A bacterial regulatory uORF senses multiple classes of ribosome-targeting antibiotics

Reviewer #1 (Public review):

Summary:

The manuscript reports that expression of the E. coli operon topAI/yjhQ/yjhP is controlled by the translation status of a small open reading frame, that authors have discovered and named toiL, located in the leader region upstream of the operon. Authors propose the following model for topAI activation: Under normal conditions, toiL is translated but topAI is not expressed because of Rho-dependent transcription termination within the topAI ORF and because its ribosome binding site and start codon are trapped in an mRNA hairpin. Ribosome stalling at various codons of the toiL ORF, prompted in this work by some ribosome-targeting antibiotics, triggers an mRNA conformational switch which allows translation of topAI and, in addition, activation of the operon's transcription because presence of translating ribosomes at the topAI ORF blocks Rho from terminating transcription. The model is appealing and several of the experimental data mainly support it. However, it remains unanswered what is the true trigger of the translation arrest at toiL and what is the physiological role of the induced expression of the topAI/yjhQ/yjhP operon.

Reviewer #2 (Public review):

Summary:

Baniulyte and Wade describe how translation of an 8-codon uORF denoted toiL upstream of the topAI-yjhQP operon is responsive to different ribosome-targeting antibiotics, consequently controlling translation of the TopAI toxin as well as Rho-dependent termination with the gene.

Strengths:

The authors used multiple different approaches such as a genetic screen to identify factors such as 23S rRNA mutations that affect topA1 expression and ribosome profiling to examine the consequences of various antibiotics on toiL-mediated regulation.

Weaknesses: Future experiments will be needed to better understand the physiological role of the toiL-mediated regulation and elucidate the mechanism of specific antibiotic sensing.

The results are clearly described, and the revisions have helped to improve the presentation of the data.

Reviewer #3 (Public review):

In this revised manuscript, the authors provide convincing data to support an elegant model in which ribosome stalling by ToiL promotes downstream topAI translation and prevents premature Rho-dependent transcription termination. However, the physiological consequences of activating topAI-yjhQP expression upon exposure to various ribosome-targeting antibiotics remain unresolved. The authors have satisfactorily addressed all major concerns raised by the reviewers, particularly regarding the SHAPE-seq data. Overall, this study underscores the diversity of regulatory ribosome-stalling peptides in nature, highlighting ToiL's uniqueness in sensing multiple antibiotics and offering significant insights into bacterial gene regulation coordinated by transcription and translation.

Author response:

The following is the authors’ response to the original reviews

Public Reviews:

Reviewer #1 (Public Review):

Summary:

The manuscript reports that expression of the E. coli operon topAI/yjhQ/yjhP is controlled by the translation status of a small open reading frame, that authors have discovered and named toiL, located in the leader region of the operon. The authors propose the following model for topAI activation: Under normal conditions, toiL is translated but topAI is not expressed because of Rho-dependent transcription termination within the topAI ORF and because its ribosome binding site and start codon are trapped in an mRNA hairpin. Ribosome stalling at various codons of the toiL ORF, caused by the presence of some ribosome-targeting antibiotics, triggers an mRNA conformational switch which allows translation of topAI and, in addition, activation of the operon's transcription because the presence of translating ribosomes at the topAI ORF blocks Rho from terminating transcription. Even though the model is appealing and several of the experimental data support some aspects of it, several inconsistencies remain to be solved. In addition, even though TopAI was shown to be an inhibitor of topoisomerase I (Yamaguchi & Inouye, 2015, NAR 43:10387), the authors suggest, without offering any experimental support, that, because ribosome-targeting antibiotics act as inducers, expression of the topAI/yjhQ/yjhP operon may confer resistance to these drugs.

Strengths:

- There is good experimental support of the transcriptional repression/activation switch aspect of the model, derived from well-designed transcriptional reporters and ChIP-qPCR approaches.

- There is a clever use of the topAI-lacZ reporter to find the 23S rRNA mutants where expression topAI was upregulated. This eventually led the authors to identify that translation events occurring at toiL are important to regulate the topAI/yjhQ/yjhP operon. Is there any published evidence that ribosomes with the identified mutations translate slowly (decreased fidelity does not necessarily mean slow translation, does it?)?

G2253 is in helix 80 of the 23S rRNA, which has been proposed to be involved in correct positioning of the tRNA. Mutations in helix 80 have been reported to cause defects in peptidyl transferase center activity, which could reduce the rate of ribosome movement along the mRNA. If ribosomes are sufficiently slowed when translating toiL, this could induce expression of topAI. G1911 and Ψ1917 are in helix 69 of the 23S rRNA, which is involved in forming the inter-subunit bridge, as well as interactions with release factors. Mutations in helix 69 cause a decrease in the processivity of translation, suggesting that the mutations we identified may increase the occupancy of ribosomes within toiL, thereby inducing expression of topAI. We have added text to the Discussion section to include this speculation.

- Authors incorporate relevant links to the antibiotic-mediated expression regulation of bacterial resistance genes. Authors can also mention the tryptophan-mediated ribosome stalling at the tnaC leader ORF that activates the expression of tryptophan metabolism genes through blockage of Rho-mediated transcriptional attenuation.

We have added a citation to a recent structural study of ribosomes translating the tnaC uORF. Specifically, we speculate in the Discussion that toiL may have evolved to sense a ribosome-targeting antibiotic, or another ribosome-targeting small molecule such as an amino acid.

Weaknesses:

The main weaknesses of the work are related to several experimental results that are not consistent with the model, or related to a lack of data that needs to be included to support the model.

The following are a few examples:

- It is surprising that authors do not mention that several published Ribo-seq data from E. coli cells show active translation of toiL (for example Li et al., 2014, Cell 157: 624). Therefore, it is hard to reconcile with the model that starts codon/Shine-Dalgarno mutations in the toiL-lux reporter have no effect on luciferase expression (Figure 2C, bar graphs of the no antibiotic control samples).

These data are for a topAI-lux reporter construct rather than toiL-lux. In our model, ribosome stalling within toiL is required to induce expression of the downstream genes; preventing translation of toiL by mutating the start codon or Shine-Dalgarno sequence would not cause ribosome stalling, consistent with the lack of an effect on topAI expression.

- The SHAPE reactivity data shown in Figure 5A are not consistent with the toiL ORF being translated. In addition, it is difficult to visualize the effect of tetracycline on mRNA conformation with the representation used in Figure 5B. It would be better to show SHAPE reactivity without/with Tet (as shown in panel A of the figure).

We have modified this figure (now Figure 6) so that we no longer show the SHAPE-seq data +/- tetracycline overlayed on the predicted RNA structure, since at best, the predicted structure likely only represents uninduced state. We have included the predicted structure together with the SHAPE-seq data for untreated cells as a separate panel because it is part of the basis for our model. We have also added a supplementary figure showing a similar RNA structure prediction based on conservation of the topAI upstream region across species (Figure 6 – figure supplement 1), and we describe this in the text.

- The "increased coverage" of topAI/yjhP/yjhQ in the presence of tetracycline from the Ribo-seq data shown in Figure 6A can be due to activation of translation, transcription, or both. For readers to know which of these possibilities apply, authors need to provide RNA-seq data and show the profiles of the topAI/yjhQ/yjhP genes in control/Tet-treated cells.

A previous study (Li et al., 2014, PMID 24766808) compared RNA-seq and Ribo-seq data for E. coli to measure normalized ribosome occupancy for each gene. However, sequence coverage for topAI was too low to confidently quantify either the RNA-seq or the Ribo-seq data. Presumably RNA levels were low because of Rho termination. Hence, we were not confident that RNA-seq would provide information on the regulation of topAI-yjhQP. Other data in our study provide strong evidence that regulation is primarily at the level of translation. And the key conclusion from Figure 6 (now Figure 7) is that tetracycline stalls ribosomes on start codons.

- Similarly, to support the data of increased ribosomal footprints at the toiL start codon in the presence of Tet (Figure 6B), authors should show the profile of the toiL gene from control and Tet-treated cells.

Figure 6B shows data for both treated and untreated cells. The overall ribosome occupancy is much lower for untreated cells, making it difficult to draw strong conclusions about the relative distribution of ribosomes across toiL.

- Representation of the mRNA structures in the model shown in Figure 5, does not help with visualizing 1) how ribosomes translate toiL since the ORF is trapped in double-stranded mRNA, and 2) how ribosome stalling on toiL would lead to the release of the initiation region of topAI to achieve expression activation.

We now show the predicted structure with only SHAPE-seq data for untreated cells. The comparison of SHAPE-seq +/- tetracycline is shown without reference to the predicted structure.

- The authors speculate that, because ribosome-targeting antibiotics act as expression inducers [by the way, authors should mention and comment that, more than a decade ago, it had been reported that kanamycin (PMID: 12736533) and gentamycin (PMID: 19013277) are inducers of topAI and yjhQ], the genes of the topAI/yjhQ/yjhP operon may confer resistance to these antibiotics. Such a suggestion can be experimentally checked by simply testing whether strains lacking these genes have increased sensitivity to the antibiotic inducers.

We thank the reviewer for pointing out these references, which we now cite. The fact that another group found that gentamycin induces topAI expression – it is one of the most highly induced genes in that paper – strongly suggests that we missed the key inducing concentrations for one or more antibiotics, meaning that topAI is induced by even more ribosome-targeting antibiotics than we realized.

We did some preliminary experiments to look for effects of TopAI, YjhQ, and/or YjhP on antibiotic sensitivity, but generated only negative results. Since these experiments were preliminary and far from exhaustive, we have chosen not to include them in the manuscript. Other studies of genes regulated by ribosome stalling in a uORF have looked at genes whose functions in responding to translation stress were already known, so the environmental triggers were more obvious. With so many possible triggers for topAI-yjhQP, it will likely require considerable effort to find the relevant trigger(s). Hence, we consider this an important question, but beyond the scope of this manuscript.

Reviewer #2 (Public Review):

Summary:

In this important study, Baniulyte and Wade describe how the translation of an 8-codon uORF denoted toiL upstream of the topAI-yjhQP operon is responsive to different ribosome-targeting antibiotics, consequently controlling translation of the TopAI toxin as well as Rho-dependent termination with the gene.

Strengths:

I appreciate that the authors used multiple different approaches such as a genetic screen to identify factors such as 23S rRNA mutations that affect topA1 expression and ribosome profiling to examine the consequences of various antibiotics on toiL-mediated regulation. The results are convincing and clearly described.

Weaknesses:

I have relatively minor suggestions for improving the manuscript. These mainly relate to the figures.

Reviewer #3 (Public Review):

Summary:

The authors nicely show that the translation and ribosome stalling within the ToiL uORF upstream of the co-transcribed topAI-yjhQ toxin-antitoxin genes unmask the topAI translational initiation site, thereby allowing ribosome loading and preventing premature Rho-dependent transcription termination in the topAI region. Although similar translational/transcriptional attenuation has been reported in other systems, the base pairing between the leader sequence and the repressed region by the long RNA looping is somehow unique in toiL-topAI-yjhQP. The experiments are solidly executed, and the manuscript is clear in most parts with areas that could be improved or better explained. The real impact of such a study is not easy to appreciate due to a lack of investigation on the physiological consequences of topAI-yjhQP activation upon antibiotic exposure (see details below).

Strengths:

Conclusion/model is supported by the integrated approaches consisting of genetics, in vivo SHAPE-seq and Ribo-Seq.

Provide an elegant example of cis-acting regulatory peptides to a growing list of functional small proteins in bacterial proteomes.

Recommendations for the authors:

Reviewing Editor Comments:

(1) Examine the consequences of mutations impeding translation of the topAI/yjhQ/yjhP operon on cell growth in the presence and absence of antibiotics.

See response to Reviewer 1’s comment.

(2) Resolve discrepancies between the SHAPE data indicating constitutive sequestration of the toiL Shine Dalgarno sequence with antibiotic-regulated translation of the toiL ORF.

See response to Reviewer 1’s comment.

(3) Reconcile published Ribo-Seq data with the model that start codon/Shine-Dalgarno mutations in the toiL-lux reporter have no effect on luciferase expression in the absence of antibiotics.

See response to Reviewer 1’s comment.

(4) Clarify whether antibiotic MIC values were employed to select antibiotic concentrations for different experiments.

The antibiotic concentrations we used are in line with reported MICs for E. coli. We now list the reported ECOFFs/MICs and include relevant citations.

(5) Provide RNA-seq data to complement the Ribo-Seq data for the topAI/yjhQ/yjhP genes in control vs. Tet-treated cells.

See response to Reviewer 1’s comment.

(6) Revise the text to address as many of the reviewers' suggestions as reasonably possible.

Changes to the text have been made as indicated in the responses to the reviewers’ comments.

Reviewer #2 (Recommendations for the Authors):

(1) Page 6: I would have liked to have more information about the 39 suppressor mutations in rho. Do any of the cis-acting mutations give support for the model proposed in Figure 8?

We only know the specific mutation for some of the strains, and we now list those mutations in the Methods section. For other mutants, we mapped the mutation to either the rho gene or to Rho activity, but we did not sequence the rho gene. Most of the specific mutations we did identify fall within the primary RNA-binding site of Rho and hence should be considered partial-loss-of-function mutations (complete loss of function would be lethal).

We identified cis-acting mutations by re-transforming the lacZ reporter plasmid into a wild-type strain. We did not sequence any of these plasmids.

(2) Page 12-13, Section entitled "Mapping ribosome stalling sites induced by different antibiotics": This section should start with a better transition regarding the logic of why the experiments were carried out and should end with an interpretation of the results.

We have added a few sentences at the start of this section to explain the rationale. We have also added two sentences at the end of this section to summarize the interpretation of the data.

(3) Page 15: The authors should discuss under what conditions the expression of TopAI (and YjhQ/YjhP might be induced? Is expression also elevated upon amino acid starvation?

We have looked through public RNA-seq data but have not identified growth conditions other than antibiotic treatment that induce expression of topAI, yjhQ or yjhP.

(4) References: The authors should be consistent about capitalization, italics, and abbreviations in the references.

These formatting errors will be fixed in the proofing stage.

(5) All graph figures: There should be more uniformity in the sizes of individual data points (some are almost impossible to see) and error bars across the figures.

We have tried to make the data points and error bars more visible for figures where they were smaller.

(6) Figure 1B: I do not think the left arrow labeling is very intuitive and suggest renaming these constructs.

We have removed the arrows to improve clarity.

(7) Figure 2A: toiL should be introduced at the first mention of Figure 2A.

We have added a schematic of the topAI-yjhQ-yjhP region as Figure 1A, including the toiL ORF, which we briefly mention in the text. We have opted to split Figure 2C into two panels. In Figure 2C we now only show data for the wild-type construct. Data for the mutant constructs are now shown in a new figure (Figure 5), alongside data for the wild-type constructs. We have simplified Figure 2A, since the mutations are not relevant to this revised figure, and we now show the schematic with the mutations as Figure 5A.

(8) Figure 3C and 3D: I suggest giving these graphs headings (or changing the color of the bars in Figure 3D) to make it more obvious that different things are measured in the two panels.

We have added headers to panels B-D make it clear that which graphs show ChIP-qPCR data which graph shows qRT-PCR data.

(9) Figure 6: It might be nice to show the topAI-yjhPQ operon here.

We now show the operon in Figure 1A.

(10) Figure 8: This figure could be optimized by adding 5' and 3' end labels and having more similarity with the model in Figure 7.

The constructs shown in Figure 7 lack most of the topAI upstream region, so they aren’t readily comparable to the schematic in Figure 8. However, we have changed the color of the ribosome in Figure 7 to match that in Figure 8. We also indicate the 5’ end of the RNA in Figure 8.

Reviewer #3 (Recommendations for the Authors):

Areas to improve:

(1) While it's important to learn about ToiL-dependent regulation of the downstream topAI-yjhQ toxin-antitoxin genes, the physiological consequence of topAI-yjhQ activation seems to be lost in the manuscript. Everything was done with a reporter lacZ/lux. In the absence of toiL translation (i.e. SD mutant) and/or ribosome stalling, does premature transcription termination result in non-stochiometric synthesis of toxin vs. antitoxin, leading to growth arrest or other measurable phenotype? Knowing the impact of ToiL in the native topAI-yjhQ context will be valuable.

See response to Reviewer 1’s comment.

(2) It was indicated in Figure 4-figure supplement 1 that toiL homologs are found in many other proteobacteria, are the UR sequences in those species also form a similar inhibitory RNA loop?? The nt sequence identity of toiL is likely to be constrained by the base pairing of the topAI 5' region.

We have added a supplementary figure panel showing an RNA structure prediction for the topAI upstream region based on sequence alignment of homologous regions from other species (Figure 6 – figure supplement 1).

What is the frequency of the MLENVII hepta-peptide in the E. coli genome-wide. Is the sequence disfavored to avoid spurious multi-antibiotic sensing?

LENVII is not found in any annotated E. coli K-12 protein. However, this is a sufficiently long sequence that we would expect few to no instances in the E. coli proteome.

(3) Figure 1A, it would be helpful to indicate the location of the toiL (red arrow as in Figure 2A) relative to the putative rut site early in the beginning of the results. Does TSS mark the transcription start site? There is no annotation of TSS in the figure legend. Was TSS previously mapped experimentally? Please include relevant citations.

We now indicate the position of the TSS relative to the topAI start codon. Similarly, we indicate the position of the start of toiL relative to the topAI start codon in Figure 2A. We now explain “TSS” in the figure legend. There is a reference in the text for the TSS (Thomason et al., 2015).

(4) Please consider rearranging the results section, perhaps more helpful to introduce the toiL in Figure 1 or earlier. The current format requires readers to switch back-and-forth between Figure 4 and Figure 2.

We have added a schematic of the topAI upstream region as Figure 1A, and we have separated Figure 2C as described in a response to a comment from Reviewer 2.

(5) Figure 2A and Figure 2-Figure Suppl 1A, for clarity, please mark the rut site upstream of the red arrow.

Rather than mark the rut on Figure 2A, which would make for a busy schematic, readers can compare the positions of the rut to those of toiL, which we have now added to Figures 1B (formerly Figure 1A) and 2A.

(6) The following conclusion seems speculative: "...but does not trigger termination until RNAP ..., >180 nt further downstream…". Shouldn't the authors already know where the termination site is based on their previous Term-seq data (see Ref 1, Adams PP et al 2021)?

Sites of Rho-dependent transcription termination cannot be mapped precisely from Term-seq data because exoribonucleases rapidly process the unstructured RNA 3’ ends.

(7) Genetic screen: Please discuss why the 23S rRNA mutations that cause translational infidelity could promote topAI translation. Wouldn't the mutant ribosome be affected in translating toiL?

See response to Reviewer 1’s comment.

(8) Although antibiotic concentrations were provided in Figure 2 legend, please provide the MIC values of each antibiotic, e.g., in Table S2, for the tested E. coli strain, to inform readers how specific subinhibitory concentrations were chosen.

See response to Reviewing Editor.

(9) Please clarify the calculation of luciferase units in the y-axis of Figure 2A, why the scale is drastically higher than that of Figure 7C using the same antibiotics?

These reporter assays use different constructs. The reporter construct used for experiments in Figure 7 includes a portion of the ermCL gene and associated downstream sequence. We have enlarged Figure 7A to highlight the difference in reporter constructs.

(10) Table S4 needs a few more details. It is unclear how those numbers in columns G-H were generated. Do those numbers correspond to ribosome density per nt/ORF?

We have added footnotes to Table S4 to indicate that the numbers in columns G and H represent sequence read coverage normalized by region length and by the upper quartile of gene expression.

(11) Figure 5, if the SHAPE results were true, the Shine Dalgarno sequence of toiL is sequestered in the hairpin structure with and without tetracycline treatment. It is inconceivable that translational initiation will occur efficiently, please discuss.

Our representation of the SHAPE-seq data was confusing since we overlayed the SHAPE-seq changes on a predicted structure that likely corresponds to the uninduced state. We hope that the new version of Figure 5 is clearer.

We presume the reviewer is referring to the Shine-Dalgarno sequence of topAI rather than toiL, since the Shine-Dalgarno sequence of toiL is predicted to be unstructured even in the absence of tetracycline treatment. The ribosome-binding site of topAI is more accessible in cells treated with tetracycline, although the SHAPE-seq data suggest that this is a transient event. The binding of the initiating ribosome may also reduce reactivity in this region under inducing conditions. We now discuss this briefly in the text.