- Research article
- Open Access
An investigation of horizontal transfer of feed introduced DNA to the aerobic microbiota of the gastrointestinal tract of rats
BMC Research Notesvolume 5, Article number: 170 (2012)
Horizontal gene transfer through natural transformation of members of the microbiota of the lower gastrointestinal tract (GIT) of mammals has not yet been described. Insufficient DNA sequence similarity for homologous recombination to occur has been identified as the major barrier to interspecies transfer of chromosomal DNA in bacteria. In this study we determined if regions of high DNA similarity between the genomes of the indigenous bacteria in the GIT of rats and feed introduced DNA could lead to homologous recombination and acquisition of antibiotic resistance genes.
Plasmid DNA with two resistance genes (npt I and aad A) and regions of high DNA similarity to 16S rRNA and 23S rRNA genes present in a broad range of bacterial species present in the GIT, were constructed and added to standard rat feed. Six rats, with a normal microbiota, were fed DNA containing pellets daily over four days before sampling of the microbiota from the different GI compartments (stomach, small intestine, cecum and colon). In addition, two rats were included as negative controls. Antibiotic resistant colonies growing on selective media were screened for recombination with feed introduced DNA by PCR targeting unique sites in the putatively recombined regions. No transformants were identified among 441 tested isolates.
The analyses showed that extensive ingestion of DNA (100 μg plasmid) per day did not lead to increased proportions of kanamycin resistant bacteria, nor did it produce detectable transformants among the aerobic microbiota examined for 6 rats (detection limit < 1 transformant per 1,1 × 108 cultured bacteria). The key methodological challenges to HGT detection in animal feedings trials are identified and discussed. This study is consistent with other studies suggesting natural transformation is not detectable in the GIT of mammals.
Relatively few studies have examined the occurrence of horizontal gene transfer (HGT) by natural transformation in the gastrointestinal tract (GIT) of mammals under in vivo conditions [1–5]. It is unclear if the lack of observable competence among bacteria in the GIT is due to experimental limitations, limited occurrence in the few model bacteria examined, or due to true lack of conditions conductive for the development of competence among members of the GIT microbiota of various mammals.
There are several requirements for natural transformation to take place with non-mobile DNA in the GIT [6, 7]. First, bacteria must be able to express a competent stage while in the GIT system. Numerous bacterial species present in the GIT are known to be able to develop natural competence when grown in vitro[8–10]. For instance, species within the genera Bacillus, Campylobacter, Helicobacter, Lactobacillus, Neisseria, Pseudomonas, Streptococcus, and Vibrio are known to express competence under specific laboratory conditions. However, so far only few in vivo studies have reported on the occurrence of bacterial competence in the GIT and those are limited to the upper GIT. The lower part of the GIT has nevertheless been suggested to be the most active site for HGT processes due to abundance of nutrients, high bacterial density and slower degradation rates of DNA [11, 12].
Second, extracellular DNA of sufficient length and concentration must be present and accessible to competent bacteria. Several experimental studies have investigated the stability of DNA in the GIT of mammals ([2–4, 13–20]) suggesting that minor fractions of DNA provided with different feed sources may remain in various GIT compartments. For a recent review, see Rizzi et al.. Moreover, that minor portions of such DNA can remain of sufficient size for the acquisition of functional traits (e.g. new protein coding sequences) by competent bacteria.
Third, chromosomal DNA fragments taken up over the cytoplasmic membrane must be able to recombine with the bacterial host chromosome for stable inheritance and vertical transmission [21–23]. In general, incoming DNA must contain regions of minimum 25-200 bp in length of high similarity to the recipient genome for homologous recombination to occur [24–28]. These prerequisites are met by ribosomal gene sequences, which are sufficient in length and degree of sequence conservation for homologous recombination events .
Several studies have investigated conditions for natural transformation in the upper GIT. The oral cavity is the first place feed introduced DNA enters the GIT and therefore also contains the highest amount of DNA ingested. In general, a highly time-limited stability of DNA and ability to transform defined model bacteria introduced to saliva or stomach fluids has been observed in vitro and in vivo[16, 30–34]. Only few studies have investigated the occurrence of natural transformation in the compartments of the lower GIT. Wilcks and colleagues [2, 3] reported some persistence of introduced DNA in the GIT of germ-free/gnotobiotic or mono-associated rats, but did not detect uptake of plasmid DNA by E. coli, Bacillus subtilis, or Streptococcus gordonii. In another study, Nordgård et al.,  examined the ability of Acinetobacter baylyi colonized gnotobiotic mice and rats for potential in vivo transformation of bacteria with feed introduced DNA. No transformants were detected in vivo or in vitro. In the only in vivo study of natural transformation of bacteria present in the GIT of humans published so far, only limited persistence of food-derived plant DNA was found . The DNA did not survive passage through the GIT of healthy human subjects as determined by PCR. Three out of these seven human volunteers (ileostomists) showed evidence of low-frequency HGT of plant DNA (the epsps transgene) to the microbiota of the small bowel, but this appeared to have occurred before feeding the experimental meal and no single microbial isolate could be obtained for further analysis. With the exception of the in vivo study in human described above, most published studies focusing on natural transformation in the GIT have used experimental models aimed at detecting HGT events occurring into single bacterial strains introduced into the GIT system.
The aim of this study was to examine if naturally occurring aerobic bacteria in the GIT of rats can undergo natural transformation and recombination with feed introduced DNA. We hypothesized that regions of high DNA similarity between the introduced DNA and the indigenous bacteria in the rat GIT would lead to stable integration of two selectable marker genes (streptomycin or kanamycin resistance) based on homologous recombination. Moreover, that such events occurred at a frequency that was detectable by direct selective plating. Plasmid DNA with 16S rRNA and 23S rRNA sequences flanking the two antibiotic resistance marker genes were mixed with standard rat feed pellets (Figure 1). A double-sided homologous recombination event with sequence-similar rRNA genes in competent bacteria in the rat GIT would lead to integration of the antibiotic resistance genes (Figure 2). Recombinant-specific primer annealing sites were chosen to enable unambiguous detection of transformants carrying chromosomal integrations of the resistance genes among the overall population of bacteria recovered from the GIT.
The plasmids (purified from E. coli) were highly potent in generating transformants when incubated with naturally competent A. baylyi cells in vitro on LB agar plates. Transformation frequencies of above 1,4 × 10-5 were obtained per 0,2 μg plasmid (data not shown): similar to previous in vitro studies with the same experimental setup [35, 36].
To address the question whether the plasmids are also sufficient for natural transformation studies of phylogenetically distant bacteria, natural transformation assays were performed with pM1, pM2 and pM3 using ~30 different marine isolates as potential recipients. These studies revealed the ability of different Gammaproteobacteria and Actinobacteria to be naturally transformed with the three DNA constructs, even in the presence of significant differences in 16S rRNA identity (Table 1).
Total CFU and resistant CFU in the gut contents
The total CFU and the number of resistant CFU in samples from the stomach, small intestine, cecum and colon of rats were determined after the 4 days of plasmid exposure. The mean numbers of bacteria in the different compartments are presented in Table 2. The bacterial counts varied between individuals, for instance, in the stomach, the total CFU ranged from < 1 × 103 to 6 × 105, in the small intestine the CFU varied from < 103 to 2 × 107. Total CFU obtained for material sampled from the cecum and colon were lower than expected, ranging form < 1 × 106 (not quantifiable due to dilution error, most of the individuals) to up to 4 × 10 7. Variability in CFU numbers is expected due to different amounts of food material present in the colon of each rat; reflecting differences in activity, eating pattern and digestion.
CFU numbers of resistant colonies determined by plating on selective media containing streptomycin and kanamycin were very low. In most animals we could not detect culturable resistant bacteria; with a detection limit of 1 CFU per gram material. However, from some compartments and individuals a few colonies emerged on the selective media (see Table 2). A total of 441 isolates were collected from plates inoculated with GIT material obtained from exposed rats, and 17 isolates from GIT material obtained from non-plasmid exposed rats. All isolates growing on selective media were picked and frozen in LB media with 15% glycerol for further characterization.
PCR analyses of antibiotic resistant colonies
None of the 441 antibiotic resistant bacteria recovered from the GIT content of DNA exposed rats produced a positive PCR signal. The total amount of bacteria plated on selective media from the six exposed individuals exceeded 1 × 108 bacteria. Thus, less than 1 antibiotic resistance gene transfer event was observed per 1,1 × 108 culturable bacteria. The transformation frequency of the culturable bacterial fraction with the feed added plasmid was therefore less than 8,8 × 10-9 transformants per culturable bacteria.
Natural transformation of microbial communities in the GIT requires the presence of bacteria that express competence in gut locations where DNA is present and accessible. Moreover, those competent bacteria must be able to integrate and vertically transfer the DNA over subsequent generations (i.e. through linkage to a replicating unit). Finally, no significant absolute or relative fitness reductions should be experienced by the transformed cells . The model system presented here was optimized such that the requirements for stable inheritance were met through providing the target DNA with flanking regions sharing homology to bacterial rRNA genes harboured by a subset of the bacteria present in the GIT of rats. Table 3 lists the bacterial species with less than 25% DNA sequence divergence at the 16S or 23S loci.
The in vivo model system used also included a selectable marker system (npt I, aad A) that facilitated unambiguous detection of HGT among the diversity of the culturable bacterial populations. Although the model system chosen attempted to circumvent some natural barriers to HGT (introduced DNA sequence similarity) and model based limitations (strong positive selection of transformants, and high levels of DNA exposure to natural bacterial communities) no transformants were observed among the 6 rats and approx. 108 cultivable, aerobic bacteria examined. Several factors relevant to these findings may explain the lack of observable uptake of plasmid DNA in this study:
Natural transformation does not occur in the GIT of rats due to lack of competence-expressing bacteria or lack of access to extracellular DNA for competent bacteria.
ii) Natural transformation occurs at frequencies below the levels detectable within a limited number of aerobic bacteria per animal GIT and days of DNA exposure. The various parts of the GIT support different inoculum densities . Although total CFU numbers were at reasonable levels in the various gut compartments in rats (103-107 CFU per gram GIT content), these population sizes may be too small to enable identification of rare HGT events in individual, single GITs [4, 38]. The detection limit per rat and GIT compartment is thus 1 transformant per 103 to 107 bacteria, in our study. A detection limit of 1 transformant per 1 × 108 bacteria is reached when the overall number of CFU tested is summarized for the 6 rats. Transformable bacterial species may nevertheless express competence below this frequency range and screening of larger bacterial population sizes should be considered for consistent detection or rare HGT events. The number of species in the GIT of rats that are expected to be competent based on in vitro data is only a fraction of overall population size and diversity. It can be estimated that it would be necessary to sample the combined bacterial populations of the GIT content of 60 to 6 000 rats to identify bacterial transformants produced at lower frequencies of 10-10 to 10-13.
iii) Natural transformation is limited in frequency due to the low concentrations or lack of accessibility of DNA substrates in the different compartments of the digestive tract. Although a continuous supply of DNA was ensured by daily administration of high amounts (100 μg plasmid DNA), purified DNA is known to rapidly fragment in vivo. However previous studies have shown that minor proportions of orally ingested plasmid DNA persist in a biologically active form in the GIT of rats [2, 3], and 1-2% of plasmid and bacteriophage DNA can survive passage through the mouse GIT and be detected in the faeces [13–15]. The latter study detected size ranges from a few hundred bp up to about 1700 bp. Similar observations of short fragment size have also been reported in other animal experiments involving fish, poultry, pig, sheep, cattle and humans [16–19, 32, 39–41]. See Rizzi et al., for a comprehensive review. However, almost all of these studies have biochemically analyzed persistence and degree of fragmentation after recovery of DNA from the gut. So far, little information is available on the extent that DNA fragments present in various gut compartments are physically accessible to bacteria as templates for natural transformation [4, 5]. The study by Nordgård et al.  indicates that gut contents may be inhibitory to natural transformation. The study examined the effects of the GIT content of both normal microbiota rats and germfree mice in in vitro transformation assays of competent cells of A. baylyi. The study showed that the presence of both types of gut contents was inhibitory to transformation . Only purified DNA added to cecum and large intestine content samples from germfree mice was able to transform A. baylyi at low frequencies in vitro. The sharply reduced DNA uptake frequencies also observed in the presence of sterile gut material from mice indicate that microbially produced DNA nucleases were not responsible for the absence of observable transformation. Bacterial nucleases have also in other studies been found to play a minor role in DNA degradation in the GIT [2, 3, 42]. Thus, host nucleases or other macromolecules present in the GIT may inhibit natural transformation.
iv) Natural transformation occurs in the GIT but is not observed in our model systems due to other technical limitations. These limitations can include the possibility that bacterial species present in the gut only express competence as a response to certain host physiological conditions or certain (feed) nutrition sources not adopted in our feeding protocol. Moreover, it cannot be excluded that the plasmid DNA used for our transformation study in vivo was not optimal for DNA uptake among all the relevant competent bacterial species in the gut. The DNA similarity, present on the plasmids used, is highest to the Acinetobacter and Psychobacter genera. However, the in vitro experiments with some marine isolates confirmed the broader applicability of these vectors for natural transformation studies. To our knowledge the results obtained from the in vitro study of the marine isolates represent the first report of natural transformation of members of the genera Photobacterium, Marinobacter and Psychrobacter (Gammaproteobacteria) and the genus Kocuria (Actinobacteria). These observations suggest that the constructed vectors can be used to detect novel transformable bacteria from various environmental samples. As most bacteria have multiple copies of rRNA genes , a lethal effect of an integration of the marker genes into a single rRNA locus is not expected.
Finally, molecular evidence indicates that as little as 10% and up to 40-50% of the GIT population can be identified by differential plating methods . The inability to isolate the major fraction of the microorganisms in the GIT, and importantly the obligate anaerobic proportion, on agar media could also contribute to the failure to detect transformants in our model system. However, it is emphasized that most GIT bacteria that are of high concern in the context of resistance development do belong to the culturable fraction (e.g. the Enterobacteriaceae). The lack of provision of an immediate selective advantage to the transformants may also have given the net result that rare transformants failed to survive and to expand to numbers that were sufficient to allow their detection .
Natural transformation occurring among members of the bacterial community in the lower GIT remains to be demonstrated. Most previous studies have examined natural transformation of single bacterial species introduced into the GIT of various mammalian species. The model system presented here allowed a subset of the aerobic microbial community to be tested for competence development; however, transformants were not detected among the 6 rat GITs and 108 bacteria tested; suggesting a transformation frequency below 8.8 × 10-9 (for the combined rat samples). As discussed here, small rodent models may harbour too few culturable bacteria per individual and gut compartment to allow realistic detection limits of natural transformation events in the lower GIT. The design of future studies must consider the opportunities and limitations inherent in the population sizes of culturable aerobic and anaerobic bacteria in the model organism used, the transformable fraction and level of competence expressed among these, and the DNA exposure rates expected in the various GIT compartments
Three chromosome integration vectors, pM1, pM2 and pM3 were derived from the E. coli vector pBIISK (ori V of pMB1) (Figure 1). pM1 (9500 bp) and pM2 (9338 bp) both carry the 16S rRNA gene (bp 331-1331) from Acinetobacter baylyi ADP1 and the 23S rRNA gene (bp 68-1326) from a marine Psychrobacter isolate. The 16S rRNA gene was amplified with primer pair F16 and F17, and the 23S rRNA gene was amplified with primer pair F13 and F14. The rRNA sequences flank genes encoding the green fluorescent protein (GFP) and streptomycin and kanamycin resistance. The two resistance markers can be used for selection of transformants. The gfp gene enables detection of transformants but was not used throughout this study since transformants were already obtained by selection alone. In pM3 (8281 bp) the resistance genes and gfp gene are flanked by two 16S rRNA gene sequences (bp 362-693 and bp 694-1331) (Figure 1). The plasmids were maintained in E. coli TOP10 cells (Invitrogen) grown on LB agar containing streptomycin (50 μg/ml) and kanamycin (100 μg/ml). Plasmid DNA was isolated using the Qiagen Plasmid Maxi Kit (Qiagen, Germany) following the manufacturer's protocol and quantified by Nanodrop ND-1000 (Nanodrop Technologies) prior to dilution and mixing in feed. The integrity of the plasmids was also confirmed by agarose gel electrophoresis.
The level of similarity between the 16S and the 23S rRNA gene fragments inserted into the pM2 and pM3 plasmids with known DNA sequences was determined. The partial nucleotide sequence (331-1331 bp) of the 16S rRNA gene of A. baylyi DSM14961 type strain (EF611407) and the partial nucleotide sequence (68-1326 bp) of the 23S rRNA gene of Psychrobacter faecalis strain DSMZ 14664 (HM236417) were used for BLASTN retrieval. The search was limited to the type strain bacteria of the most common genera found in rat and in mouse GIT system [45–49]. For the 16S rRNA fragment, the BLASTN search was to 101 different bacterial species, while for the 23S rRNA fragment it included 302 different bacterial species. The maximum target sequences parameter was set to 500.
Cells of the naturally competent A. baylyi strain BD413 were exposed to the plasmids (pM2 and pM3) to confirm that the purified plasmids were able to transform a known naturally competent bacterium at high frequencies. For the in vitro filter transformation assays, 0.2, 2, or 20 μg of purified plasmid DNA (Qiagen Miniprep kit, Germany) was mixed with 100 μl of bacterial cells, and incubated on nitrocellulose filters on LB agar plates for 24 h to allow binding and uptake of DNA and replacement recombination events of the ribosomal gene sequences (Figure 2.), before serial dilutions and selective plating on LB agar as described by Ray and Nielsen . To confirm the broader functionality of the plasmids for natural transformation, available marine isolates of the Gammaproteobacteria, Actinobacteria and Alphaproteobacteria were exposed to pM1, pM2 and pM3. Purified plasmid DNA (1 μg) was linearized with Not I and mixed with 100 μl of overnight culture of a marine isolate and incubated on nitrocellulose filters on marine broth (MB) agar (18.5 g marine broth, 10 g NaCl and 15 g agar per liter, pH 7.6) at room temperature for 24 to 48 h (until growth was visible). After serial dilutions of the cells with saline, selective plating was done on MB agar containing 100 μg/ml streptomycin and 100 μg/ml kanamycin. Transformation was confirmed by selective plating and PCR using primers E9F and K1 for amplification of the 16S rRNA gene and gfp, and primers kmup and 97ar for amplification of npt I and 23S rRNA genes (Table 4 and Figure 2).
Rat feed preparation
Food pellets were made by mixing 80 ml sterile water with 200 g of standard rat feed powder AIN-93 (Scanbur BK AS, Norway) . After solidification overnight at 4°C, the mass was cut into 1.5 × 1.5 × 1.5 cm feed pellets and left on a tray covered with baking paper to dry overnight in a sterile hood at RT. The pellets were kept at -20°C until used. A total of 100 μg plasmid DNA (pM2 and pM3 at equal amounts) was pipetted into the dry feed pellet. Control rats received the same pellets, but without DNA. Previous studies have shown that the feed source is free of DNA and that DNA mixed into the pellet does not degrade or lose its ability to transform bacteria over a 72 h incubation period at RT .
Wistar rats (Mol:WIST Han, M&B Denmark), bred at the rat facility of the Animal Department of the University of Tromsø, Norway were used. The rats (200 ± 20 grams) were referred to two different groups. Group one consisted of two rats: one female and one male and group two: three females and three males. Group one was given the pellet meal only (contained no DNA; negative control group), while group two was given the target meal (feed pellets with added plasmid DNA). The total amount of plasmid DNA ingested per rat in group two was 100 μg per day for four days (50% pM2 and 50% pM3). Plasmid DNA was provided as a purified DNA extract; with no restriction enzyme treatment. The rats were killed in a CO2 chamber 4 days after starting on the target meal, and immediately before sampling of contents from the different GI compartments (stomach, small intestine, cecum and colon). The experiments/housing procedures were approved by The National Animal Research Authority, Norway.
Enumeration of bacterial cells
Before starting the feeding trials, the overall background level of antibiotic resistance to streptomycin and kanamycin were determined for faeces from rats kept at the Animal facility used for the feeding trial. No colonies emerged on the selective plates (data not shown); indicating a low level of background resistance among the aerobic population and suitability of streptomycin or kanamycin in identifying bacterial transformants.
The content from the different sampling sites (stomach, small intestine, cecum and large intestine) were plated on LB agar-plates. Total CFUs (colony forming unit) per gram were determined under aerobic conditions on non-selective plates (100 μl of ten-fold dilutions made in saline), and on plates with streptomycin or kanamycin (both 50 μg/ml) for transformant CFU per gram. Approx. 20 transformant-selective agar plates were used per sample (of approx. 1 g) to ensure sufficient dispersal of gut material per plate and to prevent bacterial growth on insoluble feed material. CFU were determined after incubation at 37°C for 46 h. The transformation frequencies are given as the number of CFU growing on transformant-selective plates divided by the number of CFU on non-selective plates. The detection limit is given as the reciprocal of the number of CFU on non-selective plates. The colonies grown on the selective plates were picked and kept at -20°C in LB medium containing 10% glycerol and kanamycin or streptomycin at 50 μg/ml until further analysis. The total number of bacterial cells tested for potential uptake of added plasmid DNA was calculated as the sum of the numbers of bacteria (CFUs) obtained under aerobic conditions on non-selective plates for the 6 treated rats combined (1,1 × 108 CFUs); as an equal or higher amount of sample was also plated on the selective media for growth of transformed cells. Thus, the detection limit represents the potential for transformation occurring in the GIT of a rat population consisting of 6 individuals.
Bacterial DNA isolation and PCR amplification
DNA was isolated from colonies that emerged on the selective agar plates with the GenElute Bacterial Genomic Kit (Sigma) following the manufacturer's protocol. Isolated DNA was quantified with a Nanodrop ND-1000 (Thermo, USA). The DNA from single, re-streaked colonies was subjected to PCR analysis of the bacterial 16S rRNA gene to confirm the general absence of PCR inhibitors. The reactions were performed in a total volume of 25 μl containing the following: 0,75 μl of each primer at 10 μM, 12,5 μl HotStarTaqMix (Qiagen, Germany), 10 μl water and 1 μl bacterial DNA template (100 ng/μl). Primers and cycling conditions used are listed in Tables 4 and 5. The PCR products were run on 1% agarose gels before visualization.
To determine if the emerging antibiotic resistant bacteria arose from recombination (transformants), primers targeting different regions of the two plasmid constructs were used (Table 4). For each resistant bacterial colony obtained, PCR was done to define if a fragment of the plasmid (pM2 or pM3) was inserted into the genome. The reactions were performed in a total volume of 50 μl containing the following: 1 μl of each primer at 10 μM, 25 μl DYNAzyme™ II PCR Master Mix (Finnzymes Oy, Finland), 18 μl ddH2O and 2 μl DNA template (100 ng/μl). Negative PCR setup controls (no DNA template), negative rat controls (received no DNA in feed) and positive controls (plasmid dilution series) were included in each PCR set-up. Primers and cycling conditions are listed in Tables 4 and 5.
Netherwood T, Martin-Orue SM, O'Donnell AG, Gockling S, Graham J, Mathers JC, Gilbert HJ: Assessing the survival of transgenic plant DNA in the human gastrointestinal tract. Nat Biotechnol. 2004, 22: 204-209. 10.1038/nbt934.
Wilcks A, van Hoek AHAM, Joosten RG, Jacobsen BBL, Aarts HJM: 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. 10.1016/j.fct.2003.10.013.
Wilcks A, Jacobsen BB: Lack of detectable DNA uptake by transformation of selected recipients in mono-associated rats. BMC Res Notes. 2010, 3: 49-10.1186/1756-0500-3-49.
Nordgård L, Nguyen T, Midtvedt T, Benno Y, Traavik T, Nielsen KM: Lack of detectable uptake of DNA by bacterial gut isolates grown in vitro and by Acinetobacter baylyi colonizing rodents in situ. Environ Biosafety Res. 2007, 6: 149-160. 10.1051/ebr:2007029.
Rizzi A, Raddadi N, Sorlini C, Nordgård L, Nielsen KM, Daffonchio D: The stability and degradation of dietary DNA in the gastrointestinal tract of mammals - implications for horizontal gene transfer and the biosafety of GMOs. Crit Rev Food Sci Nutr. 2012, 52: 142-161. 10.1080/10408398.2010.499480.
Nielsen KM, Ray JL, Johnsen PJ: The natural uptake of extracellular DNA in bacteria. Encyclopedia of Microbiology. Edited by: Schaechter M. 2009, Oxford: Elsevier, 587-596. 3
Brigulla M, Wackernagel W: Molecular aspects of gene transfer and foreign DNA acquisition in prokaryotes with regard to safety issues. Appl Microbiol Biotechnol. 2010, 86: 1027-1041. 10.1007/s00253-010-2489-3.
Lorenz MG, Wackernagel W: Bacterial gene transfer by genetic transformation in the environment. Microbiol Rev. 1994, 58: 563-602.
De Vries J, Wackernagel W: Microbial horizontal gene transfer and the DNA release from transgenic crop plants. Plant Soil. 2004, 266: 91-104.
Van Reenen CA, Dicks LMT: Horizontal gene transfer amongst probiotic lactic acid bacteria and other intestinal microbiota: what are the possibilities? A review. Arch Microbiol. 2011, 193: 157-168. 10.1007/s00203-010-0668-3.
McCracken VJ, Lorenz RG: The gastrointestinal ecosystem: a precarious alliance among epithelium, immunity and microbiota. Cell Microbiol. 2001, 3: 1-11. 10.1046/j.1462-5822.2001.00090.x.
O'Hara AM, Shanahan F: The gut flora as a forgotten organ. EMBO Rep. 2006, 7: 688-693. 10.1038/sj.embor.7400731.
Schubbert R, Lettmann C, Doerfler W: Ingested foreign (phage M13) DNA survives transiently in the gastrointestinal tract and enters the bloodstream of mice. Mol Gen Genet. 1994, 242: 495-504. 10.1007/BF00285273.
Schubbert R, Renz D, Schmitz B, Doerfler W: Foreign (M13) DNA ingested by mice reaches peripheral leukocytes, spleen, and liver via the intestinal wall mucosa and can be covalently linked to mouse DNA. Proc Nat Acad Sci USA. 1997, 94: 961-966. 10.1073/pnas.94.3.961.
Schubbert R, Hohlweg U, Renz D, Doerfler W: On the fate of orally ingested feed-derived DNA in mice: chromosomal association and placental transmission to the fetus. Mol Gen Genet. 1998, 259: 569-576. 10.1007/s004380050850.
Duggan PS, Chambers PA, Heritage J, Forbes MJ: Fate of genetically modified maize DNA in the oral cavity and rumen of sheep. Br J Nutr. 2003, 89: 159-166. 10.1079/BJN2002764.
Reuter T, Aulrich K: Investigations on genetically modified maize (Bt-maize) in pig nutrition: fate of feed-ingested feed-derived DNA in pig bodies. Eur Food Res Technol. 2003, 216: 185-192.
Einspanier R, Klotz A, Kraft J, Aulrich K, Poser R, Schwägele F, Jahreis G, Flachowsky G: The fate of forage plant DNA in farm animals: a collaborative case-study investigating cattle and chicken fed recombinant plant material. Eur Food Res Technol. 2001, 212: 129-132. 10.1007/s002170000248.
Einspanier R, Lutz B, Rief S, Berezina O, Zverlov V, Schwarz W, Mayer J: Tracing residual recombinant feed molecules during digestion and rumen bacterial diversity in cattle fed transgene maize. Eur Food Res Technol. 2004, 3: 269-273.
Mazza R, Soave M, Morlacchini M, Piva G, Morocco A: Assessing the transfer of genetically modified DNA from feed to animal tissues. Transgenic Res. 2005, 14: 775-784. 10.1007/s11248-005-0009-5.
Bensasson D, Boore JL, Nielsen KM: Genes without frontiers. Heredity. 2004, 92: 483-489. 10.1038/sj.hdy.6800451.
Thomas CM, Nielsen KM: Mechanisms of, and barriers to, horizontal gene transfer between bacteria. Nat Rev Microbiol. 2005, 3: 711-721. 10.1038/nrmicro1234.
Ray JL, Harms K, Wikmark OG, Starikova I, Johnsen PJ, Nielsen KM: Sexual isolation in Acinetobacter baylyi is locus-specific and varies 10,000-fold over the genome. Genetics. 2009, 182: 1165-1181. 10.1534/genetics.109.103127.
Shen P, Huang HV: Homologous recombination in Escherichia coli: dependence on substrate length and homology. Genetics. 1986, 112: 441-457.
Zawadzki P, Cohan FM: The size and continuity of DNA segments integrated in Bacillus transformation. Genetics. 1995, 141: 1231-1243.
Matic I, Taddei F, Radman M: Genetic barriers among bacteria. Trends Microbiol. 1996, 4: 69-72. 10.1016/0966-842X(96)81514-9.
Majewski J, Cohan FM: DNA sequence similarity requirements for interspecific recombination in Bacillus. Genetics. 1999, 153: 1525-1533.
Majewski J: Sexual isolation in bacteria. FEMS Microbiol Lett. 2001, 199: 161-169. 10.1111/j.1574-6968.2001.tb10668.x.
Strätz M, Mau M, Timmis KN: System to study horizontal gene exchange among microorganisms without cultivation of recipients. Mol Microbiol. 1996, 22: 207-215. 10.1046/j.1365-2958.1996.00099.x.
Mercer DK, Scott KP, Bruce-Johnson WA, Glover LA, Flint HJ: Fate of Free DNA and transformation of the oral bacterium Streptococcus gordonii DL1 by plasmid DNA in human saliva. Appl Environ Microbiol. 1999, 65: 6-10.
Mercer DK, Scott KP, Melville CM, Glover AL, Flint HJ: Transformation of an oral bacterium via chromosomal integration of free DNA in the presence of human saliva. FEMS Microbiol Lett. 2001, 200: 163-167. 10.1111/j.1574-6968.2001.tb10709.x.
Duggan PS, Chambers PA, Heritage J, Forbes MJ: Survival of free DNA encoding antibiotic resistance from transgenic maize and the transformation activity of DNA in ovine saliva, ovine rumen fluid and silage effluent. FEMS Microbiol Lett. 2000, 191: 71-77. 10.1111/j.1574-6968.2000.tb09321.x.
Kharazmi M, Bauer T, Hammes WP, Hertel C: Effect of food processing on the fate of DNA with regard to degradation and transformation capability in Bacillus subtilis. Syst Appl Microbiol. 2003, 26: 495-501. 10.1078/072320203770865774.
Shedova E, Albrecht C, Zverlov VV, Schwarz WH: Stimulation of bacterial DNA transformation by cattle saliva: implications for using genetically modified plants in animal feed. World J Microbiol Biotechnol. 2009, 25: 457-463. 10.1007/s11274-008-9910-4.
Nielsen KM, Gebhard F, Smalla K, Bones AM, van Elsas JD: Evaluation of possible horizontal gene transfer from transgenic plants to the soil bacterium Acinetobacter calcoaceticus BD413. Theor Appl Genet. 1997, 95: 815-821. 10.1007/s001220050630.
Nielsen KM, van Weerelt MD, Berg TN, Bones AM, Hagler AN, van Elsas JD: Natural transformation and availability of transforming DNA to Acinetobacter calcoaceticus in soil microcosms. Appl Environ Microbiol. 1997, 63: 1945-1952.
Pettersen AK, Bøhn T, Primicerio R, Shorten PR, Soboleva TK, Nielsen KM: Modeling suggests frequency estimates are not informative for predicting the long-term effect of horizontal gene transfer in bacteria. Environ Biosafety Res. 2005, 4: 223-233. 10.1051/ebr:2006008.
Nielsen KM, Townsend JP: Monitoring and modeling horizontal gene transfer. Nat Biotechnol. 2004, 22: 1110-1104. 10.1038/nbt1006.
Martin-Orue SM, O'Donnell AG, Arino J, Netherwood T, Gilbert HJ, Mathers JC: Degradation of transgenic DNA from genetically modified soya and maize in human intestinal simulations. Br J Nutr. 2002, 87: 533-542. 10.1079/BJN2002573.
Sanden M, Bruce IJ, Rahman AM, Hemre GI: The fate of transgenic sequences present in genetically modified plant products in fish feed, investigating the survival of GM soybean DNA fragments during feeding trials in Atlantic salmon, Salmo salar L. Aquaculture. 2004, 237: 391-405. 10.1016/j.aquaculture.2004.04.004.
Nielsen KM, Berdal KG, Kruse H, Sundsfjord A, Mikalsen A, Yazdankhah S, Nes I: An assessment of potential long-term health effects caused by antibiotic resistance marker genes in genetically modified organisms based on antibiotic usage and resistance patterns in Norway. VKM-Report, Norwegian Scientific Committee for Food Safety, Oslo, Norway.
Maturin L, Curtiss R: Degradation of DNA by nucleases in intestinal tract of rats. Science. 1977, 196: 216-218. 10.1126/science.322286.
Boyer SL, Flechtner VR, Johanson JR: Is the 16S-23S rRNA internal transcribed spacer region a good tool for use in molecular systematics and population genetics? A case study in cyanobacteria. Mol Biol Evol. 2001, 18: 1057-1069. 10.1093/oxfordjournals.molbev.a003877.
McCartney AL: Application of molecular biological methods for studying probiotics and the gut flora. Br J Nutr. 2002, 88 (Suppl 1): S29-S37.
Brooks SPJ, McAllister M, Sandoz M, Kalmokoff ML: Culture-independent phylogenetic analysis of the faecal flora of the rat. Can J Microbiol. 2003, 49: 589-601. 10.1139/w03-075.
Inoue R, Ushida K: Development of the intestinal microbiota in rats and its possible interactions with the evolution of the luminal IgA in the intestine. FEMS Microbiol Ecol. 2003, 45: 147-153. 10.1016/S0168-6496(03)00134-X.
Hanske L, Hussong R, Frank N, Gerhäuser C, Blaut M, Braune A: Xanthohumol does not affect the composition of rat intestinal microbiota. Mol Nutr Food Res. 2005, 49: 868-873. 10.1002/mnfr.200500048.
Dalby AB, Frank DN, St Amand AL, Bendele AM, Pace NR: Culture-independent analysis of indomethacin-induced alterations in the rat gastrointestinal microbiota. Appl Environ Microbiol. 2006, 72: 6707-6715. 10.1128/AEM.00378-06.
Licht TR, Hansen M, Poulsen M, Dragsted LO: Dietary carbohydrate source influences molecular fingerprints of the rat faecal microbiota. BMC Microbiol. 2006, 6: 98-10.1186/1471-2180-6-98.
Ray JL, Nielsen KM: Experimental methods for assaying natural transformation and inferring horizontal gene transfer. Methods in Enzymology; methods in molecular evolution: producing the biochemical data. Edited by: Zimmer EA, White TJ, Cann RL, Wilson AC. 2005, San Diego: Academic, 224: 491-520.
Cardinale M, Brusetti L, Quatrini P, Borin S, Puglia AM, Rizzi A, Zanardini E, Sorlini C, Corselli C, Daffonchio D: Comparison of different primer sets for use in automated ribosomal intergenic spacer analysis of complex bacterial communities. Appl Environ Microbiol. 2004, 70: 6147-6156. 10.1128/AEM.70.10.6147-6156.2004.
Lane DJ: 16S/23S rRNA sequencing. Nucleic acid techniques in bacterial systematics. Modern microbiological methods. Edited by: Stackebrandt E, Goodfellow M. 1991, Chichester: J Wiley & Sons, 133-
Muyzer G: Structure, function and dynamics of microbial communities: the molecular biology approach. Advances in molecular ecology. Edited by: Carvalho GR. 1998, Amsterdam: Ios Press, 87-117.
Munson MA, Baumann P, Clark MA, Baumann L, Moran NA, Voegtlin DJ, Campbell BC: Aphid-eubacterial endosymbiosis: evidence for its establishment in an ancestor of four aphid families. J Bacteriol. 1991, 173: 6321-6324.
Reeves PG: Components of the AIN-93 diets as improvements in the AIN-76A diet. J Nutr. 1997, 127: 838S-841S.
Grønsberg IM, Nordgård L, Fenton K, Hegge B, Nielsen KM, Bardocz S, Pusztai A, Traavik T: The uptake and organ distribution of feed introduced plasmid DNA in growing or pregnant rats. Food Nutr Sci. 2011, 2: 377-386. 10.4236/fns.2011.24053.
We thank the Animal Department, Faculty of Health Sciences, University of Tromsø, for technical assistance and support. We thank Julia Eggert for assistance during the animal experiment and Iris Graf for assistance with the natural transformation experiments of marine isolates. We thank Lise Øverås (University of Bergen, Norway), Ian Joint (Plymouth Marine Laboratory, United Kingdom) and Francisco Rodriguez-Valera (Universidad Miguel Hernandez de Elche, Spain) for providing the marine samples. The work was founded by the Norwegian Research Council, GenØk - Centre for Biosafety and by the European Commission (MIRACLE project).
The authors declare that they have no competing interests.
LN participated in the design, coordination, analysis, and drafting of the manuscript. LB and NR participated the molecular and microbiological experiments and analyses, TT participated in the design and drafting of the manuscript. BA designed the plasmids, contributed the data on the marine isolates, and revised the manuscript. KMN contributed to the design, coordination, analysis, and drafting of the manuscript. All authors read and approved the final manuscript.