Microbial communities and their predicted metabolic characteristics in deep
fracture groundwaters of the crystalline bedrock at Olkiluoto, Finland

Bomberg, Malin; Lamminmäki, Tiina; Itävaara, Merja

doi:https://doi.org/10.5194/bg-13-6031-2016

Articles | Volume 13, issue 21

https://doi.org/10.5194/bg-13-6031-2016

© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/bg-13-6031-2016

© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 13, issue 21

Research article

|

03 Nov 2016

Research article |

| 03 Nov 2016

Microbial communities and their predicted metabolic characteristics in deep fracture groundwaters of the crystalline bedrock at Olkiluoto, Finland

Malin Bomberg, Tiina Lamminmäki, and Merja Itävaara

Download

Final revised paper (published on 03 Nov 2016)
Supplement to the final revised paper
Preprint (discussion started on 26 Aug 2015)
Supplement to the preprint

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

SC C4851: 'Comments after Review', Saqib Jabbar, 31 Aug 2015
- EC C6811: 'Important information for the authors', Tina Treude, 20 Oct 2015
RC C5710: 'Review: bg-2015-335', Anonymous Referee #1, 24 Sep 2015
- AC C8296: 'Author responses to Reviewer 1', Malin Bomberg, 07 Dec 2015
RC C6537: 'Referee comment', Anonymous Referee #2, 15 Oct 2015
- AC C8297: 'Author responses to Reviewer 2', Malin Bomberg, 07 Dec 2015

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

ED: Reconsider after major revisions (19 Dec 2015) by Tina Treude

AR by Malin Bomberg on behalf of the Authors (21 Mar 2016) Author's response Manuscript

ED: Referee Nomination & Report Request started (30 Mar 2016) by Tina Treude

RR by Anonymous Referee #2 (26 Apr 2016)

Suggestions for revision or reasons for rejection

The manuscript entitled, 'Estimation of microbial metabolism and co-occurrence patterns in fracture groundwaters of deep crystalline bedrock at Olkiluoto, Finland', analyzes the community composition and function of microbial communities in the Finnish subsurface (296-798 m depth). A bacterial and archaeal V6 16S rDNA dataset from 12 different boreholes is the focus of the authors’s analysis and discussion. Metadata such as the pH, the concentration of anions and cations, and relative abundance of bacteria and archaea are also reported. From this data, the authors use techniques such as NMDS, network analysis, and PICRUST to analyze the data. I do not believe that the methods outlined in Section 2.7 Statistical analyses were applied and analyzed appropriately.

Major Comments:

My first concern is about the taxonomic data matrix from which the statistical analyses were made. From what I can tell, these analyses were performed on counts of taxonomic labels that are illustrated in Figure 3. Based on the legend of Figure 3, it appears that a general taxonomic cut-off was not applied as there are genera such as `Microbacterium’ and then broad categories of phyla such as `Actinobacteria_Other’. Since Figure 7 illustrates that the V6 sequences did not have good matches to sequences in the reference database the `phylaX_Other’ counts are inflated relative to other genera. This imbalance in taxonomic labels is particularly concerning as it makes the interpretation of the statistical analyses downstream difficult. For example, two sites may have similar relative abundances of `phylaX_Other’ but the community composition of taxa within the broad category may vary dramatically. Therefore, I would recommend that OTUs or a broader taxonomic classification be used as the taxonomic data matrix in statistical analyses.

Another concern is in the co-occurrence network analysis. The cut-offs are very relaxed (R>0.7, p<0.01). This leads to an extremely large network with many spurious correlations. As presented, I don’t believe network is not informative.

In the NMDS plots, I think that there should be an indication of how the environmental variables were placed on the NMDS plot or how statements of the type ‘No clear environmental factor was identified to drive the communities’ (page 9, line 20-25) were made. As written, one is unable to interpret or reproduce their analyses.

Minor Comments:

Abstract

- I am confused as to what page 1, lines 15-17 mean

Introduction

- Page 2, line 17: did the previous efforts provide information on the relative abundance of archaea to bacteria? This is a nice point of your MS would be good to know if there are any estimates of bacteria and archaea proportions in the introduction (or if there are any general trends about the proportion of bacteria and archaea)

- Page 3, lines 4-12 it is unclear what environments the citations are from

Methods

- Page 5, line 9: what are the negative DNA isolation controls?

Section 2.7
- It is unclear what the input data for the statistical analyses were (are they taxa, OTUs, etc). Based on Figure 3, I wonder if the statistical tests were performed with taxonomic labels. I think that these analyses would be better served on the OTUs the authors generated as OTUs are a type of reference free data. Using OTUs would be especially useful as the majority of taxa (based on Figure 3) are labeled as ‘phylaX_Other’.

- Why was the Chord distance used for NMDS and the Bray-Curtis for UPGMA clustering? Is there a reason why the same similarity metric used in both?

Results

- Due to the difficulty it may be difficult to report temperature of the groundwater. However, it may be useful (for the readers) to report temperature if any information is available.

Discussion

- Page 12, lines 21-25: I am unsure what the authors are trying to say

- Page 12, lines 4-18: as presented, I am unsure how these interpretations were taken from the NMDS plots. I think that the discussion on this topic needs to be improved

- Page 14, lines 2-4: See my concerns about the network in major comments.

- Page 14, lines 15-16: I am unsure what the logic behind the statement ‘Thus, on higher taxonomical level common traits for specific groups of microorganisms may be revealed’. The writing in the paragraph could be improved to clarify this section.

- Page 15, lines 13-20: I found this section difficult to follow

- Page 15, lines 24-25: when referring to PICRUST functions, the authors should use `predicted’ or similar indication to highlight it is based on the PICRUST results and not other data (such as rates etc).

- The authors refer to methane oxidation several times but do not mention the possibility for anaerobic methane oxidation using a reversed methanogenesis pathway.

Tables and Figures

- Table 2ab, columns labeled ’Singles’ and ‘Doubles’ to ‘Singletons’ and ‘Doubletons’ as caption

- Table 2b appears twice

- Figure 3: What are minor groups (e.g. are they known members of the phyla that are colored as XXX_Other)? I feel like this figure is misleading about taxonomic composition. Based on the legend it appears that a general taxonomic cut-off was not applied as there are genera such as `Microbacterium’ and then broad categories of phyla such as `Actinobacteria_Other’. Since Figure 7 illustrates that the V6 sequences did not have good matches to sequences in the reference database the `phylaX_Other’ counts are inflated relative to other genera. This imbalance in taxonomic labels may also cause issues in the other statistical analyses downstream.

- For Figure 9, it would be easier to read if the connections not observed were faded out to emphasize which connections (pathways) were observed

- I did not see the supplementary tables referred to in the author’s response (page 51 line 31)

Hide

ED: Reconsider after major revisions (27 Apr 2016) by Tina Treude

AR by Malin Bomberg on behalf of the Authors (17 Jun 2016) Author's response Manuscript

ED: Referee Nomination & Report Request started (06 Jul 2016) by Tina Treude

RR by Anonymous Referee #2 (04 Aug 2016)

Suggestions for revision or reasons for rejection

The submitted manuscript "Microbial communities and their predicted metabolic characteristics in deep fracture groundwaters of the crystalline bedrock at Olkiluoto, Finland" provides a complete description of the microbial communities (and predicted functions) of 12 samples ranging from 296 to 798 m depth. The methodology and analyses used are sound and I only have minor comments (see below):

1) In general, when talking about the predicted metabolisms, word choice should be adjusted. For example

2) pg 8 lines 28-29, could the inflation of "Bacteria, Other" be also due to novel sequences (i.e. Microbial Dark Matter)?

3a) pg 9 lines 25-26, I'm not sure how this conclusion was made. Additional clarification would be helpful. The conclusion about physicochemical parameters is then repeated in lines 27-28.
3b) I don't understand how the conclusion that the number of "OTUs was the strongest clustering force" or what exactly this means. This part should be re-written for additional clarity

4) Section 3.5 should be rewritten in a language that emphasizes the metabolic functions are predicted. The first paragraph does a good job at conveying the metabolisms are predicted bu the second paragraph and onward says things like "Nitrate is reduced", "Ammonia is then used" --- which could be a bit misleading despite the section title. I suggest clarifying this paragraph with either an additional sentence that the predicted community function could be described as XX or changing things like "Nitrate is reduced" to "Nitrate was predicted to reduce"
--- Clarification that the functions/metabolisms are predicted should be further clarified here:
4b) pg 15 lines 30 to 31
4c) pg 16 line 3
4d) pg 16 section "Amino acid metabolism"

5) pg 12, lines 26-27. I'm not sure what this sentence means. Please clarify

6) pg 13 lines 12-25. This paragraph is difficult to follow and should be rewritten for clarity

7) Abstract, Figure 4. When introducing or describing shared OTUs as a *percentage* (axis label is fraction) of samples, it would be useful to be reminded how many samples there were

8) Figure 5. The heatmaps could be made more interpretable via a better color scheme a single label on the y-axis saying "OTUs" rather than the individual OTU names.

9a) Figure 6. C-F are not necessary in the main text, they can be moved to the supplement. If moved to the supplement, the % variance described by PCoA axis 1 and 2 could be added to the axis labels of A and B.
9b) It is hard to distinguish the shapes in A and B

10) Figure 7 is pixelated. I different image type would look better in publication.

Typos:
1) pg 3 line 10 allows should be allows
2) pg 4 line 14, remove "at"
3) pg 12 line 20, the use of "so far hidden" should be changed. Perhaps the authors meant previously?
4) pg 17, "Membrane transport" section: zink should be zinc
5) pg 17 line 30, estimated to predicted
6) Table 1: put units in (). Define NPOC, DIC, TDS, TNC

Hide

ED: Publish subject to minor revisions (Editor review) (06 Aug 2016) by Tina Treude

AR by Malin Bomberg on behalf of the Authors (14 Sep 2016) Author's response

ED: Publish as is (22 Sep 2016) by Tina Treude

AR by Malin Bomberg on behalf of the Authors (02 Oct 2016) Manuscript

Download

Article (4542 KB)
Full-text XML

Short summary

Over 95 % of the microbial communities in isolated groundwater in Olkiluoto bedrock, Finland, is composed of only a few common species. The remaining rare microbiome contains over 3- and 6-fold more microbial diversity. The rare microbiome is an unlimited source of genomic functionality in the ecosystem. These microorganisms can respond to change caused by physical or biological factors that may lead to alterations in the diversity and function of the microbial communities in crystalline bedrock.