RNase E cleavage shapes the transcriptome of Rhodobacter sphaeroides and strongly impacts phototrophic growth

This study identifies the cleavage sites of the endoribonuclease RNase E in the Rhodobacter sphaeroides transcriptome and demonstrates its effect on oxidative stress resistance and phototrophic growth.

Review Ms. Förstner et al. The authors performed comparative RNA-Seq using a Rhodobacter sphaeroides wild-type strain and a derivative strain in which the endogenous rne gene was replaced with the temperaturesensitive rne-3071 allele from E. coli. The transcriptomes were analyzed at 32{degree sign} C and at the non-permissive temperature of 42{degree sign} C. They observed 5'-ends that are enriched in the mutant strain as well as 5'-ends that are reduced. RNase E knockdown is shown to lead to a defect in the formation of photosynthetic complexes, resulting in retarded phototrophic growth. The study documents for an alpha-proteobacterium with versatile metabolic capacities how the activity of RNase E in RNA processing and decay moulds the bacterial transcriptome. Before publication, the manuscript requires major revision to improve its readability.
Major comments: 1) The manuscript needs extensive language editing; I have made numerous suggestions below.
2) The figures, and the lettering in particular, are too smallI. Use roughly equal letter size for all part figures and figures should be readable on a normal A4 printout. Also, in many figures, the panels are marked as "a", "b" etc., but are referred to in the text as panel "A", "B" etc.; please harmonize. 3) Some bioinformatic illustrations (Fig. 1C,D / Fig. 2 / Fig. S1) should be better explained in the figure legends and text (e.g. p.4, 3rd paragraph), such that the reader can fully understand the information that is provided; what means "base mean" (x-axis) in Fig. 1C, D? What is the reason for the morphology of the plots below base mean values of about 10? In the legend to Fig. 2, the authors should clearly state that panels A, C and E show the sum of all 5'-ends assumed to be RNase cleavage sites (41,000), as inferred fom the fact that they were depleted in the mutant at the nonpermissive temperature; likewise, explain that panels B, D and F are the illustrations for the enriched sites (the 41,000 sites reduced to 23,000). The difference between panels A/C/E and B/D/F should also be immediately recognizable in the figure by differential axis lettering. 4) Fig, 1A,B: explain the boxes with light blue/cyan columns, captioned "mapped 5' ends". 5) The colour code used for the strains is supoptimal; particularly in Fig. 5C, the light green and blue shades are hard to differentiate; use colours that can be better distinguished. In Fig. 5C, move the assignment of colour (WT and rne ts) to outside the box to clearly separate these dots from the data points within the box. 6) p.3, lines 22 to 26: the information on SorY is not relevant here, can be omitted. 7) p.3, line 28, make clearer what you mean here: "... mapping of RNase E cleavage site (which are reduced upon RNase E depletion) ..." 8) p.7, 2nd paragraph, discussion of data for SorX: I do not see clear coherence of NB and RNA-seq data: in the NB, pre-SorX (116 nt) is enriched in the mutant relative to WT at 32{degree sign} C, but the reverse is seen for the coverage of the corresponding 5'-end; the latter increases for the mutant at 42{degree sign} C vs. 32{degree sign} C, but the pre-SorX signal in the NB decreases for the mutant at 42 vs 32{degree sign} C. Please scrutinize the current discussion of the SorX data. 9) p.8, last sentence of the 1st paragraph appears rather unexpectedly. The authors discuss that the NB and RNA-Seq data for PcrZ do not indicate much difference between WT and mutant, and then suddenly mention that RNase E cleavage of PcrZ may have stong impact on the formation of photosynthetic complexes. Please modify. Minor comments: 10) throughout manuscript: "Gram-negative", not "gram-negative" 11) p.3, lines 17/18: I doubt that most readers understand what is meant by an "incoherent feedforward loop". Since the details of this loop are not important here, use a more general description, e.g. "regulatory circuit". 12) p.5, 1st paragraph, last sentence: I doubt that ref. 42 was the one that showed endonucleolytic J1 activity in B. subtilis. 13) p.5, 3rd paragraph, line 4: what do you mean by "within stretches of the same nucleotide"? Homonucleotide stretches? / line 5: what is meant by "in a proximity of 3 nt"? (see also p.12, line 8). Let us assume an example sequence, "5'-GCATACC" (cleavage on the 5' site of the nucleotides); when counting from the central T, do you mean to assign also cleavages up to the first G and to the last C to the same site? Please make clearer. 14) p.7, 2nd paragraph: describe first the sRNAs shown in Fig. 4 and then those documented in Fig.  S3. 15) p.7, 3rd paragraph, line 4 from bottom to end of page: complete genome position "2,565" in line 2 from bottom; I also have difficulties to follow the argumentation; I see multiple 5'-ends from position 2,565,910 to 2,565,935, indicating 5'-end heterogeneity of the "55 nt" fragment ensemble; the 5'-end signals are similar for the WT and mutant at both temperatures, but the NB indicates a somewhat lower abundance of the "55 nt" fraction in the mutant. Please clarify. The authors could add a sentence stating that the NB and RNA-seq 5' end coverage data are not always consistent in terms of quantitative trends. 16) p.8, 2nd paragraph: the first sentence can be deleted. 17) p. 8, 2nd paragraph, discussion of data for RSP_7527 is rather confusing, rewrite. 18) p.12, line 4: what do you mean by "for the remaining one" ? 19) Legend to Fig. 3: what is meant by "between position -1 and 1 or 1 and 2" and by "mutant enriched sites" ? 20) Fig. S4: change y-axis lettering to "ratio OD770 to OD660" /explain abbreviations "LO" and "PT" in the figure legend 21) Fig. S6: give growth temperature in the legend Suggestions for text changes: p.1: -Abstract, line 8: delete "in this GC rich alpha proteobacterium" (not relevant as abstract information) -Introduction, line 2 rewrite to "Successful adaptation to changing conditions ..." p.2: -3rd paragraph, line 5: "... by prior endonucleolytic cleavage of the RNA: Binding of RNase E to a ..." -4th paragraph, line 5: "...symbionts and pathogens. An important role ..." -line 6: homeostasis -line 8: delete comma after "mRNA" p.3: -line 11: "...restricted to the model organism R. sphaeroides whose genome was the first to become available in this group of bacteria [28,29]. RNA-seq analyses of R. sphaeroides were performed for the first time . To study the relevance of RNase E cleavage for the shaping of the R. sphaeroides transcriptome, ..." -2nd paragraph, line 7: "... at 32{degree sign} C (see below), demonstrating ..." -2nd paragraph, line 9: "...the two strains, which were harvested in mid-exponential growth phase at 32{degree sign} C or 20 min after ..." -3rd paragraph, line 6, rewrite: "...shift to 42{degree sign} C, which was in most cases not observed, or to a lesser extent, with RNA derived from cultures grown at 32{degree sign} ..." -3rd paragraph, line 12: comma before "may" -last paragraph: lines 5/6: what do you mean by "(about 2 500)"? -last paragraph: line 7: "... to 42{degree sign} C may have affected ..." p.5: -3rd paragraph, line 1, rewrite: "5' ends enriched in the wild type and those enriched in the mutant are often ..." -last paragraph, line 7 from bottom: "Only about 2 % of the cleavage sites mapped to 3' UTRs; p.6: -3rd paragraph, sentence starting in line 3, rewrite: "Although R. sphaeroides has a higher genomic G,C content (~69%) than E. coli (~51%), R. sphaeroides RNase E shows a preference for A,U-rich target sites as well." -4th paragraph, rewrite last sentence (cleavage site rates were not measured): "While RNase E cleavage sites were oberved to occur frequently in close proximity of stop codons in Salmonella [43], we found a relatively high incidence of cleavage sites around start codons." -5th paragraph, line 2 from bottom: comma before "demonstrating" -last paragraph, last line: comma after "processing" p.7: -2nd paragraph, line 6 from bottom: "The function of this sRNA is unknown." -3rd paragraph, first sentence: "As SorX [34], sRNA SorY has a role in the oxidative stress response [33]. Unlike ..." -3rd paragraph, line 6: "Another sRNA with known function in R. sphaeroides is PcrZ (136 nt), which is involved in the regulation of photosynthesis gene expression [31]." (most readers will not understand what is meant by an ""incoherent feed-forward loop") -3rd paragraph, line 6 from bottom: "... rneE.colie(ts) gene from E. coli. In agreement with this, the corresponding Northern blot indicates that the amount of the 136 nt transcript does not significantly vary in the two strains at the permissive or non-permissive temperature (Supplementary Figure S3D)." p.8: -3rd paragraph, line 4: "... more abundant in the mutant strain at both temperatures ..." -3rd paragraph, last line: replace "not shown" with " Fig. S3F" p.9: -line 3, rephrase: "... or phototrophic growth conditions. This corresponds to the observation that phototrophic growth of the mutant is strongly retarded ( Figure 5B). Remarkably, both strains showed identical growth behavior when grown chemotrophically. This finding underlines the important physiological role of RNase E in R. sphaeroides and further demonstrates that RNase E can have a selective impact on specific physiological processes." -2nd paragraph, line 3: "...(pufB and pufA for the light harvesting complex I and pufL and pufM for the reaction center), a protein involved in assembly of the complexes (pufX) and another one affecting early bacteriochlorophyll synthesis (pufQ) [52]." -2nd paragraph, line 7: introduce abbreviation "LHI" already in line 4 -2nd paragraph, line 10: "Again, RNase E initiates further processing by cleaving in the 5' region of the pufL coding sequence [21,23]." -2nd paragraph, line 15: "However, there are more RNase E ..." -2nd paragraph, line 18: "... to identify primary sites that limit the overall rate of processing. As in the puf operon, ..." -2nd paragraph, line 21: "...light-harvesting II complex, bch genes code for enzymes of bacteriochlorophyll synthesis and crt genes for enzymes of carotenoid synthesis) ..." -2nd paragraph, line 25: " Some mRNAs encoding important regulators ..." -2nd paragraph, line 28: "Nevertheless, the strong effect ... cannot be deduced ..." -2nd paragraph, line 32: delete comma after "elusive" p.10: -line 2: delete "which are" -line 6: "...64 and 30 per kb, respectively). All these ..." -line 19: "However, the mutant ..." -line 21: "This demonstrates that RNase E also affects other ..." -line 23: "...RNase E in RNA metabolism, despite the presence of several enzymes with ribonucleolytic activity in bacteria, and ..." -line 25: "...and decay on regulation of gene expression, an aspect that is still neglected in many studies focusing on gene regulation." -2nd paragraph, line 3: "...temperature-sensitive E. coli rne-3071 (ts) allele derived from strain E. coli N3431 [4,59]. Construction of this strain is described ..." -2nd paragraph, line 7: "...conditions in the presence of 40 W ..." 3rd paragraph, line 3: "...an extinction coefficient of 76 mM-1 cm-1 at 770 nm." p.11: -1st paragraph, line 1: 1.5 ml / lines 4 and 5: units of "100 protein" and "400 ICM buffer" ? -2nd paragraph, line 2: diameter / line 7: "...oxidative stress agent was defined as 100 % survival. The percentage of colonies grown from the treated cultures (percent survival) was referred to the 100% of the control culture." 3rd paragraph, line 3: "per 1 RNA": unit lacking / same line: "...by PCR against the gloB ..." 3rd paragraph, line 7: "For detection of SorX, a PCR product (primers listed in Tabel S1) was labeled with .. The manuscript entitled "RNase E cleavage shapes the transcriptome of Rhodobacter sphaeroides and strongly impacts phototrophic growth" by Konrad U. Förstner and colleagues reports a transcriptome-wide mapping of RNase E cleavage sites by TIER-seq in the GC-rich alpha proteobacterium R. sphaeroides. As RNase E seems essential in their model organism, the authors chose to replace the endogenous rne gene by a temperature sensitive (ts) E. coli rne gene and to compare their transcriptomes at permissive and non-permissive temperatures. The relevance of the genome-wide data produced was assessed on the processing of two UTR-derived sRNAs, UpsM [PMID: 27802301] and SorX [PMID: 27420112], known to be RNase E-dependent. More specifically, the TIER-seq strategy developed here allowed the identification of new UTR-derived sRNAs generated by RNase E processing, confirmed the key role played by RNase E on the formation of photosynthetic complexes by initiating a differential degradation of the pufQBALMX polycistronic mRNA in Rhodobacterales and established the role of RNase E on oxidative stress response by identifying mRNAs coding sigma factors as RNase E targets. The manuscript falls into the scope of Life Science Alliance. The TIER-seq strategy that was implemented by the authors as developed in Chao et al. (2017) [PMID: 28061332] seems sound and provides resources that should be valuable for people working on RNA metabolism in bacteria and on metabolic switches in Rhodobacterales (e.g. stress response, phototrophic vs chemotrophic growth). Nonetheless, I have some major concerns that should be addressed by the authors before considering it for publication. I will restrict my comments as general ones, as more specific details can be addressed in the second round of revision. Nevertheless, I advise the authors to seriously improve the writing of the manuscript before resubmission (if accepted by Life Science Alliance). The punctuation marks are mostly missing, making it difficult to read. The introduction should be restricted to what is actually discussed in the "Results and Discussion" section. Repetitions between the "Introduction" and "Results and Discussion" should be avoided when unnecessary. Some long paragraph would benefit from being sectioned. As written, the main conclusions of the section "Results and Discussion" are difficult to extract and are not put in perspective to what is known for other organisms. Many typos remain, especially in the section "Materials and Methods", should be corrected (mutant strain with two different writings, diameter with an r missing...). My main concern is the choice of comparing transcriptome data from R. sphaeroides cells expressing the endogenous RNase E or an E. coli ts RNase E. As classified in Aït- Bara and Carpousis (2015) [PMID: 26096689], E. coli and R. sphaeroides RNase E are type I and type II, respectively. While type II resembles type I in the organization of their catalytic region (with the exception of an insertion in the S1 domain), type I RNase E members have a membrane targeting sequence that has not been discerned in any type II RNase E. In E. coli (type I) and Caulobacter crescentus (another alpha-proteobacterium; type II), RNase E molecules have different cellular locations that seem to result in different spatial organization of their transcriptomes [see: PMID: 27198188; PMID: 29610352; PMID: 20562858]. Therefore, one can wonder if the genome-wide data produced in this study are physiologically relevant. In my opinion, it would have been better to perform a depletion of the endogenous RNase E in the cells as performed in Helicobacter pylori for RNase J using a plasmid to conditionally express RNase E [PMID: 26726773]. Although the authors cannot change their data sets, it should be more extensively discussed in the manuscript. While the authors discussed how the differences in the composition of E. coli and R. sphaeroides degradosomes could impact the results, the impact of potentially different RNase E localizations is not addressed. A second concern is an apparent confusion the authors make between RNA abundance and RNA stability as suggested by some result interpretations in the section "The effect of RNase E on maturation / stability of selected substrates" in the "Results and Discussion". While the authors clearly state in the abstract that "Bacteria adapt to changing environmental conditions by rapid changes in the transcriptome. This is achieved by adjusting rates of transcription but also by processing and degradation of RNAs.", they conclude that variation in RNA abundances on Northern blot results from a variation in stability without proving it. Indeed, at no point in the manuscript, variations in RNA transcription or RNA stability are experimentally addressed. Therefore, the title of the section should be changed and conclusion like "(...) the lack of RNase E reduces the 5' end-dependent pathway of degradation and stabilizes pre-SorX as also indicated by the Northern blot (...)" should be corrected.
Finally, the authors put emphasis that their strategy will, additionally to determining RNase E cleavage sites, give the 5' RNA ends. I am not sure to understand the difference between the two, as it is the same for me. The authors should try to better define what they imply by this distinction and draw specific conclusions. Otherwise, I would suggest to simply their manuscript by only referring to RNase E cleavage sites.
1st Authors' Response to Reviewers: July 8, 2018 Reviewer #1 (Comments to the Authors (Required)): The authors globally identified the RNase E cleavage sites and 5' ends in Rhodobacter sphaeroides by applying the TIER-seq method. In this method the endoribonucleases are inactivated for a short time and this is followed by RNA-seq. The results demonstrate the importance of RNase E to different aspects in the life of R. sphaeroides such as: maturation and turn-over of RNAs or coping with oxidative stress. The paper is well-written and is easy to read. It also provides an important resource for researchers studying Rhodobacter sphaeroides. However, the paper could be improved by the following: 1. The authors state that "RNase E shows a preference for AU rich target sites". The only support they provide for this statement is a weak motif shown in Figure 3(a). The authors should consider additional experiments to support this statement or to better address this issue. For example, replacing the AU in one the RNase E binding sites and test how it affects the cleavage.

Authors' response:
We thank the reviewer for the suggestions. It was shown in previous publications from our group that RNase E also prefers AU-rich sequences at selected sites in Rhodobacter capsulatus. We added this information and the references.
2. The manuscript is missing a more comprehensive discussion about the role of RNase E in bacteria, what do the authors consider is its role in comparison to other RNases? For example, the last sentence in the Results and Discussion section can be expanded to a paragraph. Minor comments: 4. Page 4, second paragraph: Adding a comment about the degree of similarity between RNase E from E. coli and RNase E from R. sphaeroides will be valuable.
Authors' response: This information was included.
5. Page 7, bottom paragraph: "it is transcribed from an own promoter..." should be changed to "it is transcribed from its own promoter" Authors' response: The text was changed.
6. Figure 1 Authors' response: The 80 were chosen as there are genes with up to 80 sites but in very low frequency. We have adapted the image and the range goes now up to 40 sites per gene while keeping as zoomed extract to the frequencies of 1 -10 sites.
8. Figure 5: The graphs' formatting is not uniform. For example the graphs in (b) are missing a boarder that is present for the graphs in (a) and (c).

Authors' response:
We understand the potential disbalance that the reviewer sees here intuitively. But all panels have a grid in the actual graph canvas. Adding an additional border around (B) would be non-functional visual noise that distracts from the actual information (see Tufte, Edward R. (1983). The Visual Display of Quantitative Information. Cheshire).
The authors performed comparative RNA-Seq using a Rhodobacter sphaeroides wild-type strain and a derivative strain in which the endogenous rne gene was replaced with the temperature-sensitive rne-3071 allele from E. coli. The transcriptomes were analyzed at 32{degree sign}C and at the non-permissive temperature of 42{degree sign}C. They observed 5'-ends that are enriched in the mutant strain as well as 5'-ends that are reduced. RNase E knockdown is shown to lead to a defect in the formation of photosynthetic complexes, resulting in retarded phototrophic growth. The study documents for an alpha-proteobacterium with versatile metabolic capacities how the activity of RNase E in RNA processing and decay moulds the bacterial transcriptome. Before publication, the manuscript requires major revision to improve its readability.
Major comments: 1) The manuscript needs extensive language editing; I have made numerous suggestions below.

Authors' response:
We are thankful for the detailed and helpful comments and changed the text accordingly.
2) The figures, and the lettering in particular, are too smallI. Use roughly equal letter size for all part figures and figures should be readable on a normal A4 printout. Also, in many figures, the panels are marked as "a", "b" etc., but are referred to in the text as panel "A", "B" etc.; please harmonize.

Authors' response:
We thank reviewer for pointing to this and have addressed this in the whole manuscript as well as the supplementary material.
3) Some bioinformatic illustrations (Fig. 1C,D / Fig. 2 / Fig. S1) should be better explained in the figure legends and text (e.g. p.4, 3rd paragraph), such that the reader can fully understand the information that is provided; what means "base mean" (x-axis) in Fig. 1C, D? What is the reason Authors' response: We modified the text.