ABSTRACT
CTnDOT is a 65-kb conjugative transposon present in Bacteroides spp. that confers resistance to erythromycin [erm(F)] and tetracycline [tet(Q)]. An interesting feature of CTnDOT is that both excision from the chromosome and transfer of CTnDOT are stimulated by exposure to tetracycline. However, when no tetracycline is present, transfer of CTnDOT is not detectable. Previous studies suggested that a region containing a small RNA, RteR, appeared to mediate repression of CTnDOT transfer; however, virtually nothing was known about RteR. We have demonstrated that RteR is a 90-nucleotide transcript that is not further processed. RteR inhibits conjugative transfer of CTnDOT by targeting the transfer region, a 13-kb operon that encodes the tra genes required to assemble the mating apparatus. We report here that RteR interacts with the region downstream of traA. Levels of the downstream tra mRNA are dramatically reduced when RteR is present. Further, RteR does not appear to decrease the half-life of the tra mRNA transcript, suggesting that RteR does not bind to the transcript to initiate RNase-dependent decay, similar to other trans-acting small RNAs. We predict that RteR may act to enhance termination of the tra operon within traB, which could account for the decreased abundance of the tra transcript downstream of traA and explain why the tra mRNA has the same half-life whether or not RteR is present. RteR is the only small RNA that has been characterized so far within the Bacteroidetes phylum.
INTRODUCTION
Bacteroides are Gram-negative obligate anaerobes that are a major component of the human intestinal microbiota (5, 11). Bacteroides can act as reservoirs of antibiotic resistance genes within the gastrointestinal tract, by transferring these resistance genes to other organisms via mobile genetic elements, such as conjugative transposons (14, 18, 20, 27). One such example is CTnDOT, a 65-kb conjugative transposon that confers resistance to the antibiotics erythromycin and tetracycline.
Exposure to low levels of tetracycline stimulates both the excision of CTnDOT from the chromosome and conjugative transfer of CTnDOT (14, 19, 21). Upon tetracycline induction, translation of the tetQ-rteA-rteB operon is allowed to continue. tetQ encodes a ribosomal protection type of tetracycline resistance; rteA and rteB encode a two-component signal transduction system, with RteA recognizing an external signal and RteB being the response modulator (21). It is not known what signal RteA senses, although it is not tetracycline (13). RteB then activates transcription of RteC, which then activates transcription of the excision operon, containing xis2c-xis2d-exc. The proteins encoded by this operon are not only important for excision and integration of CTnDOT, but they also serve a regulatory role by activating transcription of the transfer (tra) operon (7).
In the absence of tetracycline, however, there is no detectable transfer of CTnDOT (26). Previous studies demonstrated that a 500-bp region containing a small RNA (sRNA), RteR, is sufficient for mediating repression of the self-transmissible plasmid pLYL72 (9) when there is no tetracycline present (7, 26). Western blotting and subcloning analysis demonstrated that the target of RteR was within the 13-kb tra operon (26).
In this study, we have identified the 5′ and 3′ ends of RteR by using rapid amplification of cDNA ends (RACE) analysis, and we demonstrate that RteR is a 90-nucleotide (nt) noncoding RNA that is not further processed into a smaller transcript. We have also identified the promoter of rteR and localized rteR to the region immediately downstream of the excision gene exc. Interestingly, rteR is transcribed constitutively in response to tetracycline exposure, unlike the other CTnDOT regulatory genes, which have proven to be expressed only after tetracycline induction.
We report here that we have narrowed the target of RteR to an approximately 1-kb region between traA and traC and demonstrate that downstream of traA there is little mRNA detectable in the presence of RteR. We also show that the half-life of the tra mRNA is not affected by the presence of RteR. These observations suggest that RteR may act to initiate premature transcription termination early in the tra operon, thus resulting in inhibition of CTnDOT conjugative transfer.
MATERIALS AND METHODS
Strains, plasmids, and growth conditions.All bacterial strains and plasmids used in this study are summarized in Table 1. Unless otherwise noted, all Bacteroides strains were initially cultured in chopped meat broth (Remel) and then transferred to anaerobic Trypticase-yeast extract-glucose (TYG) medium (16) for overnight growth with antibiotics when appropriate. Subculturing of Bacteroides strains was also done in TYG liquid medium, and all Bacteroides culturing was performed anaerobically at 37°C. Escherichia coli was grown aerobically in Luria-Bertani (LB) broth at 37°C, with antibiotics when appropriate. Unless otherwise noted, antibiotic concentrations used for culturing were as follows: ampicillin at 100 μg/ml; cefoxitin at 10 μg/ml; chloramphenicol at 10 μg/ml; erythromycin at 10 μg/ml; gentamicin at 200 μg/ml; kanamycin at 100 μg/ml; rifampin at 2 μg/ml; streptomycin at 100 μg/ml; tetracycline at 1 μg/ml.
Bacterial strains and plasmids
Bacterial mating assays.E. coli strain HB101 containing the IncPα plasmid RPI was used to mobilize plasmids from an E. coli donor strain to Bacteroides recipients, as described previously (16).
For assays measuring the transfer frequency of pLYL72 from a Bacteroides donor to the recipient E. coli strain HB101, the procedure was performed as described previously (9). The transfer frequency is expressed as the ratio of transconjugants per recipient.
Site-directed mutagenesis.Site-directed mutagenesis was performed using the Stratagene QuikChange XL II kit (Agilent Genomics). To create the mutation in the putative RteR promoter, pJW305 plasmid DNA was used as a template, with primers SDM 5740 top and SDM 5740 bottom. To create the substitutions within the sequence of rteR that changed nucleotides 49 to 52 from GGAU to ACCA, plasmid pJW305 was used with primers SDM RteR 49–52 top and bottom. All primers used in this study are summarized in Table 2. The PCR-generated products were then treated with DpnI at 37°C for 1 h and subsequently transformed into chemically competent E. coli DH5α. The resulting plasmid was mobilized into Bacteroides strain BT4001containing pLYL72 via a triparental mating as described above.
Primers used in this study
PCR amplification and sequencing of RteR from other CTnDOT-like elements.A 500-bp region containing the 3′ end of exc and rteR from various strains containing CTnDOT-like conjugative transposons was amplified using primers JLW10F and JLW11R (Table 2). These PCR products were then cloned into a commercial cloning vector per the manufacturer's protocol (Promega) and submitted for sequencing at the University of Illinois Core Sequencing Facility.
In addition, BLAST searches were performed to determine whether other exc-rteR sequences could be found in the nucleotide databases. All sequences were then aligned using ClustalW2 analysis (6, 8) to compare sequence similarities among RteR homologues. Secondary structure predictions of RteR homologues were also performed, using MFOLD (29).
RNA isolation.Bacteroides strains were grown in anaerobic TYG medium to an optical density at 650 nm (OD650) of 0.4 to 0.6. RNA was then isolated using the hot phenol method as described previously (1). RNA concentrations were quantified using a nanodrop spectrophotometer.
RACE analysis.5′ rapid amplification of cDNA ends (RACE) analysis of RteR was performed as previously outlined (23) using the rteR-specific primer JLW11R. RNA was isolated from strain BT4001 ΩQAB pJW305. Cells were grown anaerobically in TYG to an OD650 of 0.4 to 0.6.
3′ RACE analysis of RteR was performed as previously outlined (2), using the rteR-specific primer JLW12F. RNA was isolated from BT4001 ΩQAB pJW305 and grown anaerobically in TYG to an OD650 of 0.4 to 0.6.
Northern analysis.Approximately 2 μg of total RNA was electrophoresed on an 8% acrylamide gel (Sequagel, ureagel-8) at 100 V for 1 h. A radiolabeled Decade marker (Ambion) was used as a size standard. RNA was electrotransferred to a Nytran+ membrane (Whatman) at 200 mA for 1 h, followed by UV cross-linking at 1,200 μJ × 1,000. Prehybridization was performed at 42°C for at least 2 h in Ultrahyb solution (Ambion). A 200-bp RteR-specific probe was synthesized using the Maxiscript T7 kit (Ambion), with the primers JLW12F and T7JLW22R. Following prehybridization, the membrane was incubated with the probe overnight at 42°C. The membrane was then washed for 30 min at 42°C in 2× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate) and 0.1% SDS, followed by another 30-min wash at 42°C in 0.1× SSC and 0.1% SDS.
Qualitative RT-PCR.Bacteroides strains were grown in anaerobic TYG medium to an OD650 of 0.4 to 0.6, and RNA was isolated as described above. RNA samples were then treated with Ambion Turbo DNase and quantified using a nanodrop spectrophotometer. Reverse transcription-PCR (RT-PCR) was performed using the Access RT-PCR system (Promega). Reaction mixtures with reverse transcriptase omitted were tested to ensure that all genomic DNA was adequately removed. Samples were electrophoresed on a 1.5% agarose gel for analysis.
RT-qPCR.Bacteroides strains were grown in anaerobic TYG medium to an OD650 of 0.4 to 0.6, and RNA was isolated as described above. RNA samples were then treated with Ambion Turbo DNase, and RT-PCR was subsequently performed to ensure complete removal of genomic DNA. cDNA was synthesized using SuperScript III reverse transcriptase as outlined in the manufacturer's protocol (Invitrogen). Quantitative PCR (qPCR) was performed on a Realplex2 Mastercycler (Eppendorf), using Ssofast EvaGreen as a signal reporter (Bio-Rad). Unless otherwise noted, rpoD was used as a reference marker. The final concentrations of reaction mixture components were as follows: 1 μM primers, 5 μl cDNA, and 10 μl Ssofast EvaGreen supermix (Bio-Rad). The reaction conditions were as follows: initial denaturation for 2 min at 98.0°C; amplification and quantification at 98.0°C for 5 s and 55°C for 10 s for 40 cycles. A melting curve was determined at 95.0°C for 15 s, followed by 55.0°C for 15 s with subsequent heating to 95.0°C over the course of 20 min, with continuous fluorescence measurement, followed by a final incubation at 95.0°C for 15 s. Each measurement was performed in triplicate, and relative quantification (RQ) was performed using the equation RQ = 1/ECT, with E representing amplification efficiency and CT the threshold cycle (12).
tra mRNA half-life assays.To further narrow the target of RteR within the tra region, quantitative RT-PCR was implemented to determine where changes in transcript levels were occurring. Bacteroides strain BT4100 with pLYL72 and pAFD1 (empty vector) or pJW305 (with RteR) was grown to an OD650 of ∼0.5. Aliquots of cultures were then taken 0, 5, and 10 min post-rifampin (400 μg/ml) addition. RNA was isolated as described previously and subsequently treated with Ambion Turbo DNase. cDNA was then synthesized using SuperScript III reverse transcriptase (Invitrogen) as described in the manufacturer's guidelines. RT-qPCR was then performed to detect the relative abundance of transcript over time, as described above. Quantification was performed by normalizing the amount of transcript remaining post-rifampin induction, relative to time zero, which was set at 100%.
In order to confirm that the concentration of rifampin used was sufficient to inhibit RNA polymerase, a set of cultures with no added rifampin was also prepared at the same time points. Qualitative RT-PCR was performed, and a similar amount of transcript was detected after 30 min for the cells without rifampin treatment; however, very little transcript was detected in cells treated with rifampin. This observation suggested that the quantity of rifampin used was sufficient for RNA polymerase inhibition.
RESULTS
RteR inhibits conjugative transfer of the self-transmissible plasmid pLYL72.The transfer (tra) operon, mobilization (mob) operon, and OriT of CTnDOT were previously cloned into a vector, resulting in the 35-kb plasmid pLYL72, which transferred constitutively at a rate of approximately 10−5 to 10−6 transconjugants per recipient (9). However, when CTnDOT regulatory genes were provided in trans, transfer of pLYL72 was then regulated upon tetracycline induction (26). Previous studies demonstrated that a 500-bp region containing rteR was sufficient for reducing pLYL72 transfer frequency to levels below 10−8 transconjugants per recipient. To further minimize the region containing rteR, various subclones were produced to generate 5′ and/or 3′ truncations of this 500-bp region. The resulting subclones were placed in trans to pLYL72 to determine whether the truncated region was still sufficient for inhibition of pLYL72 conjugative transfer. By using these subclones, we were able to minimize the region containing rteR to a 200-bp fragment (Fig. 1).
Localization of the region containing rteR that is required for inhibition of pLYL72 conjugative transfer. A putative transcription start site (TSS) was identified in a previous study, yet that identification was not consistent with other subcloning results (7). We constructed more subclones to further minimize the region containing rteR that is sufficient for pLYL72 transfer inhibition. The transfer frequency of pLYL72, expressed as transconjugants per recipient, is shown to the left of each plasmid construct. Plasmids labeled R+ were able to inhibit conjugative transfer of pLYL72, suggesting that an intact copy of RteR was present. Plasmids labeled R- were no longer able to mediate regulation of pLYL72 transfer.
Further, these results suggested that RteR does not control the excision operon containing xis2c, xis2d, and exc, because RteR can still inhibit conjugative transfer of pLYL72 without the excision operon present. After localizing the region containing RteR that is required for pLYL72 transfer regulation, we confirmed that no open reading frames are present, thus suggesting that RteR is a noncoding RNA.
The rteR promoter flanks the 3′ end of exc and produces a 90-nt RteR transcript that is constitutively transcribed.Previous results suggested that rteR is transcribed constitutively (7), unlike many other CTnDOT regulatory genes that are transcribed and/or translated only upon tetracycline induction (26). However, that preliminary study localized rteR by using primer extension to a region that lay within the 3′ end of exc (Fig. 1), which we report here is not correct. The results of subcloning experiments conflicted with their proposed transcription start site of rteR. It was possible then that RteR was not the transcript detected in their Northern analysis. To confirm that RteR is indeed the correct transcript and is constitutively transcribed, we repeated the Northern analysis using a probe specific to the recently localized 200-bp fragment containing RteR that was sufficient for inhibition of pLYL72 conjugative transfer. Northern blotting was performed using RNA harvested from Bacteroides thetaiotaomicron strain BT4104 grown either in the absence or presence of tetracycline (1 μg/ml). As shown in Fig. 2, the relative abundance of RteR was the same regardless of whether or not tetracycline was present in the growth medium. This result confirmed that rteR is constitutively transcribed with respect to tetracycline exposure.
Northern analysis of RteR. For Northern analysis of the RteR transcript, a radiolabeled probe was created that was specific to the 200-bp fragment demonstrated to inhibit conjugative transfer of pLYL72. Approximately 2 μg of RNA was loaded per well. Size standards used (M) are shown in the center, with sizes of fragments shown. No RteR was detectable in the empty vector control, BT4001 pLYL72 pAFD1, or in a strain containing mutations that abolish transcription from the rteR promoter (BT4001 pLYL72 pJW310). Similar levels of RteR were detected from BT4104 pJW305 whether or not tetracycline (Tc) was present, suggesting that rteR is constitutively transcribed.
The fact that rteR is constitutively expressed suggests that rteR is not part of the operon containing xis2c-xis2d-orf3-exc, which is only transcribed upon tetracycline stimulation. Instead, rteR appears to be transcribed from an independent promoter. Sequence analysis revealed what appeared to be a consensus Bacteroides promoter (3, 10) that flanked the 3′ end of exc, which was in the vicinity where we expected to find the rteR promoter (Fig. 3). To confirm that this was in fact the promoter directing transcription of rteR, site-directed mutations were made in the putative −7 portion that changed the TTTG of the conserved TnTAnTTTG to an AAAC. Mutations such as this have been previously determined to abolish activity of other Bacteroides promoters (3, 10). Northern analysis was performed to see whether the RteR transcript was detectable in this mutant, BT4001 pLYL72 pJW310. RteR was not detected (Fig. 2) even after prolonged exposure (data not shown), suggesting that this mutation was sufficient to completely abolish activity from what we have now identified as the rteR promoter.
The rteR promoter flanks the 3′ end of exc. A putative promoter was identified near the 3′ end of exc. The −7 and −33 regions of the rteR promoter are shown in uppercase black text, with the relevant regions of the promoter noted above the sequence. The stop codon of exc is underlined. In the illustration on top, gray arrows represent the genes that are part of the excision operon, whereas rteR, which is a negative regulator, is shown as a dotted arrow.
To further demonstrate that this mutation abolished activity of the rteR promoter, we also placed this mutation in trans with the self-transmissible plasmid pLYL72, to determine whether this mutant was still capable of repressing pLYL72 conjugative transfer. The transfer frequency of pLYL72 with pJW310 in trans was similar to that of the constitutive levels of transfer, further demonstrating that an intact copy of RteR was no longer present to mediate inhibition of conjugative transfer. Together, these results confirm that we have identified the rteR promoter.
Both 5′ and 3′ RACE analyses were performed to definitively identify the extent of the RteR sequence. 5′ RACE was performed not only to map the 5′ end of RteR but also to identify whether RteR is the primary transcription product or if RteR undergoes internal 5′ processing. The primary transcription start site was mapped to the G nucleotide located 7 bp downstream of the −7 element of the rteR promoter, which is a common transcription start site relative to other characterized Bacteroides promoters (3). Further, 5′ RACE also indicated that RteR does not undergo processing at the 5′ end of the transcript, but is rather the entire transcription product. The 3′ end of RteR, as shown in Fig. 4, is 90 nt downstream of the transcription start site. The secondary structure of RteR, predicted using MFOLD (29), is shown below in Fig. 6A.
RACE analysis to identify the 5′ and 3′ ends of RteR. RACE analysis was performed to confirm the ends of the RteR transcript. The 5′ end of RteR was mapped to the G nucleotide, as indicated by the larger arrow, since this end was identified in 10 out of 15 sequences. Other possible transcription start sites were detectable, as indicated by the shorter arrows at the 5′ end. The 3′ end is indicated by a large black arrow and was detected in 8 out of 12 samples. The other 3′ ends that were detected are indicated by short black arrows. The nucleotides that we have identified as part of the RteR sequence are shown in black text. The rteR promoter identified in this study is boxed, with labels designating the −7 and −33 regions.
The rteR region is present on other CTnDOT-like elements, and both the primary sequence and predicted secondary structure are well conserved.We wanted to determine whether RteR is an sRNA regulator that is specific to CTnDOT, or whether RteR is present on other Bacteroides conjugative transposons (CTns) found in Bacteroides. Not only were we able to find rteR in other Bacteroides spp., but we also observed that rteR was present on other conjugative transposons and was also present in other members of the Bacteroidetes phylum. What was even more surprising is that rteR is very well conserved, with most homologues found to have 98% or greater identity, and many were 100% identical. We were able to find, however, a few sequences that were only about 73% identical to rteR present on CTnDOT. An alignment of these sequences was performed using ClustalW2 analysis (6, 8), and this is shown in Fig. 5. An interesting observation was that despite the substitutions in these rteR sequences, the predicted secondary structure remained similar. Any of the nucleotide substitutions within RteR were either in a single-stranded region of RteR or resulted in a substitution where the substitution could still base-pair with the adjacent nucleotide in the predicted secondary structure (data not shown).
A database search revealed that rteR homologues are well conserved. ClustalW2 analysis (6, 8) was performed to align sequences of rteR homologues identified using basic local alignment search tool (BLAST) analysis. This analysis demonstrated that rteR is present among other members of the Bacteroidetes phylum. The sequence of rteR is shown in the top row for comparison. If the sequence is a match to rteR, the nucleotide is labeled as a dot, whereas any substitutions are shown with the corresponding change in nucleotide. Many homologues had 100% sequence identity. Of the homologues that did have substitutions in the rteR sequence, most had approximately 97% or greater sequence identity, but some sequences had as little as 73% identity. All of these homologues have nearly identical secondary structures, which were predicted using MFOLD analysis (29).
The secondary structure of RteR appears to be important for regulatory function.As previously mentioned, rteR is very well conserved, and these homologues all share nearly identical predicted secondary structures. This observation led us to suggest that perhaps it is the secondary structure of RteR that is important for RteR-mediated negative regulation of conjugative transfer. In the first mutant we constructed, we made site-directed mutations in 4 nucleotides (pJW312) (Fig. 6B) that changed the predicted secondary structure of RteR in a way that would form a stem-loop in what is predicted to be a single-stranded bulge (Fig. 6A). When pJW312 was placed in trans to pLYL72, conjugative transfer was no longer inhibited, demonstrating that this mutation was sufficient to abolish the regulatory function of RteR.
The secondary structure of RteR appears to be important for RteR-mediated repression of conjugative transfer. (A) The secondary structures of RteR clones were predicted using MFOLD analysis (29). Beneath each structure of RteR is the transfer frequency of pLYL72 when provided in trans, expressed as the number of transconjugants per recipient. Clones that were able to repress transfer of pLYL72 are labeled as R+, and the others were labeled as R- if the clone could no longer regulate pLYL72 transfer. (B) The changes in the primary nucleotide sequence of each of these mutants are shown, relative to wild-type RteR.
We also created a deletion mutant that contained an internal deletion within rteR, yet also contained the excision operon containing xis2C, xis2D, and exc, to ascertain if there is an indirect interaction between RteR and the excision proteins. Based on our results demonstrating that a 4-nt substitution could abolish RteR-mediated repression, we expected that a 30-nt deletion (pJW318) (Fig. 6A and B) within rteR would also be sufficient to abolish activity. However, this mutant, which deletes approximately 33% of RteR, was still able to inhibit conjugative transfer of pLYL72. The observation that pJW318 is still active may actually reveal more about the previous mutant, pJW312. The deletion in pJW318 actually removed the nucleotides that had site-directed mutations in pJW312. This suggests that perhaps the mutations in pJW312 did not change critical nucleotides needed for base pairing, because these nucleotides are not even present in pJW318. It suggests then that perhaps the site-directed mutations in pJW312 altered the secondary structure in a way that abolished the ability of RteR to interact with the target (Fig. 6A).
The level of tra mRNA transcript downstream of traA is dramatically decreased when RteR is present.If RteR were preventing transcription entirely, as suggested in a previous study (7), no tra mRNA should be detected at other points of the tra transcript. We tested this by using qualitative RT-PCR (data not shown) and demonstrated that there is in fact detectable transcript throughout the length of the tra operon. In the case of traA, a band of similar intensity was seen whether RteR was present or not, which suggested that in this portion of the operon, RteR had no effect. Beyond traA, however, the relative level of transcript rapidly decreased. RT-qPCR was performed to confirm our qualitative RT-PCR results. A similar relative level of transcript was observed in the traA portion; however, immediately downstream of traA the relative percentage of transcript was dramatically reduced (Fig. 7). This observation demonstrates that RteR does not prevent transcription, as previously suggested, and the target of RteR is downstream of traA. We predict that the target of RteR may in fact be in the region between traA and traC.
RT-qPCR analysis was used to measure the relative percentage of tra mRNA in the presence of RteR. RT-qPCR was performed to assess the levels of tra transcript detected when RteR was present, relative to the amount of transcript when RteR was absent. Samples containing RteR were normalized against samples without RteR to obtain the relative percentage of transcript remaining in the presence of RteR; these values are shown above each primer set. The percentage shown for each primer set is an average value of relative transcript present.
The half-life of tra mRNA is not affected by the presence of RteR.The half-life of the tra mRNA transcript was measured to better ascertain how RteR might be working to inhibit conjugative transfer. One possibility is that RteR exhibits an effect on the stability of the tra transcript. If the tra mRNA half-life decreased when RteR was present, this would suggest that RteR works to recruit ribonucleases to the RteR-tra mRNA duplex. This type of interaction is often mediated by binding of the sRNA to the target mRNA transcript via a chaperone protein, such as Hfq (22). The complex is then targeted for RNase-mediated decay, resulting in a decreased half-life of the target mRNA (25).
The half-life of pLYL72 tra mRNA was measured in both the presence and absence of RteR. At traA, traB-traC, and traD, the half-life of the tra transcript was approximately 1.63 min, whether or not RteR was present, demonstrating that RteR does not act to destabilize the tra mRNA transcript by means of RNase-dependent decay (Fig. 8).
The half-life of tra mRNA is similar whether or not RteR is present. A rifampin chase assay was performed to assess the stability of the tra mRNA transcript in both the presence and absence of RteR. Rifampin at 400 μg/ml was added to mid-log-phase Bacteroides cultures, and RNA samples were taken at 0, 5, and 10 min post-rifampin addition. Relative RNA levels were then measured using RT-qPCR and were quantified by normalizing the abundance of RNA remaining over time relative to that at time zero. Primer sets were used to detect the message at traA, traB-traC, and traD. At all points of the transcript, the half-life was approximately 1.6 min whether or not RteR was present, suggesting that RteR does not destabilize the tra mRNA transcript.
RteR has no effect on excision from the chromosome and does not regulate the mob genes.Many steps are required for transfer of conjugative transposons, such as excision from the chromosome, nicking of the OriT, and assembly of the mating bridge (28). To investigate the effect of RteR on excision from the chromosome, we introduced rteR on a plasmid into a strain containing CTnERL. CTnERL is essentially identical to CTnDOT but lacks the erm(F) region. We used this strain rather than a strain containing CTnDOT, so that we could use erythromycin as a selectable marker for the plasmid containing rteR, pJW305. Although CTnERL already contains a single copy of rteR, by placing rteR on a plasmid in trans we could determine whether there was a reduction of the excised element associated with increasing the copy number of RteR. Using RT-PCR, we found that the copy number of RteR did not affect the amount of excision from the chromosome, suggesting that RteR does not exhibit any regulatory effect on excision (data not shown).
We also used RT-PCR to determine if RteR had any regulatory effect on expression of the mob genes. We observed similar levels of mobA, mobB, and mobC transcript whether RteR was present or not, confirming that the mob genes are not a target of RteR (data not shown).
DISCUSSION
CTnDOT has proved to be an interesting mobile genetic element, with a complex regulatory system in which stimulation of excision and transfer is dependent upon low levels of tetracycline (21). A small RNA, RteR, was previously suggested as a possible negative regulator that is responsible for inhibition of conjugative transfer in the absence of tetracycline (7, 26). Previous studies had indicated that RteR was an approximately 100-nt sRNA (7), but these studies had only localized RteR to a 500-bp region (27). We report here that the region required for mediating repression of conjugative transfer contains a 90-nt noncoding RNA, RteR. We have identified both the 5′ and 3′ ends of the RteR transcript, and we have also confirmed that the promoter is located at the 3′ end of exc.
RteR was previously reported to act within the 13-kb transfer (tra) operon (7, 26). Previous studies suggested that RteR prevented transcription of the entire tra operon (7), but we demonstrated here that this is not the case. The previous study had taken a similar approach as we present here, detecting the tra transcript with and without RteR present. However, that study only investigated traG, assuming that because the tra genes are organized in an operon that any effect at traG would be representative of the entire tra operon. Their results demonstrated that there was barely any traG transcript detectable when RteR was present, leading to the conclusion that RteR was preventing transcription of the entire operon. We report that RteR acts to prevent efficient transcription downstream of traA, but traA itself is still transcribed.
Our observation that there is a significant decrease in the amount of transcript downstream of traA, coupled with the finding that RteR does not result in a decreased half-life of the tra mRNA, suggests that RteR might enhance premature transcription termination. To support this finding, we analyzed the tra sequence downstream of traA and found a region within traB that has the potential to form an intrinsic terminator. If this terminator structure within traB is what forms to mediate premature transcription termination, we propose that RteR may be binding the elongating tra mRNA, thus altering the tra secondary structure. This alteration then results in the formation of this intrinsic terminator. The result would be to abolish transcription of the remaining tra operon and ultimately lead to inhibition of CTnDOT conjugative transfer.
Upstream of this predicted terminator is a region that is complementary to RteR. Coincidentally, this region of RteR is within the 8-nt single-stranded loop and is also one of the only regions of RteR that was conserved in all of the homologues identified. This could explain why the mutant pJW312 (Fig. 6A and B) can no longer repress pLYL72 transfer, because the site-directed mutations resulted in this region becoming double stranded, so it may not have been able to bind the traB transcript efficiently. Conversely, our mutant pJW318 (Fig. 6A and B) may still be able to bind to traB because there is still a largely single-stranded loop that has the potential to interact with traB despite the change in a part of the single-stranded loop.
Our results do not suggest how CTnDOT is able to overcome RteR-mediated negative regulation upon tetracycline induction. We posit that perhaps the activation of the tra operon by Xis2c, Xis2d, and Exc overrides the effects of RteR. Another possibility is that upon tetracycline induction, a protein may sequester RteR, thereby overriding any negative regulation of RteR.
We have demonstrated that rteR is not only present in Bacteroides spp. but also is present in other members of the Bacteroidetes phylum. This finding is consistent with the observation that CTnDOT is readily transferred within the phylum. CTnDOT-like elements have been observed in close relatives of Bacteroides, such as Prevotella, Porphporymonas, Tannerella, and Cytophaga species. Our database search revealed that rteR is also present on a related conjugative transposon, CTn341, and is found in Bacteroides fragilis strain YCH46 that contains a CTnERL-like element in the chromosome (24). We also report that we were able to PCR amplify and sequence rteR from other Bacteroides strains harboring the following conjugative transposons: CTnERL, CTn12256, and CTnV479. RteR is the only sRNA thus far described in this phylum. RteR is also the first regulatory sRNA to be found on a conjugative transposon.
ACKNOWLEDGMENTS
We are grateful to Carin Vanderpool for thoughtful discussion and to Nadja Shoemaker for reading of the manuscript. We also thank Jennifer Rice and Divya Balasubramanian for comments and advice with RNA techniques. We thank Erika Bongen for assistance with the database search for RteR homologues.
This work was supported by grant AI 22383 from the National Institutes of Health.
FOOTNOTES
- Received 1 June 2012.
- Accepted 14 July 2012.
- Accepted manuscript posted online 20 July 2012.
- Copyright © 2012, American Society for Microbiology. All Rights Reserved.
