RclS Sensor Kinase Modulates Virulence of Pseudomonas capeferrum
and Branko Jovčić 1,2Round 1
Reviewer 1 Report
The manuscript “RclS sensor kinase modulates virulence of Pseudomonas capeferrum“ of Katarina Novovic et al. aims to elucidate the role (regulon) of the sensor kinase RclS of the three-component system RclSAR of Pseudomonas capeferrum. The investigation is based on a comparison of the transcriptomes of the wild-type strain with a rclS deletion strain in the exponential and the stationary growth phase. The identified high number of differentially regulated genes (about 1600 in exponential phase and about 600 in stationary phase) suggests that RclS plays a central role in regulating metabolic activities of the bacterium. Accordingly, the number of metabolic pathways and cellular functions affected by the deletion is very high. Phenotypic analyses revealed an impact of the deletion on antibiotic sensitivity, biofilm formation, adhesion, cytotoxicity and other phenomena.
I have the following queries:
1. Despite the large number of experiments and data presented in the manuscript, the precise function of RcdS remains a mystery. All that can be said is that the deletion of rcdS has very complex effects on multiple cellular functions. Which function is affected in particular remains unclear. Direct target genes of the 3-component system are not identified. Thus, the gain in knowledge is small.
2. The transcriptome analysis and the phenotypic analyses seem like two parts that are not directly related to each other. Why exactly were these phenotypic analyses chosen and what connection do they have to the transcriptome analysis?
3. The interactions of RcdS shown in Fig. 8 are entirely speculative with the data presented and are based on coincidences only. Evidence on the mechanism of the interactions (direct, indirect) of RcdS with the other proteins is lacking. Also, the authors included lasI in the model, although this is the only gene for which RT-qPCR did not confirm the transcriptome result. How is this explained?
4. The authors suggest that RcdS is also involved in prostranscriptional regulation. The evidence for this suggestion is very week and needs more investigations.
5. In Fig. 8, RcdS spans the entire cell wall with domains integrated into the inner AND outer membrane. What is the evidence for this membrane topology? I do not know of a polypeptide that exhibits such a membrane topology.
6. The transcriptome data sets should be made available in the appendix.
7. Refences need to be checked, e.g., 7 and 8.
8. The manuscript will benefit from an English editing service.
Author Response
Response to Reviewer 1 Comments
Point 1: Despite the large number of experiments and data presented in the manuscript, the precise function of RclS remains a mystery. All that can be said is that the deletion of rclS has very complex effects on multiple cellular functions. Which function is affected in particular remains unclear. Direct target genes of the 3-component system are not identified. Thus, the gain in knowledge is small.
Response 1: To some extent we have to agree with you, but also current knowledge about the activity of RclSAR system indicates that this system could be described as global regulatory system. In our previous study we observed that RclSAR system stimulates expression of the rpoS gene (https://doi.org/10.1016/j.resmic.2021.103885). It is well known that RpoS, the alternative σ subunit of RNA polymerase, is a major transcriptional regulator of the genes required for adaptation to the stationary phase in bacteria. Accordingly, it was expected that all cellular processes regulated by RpoS were also regulated by RclSAR, as a higher ranged regulator in this network, but indirectly. Also, in the present study we showed that this system positively regulates AHLs production (phenotypic test, RNA-seq and RT-qPCR in stationary growth phase-lasI and lasR genes). As we mentioned for RpoS, regulation of system included in AHLs production could initiate a downstream cascade of regulation also, having in mind the role of QS signal molecules in changes of gene expression pattern due to adaptation.
On the one hand, RNA-seq pointed out that this system has potential role in regulation of diverse metabolic and biocontrol processes in cell, but these observations are not confirmed experimentally. That would be the subject of further research. The lists of significant DEGs revealed by RNA-seq are now available in Tables S1 and S2. On the other hand, the phenotypic tests revealed the involvement of this system in virulence and antibiotic resistance regulation. Since we used knock-out in gene encoding for sensor kinase we could not definitely to claim which genes are the targets of this system, but in this and our previous study we definitely confirmed the changes in transcription of certain genes in DrclS mutant by RT-qPCR. Certainly our plans for the future are to examine whether the DNA-binding response regulator could interact with promoters of the rpoS, lasI and lasR genes.
Point 2: The transcriptome analysis and the phenotypic analyses seem like two parts that are not directly related to each other. Why exactly were these phenotypic analyses chosen and what connection do they have to the transcriptome analysis?
Response 2: You have the right that at first glance it seems as if these two parts of the study are not connested. However, our aim was to investigate if this system modulates virulence and antibiotic resistance of Psedomonas capeferrum since described three-component systems mostly were included in regulation of these cellular processes. Also, previously we observed that our system has genetic and domain organization as the Roc1 system from Pseudomonas aeruginosa, system included in surface attachment and host colonization, so we wanted to test if RclSAR system could regulate similar processes. In addition, we recognized this system as global regulatory system since it is stimulated activity of the rpoS gene promoter what was analyzed in our previous study. Accordingly, we wanted to get a broader picture of the regulatory potential of this system, and that is the reason why RNA-seq of the wild type and DrclS mutant strain was performed. Connecting of these two parts of the study was not simple since the function of most of the virulence and antibiotic resistance genes in WCS358 genome were not experimentally confirmed. Anyhow, our further research will be based on determination of specific genes included in this phenotype in WCS358 which are regulated by RclSAR. A few genes selected for RT-qPCR could explain the obtained results, but this is too speculative to be commented in the manuscript (bepC1-outer membrane efflux protein, mdtB1-multidrug resistance protein, ptlH-type IV secretion system protein).
Point 3: The interactions of RclS shown in Fig. 8 are entirely speculative with the data presented and are based on coincidences only. Evidence on the mechanism of the interactions (direct, indirect) of RclS with the other proteins is lacking. Also, the authors included lasI in the model, although this is the only gene for which RT-qPCR did not confirm the transcriptome result. How is this explained?
Response 3: Thank you for your comment. The aforementioned model included sensor kinase RclS, since the knock-out in this gene was used, and it is a part of three-component system with two response regulators. According to described mechanisms of function for these type of systems, after perceiving the signal RclS acts through interaction with regulators which further affect cellular changes. Accordingly, kinase cannot directly regulate transcription of target genes. In this model is shown the situation in stationary growth phase and results obtained for the lasI and lasR genes by RNA-seq and RT-qPCR were comparable and decreased in DrclS mutant in this phase of growth. Also, the production of AHLs after overnight incubation was decreased in DrclS mutant. Similarly, in our previous study (https://doi.org/10.1016/j.resmic.2021.103885) was noticed that promoter activity of the rpoS gene was reduced in mutant strain. Our aim was to supplement the model proposed in previous study by Bertani and Venturi and we hope that further research will improve this model additionally.
Point 4: The authors suggest that RclS is also involved in post-transcriptional regulation. The evidence for this suggestion is very week and needs more investigations.
Response 4: Thank you for the suggestion, we introduced changes according to your comment. In Discussion section we wrote “GO enrichment analysis, method for defining biological functions of detected DEGs, revealed that RclS sensor kinase is included in regulation of DEGs classified in RNA binding GO term. This observation is of special importance because it leads to conclusion that RclS sensor kinase, besides regulation on transcriptional level, could act as post-transcriptional regulator of gene expression.” In this statement we pointed out the result obtained by GO enrichment analysis which is very important in the context of determining the regulatory potential of this system. Since this observation is not experimentally confirmed, we changed this statement to potentially possible, but it is certainly not to be neglected.
Point 5: In Fig. 8, RclS spans the entire cell wall with domains integrated into the inner AND outer membrane. What is the evidence for this membrane topology? I do not know of a polypeptide that exhibits such a membrane topology.
Response 5: Thank you for your suggestion. In the revised model, RclS sensor kinase spans only through inner membrane. In our previous study we predicted domain organization of RclS sensor kinase (https://doi.org/10.1016/j.resmic.2021.103885). This sensor kinase is consisted of noncytoplasmic, transmembrane and cytoplasmic regions, so we predicted that this sensor kinase spans through membrane and are not cytoplasmic kinase.
Point 6: The transcriptome data sets should be made available in the appendix.
Response 6: Thank you for your suggestion. The lists of significant DEGs in exponential and stationary growth phase are available in Tables S1 and S2 now.
Point 7: Refences need to be checked, e.g., 7 and 8.
Response 7: Unfortunately, the doi of the citation 3 was moved in new row and was recognized as separate reference, so all reference after was moved by 1. Now, this problem is solved.
Point 8: The manuscript will benefit from an English editing service.
Response 8: The manuscript was edited by a native English speaker, as you suggested.
Reviewer 2 Report
The main goal of this work was to perform the identification and evaluation of the RclSAR system regulon in P. capeferrum WCS358 using a rclS knock-out mutant. Several techniques were used, namely: RNA sequencing, transcriptional analysis, antimicrobial, virulence and cytotoxicity evaluation. Then, at the end, they proposed an interconnection model of regulatory systems (RclSAR, RpoS and LasIR) in P. capeferrum WCS358.
General comments: The work is very well design. I suggest an English revision, focused on the Introduction and abstract sections.
Specific comments:
• The last paragraph of the introduction, in Page 2, you show results, which is not correct. In this paragraph you should only include the goals of the work and not results.
• Please do not used abbreviations before the extended word is presented (Page 3 lines 4 and 6 – Q20, Q30 and GC).
• Page 3 Table 1 – please include the legend of the table.
• Page 4 line 2 – you mention “corrected p value of 0.005 as parameters of significance”. I think you mean 0.05. Please verify.
• Page 5 – “GO and KEGG enrichment analysis”. Abbreviations may be included after the extended word is presented.
• Page 8 – Table 2 – please include the reference breakpoints used and respective reference.
• Page 12 – AHL stock solutions. Include the meaning
• Page 13 – M9 minimal medium - include meaning. Correct 0.05 McFarland to 0.5 McFarland in the 4.8 section.
• It is necessary to include several references in the manuscript that supports the sentences
o Page 1 – lines 2 and 4 of introduction
o Page 2 – lines 3, 20, 23, 34
o Page 9 – Discussion lines 2, 10, 11, 14 and 20
o Page 10 line 24
Author Response
Response to Reviewer 2 Comments
General comments: The work is very well design. I suggest an English revision, focused on the Introduction and abstract sections.
Thank you for your suggestions. We tried to answer to all your suggestions and comments. Also, the English is now improved.
Specific comments:
Point 1: The last paragraph of the introduction, in Page 2, you show results, which is not correct. In this paragraph you should only include the goals of the work and not results.
Response 1: Thank you for your suggestion. In most papers, as well as those cited in this manuscript (e.g.https://doi:10.1111/j.1365-2958.2004.04402.x;https://doi.org/10.1128/AEM.00712-17;https://doi.org/10.1128/AEM.70.9.5493-5502.2004), the last paragraph of introduction section summarizes the main results of the study. Nevertheless, we agree that this paragraph was too lengthy, so it is reduced to the concise outputs of this study now.
Point 2: Please do not used abbreviations before the extended word is presented (Page 3 lines 4 and 6 – Q20, Q30 and GC).
Response 2: The terms Q20, Q30 and GC content are established parameters used in RNA-seq analysis and represent percentages of bases whose correct base recognition rates are greater than 99% in total bases, percentages of bases whose correct base recognition rates are greater than 99.9% in total bases and percentages of G and C in total bases, respectively. Since those explanations are too long in most papers these parameters are represented as abbreviations (e.g. https://doi:10.3390/ijms19030717; https://doi.org/10.1002/mbo3.357; https://doi.org/10.1534/g3.113.009779; https://doi.org/10.3389/fgene.2022.782709). In line 6 of section 2.1. we indicated that Q20 and Q30 are base accuracy parameters.
Point 3: Page 3 Table 1 – please include the legend of the table.
Response 3: The legend of the Table has been included and named Summary of the RNA-seq data.
Point 4: Page 4 line 2 – you mention “corrected p value of 0.005 as parameters of significance”. I think you mean 0.05. Please verify.
Response 4: Thank you for the comment. We verify that p value of 0.005 is correct, because for the RNA-seq analysis of one biological replicate p value of significance is more stringent and this value is 0.005 instead of 0.05.
Point 5: Page 5 – “GO and KEGG enrichment analysis”. Abbreviations may be included after the extended word is presented.
Response 5: The complete names of these abbreviations were mentioned in Materials and Methods in the first version of the manuscript, but now are moved to Results according to your suggestion.
Point 6: Page 8 – Table 2 – please include the reference breakpoints used and respective reference.
Response 6: Thank you for the suggestion. Antibiotic susceptibility testing and results interpretation were performed according to EUCAST Breakpoint tables for interpretation of MICs and zone diameters Version 12.0, valid from 2022-01-01. It is important to emphasize that selection of antibiotics used in this study was made according to their belonging to different classes (structurally and functionally), and not according to their clinical relevance. Breakpoint values in EUCAST for Pseudomonas spp. Are given mostly for the antibiotics that are used in clinical practice for the treatment of infections predominantly caused by Pseudomonas aeruginosa as the most important representative of this genera from the clinical point of view. Thus breakpoint values for the number of antibiotic that we used are not given or have designation IE, meaning there is insufficient evidence that the organism or group is a good target for therapy with the agent. In these situations EUCAST suggests that an MIC with a comment but without an accompanying S, I or R categorisation may be reported, as we done in our study.
Point 7: Page 12 – AHL stock solutions. Include the meaning
Response 7: As for the enrichment analysis, the complete name of this abbreviation was mentioned in Materials and Methods, but now is moved to the Results.
Point 8: Page 13 – M9 minimal medium - include meaning. Correct 0.05 McFarland to 0.5 McFarland in the 4.8 section.
Response 8: In section 4.7. we now introduced the recipe for M9 minimal medium preparation.
The cells were diluted to the optical turbidity equivalent to 0.5 McFarland standard and then additionally diluted 10 times, so final optical turbidity was 0.05 McFarland. In revised manuscript we reformulated this sentence.
- It is necessary to include several references in the manuscript that supports the sentences
Point 9: Page 1 – lines 2 and 4 of introduction
Response 9: Since the cited Review (https://doi.org/10.1111/j.1574-6976.2008.00135.x) has contained enough basic knowledge about signal transduction systems we used this reference for the starting of introduction part.
Point 10: Page 2 – lines 3, 20, 23, 34
Response 10:
Page 2 – lines 3-The citation was added now.
Lines 20, 23-The references for indicted sentences were summarized in the following sentence [1,3,5,9,10].
Line 34-The citation was added now.
Point 11: Page 9 – Discussion lines 2, 10, 11, 14 and 20
Response 11:
Line 2-The citation was added now.
Line 10-The citations were added now.
Line 11-The references mentioned in the following sentence include this sentence also.
Line 14-The citations were added now.
Line 20-The citation was added now.
Point 12: Page 10 line 24
Response 12: The references which support this statement were indicated in the following sentence through citing corresponding examples.
Round 2
Reviewer 1 Report
I am still not happy with Fig. 8. It suggests that the sensor kinase RclS directly interacts with (e.g., phosphorylates) the response regulator LasR and the acyl homoserine lactone synthase LasI. This is clearly not the case (there is at least no evidence for this in the manuscript). Instead, the RclS-dependent 3-component system affects somehow the expression of lasR and lasI at the transcriptional level. This needs to be made clear in the model.
Author Response
Response to Reviewer 1 Comments
Point: I am still not happy with Fig. 8. It suggests that the sensor kinase RclS directly interacts with (e.g., phosphorylates) the response regulator LasR and the acyl homoserine lactone synthase LasI. This is clearly not the case (there is at least no evidence for this in the manuscript). Instead, the RclS-dependent 3-component system affects somehow the expression of lasR and lasI at the transcriptional level. This needs to be made clear in the model.
Response: Thank you for your suggestion. In the revised manuscript we changed our model in order to be less confusing. Now, the regulators are not presented as proteins in order not to misinterpret that regulation is performed at posttranslational level. In Figure legend is indicated that activation of signal transduction pathway starting with RclS activation leads to increased expression of the lasI, lasR and rpoS genes. Also, since transcriptional regulation by RclS is indirect it is represented in the form of dashed lines.
