Escherichia coli Aggregates Mediated by Native or Synthetic Adhesins Exhibit Both Core and Adhesin-Specific Transcriptional Responses

ABSTRACT Bacteria can rapidly tune their physiology and metabolism to adapt to environmental fluctuations. In particular, they can adapt their lifestyle to the close proximity of other bacteria or the presence of different surfaces. However, whether these interactions trigger transcriptomic responses is poorly understood. We used a specific setup of E. coli strains expressing native or synthetic adhesins mediating bacterial aggregation to study the transcriptomic changes of aggregated compared to nonaggregated bacteria. Our results show that, following aggregation, bacteria exhibit a core response independent of the adhesin type, with differential expression of 56.9% of the coding genome, including genes involved in stress response and anaerobic lifestyle. Moreover, when aggregates were formed via a naturally expressed E. coli adhesin (antigen 43), the transcriptomic response of the bacteria was more exaggerated than that of aggregates formed via a synthetic adhesin. This suggests that the response to aggregation induced by native E. coli adhesins could have been finely tuned during bacterial evolution. Our study therefore provides insights into the effect of self-interaction in bacteria and allows a better understanding of why bacterial aggregates exhibit increased stress tolerance. IMPORTANCE The formation of bacterial aggregates has an important role in both clinical and ecological contexts. Although these structures have been previously shown to be more resistant to stressful conditions, the genetic basis of this stress tolerance associated with the aggregate lifestyle is poorly understood. Surface sensing mediated by different adhesins can result in various changes in bacterial physiology. However, whether adhesin-adhesin interactions, as well as the type of adhesin mediating aggregation, affect bacterial cell physiology is unknown. By sequencing the transcriptomes of aggregated and nonaggregated cells expressing native or synthetic adhesins, we characterized the effects of aggregation and adhesin type on E. coli physiology.

The article raises an interesting question and takes an imaginative approach to answering it.
My biggest criticism of the study comes from the way the bacteria are grown and harvested for RNA purification and transcriptome analysis. The authors grow the bacteria in static conditions and purify the aggregates that settle to the bottom of the tube. Non-aggregating bacteria also settle to the bottom, but to a lesser extent, and given that the bacteria are grown to an OD of 0.5, the difference in time spent at the bottom of the tube by aggregating bacteria compared to non-aggregating bacteria may influence the results obtained. I understand that Ag43-mediated and also nanobody induced aggregation also occurs when bacteria are incubated under agitated conditions. If this is the case, it would be interesting to compare the transcriptome of the same bacterial strains grown under shaking conditions. This would help to know whether the results obtained are influenced by the anoxic conditions (microaerophilia) that cause sedimentation and are inherent to the aggregation process.
Reviewer #2 (Comments for the Author): Chekli et al. present a manuscript examining the effects of autoaggregation of bacteria on the physiology of the bacteria within aggregates. They do this elegantly by transcriptomics, cleverly using a separation funnel to isolate aggregated bacteria. The results show how the bacteria become more resistant to a variety of stresses, a phenomenon that has been known for bacterial aggregates for some time. The authors further examine the effects of native (Ag43) and synthetic (nanobody-based) autoagglutinins and find that the native autoagglutinin shows an enhanced response. As the authors rightly point out, how general this enhanced response is remains to be seen. Other open questions include finding out how the autoaggregation signal is transduced from the cell surface to elicit the transcriptional responses. Another question is where in the autoaggregates these changes take place -throughout or mainly towards the centre of the aggregate. These questions are clearly beyond the scope of the current manuscript, which is well written and nicely presented. I only have some minor comments to address: 1. The authors could provide an explanation for why amikacin was specifically chosen for the survival assays. 2. In figure 1, the authors should follow the nomenclature suggested by Drobnak et al. (2015) regarding autotransporter regions. So 'passenger' rather than alpha domain, etc. The authors should also note that Intimin includes a D00 domain between the beta-barrel and the D0 domain (see Weikum et al., 2020). 3. The authors used spent medium to prevent growth during their autoaggregation experiments. However, spent medium may contain compounds that are toxic or have an effect on cellular physiology (e.g. Quorum sensing molecules). The authors do of course have controls, but even so, they could justify the use of spent medium over say PBS, which would also prevent growth. 4. For the microscopy to determine aggregate size, it is not clear to me how the cells were made fluorescent. Please add this information.
Staff Comments:

Preparing Revision Guidelines
To submit your modified manuscript, log onto the eJP submission site at https://spectrum.msubmit.net/cgi-bin/main.plex. Go to Author Tasks and click the appropriate manuscript title to begin the revision process. The information that you entered when you first submitted the paper will be displayed. Please update the information as necessary. Here are a few examples of required updates that authors must address: • Point-by-point responses to the issues raised by the reviewers in a file named "Response to Reviewers," NOT IN YOUR COVER LETTER. • Upload a compare copy of the manuscript (without figures) as a "Marked-Up Manuscript" file. • Each figure must be uploaded as a separate file, and any multipanel figures must be assembled into one file. For complete guidelines on revision requirements, please see the journal Submission and Review Process requirements at https://journals.asm.org/journal/Spectrum/submission-review-process. Submissions of a paper that does not conform to Microbiology Spectrum guidelines will delay acceptance of your manuscript. " Please return the manuscript within 60 days; if you cannot complete the modification within this time period, please contact me. If you do not wish to modify the manuscript and prefer to submit it to another journal, please notify me of your decision immediately so that the manuscript may be formally withdrawn from consideration by Microbiology Spectrum.
If your manuscript is accepted for publication, you will be contacted separately about payment when the proofs are issued; please follow the instructions in that e-mail. Arrangements for payment must be made before your article is published. For a complete list of Publication Fees, including supplemental material costs, please visit our website.

Reviewer #1 (Comments for the Author):
The article by Chekli et al. studies how aggregation alters gene expression in E. coli. To do this, they design two alternative aggregation systems: one based on the constitutive expression of the main E. coli adhesin (ag43), and another based on the expression in some bacteria of an antigen and in other cells of a nanobody that recognises this antigen. The aggregates formed by these bacteria are compared at the transcriptomic level with bacteria that do not express Ag43 or that express a nanobody and an antigen that is not recognised by the nanobody. The results show much more expression changes when aggregation is induced by the native protein than when bacteria aggregate expressing the nanobody. The article raises an interesting question and takes an imaginative approach to answering it.

Response:
Thank you to the reviewer for the positive feedback.
My biggest criticism of the study comes from the way the bacteria are grown and harvested for RNA purification and transcriptome analysis. The authors grow the bacteria in static conditions and purify the aggregates that settle to the bottom of the tube. Non-aggregating bacteria also settle to the bottom, but to a lesser extent, and given that the bacteria are grown to an OD of 0.5, the difference in time spent at the bottom of the tube by aggregating bacteria compared to non-aggregating bacteria may influence the results obtained. I understand that Ag43-mediated and also nanobody induced aggregation also occurs when bacteria are incubated under agitated conditions. If this is the case, it would be interesting to compare the transcriptome of the same bacterial strains grown under shaking conditions. This would help to know whether the results obtained are influenced by the anoxic conditions (microaerophilia) that cause sedimentation and are inherent to the aggregation process.

Response:
The bacteria were not grown in static conditions but in shaking conditions (180 rpm). We apologize because this was not clearly indicated, and we have now added this information in the M&M section (see below) Line 587: "Bacterial cultures were grown overnight in LB medium at 37°C under agitation (180 rpm). This is true that aggregation can occur when cells are under agitation, but this aggregation remains limited, and in this condition (without settlement) it would have been impossible to properly collect the aggregating cells. In our hands, clearly, the aggregation is maximal when bacteria are left to settle for some time under static conditions. This is why we chose to limit our analysis on bacteria grown under agitation then left to settle for 3h. In this condition, the deposition of non-aggregative cells due to gravity is very limited, if not inexistent, as shown in the funnels presented in Figure S3A compared to aggregative cells in Figure S3B.
Since the sampling of both aggregating and non-aggregating cells have been performed in a funnel, if some sedimentation of non-aggregating cells occurred, then the sedimented cells have also been collected. Consequently, our analysis reflects the transcriptional response of actively aggregating cells as compared to (very limited) gravity sedimented cells.
We have also added more details on these different aspects in the M&M section: Line 608-614: "Bacteria were grown at 37°C under agitation (180 rpm) in LB medium until OD 600 = 0.5, then were transferred to separating funnels and left at 37°C for 3 hours in static conditions (Supplementary Fig. S3). In these conditions, the expected aggregation of cells was strong while gravity sedimentation of non-aggregating cells was very limited (supplementary Fig. S3). Then, by opening the tap of the separating funnel, we collected 1mL of the lower part of the culture, corresponding to the aggregated cells for strains forming aggregates or to the sedimented cells for non-aggregating strains. »

Reviewer #2 (Comments for the Author):
Chekli et al. present a manuscript examining the effects of autoaggregation of bacteria on the physiology of the bacteria within aggregates. They do this elegantly by transcriptomics, cleverly using a separation funnel to isolate aggregated bacteria. The results show how the bacteria become more resistant to a variety of stresses, a phenomenon that has been known for bacterial aggregates for some time. The authors further examine the effects of native (Ag43) and synthetic (nanobody-based) autoagglutinins and find that the native autoagglutinin shows an enhanced response. As the authors rightly point out, how general this enhanced response is remains to be seen. Other open questions include finding out how the autoaggregation signal is transduced from the cell surface to elicit the transcriptional responses. Another question is where in the autoaggregates these changes take placethroughout or mainly towards the centre of the aggregate. These questions are clearly beyond the scope of the current manuscript, which is well written and nicely presented. I only have some minor comments to address:

Response:
Thank you to the reviewer for the positive feedback and for taking up the questions that our work raises.
1. The authors could provide an explanation for why amikacin was specifically chosen for the survival assays.

Response:
To validate that the aggregates formed in our study display the characteristic enhanced survival to external stress, we decided to use a bactericidal antibiotic. 2. In figure 1, the authors should follow the nomenclature suggested by Drobnak et al. (2015) regarding autotransporter regions. So 'passenger' rather than alpha domain, etc. The authors should also note that Intimin includes a D00 domain between the beta-barrel and the D0 domain (see Weikum et al., 2020).

Response:
We have modified the nomenclature accordingly.
3. The authors used spent medium to prevent growth during their autoaggregation experiments. However, spent medium may contain compounds that are toxic or have an effect on cellular physiology (e.g. Quorum sensing molecules). The authors do of course have controls, but even so, they could justify the use of spent medium over say PBS, which would also prevent growth.

Response:
We used spent medium to dilute the cultures when evaluating the capacity of aggregation of cells carrying the different constructions ( Fig. 1C aggregation curves). Despite the possible presence of some molecules that can impact on bacterial physiology we consider that this is still more physiological than resuspending the cells in PBS. This has been justified in the text on line 594: 1 mL of overnight cultures for each strain were diluted to OD 600 = 3 in spent LB to prevent growth during the experiment under physiologically relevant conditions. However, dilution in spent medium was not performed for the main experiments of this manuscript where we collected the aggregated vs non-aggregated cells and performed the transcriptomic analysis. In this later case aggregate formation occurred directly in cultures at OD= 0.5 to avoid any interference or influence of non-desired molecules.
4. For the microscopy to determine aggregate size, it is not clear to me how the cells were made fluorescent. Please add this information.

Response:
The aggregates were observed using an epifluorescence microscope but using the phase contrast mode. The cells used were not made fluorescent.
We have now added this information in the M& M section: "line 637-638: "20 pictures of each type of aggregates were then taken using epifluorescence microscope with the phase contrast mode (EVOS M7000, Invitrogen). Images were then analyzed using FIJI software (Version 2.9.0) ».