JAC Advance Access originally published online on January 31, 2008
Journal of Antimicrobial Chemotherapy 2008 61(4):845-852; doi:10.1093/jac/dkn033
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Original research |
Selective pressure affects transfer and establishment of a Lactobacillus plantarum resistance plasmid in the gastrointestinal environment
1 Department of Microbiology and Risk Assessment, National Food Institute, Technical University of Denmark, Mørkhøj Bygade 19, DK-2860 Søborg, Denmark 2 Department of Bacteriology, Mycology and Parasitology, Statens Serum Institut, Artillerivej 5, DK-2300 Copenhagen S, Denmark 3 Center for Microbial Biotechnology, BioCentrum-DTU, Technical University of Denmark, DK-2800 Lyngby, Denmark 4 Chr. Hansen A/S, Bøge Allé 10-12, DK-2970 Hørsholm, Denmark
* Corresponding author. Tel: +45-7234-7185; Fax: +45-7234-7698; E-mail: anw{at}food.dtu.dk
Received 2 November 2007; returned 28 November 2007; revised 8 January 2008; accepted 9 January 2008
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
Objectives and methods: A Lactobacillus plantarum strain recently isolated from French raw-milk cheese was tested for its ability to transfer a small plasmid pLFE1 harbouring the erythromycin resistance gene erm(B) to Enterococcus faecalis. Mating was studied in vitro and in different gastrointestinal environments using gnotobiotic rats as a simple in vivo model and streptomycin-treated mice as a more complex model. Transfer and establishment of transconjugants in the intestine were investigated with and without selective pressure.
Results: Compared with the relatively low transfer frequency of 
5.7 x 10–8 transconjugants/recipient obtained in vitro by filter mating, a surprisingly high number of transconjugants (10–4 transconjugants/recipient) was observed in gnotobiotic rats even without antibiotic treatment. When erythromycin was administered, a transfer rate of 100% was observed, i.e. the recipient population turned completely into transconjugants (3 x 109 cfu/g faeces). Additionally, the time to reach a stable transconjugant population level was much faster in the erythromycin-treated gnotobiotic rats (1 day) than in the untreated animals (4–5 days). Transconjugants persisted in the gut in relatively stable numbers at least 12 days after termination of antibiotic treatment. In the streptomycin-treated mice, no transfer was observed either with or without erythromycin treatment.
Conclusions: The overall results imply that the gastrointestinal tract may comprise a more favourable environment for antibiotic resistance transfer than conditions provided in vitro. However, the indigenous gut microbiota severely restricts transfer, thus minimizing the number of detectable transfer events. Treatment with erythromycin strongly favoured transfer and establishment of pLFE1.
Keywords: L. plantarum , antibiotic resistance , horizontal gene transfer , gastrointestinal tract
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
During recent years, the question of whether lactobacilli can function as a pool for antibiotic resistance genes has gained much attention.1,2 Lactobacilli are industrially valuable bacteria, which are augmented in large viable numbers both in probiotic products and as starter cultures in a range of fermented milk and meat products. Lactobacilli also occur frequently in food produced from undefined starter cultures and as unintended contaminations. Many studies have selected and identified antibiotic-resistant species of lactobacilli from various environments encompassing different parts of the food chain, e.g. probiotics and starter cultures,3,4 fermented milk products,5–7 cheese,8,9 meat products10,11 and human isolates.12 When associated with mobile genetic elements such as plasmids or transposons, antibiotic resistance determinants can potentially be transferred to the commensal intestinal microbiota or pathogens transiently present in humans. Literature on conjugative transfer of naturally occurring antibiotic-resistance determinants in lactobacilli is nevertheless rare. Previously, transfer of native tetracycline and erythromycin resistance plasmids from two food-associated Lactobacillus plantarum strains to Enterococcus faecalis was observed in gnotobiotic rats.13 Gevers et al.14 showed in vitro conjugation from different species of lactobacilli to E. faecalis and Lactococcus lactis recipients but not to a Staphylococcus aureus recipient. In contrast, two other studies found no indications of intraspecies (Lactobacillus donor–recipient combinations) or interspecies (from lactobacilli to E. faecalis, Enterococcus faecium or L. lactis) resistance transfer.3,4 However, extrapolation of in vitro data to naturally occurring environments in the gut is not straightforward due to the limited knowledge of transfer mechanisms and factors affecting transferability within the gastrointestinal tract.
In this study, we examined the transfer of a native erythromycin resistance plasmid (pLFE1) from a L. plantarum strain isolated from a French raw-milk cheese to E. faecalis, a common inhabitant of the normal human gut microbiota, and also an opportunistic pathogen. Mating was investigated in two in vivo models with no or decreased colonization resistance represented by germ-free rats and streptomycin-treated mice, respectively, and in vitro by filter-mating. Additionally, the effect of different concentrations of erythromycin on the transfer rate and the persistence of pLFE1-carrying transconjugants in the gastrointestinal tract was investigated.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
Bacterial strains
The donor strain used in this study was L. plantarum M345 isolated in 2005 from a French raw-milk cheese (Tomme de Savoie) according to IDF standard 122C:1996.15 L. plantarum M345 harbours a small mobile plasmid (pLFE1) of 4 kb, which, among others, contains an erm(B) erythromycin resistance gene. For the in vivo experiment with streptomycin-treated mice, a spontaneously streptomycin-resistant mutant M345S of M345 was isolated and used as the donor. Before the experiment was initiated, the plasmid profile of both variants and in vitro transfer rate of pLFE1 was checked and no differences were observed between the streptomycin-resistant mutant and the wild-type strain. E. faecalis JH2SS showing resistance to streptomycin and spectinomycin (Strepr, Specr)16 was used as recipient strain. E. faecalis JH2-217 isogenic to E. faecalis JH2SS and resistant to rifampicin and fusidic acid was used as a recipient to test for secondary transfer of pLFE1 from transconjugants of E. faecalis JH2SS in vitro.
In vivo experiment with gnotobiotic rats and in vitro analyses. For all in vitro analyses and preparation of inoculation cultures for the gnotobiotic rat experiment, the donor strain was grown anaerobically at 37°C for 24–48 h in de Man, Rogosa and Sharpe (MRS) medium (Oxoid, Hampshire, UK) supplemented with 50 mg/L erythromycin (Sigma, Bornem, Belgium). For determination of donor cfu from the rat samples, Rogosa agar (Oxoid) supplemented with 50 mg/L erythromycin was used. The recipient was grown aerobically at 37°C for 24 h in brain heart infusion (BHI) (Oxoid) supplemented with 500 mg/L streptomycin and 500 mg/L spectinomycin (Sigma) for the in vitro analyses and inoculation culture for the gnotobiotic rat experiment. Selection of the recipient from the rat samples was performed on Slanetz and Bartley agar (Oxoid) supplemented with the appropriate antibiotics and incubated at 42°C for 48 h. Transconjugants were selected from the rat samples on Slanetz and Bartley agar (Oxoid) supplemented with streptomycin and spectinomycin (both at 500 mg/L) and 16 mg/L erythromycin and incubated under aerobic conditions at 42°C for 72 h [limit of detection (l.d.) = 20 cfu/g faeces]. In subsequent analyses, transconjugants were grown in BHI media at 37°C for 24–48 h.
In vivo experiment with streptomycin-treated mice. The donor strain was selected from mice faecal samples on MRS agar plates supplemented with 100 mg/L streptomycin and 50 mg/L erythromycin. The recipient was selected on Enterococcosel agar (BD Diagnostics) supplemented with 500 mg/L streptomycin and 500 mg/L spectinomycin. Plates selective for transconjugants were Enterococcosel agar containing 500 mg/L streptomycin, 500 mg/L spectinomycin and 50 mg/L erythromycin (l.d. = 20 cfu/g faeces), and Enterococcosel agar containing 500 mg/L spectinomycin and 50 mg/L erythromycin (l.d. = 10 cfu/g faeces). The plates were incubated at 37°C for 48 h (donor and recipient) or 72 h (transconjugants). The donor was incubated under anaerobic conditions and the recipient and transconjugants were incubated under aerobic conditions.
Animal management and in vivo set-up
The animal experiments were approved and conducted according to Danish national legislation.
Gnotobiotic rats.
Fifteen germ-free Sprague–Dawley rats (13 males and 2 females) were bred at the National Food Institute from parents originally supplied by Taconic (Germantown, NY, USA). The rats were 6 weeks old at the beginning of the experiment. Housing and feeding were as previously described,18 except the rats were caged two or three (belonging to the same group) together and walking on gratings. The rats were placed in three groups of five animals: group A, a control group (including the two female rats), which did not receive erythromycin; group B, a group treated with a low concentration of erythromycin; and group C, a group treated with a higher concentration of erythromycin. The high concentration of erythromycin used in the present study was based on the recommended clinical dose for children, which is calculated on the basis of body weight. This concentration was then scaled down to the size of the rats and mice. Furthermore, in order to mimic the intestinal absorption ways as best as possible, a therapeutic preparation of erythromycin (ABBOTICIN, Abbott Laboratories Ltd, Queensborough, UK) was used, which contains erythromycin in the form of erythromycinethylsuccinate (for concentrations, see below).
At day 0, all rats received 4 x 108 cfu of the recipient strain E. faecalis JH2SS per os by gavage. Six days after introduction of the recipient, the donor strain was introduced. During five consecutive days, bacteria from a fresh overnight culture of the donor strain were harvested and resuspended in phosphate-buffered saline (PBS) (Oxoid) and offered instead of drinking water. The rats were allowed free access to the solution, which contained 3 x 1010 cfu/mL donor bacteria.
During the period of donor dosing, the rats in groups B and C received two daily doses each of 0.25 mg/animal and 2.5 mg/animal erythromycin, respectively. The erythromycin was given per os by gavage. Fourteen days after the end of erythromycin treatment and donor dosing, all rats were euthanized at day 24.
Streptomycin-treated mice. Twenty-four outbred albino female NMRi and 24 inbred BALB/c mice aged 6–8 weeks were supplied from Taconic (Ejby, Denmark). The mice were caged in pairs and the cages were changed daily. From day ‘–1’, the mice continuously received drinking water containing 1 mg/mL streptomycin sulphate (Sigma, St Louis, MO, USA). The animals from each breed were placed in four groups of six mice: group A, only received the recipient; group B, received the recipient and erythromycin treatment; group C, received the recipient and the donor; and group D, received the recipient, the donor and erythromycin treatment.
The selectivity of the applied agar plates for donor, recipient and transconjugants cultivation from faecal samples was confirmed prior to recipient inoculation (day –1). Owing to sporadic growth of faecal bacteria on the recipient plates, two mice were excluded from the experiment. The recipient (108 cfu) was inoculated to all mice at day 0 and re-inoculated again at day 10. Starting at day 7, mice belonging to groups C and D were inoculated daily with 5.0 x 108 cfu of the donor strain. During the period of donor dosing, mice in groups B and D were given one daily dose of 0.025 mg erythromycin (ABBOTICIN). All bacterial inoculums were overnight cultures washed and resuspended in 20% (w/v) sucrose (Sigma) and all inoculations were done per os. On day 17, all mice were euthanized.
Collection and treatment of in vivo samples
Gnotobiotic rats. To verify the germ-free status of the rats, faecal samples were collected and analysed for aerobic and anaerobic growth of bacteria and yeasts before the experiment was initiated. Faecal samples were collected directly from the rats by careful squeezing of their abdomen. At euthanization, samples from duodenum, ileum, caecum and colon were taken. All samples were homogenized by whirly mixing in PBS. Ten-fold dilution series were prepared in PBS and incubated on the appropriate selective agar-plates for enumeration of donors, recipients and transconjugants as described above.
Streptomycin-treated mice. Faecal samples (0.5 g/cage, originating from two animals) shed within 24 h were collected. Numbers of cfu of donors, recipients and transconjugants were determined on the selective plates as described above.
Verification of transconjugants by PCR
Selection of transconjugants was based on their phenotypic resistance profile. In order to confirm that these isolates were in fact true transconjugants and not mutants, PCR assays with primers species-specific for E. faecalis and targeting the erm(B) resistance gene were carried out as previously described.13 Altogether, 45 transconjugant isolates from different animals, groups and days during the study were tested. The donor and recipient strain were included as positive and negative controls.
The MIC of erythromycin was determined for nine transconjugants isolated from the animals as well as for L. plantarum M345 and E. faecalis JH2SS. Overnight cultures were prepared in broth according to standard conditions. The cultures were streaked onto Mueller–Hinton (BBL, Sparks, USA) agar plates for transconjugants and the recipient and on MRS agar plate for the donor, respectively. Etest strips (AB Biodisk, Solna, Sweden) were applied and the plates were read after overnight incubation at appropriate conditions for donor, recipient and transconjugants.
Plasmid profiles and Southern blotting
Plasmid DNA was isolated from nine transconjugants (representing nine rats and three groups) and the donor and recipient strain. We used the QIAprep Spin Miniprep Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol, except that an initial lysozyme step (2 mg/mL lysozyme in P1 buffer for 25 min at 37°C) was included. The plasmid extractions were cut with XcmI (New England Biolabs), which cuts pLFE1 uniquely outside the erm(B) probe (sequencing results, unpublished results) and both undigested and digested profiles were run on a 0.7% agarose gel. The gel was blotted and hybridized with an erm(B) probe as previously described.13
Secondary transfer of erm(B) plasmid pLFE1 from transconjugants
The ability of E. faecalis JH2SS transconjugants obtained from the animals to function as new donors of the erm(B) gene was tested by a filter mating assay. Twenty transconjugant isolates verified by PCR were assessed as donors and E. faecalis JH2-2 (Rifr, Fusr) was used as an isogenic recipient. The strains were grown in BHI broth supplemented with the respective antibiotics to an OD600 = 0.7–1.0. Four donor mixtures were then prepared from the 20 transconjugants by mixing each of five transconjugant cultures together in equal volumes. Thereafter, donor mixture and recipient culture were mixed in equal volumes and vacuum-filtered onto sterile filters (HAWP04 700, Millipore, Bedford, MA, USA), which were incubated under aerobic conditions on non-selective BHI agar plates at 37°C overnight. The bacteria were washed off the filters with PBS and appropriate dilutions spread onto plates selective for the donor (BHI + 500 mg/L streptomycin, 500 mg/L spectinomycin and 16 mg/L erythromycin), recipient (BHI + 50 mg/L rifampicin and 25 mg/L fusidic acid) and transconjugants (BHI+16 mg/L erythromycin, 50 mg/L rifampicin and 25 mg/L fusidic acid). As control, the donor mixture and the recipient culture were each mixed with PBS and placed separately on filters.
Effect of erythromycin on transfer frequency in vitro
The effect of erythromycin concentration on erm(B) plasmid transfer frequency from L. plantarum M345 to E. faecalis JH2SS in vitro was tested. Four different concentrations 0.05, 0.5, 1 and 5 mg/L erythromycin were added to the BHI mating plates. A control without erythromycin was also included. The donor and recipient strains were grown in broth supplemented with the respective antibiotics to OD600 = 0.9–1.1 The bacteria were mixed in equal volumes and vacuum filtered onto sterile filters (HAWP04 700), which were placed on the mating plates containing the various concentrations of erythromycin. The plates were incubated under aerobic conditions at 37°C overnight and the bacteria subsequently washed off and prepared on plates selective for the donor, recipient and transconjugants. Controls of separate donor and recipient plates were also prepared. The experiment was performed in five replicates.
The effect of erythromycin treatment on the number of recipients and transconjugants in the in vivo experiments was calculated using the analysis of variance with interaction. The number of recipients or transconjugants was set as the response variable, and for the gnotobiotic rat experiment, rats were specified as random effect, and group and time as fixed effects. A similar analysis was made for the streptomycin-treated mice, where cages were specified as random effect and mouse-breed, time and donor inoculation as fixed effects. For pair-wise testing, Student's t-test was used.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
Conjugal transfer in gnotobiotic rats
Transferability of a wild-type erm(B) plasmid from L. plantarum M345 to E. faecalis JH2SS was studied in gnotobiotic rats. Strain E. faecalis JH2SS was given as a single dose to all rats at day 0. The strain readily colonized the rats at levels of 5 x 109 cfu/g faeces (Figure 1). At days 6–10, the donor strain was introduced via the drinking water. During the same period, rats belonging to groups B and C received erythromycin in medium and high concentration, respectively, whereas rats belonging to group A received no antibiotic treatment. Erythromycin had a major effect on the rate of establishment as well as on the size of transconjugant populations obtained. Two days after introduction of the donor strain, transconjugants were detected in rats from all groups (Figure 1). When erythromycin was administered, the increase in numbers of transconjugants occurred much faster (1 day) than that observed in animals without antibiotic treatment (4–5 days), after which a stable population level was reached. Furthermore, the size of the transconjugant population, which was established in the rats, was approximately 4 log units higher in groups B and C (3 x 109 cfu/g faeces) than in group A (5 x 105 cfu/g faeces), the former being an almost complete turnover of recipients into transconjugants (Figure 1).
|
Not surprisingly, erythromycin affected the recipients. After initiation of erythromycin treatment, the number of recipients decreased significantly both in groups B (P < 0.01) and C (P = 0.001). A significantly larger decrease (P < 0.05) in the number of recipients was observed in the rats treated with a high concentration of erythromycin than in rats treated with a medium concentration of the antibiotic. However, after establishment, no significant (P = 0.73) differences in transconjugant numbers could be seen between rats treated with high and medium concentrations of erythromycin (Figure 1). Erythromycin did not seem to have any effect on colonization of the donor strain. Thus, the donor was retrieved from rat faeces in numbers of
109 cfu/g faeces, irrespective of erythromycin treatment (Figure 1). After discontinuation of donor and erythromycin dosing at day 10, the number of donors decreased to 2 x 108 cfu/g faeces in all rats and remained relatively constant throughout the rest of the study. Cultivation of bacteria from intestinal segments
The numbers of donors, recipients and transconjugants were higher in the upper end of the intestine (duodenum and ileum) than in the lower intestinal segments (caecum and colon) (Table 1). Transconjugants were present in all segments indicating that transfer occurred relatively soon in the gastrointestinal tract. The ratio between donors, recipients and transconjugants was approximately the same in all segments.
|
Verification of transconjugants
PCR was carried out using primers specific for the erm(B) gene and for E. faecalis to verify the transconjugants that were selected during the in vivo experiments solely by their resistance phenotype. All 45 transconjugants subjected to PCR analysis gave positive products both with the erm(B) and the E. faecalis specific primers (data not shown). As expected, L. plantarum M345 and E. faecalis JH2SS gave positive and negative reactions, respectively, with the erm(B) PCR primers. The opposite result was obtained with E. faecalis specific primers, confirming the reliability of the analysis. MIC of erythromycin was determined for L. plantarum M345, E. faecalis JH2SS and nine transconjugants from the rats. All transconjugants were resistant to erythromycin with an MIC >256 mg/L. The recipient strain was sensitive (MIC 0.25 mg/L) and the donor was resistant (MIC > 256 mg/L).
Plasmid profiles and Southern blotting
Plasmids were extracted from the donor L. plantarum M345, the recipient E. faecalis JH2SS and nine transconjugant isolates isolated from the gnotobiotic rat samples. The plasmid profiles were digested with XcmI, blotted and hybridized with an erm(B)-specific probe. In the recipient strain, no plasmids were identified. In contrast, the donor and all transconjugants contained a plasmid band of 4 kb, which hybridized with the erm(B) probe (Figure 2). In the digested profiles, the 4 kb band appears sharp and distinct, whereas the undigested profiles show a larger smear. This smear is presumably due to various forms of the plasmid migrating with different velocities during electrophoresis.
|
A second band of
5 kb, which the erm(B)-specific probe did not hybridize with, was clearly present in the transconjugant-digested profiles. As the donor harbours several other plasmids in addition to pLFE1 (Figure 2), we believe that the 5 kb band originates from another co-transferred plasmid. Secondary transfer of erm(B) plasmid in vitro
The potential of transconjugants obtained in vivo to transfer the erm(B) plasmid to new recipients was investigated by filter mating. However, under the experimental conditions applied (detection limit was 10–9 transconjugants/recipient), no transfer was detected to the recipient E. faecalis JH2-2 (data not shown).
Effect of erythromycin on in vitro transfer frequency
The effect of erythromycin on selection of transconjugants was investigated in vitro by application of different concentrations of erythromycin to the agar plates used during mating. The number of transconjugants increased significantly when low concentrations of erythromycin were added (Table 2). The highest transfer frequency of 2.0 x 10–6 transconjugants/recipient was observed at 0.50 mg/L erythromycin. At higher erythromycin concentrations, the frequency decreased again to levels comparable to the control. The recipient was significantly (P < 0.05) inhibited by 
0.50 mg/L erythromycin, as also expected from the MIC value of 0.25 mg/L. In contrast, the donor strain was unaffected by erythromycin.
|
Conjugation study in streptomycin-treated mice
Two different breeds of mice NMRi and BALB/c were used to study conjugal transfer in a gastrointestinal environment with a complex microbiota, where the colonization barrier was reduced by streptomycin treatment. Similar colonization levels of the donor and recipient strain were observed for the two breeds (data not shown). The recipient was introduced in all mice at day 0 and colonized at least transiently at relatively high levels. At day 1, 109 cfu/g faeces were recovered from faeces. During the next 1–2 days, the cfu numbers decreased 1 log unit. For the remaining part of the study, the population number was constant to slightly decreasing; thus at the end of the experiment (day 17), 107 cfu/g faeces of the recipient was recovered from mice not treated with erythromycin (groups A and C, data not shown). In contrast, treatment with erythromycin (groups B and D) from day 7 resulted in a significant (P < 0.001) decrease in the recipient of 2–3 log units. Despite an increase after re-inoculation with the recipient at day 10, the numbers decreased to 105 cfu/g faeces at day 17 in the erythromycin-treated groups (data not shown). The donor was introduced in the mice (groups C and D) at day 7 and was repeatedly inoculated each day throughout the experiment. As expected, erythromycin treatment did not affect the donor colonization, which remained relatively constant between 106 and 107 cfu/g faeces in all mice. No transconjugants were detected in this study (l.d. = 10 cfu/g faeces) (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
In the present study, mobilization of a small, naturally occurring erythromycin resistance plasmid from L. plantarum was studied in vitro, as well as in two different gastrointestinal in vivo models. Gnotobiotic rats were used as a high transfer model to evaluate the transfer potential of the L. plantarum donor strain in a gastrointestinal environment without a colonization barrier. In this model, a high concentration of transconjugants was observed even when no selective pressure was applied (Figure 1, group A). Interestingly, the ratio of transconjugants to recipients was 3–4 log units higher in the gnotobiotic model than in vitro. This result is in agreement with a recent study of Dahl et al.,19 investigating conjugation between different strains of E. faecium. In 9 out of 12 mating pairs, they observed higher ratios of transconjugants to recipients in gnotobiotic mice than in a filter-mating assay. They therefore suggested that transfer rates determined in vitro underestimate the transfer potential in the mammalian gastrointestinal tract. However, in a recent study, the opposite relationship between in vitro and in vivo transconjugants to recipient ratios was observed. In vitro transfer frequencies of plasmids harbouring tet(M) and erm(B) resistance genes from two strains of L. plantarum to E. faecalis were 3–4 log units higher than the ratio observed in gnotobiotic rats.13 A possible explanation for these differences could be the colonization ability of the donor strain, which in the current study reached a population level of 108–109 cfu/g faeces, whereas in the earlier study, only 106–107 cfu/g faeces of the donor strains were observed.13 A high donor density will in theory increase the encounters of donors and recipients, and thus increase mating events.20 However, a vast number of other factors have also been proposed to affect the development of a specific plasmid-carrying population in the gut; for instance, the donor potential or the genetic advantages/disadvantages of the specific plasmid to the recipient population.21 Thus, the efficiency of plasmid dissemination in the gastrointestinal tract cannot be predicted from extrapolation of in vitro mating experiments but requires more complex environments.
Erythromycin treatment had a very pronounced effect on dissemination of the erm(B) resistance plasmid pLFE1 in the gnotobiotic gut (Figure 1, groups B and C). We hypothesize that the increase in plasmid-carrying recipients was not only the result of a growth advantage during selective pressure, but also a direct effect on the number of transfer events. The transconjugants found in the in vivo samples could have evolved both horizontally from transfer and vertically by growth of the transconjugants. However, the relative contribution from each of these factors cannot be deduced from the present experimental set-up. Nevertheless, there is an upper limit to the rate with which the bacteria can grow and thus to the maximum contribution from the vertical factor. In the erythromycin-treated gnotobiotic rats, the initial increase in transconjugant numbers occurred very fast compared with the untreated rats (Figure 1). Calculating the maximum generation time (disregarding transfer and the intestinal bacterial dilution rate) for the first day where transconjugants are detected, it would be 60 and 300 min for the erythromycin-treated and untreated rats, respectively. If a selective growth advantage was the only factor responsible for the increase in transconjugant numbers, the generation time in the erythromycin-treated rats should be well below 1 h when the bacterial dilution rate is taken into account. Therefore, it is speculated that an increased plasmid transfer rate has contributed to the fast increase in transconjugant numbers in the erythromycin-treated rats.
There are only few reports on the effect of erythromycin on transfer of erythromycin resistance determinants. Nevertheless, the presented results are in good agreement with those of Igimi et al.,22 who examined the transfer of pAMβ1 from L. lactis to bacteria belonging to the ‘Schaedler flora’ in previously germ-free mice. When treatment with erythromycin was initiated, an increase in E. faecalis harbouring pAMβ1 from below detection limit to 107 cfu/g faeces within 1 day was observed.22 Likewise, Morelli et al.23 observed an improved plasmid dissemination efficiency from 10–7 to 10–4 transconjugants/donor after initiation of erythromycin treatment in gnotobiotic mice colonized with pAMβ1-bearing Lactobacillus reuteri donor strain and E. faecalis recipient. However, to our knowledge, there are no reports describing selection-induced transfer by erythromycin, as is known for conjugative transposons carrying tet(M).24–26
Also in vitro, an effect of low concentration of erythromycin was detected, which was abolished at high concentration. Addition of 0.5 mg/L erythromycin to the in vitro mating plates resulted in a factor of 30 increase in the ratio of transconjugants to recipients compared with the control, whereas 5 mg/L erythromycin in the mating plates resulted in a ratio similar to the control (Table 2). The selection for transconjugants by low concentrations of erythromycin in the mating medium could again be due to two things: (i) increased transfer frequency and/or (ii) a growth advantage of transconjugants. We believe that inhibition of the recipient at higher erythromycin concentrations restricts transfer, as reported previously,27,28 and thus counteracts the selective effects of erythromycin on transconjugants exerted at lower concentrations. In contrast to the results described here, Morelli et al.23 did not see any effect of adding 1 mg/L erythromycin to the mating plates in the transfer of pAMβ1 from L. reuteri to E. faecalis JH2SS. These diverging results could possibly be explained by (i) the donor/plasmid examined by Morelli et al. did not respond to erythromycin by induction of transfer or (ii) the selection for transconjugants over recipients depends on their respective growth rates, which are affected by the specific R-plasmid and the erythromycin concentration. The latter has, for instance, been shown for different R-plasmids, where the equilibrium between susceptible and resistant populations in the gut was changed with different concentrations of the antibiotic.29
In the gnotobiotic gut, the transconjugants persisted in stable numbers for at least 12 days after the end of erythromycin treatment. This indicates that either the resistance plasmid did not represent a noteworthy competitive disadvantage in the absence of antibiotic selection and/or the plasmid had a high rate of transfer. In other in vivo studies looking at transconjugant persistence without selective pressure, both relatively stable30–33 and rapidly decreasing,34,35 transconjugant populations have been reported. The fact that the recipient population was well established in the intestine prior to donor introduction could be an important competitive parameter for the newly developed transconjugant population. Thus, the recipients who acquire the plasmid do not need to compete for available niches since they are already attached to the intestinal mucosa.
The ability to conjugate is another parameter shown to be important for the maintenance of a plasmid-carrying population in the gut.36 Whether secondary transfer of the erm(B) plasmid from transconjugants to recipients contributed to the stability of the plasmid in the gnotobiotic model is not known. It was tried to elucidate the potential of the transconjugants to act as new donors by in vitro mating to an isogenic recipient. Although no transfer was observed by this method, it cannot be excluded that the transfer can take place in the gastrointestinal tract. Nevertheless, it is believed that the L. plantarum donor strain harbours genes for conjugation of the small mobilizable erm(B) plasmid either on the chromosome or on a larger plasmid, which either are not co-transferred or are co-transferred but not maintained in the E. faecalis recipient.
In the gnotobiotic rats, a frequency of up to 1 transconjugant/recipient was observed, demonstrating that the L. plantarum strain maintains an active transfer apparatus after passage of the stomach and upper intestine. In contrast, no transconjugants were detected in the more complex model using streptomycin-treated mice, where transfer was assessed in the presence of a surrounding microbiota. This indicates that the presence of the indigenous intestinal bacteria efficiently reduces the formation of effective mating pairs. This finding is in agreement with previous investigations that have found a lower transfer rate in animals with a colonization barrier than in a gnotobiotic intestinal environment.37 However, lacking detection of transfer in the streptomycin-treated mice does not exclude that transfer took place, but suggests that it was a rare event.
Despite the fact that a relatively low transfer frequency between L. plantarum and E. faecalis was observed in vitro, this study shows that a surprisingly high concentration of transconjugants can be obtained in the gastrointestinal tract of gnotobiotic rats. This concentration can be considerably further raised during treatment with erythromycin.
|
Funding |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
This work was supported by the European Commission grant CT-2003-506214 (ACE-ART) under the 6th framework programme.
|
Transparency declarations |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
None to declare.
|
Acknowledgements |
|---|
Part of this work was presented at the Fourth Probiotics, Prebiotics & New Foods, Rome, 2007. We wish to thank Helle Lindgaard Madsen who originally isolated the L. plantarum M345 donor strain. Furthermore, we thank Anne Ørngren and her department, especially Kenneth Worm, for professional handling of the gnotobiotic rats. We are grateful for the excellent technical assistance offered by Kate Vibefeldt and Bodil Madsen, and for statistical calculations provided by Tina Beck Hansen.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionFundingTransparency declarationsReferences |
|---|
1 Teuber M, Meile L, Schwarz F. Acquired antibiotic resistance in lactic acid bacteria from food. Antonie Van Leeuwenhoek (1999) 76:115–37.[CrossRef][Web of Science][Medline]
2 Ammor MS, Floréz AB, Mayo B. Antibiotic resistance in non-enterococcal lactic acid bacteria and bifidobacteria. Food Microbiol (2007) 24:559–70.[CrossRef][Web of Science][Medline]
3
Hummel AS, Hertel C, Holzapfel WH, et al. Antibiotic resistances of starter and probiotic strains of lactic acid bacteria. Appl Environ Microbiol (2007) 73:730–9.
4
Klare I, Konstabel C, Werner G, et al. Antimicrobial susceptibilities of Lactobacillus, Pediococcus and Lactococcus human isolates and cultures intended for probiotic or nutritional use. J Antimicrob Chemother (2007) 59:900–12.
5 Katla AK, Kruse H, Johnsen G, et al. Antimicrobial susceptibility of starter culture bacteria used in Norwegian dairy products. Int J Food Microbiol (2001) 67:147–52.[CrossRef][Web of Science][Medline]
6 Temmerman R, Pot B, Huys G, et al. Identification and antibiotic susceptibility of bacterial isolates from probiotic products. Int J Food Microbiol (2003) 81:1–10.[CrossRef][Web of Science][Medline]
7 D'Aimmo MR, Modesto M, Biavati B. Antibiotic resistance of lactic acid bacteria and Bifidobacterium spp. isolated from dairy and pharmaceutical products. Int J Food Microbiol (2007) 115:35–42.[CrossRef][Web of Science][Medline]
8 Herrero M, Mayo B, Gonzalez B, et al. Evaluation of technologically important traits in lactic acid bacteria isolated from spontaneous fermentations. J Appl Bacteriol (1996) 81:565–70.
9 Floréz AB, Delgado S, Mayo B. Antimicrobial susceptibility of lactic acid bacteria isolated from a cheese environment. Can J Microbiol (2005) 51:51–8.[Medline]
10 Aquilanti L, Garofalo C, Osimani A, et al. Isolation and molecular characterization of antibiotic-resistant lactic acid bacteria from poultry and swine meat products. J Food Prot (2007) 70:557–65.[Web of Science][Medline]
11 Gevers D, Huys G, Devlieghere F, et al. Isolation and identification of tetracycline resistant lactic acid bacteria from pre-packed sliced meat products. Syst Appl Microbiol (2000) 23:279–84.[Web of Science][Medline]
12 Catatolok O, Gogebakan B. Presence of drug resistance in intestinal lactobacilli of dairy and human origin in Turkey. FEMS Microbiol Lett (2004) 236:7–12.[Web of Science][Medline]
13 Jacobsen L, Wilcks A, Hammer K, et al. 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. FEMS Microbiol Ecol (2007) 59:158–66.[CrossRef][Medline]
14 Gevers D, Huys G, Swings J. In vitro conjugal transfer of tetracycline resistance from Lactobacillus isolates to other Gram-positive bacteria. FEMS Microbiol Lett (2003) 225:125–30.[CrossRef][Web of Science][Medline]
15 International Dairy Federation. International IDF standard 122C:1996. Milk and milk products, preparation of samples and dilutions for microbiological examination. (1996).
16
Tomich PK, An FY, Clewell DB. Properties of erythromycin-inducible transposon Tn917 in Streptococcus faecalis. J Bacteriol (1980) 141:1366–74.
17 Jacob AE, Hobbs SJ. Conjugal transfer of plasmid-borne multiple antibiotic resistance in Streptococcus faecalis var. zymogenes. J Bacteriol (1974) 117:360–72.
18 Wilcks A, van Hoek AHAM, Joosten RG, et al. Persistence of DNA studied in different ex vivo and in vivo rat models simulating the human gut situation. Food Chem Toxicol (2004) 42:493–502.[CrossRef][Web of Science][Medline]
19
Dahl KH, Mater DD, Flores MJ, et al. Transfer of plasmid and chromosomal glycopeptide resistance determinants occurs more readily in the digestive tract of mice than in vitro and exconjugants can persist stably in vivo in the absence of glycopeptide selection. J Antimicrob Chemother (2007) 59:478–86.
20
Licht TR, Christensen BB, Krogfelt KA, et al. Plasmid transfer in the animal intestine and other dynamic bacterial populations: the role of community structure and environment. Microbiology (1999) 145:2615–22.
21 Licht TR, Wilcks A. Conjugative gene transfer in the gastrointestinal environment. Adv Appl Microbiol (2005) 58C:77–95.[Web of Science]
22 Igimi S, Ryu CH, Park SH, et al. Transfer of conjugative plasmid pAMβ1 from Lactococcus lactis to mouse intestinal bacteria. Lett Appl Microbiol (1995) 23:31–5.[Web of Science]
23 Morelli L, Sarra PG, Bottazzi V. In vivo transfer of pAMβ1 from Lactococcus reuteri to Enteroccus faecalis. J Appl Bacteriol (1988) 65:371–5.[Medline]
24
Doucet-Populaire F, Trieu-Cuot P, Dosbaa I, et al. Inducible transfer of conjugative transposon Tn1545 from Enterococcus faecalis to Listeria monocytogenes in the digestive tracts of gnotobiotic mice. Antimicrob Agents Chemother (1991) 35:185–7.
25 Showsh SA, Andrews RE Jr. Tetracycline enhances Tn916-mediated conjugal transfer. Plasmid (1992) 28:213–24.[CrossRef][Web of Science][Medline]
26 Torres OR, Korman RZ, Zahler SA, et al. The conjugative transposon Tn925: enhancement of conjugal transfer by tetracycline in Enterococcus faecalis and mobilization of chromosomal genes in Bacillus subtilis and E. faecalis. Mol Gen Genet (1991) 225:395–400.[CrossRef][Web of Science][Medline]
27
Sasaki Y, Taketomo N, Sasaki T. Factors affecting transfer frequency of pAMβ1 from Streptococcus faecalis to Lactobacillus plantarum. J Bacteriol (1988) 170:5939–42.
28 Licht TR, Struve C, Christensen BB, et al. Evidence of increased spread and establishment of plasmid RP4 in the intestine under sub-inhibitory tetracycline concentrations. FEMS Microbiol Ecol (2003) 44:217–23.[CrossRef][Medline]
29
Corpet DE, Lumeau S, Corpet F. Minimum antibiotic levels for selecting a resistance plasmid in a gnotobiotic animal model. Antimicrob Agents Chemother (1989) 33:535–40.
30
Bahl MI, Sørensen SJ, Hansen LH, et al. Effect of tetracycline on transfer and establishment of the tetracycline-inducible conjugative transposon Tn916 in the guts of gnotobiotic rats. Appl Environ Microbiol (2004) 70:758–64.
31 Jacobsen BL, Skou M, Hammerum AM, et al. Horizontal transfer of the satA gene encoding streptogramin A resistance between isogenic Enterococcus faecium strains in the gastrointestinal tract of gnotobiotic rats. Microb Ecol Health Dis (1999) 11:241–7.[CrossRef]
32
Licht TR, Laugesen D, Jensen LB, et al. Transfer of the pheromone-inducible plasmid pCF10 among Enterococcus faecalis microorganisms colonizing the intestine of mini-pigs. Appl Environ Microbiol (2002) 68:187–93.
33 Alpert CA, Mater DD, Muller M, et al. Worst-case scenarios for horizontal gene transfer from Lactococcus lactis carrying heterologous genes to Enterococcus faecalis in the digestive tract of gnotobiotic mice. Environ Biosafety Res (2003) 2:173–80.[CrossRef][Medline]
34
Lester CH, Frimodt-Moller N, Sorensen TL, et al. In vivo transfer of the vanA resistance gene from an Enterococcus faecium isolate of animal origin to an E. faecium isolate of human origin in the intestines of human volunteers. Antimicrob Agents Chemother (2006) 50:596–9.
35 Gruzza M, Fons M, Ouriet MF, et al. Study of gene transfer in vitro and in the digestive tract of gnotobiotic mice from Lactococcus lactis strains to various strains belonging to human intestinal biota. Microb Releases (1994) 2:183–9.[Medline]
36
Bahl MI, Hansen LH, Licht TR, et al. Conjugative transfer facilitates stable maintenance of IncP-1 plasmid pKJK5 in Escherichia coli cells colonizing the gastrointestinal tract of the germfree rat. Appl Environ Microbiol (2007) 73:341–3.
37 Schlundt J, Saabye P, Lohmann B, et al. Conjugal transfer of plasmid DNA between Lactococcus lactis strains and distribution of transconjugants in the digestive tract of gnotobiotic rats. Microb Ecol Health Dis (1994) 7:59–69.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. E. Galen, J. Y. Wang, M. Chinchilla, C. Vindurampulle, J. E. Vogel, H. Levy, W. C. Blackwelder, M. F. Pasetti, and M. M. Levine A New Generation of Stable, Nonantibiotic, Low-Copy-Number Plasmids Improves Immune Responses to Foreign Antigens in Salmonella enterica Serovar Typhi Live Vectors Infect. Immun., January 1, 2010; 78(1): 337 - 347. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||