Two enhancer binding proteins activate σ54-dependent transcription of a quorum regulatory RNA in a bacterial symbiont
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
The manuscript has the potential to transform the field of sensory transduction and gene regulation in the Vibrio genus by uncovering a previously undescribed enhancer binding protein and its role in the regulation of quorum sensing and physiology in the Vibrio - squid symbiosis. However, in its present form, several experiments are required to support the claims of the manuscript.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Microbiology and Infectious Disease (eLife)
Abstract
To colonize a host, bacteria depend on an ensemble of signaling systems to convert information about the various environments encountered within the host into specific cellular activities. How these signaling systems coordinate transitions between cellular states in vivo remains poorly understood. To address this knowledge gap, we investigated how the bacterial symbiont Vibrio fischeri initially colonizes the light organ of the Hawaiian bobtail squid Euprymna scolopes . Previous work has shown that the small RNA Qrr1, which is a regulatory component of the quorum-sensing system in V. fischeri , promotes host colonization. Here, we report that transcriptional activation of Qrr1 is inhibited by the sensor kinase BinK, which suppresses cellular aggregation by V. fischeri prior to light organ entry. We show that Qrr1 expression depends on the alternative sigma factor σ 54 and the transcription factors LuxO and SypG, which function similar to an OR logic gate, thereby ensuring Qrr1 is expressed during colonization. Finally, we provide evidence that this regulatory mechanism is widespread throughout the Vibrionaceae family. Together, our work reveals how coordination between the signaling pathways underlying aggregation and quorum-sensing promotes host colonization, which provides insight into how integration among signaling systems facilitates complex processes in bacteria.
Article activity feed
-
-
Author Response
Reviewer #1 (Public Review):
The authors present data identifying the role of the bacterial enhancer binding protein (bEBP) SypG in the regulation of the Qrr1 small RNA, which is known to be a key regulator of Vibrio fischeri bioluminescence production and squid colonization. Previously, only the bEBP LuxO was known to activate Qrr1 expression. LuxO and Qrr1 are conserved in the Vibrionaceae, and the authors show that SypG is conserved in ~half of the Vibrio family, suggesting that this Qrr1 regulatory OR gate controlled by LuxO or SypG may play important roles in physiology processes in other species.
Successful squid colonization by Vibrio fischeri is a complex process, known to be influenced by several factors, including the formation of and dispersal from cellular aggregates prior to entering squid pores, and …
Author Response
Reviewer #1 (Public Review):
The authors present data identifying the role of the bacterial enhancer binding protein (bEBP) SypG in the regulation of the Qrr1 small RNA, which is known to be a key regulator of Vibrio fischeri bioluminescence production and squid colonization. Previously, only the bEBP LuxO was known to activate Qrr1 expression. LuxO and Qrr1 are conserved in the Vibrionaceae, and the authors show that SypG is conserved in ~half of the Vibrio family, suggesting that this Qrr1 regulatory OR gate controlled by LuxO or SypG may play important roles in physiology processes in other species.
Successful squid colonization by Vibrio fischeri is a complex process, known to be influenced by several factors, including the formation of and dispersal from cellular aggregates prior to entering squid pores, and inoculation of the light organ crypts, and biofilm formation within the crypts. Previously, it was shown that strains lacking qrr1 were at a deficit for crypt colonization in the presence of wild-type V. fischeri. Conversely, cells lacking binK, which encodes a hybrid histidine kinase, were at an advantage for crypt colonization in the presence of wild-type cells. However, the authors identified BinK as a negative regulator of Qrr1 expression in a transposon screen. The authors used genetic epistasis experiments and found that Qrr1 transcription can be activated by either phosphorylated LuxO at low cell densities (in the absence of quorum sensing signals) or by SypG, presumably by binding to the two upstream activation sequences in the promoter of qrr1 to activate transcription by the required alternative sigma factor sigma-54. The competition between these bEBPs has not been tested. The model proposed is an OR gate through which quorum sensing and aggregation signals control Qrr1. However, there are several untested aspects of this model. First, the role of phosphorylation in SypG activity, and the connection to BinK, are not addressed in this manuscript, which may confound the observed effects observed on qrr1 transcription. Further, the authors did not test whether SypG directly binds to the qrr1 promoter, nor did they assess the individual role of LuxO binding to the two LuxO binding sites in the absence of SypG. The study is lacking an in vivo assessment of SypG and LuxO binding/competition at the Qrr1 promoter based on the authors' model of the OR gate.
Major comments:
• What is known about the connection between BinK and SypG? BinK is a hybrid HK (intro states this). Does BinK phosphorylate/dephosphorylate SypG - directly or indirectly? I saw a published paper (Ludvik et al 2021) with a diagram suggesting BinK does inhibit SypG, but the connection is unclear. This diagram also suggested that SypG needs to be phosphorylated. Can the authors comment - does SypG need to be phosphorylated to be active? Because SypG has the same sequence as the LuxO linker (Fig. S2), then I presume that SypG would also need to be phosphorylated to be active (like LuxO)? The authors utilize a phosphomimic of LuxO to test function under constitutive activity (Fig. S3), but they do not use a phosphomimic of SypG (Fig 4). If the authors used a constitutive allele, would those assays reveal more about the competition between SypG and LuxO, in the presence of phosphorylated LuxO at low cell density? The authors should include a putative cartoon model for how BinK HK activity connects to SypG, based on what is already in the literature, to aid the reader.
We have added information & corresponding cartoon model in the results section about the signaling pathway involving BinK and SypG, including that SypG must be phosphorylated to be active and that BinK acts as a phosphatase towards SypG. We have also generated a SypGD53E mutant and found increased Pqrr1 activity, which suggests that phosphorylation of SypG has a major impact on SypG-dependent activation of Pqrr1.
• Line 246: Figure S3: nucleotide substitutions in both UAS regions showed loss of Pqrr1-gfp, but this could be due to binding/activation by SypG or LuxO. This should be tested in a sypG- strain to determine the sole effect of LuxO binding to these two UASs. In Figures 4G and 7, the luxO- sypG- Ptrc-sypG strain backgrounds allow the independent analysis of the two bEBPs. It is important to test which of these two sites is critical for LuxO-dependent activation of Pqrr1, given the conservation of the LuxO-Qrr1 region in other Vibrios (line 327, Fig. S5). Thus, the authors could also discuss whether these two proteins would compete at both sites. Further, the authors should comment that they have not shown biochemical evidence that SypG binds to the two UASs in the Qrr1 promoter. The regulation of this locus by SypG is only shown by genetic assays in this manuscript.
We have added a paragraph in the discussion highlighting how useful protein-DNA assays would be to address competition along with the barriers encountered with approaches to purify SypG. Regarding the contribution of each UAS to LuxO-dependent activation, we refer to the phosphomimic data of LuxO (Fig. S4) in the supplement that highlight G-131 and G-97 do not affect LuxO-dependent activation (as pointed out by reviewer #2), which has contributed to our test of a G-131T mutant in the co-colonization experiment.
• Examination of the binding of LuxO and SypG (e.g., ChIP-seq) in combination with their transcriptional reporter under varying conditions (low cell density vs high cell density, with or without rscS* overexpression) would be extremely beneficial in testing the model proposed.
We agree but have not had success in our attempts to perform ChIP due to protein instability. For example, we have tried SypG with a C-terminal TAP tag, which my colleague Dr. Lu Bai at Penn State has used extensively for ChIP, ChIP-seq, and ChIP-exo, but we could not observe a signal even when RscS* allele was included in the strain.
Reviewer #2 (Public Review):
The study by Surrett et al. uncovers a novel regulatory axis in Vibrio fischeri that controls the expression of the qrr1 small RNA, which post-transcriptionally controls various quorum-dependent outputs. This study is timely and addresses a major question about the physiology of this important model symbiosis and potentially other Vibrio species. The results should be of broad interest within the field of microbiology.
While it was previously believed that qrr1 expression is under the strict control of the LuxO-dependent quorum sensing cascade, the authors demonstrate that qrr1 expression can be induced by another bEBP, SypG, in a manner that is quorum-independent. It was previously shown that qrr1 is important for colonization, and the authors recapitulate and extend this finding here. However, bacteria are likely at high cell density prior to entry into the crypts, which would repress qrr1 expression. Thus, despite the importance of qrr1 expression for crypt colonization, it would counterintuitively be repressed. The discovery of the SypG quorum-independent induction of qrr1 in this study may help resolve this conundrum. The authors take a largely genetic approach to characterize this novel regulatory pathway in combination with a squid colonization model. The experiments performed are generally well controlled and the data are clearly presented. The authors, however, fail to provide experimental evidence to support the physiological relevance of SypG-dependent control of qrr1 expression during host colonization.
Fig. 2 - It is unclear why there is a disconnect between qrr1 expression and qrr1-dependent effects. Data in 2B, indicate that qrr1 is induced in the ∆binK mutant according to the Pqrr1-gfp reporter but this expressed qrr1 does not have any effect on phenotypes like bioluminescence according to the data presented in 2C. While the authors reveal an effect of the binK deletion when rscS is overexpressed, it is unclear why this is necessary since simple deletion of bink without rscS* is sufficient to induce qrr1 in 2B. Could this discrepancy be due to the fact that experiments in 2B are done on solid media while the experiments in 2C are performed in liquid media? Do cells in liquid not express qrr1? Or conversely, perhaps testing the bioluminescence of cells scraped off of plates could reveal a phenotype for the binK mutant similar to those seen in the rscS* background in liquid. Or alternatively, if cells in a liquid culture still express qrr1, perhaps there is a posttranscriptional mechanism that prevents qrr1 from exerting an effect on bioluminescence? The latter possibility would alter the proposed model.
To help explain why we chose to overexpress RscS, we have added the cartoon in Fig. 2C, which highlights how BinK dephosphorylates SypG. We believe that the conditions used in the bioluminescence assay do not phosphorylate SypG, which prevents an effect by BinK. However, overexpression of RscS permits phosphorylation of SypG, which enables a phenotype to emerge in a binK mutant. We have tested the bioluminescence of cells within spots but did not detect a difference.
The authors propose a model in which sypG dependent activation of qrr1 is required for appropriate temporal regulation of this small RNA and contributes to optimal fitness of V. fischeri during colonization, however, this was not directly tested, and experimental evidence to support a physiological role for spyG-dependent regulation of qrr1 remains lacking. Data in Fig. S3 and Fig. 4G-H suggest that the Gs at -131 and -97 in Pqrr1 are largely dispensable for LuxO-dependent activation, but are important for SypG-dependent activation of Pqrr1. Also, the Pqrr1 mutations at C -130 and -96 completely prevent sypG-dependent activation while only partially reducing LuxO-dependent activation. If SypG-dependent activation of qrr1 is critical for the fitness of V. fischeri, a strain harboring these Pqrr1 promoter mutations should be attenuated in a manner that resembles the qrr1 deletion mutant as shown in Fig. 3C.
We thank the reviewer for this suggestion, which led us to generate and test a G-131T mutant in vivo.
Fig. S4 - these data suggest that LuxO cannot enhance transcription of PsypA and PsypP at native expression levels. But sypG-dependent induction of qrr1 was largely tested with Ptrc-dependent overexpression of SypG. Would overexpression of LuxO induce PsypA and PsypP? The authors should at least acknowledge this possibility in the text.
As requested, we have added text that acknowledges this possibility.
The authors adopt three distinct strategies to induce sypG-dependent activation of qrr1 in distinct figures throughout the manuscript: deletion of binK, overexpression of rscS (rscS*), and direct overexpression of sypG. It is not entirely clear why distinct approaches are used in different figures. This is particularly true for Fig. 5 since the authors already demonstrated that the direct overexpression of sypG can be used, which is a more direct way of addressing this question. Similarly, sypG overexpression should inhibit bioluminescence in Fig. 2 based on the proposed model, which would have tested the claims made more directly. Additional text to clarify this would be helpful.
As requested, we have added Fig. 2C and text to describe how SypG is regulated, which provides ways to test SypG-dependent activation of qrr1.
The Fig. 5D legend indicates that the strains harbor a Ptrc-GFP reporter. However, the text would suggest that these strains should harbor a Pqrr1-GFP reporter to test the question posed.
This has been corrected.
The conclusion that SypG and LuxO share UASs in the qrr1 promoter is based on fairly limited genetic evidence where point mutations were introduced into 3 bp of the predicted LuxO UASs within the qrr1 promoter. This conclusion needs to be qualified in the text or additional experimental evidence is needed to support this claim. For example, in vivo ChIP-exo could be used to map the SypG and LuxO binding sites. Or SypG and LuxO could be purified to assess binding to the qrr promoter in vitro (to map binding sites or test competitive interactions of these proteins to the qrr promoter).
As described above and in the text, we have not been able to construct a functional tagged SypG that would enable these types of studies.
On a related note, SypG binding to the qrr1 promoter is speculated based on indirect genetic evidence. But the authors do not directly demonstrate this. This should be acknowledged in the text or additional experimental evidence should be provided to support this claim.
As requested, we have added text in the discussion that highlights this problem.
Reviewer #3 (Public Review):
In this manuscript, Surrett and coworkers aimed to identify the mechanism that regulates the transcription of Qrr1 sRNA in the squid symbiont Vibrio fischeri. In many Vibrio species, Qrr1 transcription is regulated by quorum sensing (QS) and activated only at low cell density. Qrr1 is important for V. fischeri to colonize the squid host. In the QS systems that have been studied so far, LuxO is the only known response regulator that activates Qrr sRNA transcription. However, the authors argued that since V. fischeri forms aggregates before entering into the light organ of the squid, Qrr1 would not be made as high cell density QS state is likely induced within the aggregates. Therefore, they hypothesized that additional regulatory systems must exist to allow Qrr1 expression in V. fischeri to initiate colonization of the light organ. In turn, the authors identified that disruption of the function of the sensor kinase BinK allowed Qrr1 expression even at high cell density. Through a series of cell-based reporter assays and an in vivo squid colonization assay, they concluded that BinK is also involved in Qrr1 regulation within the squid light organ. They went on to show that another sigma54-dependent response regulator SypG is also involved in controlling Qrr1 expression. The authors propose dual regulation of LuxO and SypG on Qrr could be a common regulatory mechanism on Qrr expression in a subset of Vibiro species.
Overall, the experiments were carefully performed and the findings that BinK and SypG are involved in Qrr1 regulation are interesting. This paper is of potential interest to an audience in the field of QS and Vibrio-host interaction. However, experimental deficiencies and alternative explanations of the results have been identified in the manuscript that prevents a thorough mechanistic understanding of the interplay between QS and these new regulators.
- The premise that Qrr1 expression in the light organ has to be regulated by systems other than QS is unclear. In lines 108-109, it was stated that "...prior to entering the light organ, bacterial cells are collected from the environment and form aggregates that are densely packed", however, in lines 184-185, it was stated that "The majority of crypt spaces each contained only one strain type (Fig. 3B), which is consistent with most populations arising from only 1-2 cells that enter the corresponding crypt spaces". So, if the latter case is true (i.e., 1-2 cells/crypt), why Qrr1 could not be made at that time point as predicted by a QS regulation model?
We have not changed this section because if Qrr1 is expressed only after the cells have already entered the crypt space, then the Δqrr1 mutant would colonize a number of crypt spaces comparable to that of wild type cells.
- The involvement of the rscS* allele for the ∆binK mutant to show an altered bioluminescence phenotype is confusing. It is unclear why a WT genetic background was sufficient to show the high Qrr1 phenotype in the original genetic screen that identified BinK (Fig. 2A-B), while the rcsS* allele is now required for the rest of the experiments to show the involvement of BinK in bioluminescence regulation (Fig 2C). Is the decreased bioluminescence phenotype observed in rcsS* ∆binK mutant (fig. 2C) dependent on LuxU/LuxO/Qrr1/LitR? Could it be through another indirect mechanism (e.g., SypK as discussed in line 403)? A better explanation of the connection between RcsS/Syp and BinK and perhaps additional mutant characterization are necessary to interpret the observed phenotypes.
As described above, we have added a cartoon that illustrates the pathway involving BinK (Fig. 2C) and additional justification in the results section, which better explains why RscS overexpression was used.
- In squid colonization competition assays (Fig. 3), it was concluded that the ∆qrr1 allele is epistatic to the ∆binK allele (line 204), and the enhanced colonization of the ∆binK mutant is dependent on Qrr1 (section title, line 162). This conclusion is hard to interpret. The results can be interpreted as ∆qrr1 mutation lowers the colonization efficiency of the ∆binK mutant which could imply BinK regulates Qrr1 in vivo. Alternatively, it could be interpreted that the ∆binK mutation increases the colonization efficiency of the ∆qrr1 mutant. Direct competition between single and double mutants in the same animals may resolve the complexity. And direct comparison of Qrr1 expression of WT and ∆binK mutants inside the animals, if possible, will also help interpret these results.
We thank the reviewer for the suggestion and were able to test the ΔbinK and ΔbinK Δqrr1 mutants directly (Fig. S2). We were unable to interpret the data using the Pqrr1 reporter due to unexpected heterogeneity in Pqrr1 activity throughout the crypt spaces.
- Similar concern to above (#2), in Fig. 4, the link between BinK and Qrr1 regulation is not fully explored. What connects BinK and Qrr1 expression? Does BinK function via LuxU (or other HPT) to control SypG like the other QS kinases? And what is the role of other known kinases (e.g., SypF) in the signaling pathway? And did the authors test other bEBPs found in V. fischeri for their role in Qrr1 regulation?
We have added to the discussion content that highlights examining LuxU as a direction worthwhile to pursue to understand how BinK affects signaling that activates Qrr1.
- In addition to the genetic analysis, additional characterization of SypG is required to demonstrate the proposed regulatory mechanism: What is the expression level (and phosphorylation state) of SypG and LuxO at different cell densities? Does purified SypG directly bind to the qrr1 promoter region? c. How do these two bEBPs compete with each other if they are both made and active?
We agree that these are interesting questions, but as described above, we were unable to purify SypG to address the biochemistry.
- The molecular OR logic gate is used to describe the relationship between LuxO and SypG, but this logic relationship is not always true in all conditions (if at all). In WT, deletion of luxO completely abolished Qrr1 expression (Fig. 4C). Even in the binK mutant, LuxO still seems to be the more prominent regulator (Fig. 4D) as deletion of luxO already caused a smaller but significant drop in Qrr1 expression. The authors may need to use this term more precisely.
We note that in wild-type cells, SypG is not active under the conditions tested, so SypG would not contribute to activating Qrr1 expression. The level of Pqrr1 activity by the SypG(D53E) variant surpasses the basal level of LuxO, which suggests that LuxO does not always serve as the prominent regulator. We have added content to the discussion to highlight how LuxO may contribute more to the regulation.
-
Evaluation Summary:
The manuscript has the potential to transform the field of sensory transduction and gene regulation in the Vibrio genus by uncovering a previously undescribed enhancer binding protein and its role in the regulation of quorum sensing and physiology in the Vibrio - squid symbiosis. However, in its present form, several experiments are required to support the claims of the manuscript.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. The reviewers remained anonymous to the authors.)
-
Reviewer #1 (Public Review):
The authors present data identifying the role of the bacterial enhancer binding protein (bEBP) SypG in the regulation of the Qrr1 small RNA, which is known to be a key regulator of Vibrio fischeri bioluminescence production and squid colonization. Previously, only the bEBP LuxO was known to activate Qrr1 expression. LuxO and Qrr1 are conserved in the Vibrionaceae, and the authors show that SypG is conserved in ~half of the Vibrio family, suggesting that this Qrr1 regulatory OR gate controlled by LuxO or SypG may play important roles in physiology processes in other species.
Successful squid colonization by Vibrio fischeri is a complex process, known to be influenced by several factors, including the formation of and dispersal from cellular aggregates prior to entering squid pores, and inoculation of the …
Reviewer #1 (Public Review):
The authors present data identifying the role of the bacterial enhancer binding protein (bEBP) SypG in the regulation of the Qrr1 small RNA, which is known to be a key regulator of Vibrio fischeri bioluminescence production and squid colonization. Previously, only the bEBP LuxO was known to activate Qrr1 expression. LuxO and Qrr1 are conserved in the Vibrionaceae, and the authors show that SypG is conserved in ~half of the Vibrio family, suggesting that this Qrr1 regulatory OR gate controlled by LuxO or SypG may play important roles in physiology processes in other species.
Successful squid colonization by Vibrio fischeri is a complex process, known to be influenced by several factors, including the formation of and dispersal from cellular aggregates prior to entering squid pores, and inoculation of the light organ crypts, and biofilm formation within the crypts. Previously, it was shown that strains lacking qrr1 were at a deficit for crypt colonization in the presence of wild-type V. fischeri. Conversely, cells lacking binK, which encodes a hybrid histidine kinase, were at an advantage for crypt colonization in the presence of wild-type cells. However, the authors identified BinK as a negative regulator of Qrr1 expression in a transposon screen. The authors used genetic epistasis experiments and found that Qrr1 transcription can be activated by either phosphorylated LuxO at low cell densities (in the absence of quorum sensing signals) or by SypG, presumably by binding to the two upstream activation sequences in the promoter of qrr1 to activate transcription by the required alternative sigma factor sigma-54. The competition between these bEBPs has not been tested. The model proposed is an OR gate through which quorum sensing and aggregation signals control Qrr1. However, there are several untested aspects of this model. First, the role of phosphorylation in SypG activity, and the connection to BinK, are not addressed in this manuscript, which may confound the observed effects observed on qrr1 transcription. Further, the authors did not test whether SypG directly binds to the qrr1 promoter, nor did they assess the individual role of LuxO binding to the two LuxO binding sites in the absence of SypG. The study is lacking an in vivo assessment of SypG and LuxO binding/competition at the Qrr1 promoter based on the authors' model of the OR gate.
Major comments:
• What is known about the connection between BinK and SypG? BinK is a hybrid HK (intro states this). Does BinK phosphorylate/dephosphorylate SypG - directly or indirectly? I saw a published paper (Ludvik et al 2021) with a diagram suggesting BinK does inhibit SypG, but the connection is unclear. This diagram also suggested that SypG needs to be phosphorylated. Can the authors comment - does SypG need to be phosphorylated to be active? Because SypG has the same sequence as the LuxO linker (Fig. S2), then I presume that SypG would also need to be phosphorylated to be active (like LuxO)? The authors utilize a phosphomimic of LuxO to test function under constitutive activity (Fig. S3), but they do not use a phosphomimic of SypG (Fig 4). If the authors used a constitutive allele, would those assays reveal more about the competition between SypG and LuxO, in the presence of phosphorylated LuxO at low cell density? The authors should include a putative cartoon model for how BinK HK activity connects to SypG, based on what is already in the literature, to aid the reader.
• Line 246: Figure S3: nucleotide substitutions in both UAS regions showed loss of Pqrr1-gfp, but this could be due to binding/activation by SypG or LuxO. This should be tested in a sypG- strain to determine the sole effect of LuxO binding to these two UASs. In Figures 4G and 7, the luxO- sypG- Ptrc-sypG strain backgrounds allow the independent analysis of the two bEBPs. It is important to test which of these two sites is critical for LuxO-dependent activation of Pqrr1, given the conservation of the LuxO-Qrr1 region in other Vibrios (line 327, Fig. S5). Thus, the authors could also discuss whether these two proteins would compete at both sites. Further, the authors should comment that they have not shown biochemical evidence that SypG binds to the two UASs in the Qrr1 promoter. The regulation of this locus by SypG is only shown by genetic assays in this manuscript.
• Examination of the binding of LuxO and SypG (e.g., ChIP-seq) in combination with their transcriptional reporter under varying conditions (low cell density vs high cell density, with or without rscS* overexpression) would be extremely beneficial in testing the model proposed.
-
Reviewer #2 (Public Review):
The study by Surrett et al. uncovers a novel regulatory axis in Vibrio fischeri that controls the expression of the qrr1 small RNA, which post-transcriptionally controls various quorum-dependent outputs. This study is timely and addresses a major question about the physiology of this important model symbiosis and potentially other Vibrio species. The results should be of broad interest within the field of microbiology.
While it was previously believed that qrr1 expression is under the strict control of the LuxO-dependent quorum sensing cascade, the authors demonstrate that qrr1 expression can be induced by another bEBP, SypG, in a manner that is quorum-independent. It was previously shown that qrr1 is important for colonization, and the authors recapitulate and extend this finding here. However, bacteria are …
Reviewer #2 (Public Review):
The study by Surrett et al. uncovers a novel regulatory axis in Vibrio fischeri that controls the expression of the qrr1 small RNA, which post-transcriptionally controls various quorum-dependent outputs. This study is timely and addresses a major question about the physiology of this important model symbiosis and potentially other Vibrio species. The results should be of broad interest within the field of microbiology.
While it was previously believed that qrr1 expression is under the strict control of the LuxO-dependent quorum sensing cascade, the authors demonstrate that qrr1 expression can be induced by another bEBP, SypG, in a manner that is quorum-independent. It was previously shown that qrr1 is important for colonization, and the authors recapitulate and extend this finding here. However, bacteria are likely at high cell density prior to entry into the crypts, which would repress qrr1 expression. Thus, despite the importance of qrr1 expression for crypt colonization, it would counterintuitively be repressed. The discovery of the SypG quorum-independent induction of qrr1 in this study may help resolve this conundrum. The authors take a largely genetic approach to characterize this novel regulatory pathway in combination with a squid colonization model. The experiments performed are generally well controlled and the data are clearly presented. The authors, however, fail to provide experimental evidence to support the physiological relevance of SypG-dependent control of qrr1 expression during host colonization.
Fig. 2 - It is unclear why there is a disconnect between qrr1 expression and qrr1-dependent effects. Data in 2B, indicate that qrr1 is induced in the ∆binK mutant according to the Pqrr1-gfp reporter but this expressed qrr1 does not have any effect on phenotypes like bioluminescence according to the data presented in 2C. While the authors reveal an effect of the binK deletion when rscS is overexpressed, it is unclear why this is necessary since simple deletion of bink without rscS* is sufficient to induce qrr1 in 2B. Could this discrepancy be due to the fact that experiments in 2B are done on solid media while the experiments in 2C are performed in liquid media? Do cells in liquid not express qrr1? Or conversely, perhaps testing the bioluminescence of cells scraped off of plates could reveal a phenotype for the binK mutant similar to those seen in the rscS* background in liquid. Or alternatively, if cells in a liquid culture still express qrr1, perhaps there is a posttranscriptional mechanism that prevents qrr1 from exerting an effect on bioluminescence? The latter possibility would alter the proposed model.
The authors propose a model in which sypG dependent activation of qrr1 is required for appropriate temporal regulation of this small RNA and contributes to optimal fitness of V. fischeri during colonization, however, this was not directly tested, and experimental evidence to support a physiological role for spyG-dependent regulation of qrr1 remains lacking. Data in Fig. S3 and Fig. 4G-H suggest that the Gs at -131 and -97 in Pqrr1 are largely dispensable for LuxO-dependent activation, but are important for SypG-dependent activation of Pqrr1. Also, the Pqrr1 mutations at C -130 and -96 completely prevent sypG-dependent activation while only partially reducing LuxO-dependent activation. If SypG-dependent activation of qrr1 is critical for the fitness of V. fischeri, a strain harboring these Pqrr1 promoter mutations should be attenuated in a manner that resembles the qrr1 deletion mutant as shown in Fig. 3C.
Fig. S4 - these data suggest that LuxO cannot enhance transcription of PsypA and PsypP at native expression levels. But sypG-dependent induction of qrr1 was largely tested with Ptrc-dependent overexpression of SypG. Would overexpression of LuxO induce PsypA and PsypP? The authors should at least acknowledge this possibility in the text.
The authors adopt three distinct strategies to induce sypG-dependent activation of qrr1 in distinct figures throughout the manuscript: deletion of binK, overexpression of rscS (rscS*), and direct overexpression of sypG. It is not entirely clear why distinct approaches are used in different figures. This is particularly true for Fig. 5 since the authors already demonstrated that the direct overexpression of sypG can be used, which is a more direct way of addressing this question. Similarly, sypG overexpression should inhibit bioluminescence in Fig. 2 based on the proposed model, which would have tested the claims made more directly. Additional text to clarify this would be helpful.
The Fig. 5D legend indicates that the strains harbor a Ptrc-GFP reporter. However, the text would suggest that these strains should harbor a Pqrr1-GFP reporter to test the question posed.
The conclusion that SypG and LuxO share UASs in the qrr1 promoter is based on fairly limited genetic evidence where point mutations were introduced into 3 bp of the predicted LuxO UASs within the qrr1 promoter. This conclusion needs to be qualified in the text or additional experimental evidence is needed to support this claim. For example, in vivo ChIP-exo could be used to map the SypG and LuxO binding sites. Or SypG and LuxO could be purified to assess binding to the qrr promoter in vitro (to map binding sites or test competitive interactions of these proteins to the qrr promoter).
On a related note, SypG binding to the qrr1 promoter is speculated based on indirect genetic evidence. But the authors do not directly demonstrate this. This should be acknowledged in the text or additional experimental evidence should be provided to support this claim.
-
Reviewer #3 (Public Review):
In this manuscript, Surrett and coworkers aimed to identify the mechanism that regulates the transcription of Qrr1 sRNA in the squid symbiont Vibrio fischeri. In many Vibrio species, Qrr1 transcription is regulated by quorum sensing (QS) and activated only at low cell density. Qrr1 is important for V. fischeri to colonize the squid host. In the QS systems that have been studied so far, LuxO is the only known response regulator that activates Qrr sRNA transcription. However, the authors argued that since V. fischeri forms aggregates before entering into the light organ of the squid, Qrr1 would not be made as high cell density QS state is likely induced within the aggregates. Therefore, they hypothesized that additional regulatory systems must exist to allow Qrr1 expression in V. fischeri to initiate …
Reviewer #3 (Public Review):
In this manuscript, Surrett and coworkers aimed to identify the mechanism that regulates the transcription of Qrr1 sRNA in the squid symbiont Vibrio fischeri. In many Vibrio species, Qrr1 transcription is regulated by quorum sensing (QS) and activated only at low cell density. Qrr1 is important for V. fischeri to colonize the squid host. In the QS systems that have been studied so far, LuxO is the only known response regulator that activates Qrr sRNA transcription. However, the authors argued that since V. fischeri forms aggregates before entering into the light organ of the squid, Qrr1 would not be made as high cell density QS state is likely induced within the aggregates. Therefore, they hypothesized that additional regulatory systems must exist to allow Qrr1 expression in V. fischeri to initiate colonization of the light organ. In turn, the authors identified that disruption of the function of the sensor kinase BinK allowed Qrr1 expression even at high cell density. Through a series of cell-based reporter assays and an in vivo squid colonization assay, they concluded that BinK is also involved in Qrr1 regulation within the squid light organ. They went on to show that another sigma54-dependent response regulator SypG is also involved in controlling Qrr1 expression. The authors propose dual regulation of LuxO and SypG on Qrr could be a common regulatory mechanism on Qrr expression in a subset of Vibiro species.
Overall, the experiments were carefully performed and the findings that BinK and SypG are involved in Qrr1 regulation are interesting. This paper is of potential interest to an audience in the field of QS and Vibrio-host interaction. However, experimental deficiencies and alternative explanations of the results have been identified in the manuscript that prevents a thorough mechanistic understanding of the interplay between QS and these new regulators.
1. The premise that Qrr1 expression in the light organ has to be regulated by systems other than QS is unclear. In lines 108-109, it was stated that "...prior to entering the light organ, bacterial cells are collected from the environment and form aggregates that are densely packed", however, in lines 184-185, it was stated that "The majority of crypt spaces each contained only one strain type (Fig. 3B), which is consistent with most populations arising from only 1-2 cells that enter the corresponding crypt spaces". So, if the latter case is true (i.e., 1-2 cells/crypt), why Qrr1 could not be made at that time point as predicted by a QS regulation model?
2. The involvement of the rscS* allele for the ∆binK mutant to show an altered bioluminescence phenotype is confusing. It is unclear why a WT genetic background was sufficient to show the high Qrr1 phenotype in the original genetic screen that identified BinK (Fig. 2A-B), while the rcsS* allele is now required for the rest of the experiments to show the involvement of BinK in bioluminescence regulation (Fig 2C). Is the decreased bioluminescence phenotype observed in rcsS* ∆binK mutant (fig. 2C) dependent on LuxU/LuxO/Qrr1/LitR? Could it be through another indirect mechanism (e.g., SypK as discussed in line 403)? A better explanation of the connection between RcsS/Syp and BinK and perhaps additional mutant characterization are necessary to interpret the observed phenotypes.
3. In squid colonization competition assays (Fig. 3), it was concluded that the ∆qrr1 allele is epistatic to the ∆binK allele (line 204), and the enhanced colonization of the ∆binK mutant is dependent on Qrr1 (section title, line 162). This conclusion is hard to interpret. The results can be interpreted as ∆qrr1 mutation lowers the colonization efficiency of the ∆binK mutant which could imply BinK regulates Qrr1 in vivo. Alternatively, it could be interpreted that the ∆binK mutation increases the colonization efficiency of the ∆qrr1 mutant. Direct competition between single and double mutants in the same animals may resolve the complexity. And direct comparison of Qrr1 expression of WT and ∆binK mutants inside the animals, if possible, will also help interpret these results.
4. Similar concern to above (#2), in Fig. 4, the link between BinK and Qrr1 regulation is not fully explored. What connects BinK and Qrr1 expression? Does BinK function via LuxU (or other HPT) to control SypG like the other QS kinases? And what is the role of other known kinases (e.g., SypF) in the signaling pathway? And did the authors test other bEBPs found in V. fischeri for their role in Qrr1 regulation?
5. In addition to the genetic analysis, additional characterization of SypG is required to demonstrate the proposed regulatory mechanism: What is the expression level (and phosphorylation state) of SypG and LuxO at different cell densities? Does purified SypG directly bind to the qrr1 promoter region?
c. How do these two bEBPs compete with each other if they are both made and active?6. The molecular OR logic gate is used to describe the relationship between LuxO and SypG, but this logic relationship is not always true in all conditions (if at all). In WT, deletion of luxO completely abolished Qrr1 expression (Fig. 4C). Even in the binK mutant, LuxO still seems to be the more prominent regulator (Fig. 4D) as deletion of luxO already caused a smaller but significant drop in Qrr1 expression. The authors may need to use this term more precisely.
-
