a subsequent decline in numbers of transconjugants in the rhizosphere and on leaf surfaces were correctly predicted.     

Recombinant plasmid associated cell aggregation and high-frequency conjugation of Streptococcus lactis ML3

Lactose-positive (Lac+) transconjugants resulting from matings between Streptococcus lactic ML3 and S. lactis LM2301 possess a single plasmid of approximately 60 megadaltons (Mdal) which is nearly twice the size of the lactose plasmid of the donor. The majority of these Lac+ transconjugants aggregated in broth and were able to transfer lactose-fermenting ability at a frequency higher than 10(-1) per donor on milk agar plates or in broth. Lac+ transconjugants which did not clump conjugated at a much lower frequency. Lactose-negative derivatives of Lac+ clumping transconjugants did not aggregate in broth and were missing the 60-Mdal plasmid. The ability to aggregates in broth was very unstable. Strains could lose the ability to clump but retain lactose-fermenting ability. The majority of these Lac+ nonclumping derivatives of clumping transconjugants contained a plasmid of approximately 33 Mdal, the size of the lactose plasmid of the original donor ML3. These strains transferred lactose-fermenting ability at a frequency of approximately 10(-6) per donor, resulting in both Lac+ clumping transconjugants which contained a 60-Mdal plasmid and Lac+ nonclumping transconjugants which possessed a 33-Mdal plasmid. Our results suggest that the genes responsible for cell aggregation and high-frequency conjugation are on the segment of deoxyribonucleic acid which recombined with the 33-Mdal lactose plasmid in S. lactis ML3.     

Effect of Genomic Location on Horizontal Transfer of a Recombinant Gene Cassette Between Pseudomonas Strains in the Rhizosphere and Spermosphere of Barley Seedlings

The use of genetically engineered bacteria in natural environments constitutes a risk of transfer of recombinant DNA to the indigenous bacteria. However, chromosomal genes are believed to be less likely to transfer than genes on mobilizable and conjugative plasmids. To study this assumption, horizontal transfer of a recombinant gene cassette inserted into the chromosome of a Pseudomonas stutzeri strain, into a mobilizable plasmid (pAGM42), and into a conjugative plasmid (pKJK5) isolated from barley rhizosphere was investigated. Horizontal transfer efficiencies of the gene cassette inserted into a conjugative plasmid was 8.20 × 10⁻³ transconjugants/(donors × recipients)^1/2 in the rhizosphere and 4.57 × 10⁻² transconjugants/(donors × recipients)^1/2 in the spermosphere. Mobilization of the plasmid pAGM42 by the plasmids RP4 and pKJK5 was also detected at high levels in the microcosms, transfer efficiencies were up to 4.36 × 10⁻³ transconjugants/(donors × recipients)^1/2. Transfer of chromosomal encoded genes could not be detected in the microcosms by conjugation or transformation. However, transformation did occur by using the same bacterial strains under laboratory conditions. The rhizosphere and especially the spermosphere thus proved to be hot spot environments providing favorable conditions for gene transfer by mobilization and conjugation, but these environments did not support transformation at a detectable level. Received: 21 July 2000 / Accepted: 21 August 2000     

Conjugal transfer at natural population densities in a microcosm simulating an estuarine environment

Abstract Estuarine microcosms were used to follow conjugal transfer of a broad host range IncP1 plasmid from a Pseudomonas putida donor to indigenous bacteria. Donor cells were added at a concentration similar to the natural abundance of bacteria in the water column (10⁶ cells ml⁻¹). Transfer was not detected in any of the test microcosms (calculated limit of detection of 10⁻⁷ and 10⁻⁴ transconjugants donor⁻¹ in water column and sediment, respectively), with the exception of transfer to an isogenic recipient (added at 10⁵ cells ml⁻¹) in sediments of controls that had been inoculated with both donors and recipients. The same plasmid was transferred with high efficiencies (10⁻¹ to 10⁻³) to a variety of recipients in filter and broth matings. These results suggest that if conjugal gene transfer occurred, it was at efficiencies that were not detectable in estuarine microcosms simulating natural population densities.     

Gene transfer of Alcaligenes eutrophus JMP134 plasmid pJP4 to indigenous soil recipients.

This study evaluated the potential for gene transfer of a large catabolic plasmid from an introduced organism to indigenous soil recipients. The donor organism Alcaligenes eutrophus JMP134 contained the 80-kb plasmid pJP4, which contains genes that code for mercury resistance. Genes on this plasmid plus chromosomal genes also allow degradation of 2,4-dichloruphenoxyacetic acid (2,4-D). When JMP134 was inoculated into a nonsterile soil microcosm amended with 1,000 micrograms of 2,4-D g-1, significant (10(6) g of soil-1) populations of indigenous recipients or transconjugants arose. These transconjugants all contained an 80-kb plasmid similar in size to pJP4, and all degraded 2,4-D. In addition, all transconjugants were resistant to mercury and contained the tfdB gene of pJP4 as detected by PCR. No mercury-resistant, 2,4-D-degrading organisms with large plasmids or the tfdB gene were found in the 2,4-D-amended but uninoculated control microcosm. These data clearly show that the plasmid pJP4 was transferred to indigenous soil recipients. Even more striking is the fact that not only did the indigenous transconjugant population survive and proliferate but also enhanced rates of 2,4-D degradation occurred relative to microcosms in which no such gene transfer occurred. Overall, these data indicate that gene transfer from introduced organisms is an effective means of bioaugmentation and that survival of the introduced organism is not a prerequisite for biodegradation that utilizes introduced biodegradative genes.     

Root growth and exudate production define the frequency of horizontal plasmid transfer in the Rhizosphere

To identify the main drivers of plasmid transfer in the rhizosphere, conjugal transfer was studied in the rhizospheres of pea and barley. The donor Pseudomonas putida KT2442, containing plasmid pKJK5::gfp, was coated onto the seeds, while the recipient P. putida LM24, having a chromosomal insertion of dsRed, was inoculated into the growth medium. Mean transconjugant-to-donor ratios in vermiculite were 4.0+/-0.8 x 10(-2) in the pea and 5.9+/-1.4 x 10(-3) in the barley rhizospheres. In soil, transfer ratios were about 10 times lower. As a result of a 2-times higher root exudation rate in pea, donor densities in pea (1 x 10(6)-2 x 10(9) CFU g(-1) root) were about 10 times higher than in barley. No difference in recipient densities was observed. In situ visualization of single cells on the rhizoplane and macroscopic visualization of the colonization pattern showed that donors and transconjugants were ubiquitously distributed in the pea rhizosphere, while they were only located on the upper parts of the barley roots. Because the barley root elongated about 10 times faster than the pea root, donors were probably outgrown by the elongating barley root. Thus by affecting the cell density and distribution, exudation and root growth appear to be key parameters controlling plasmid transfer in the rhizosphere.     

Transfer of plasmid RP4 in the spermosphere and rhizosphere of barley seedling

Transfer of plasmid RP4 to indigenous bacteria in bulk soil could only be detected in soil with nutrient amendment. Lack of physiological active donor and recipient cells was apparently one of the limiting factors in un-amended bulk soil. Plasmid transfer was detected both in the spermosphere and rhizosphere of barley seedlings. Transfer occured from seed coated donor bacteria (i) to introduced recipient bacteria and (ii) to indigenous bacteria present in soil. Plasmid transfer was also detected from donor bacteria introduced to the soil to seed coated recipient bacteria. Transfer efficiencies in the rhizosphere were significantly below the transfer efficiencies obtained in the spermosphere. The transfer efficiencies detected in the barley spermosphere were among the highest reported from any natural environment.     

Role of the intraspecific competition in the regulation of Agrobacterium tumefaciens transconjugant population level in soil experiments

Abstract In microcosms of sterilized soil simultaneously inoculated with Pseudomonas aeruginosa carrying the plasmid R68-45 and the plasmid-free Agrobacterium tumefaciens , transconjugants were detectable after two days of incubation and their number remained constant thereafter. The growth of a transconjugant strain was monitored in sterile soil. When mixed together with the parental strains at high inoculum or when the soil was previously colonized by the donor, the transconjugant was able to grow. If the recipient was the first soil colonizer, the challenging population of transconjugant remained stable at its initial level. We demonstrated the possible role of intraspecific competition in the limitation of transconjugant numbers.     

Genetic transfer of the mcd gene in soil

AIMS: To investigate the role of horizontal gene transfer of mcd (methylcarbamate-degrading) gene in high genetic diversity of carbofuran-degrading bacteria. METHODS AND RESULTS: The actuality of genetic transfer from degraders to an Agrobacterium tumefaciens strain was determined in liquid medium. The mcd gene was chosen for transfer experiments. Transconjugants were obtained irrespective of the type of the donor strain (Gram-positive or Gram-negative), size of the inoculum, or nature and concentration of the pesticide in the medium. Soil microcosms, inoculated with or without the donor and/or recipient strains were used. The size of the initial degrading population (treated or untreated soil) and the nature of the inoculated donor strains were considered. More transconjugants were isolated in the previously treated soil than in the untreated soil. Agrobacterium transconjugants were isolated even when the donor strain was not inoculated, probably as a result of gene transfer from indigenous degrading population to the recipient strain. Moreover, potential transconjugants belonging to the Pseudomonas genus were isolated. CONCLUSIONS: Our results seem to demonstrate that the mcd gene is transferable in soil among bacterial populations. SIGNIFICANCE AND IMPACTS OF THE STUDY: The transfer of the mcd gene is partly responsible for the high genetic diversity of micro-organisms able to catabolize carbofuran.     

Detection and Characterization of Plasmid pJP4 Transfer to Indigenous Soil Bacteria

Prior to gene transfer experiments performed with nonsterile soil, plasmid pJP4 was introduced into a donor microorganism, Escherichia coli ATCC 15224, by plate mating with Ralstonia eutropha JMP134. Genes on this plasmid encode mercury resistance and partial 2,4-dichlorophenoxyacetic acid (2,4-D) degradation. The E. coli donor lacks the chromosomal genes necessary for mineralization of 2,4-D, and this fact allows presumptive transconjugants obtained in gene transfer studies to be selected by plating on media containing 2,4-D as the carbon source. Use of this donor counterselection approach enabled detection of plasmid pJP4 transfer to indigenous populations in soils and under conditions where it had previously not been detected. In Madera Canyon soil, the sizes of the populations of presumptive indigenous transconjugants were 10⁷ and 10⁸ transconjugants g of dry soil⁻¹ for samples supplemented with 500 and 1,000 μg of 2,4-D g of dry soil⁻¹, respectively. Enterobacterial repetitive intergenic consensus PCR analysis of transconjugants resulted in diverse molecular fingerprints. Biolog analysis showed that all of the transconjugants were members of the genus Burkholderia or the genus Pseudomonas. No mercury-resistant, 2,4-D-degrading microorganisms containing large plasmids or the tfdB gene were found in 2,4-D-amended uninoculated control microcosms. Thus, all of the 2,4-D-degrading isolates that contained a plasmid whose size was similar to the size of pJP4, contained the tfdB gene, and exhibited mercury resistance were considered transconjugants. In addition, slightly enhanced rates of 2,4-D degradation were observed at distinct times in soil that supported transconjugant populations compared to controls in which no gene transfer was detected.     

Plasmid Transfer Detection in Soil using the Inducible lPR System Fused to Eukaryotic Luciferase Genes

We report a model system for plasmid transfer analysis using the regulated lambda phage right promoter, λPr, fused to luc and lucOR as repoter genes. We have demonstrated that the systems cI857-λPr::luc and cI857-λPr::lucOR are temperature-inducible in Escherichia coli but not in other Gram-negative bacteria analyzed, enabling detection of luminescence when plasmids were mobilized from E. coli to those Gram-negative backgrounds. Using light for the detection, we have observed plasmid transfer from E. coli harboring RK2 and R388 derived plasmids to Pseudomonas putida KT2440 (co-introduced with donors) and to indigenous microorganisms, in vitro and in nonsterile soil microcosms. The importance of nutrients for an efficient plasmid transfer in nonsterile soil microcosms has been confirmed. When plasmid transfer experiments were carried out into nonsterile soil microcosms, significant populations of indigenous transconjugants arose. This system provides efficient marker genes and avoids the use of antibiotics for the selection of transconjugants.     

Methods for Detection of Conjugative Plasmid Transfer in Aquatic Environments

Donor and recipient counter selection was evaluated by selecting bacteria that received plasmid RP4 by conjugation on filters and in lake water microcosms. Three counter selection systems were compared; (i) Use of antibiotic-resistant recipients, (ii) use of an auxotrophic donor, and (iii) use of a donor with chromosomal suicide genes. Transfer efficiencies of transconjugants per recipient obtained with the three different counter selection systems in filter-matings were not significantly different. Some nalidixic acid-resistant recipients became partly sensitive to nalidixic acid after receiving the plasmid. Use of an auxotrophic donor was a feasible and easy way to recover indigenous transconjugants. A strain with two copies of the suicide gene gef was successfully eliminated in filter-matings, but elimination of the donor in microcosms by induction of the suicide genes did not succeed. Thus, this counter selection system was not usable in microcosm experiments. Received: 3 March 1998 / Accepted: 15 May 1998