Sexual Selection and Proteinaceous Diversity in the Femoral Gland Secretions of Lacertid Lizards

Round 1

Reviewer 1 Report

I have read with interest the study entitled “Sexual selection and proteinaceous diversity in the femoral gland secretions of lacertid lizards”. Studies on lizard femoral glands have been focused mainly on the lipid fraction while proteins have been for long time neglected. Although there is an increasing interest on proteins from reptile glands, we know very little about the evolution of these molecules. I think the main idea of the study to test the effect of sexual dimorphism on protein profile variation is interesting, and the authors find a relationship between the degree of sexual dimorphism and some protein bands. In addition this study uses mass spectrometry proteomics in an attempt to identify proteins present in each band (even protein identification is very poor, see below). However, under my opinion, the research has clear limitations. My two main concerns are 1) inter-individual variation on protein expression and 2) identification of proteins.

Related to inter-individual variation, the authors assume that each protein profile is stable across individuals in each species. However, there is no data or statistics showing how large is the variation in the protein bands within species. I would expect some degree of variation between individuals on protein expression. This could affect the results, since in their phylogenetic analysis they take one protein profile for each species. This point is important and must be clarified, and data needs to be added to show how strong the results and conclusions are. This is related to the “virtual gel lane”. I am not sure to have understood how this was created. Is this an average between individuals? Even if data is taken from another publication, I would add some statistics. For instance, it would be good to add this data (e.g. in a table), showing the relative amount of protein clusters for each species (and add a measure of variation within species).

The second point concerns protein identification. The number of proteins identified is very low. Indeed, mass spectrometry-based proteomics in this study could not identify most of the proteins (but just one, carbonic anhydrase). I have several comments here. First, each band could have one or more proteins, therefore, it is a big assumption to say that one band is just one protein. This point needs to be acknowledged. More important, it is unclear in the manuscript which species/ samples were used for identification. However, I went to the supplementary material and there is clear that two species were analysed (Acanthodactylus scutellatus and Gallotia stehlini). I suggest clarifying this in the manuscript also – mention how many samples were analysed and which species of lizards. And also would be necessary to justify the species of choice, why were only these two species selected here? Are other samples or species available for proteomics that could improve protein identification?

In the gels shown in supplementary material, it seems that bands B and A are only present in Acanthodactylus and C, D and E are only present in Gallotia. However, what about other species?

The addition of more species in this part of the study could allow the identification of more proteins and also corroborate whether the same proteins appear in the same band in all species.   

I recognize the effort of the authors in an attempt to identify the proteins, and I think it is necessary to do so and not just relying on protein bands. Perhaps I would suggest that for future studies the authors consider performing mass spectrometry proteomics on the femoral gland secretions directly (not just using the bands of the gels) to get a better characterization of the proteins.

As discussed by the authors, the identification of carbonic anhydrase in one of the bands influenced by sexual size dimorphism may show an indirect effect of sexual selection on this protein. This could be an argument showing that proteins are not involved in chemical communication (at least directly). The authors suggest that the proteins could still be involved in communicative processes by an indirect effect. I suggest to expand this explanation, how this indirect effect could have evolved? In addition, FG proteins could have other functions in lizards, and would be worth to mention that.  

Author Response

Attached you can find the point-to-point-reply

Author Response File: Author Response.pdf

Reviewer 2 Report

I have reviewed a previous version of this manuscript for a different journal, but the manuscript is quite different from its previous form. I believe this new version is a drastic improvement on the previous version I read, as it has refocused the manuscript to be of a broader appeal to researchers interested in sexual selection (and dimorphism), animal communication, and evolution. The authors explore one possible explanation for which chemical signal elements may be targeted by sexual selection in Lacertid lizards as a follow up to a previous study by two of the same authors on the present manuscript (Baeckens and Martín).

The strengths of the study include a large sample size, a thorough chemical analysis of both the lipid and protein fractions of lizard glandular secretions (via combining previously compiled data sets), and an honest and conservative interpretation of the data. The approach is novel in that most chemical signal studies analyze lipid or volatile fractions, and few have analyzed protein fractions, but to my knowledge, analysis of both the protein and lipid fractions has not been done. Thus, this manuscript provides a valuable perspective on sexual dimorphism/sexual selection as a factor in the evolution of multi-component chemical signals. Furthermore, studies that characterize proteins in femoral gland secretions are rare, and the present study identifies at least one protein that may be significant to chemical communication in Lacertids. This topic is relevant to the scope of the journal Diversity because of the phylogenetically-informed analysis targeting chemical signal evolution. I have only a few minor comments for improvement, which I detail below.

 

The introduction does an excellent job justifying the need for this study, and outlines a scientifically sound and logical argument for addressing an actionable gap in the literature (i.e., that the previous study which did not detect a signal of sexual selection in the lipid content of lizard secretions may not have analyzed the signal features being targeted by sexual selection). There are some minor grammatical errors and editing errors that need to be addressed before publication, as well as one clarification on Figure 1 regarding which bars are considered sex-biased in the 9 and 1 counts.

 

Minor Comments

Line 49: insert “have” and “into”; “decade have grown into”

Line 56: “combined with an activity”

Line 57-58: Consider the following edit: “Together, all these observations lead to the hypothesisze that sexual selection may play a pivotal role in the evolution of lizard chemical signals.”

Line 62: remove “in”

Line 66-68: awkward and confusing phrasing: “Lipids, being more volatile and easily associatedble to with male quality- and condition-related traits, have been historically considered of superior in importance in functioning as semiochemicals compared to proteins”

Line 69: replace “base” with “basis”

Line 72: consider replacing “showed no covariation pattern with” with “did not covary with”

Line 82: What precisely is meant by “seasonal waving in the relative abundance”? Are the authors referring to the seasonal variation in abundances?

Line 84-87: Consider replacing, “temporal association of proteins and lipids” to “synchronized variation in abundances between proteins and lipids”

Line 85-87: This sentence has a few awkward phrases that make it confusing:

“The temporal association of proteins and lipids leads us to hypothesize that proteins may be involved in sexual signaling, therefore offering a potential target to sexual selection, which eventually contributes to the observed FG protein diversity.”

Consider the following edits:

“This synchronized variation in protein abundances with lipids leads us to hypothesize that proteins may be involved in sexual signaling, therefore also being a potential target for sexual selection to act on, which eventually contributes to the observed diversity of FG proteins.”

Line 88: Replace “at assessing” with “to assess”

Line 102: Replace “in” with “of”

Line 108: Although Alberts et al., (1993) did not characterize each protein bands and only the size of protein bands, it might be good to cite this paper as a “but see” because they do compare proteins.

Allison C. Alberts, John A. Phillips and Dagmar I. Werner Source: Copeia, Aug. 18, 1993, Vol. 1993, No. 3 (Aug. 18, 1993), pp. 775-781

Lines 116, 148, 196, 273, 278: “associated to” should be “associated with”

Line 118: consider placing “also” immediately before the verb “affects”

Line 119: Similarly, consider moving “also” to immediately before the verb “be”

Line 146: change “run” to “ran”

Line 171-173: This states that there are 9 cases of female-biased SSD and 1 of SHD, but Fig. 1 has the SHD tree on the right and it seems to have many more blue bars than just 1. Additionally, the SSD portion of Fig. 1 on the left is difficult to tell which bars are included in these 9 cases because many more are a lighter blue. It would be helpful to put a symbol of some kind on the bars that are considered female-biased and male-biased if light blue and light red are not also considered biased.

Line 188:  Consider defining abbreviations in the figure so that the figure is understandable on its own without having to search through the main text to remember abbreviations. This will improve overall readability of the manuscript.

Line 201: “, and” should be deleted

Table 1: Caption needs proof reading. Edit “Figg. 3” to remove second “g.” Insert spaces between “error=difference.” Also, I do not understand what is meant by “gel region originating the sample;” please edit for clarity.

Line 240: change “correlates to” to “correlates with”

Line 246: change “relation” to “relationship”

Line 247: By “found FG secretions to be effective in mate choice…” do the authors mean that FG secretions were found to impact mate choice, etc? The use of “be effective” is too vague.

Line 260: change “base” to “basis”

Line 263: Edit “not-reproductive” to “non-reproductive”

Line 274: Switch the order of be and indeed

Line 290: Insert “an” between “such” and “indirect” 

 The manuscript is generally well written and clear, but there are some awkward phrases throughout that need to be edited. For example, there is one particular phrase that the authors use repeatedly, "x associates to y." I believe this should be edited to "x is associated with y." I have outlined other examples in the minor comments above. There are some other minor edits that I detailed extensively in my minor comments, but the manuscript would benefit from further editing.

 

Author Response

attached you can find the point-to-point reply

Author Response File: Author Response.pdf

Reviewer 3 Report

The ms “Sexual selection and proteinaceous diversity in the femoral gland secretions of lacertid lizards” explores whether sexual selection may modulate the proteinaceous compositions of femoral glands. The study is an important new step in investigating different aspects associated with the proteinaceous compounds of the femoral glands. Moreover, as the authors mention, most studies investigating the compounds of femoral secretions, their functionality, and evolution have mainly focused on lipids, while proteins have only begun to receive attention in recent years.

Below are my few comments/questions.

 It is necessary to discuss the head size dimorphism results, including comments on why the two indexes showed different results. Do the authors have any hypothesis to explain the lack of relationship with the proteinaceous profile? Apparently, SSD and HSD do not correlate (Fig. 1). Therefore, these indexes may provide different information involving different associated behaviors, which may explain the results. On the other hand, it seems appropriate that authors discuss their recent results on the protein-lipid association of femoral glands, among others, considering the hypothesis that proteins may improve scent stability.

 L 101-102. It seems relevant to indicate in the text that secretions were only from males. Considering that females can also produce secretions, what authors would predict concerning the presence of the five protein regions that show correlation with SSD if female secretions were included?

 For the figures and tables, avoid some acronyms and replace them with their meaning; this will make it easier for the reader to concentrate on getting the information presented. For example, for Fig 1, the suggestion is to indicate sexual size dimorphism and head size dimorphism; for Fig. 2, replace the acronyms in “X” exe, and in Fig. 3, defined in the legend sexual size dimorphism (SSD). Similarly, for Tables 1 and 2, define FDR: false detection rate.

L 273- 276. Based on this, the proposition seems to be that proteinaceous content may inform on the quality, not the identify, which seems contradictory with what is indicated above. In this context, is the final proposition that more compounds will make it easier to assess individual quality when sexual selection is strong?

 L 183. Concentration or diversity of proteins? On the other hand, can the authors provide a simple figure illustrating the relationship between SSD and protein diversity?

L 272. These studies seem not to explore whether proteins provide information on quality.

L282-284. How can this be linked to SSD? Further,

 

L285-291. From those regions with high protein diversity, is there any evidence that components may act as a supporting structure to lipids?

Author Response

attached you can find the point-to-point reply

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have addressed most of the comments in a satisfactory way and changed the manuscript accordingly. I have just a couple of suggestions.

The authors used previously published data from another study as clearly stated. I am aware that the data is easy and free to access, but the manuscript would benefit from adding some basic information about inter-individual variation on protein profiles. My suggestion is to (at least) add a sentence in the methods to this respect.

Concerning the assumption of one band equals one protein, the authors mention that a sentence about this has been added in the methods (line 106), but I cannot find this. I believe that since identification is very low, the authors cannot rule out the possibility that there could be more proteins unidentified in each band. Although a sentence has been added at the end of the discussion, including this information in the methods would be necessary.

Author Response

Reply is in the attached file

Author Response File: Author Response.pdf