Genome-Wide Identification, Phylogenetic Analysis and Expression Pattern Profiling of the Aquaporin Family Genes in Leuciscus waleckii

Round 1

Reviewer 1 Report

The manuscript investigates the repertoire of aquaporins in the Amur ide (Leuciscus waleckii) as potential candidates for population-specific adaptations to the alkaline water of the land-locked Dali Lake in northern China. The study includes analyses of the SNP differences and expression levels of the 20 identified aquaporins in the different populations and is potentially interesting. However, there are a large number of major and minor issues that need to be addressed prior to being considered further for publication. These issues are outlined below:

Introduction

1. MIP is not the correct acronym for the superfamily. The term MIP was given to the proteins when investigators did not know what they were, but observed that they were the major proteins intrinsic to the membranes of lens fibres. However with the discovery of their function, which subsequently led to the Nobel Prize in Chemistry, the superfamily was termed aquaporins (see PMID: 7692747).
2. Ref 3 did not discover the 17 subfamilies of aquaporins separated in four grades.
3. With regard to function, it is more relevant to introduce the function of piscine orthologs that are known to transport, water, urea, glycerol, ammonia, hydrogen peroxide or other solutes, rather than the distantly related mammalian orthologs. For example, the transport properties of the closely cyprinid zebrafish are known (PMID: 20149227; 19939263; 31840107), as well as those of Atlantic salmon (PMIDs (21941512; 23868847; 25667219; 31840107; 32664262), and other species of teleost (PMID: 20717996; 25586329). In terms of the subject of the study, it would seem that ammonia/urea-transporting aquaporins, e.g. Aqp8, Aqp14, and all of the aquaglyceroporins should be of interest. It is in fact mostly these channels that apparently show changes in expression levels in this manuscript. 
4. Common carp has 38 aquaporin genes (see PMID: 28258572)
5. NH₄⁺ has not been shown to permeate aquaporins.
6. The proper acronyms need to be used for the teleost aquaporins. Protein = Aqp (first letter capitalised, other letters small case), nucleotide = aqp (small case in italics).
7. A full nomenclature for diploid teleost aquaporin proteins is established as Aqp0a, Aqp0b, Aqp1aa, Aqp1ab, Aqp3a, Aqp3b, Aqp4a, Aqp4b, Aqp7, Aqp8aa, Aqp8ab, Aqp8ba, Aqp8bb, Aqp9a, Aqp9b, Aqp10aa, Aqp10ab, Aqp10ba, Aqp10bb, Aqp11a, Aqp11b, Aqp12, Aqp14, Aqp15 (see PMIDs 25426855; 31840107 and 32664262). Note also, that several channels may show lineage-specific duplications as outlined in the PMID 25426855 and 32664262 papers.
8. Several aquaporins of the Amur ide have previously been analysed phylogenetically (see PMIDs: 31840107 and 32664262). In this respect, the Aqp10 channels are in fact Aqp10aa and Aqp10bb.

Methods

9. The genome on which this study has been based is not available. This needs to be submitted to GenBank for validation purposes. In particular, the three aqp3a-type genes each located on a separate chromosome, are not present in the Amur ide genome available at NCBI. Only one aqp3a gene that is orthologus to zebrafish and other cyprined aqp3a genes is found.
10. Lyu et al. not in the reference list. 
11. Wang et al., transcriptome ref not cited
12. Definition and proof of sample populations. There needs to be an analysis of population-specific gene markers to demonstrate that the results are valid and not obtained from closely related species.
13. How were the transcriptomes prepared and sequenced – from what tissues
14. How were 3D PDB models generated?

Results

15. No alignment is available for validation and repeatability assessment. The alignments for the phylogenetic analyses need to included as supplementary information.
16. What are the statistical values on the tree nodes?
17. Where is the second phylogenetic tree?
18. Categories are termed grades (see PMID: 26338866). Better to refer to the grades as classical aquaporins, unorthodox aquaporins, aquaglyceroporins and AQP8-type aquaporins, rather than coded categories that the reader needs to contantly lookup.
19. NP_001137369.1 is AQP6ub
20. Where are the accession numbers for the Amur ide aquaporins? These sequences need to be deposited in GenBank.
21. Could the Aqp8ab sequence be Aqp8bb?
22. The MEME analysis is superflous – better to provide the alignment with TMDs ar/R, NPA and SNPs annotated. We learn nothing structural from the motif analysis, just that undefined motifs apparently exist.
23. There are substantial expression differences between the DL and WSL populations, which is potentially interesting. However, since the data sets are derived from different studies and different investigators, it is essential that gene expression levels are independently validated by qPCR.
24. The experiment shown in Fig. 4. does not explain the treatment of the fish during and after transfer. What steps were taken to avoid stress and acclimate the fish? There are no statements concerning the ethical treatment of the animals. What accepted guidelines and protocols were followed?
25. The results shown in Fig.4a and 4b should to be arranged in the same order of aquaporins in order to easily cross-reference the results of the experiments. These data should also be validated by qPCR.
26. SNP percentages should not exceed more than one decimal place
27. For the SNP analyses, it seems that the data are derived from different studies. What steps were taken to validate the accuracy of the transcriptomes prior to and after the experiments. For example, there appears to be no independent validation that the SNPs actually exist and are not simply the result of sequencing errors?
28. The SNP results shown in Fig. 7 are interesting, but it would be helpful to show which part of the proteins in which the substitution occurs, in order to understand the possible functional implications. Fig. 9 a and b are not very clear. The other SNPs could be shown in supplementary information for cross-referencing in the future.
29. The identified amino acid substitution in Figs 8 and 9 does not line the pore, but exists on the outside of TMD2. How do the authors expect that this site has any functional significance?
30. The three dimensional analyses of pore diameter are not valid without crystallographic data. Models based on distantly related homologs do not provide reliable structal data at the angstrom level. Consequently the conclusions based upon such in silico models are overstated in Fig. 9. Such an analysis could be validated if functional permeability experiments were conducted to show that indeed there are significant differences in the pore dimensions. Without such evidence, the speculation is not valid.
31. The manuscript still needs some english editing.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The present work provides us with valuable information about the adaptive mechanisms of salt-alkaline tolerance. The authors carefully examined the responses of L. waleckii AQP genes to the alkaline pH of Dali Lake. The authors found that AQP3a-1 was one of the key genes involved in regulating L. waleckii adaptation to alkaline environments. The text is well written, although careful proofreading in the English language is needed. The introduction contains the theoretical basis that supports the hypotheses of the work. The materials and methods are well detailed. Although an ethical note on the use of animals in experimentation is lacking. The results are robust and consistent. However, the discussion mas redundant phrases (eg, see links 377 through 387) and still lacks a mechanistic explanation in an ecophysiological context. For example, what impacts on AQP gene expression on a bidirectional movement: fresh water to alkaline and vice versa?

Author Response

Dear reviewers:

On behalf of my co-authors, we are very grateful to you for giving us opportunity to revise our manuscript. We appreciate you very much for your positive and constructive comments and suggestions on our manuscript entitled “Genome-wide identification, phylogenetic analysis and expression pattern profiling of the Aqp family genes in Leuciscus waleckii” (Manuscript ID: fishes-2127440)

We have studied reviewers' comments carefully and tried our best to revise our manuscript according to the comments. The following are the responses and revisions. I have made in response to the reviewers' questions and suggestions on an item-by-item basis. Thanks again to the hard work of you!

1. In fact, our manuscript has been approved by aquatic animal expert Dr. Liew (Honjung Liew). If necessary, the paper can be sent to the organization recommended by Fishes for English polishing after revision
2. The answer of the question “what impacts on AQP gene expression on a bidirectional movement: fresh water to alkaline and vice versa?” has been revised in line613-644

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

A response to the authors revisions is provided on a point-by-point bases below 

Title

Change “Aqp” to “aquaporin”

Abstract

When truncating gene symbols in a list, the truncated text needs to be replaced with a “-“. For example change: “whereas, the aqp3a, 3ap1, 7, 9a expression ...” to “whereas, the aqp3a, -3ap1, -7, -9a expression ...” (make sure to keep the symbols italicised

Introduction

1. ok - But the authors use the acronym Aqp for aquaporin. This is only necessary in the case of protein and gene names, but not for the term aquaporin. The term aquaporin should be written out in full. It should be noted, however, that the first aquaporin isolated was AQP0 from bovine lens fibres, but the first demonstration of the function of an aquaporin was AQP1 (see Finn & Cerda, 2018, Aquaporin in Encyclopedia of Signaling Molecules2nd Edition. Springer, New York, p 1-18). The authors need to check gene and protein nomenclature conventions. For example for tetrapod proteins, it is AQP1 (all capitals), but a teleost orthologue is annotated Aqp1aa (only the first letter is capitalised). For the respective nucleotides the symbols are AQP1 (all capitals, italicised) and aqp1aa (all small case, italicised). It seems that this is mostly correct in the manuscript, but not when referring to the human or mammalian orthologues. Finally, the first sentence of the introduction is not actually correct. Aquaporins do more that transport water and other small solutes between cells. For example, they also function in the inctracellular membranes of vesicles and mitochondria.
2. ok
3. ok 
4. In the phylogenetic tree of PMID: 28258572 (Figure 10.1) there are 38 Common carp aquaporins
5. ok
6. see point 1 above
7. This only partially ok. Many of the symbols for other species remain incorrect throughoutt the text. Since the literature reference were provided in my previous review, this needs to be thoroughly addressed in any future revision.
8. The sentence correcting the Aqp10aa and Aqp10bb nomenclature is not accurate. It requires redrafting.

Methods

9. At the very least an accession number must be provided that demonstrates that the genome will become available.
10. ok 
11. ok, but as for point 9, an accession number for the reference genome is necessary.
12. ok
13. ok
14. ok

Results

15. ok
16. ok
17. ok
18. ok.
19. ok
20. I could not find any accession numbers for the Amur ide aquaporins.
21. ok
22. This point has not been addressed, and the MEME analysis remains superflous.
23. ok
24. The authors have misunderstood this point. What steps were taken to avoid stress and acclimate the fish? Were they treated humanely? Were they anaesthetised? There are no statements concerning the ethical treatment of the animals. What accepted guidelines and protocols were followed?
25. ok
26. ok
27. ok
28. ok
29. As stated, the identified amino acid substitution in Figs 8 and 9 does not line the pore, but exists on the outside of TMD2. Abetter explanation of how the authors think that such a substitution would have functional significance? There is no His63 in LwAqp3a, or His182 in human AQP3. Moreover, they are not the same residues, and neither form part of the ar/R residues in aquaporin-3 channels.
30. This point has also not been addressed. Without crystallographic or experimental evidence, it is simply speculation.
31. The manuscript still needs substtantial English editing.

Other comments

32. Two references are combined in reference 39.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf