Usage of Cultured Human Fecal Microbiota for Colonization of Caenorhabditis elegans to Study Host–Microbe Interaction

Round 1

Reviewer 1 Report

 

 

This manuscript discusses an interesting topic in the field, specifically with the use of non-mammalian, easy to perform, and high throughput, in vivo model with the use of C. elegans. Authors tried to culture fecal slurry, then grow worms with this nutrient source instead of traditional Ecoli OP50. They noticed a beneficial effect on worm growth and egg hatching. Finally, they proposed this model could well be used for models using FMT strategies.

 

 

Overall there was no attempt to grow fecal slurry under anaerobic conditions. Overall, it seems this model simply reflect the aerotolerant part of the gut microbiota.That is quite well illustrated by the 16S analsis in Fig1. Boacteroidota (mostly anaerobes) are lost in the culture overnight and the remaining Firmicutes are rather Enterococcus.

 Is there anything known on the interaction of nematodes with human anaerobic strains ?

 

I am very dubious regarding experimental condition regarding the total amount of diversity richness in the samples. Please provide additional figures on the TOP15/TOP30 at the Family rank. At this step of reviewing, I have not reprocessed ( and I have not the time to dedicate to that) the entire raw dataset but the results are surprising enough to justify. Please provide additional information regarding theworkflow for giving indication on validity of the ASVs annotation.

 

Regarding worm phenotyping. Results are interesting, suggesting a nutritional effect of complex microbiota on worms, compared to OP50 only. Isn't that suspected in essence? Is the food source related or not to the viability of overall bacteria ? There is another phenotype that I would have been very interested to see, is specifically the "dormant" stage or dauer. Other reports in the literature with authors using complex microbiota to feed worms , seemed to reveal some interesting findings in this regard. Please comment.

 

Related to last part of the findings : the feasibility of FMT with worms.  What about doing FMT with a "pro-inflammatory" / "dysbiotic" microbiota in this model to see whether or not it can really be used as a model for causative mechanistic studies ? This question is not a dirty one, but is related to the fact that most cultured bacteria in the model are rather pathobionts strains in the context of intestinal inflammation (see. E.coli/Shigella or Pseudomonas).

 

Discussion is quite surprising in its overall organization. That said, it is short. However, it is not enough related with the findings specifically, and the limitations of this model/study did not seem to be well acknowledged.

 

Minor :

line 118: It is unclear the meaning of "blood donor"

 

The color legend in Fig1 panel G is confusing. I expect this Figure to be the same order as previously : Stock then ON cultures. Unless there is a confusion in the graphic as presented? Please clarify

 

In addition to the Cheng study (ref13), even more promising results have been put forward on the OligoMM12 (12 members) or Oligomm19 (19) models. this is even more reduced and still promote colonization resistance against known enteropathogens.

Author Response

This manuscript discusses an interesting topic in the field, specifically with the use of non-mammalian, easy to perform, and high throughput, in vivo model with the use of C. elegans. Authors tried to culture fecal slurry, then grow worms with this nutrient source instead of traditional Ecoli OP50. They noticed a beneficial effect on worm growth and egg hatching. Finally, they proposed this model could well be used for models using FMT strategies.

Overall there was no attempt to grow fecal slurry under anaerobic conditions. Overall, it seems this model simply reflect the aerotolerant part of the gut microbiota. That is quite well illustrated by the 16S analsis in Fig1. Boacteroidota (mostly anaerobes) are lost in the culture overnight and the remaining Firmicutes are rather Enterococcus. Is there anything known on the interaction of nematodes with human anaerobic strains ?

We agree that studying anaerobic conditions would be interesting to the extent that this is possible given that the worms cannot be cultured anaerobically. In fact, we have already commented on this in the text. This would be an interesting follow-up study but beyond the scope of the current manuscript.

I am very dubious regarding experimental condition regarding the total amount of diversity richness in the samples. Please provide additional figures on the TOP15/TOP30 at the Family rank. At this step of reviewing, I have not reprocessed (and I have not the time to dedicate to that) the entire raw dataset but the results are surprising enough to justify. Please provide additional information regarding theworkflow for giving indication on validity of the ASVs annotation.

We have included TOP15 at the family rank in the supplemental material.

We have already given clear references to the workflow used in the materials and methods section. The raw sequencing data were summarized into Zero-radius Operational Taxonomic Unit (ZOTU), commonly referred to as amplicon sequence variants (ASVs), using an in-house pipeline, AmpProc v5.1.0.beta2.11.0 [33] which operates primarily on the USEARCH11 [34] workflow. The ASVs were assigned taxonomy using the ribosomal RNA database, SILVA v138 and with 99% sequence identity.

Regarding worm phenotyping. Results are interesting, suggesting a nutritional effect of complex microbiota on worms, compared to OP50 only. Isn't that suspected in essence? Is the food source related or not to the viability of overall bacteria ? There is another phenotype that I would have been very interested to see, is specifically the "dormant" stage or dauer. Other reports in the literature with authors using complex microbiota to feed worms , seemed to reveal some interesting findings in this regard. Please comment.

As stated in the previous rebuttal – we have not observed any dauers. This is consistent with the finding that the worms develop normally on the cHFM. We have added this information to the manuscript. Since dauers are non-feeding we are not sure if they are suitable for gut microbiota studies. However, further studies are needed to determine if the observed phenotypes follow from differences in nutritional value or stimulation of specific host signaling pathways. We believe that our platform offers the possibility to address that and many other interesting remaining questions in future studies.

Related to last part of the findings : the feasibility of FMT with worms.  What about doing FMT with a "pro-inflammatory" / "dysbiotic" microbiota in this model to see whether or not it can really be used as a model for causative mechanistic studies ? This question is not a dirty one, but is related to the fact that most cultured bacteria in the model are rather pathobionts strains in the context of intestinal inflammation (see. E.coli/Shigella or Pseudomonas).

We do not consider this a dirty question. Rather we completely agree that studying dysbiotic microbiomes would be very interesting. Furthermore, we agree with the reviewer that our platform can and should be used for such studies. That is exactly the reason why we want to share our approach. However, including such studies is not a small undertaking. We feel that would warrant a new publication and find it is beyond the scope of the current manuscript.  

Discussion is quite surprising in its overall organization. That said, it is short. However, it is not enough related with the findings specifically, and the limitations of this model/study did not seem to be well acknowledged.

We have made added a few more topics to the discussion and made minor changes. However, it is not clear to us what additional topics we could add without the discussion becoming too speculative. In our opinion the main limitations are acknowledged: 1. The inability to directly study anaerobic bacteria, 2. The difference in temperature found in the human gut versus the worm gut, and 3. The selection following from the type of growth media used.

Minor :

line 118: It is unclear the meaning of "blood donor"

We have rewritten the sentence.  

The color legend in Fig1 panel G is confusing. I expect this Figure to be the same order as previously : Stock then ON cultures. Unless there is a confusion in the graphic as presented? Please clarify

We completely agree and have changed the figure coloring

In addition to the Cheng study (ref13), even more promising results have been put forward on the OligoMM12 (12 members) or Oligomm19 (19) models. this is even more reduced and still promote colonization resistance against known enteropathogens.

Great suggestion also related to anaerobic studies. We have included a comment about the OligoMM12 and OligoMM19 synthetic microbiomes in our discussion.

Reviewer 2 Report

 

-I had some issues with my computer which made all figures black-white. Please disregard my color-related comments

 

-Figure 2 legend: What is the statistical test used? The legend states two-way ANOVA t-test…is it ANOVA or t-test? ANOVA does not seem like the right statistical analysis for a set that compares means between two groups.

 

-Figure 2, panel D: The authors should double check t-test analysis between OP50 and cHFM. 

-The authors stated that the data were deposited into ENA; however, no record is found

Author Response

-I had some issues with my computer which made all figures black-white. Please disregard my color-related comments

 

-Figure 2 legend: What is the statistical test used? The legend states two-way ANOVA t-test…is it ANOVA or t-test? ANOVA does not seem like the right statistical analysis for a set that compares means between two groups.

 

Indeed, we have specified that it is a t-test.

 

 

-Figure 2, panel D: The authors should double check t-test analysis between OP50 and cHFM. 

 

The test is correct.

-The authors stated that the data were deposited into ENA; however, no record is found

We apologize we did not realize that it was not made publicly available – it is now, and we have included appropriate information in the manuscript.

Round 2

Reviewer 1 Report

Dear authors,

 

Thank you for your point by point response. I appreciate most of the replies and believe you justify enough your point of view. That said, I would have appreciate you include additional details on bioinformatic workflow in the paper directly (I noticed a new link to raw dataset, this is well appreciated).

Congratulations for an interesting study in the field.