Effects of Reduced Seawater pH and Oil Contamination on Bacterial Communities and Biochemical Markers of Estuarine Animal Hosts

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

the presented study is of very high quality and interest to the readers.

The introduction provides sufficient information to state the issue and the aim of the study is clearly specified.

My suggestions for improvements refer to the Materials and Methods section:

 

The treatment variants could be presented and described in table. This will allow for a more clear presentation.

l. 233, 234, 236, 241, 243, 250, 252 – the empty brackets suggest that these should have been some information provided, but it hasn’t – is this correct?

Another suggestion refers to Figure 7:

 

Figure 7 is very complex and for this reason, I suggest that the Authors could take into consideration providing them in e.g. full page version.

 

The explanation of bacterial community composition is very well conducted, based on a reasonable arguments and confirmed by proper citations.

The analysis of bacterial community composition and their functions is robust and impressive.

The conclusions are drawn based on the thorough examination and discussion of the results.

Author Response

Reviewer #1

Comments and Suggestions for Authors

 

Dear Authors,

 

 

1. the presented study is of very high quality and interest to the readers.

 

2. The introduction provides sufficient information to state the issue and the aim of the study is clearly specified.

 

 

My suggestions for improvements refer to the Materials and Methods section:

 

3. The treatment variants could be presented and described in table. This will allow for a more clear presentation.

 

Table S1 was added in the revised manuscript (See line 101). Subsequent changes were made to the numbering of Tables throughout the manuscript.

 

4. 233, 234, 236, 241, 243, 250, 252 – the empty brackets suggest that these should have been some information provided, but it hasn’t – is this correct?

 

The empty brackets the reviewer refers (e.g., rarefy(), anova(), etc..) are functions used in the analysis. These brackets are intentionally left empty to indicate a specific algorithm from a specific R package. No changes were made.

 

5. Another suggestion refers to Figure 7

 

6. Figure 7 is very complex and for this reason, I suggest that the Authors could take into consideration providing them in e.g. full page version.

The option to publish a figure as a full-page version is editorial. Manuscript submission does not provide this option.

 

7. The explanation of bacterial community composition is very well conducted, based on a reasonable arguments and confirmed by proper citations.

 

 

8. The analysis of bacterial community composition and their functions is robust and impressive.

 

9. The conclusions are drawn based on the thorough examination and discussion of the results.

 

We thank the reviewer for his/her availability in reviewing the manuscript in a timely manner. Below we provide detailed response for reviewers’ remarks.

 

--------------------------------------------------------------------------------------------------------------------------------------

 

 

Reviewer 2 Report

Comments and Suggestions for Authors

The paper aims at elucidating the effects of multiple stressors on benthic invertebrates by targeting changes in the associated microbial communities and at the biochemical level. The two stressors here investigated were the acidification level and the oil contamination, and the responses were deciphered by defining the taxonomic composition of associated microbiomes and chemical biomarkers under controlled conditions.

The paper is well written and structured. The experimental design properly described, but confused and unordered in the results, probably due to the large amount of results obtained. In my opinion, the paper deserve to be improved, by adjusting the presentation of results.

Moderate English revisions are required. Below some suggestions:

Title: I would suggest to shorten the title in a more concise version.

Introduction

Line 61. Delete the round bracket

I suggest to improve information about the two target species of invertebrates. why were they chosen? is there evidence of specific responses or sensitivity to the selected stressors?

Results

Overall, I would suggest to report firstly results on diversity indices and taxonomic composition of the microbial communities in the different treatment. Then I would report statistical differences between groups, and evidences obtained from statistical analysis.

Line 296-297 'There were no other significant differences in the diversity indices of both species (Fig. 1; Tables S1 and S2)'. In the Table S2 you show the statistical differences intragroup, but what about differences between the different treatment?

Lines 302-303. 'A DGGE analysis using sediment and host DNA samples was executed to determine the efficiency of the protocol in removing sediment-bound microbes from the host samples'. This step is not clear to me, and it is not clear the aim of the DGGE. Couldn't the differences between sediments and hosts already be revealed by NGS analysis?

Line 312-332. I would move the entire section after reporting taxonomic composition of bacterial communities

From line 342 onwards: all bacterial taxonomic group should be in italics.

For greater order and clarity, I would divide the results into sub-sections for polychaetes and gastropods.

How the authors detect a factor as significant predictor of taxonomic group abundance? Did they calculate correlation between the parameter and the relative abundance?

The numerical values of most abundant phyla and orders should be reported. How much Proteobacteria were abundant in the Oil or pH treatment? How much differently from the control treatment? This should be stated.

The taxonomic composition of bacterial communities hosted by Hediste diversicolor appeared different within the same treatment in the replicates, while in the Peringia ulvae it is more homogeneus. How do you explain this?

Line 504-506. 'Pairwise comparisons of predictive functional profiles revealed that treatments Cont, OnOi, and OnpH were significantly enriched with KEGG pathways in the polychaete (Fig. 505 S3) and gastropod (Fig. S4) bacterial communities'. Please, move the statistical evidence after reporting results of the analysis.

Figure 7. Improve quality and readability.

Figure 6. The phyla must be written in italics.

Figures S1-S2. I would move as part of the main manuscript, not in supplementary.

Add some figures/tables reporting relative abundances at order/genus level

 

I suggest publication after major revisions. 

Comments on the Quality of English Language

Moderate revisions are required

Author Response

Reviewer #2

Comments and Suggestions for Authors

The paper aims at elucidating the effects of multiple stressors on benthic invertebrates by targeting changes in the associated microbial communities and at the biochemical level. The two stressors here investigated were the acidification level and the oil contamination, and the responses were deciphered by defining the taxonomic composition of associated microbiomes and chemical biomarkers under controlled conditions.

 

The paper is well written and structured. The experimental design properly described, but confused and unordered in the results, probably due to the large amount of results obtained. In my opinion, the paper deserve to be improved, by adjusting the presentation of results.

 

We agree with the reviewer that the Results and Discussion section deserve an improvement. A thorough review of this section was performed in this revision and some rearrangements were made in the structure of the section:

1. Complete revision of the results and discussion
2. When appropriate, we divided the discussion into sections for each host (e.g., Biomarkers section), for a better comprehension by the reader
3. Elimination of redundant information.

By doing this, we hope this will improve the final manuscript.

 

Moderate English revisions are required. Below some suggestions:

 

The revised manuscript has been carefully reviewed by a native English speaker, who is also a co-author of the study, in response to the reviewer's request.

 

Title: I would suggest to shorten the title in a more concise version.

Title was changed to: “Effects of reduced pH and oil contamination on bacterial communities and biochemical markers of estuarine animal hosts”.

 

Introduction

 

Line 61. Delete the round bracket

 

The bracket has been deleted.

 

I suggest to improve information about the two target species of invertebrates. why were they chosen? is there evidence of specific responses or sensitivity to the selected stressors?

 

We appreciate the suggestion from the reviewer and have made changes accordingly. Justification of target species was originally presented in the material and methods section (see lines 143-146 of the original manuscript). However following reviewers remarks this justification was transferred to the Introduction section and edited (See lines 75-79 of the revised manuscript: “….These species were chosen because they 1) are typically abundant at temperate coastal environments, including at at the sampling site of this study; 2) are well-studied organisms in the Ria de Aveiro ecosystems having a vast publication record; and 3) represent individuals with two distinct benthic lifestyles, viz. sediment-dwelling (H. diversicolor) and epibenthic grazing (P. ulvae).…”)

Furthermore, in response to the second question raised by the reviewer, we have included an additional paragraph reporting on the effects of each stressor on the respective hosts, see lines 84-94 of the revised manuscript:

“…Under controlled microcosm conditions, numerous studies consistently demonstrated the sensitivity of H. diversicolor to changes of seawater pH [21–24]. Specifically, lower seawater pH values were observed to impact the osmotic regulation of the polychaete, augment its metabolic activity, and elevate oxidative stress biomarkers, thereby inducing cellular damage [24–27]. Furthermore, reduced seawater pH was also found to enhance the toxicity of certain xenobiotics, such as pharmaceuticals [24][22]. Although the impact of reduced seawater pH on P. ulvae remains unexplored, ocean acidification alters the marine carbonate chemistry, potentially diminishing calcification rates in shelled organisms. Notably, seawater pH has been demonstrated to exert a mild impact on the food consumption and survival of the marine gastropod Gibbula umbilicalis [28]. Despite their crucial roles in remediating polluted environments, both H. diversicolor and P. ulvae have exhibited susceptibility to the deleterious effects of oil hydrocarbon pollution, as documented in previous studies [28,29]…”.

 We believe these adjustments enhance the clarity and context of our study.

 

Results

 

Overall, I would suggest to report firstly results on diversity indices and taxonomic composition of the microbial communities in the different treatment. Then I would report statistical differences between groups, and evidences obtained from statistical analysis.

 

The reviewer suggested a rearrangement of the results and discussion in which the beta-diversity analysis is placed after the specific taxonomic analysis. Despite respecting the reviewer’s opinion, we prefer our current form of presentation. The rationale behind our decision lies in our belief that the combination of alpha- and beta-analyses offers a comprehensive understanding of the dynamics that occurred, whereas taxa-specific analysis provides a more focused perspective on these dynamics. Given this consideration, we have opted not to make changes to the manuscript in response to this suggestion. We hope that this explanation clarifies our decision and aligns with the overall goals of the study.

 

Line 296-297 'There were no other significant differences in the diversity indices of both species (Fig. 1; Tables S1 and S2)'. In the Table S2 you show the statistical differences intragroup, but what about differences between the different treatment?

 

Tables S2 and S3 only present the results of the factorial ANOVA for each diversity index analysis. A pairwise comparison between treatments was not performed. Following the reviewer’s remarks we performed a Wilcoxon sum rank test for all combinations of treatment from the same host (See Lines 218-220 in the revised manuscript: “…Wilcoxon rank sum test was used for pairwise comparison of diversity indexes between treatments through the pairwise.wilcox.test() algorithm in stats package in R with p-value correction using the less conservative false-discovery rate…”). No difference between treatments was detected in this analysis and this was reported in the Results and Discussion section (See Lines 277-278 in the revised manuscript: “…Pairwise comparison of diversity indexes among treatments revealed no significant differences between treatments (WILCOXON: P>0.05). …”).

 

 

Lines 302-303. 'A DGGE analysis using sediment and host DNA samples was executed to determine the efficiency of the protocol in removing sediment-bound microbes from the host samples'. This step is not clear to me, and it is not clear the aim of the DGGE. Couldn't the differences between sediments and hosts already be revealed by NGS analysis?

The objective of this study was to determine the effect of two stressors on host-associated microbiome and host health. Hosts used in this study included the bottom-dweller Hediste diversicolor and the epibenthic grazer Peringia ulvae. During the extensive sampling event and afterward, in the preparation for DNA extraction, an effort was made to remove adherent particles. However potential carryover of the sediment microbiome host samples was targeted as a potential issue. Therefore, we performed a simple electrophoretic analysis to compare microbial profiles based on 16S amplicons. The objective of this analysis is to show that sediment and host-microbiome communities are indeed distinct, and any detected effect is specific to the host microbiome. Other studies of our authorship have studied the effects of these and other stressors on the sediment microbiome (see: doi.org/10.1111/jam.14261, doi.org/10.1007/s11356-018-3214-5,  doi.org/10.1111/mec.13779, doi.org/10.1111/gcb.12801).

 

 

 

Line 312-332. I would move the entire section after reporting taxonomic composition of bacterial communities

 

We appreciate the reviewer's suggestion regarding the separation of the beta-analysis section. However, after careful consideration, we have decided to maintain our preferred structure for the microbiome section in the results and discussion. Our rationale is to present the information in a logical progression, moving from broader insights (alpha- and beta-analyses) to more detailed results (taxonomy, specific ASVs, and functional prediction). We believe this approach enhances the clarity and flow of our findings. We hope this explanation helps in understanding our decision.

 

From line 342 onwards: all bacterial taxonomic group should be in italics.

 

The reviewer’s suggestion is in line with recent guidelines (doi.org/10.1186/s43008-020-00048-6) and we have duly incorporated this recommendation into the revised manuscript. To ensure consistency, a systematic review of the entire manuscript, supplementary material, and figures was conducted. We trust that these adjustments align with the latest standards and address the concerns raised by the reviewers.

 

 

For greater order and clarity, I would divide the results into sub-sections for polychaetes and gastropods.

 

We appreciate the constructive feedback from the reviewer and have carefully considered the suggestion to reorganize the results and discussion. Given the similarity of some results between the polychaete and gastropod (such as alpha and beta-diversity and the description of most abundant ASVs), we have opted for a combined discussion of both hosts rather than separate sub-sections for each. We believe this approach enhances the coherence of our narrative and provides a more integrated perspective on the shared aspects of the host microbiomes in the microcosm system.

However, in response to the overall evaluation, we have undertaken significant improvements to enhance clarity and organization throughout the manuscript. Specifically, we have divided the taxonomic composition analysis into paragraphs dedicated to each host and divided the ‘Biochemical markers’ subsection into two. We believe these changes address the concerns raised by the reviewer and contribute to a more coherent and reader-friendly presentation of our findings.

 

How the authors detect a factor as significant predictor of taxonomic group abundance? Did they calculate correlation between the parameter and the relative abundance?

 

Factors influencing taxonomic group abundance were identified using a two-way ANOVA applied to generalized linear models (GLMs) of respective data (relative abundances * factor). GLMs were generated using the glm() function from the stats package in R. Categorical variables used in the GLM were oil addition (i.e., ‘oil’), seawater pH (‘pH’), and their interaction (oil*pH). Significant results (p<0.05) about a given factor were determined using the two-anova F test, using the function anova() from the vegan package. Results from this multi-categorical ANOVA were listed in Tables S2-S3 (diversity indexes), S6-S9 (bacterial phyla and orders) and S12-S13 (ASV). This approach was initially outlined in the material and methods section (See Lines: 234-237: “…Diversity parameters and relative abundances of selected bacterial higher taxa, as well as the most abundant ASVs (50 most abundant) were tested for significant variation among factors and their interaction (Oil*pH) using the anova() function applied to a GLM model in R [29]….”). In the revised manuscript, this description was refined for clarity (See Lines 215-219: “…Diversity parameters and relative abundances of selected bacterial higher taxa, as well as the most abundant ASVs (50 most abundant), were tested for significant variation among categorical factors ‘oil’ and ‘pH’ and their interaction (Oil*pH) through a two-way ANOVA f-test. The anova() function in R was applied to generalized linear models (GLM) created using the glm() function in R [39]…”).

 

The numerical values of most abundant phyla and orders should be reported. How much Proteobacteria were abundant in the Oil or pH treatment? How much differently from the control treatment? This should be stated.

 

We agree with the reviewer and numerical values of the mean relative abundance and standard deviation of each taxonomic group discussed were added as Table S5 to the Supporting Information file 1. Subsequent table numbering was updated both in the Supporting Information file and in the revised manuscript. Additional discussion was added in the revised manuscript. (See Lines 306-311: “...In our study, a taxonomic analysis of the bacterial composition of both invertebrates showed that Proteobacteria, Cyanobacteria, Planctomycetota, and Bacteroidota were the four most abundant phyla in the whole dataset (Fig. 3; Table S5). Taxonomic profiles evidence clear distinctions between the bacterial communities associated with both hosts. Despite both having similar abundances of proteobacterial ASV, cyanobacterial were predominant in polychaete samples (mostly in the OnpH treatment) and Bacteroidota ASV in the gastropod samples.”).

 

The taxonomic composition of bacterial communities hosted by Hediste diversicolor appeared different within the same treatment in the replicates, while in the Peringia ulvae it is more homogeneus. How do you explain this?

 

Indeed, variability between replicates is higher in the polychaete than in the gastropod and this is transversal to all data analysis reported (alpha-, beta- and taxonomic composition). The justification for this high variability can be read in Lines 320-327 of the original manuscript: “….For the polychaete, a greater variation in ASV composition was observed within treatments compared to the gastropod (Fig. 2a and 2c). This could be attributed to the unique characteristics of the individual worm’s bacterial community. This feature appears to be supported by the previous study which found a higher variation in the overall bacterial communities of the gut tracts of the polychaete H. diversicolor compared to individual burrow systems [17]. Hochstein et al., 2019 also found a similar trend in the polychaete Capitella teleta, with 66% of the identified OTUs from the gut unique to the worm but displaying a high inter-individual variability [45]….”.  Therefore, we believe it to be redundant to repeat the same rationale throughout the article. We have nonetheless restated the higher variability among treatment replicates in the paragraph discussing Figures 3 and 4 (See lines 305-307: “…. As in the PCoA and the alpha-diversity, the taxonomic composition of the polychaetes bacterial community exhibited a higher variability among treatment replicates than that of the gastropod, yet this is in line with previous studies [46,47]. ….”).

 

Line 504-506. 'Pairwise comparisons of predictive functional profiles revealed that treatments Cont, OnOi, and OnpH were significantly enriched with KEGG pathways in the polychaete (Fig. 505 S3) and gastropod (Fig. S4) bacterial communities'. Please, move the statistical evidence after reporting results of the analysis.

 

This phrase was rewritten (See lines 421-422: “…Differential features in the predicted functional metagenome of the bacterial communities associated with the polychaete and gastropod are plotted in Fig. S3 and S4…”).

 

Figure 7. Improve quality and readability.

 

Figure 7 was improved.

 

Figure 6. The phyla must be written in italics.

 

Taxonomic nomenclature in Figures 3,4,5 and 6 were italicized.

 

Figures S1-S2. I would move as part of the main manuscript, not in supplementary.

 

The primary objective of this article is to assess the effects of reduced seawater pH and oil contamination on the host microbiome. Figures S1 and S2, which present DGGE analyses between host and sediment samples, were performed to confirm the distinctiveness of the host microbiome from that of the surrounding environment. Given their relevance to the specific host-environment relationship rather than the main objective of the manuscript, we propose these figures be included as supplementary material.

 

Add some figures/tables reporting relative abundances at order/genus level

 

In the original manuscript taxonomic analysis, we chose to focus our analysis of taxonomic groups at the phyla (Fig 3) and order (Fig 4) ranks. We think that additional information at the genus level would be redundant considering that the most significant deviations are related to taxonomic groups already discussed at the order or related to the most abundant ASVs. Therefore, no changes were made in this respect.