Rescue of Escherichia coli auxotrophy by de novo small proteins
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
This manuscript describes randomly generated small proteins of <50 amino acids that can rescue the growth of an auxotrophic mutant of Escherichia coli. The authors suggest that these proteins function by binding specifically to a regulatory element in the 5' UTR of the his operon RNA, altering RNA structure to increase expression. The study suggests that functional small proteins can evolve de novo and that newly evolved small proteins can function as regulators by binding RNA. This is an exciting idea, but the suggested mechanism involving the binding of the small proteins to RNA requires additional experimental support.
(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. Reviewer #1 and Reviewer #3 agreed to share their name with the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Evolutionary Biology (eLife)
Abstract
Increasing numbers of small proteins with diverse physiological roles are being identified and characterized in both prokaryotic and eukaryotic systems, but the origins and evolution of these proteins remain unclear. Recent genomic sequence analyses in several organisms suggest that new functions encoded by small open reading frames (sORFs) may emerge de novo from noncoding sequences. However, experimental data demonstrating if and how randomly generated sORFs can confer beneficial effects to cells are limited. Here, we show that by upregulating hisB expression, de novo small proteins (≤50 amino acids in length) selected from random sequence libraries can rescue Escherichia coli cells that lack the conditionally essential SerB enzyme. The recovered small proteins are hydrophobic and confer their rescue effect by binding to the 5′ end regulatory region of the his operon mRNA, suggesting that protein binding promotes structural rearrangements of the RNA that allow increased hisB expression. This study adds RNA regulatory elements as another interacting partner for de novo proteins isolated from random sequence libraries and provides further experimental evidence that small proteins with selective benefits can originate from the expression of nonfunctional sequences.
Article activity feed
-
-
Author Response
Reviewer #1 (Public Review):
The authors screen large libraries of small proteins to identify three proteins of <50 aa that rescue the growth of an auxotrophic serB deletion Escherichia coli strain. They convincingly show that the growth rescue is due to the small proteins increasing expression of the his operon by reducing transcriptional attenuation. The authors argue that the small proteins function by directly binding the his RNA 5' UTR to alter RNA secondary structure.
The conclusion that the three small proteins reduce his operon attenuation is well supported by the data. A previous study suggested this mechanism for a somewhat larger, randomly selected protein, but the current study extends this prior work by firmly establishing that the proteins modulate attenuation. The suggestion that the small proteins …
Author Response
Reviewer #1 (Public Review):
The authors screen large libraries of small proteins to identify three proteins of <50 aa that rescue the growth of an auxotrophic serB deletion Escherichia coli strain. They convincingly show that the growth rescue is due to the small proteins increasing expression of the his operon by reducing transcriptional attenuation. The authors argue that the small proteins function by directly binding the his RNA 5' UTR to alter RNA secondary structure.
The conclusion that the three small proteins reduce his operon attenuation is well supported by the data. A previous study suggested this mechanism for a somewhat larger, randomly selected protein, but the current study extends this prior work by firmly establishing that the proteins modulate attenuation. The suggestion that the small proteins function by directly binding the his RNA is less well supported by the data. The RNase T1 mapping data are not straightforward to interpret, and there is no assessment of protein-RNA interactions in vivo.
Major comments:
- The RNase T1 probing data are not straightforward to interpret, and hence are insufficient to conclude that Hdp1 binding to the his 5' UTR is the mechanism by which it reduces attenuation. Specifically, G96 has reduced cleavage in the presence of Hdp1, inconsistent with the antiterminator conformation. The authors argue that G96 could be within the site of Hdp1 binding. This is certainly possible but would require additional experimental evidence to draw a confident conclusion. Also, the increased cleavage of bases around the start codon and Shine-Dalgarno sequence is inconsistent with a shift from the terminator to the antiterminator conformation. One confounding issue here is the lack of replicates and the lack of quantification. Additional probes could be tested, which would provide complementary structural information.
We agree that the RNase T1 probing data alone does not provide sufficient resolution to fully assess changes in terminator/anti-terminator conformations. Therefore, we have clarified our interpretation of the data, addressed its limitations, and have softened the conclusions that can be drawn from it in the text (lines 419-431). We have also included two additional T1 probing experimental replicates in Supplementary Fig. S11 which are in agreement with the cleavage patterns presented in the main text Figure 3D. Based on the revised conclusions and the consistency of the cleavage patterns between the experimental replicates, we do not think that quantification of the probing data would provide any additional information.
- There are no experiments to test whether Hdp1 binds the his RNA in vivo. The in vitro data show that Hdp1 can bind the his RNA, but they do not show that this occurs in vivo, or that this is the mechanism by which Hdp1 regulates the expression of the his operon.
As addressed in the Essential Revisions section, we have now performed and included data from co- immunoprecipitation assays, in which we were able to successfully detect and demonstrate enrichment of his operator-regulated RNA transcripts in HA-tagged Hdp1 pull-down samples. We were also able to demonstrate less enrichment (i.e. reduced interaction/specificity) for thr operator-regulated RNA transcripts in the Hdp1 pull-downs as well as lower enrichment for all his operator-regulated target RNA transcripts in pull-downs performed with the HA-tagged Hdp1 L27Q mutant. These data are presented in Fig. 3A and discussed in lines 313-337.
Reviewer #2 (Public Review):
In this work, Babina et al. address a central question in molecular evolution that is only partially answered: how does cellular novelty emerge in evolution? The authors focus here on small proteins, whose importance to various cellular functions has become more appreciated recently. Babina et al. ask if functional small proteins can emerge from random sequences, a question that is mostly unresolved with only a small number of examples in the published literature for such functions. In this study, the authors demonstrate that proteins selected from random, synthetic libraries can rescue auxotrophy in E. coli. Namely, the authors find three small, random proteins (<50 amino acids) that allow E. coli cells with a ΔserB genetic background to grow in a medium without the amino-acid serine. They then show that this rescue is based on the up-regulation of HisB, an enzyme that can compensate for the serB deletion. Finally, using different molecular biology techniques, the authors propose a model in which up-regulation of HisB is achieved by physical interactions between the random proteins and the his operator that regulates the transcription of the his operon in E. coli.
Notably, as the authors themselves point out, a previous study has already shown that semi-random proteins can result in up-regulation of HisB levels to rescue ΔserB cells. Thus, most of the novelty comes from the attempt to figure out the molecular mechanism of the three random proteins. The idea that a random protein binds the 5' of an mRNA which results in up-regulated expression levels is interesting and can benefit the field. However, some clarification on existing data and additional control experiments are needed to support the authors' claims:
- Growth data are not presented in the current form of the manuscript, which makes it impossible to evaluate many of its claims. Especially, the extent of rescue and fitness gain achieved by these random proteins compared to cells harboring the serB gene.
We thank the reviewer for pointing out this discrepancy. We have now added all relevant growth data under non-permissive conditions (Figure 1G, Supplementary Figures S2, S3, S5) and have also included data on the fitness effects exerted by Hdp expression in cells harboring serB under permissive conditions (LB medium), to allow for comparison with the empty plasmid control strain (Supplementary Figure S1).
- The authors have screened their library on other auxotrophic strains, however, they could only find random proteins that rescue growth in the ΔserB background. Currently, they do not address this point, but it might be relevant to the molecular mechanism of those random proteins.
The reviewer raises an interesting point. We have added a paragraph to our Discussion addressing why we believe that the serB-model with a complementary enzyme is an ideal target for the selection of de novo genes (lines 536-543).
- Central to the authors' claims is the up-regulation of HisB, however, they mostly work with an alternative LacZ system to assess the effects of the random proteins on expression. The paper will benefit from some more work measuring actual HisB levels as expressed by the various constructs used along the paper. The authors did provide an important proteomic analysis to show that HisB (along with other proteins in the his operon) is up- regulated as a result of the expression of one of the random proteins. However, it is unclear if the reported ~3- fold increase in HisB levels is enough to allow the growth of ΔserB cells in a medium without serine.
We thank the reviewer for raising this concern and allowing the opportunity to clarify. It is well established that upregulation of HisB can rescue growth of a SerB-deficient strain on minimal medium (for examples, see Patrick, et al. PMID: 17884825, Digianantonio and Hecht PMID: 26884172). We have now performed additional proteomics analyses that show a specific upregulation of the his operon upon expression of Hdp1 and Hdp3. We have also added a control experiment overexpressing HisB from our expression vector, showing that it restores growth of the auxotrophic ΔserB mutant. It is also clear that histidine starvation itself does not de-repress HisB sufficiently to allow growth of a ΔserB mutant, as this strain does not grow on minimal medium lacking histidine (such as M9 minimal medium that was used for the functional selection in our study). In addition to upregulation of HisB, we show that the rescue is dependent on presence of HisB and provide additional experiments showing a specific interactions in vitro and in vivo of Hdp1 with the his operator RNA. Our results clearly show that rescue depends on HisB and that Hdp expression upregulates HisB, and we do believe our central claim is substantiated beyond reasonable doubt. The reviewer’s main concern, that it is unclear if expression levels of HisB are high enough to allow growth is, in our opinion, resolved by the observation that Hdp-dependent upregulation of HisB does restore growth.
We respectfully disagree with the reviewer’s suggestion that an exact determination of the level of upregulation is relevant and needed, as outlined above. In addition, we would like to point out that it is not possible to measure HisB upregulation compared to an empty plasmid control strain under non- permissive conditions. Comparing HisB levels in a ΔserB strain expressing Hdp vs. the empty plasmid control in minimal medium is not possible, since the empty plasmid control strain is not able to grow, and the corresponding baseline of HisB expression cannot be determined in a non-growing strain. To circumvent this, we determined HisB levels in rich medium, which does not necessarily reflect the exact amount of upregulation occurring under non-permissive conditions, but still allows us to detect a physiological activity. Alternative experimental setups, such as comparing HisB levels in a strain carrying serB in minimal medium also suffer severe shortcomings as it no longer reflects the cellular physiology of the auxotoph under non-permissive conditions, where growth is dependent on HisB upregulation.
- It is unclear how noisy and statistically significant some of the critical experiments in the manuscript are, especially the EMSA and T1-digestion experiments. The authors should try to find a different operator with a similar RNA structure and attenuation function, but a different nucleotide sequence, to the his operator, and show that this control operator is unaffected by the random proteins. Demonstrating the lack of phenotypes using the LacZ system, EMSA experiments, and T1-digestion patterns will much support the authors' claims.
We thank the reviewer for suggesting this important control and agree that its inclusion significantly strengthens our claims. We used the threonine operon (thr) operator, which is regulated by terminator/anti-terminator formation similar to that of to the his operon with the his operator. We show that Hdp1 does not cause de-repression of this operator using a lacZ reporter construct. Strongly supporting this is the fact that our whole proteome analysis showed specific upregulation of the his operon. Any other off target de-repression would be detected in this assay. Furthermore, we now include the thr operator RNA as a control in the EMSAs, which demonstrates reduced binding with Hdp1 in comparison to the his operator RNA. We also added an in vivo pull-down experiment using tagged Hdp1, showing marked enrichment of his operator-regulated RNA transcripts, whereas the observed enrichment of the control thr RNA transcripts is substantially less.
-
Evaluation Summary:
This manuscript describes randomly generated small proteins of <50 amino acids that can rescue the growth of an auxotrophic mutant of Escherichia coli. The authors suggest that these proteins function by binding specifically to a regulatory element in the 5' UTR of the his operon RNA, altering RNA structure to increase expression. The study suggests that functional small proteins can evolve de novo and that newly evolved small proteins can function as regulators by binding RNA. This is an exciting idea, but the suggested mechanism involving the binding of the small proteins to RNA requires additional experimental support.
(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. Reviewer #1 and …
Evaluation Summary:
This manuscript describes randomly generated small proteins of <50 amino acids that can rescue the growth of an auxotrophic mutant of Escherichia coli. The authors suggest that these proteins function by binding specifically to a regulatory element in the 5' UTR of the his operon RNA, altering RNA structure to increase expression. The study suggests that functional small proteins can evolve de novo and that newly evolved small proteins can function as regulators by binding RNA. This is an exciting idea, but the suggested mechanism involving the binding of the small proteins to RNA requires additional experimental support.
(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. Reviewer #1 and Reviewer #3 agreed to share their name with the authors.)
-
Reviewer #1 (Public Review):
The authors screen large libraries of small proteins to identify three proteins of <50 aa that rescue the growth of an auxotrophic serB deletion Escherichia coli strain. They convincingly show that the growth rescue is due to the small proteins increasing expression of the his operon by reducing transcriptional attenuation. The authors argue that the small proteins function by directly binding the his RNA 5' UTR to alter RNA secondary structure.
The conclusion that the three small proteins reduce his operon attenuation is well supported by the data. A previous study suggested this mechanism for a somewhat larger, randomly selected protein, but the current study extends this prior work by firmly establishing that the proteins modulate attenuation. The suggestion that the small proteins function by directly …
Reviewer #1 (Public Review):
The authors screen large libraries of small proteins to identify three proteins of <50 aa that rescue the growth of an auxotrophic serB deletion Escherichia coli strain. They convincingly show that the growth rescue is due to the small proteins increasing expression of the his operon by reducing transcriptional attenuation. The authors argue that the small proteins function by directly binding the his RNA 5' UTR to alter RNA secondary structure.
The conclusion that the three small proteins reduce his operon attenuation is well supported by the data. A previous study suggested this mechanism for a somewhat larger, randomly selected protein, but the current study extends this prior work by firmly establishing that the proteins modulate attenuation. The suggestion that the small proteins function by directly binding the his RNA is less well supported by the data. The RNase T1 mapping data are not straightforward to interpret, and there is no assessment of protein-RNA interactions in vivo.
Major comments:
1. The RNase T1 probing data are not straightforward to interpret, and hence are insufficient to conclude that Hdp1 binding to the his 5' UTR is the mechanism by which it reduces attenuation. Specifically, G96 has reduced cleavage in the presence of Hdp1, inconsistent with the antiterminator conformation. The authors argue that G96 could be within the site of Hdp1 binding. This is certainly possible but would require additional experimental evidence to draw a confident conclusion. Also, the increased cleavage of bases around the start codon and Shine-Dalgarno sequence is inconsistent with a shift from the terminator to the antiterminator conformation. One confounding issue here is the lack of replicates and the lack of quantification. Additional probes could be tested, which would provide complementary structural information.
2. There are no experiments to test whether Hdp1 binds the his RNA in vivo. The in vitro data show that Hdp1 can bind the his RNA, but they do not show that this occurs in vivo, or that this is the mechanism by which Hdp1 regulates the expression of the his operon.
-
Reviewer #2 (Public Review):
In this work, Babina et al. address a central question in molecular evolution that is only partially answered: how does cellular novelty emerge in evolution? The authors focus here on small proteins, whose importance to various cellular functions has become more appreciated recently. Babina et al. ask if functional small proteins can emerge from random sequences, a question that is mostly unresolved with only a small number of examples in the published literature for such functions. In this study, the authors demonstrate that proteins selected from random, synthetic libraries can rescue auxotrophy in E. coli. Namely, the authors find three small, random proteins (<50 amino acids) that allow E. coli cells with a ΔserB genetic background to grow in a medium without the amino-acid serine. They then show that …
Reviewer #2 (Public Review):
In this work, Babina et al. address a central question in molecular evolution that is only partially answered: how does cellular novelty emerge in evolution? The authors focus here on small proteins, whose importance to various cellular functions has become more appreciated recently. Babina et al. ask if functional small proteins can emerge from random sequences, a question that is mostly unresolved with only a small number of examples in the published literature for such functions. In this study, the authors demonstrate that proteins selected from random, synthetic libraries can rescue auxotrophy in E. coli. Namely, the authors find three small, random proteins (<50 amino acids) that allow E. coli cells with a ΔserB genetic background to grow in a medium without the amino-acid serine. They then show that this rescue is based on the up-regulation of HisB, an enzyme that can compensate for the serB deletion. Finally, using different molecular biology techniques, the authors propose a model in which up-regulation of HisB is achieved by physical interactions between the random proteins and the his operator that regulates the transcription of the his operon in E. coli.
Notably, as the authors themselves point out, a previous study has already shown that semi-random proteins can result in up-regulation of HisB levels to rescue ΔserB cells. Thus, most of the novelty comes from the attempt to figure out the molecular mechanism of the three random proteins. The idea that a random protein binds the 5' of an mRNA which results in up-regulated expression levels is interesting and can benefit the field. However, some clarification on existing data and additional control experiments are needed to support the authors' claims:
Growth data are not presented in the current form of the manuscript, which makes it impossible to evaluate many of its claims. Especially, the extent of rescue and fitness gain achieved by these random proteins compared to cells harboring the serB gene.
The authors have screened their library on other auxotrophic strains, however, they could only find random proteins that rescue growth in the ΔserB background. Currently, they do not address this point, but it might be relevant to the molecular mechanism of those random proteins.
Central to the authors' claims is the up-regulation of HisB, however, they mostly work with an alternative LacZ system to assess the effects of the random proteins on expression. The paper will benefit from some more work measuring actual HisB levels as expressed by the various constructs used along the paper. The authors did provide an important proteomic analysis to show that HisB (along with other proteins in the his operon) is up-regulated as a result of the expression of one of the random proteins. However, it is unclear if the reported ~3-fold increase in HisB levels is enough to allow the growth of ΔserB cells in a medium without serine.
It is unclear how noisy and statistically significant some of the critical experiments in the manuscript are, especially the EMSA and T1-digestion experiments. The authors should try to find a different operator with a similar RNA structure and attenuation function, but a different nucleotide sequence, to the his operator, and show that this control operator is unaffected by the random proteins. Demonstrating the lack of phenotypes using the LacZ system, EMSA experiments, and T1-digestion patterns will much support the authors' claims.
-
Reviewer #3 (Public Review):
In this paper, the authors investigate the evolutionary origin of small proteins. These proteins could arise from the de novo evolution of new genes through mutations that create a new short open reading frame, or through the truncation of once larger protein-encoding open reading frame. The authors seek to demonstrate that small proteins that are accidentally translated from a randomly-occurring sequence could confer evolutionary advantage, which would then fix the new gene in the species' genome if that evolutionary pressure persists. The authors seek to replicate these conditions by generating a random library of short protein-encoding sequences and find sequences that could rescue the ability of E. coli to grow in an otherwise unfavorable environment. In this case, the authors use an E. coli serB …
Reviewer #3 (Public Review):
In this paper, the authors investigate the evolutionary origin of small proteins. These proteins could arise from the de novo evolution of new genes through mutations that create a new short open reading frame, or through the truncation of once larger protein-encoding open reading frame. The authors seek to demonstrate that small proteins that are accidentally translated from a randomly-occurring sequence could confer evolutionary advantage, which would then fix the new gene in the species' genome if that evolutionary pressure persists. The authors seek to replicate these conditions by generating a random library of short protein-encoding sequences and find sequences that could rescue the ability of E. coli to grow in an otherwise unfavorable environment. In this case, the authors use an E. coli serB auxotrophic mutant grown in minimal media since the serB mutant strain cannot grow in minimal media because it is incapable of synthesizing serine. They identified three small protein sequences that allowed survival on minimal media. The authors use classic genetics experiments (deletions and reporter fusions) and modern proteomics approaches to uncover that rescue likely occurs via the increased expression of the HisB protein, which is able to compensate for the serB mutation and rescue serine biosynthesis. Biochemical experiments demonstrate that this small protein can bind to the operator of the his operon, causing structural changes in the terminator that may affect the expression of operon-encoded proteins.
The general approach described in this paper was previously successfully used by this research group in two other publications to identify small protein sequences that affect other E. coli phenotypes. Overall, the experiments were well-designed and thorough and support the authors' claims. However, the inclusion of some data that were discussed in the text but not shown and the inclusion of some additional control experiments would strengthen the author's conclusions.
These findings not only support the idea that randomly expressed short protein sequences could provide evolutionary advantages to an organism but also suggest the existence of a class of small proteins that regulate gene expression by directly binding to mRNA. To my knowledge, this function has not been reported for any of the naturally encoded small proteins and it would be very exciting to observe this mechanism occurring in nature. Given the wealth of sORFs that have been newly identified and the dearth of characterized sORF products, it is likely that at least one such small protein exists.
-
