Composition and Patterns of Taxa Assemblages in the Western Channel Assessed by 18S Sequencing, Microscopy and Flow Cytometry

Round 1

Reviewer 1 Report (Previous Reviewer 1)

I have no fundamental comments on the article and it can be published.

Author Response

Thank you for your time and contribution, we feel this is a valuable study and on direction from other reviewers have made significant improvements.

Rowena Stern

Reviewer 2 Report (New Reviewer)

The manuscript by Rowena et al. presents results of a study that must be considered highly necessary today. It uses genetic identification to reveal the species that are in the pico- and nanoplankton fractions, compares the information provided by genetic identification with that obtained by microscopy, and uses the genetic approach (quantitative PCR) to infer abundances of species hardly identified by microscopy.

In my opinion, the study is relevant, though the results obtained are already known. That is, genetic identification reveals as useful approach in pico- and nanoplankton sizes (where optical microscopy fails to identify), although species identification occurs in very few cases. In plankton >20µm, the comparison between genetic and microscopic identification does not reach a level to be considered acceptable, and the quantitative estimate of two HAB species does not seem to be satisfactory. It is true that the two species chosen to carry out this quantitative estimation are of difficult identification by microscopy. Perhaps it would have been more appropriate to choose 2 species with unambiguous microscopic identification, even though these species were not HAB species. In this way, calibration of the quantitative genetic estimate could be better, if the aim is to apply this molecular technique for plankton identification and counting.

Being an interesting manuscript, however its reading becomes very difficult. It is not clear whether the Pseudo-nitzschia and Aureoccocus assays belong to material and methods or to results. There is supplementary material incorrectly cited. The manuscript also needs editing work and the following are only some examples:

I think there is a comma missing between structure and the drivers in the first sentence of the introduction.

Environmental sequencing is somewhat unclear for people who do not work directly with molecular

Line 39, why pico- and nanoplankton is here <5µm and later (line 52) <20µm?

Lines 57-59. Large phytoplankton are composed primarily of diatoms and dinoflagellates, but not all diatoms and dinos are >20µm. Planktologist know this very well.

Lines 209-210. The filters were removed with forceps; from where?

Line 211. When was ethanol added?

I think the manuscript needs substantial editing work before publication.

The manuscript by Rowena et al. presents results of a study that must be considered highly necessary today. It uses genetic identification to reveal the species that are in the pico- and nanoplankton fractions, compares the information provided by genetic identification with that obtained by microscopy, and uses the genetic approach (quantitative PCR) to infer abundances of species hardly identified by microscopy.

In my opinion, the study is relevant, though the results obtained are already known. That is, genetic identification reveals as useful approach in pico- and nanoplankton sizes (where optical microscopy fails to identify), although species identification occurs in very few cases. In plankton >20µm, the comparison between genetic and microscopic identification does not reach a level to be considered acceptable, and the quantitative estimate of two HAB species does not seem to be satisfactory. It is true that the two species chosen to carry out this quantitative estimation are of difficult identification by microscopy. Perhaps it would have been more appropriate to choose 2 species with unambiguous microscopic identification, even though these species were not HAB species. In this way, calibration of the quantitative genetic estimate could be better, if the aim is to apply this molecular technique for plankton identification and counting.

Being an interesting manuscript, however its reading becomes very difficult. It is not clear whether the Pseudo-nitzschia and Aureoccocus assays belong to material and methods or to results. There is supplementary material incorrectly cited. The manuscript also needs editing work and the following are only some examples:

I think there is a comma missing between structure and the drivers in the first sentence of the introduction.

Environmental sequencing is somewhat unclear for people who do not work directly with molecular

Line 39, why pico- and nanoplankton is here <5µm and later (line 52) <20µm?

Lines 57-59. Large phytoplankton are composed primarily of diatoms and dinoflagellates, but not all diatoms and dinos are >20µm. Planktologist know this very well.

Lines 209-210. The filters were removed with forceps; from where?

Line 211. When was ethanol added?

I think the manuscript needs substantial editing work before publication.

Author Response

Author responses to reviewer 2 Composition and patterns of taxa assemblages in the Western Channel assessed by 18S sequencing, microscopy and flow cytometry.

We thanks reviewer 2 for their critical appraisal that has allowed this manuscript to be improved. The main aspects have been removal of much methodology, re-organising section titles and discussion sections are more directly related to aims. Answers in italics.

In my opinion the results are already known-… genetic identification reveals useful approaches in pico- and nanoplankton sizes…although species identification occurs in few cases.

The authors concur these conclusions, but this study  provides a genetic eukaryotic surveys of the whole Western channel transect that has not been measured before in this region. We show seasonality and add phytoplankton taxonomic resolution to flow cytometry  phytoplankton aggregate groups. Further, it compares multiple sampling and detection methods namely, microscopy,  genetic and flow cytometry at single station and across a transect providing unique spatio-temporal insight into diversity changes.  The literature review that was undertaken by Stern et al., 2018 (doi:10.1093/plankt/fby035) shows there are very few comparisons with multiple sampling or detection methods, and where it has been done, it is usually to one or a few taxa groups. The aim of the study is to understand to what extent can genetic detection augment optically acquired data, which been further clarified in the introduction. The three methods offer complementary information for complete monitoring. This study also provides a unique assessment of this state of knowledge and data point to compare with others to assess genetic detection consistency.

The authors have augmented this paper by linking genetic taxonomic resolution to flow cytometry detected groups in section 3.4 having adding in taxa size categories to Table S5 to assess likely pico- and nano-sized phytoplankton. Fig. 4 in section 3.4 has been augmented to show spatially resolved phytoplankton groups by flow cytometry that additionally reveal spatial-specific seasonality. In section 3.5 the authors have added Fig. 6B to summarise HTS-derived phytoplankton taxa from WaMS. Finally a synthesis figure (Fig. 10) has been added to summarise information and benefits of each method.

In plankton >20µm, the comparison between genetic and microscopic identification does not reach the level to be considered acceptable

It is true small sized organisms are detected less by microscopic methods, however genetic detection is not clearly size-biased as we found many metazoans in our samples. With additional size data in Table S5 we also find larger diatoms/dinoflagellates/ciliate taxa. It is likely the number and quality of target gene markers encountered by PCR/sequencing processes and the chemical optimisations of these processes. This comparison is still important to quantify the taxa that can be measured by each method.

And the quantitative aspect of the two HAB species does not seem to be satisfactory… perhaps it would be better to choose 2 species with unambiguous identification even though these species were not HAB species… if the aim is to apply this method for plankton identification and counting

It is somewhat unclear what aspect satisfactory means, so I hope this addresses the comment. The authors acknowledge the aims of the project are not sufficiently clear. There were technical issues with qRT-PCR that we address as one of our aims (aim 2) in introduction L144-147. We do think there is validity in the reviewers comments comparing morphologically unambiguous species as a meaningful comparison and add this suggestion to our discussion L621-625.

The primary objective was not to apply this method for plankton identification and counting, qRT-PCR has already been validated as a method to count plankton and we used published, validated qRT-PCR assays (Andree et al. 2011, Popels et al. 2003) showing that detection was species-specific. Harmful algae are the only taxa where species-specific assays are lacking so it’s unclear how toxins are linked to species. Our aim was to compare as harmful algae species in the Western channel that have not been sufficiently characterised at species-level and then to compare them with phytoplankton groups that share some similar traits (size-based for A.anophagefferens or belonging to same genus for P. delicatissima).  The two contrasting species were chosen as they were present in the Western channel and had validated assays with potentially. We have clarified this in the introduction L99-101 and L126-127.

The authors note reviewers views on unsatisfactory HAB measurements. We reviewed HAB qRT-PCR data again and reassessed P.delicatissima cell concentrations using just copy number method  that has now improved cell concentration estimates to levels seen at station WCO L4. WaMS seasonality didn’t match with that of WCO L4 so we included CPR Pseudo-nitzschia data that showed a better match for the earlier appearance of P. delicatissima.

We reviewed A. anophagefferens dataset and removed qRT-PCR AA2013 run data which was of insufficient quality allowing better quality comparison with that of picoeukaryotes.

Being an interesting manuscript it becomes difficult. It is not clear if Pseudo-nitzschia or Aureococcus assays belong to material and methods or results. There is  supplementary material incorrectly cited.

The  JMSE guidelines expressed full descriptions of methods and results however this then becomes difficult to navigate with so much technical information. The assay development for these species were described in under section 2.4 containing several subsections in Materials and Methods and the results for validation of qRT-PCR assays in section 3.2, whilst seasonal trends were in results section 3.3.

The authors have followed editors advice and have shortened the paper to make it easier to follow the main aims. The technical validation sections of qRT-PCR in materials and method has been transferred to Appendix 2 so essential quantitative real-time PCR methods are summarised  2.5. The results relating to detection and species specific performance assays are in results section 3.2, whilst seasonal trends of P. delicatissima and A. anophagefferens and their comparisons to other phytoplankton datasets are in results section 3.3.

Supplementary tables were checked, excel spreadsheet tabs were relabelled to table number and we identified Table S12 was incorrect and an additional Table S13 was not mentioned.  These have been corrected.

Specific comments

 The manuscript also needs editing work and the following are only some examples:

I think there is a comma missing between structure and the drivers in the first sentence of the introduction.

Thank you -this has been correct

Environmental sequencing is somewhat unclear for people who do not work directly with molecular

L59 has been amended to explain what this is.

Line 39, why pico- and nanoplankton is here <5µm and later (line 52) <20µm?

We were describing two different components de Vargas et al 2015 defined picoplankton as phytoplankton under 5 microns and for L52 describing pico- and nanoplankton (2-20 microns). We have used consistent definitions L56, 64, 65 according to Finkel et al. 2009.

Lines 57-59. Large phytoplankton are composed primarily of diatoms and dinoflagellates, but not all diatoms and dinos are >20µm. Planktologist know this very well.

Noted- this gives the wrong impression of the community, the line has been removed.

Lines 209-210. The filters were removed with forceps; from where?

This line was removed in shortening the manuscript as more detailed description was provided in our validation study of WaMS DNA samples in Stern et al. 2015. Forceps were used when transferring filters in microfuge tubes used in storage to new tubes for DNA extractions to avoid contaminating or damaging them.

Line 211. When was ethanol added?

This line has also been removed as methods are described in Stern et al. 2015. It described collection of plankton filters and dislodged plankton from their ethanol storage tubes.  They are fairly standard so text was minimised here in an effort to shorten text under editorial instruction. The filters were stored in tubes of ethanol, then removed to clean, dry tubes to let it evaporate before DNA extraction commenced. Sometimes plankton got dislodged from the filter and stayed in the ethanol storage tubes so to collect these, additional centrifugation was carried out, ethanol aspirated away leaving a small amount that would evaporate so the plankton pellet could be collected and placed in the DNA extraction tube with the filter.

I think the manuscript needs substantial editing work before publication.

The authors take this on board and have made substantial changes. The technical and ecological aspects have been separated more clearly to assist the reader focusing on study objectives.

1. The manuscript was shortened
2. The overall aims were aligned more with results and discussion with corresponding improved section labels to guide readers.
3. The qRT-PCT results for P. delicatissima were improved with new methodology and seasonal occurrence improved by the inclusion of CPR Pseudo-nitzschia in Fig. 3 revealing different spatial timing of P. delicatissima and its species complex.
4. An additional summary taxa composition panel was added to Fig.6 (panel B) that were harder to see earlier with heterotrophic taxa included.
5. In section 3.4, with size data added to taxa in Table S5, the phytoplankton taxa composition of pico- and nanoplankton was assessed for flow cytometry phytoplankton that revealed consistent and seasonal taxa components. This has not been done before with these datasets.
6. The discussion is divided into four sections discussion aim (3) sequencing results which was the most effective improvement to existing monitoring. An additional section on technical performance of qRT-PCR from WaMS was added to relate to aim 2 and finally a comparison of P. delicatissima/A.anophagefferens was added to relate to aim 1. A synthesis paragraph and summary fig. 10 of the complementary findings of each method with ways to improve and harmonise these methods

The manuscript by Rowena et al. presents results of a study that must be considered highly necessary today. It uses genetic identification to reveal the species that are in the pico- and nanoplankton fractions, compares the information provided by genetic identification with that obtained by microscopy, and uses the genetic approach (quantitative PCR) to infer abundances of species hardly identified by microscopy.

Thank you. We hope our amendments have made the paper more clear and thank you for your time in improving this paper.

Reviewer 3 Report (New Reviewer)

General comment

This study (title: Composition and patterns of taxa assemblages in the Western Channel assessed by 18S sequencing, microscopy and flow cytometry) is a meaningful study that analyzed the phytoplankton community through the recent molecular and traditional microscopic methods, a hot topic related to phytoplankton monitoring. However, compared to the clear overall research purpose (comparison of the two methods), the consistency of the data and methods used is low, and the composition is not concise. In addition, the introductory part, which should explain the purpose of the research from the background of the research, does not feel logically organized. In addition, the methods are not written concisely with key contents, but feel like they are written sporadically, and most of the important contents such as used primers and conditions are located in the Appendix, making it difficult to convey information to readers. It is thought that the discussion needs to be written based on the results to suit the research title and purpose. For these reasons, I think that this paper is not suitable for publication in JMSE.

 

Specific comments.

-          Is it possible to monitor with 270um mesh CPR? I think it is too big to monitor phytoplankton.

-          I understand the logic that diversity is underestimated because there are many taxa missed by microscopic analysis in the introductory part, but I think I need an example or explanation why diversity is important.

-          In particular, the parts of Pseudo-nitzschia delicatissima and Aureococcus anophageferens in the introduction feel forced to be added to include the analyzed data, and are rather suitable for discussion.

Author Response

Author responses to reviewer 3 Composition and patterns of taxa assemblages in the Western Channel assessed by 18S sequencing, microscopy and flow cytometry.

We thank reviewer 3 for noting the study is meaningful and note their comments that must be improved, below and our responses. We point out to reviewer 3 the editor has requested the manuscript be shortened

1. Does the introduction provide sufficient background and include all relevant references?

The introduction has been shortened and reorganised to improve clarity -see comments below with 14 references, although recognising the editors comment to be concise and shorten the manuscript.

2. Is the research design appropriate?

The authors note comments regarding discussion based on results, the aims, results and discussion have been labelled and aligned better with discussion sections relating to aims. Re-analysis of qRT-PCR has improved quality of data and cell abundances relating to P. delicatissima. Cell concentrations are of the same order as WCO L4 microscopic ones, whilst CPR P. delicatissima dataset comparison confirms earlier occurrence of P. delicatissima found in WaMS but not visible in WCO L4 dataset. Fig. 4 shows additional spatial variation pico- and nanoplankton by flow cytometry providing insights into taxonomic assemblages in coastal and open waters. Fig. 6 has been improved to summarise phytoplankton and WaMS derived phytoplankton taxa asssemblages are now related to flow cytometry dataset to provide taxonomic resolution.

3. Are the methods adequately described?

We have made a compromise between detail and clarity based on editorial and other reviewers comment to focus on aims of the project. Methods are described in detail in appendices 1 and 2 (appendix 2 is eight pages long) and we have cited more clearly our earlier validation study (Stern et al. 2015) that contains detailed methods of DNA extraction, flow cytometry and bioinformatics methods

4. Are the conclusions supported by the results?

The discussion/conclusions have been re-organised to relate to the original aims more clearly

Additionally areas that can be improve are the presentation of results.

 

However, compared to the clear overall research purpose (comparison of the two methods) feels the consistency of data and methods used is low, and composition is not concise.

The authors recognise the length of the manuscript and inclusion of multiple datasets and technical validation makes the flow difficult to follow. The editor has requested the text is shortened, so we have shortened the text, especially in the methods, and re-organised as below

The abstract states overall study aim to determine how genetic methods can augment optical phytoplankton monitoring methods for improved diversity assessments.

The Introduction has been re-organised- see next question and fully addressed three methods as flow cytometry comparisons were a minor focus in manuscript. Additionally, the technical performance of WaMS samples at genetic quantification are stated so these can be addressed separately, rather than with ecological interpretation in the discussion as was previously done.

Methods: section titles were revised to match results more and removed confusing subsections. In this revision, the methods have been shortened with SST and nutrient methods in appendix 1, all detailed qRT-PCR (section 2.5) in appendix 2 and as our earlier validation study described DNA extraction and flow cytometry in detail, we have shortened these sections and referred to that source more fully. We appreciate the detail is useful but detracts from the main aims so we summarised important aspects such as sequence validation of HRM curves to show species specific detection, quality assurance of different qRT-PCR runs and their inter-comparison.

The qRT-PCR sections were reassessed to improve P. delicatissima cell concentration estimates. We removed A. anophagefferens 2013 sample runs with reduced reaction quality.

Results:  Section titles are revised to reflect aims of the study and extraneous information removed, and additional results to augment that of other phytoplankton surveys. Pseudo-nitzschia comparisons are extended and better phytoplankton seasonal descriptions from WaMS data. Fig. 4 shows spatial separation of plankton groups across the channel. Section 3.4 now relates HTS derived taxa to flow cytometry phytoplankton groups, allowing for greater resolution of assemblages contained within FC groups. This has not been reported over a transect in the Western channel. 

Discussion: The discussion is divided according to the aims with text directly links them to the aim. Aim 3 was discussed first as HTS biodiversity data from WaMS had the biggest improvement on existing monitoring - the first overall on biodiversity. A labelled section of performance of qRT-PCR assays on WaMS samples linked to aim 2 is next and finally aim 1 is labelled seasonal dynamics of harmful algae alongside other datasets. Finally a synthesis paragraphs finishes the discussion including figure 10 to summarise the results and overall conclusion that genetics augments other monitoring approaches that when combined lead to more complete and accurate picture of phytoplankton diversity. 

In addition the introductory part, which should explain the purpose of the research from the background of the research, does not feel logically organized.

The introduction has been re-written to more clearly define the purpose of the study is to determine how useful genetically-acquired taxa for routine monitoring in order to improve legal assessments of marine health with respect to plankton. The background cites importance of phytoplankton, then legal requirements to measure phytoplankton diversity, most of which also requires abundance (hence the inclusion of qRT-PCR trial on WaMS samples). The introduction focuses on pico- to nano-sized plankton as this is where there is the greatest phytoplankton knowledge gap exists. We discuss strength and weakness in current methods used to assess these sized phytoplankton including genetic methods and the need for cross comparison to meet needs of official monitoring and assessments. The introduction then discusses the western channel and knowledge gaps that include smaller phytoplankton and harmful algae and provide background to choice of harmful algae: ie they are not sufficiently characterised to species and are present in the western channel so a good test choice. We cite studies on why genetics can resolve these gaps and finally we summarise aims.

The methods are not written concisely with key contents, but feel like they are written sporadically, and most of the important contents such as used primers and conditions are located in the Appendix, making it difficult to convey information to the readers.

The level of technical information here and many different datasets have meant it is difficult to navigate. We have shortened the methods and labelled subsections to assist readers navigate the methods in line with editorial direction. Primers and assays are summarised in Table 2. DNA sequencing methods we refer to our earlier validation study that provides these details but the level of detail for bioinformatics is fairly standard compared to similar studies. The HRM- qRT-PCT assays were shared in a way in the methods and results that made it look sporadic. So here most of the detail is now in Appendix 2 that is 8 pages long, but only essential methodological aspects are mentioned in the main manuscript relating to confirmation of species specific detection using HRM, methods to asses quality assurance of qRT-PCR assays including inter-run variability.

It is thought that the discussion needs to be written based on the results to suit the research title and purpose

This is unclear, do you mean the manuscript discussion is retrospectively based on findings and not based on independent aims? The aims align with the original project aims of AtlantOS and DEFRA that funded it to deliver an advanced framework for the development of an integrated Atlantic Ocean Observing System. Hence the use of the automated water sampling platform, the development of new biological ocean variables and comparisons with existing ones to create baseline information with which to use for integrated observing.

The discussion is based on the results but we have clarified their link to the original aims (see earlier comments) and added additional aims and we critically evaluate the performance of WaMS in capturing diversity by HTS and harmful algae phytoplankton by qRT-PCR. Finally we discuss integrated monitoring benefits, future challenges and improvements.

Specific comments

Is it possible to monitor with 270um mesh CPR? I think it is too big to monitor phytoplankton

Yes, the CPR survey monitors over 250 types of phytoplankton routinely since 1930s and there are numerous publications linking phytoplankton to environmental drivers. The mesh size is more suited to zooplankton but the silk weave is fibrous increasing the surface area. It captures phytoplankton down to 10µm and phytoplankton capture is reviewed by Richardson et al. (2006), Prog. Oceanogr. 68, 27–74. doi: 10.1016/j.pocean.2005.09.011). We have noted phytoplankton capture now in the introduction with the above cited review, as this was not mentioned in the manuscript. However, smaller phytoplankton capture may not be as efficient hence the development of WaMS device to go inside the CPR payload to capture smaller phytoplankton.

I understand the logic that diversity is underestimated because there are too many taxa missed by microscopic analysis in the introductory part, but I think I need an example or explanation why diversity is important

The first paragraph L43-52 addresses the importance of diversity more and L90-93  cite an English Channel study a relationship between phytoplankton diversity and primary production now

In particular, the parts of the Pseudo-nitzschia delicatissima and Aureococcus anophageferens in the introduction feel forced to be added to include the analyzed data, and are rather suitable for discussion.

The manner of their inclusion in the introduction was clumsy and linked to earlier comments on logical progression. The introduction here improves the linkage with the general study aims and justifies their inclusion because they are of ecological significance, they were previously identified in this region and generally but insufficiently characterised at the species-level.  

Round 2

Reviewer 3 Report (New Reviewer)

Through revisions, this paper has been greatly improved.

In particular, the structure and logic of the text have become clearer.

I now consider this paper suitable for publication in JMSE.

Just need to edit a few things below.

- Line 398, change the font to italic for species name.

- Line 411-413, unified font. 

- You need to increase the figure and the characters within it. (Especially Fig. 1, 2, 6). 

- Species names in the figure should also be changed to italic.

 

 

 

 

Author Response

Please see the attachment

Author Response File: Author Response.docx

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

This article is devoted to the seasonal dynamics of phytoplankton in the Western Channel of UK. The paper focuses on the dynamics of species of nanoeukaryotes of the genus Pseudo-nitzschia and one species of the genus Aureococcus using various identification methods: traditional identification using light microscopy, genetic identification and flow cytometry. The most important result is that these methods do not give complete matches of the number of cells. Each method has its own disadvantages and advantages. The combined application of these methods gives the most complete picture of the dynamics of phytoplankton species. Genetic methods have great prospects in assessing the trophological status of phytoplankton, as well as assessing the role of bacteria in the functioning of the ecosystem. Therefore, these methods should be developed and in the future they should become an integral part of monitoring marine ecosystems. I believe that this article deserves attention as a promising development and can be published.

Reviewer 2 Report

Please see the attached fie. 

Comments for author File: Comments.pdf