- Open Access
Tn5 transposition in Escherichia coli is repressed by Hfq and activated by over-expression of the small non-coding RNA SgrS
Mobile DNA volume 5, Article number: 27 (2014)
Hfq functions in post-transcriptional gene regulation in a wide range of bacteria, usually by promoting base pairing of mRNAs with trans-encoded sRNAs. It was previously shown that Hfq down-regulates Tn10 transposition by inhibiting IS10 transposase expression at the post-transcriptional level. This provided the first example of Hfq playing a role in DNA transposition and led us to ask if a related transposon, Tn5, is similarly regulated.
We show that Hfq strongly suppresses Tn5 transposition in Escherichia coli by inhibiting IS50 transposase expression. However, in contrast to the situation for Tn10, Hfq primarily inhibits IS50 transposase transcription. As Hfq does not typically function directly in transcription, we searched for a transcription factor that also down-regulated IS50 transposase transcription and is itself under Hfq control. We show that Crp (cyclic AMP receptor protein) fits these criteria as: (1) disruption of the crp gene led to an increase in IS50 transposase expression and the magnitude of this increase was comparable to that observed for an hfq disruption; and (2) Crp expression decreased in hfq−. We also demonstrate that IS50 transposase expression and Tn5 transposition are induced by over-expression of the sRNA SgrS and link this response to glucose limitation.
Tn5 transposition is negatively regulated by Hfq primarily through inhibition of IS50 transposase transcription. Preliminary results support the possibility that this regulation is mediated through Crp. We also provide evidence that glucose limitation activates IS50 transposase transcription and transposition.
Transposase proteins catalyze the chemical steps in bacterial transposition reactions. It follows that the regulation of expression of these genes is a critical feature in dictating the transposition frequency of most transposons. In many instances, including Tn10/IS10 and Tn5/IS50, transposase gene promoters are inherently weak. In addition, DNA adenine methylase (DAM) limits initiation of IS10 and IS50 transposase gene transcription by methylating promoter elements ,. These factors together make transcription initiation a limiting step in Tn10/IS10 and Tn5/IS50 transposition reactions ,. There are also examples where translation of transposase transcripts is subject to both intrinsic and host levels of regulation. In the case of IS10 transposase, the ribosome binding site is inherently weak and the transposon encodes an antisense RNA that binds the translation initiation region (TIR), blocking ribosome binding ,. There is also evidence that the ‘host’ protein Hfq helps mediate the pairing interaction between the antisense RNA and the IS10 transposase transcript ,.
Hfq is a global regulator of gene expression in bacteria. It typically functions at the post-transcriptional level, influencing translation initiation and/or transcript stability by catalyzing the pairing of small RNAs (sRNA) and their mRNA targets (Figure 1B and reviewed in ). In contrast to the many examples of Hfq acting in a post-transcriptional capacity to impact gene expression, there is (to our knowledge) only one example in the literature of Hfq acting at the level of transcription to influence gene expression. In the case of ribosomal proteins rpsO, rpsT and rpsB-tsf, Hfq was shown to increase transcript levels without influencing transcript stability. It was suggested that this is accomplished through Hfq binding to secondary structure elements in the respective transcripts that form early in the elongation phase of transcription and that this interaction reduces RNA polymerase pausing .
As noted above, Hfq has been implicated in the regulation of Tn10/IS10 transposition. Under conditions of hfq deficiency, a large increase in both Tn10/IS10 transposition (up to 80-fold) and transposase expression (up to 7-fold) were observed. The existing evidence is consistent with Hfq acting as a negative regulator of IS10 transposase expression by both antisense dependent and independent pathways. In support of the latter, it was found that hfq deficiency (or hfq−) had a significant impact on Tn10 transposition even when the level of antisense RNA was insufficient to impact on transposase expression (that is when Tn10 is present in single copy in the bacterial chromosome). In addition, there was a synergistic increase in transposase expression when both hfq and the antisense RNA were knocked out, implying that Hfq does not function exclusively in the same pathway as the antisense RNA .
Taking the above results into account, and considering that most bacterial transposition systems are not regulated by antisense RNAs, we wondered if Hfq might play a more general role in regulating transposition systems. In the current work, we tested this hypothesis by asking if Tn5 transposition is also regulated by Hfq. Like Tn10, Tn5 is a composite transposon (Figure 1A). The two transposons are closely related but Tn5 lacks an antisense RNA regulatory system and consequently if Hfq were to regulate this system at the post-transcriptional level, it is likely that a trans-encoded sRNA would play a role -. Tn5 does encode an inhibitor protein that limits Tn5/IS50 transposition by dimerizing with the transposase protein, forming an inactive complex . Transposase and the inhibitor protein are expressed from overlapping promoters, P1 and P2 (color coded in Figure 1A), with the inhibitor transcript (T2) being expressed at a higher level than the transposase transcript (T1). T1 expression is down-regulated by DAM (reviewed in ). There is some evidence that P1 is also negatively regulated by LexA, an SOS-inducible transcriptional repressor . However, there is little else known with regard to host proteins that influence either transposase transcription or translation.
In the current work, we show that both Tn5 transposition and IS50 transposase expression increase significantly in E. coli under conditions of hfq deficiency. However, unlike the situation in Tn10/IS10 transposition, the up-regulation of IS50 transposase expression appears mainly to be due to an increase in transposase gene transcription. As Hfq does not typically function directly in transcription, we looked at the possibility that Hfq regulates IS50 transposase expression by controlling the expression of a transcription factor. Towards this end, we provide evidence that Hfq acts in a regulatory network with Crp (cyclic AMP receptor protein) to down-regulate IS50 transposase transcription. Finally, we demonstrate that over-expression of an sRNA (SgrS) activates expression of the IS50 transposase gene specifically when cells are grown with glucose as the sole carbon source. Evidence is presented that this up-regulation is a consequence of glucose limitation, demonstrating that the IS50 transposase promoter (and Tn5 transposition) is responsive to the nutrient status of the cell.
Hfq is a potent negative regulator of Tn5 transposition
We asked if Hfq regulates Tn5 transposition in E. coli by measuring the frequency of Tn5 transposition under conditions of hfq deficiency using the ‘mating out’ assay. In this assay, an F+ donor strain harboring a chromosomal copy of Tn5 was mated to an F− recipient strain and the mating efficiency and number of transposition events were measured by plating mating mixes on the appropriate selective media (see Methods). We show in Figure 2A that in one donor strain background (DBH179) Tn5 transposition increased by close to 75-fold under conditions of hfq deficiency. Note that we did not have a defective copy of Tn5 to act as a negative control in this experiment. In lieu of this, we carried out physical mapping on a sampling of colonies present on ‘hop’ plates to ensure that bona fide transposition events were being measured in both wt and hfq− strains (Additional file 1).
We also performed a complementation assay in the DBH179 strain background to further test that the increase in transposition reported above in hfq− was actually due to the absence of Hfq, as opposed to possible polar effects of the hfq disruption allele. Towards this end, we introduced hfq on a low-copy plasmid (pDH700) into the hfq− strain and measured Tn5 transposition as above. We observed nearly complete complementation by plasmid-borne hfq, as transposition was reduced approximately 45-fold relative to when no hfq was present (Figure 2A). Furthermore, plasmid-encoded variants of hfq, including K56A and Y25A, which are impaired for RNA-binding at the ‘proximal’ and ‘distal’ surface, respectively, failed to complement hfq deficiency . This confirms that specific functions of Hfq, namely interaction with RNA via known RNA-binding surfaces, are required for effective repression of Tn5 transposition.
We also tested the impact of hfq deficiency on Tn5 transposition in a second donor strain background (DBH261) via the ‘mating out’ assay (Figure 2B). In this experiment hfq− also caused an increase in Tn5 transposition, although the magnitude of the effect was smaller (approximately 9-fold) than reported for the DBH179 strain background.
IS50 transposase expression increases in hfq−cells
We next asked if hfq status influenced IS50 transposase expression. In one approach, we measured transposase expression by constructing IS50-lacZ transcriptional and translational fusions (‘TCF’ and ‘TLF’, respectively; see Figure 3A for schematics), integrating these reporters into the chromosome of a lac−E. coli strain (DBH107), and then performing β-galactosidase assays. This was done for each reporter in isogenic strains that were either wt, dam− or hfq−. As expected for a promoter that is DAM-sensitive, transposase expression increased in the context of both transcriptional and translational fusions in the dam− strain relative to wt (approximately19- and 25-fold, respectively; Figure 3B). The increase in transposase expression for both constructs in dam− is indicative of expression coming predominantly from the P1 promoter . Transposase expression in TCF and TLF constructs also increased in hfq− cells (11-fold and 7.4-fold, respectively), indicating that Hfq (or a factor under Hfq control) represses IS50 transposase expression. As the TCF encodes only 15 nucleotides of the transposase transcript (T1), it seemed most likely that up-regulation of transposase expression in hfq− was primarily due to enhanced transcription in both TCF and TLF constructs.
To further test this possibility, we constructed a TLF reporter (pDH908) wherein the IS50 transposase promoter was replaced by a heterologous promoter (from the lpp gene) whose regulation is not sensitive to hfq status . An isogenic plasmid (pDH795) in which the TLF contained the IS50 transposase promoter was also constructed. Cells (wt or hfq−) were transformed with plasmids containing these constructs and reporter expression was measured as above. The results presented in Figure 4 show that transposase expression increased approximately 9-fold for the construct containing the IS50 promoter and less than 2-fold for the construct containing the lpp promoter. These results support our contention that Hfq-directed regulation of IS50 transposase expression occurs at the transcriptional level because the absence of the IS50 promoter and not the presence of the IS50 5′ UTR was the dominant factor in observing strong up-regulation of reporter expression under conditions of hfq deficiency.
Hfq impacts steady-state levels of full-length IS50 transposase mRNA
To further assess the impact of hfq deficiency on transposase gene expression, we looked at both the steady-state level and the stability of the transposase transcript (T1) in hfq+ and hfq− cells. For the steady-state analysis, total RNA was isolated from various strains (wt or hfq−) (DBH33 background) containing a multi-copy plasmid encoding the full-length transposase gene under the control of its native promoter. In addition to the wt version of this plasmid (pDH533), we also analyzed a mutant form containing mutations in the overlapping dam methylation sites in the transposase promoter (pDH752) (see Figure 1A); these mutations make this construct DAM insensitive. Primer extension was used to detect both T1 and T2 transcripts, as well as the lpp transcript (loading control). As expected for a dam-sensitive promoter, levels of T1 increased substantially (approximately 8-fold) in wt cells containing the plasmid with the dam-insensitive promoter versus wt cells containing the wt promoter (compare lanes 3 to 7 with lanes 8 to 12 in Figure 5A and bar graph in Figure 5B). In contrast, there was no significant change in T2 levels in the above samples. In hfq− (wt promoter) there was also a substantial increase in T1 levels (11-fold) versus the wt strain (compare lanes 3 to 7 with lanes 14 to 18) and no significant change in T2 levels. Thus in an hfq− background there was an increase in the steady-state level of transposase transcript and this increase was slightly greater than that observed when methylation of the transposase promoter was blocked.
We also looked at the combined impact of knocking out Hfq and blocking DAM methylation on T1 levels (lanes 19 to 23 in Figure 5A). In comparison to wt, the ‘double mutant’ situation resulted in a 45-fold increase in T1 levels. Based on the observed synergy, we think it unlikely that the observed impact of deleting hfq is linked to the regulation of dam expression.
To directly test if a component of Hfq-directed repression of IS50 transposase expression is post-transcriptional, we compared the stability of the IS50 transposase mRNA (T1) in isogenic wt and hfq− strains. Total RNA was isolated from a pair of rifampicin-sensitive strains (TM338 and TM618) containing a plasmid encoding IS50 transposase (pDH533) before and after rifampicin treatment as shown in Figure 6. Transposase mRNA was detected by primer extension. In the hfq− strain the half-life of the T1 transcript increased by approximately 1.7-fold, revealing that hfq status does impact on transposase mRNA stability.
Taken together, the results from Figures 3, 4, 5 and 6 show that IS50 transposase expression is substantially reduced in an hfq+ relative to an hfq− strain and that hfq status primarily affects transposase transcription.
Regulation of Tn5 transposase expression by global transcriptional regulators
As Hfq does not typically function directly in transcription, we set out to define a transcription factor that down-regulates IS50 transposase transcription and is itself regulated by Hfq. Toward this end, we asked if disrupting genes for two global transcription factors, Crp and Lrp , had an impact on IS50 transposase expression. Note that we had to construct new TCF reporter strains for this work because the available crp and lrp disruption strains we used to transduce DBH107 to either crp− or lrp− were marked with the same antibiotic resistance gene used to select for a lysogen with a chromosomal copy of the TCF. We show in Figure 7A that crp− but not lrp− had a substantial impact on transposase expression. For example, in cells grown in exponential phase in Luria broth (LB), there was up-regulation of transposase expression (approximately 4-fold) in both crp− (DBH307) and hfq− (DBH306) strains but not in the lrp− strain (DBH315). We also performed semi-quantitative RT-PCR and show that transposase-lacZ transcript levels increased similarly in crp− and hfq− strains (Figure 7B). These results are consistent with Crp being a negative regulator of IS50 transposase transcription.
We next asked if Crp expression was regulated by Hfq. Notably, work done in Yersinia pestis has shown that Hfq positively regulates Crp expression at the post-transcriptional level . Towards this end we performed Western blot analysis with a Crp antibody on E. coli cell extracts from wt (DBH303), hfq− (DBH306) and crp− (DBH307) strains (Figure 7C). The results show that lower levels of Crp are present in the hfq− strain, which is consistent with Hfq also being a positive regulator of Crp expression in E. coli.
Finally, we assessed the impact of knocking out crp on Tn5 transposition frequency using the ‘mating out’ assay (Figure 7D). In the absence of crp, Tn5 transposition increased 7-fold, which is consistent with results from the transposase expression experiments.
IS50 transposase expression and Tn5 transposition are up-regulated by over-expression of the sRNA SgrS
Over-expression of sRNAs can alter Hfq-regulated networks by limiting the availability of Hfq ,. Given our findings that Tn5 transposition and transposase gene expression are affected by hfq status, we asked if IS50 transposase expression might be sensitive to Hfq-titration. Towards this end, we measured transposase expression from the TLF under conditions where a single sRNA was over-expressed from an IPTG inducible promoter (pLlacO) in DBH33, which is lacIq. Our initial screen included four different Hfq-dependent sRNAs, including RybB, RyeB, MicC and SgrS, all of which are expected to tightly bind Hfq in vivo; apparent equilibrium dissociation constants of approximately 3.3 nM and < 20 nM have been measured for MicC and SgrS, respectively -. Cells were grown in M9 glucose and sRNA expression was induced for 4 hours in exponential phase. We show in Figure 8A that only one of the sRNAs tested, SgrS, had a significant impact on transposase expression. Induction of SgrS increased transposase expression just over three-fold. Given the comparable Hfq binding affinities of the sRNAs tested, it seemed unlikely that SgrS expression was increasing transposase expression through an Hfq-titration mechanism.
SgrS down-regulates the expression of several known targets, including the primary glucose transporter encoded by the ptsG gene, a mannose transporter encoded by manXYZ and it up-regulates the expression of yigL, a phosphatase involved in phospho-sugar detoxification . As we observed up-regulation of IS50 transposase expression in cells over-expressing SgrS in M9 glucose media, we considered the possibility that this effect was a response to glucose limitation. In fact, we show in Additional file 2 that induction of SgrS in M9 glucose resulted in a substantial slowing of bacterial growth, as would be expected if nutrients had become growth-rate limiting. To further test the glucose limitation hypothesis, we performed a similar experiment in rich media (LB) and in M9 glucose supplemented with glycerol, a carbon source whose import is not dependent on glucose transporters . We also tested the response of the reporter to over-expression of an SgrS mutant, sgrS1, that is incapable of down-regulating glucose import . In these experiments we used a Tn5 TCF as a reporter in the DBH107 strain background; DBH107 has a complete deletion of the lac operon and consequently the plasmid-encoded sRNA genes are constitutively expressed. To avoid problems in growing these cells, cultures were initially propagated in either LB or M9 glucose/glycerol and then where indicated, switched to other media.
We show in Figure 8B that after approximately 4 hours of SgrS over-expression in M9 glucose, reporter expression increased close to 5-fold relative to a ‘vector’ control. In contrast, over-expression of SgrS1 was incapable of up-regulating reporter expression under these same conditions, suggesting that SgrS must be able to down-regulate glucose import and or retention in order to increase transposase transcription. When cells were grown in M9 glucose supplemented with glycerol, expression of SgrS as above caused only an approximately 2-fold increase in transposase expression. Importantly, the reduced effects of SgrS on transposase expression under ‘glycerol’ conditions cannot be explained by differential expression of the respective sRNAs, as levels of SgrS and SgrS1 were similar in M9 glucose with or without glycerol (Figure 8C). Also, we failed to see significant transposase induction when SgrS was over-expressed in LB media where there are multiple carbon sources. Finally, consistent with the glucose limitation hypothesis, we also show in Figure 8B that increased transposase expression resulting from SgrS expression in M9 glucose was the only condition that inhibited cell growth.
Given that transposition frequency is expected to be roughly proportional to transposase expression, we also asked if glucose limitation had an impact on Tn5 transposition. Cells encoding a chromosomal copy of Tn5 were transformed with an SgrS-expressing plasmid (or vector only control) and the frequency of Tn5 transposition was measured using the ‘mating out’ assay. Note that cells were grown in M9 glucose media and SgrS expression was induced only when donor strains were subcultured on the day of mating. We show in Figure 9 that induction specifically of SgrS resulted in a 5-fold increase in Tn5 transposition relative to the vector only control. Notably, when cells were grown in M9 supplemented with glucose and glycerol, induction of SgrS did not result in a significant increase in Tn5 transposition. Also, we observed a reduced growth rate only in cultures where SgrS was induced in M9 glucose media (data not shown). The results of the ‘mating out’ analysis are thus entirely consistent with the gene expression experiments presented in Figure 8.
Hfq is a global regulator of gene expression in bacteria. However, until recently, Hfq had not been linked to the control of transposable elements. Work in the Tn10/IS10 system provided the first example of Hfq inhibiting a transposon . In the current work, we asked if the transposition of a related element, Tn5/IS50, is also regulated by Hfq. We show that Tn5 transposition and IS50 transposase expression are repressed by Hfq; however, the mechanism of repression is atypical for Hfq, involving predominantly a block in IS50 transposase transcription. Preliminary evidence is presented that is consistent with Hfq modulating IS50 transposase transcription through regulation of Crp. We also show that transposase transcription and Tn5 transposition are activated by over-expression of the sRNA SgrS and provide evidence that this is a transcriptional response to glucose limitation.
Hfq negatively regulates Tn5 transposition
The results of ‘mating out’ experiments were consistent with Hfq acting as a strong negative regulator of Tn5 transposition. Tn5 transposition increased close to 75-fold in one hfq− strain (DBH179 background). The magnitude of this increase was somewhat surprising given that up-regulation of Tn10 in hfq−, under essentially antisense-minus conditions, was about 7-fold . However, in a different hfq− strain (DBH261 background) Tn5 transposition increased only 9-fold. At this point it is unclear why there was such a large discrepancy in the ‘mating out’ values for the two strains. One possibility is that colony counts in the DBH179 ‘mating out’ (hfq−) included clones that had ‘jack-pot’ events. That is, colonies were counted that did not derive from independent transposition events. This could explain the high standard error associated with the transposition frequency in the hfq− column in Figure 2A. If, for example, we removed the 3 most prominent outliers from the (DBH179) hfq− data set, the fold increase in transposition dropped to 15-fold, which is more in line with what we observed in the DBH261 strain background and for Tn10 in single copy .
A trans-complementation (Figure 2A) experiment provided definitive proof that the increase in Tn5 transposition detected in one of our hfq− ‘mating out’ strains (DBH179 background) was in fact due to hfq deficiency. In addition, the failure of two Hfq RNA-binding face mutants to provide complementation was consistent with Hfq-directed inhibition of Tn5 transposition relying on functions of Hfq required in canonical Hfq-directed regulatory pathways . That is, Hfq must retain the ability to bind both mRNAs and sRNAs to influence Tn5 transposition.
Hfq, Crp and IS50 transposase gene expression
Evidence that hfq status influences IS50 transposase expression came from two types of experiments. First, the expression of transposase-lacZ reporter genes in both transcriptional and translational fusion constructs increased significantly under conditions of hfq deficiency. Second, the steady-state level of the native transposase transcript also increased significantly in hfq−. Importantly, the large increase in steady-state transcript level (11-fold) coincided with a less substantial increase in transposase mRNA stability (less than 2-fold increase in half-life). In addition, up-regulation of reporter expression in hfq− for a TLF was almost completely abrogated when the IS50 transposase promoter was replaced by a heterologous promoter. Taken together, these results are consistent with Hfq (or a factor regulated by Hfq) suppressing IS50 transposase expression predominantly at the level of transcription. Notably the suppressive effect of Hfq on IS50 transposase transcription was remarkably specific, as the level of a second transcript (T2) encoded by IS50 was not affected by hfq status.
As Hfq does not typically act directly in gene transcription, we think it likely that Hfq acts indirectly on the IS50 transposase promoter. In addition to DAM, only one other transcription factor, LexA, has been implicated as a regulator of transposase transcription. There is a weak LexA-binding site in the transposase promoter (Figure 1A); however, lexA deficiency was shown to increase transposase transcription only two to three-fold in a TCF . As we have seen increases in transposase expression of up to 11-fold for a TCF in hfq−, it seems unlikely that Hfq would be working through LexA. In contrast, transposase expression increased in dam− to a level more in line with that observed in hfq− (less than two-fold difference in the TCF). However, the observed synergy between hfq− and mutations that rendered the IS50 transposase promoter DAM-insensitive led us to conclude that Hfq does not regulate IS50 transcription by impacting DAM levels (and, therefore, promoter methylation). These results provided motivation to search for other targets of Hfq that impinge on IS50 transposase transcription. This search identified Crp as an additional negative regulator of IS50 transposase transcription. Notably, transposase expression increased to approximately the same level in crp− and hfq− in the experiment in Figure 7. The similar magnitude of up-regulation of transposase expression in hfq− and crp− could be indicative of Hfq acting upstream of Crp to inhibit transposase expression. We did in fact find evidence of Hfq positively regulating Crp protein levels (Figure 7C). This observation is consistent with work recently published in the Y. pestis system where it was found that Crp protein levels decreased approximately five-fold in an hfq disruption strain .
Crp is a known activator/repressor of transcription  and, therefore, more likely than Hfq to be directly involved in regulating IS50 transposase expression at the transcriptional level. Given our evidence that Hfq positively regulates crp expression, a plausible scenario explaining our expression data is that the observed up-regulation of IS50 transposase transcription in hfq− is a result of decreased Crp protein levels. Crp may act either directly or indirectly on the IS50 transposase promoter to repress transcription. This is currently a working model as we have not yet tested the possibility that Crp binds the IS50 transposase promoter and it may only be coincidental that transposase expression increased to similar levels in hfq− and crp− strains. Notably, we also found that Tn5 transposition increased when the crp gene was disrupted, although the extent of the increase was smaller than that observed in the isogenic hfq disruption strain. This could be indicative of additional factors in the Hfq regulon impinging on Tn5 transposition.
There is precedent for Crp down-regulating the transcription of a transposase gene. In the case of IS2, transposase transcription increased close to 200-fold in crp−. It was also shown through protein-DNA footprinting that Crp binds directly to the IS2 transposase promoter . Interestingly, based on the consensus binding sequence for Crp, the authors of the above study predicted that Crp would bind to the IS50 transposase gene. However, the predicted crp binding site is located downstream of the transposase promoter and is not present in our TCF (where we detected increased transposase expression in crp−). Nevertheless, it would be worthwhile to test for Crp binding to the IS50 transposase promoter as the results of Crp ChIP-chip studies revealed the presence of thousands of weak crp binding sites scattered throughout the E. coli genome . It is also possible that Crp acts indirectly on the IS50 transposase promoter by regulating the expression of another transcription factor.
Tn5 transposition and metabolic stress
We also identified conditions that activate transposase expression and transposition; over-expression of the sRNA SgrS increased transposase expression and transposition approximately five-fold. We favor the possibility that this induction is a consequence of glucose limitation but cannot formally rule out the possibility that SgrS targets an as yet undefined regulatory pathway that impinges on transposase expression. Our reasoning for this is that we observed induction of transposase expression and transposition specifically when cells were grown with glucose as the major carbon source and SgrS is known to prevent expression and function of the major glucose transporter encoded by the ptsG gene . Consistent with this idea, we found that transposase induction levels correlated with a reduced growth rate. Furthermore, we demonstrated that: (i) an allele of SgrS (sgrS1) that is incapable of down-regulating ptsG expression failed to induce transposase expression in M9 glucose; (ii) under conditions where SgrS was expressed in M9 glucose media supplemented with glycerol, we failed to see induction of transposase expression to the same extent as when glycerol was absent; (iii) SgrS expression did not impact transposase expression when cells were grown in rich media (LB) and (iv) over-expression of 3 other sRNAs (RybB, RyeB and MicC) that are not expected to influence glucose transport did not increase transposase expression in M9 glucose -. Precedent for nutritional stress influencing transposition comes from earlier work in the IS903 system where mutations in a gene (aspA) required for fermentative metabolism during anaerobic growth caused transposition to occur at an accelerated rate .
At this point it is unclear as to what factors are driving the induction of the transposase gene under SgrS over-expression conditions. With regard to further defining the mechanism of IS50 transposase up-regulation under SgrS over-expression conditions, it would also be advantageous to find alternative experimental conditions for achieving this increased expression. If, for example, simply starving cells by restricting a carbon source during growth achieves the same end as over-expressing SgrS in M9 glucose media, an unbiased screen to search for genetic factors that are necessary for the up-regulation of transposase expression could be performed to reveal the regulatory network impinging on the transposase promoter. As it stands, any factors that influence SgrS expression would interfere with the outcome of such a screen. Alternatively, if it was found that restricting glucose is not sufficient for inducing transposase expression, the possibility that SgrS plays a more direct role in controlling transposase expression would have to be considered.
In this work, we have identified several genes that impact on IS50 transposase expression, including hfq, crp and sgrS. Hfq and Crp proteins are negative regulators and SgrS RNA (under specific growth conditions) is a positive regulator of transposase gene expression. Exactly how these factors impinge on transposase expression remains to be worked out and at this point it is not clear if we are seeing modulation of the same regulatory network in opposite directions when hfq and crp genes are disrupted and SgrS RNA is over-expressed. Tn5/IS50 is the second transposon identified that is affected by disruption of the hfq gene and the first that does not encode an antisense RNA. This raises the possibility that Hfq influences the transposition frequency of many other bacterial transposons.
Plasmids, bacteriophage and strains
The IS50 translational fusion plasmid (pDH798) is a pWKS30-derivative containing base pairs 1 to 431 of IS50 (nucleotides 1 to 366 of T1) fused to codon 10  of the E. coli lacZ gene. The IS50 transcriptional fusion plasmid (pDH682) is a pUC18-derivative containing base pairs 1 to 80 of IS50 (nucleotides 1 to 15 of T1) fused to nucleotide -16 (relative to the translational start codon) of lacZ. Plasmids encoding sRNAs (pDH764, sgrS; pDH766, rybB; pDH768, micC; pDH772, ryeB) and the corresponding empty vector control (pDH763) were kindly provided by S Gottesman. The plasmid encoding sgrS1 (pDH895) was kindly provided by C Vanderpool. Plasmids encoding Hfq (pDH700, wt) and mutant derivatives (pDH701, K56A; pDH713, Y25A) are described in Ross et al. Details of plasmid constructions are provided in Additional file 3 and a list of oligonucleotides used in this work is provided in Additional file 4.
Lambda phages encoding IS50 transcriptional (λDBH849 and λDBH888) and translational (λDBH812) reporters were generated by cloning IS50 expression cassettes marked with an antibiotic resistance gene (either kanR or cmR) into the his operon of pNK81 and then infecting a strain harboring one of these plasmids with λNK1039, which also contains the his operon. Antibiotic resistant lysogens from the above crosses were selected by replica plating and subsequently phage released from the lysogens were purified, giving rise to λDBH849 (IS50-lacZ-kanR TCF), λDBH888 (IS50-lacZ-CmR TCF) and λDBH812 (IS50-lacZ-KanR TLF).
E. coli strains for the ‘mating out’ assay were constructed by P1 transduction of Tn5 from ER2507 (NEB) into DBH33, DBH344 and DBH259. Strains containing chromosomal IS50-lacZ fusions were generated by lysogenizing DBH107 with λDBH849 (DBH265), λDBH888 (DBH303) or λDBH812 (DBH281). Mutant derivatives of these strains were generated by P1 transduction. A list of all of the strains, plasmids and bacteriophage used in this work is presented in Table 1.
‘Mating out’ assay
Conjugal ‘mating out’ experiments were performed essentially as described for single-copy chromosomal transposons in Ross et al. , except that for measuring transposition in hfq− versus wt, donor growth was carried out in M9 glucose media supplemented with kanamycin (25 μg/mL) and amino acids, instead of LB. DBH13 was used as the recipient. Total exconjugants and transposition events with DBH179 and derivatives were scored by plating mating mixes on M9 glucose plates supplemented with leucine, thiamine and streptomycin (150 μg/mL) or streptomycin and kanamycin (25 μg/mL), respectively. Total exconjugants and transposition events with DBH261 and derivatives were scored by plating mating mixes on M9 glucose plates supplemented with leucine, thiamine, streptomycin (150 μg/mL) and gentamicin (12.5 μg/mL) or streptomycin, gentamicin and kanamycin (25 μg/mL), respectively.
Cells were grown in M9 glucose (with arginine and thiamine) or LB. In situations where strains contained plasmids, plasmids were maintained by including the appropriate antibiotic. Overnight cultures (0.05 mL) were used to seed subcultures (1.5 mL), which typically were grown to mid-log phase before being processed for the Miller assay as previously described .
RNA isolation, primer extension and Northern blot analysis
Total RNA was isolated essentially as described in . For steady-state analysis, cells were grown to mid-log phase in LB before RNA isolation. For half-life analysis, rifampicin (dissolved in dimethyl sulfoxide; DMSO) was added to cell cultures (to 200 μg/mL) to arrest transcription and RNA was isolated immediately before and after rifampicin addition at the indicated time intervals. Primer extension analysis was carried out using 32P-labeled primers oDH230 and oDH390, end-labeled with OptiKinase (USB, Cleveland, OH, USA) according to manufacturer’s instructions. Extension reactions used 5 μg of RNA, and Superscript III reverse transcriptase essentially as described in , except that annealing was performed at 65°C (with no ice treatment) before extending at 55°C for 45 minutes. Extension products were resolved on 6% and 10% denaturing polyacrylamide gels. For Northern blot analysis, 2 μg of RNA was mixed with an equal volume of denaturing load dye (95% deionized formamide [v/v], 10 mM EDTA, 0.5× TBE, 3% xylene cyanol [w/v]), heated to 95°C for 2 minutes, and resolved on a 6% polyacrylamide gel containing 7 M urea. Separated RNAs were electro-transferred to Hybond N (GE Healthcare, Mississauga, ON, Canada) in 0.5× TBE and fixed with UV. Annealing and washing was performed in ULTRAhyb buffer (Ambion, Burlington, ON, Canada) according to the manufacturer’s instructions, using RNA probes complimentary to SgrS or the 5S rRNA (internal standard). To construct the radiolabeled RNA probes, DNA templates for in vitro transcription were made by PCR with primers oDH232/233 (SgrS) and oDH234/235 (5S rRNA) - note that, for each primer pair, the forward primer includes the T7 core promoter. These templates were transcribed in vitro in the presence of 32P-UTP to generate uniformly labeled RNA probes. In vitro transcription reactions were performed in 25 μL volumes with approximately 1 μg DNA template, 1 × T7 RNA polymerase buffer (NEB, Beverly, MA, USA), 20 units RNasin (Promega, Madison, WI, USA), 4 mM dithiothreitol (DTT), 0.16 mg/mL BSA, 0.4 mM each of GTP, CTP and ATP, 0.01 mM UTP, 50 μCi [α-32P]UTP, and 100 units of T7 RNA polymerase.
Cells were centrifuged (2 minutes at 21,000 × g), resuspended in SDS load mix (2% [w/v] SDS, 10% [v/v] glycerol, 50 mM Tris-HCl pH 6.8, 0.25% [w/v] bromophenol blue, 0.8 M β-mercaptoethanol) and heated at 95°C for 5 minutes. To normalize for differences in growth between the various samples, the OD600 of each sample was measured and the volume spun normalized to give an equivalent to OD600 approximately equal to 0.35. The resulting lysates were subjected to SDS-PAGE on a 12% polyacrylamide gel, proteins transferred to PVDF (Roche, Indianapolis, IN, USA) and Crp was detected by Western blot with a polyclonal rabbit anti-Crp antibody (kind gift of H Aiba). The primary antibody was diluted 1:20,000 in TBST; the secondary antibody (anti-rabbit IgG-horseradish peroxidase (HRP) conjugate; Promega, Madison, WI, USA) was used at 1:5,000. Crp was visualized with a Pierce ECL 2 Western blotting substrate (Thermo Scientific, Rockford, IL, USA) and PhosphorImager (GE Healthcare). The membranes were stripped and GroES detected (rabbit anti-GroES antibody from Sigma-Aldrich (St Louis, MO, USA) at 1:10,000) for use as an internal standard; GroES is not sensitive to hfq status . Bands were quantified using ImageQuant software (GE Healthcare) and Crp levels plotted relative to GroES.
bovine serum albumin
cyclic AMP-receptor protein
DNA adenine methylase
polyacrylamide gel electrophoresis
polymerase chain reaction
reverse transcription polymerase chain reaction
sodium dodecyl sulfate
Tris borate-EDTA buffer
20 mM Tris-HCl (pH 7.5), 150 mM sodium chloride, 0.5% Tween-20
translation initiation region
Roberts D, Hoopes BC, McClure WR, Kleckner N: IS10 transposition is regulated by DNA adenine methylation. Cell 1985,43(1):117-130. 10.1016/0092-8674(85)90017-0
Yin JC, Krebs MP, Reznikoff WS: Effect of dam methylation on Tn5 transposition. J Mol Biol 1988,199(1):35-45. 10.1016/0022-2836(88)90377-4
Raleigh EA, Kleckner N: Quantitation of insertion sequence IS10 transposase gene expression by a method generally applicable to any rarely expressed gene. Proc Natl Acad Sci U S A 1986,83(6):1787-1791. 10.1073/pnas.83.6.1787
Krebs MP, Reznikoff WS: Transcriptional and translational initiation sites of IS50. Control of transposase and inhibitor expression. J Mol Biol 1986,192(4):781-791. 10.1016/0022-2836(86)90028-8
Simons RW, Kleckner N: Translational control of IS10 transposition. Cell 1983,34(2):683-691. 10.1016/0092-8674(83)90401-4
Ma C, Simons RW: The IS10 antisense RNA blocks ribosome binding at the transposase translation initiation site. EMBO J 1990,9(4):1267-1274.
Ross JA, Wardle SJ, Haniford DB: Tn10/IS10 transposition is downregulated at the level of transposase expression by the RNA-binding protein Hfq. Mol Microbiol 2010,78(3):607-621. 10.1111/j.1365-2958.2010.07359.x
Ross JA, Ellis MJ, Hossain S, Haniford DB: Hfq restructures RNA-IN and RNA-OUT and facilitates antisense pairing in the Tn10/IS10 system. RNA 2013,19(5):670-684. 10.1261/rna.037747.112
Vogel J, Luisi BF: Hfq and its constellation of RNA. Nat Rev Microbiol 2011,9(8):578-589. 10.1038/nrmicro2615
Le Derout J, Boni IV, Regnier P, Hajnsdorf E: Hfq affects mRNA levels independently of degradation. BMC Mol Biol 2010, 11: 17. 10.1186/1471-2199-11-17
Mahillon J, Chandler M: Insertion sequences. Microbiol Mol Biol Rev 1998,62(3):725-774.
Kennedy AK, Guhathakurta A, Kleckner N, Haniford DB: Tn10 transposition via a DNA hairpin intermediate. Cell 1998,95(1):125-134. 10.1016/S0092-8674(00)81788-2
Bhasin A, Goryshin IY, Reznikoff WS: Hairpin formation in Tn5 transposition. J Biol Chem 1999,274(52):37021-37029. 10.1074/jbc.274.52.37021
Mahnke Braam LA, Goryshin IY, Reznikoff WS: A mechanism for Tn5 inhibition. carboxyl-terminal dimerization. J Biol Chem 1999,274(1):86-92. 10.1074/jbc.274.1.86
Reznikoff WS: Transposon Tn5. Annu Rev Genet 2008, 42: 269-286. 10.1146/annurev.genet.42.110807.091656
Kuan CT, Tessman I: LexA protein of Escherichia coli represses expression of the Tn5 transposase gene. J Bacteriol 1991,173(20):6406-6410.
Mikulecky PJ, Kaw MK, Brescia CC, Takach JC, Sledjeski DD, Feig AL: Escherichia coli Hfq has distinct interaction surfaces for DsrA, rpoS and poly(A) RNAs. Nat Struct Mol Biol 2004,11(12):1206-1214. 10.1038/nsmb858
Martinez-Antonio A, Collado-Vides J: Identifying global regulators in transcriptional regulatory networks in bacteria. Curr Opin Microbiol 2003,6(5):482-489. 10.1016/j.mib.2003.09.002
Guisbert E, Rhodius VA, Ahuja N, Witkin E, Gross CA: Hfq modulates the sigmaE-mediated envelope stress response and the sigma32-mediated cytoplasmic stress response in Escherichia coli . J Bacteriol 2007,189(5):1963-1973. 10.1128/JB.01243-06
Lathem WW, Schroeder JA, Bellows LE, Ritzert JT, Koo JT, Price PA, Caulfield AJ, Goldman WE: Posttranscriptional regulation of the Yersinia pestis cyclic AMP receptor protein Crp and impact on virulence. mBio 2014,5(1):e01038-01013. 10.1128/mBio.01038-13
Hussein R, Lim HN: Disruption of small RNA signaling caused by competition for Hfq. Proc Natl Acad Sci U S A 2011,108(3):1110-1115. 10.1073/pnas.1010082108
Moon K, Gottesman S: Competition among Hfq-binding small RNAs in Escherichia coli . Mol Microbiol 2011,82(6):1545-1562. 10.1111/j.1365-2958.2011.07907.x
Wassarman KM, Repoila F, Rosenow C, Storz G, Gottesman S: Identification of novel small RNAs using comparative genomics and microarrays. Genes Dev 2001,15(13):1637-1651. 10.1101/gad.901001
Ishikawa H, Otaka H, Maki K, Morita T, Aiba H: The functional Hfq-binding module of bacterial sRNAs consists of a double or single hairpin preceded by a U-rich sequence and followed by a 3′ poly(U) tail. RNA (New York, NY) 2012,18(5):1062-1074. 10.1261/rna.031575.111
Fender A, Elf J, Hampel K, Zimmermann B, Wagner EG: RNAs actively cycle on the Sm-like protein Hfq. Genes Dev 2010,24(23):2621-2626. 10.1101/gad.591310
Papenfort K, Vogel J: Small RNA functions in carbon metabolism and virulence of enteric pathogens. Front Cell Infect Microbiol 2014, 4: 91. 10.3389/fcimb.2014.00091
Agre P, Bonhivers M, Borgnia MJ: The aquaporins, blueprints for cellular plumbing systems. J Biol Chem 1998,273(24):14659-14662. 10.1074/jbc.273.24.14659
Rice JB, Vanderpool CK: The small RNA SgrS controls sugar-phosphate accumulation by regulating multiple PTS genes. Nucleic Acids Res 2011,39(9):3806-3819. 10.1093/nar/gkq1219
Hu ST, Wang HC, Lei GS, Wang SH: Negative regulation of IS2 transposition by the cyclic AMP (cAMP)-cAMP receptor protein complex. J Bacteriol 1998,180(10):2682-2688.
Grainger DC, Hurd D, Harrison M, Holdstock J, Busby SJ: Studies of the distribution of Escherichia coli cAMP-receptor protein and RNA polymerase along the E. coli chromosome. Proc Natl Acad Sci U S A 2005,102(49):17693-17698. 10.1073/pnas.0506687102
Johansen J, Eriksen M, Kallipolitis B, Valentin-Hansen P: Down-regulation of outer membrane proteins by noncoding RNAs: unraveling the cAMP-CRP- and sigmaE-dependent CyaR-ompX regulatory case. J Mol Biol 2008,383(1):1-9. 10.1016/j.jmb.2008.06.058
Frohlich KS, Papenfort K, Berger AA, Vogel J: A conserved RpoS-dependent small RNA controls the synthesis of major porin OmpD. Nucleic Acids Res 2012,40(8):3623-3640. 10.1093/nar/gkr1156
Chen S, Zhang A, Blyn LB, Storz G: MicC, a second small-RNA regulator of Omp protein expression in Escherichia coli . J Bacteriol 2004,186(20):6689-6697. 10.1128/JB.186.20.6689-6697.2004
Twiss E, Coros AM, Tavakoli NP, Derbyshire KM: Transposition is modulated by a diverse set of host factors in Escherichia coli and is stimulated by nutritional stress. Mol Microbiol 2005,57(6):1593-1607. 10.1111/j.1365-2958.2005.04794.x
Wang RF, Kushner SR: Construction of versatile low-copy number vectors for cloning, sequencing and gene expression in Escherichia coli . Gene 1991, 100: 195-199. 10.1016/0378-1119(91)90366-J
Bolivar F, Backman K: Plasmids of Escherichia coli as cloning vectors. Methods Enzymol 1979, 68: 245-267. 10.1016/0076-6879(79)68018-7
Cuzon G, Naas T, Nordmann P: Functional characterization of Tn4401, a Tn3-based transposon involved in blaKPC gene mobilization. Antimicrob Agents Chemother 2011,55(11):5370-5373. 10.1128/AAC.05202-11
Hoffman EP, Wilhelm RC: Genetic mapping and dominance of the amber suppressor, Su1 (supD), in Escherichia coli K-12. J Bacteriol 1970,103(1):32-36.
Casadaban MJ: Transposition and fusion of the lac genes to selected promoters in Escherichia coli using bacteriophage lambda and Mu. J Mol Biol 1976,104(3):541-555. 10.1016/0022-2836(76)90119-4
Foster TJ, Davis MA, Roberts DE, Takeshita K, Kleckner N: Genetic organization of transposon Tn10. Cell 1981,23(1):201-213. 10.1016/0092-8674(81)90285-3
Morita T, Kawamoto H, Mizota T, Inada T, Aiba H: Enolase in the RNA degradosome plays a crucial role in the rapid decay of glucose transporter mRNA in the response to phosphosugar stress in Escherichia coli . Mol Microbiol 2004,54(4):1063-1075. 10.1111/j.1365-2958.2004.04329.x
Morita T, Maki K, Aiba H: RNase E-based ribonucleoprotein complexes: mechanical basis of mRNA destabilization mediated by bacterial noncoding RNAs. Genes Dev 2005,19(18):2176-2186. 10.1101/gad.1330405
Whitfield CR, Wardle SJ, Haniford DB: The global bacterial regulator H-NS promotes transpososome formation and transposition in the Tn5 system. Nucleic Acids Res 2009,37(2):309-321. 10.1093/nar/gkn935
Way JC, Davis MA, Morisato D, Roberts DE, Kleckner N: New Tn10 derivatives for transposon mutagenesis and for construction of lacZ operon fusions by transposition. Gene 1984,32(3):369-379. 10.1016/0378-1119(84)90012-X
Guillier M, Gottesman S: Remodelling of the Escherichia coli outer membrane by two small regulatory RNAs. Mol Microbiol 2006,59(1):231-247. 10.1111/j.1365-2958.2005.04929.x
Vanderpool CK, Gottesman S: Involvement of a novel transcriptional activator and small RNA in post-transcriptional regulation of the glucose phosphoenolpyruvate phosphotransferase system. Mol Microbiol 2004,54(4):1076-1089. 10.1111/j.1365-2958.2004.04348.x
Wadler CS, Vanderpool CK: Characterization of homologs of the small RNA SgrS reveals diversity in function. Nucleic Acids Res 2009,37(16):5477-5485. 10.1093/nar/gkp591
Mandin P, Gottesman S: Integrating anaerobic/aerobic sensing and the general stress response through the ArcZ small RNA. EMBO J 2010,29(18):3094-3107. 10.1038/emboj.2010.179
Haniford DB, Chelouche AR, Kleckner N: A specific class of IS10 transposase mutants are blocked for target site interactions and promote formation of an excised transposon fragment. Cell 1989,59(2):385-394. 10.1016/0092-8674(89)90299-7
Garrey SM, Mackie GA: Roles of the 5′-phosphate sensor domain in RNase E. Mol Microbiol 2011,80(6):1613-1624. 10.1111/j.1365-2958.2011.07670.x
Wilkinson KA, Merino EJ, Weeks KM: Selective 2′-hydroxyl acylation analyzed by primer extension (SHAPE): quantitative RNA structure analysis at single nucleotide resolution. Nat Protoc 2006,1(3):1610-1616. 10.1038/nprot.2006.249
We thank Michael Ellis for providing comments on the manuscript and for useful discussions, and Claire Young for assistance with P1 transductions. We also thank S Gottesman and C Vanderpool for providing pLlacO-sRNA expression plasmids and the sgrS1 allele, respectively, and H Aiba for providing an anti-Crp antibody. Finally, we thank W Reznikoff for providing IS50 transposase plasmids and T Naas for providing an E. coli strain harboring pOX38-Gen. This work was supported by a grant to DBH (MOP 11281) from Canadian Institutes of Health Research. JR and CM were supported by OGS and NSERC scholarships.
The authors declare that they have no competing interest.
JR performed ‘mating out’ assays, β-galactosidase assays in Figures 3, 7 and 8, RT-PCR assays, Western blot assays, participated in the design of the study and helped draft the manuscript. RT performed the β-galactosidase assay in Figure 4, steady-state transcript measurements and RNA half-life measurements and prepared figures. MDB helped construct chromosomal reporter strains. CRM made the initial discovery that Tn5 transposition is up-regulated under conditions of hfq deficiency and constructed IS50-reporter plasmids and expression plasmids. DBH participated in the design of the study, helped in constructing strains and plasmids, performed the ‘mating out’ experiment in Figure 7D and drafted the manuscript. All authors read and approved the final manuscript.
Electronic supplementary material
Authors’ original submitted files for images
Below are the links to the authors’ original submitted files for images.
About this article
Cite this article
Ross, J.A., Trussler, R.S., Black, M.D. et al. Tn5 transposition in Escherichia coli is repressed by Hfq and activated by over-expression of the small non-coding RNA SgrS. Mobile DNA 5, 27 (2014) doi:10.1186/s13100-014-0027-z
- DNA transposition