Gamete Recognition Gene Divergence Yields a Robust Eutherian Phylogeny across Taxonomic Levels

Round 1

Reviewer 1 Report

The authors present a phylogeny for placental mammals based on nucleotide sequences of the Onadhesin (Zan) gene, gene encoding a sperm protein that is species-specific. The authors compare their results to on a phylogeny derived using multiple gene sequences. Based on the authors’ analyses, the phylogeny based on the Zan gene is similar to the Supertree produced by a much larger dataset based on 31 genes.

Introduction

Lines 49, 50, 58, 62, etc. – eutherian tree – alternatively use Eutheria or placental mammals

Line 100 – You might provide a citation for why neutral genes are more likely to be phylogenetically informative.

Line 108 – You should provide a citation for the idea of speciation genes and phylogenetics. Citation – Wolf, J.B.W., Lindell J., and Backstrom, N. 2010. Speciation genetics: current status and evolving approaches. Phil. Trans. R. Soc. Lond. B Biol Sci. 365:1717-1733. > Phifer-Rixley, M. Searching for the genes that separate species. eLife 2014;3:e05377. (2014) DOI: 10.7554/eLife.05377 > Ting et al. 2000. PNAS 97:5313-5316 https://doi.org/10.1073/pnas.09054159

Line 120 – phylogenetic studies

Methods and Materials

Line 126 – What were the two prototherian mammals? I assume they are echidna because you use the duckbilled platypus as the outgroup.

Phylogenetic utility section beginning on line 130 – What “selection analyses” did you use?

Line 163 – I understand the idea of using a tanglegram to compare two phylogenies, but what type of topological analyses were used?

Line 182 – Did you measure incongruence by the number of disagreements associated with lineage relationships?

Line 188 – extensive gene sequences

General Question and Statement – I looked at the phylogeny in Figure 1, and the phylogeny is resolved. What I don’t see is any value associated with the different branches. Therefore, I wonder how much support there is for a particular branching pattern. Bootstrap values and posterior probabilities are one way to show support, but low values are not considered very informative. In addition, you talk about “statistical support” for nodes, but the term is somewhat misused because this is not statistics in the sense of have p-values but rather non-statistical approximations of an infinite number of possibilities.

Another question – I looked at the previous Zan paper and noted that in terms of amino acid replacements among divergent lineages, there are very few. This suggests that most substitutions are synonymous and occur at the third position. Is this true? What is the rate of synonymous and non-synonymous substitutions for Zan? The overall substitution rate must be reasonable slow.

Lines 224-225 – I am a little confused on the idea of “overall statistical support.” Is this an estimate of the number of bifurcations versus polytomies?

Lines 234-235 – You note that Zan and Tex14 provide the highest support at deep nodes and are under the highest positive selection. This is interesting because these results that these two genes evolve slowly, which is surprising Zan’s support for more recent speciation events. Have you looked at overall patterns of divergence across Zan? Are the sites informative for more recent bifurcations more likely to be neutral? Alternatively, are their non-variable and variable regions across Zan?

Overall, I think that this paper makes a significant contribution to our understanding of mammalian phylogenetics. I must admit, however, that the details in the results (e.g., comparison of result of various comparisons) was long and burdensome. Unfortunately, I have no specific ideas as to how one fixes this, other than placing some of the details in the supplementary section. One item, which I consider very important, is the overall need to understand more about the pattern of sequence divergence in the Zan gene. For example, which regions are likely to be under strong selection? Which types of substitutions provide the most support for various mammalian relationships? I see an overall need for a more detailed assessment of the molecular evolution of this gene, which is beyond the scope of this paper.

Author Response

Reviewer #1

The authors present a phylogeny for placental mammals based on nucleotide sequences of the Onadhesin (Zan) gene, gene encoding a sperm protein that is species-specific. The authors compare their results to on a phylogeny derived using multiple gene sequences. Based on the authors’ analyses, the phylogeny based on the Zan gene is similar to the Supertree produced by a much larger dataset based on 31 genes.

Thank you for your thoughtful comments. We believe that they helped strengthen the manuscript.

Introduction

Lines 49, 50, 58, 62, etc. – eutherian tree – alternatively use Eutheria or placental mammals

Response: We either deleted “Eutherian” or we replaced “Eutherian tree” with “Eutherian phylogenetic tree” or “phylogenetic tree”.

Line 100 – You might provide a citation for why neutral genes are more likely to be phylogenetically informative.

Response: Thank you, we added citations Molloy and Warnow 2018 and Latrille et al. 2023.

Line 108 – You should provide a citation for the idea of speciation genes and phylogenetics. Citation – Wolf, J.B.W., Lindell J., and Backstrom, N. 2010. Speciation genetics: current status and evolving approaches. Phil. Trans. R. Soc. Lond. B Biol Sci. 365:1717-1733. > Phifer-Rixley, M. Searching for the genes that separate species. eLife 2014;3:e05377. (2014) DOI: 10.7554/eLife.05377 > Ting et al. 2000. PNAS 97:5313-5316 https://doi.org/10.1073/pnas.09054159

Response: Thank you, we added citations Wolf et al. 2010, Phifer-Rixley 2014, and Ting et al. 2000.

Line 120 – phylogenetic studies

Response: Fixed.

Methods and Materials

Line 126 – What were the two prototherian mammals? I assume they are echidna because you use the duckbilled platypus as the outgroup.

Response: Yes, we added in the genus name Tachyglossus as the other prototherian mammal used in the Zan phylogenetic analysis.

Phylogenetic utility section beginning on line 130 – What “selection analyses” did you use?

Response: We added in that we performed corresponding “PAML” selection analyses.

Line 163 – I understand the idea of using a tanglegram to compare two phylogenies, but what type of topological analyses were used?

Response: We added in that we used topological comparison “by maximum likelihood methodologies”.

Line 182 – Did you measure incongruence by the number of disagreements associated with lineage relationships?

Response: Yes, we say that connector lines are drawn between corresponding taxa to provide a level of entanglement (or lack of) between phylogenies.

Line 188 – extensive gene sequences

Response: Fixed.

General Question and Statement – I looked at the phylogeny in Figure 1, and the phylogeny is resolved. What I don’t see is any value associated with the different branches. Therefore, I wonder how much support there is for a particular branching pattern. Bootstrap values and posterior probabilities are one way to show support, but low values are not considered very informative. In addition, you talk about “statistical support” for nodes, but the term is somewhat misused because this is not statistics in the sense of have p-values but rather non-statistical approximations of an infinite number of possibilities.

Response: In our Bayesian analysis, each node in the phylogeny is evaluated for Bayesian posterior probability support. This is a way to determine whether the node shows Bayesian statistical support of above or below 0.95. Each supported node indicates greater than or equal to 0.95 Bayesian statistical support (= a p-value of 0.05 or below). All other unsupported nodes indicate less than 0.95 Bayesian statistical support (= a p-value of above 0.05).

Another question – I looked at the previous Zan paper and noted that in terms of amino acid replacements among divergent lineages, there are very few. This suggests that most substitutions are synonymous and occur at the third position. Is this true? What is the rate of synonymous and non-synonymous substitutions for Zan? The overall substitution rate must be reasonable slow.

Response: In our previous study, we only showed very limited amino acid sequence alignments among very closely related rodent species. For most genes, such a limited sequence comparison would typically yield not even one amino acid substitution. Therefore, the changes shown in that figure actually do reflect rapid evolution of Zan. To clarify this point, we included a new supplemental figure (S2) showing pairwise sequence identities among individual species from the 17 Orders represented in this study. Zan orthologous amino acid sequence identities between distantly related species were barely above 50% reflecting the rapid evolution of zonadhesin protein sequence by positive selection.

Lines 224-225 – I am a little confused on the idea of “overall statistical support.” Is this an estimate of the number of bifurcations versus polytomies?

Response: Yes, “overall statistical support” is a measure of number of nodes with Bayesian statistical support equal to or greater than 0.95 versus the number of polytomies, which by definition have a Bayesian statistical support to less than 0.95. If a node has a value less than 0.95, then we must collapse that node into a polytomy. Then, for each tree, we count the number of supported nodes and then use the percent as a measure of overall statistical support for that tree.

Lines 234-235 – You note that Zan and Tex14 provide the highest support at deep nodes and are under the highest positive selection. This is interesting because these results that these two genes evolve slowly, which is surprising Zan’s support for more recent speciation events. Have you looked at overall patterns of divergence across Zan? Are the sites informative for more recent bifurcations more likely to be neutral? Alternatively, are their non-variable and variable regions across Zan?

Response: Zan and Tex14 are evolving under intense positive selection; thus, they evolve very rapidly. The reason these genes can track the evolutionary history of placental mammal species is not due to their evolving slowly as is the case with other genes, rather it is due to their function as speciation genes. The molecular evolution of these genes, i.e., the variation in their gene sequences, directly contribute to their function in the process of speciation. In addition, Zan and Tex14 are unique in that they have shown an ability to track the evolutionary history of species at both the terminal branches (closely-related species) and base of the tree (distantly-related species), a task where other genes fall short.

Overall, I think that this paper makes a significant contribution to our understanding of mammalian phylogenetics. I must admit, however, that the details in the results (e.g., comparison of result of various comparisons) was long and burdensome. Unfortunately, I have no specific ideas as to how one fixes this, other than placing some of the details in the supplementary section. One item, which I consider very important, is the overall need to understand more about the pattern of sequence divergence in the Zan gene. For example, which regions are likely to be under strong selection? Which types of substitutions provide the most support for various mammalian relationships? I see an overall need for a more detailed assessment of the molecular evolution of this gene, which is beyond the scope of this paper.

Response: Thank you for that suggestion. We placed the detailed results in the SI. Relevant to Zan sequence divergence, we have multiple ongoing studies right now on characterizing positive selection, both pervasive and episodic, across not only sites within the gene itself but also across different mammalian taxa. In those manuscripts, we will address different regions under varying positive selective forces, which substitutions and their corresponding change(s) in function contribute to speciation in multiple species groups, and any other relevant findings on Zan molecular evolution.

Reviewer 2 Report

This paper attempts to use the Zan gene, which codes for a protein involved in reproductive sperm-egg recognition, to resolve various levels of phylogenetic relationships within Eutheria. While the idea is novel and the overall study is sound, I would point out some issues which could be addressed in a revision.

L52: Maybe the identity of the 4 major clades (Afrotheria, Xenarthra, Euarchontoglires, Laurasiatheria) could be mentioned here to avoid confusion.

L98: "Genes that evolve under neutrality" should be more "genomic regions ..." because "genes" imply that they are somehow functional. One may also want to mention that neutral sequences are useful only for more closely related species because of (a) mutation saturation and (b) alignment difficulty caused by structural variation.

L126: What programs were used for making the trees? Also, Table S1 is incorrectly referenced here, as the actual Table S1 is the list of "Orders with limited taxon representation and therefore no tanglegrams".

L163: Can you explain what is "tanglegram"?

L171-180: What is the "n" in (n=x)? Is it number of trees, or number of taxa/species?

Figure 1: Why is it "Atlantogeneta" on the right side, when the tree supports Boreoeutheria and Afrotheria together as a clade (Epitheria)? Figure S1 also has armadillo and sloth at the base of the tree.

Figure 2, L275-276: The explanation for "difference in support" is confusing. Do you mean how much the family-level or deeper nodes are supported, compared to the average of all nodes, within the same tree?

L290: As above. Figure 1 and S1 has Xenarthra at the base of Eutheria to the exclusion of all other eutherians, which is inconsistent with the text here.

L301: Should be Figures 3-4.

L356: Anomaluromorpha is not an Order, but a Suborder within Rodentia.

Figure 9: I have never heard of Caniformia no longer being a monophyletic group. Is there any other literature supporting this? If not, this would be an extraordinary finding and requires more explanation.

Figure 12: The Zan tree placing Gorilla with Nomascus is highly irregular, and breaks the monophyly of Hominidae. As overwhelming evidence suggest the current consensus of Hominidae tree (Homo+Pan+Gorilla clade), almost certainly the Zan tree is "wrong" here. This could put doubt in accuracy of the Zan tree in general. This should be talked about in the discussion near L840.

L629: All but two trees (2~54)?



Major comments:

(1) The main problem of using a single gene is of course incomplete lineage sorting (ILS). The idea of using a reproductive recognition gene is that this gene would evolve quickly during a speciation event and keep mostly unchanged without speciation events. The problem is, speciation events do not have to occur "one-by-one". For example, suppose there is one species with very wide geographic distribution, and speciation occurs at two edges of distribution at similar time point (e.g., due to a glacial event splitting populations). The three resulting species could still be an unresolvable polytomy, and ILS could cause different genes (even ones like Zan) to produce different gene trees.

(2) Similarly, not all speciation events results in clear reproductive barrier. In mammals there are many genera within which interspecific hybridization is possible and introgression has been observed. That would likely mean that these speciation events do not cause a directional divergence in reproduction recognition genes like Zan. This should be mentioned in the paper.

(3) Please split Discussion into subsections. 300 lines of continuous text is not very readable.

(4) In general, the Discussion section has much of its content describing the Orders one by one, with some content a bit redundant with Results. Maybe the authors should focus here more on comparison of the tree topologies here with the literature, particularly when there is a clear incongruence with commonly accepted phylogeny (see: Canidae at base of Carnivora, placement of Gorilla etc). There should be also more discussion on methodology, particularly how to minimize effects of ILS and introgression.

Author Response

Reviewer #2

This paper attempts to use the Zan gene, which codes for a protein involved in reproductive sperm-egg recognition, to resolve various levels of phylogenetic relationships within Eutheria. While the idea is novel and the overall study is sound, I would point out some issues which could be addressed in a revision.

Thank you for your thoughtful comments. We believe that they helped strengthen the manuscript.

L52: Maybe the identity of the 4 major clades (Afrotheria, Xenarthra, Euarchontoglires, Laurasiatheria) could be mentioned here to avoid confusion.

Response: Added.

L98: "Genes that evolve under neutrality" should be more "genomic regions ..." because "genes" imply that they are somehow functional. One may also want to mention that neutral sequences are useful only for more closely related species because of (a) mutation saturation and (b) alignment difficulty caused by structural variation.

Response: Language added as suggested.

L126: What programs were used for making the trees? Also, Table S1 is incorrectly referenced here, as the actual Table S1 is the list of "Orders with limited taxon representation and therefore no tanglegrams".

Response: We specified how all of the trees were constructed, including the Zan Species-level tree, Family-level tree, and all tanglegram comparisons. Table S1 and S2 are now correctly referenced.

L163: Can you explain what is "tanglegram"?

Response: Venkatachalam et al. (2010) describes a tanglegram as a “pair of trees on the same set of leaves with matching leaves in the two trees joined by an edge”. In phylogenetics, two rooted phylogenetic trees are drawn opposite each other, using straight lines to connect matching taxa (Scornavacca et al. 2011). Essentially, a tanglegram visually depicts incongruities (“tangles”) between two phylogenies. We have included the description above incorporated into the relevant paragraph.

L171-180: What is the "n" in (n=x)? Is it number of trees, or number of taxa/species?

Response: “n” is number of trees for that taxonomic rank, i.e., Super-ordinal level, Ordinal level, Family level, etc. We specified this in that paragraph.

Figure 1: Why is it "Atlantogeneta" on the right side, when the tree supports Boreoeutheria and Afrotheria together as a clade (Epitheria)? Figure S1 also has armadillo and sloth at the base of the tree.

Response: Yes, Afrotheria and Boreoeutheria group together, so in Figure 1 we deleted “Atlantogenata” and changed Boreoeutheria to “Epitheria”. Armadillo and sloth are located at the base of the placental tree with platypus and echidna as the outgroup “basal” species.

Figure 2, L275-276: The explanation for "difference in support" is confusing. Do you mean how much the family-level or deeper nodes are supported, compared to the average of all nodes, within the same tree?

Response: In our Bayesian analysis, each node in the phylogeny is evaluated for Bayesian posterior probability support. This is a way to determine whether the node shows Bayesian statistical support of above or below 0.95 where a supported node indicates greater than or equal to 0.95 Bayesian statistical support (= a p-value of 0.05 or below) and an unsupported node indicates less than 0.95 Bayesian statistical support (= a p-value of above 0.05). To determine support for a given tree, we are counting up the number of nodes that have a Bayesian posterior probability of 0.95 or greater. We are requiring a Bayesian posterior probability of 0.95 or greater because it is a very stringent criterion for determining if a node is supported in phylogenetics. Then we compare the number of supported nodes for a given tree as an indicator of resolution. The higher the nodal support, the more resolved the tree. In addition, you can use this method to compare trees at different Family-level or deeper taxonomic levels to determine if a given gene resolves basal vs terminal nodes, or both, better than another gene. This method demonstrates the extraordinary resolution of the Zan tree (>95% of nodes supported).

L290: As above. Figure 1 and S1 has Xenarthra at the base of Eutheria to the exclusion of all other eutherians, which is inconsistent with the text here.

Response: As above, we changed the text to indicate that Xenarthra is at the base of Eutheria.

L301: Should be Figures 3-4.

Response: Changed, thank you for catching that.

L356: Anomaluromorpha is not an Order, but a Suborder within Rodentia.

Response: We deleted Anomaluromorpha from that sentence.

Figure 9: I have never heard of Caniformia no longer being a monophyletic group. Is there any other literature supporting this? If not, this would be an extraordinary finding and requires more explanation.

Response: Our Family-level Zan tree erroneously showed Hyaenidae and Herpestidae as members of Caniformia when in fact they are members of Feliformia. The Zan tree does confirm the monophyly of Caniformia. Thank you very much for catching that mistake.

Figure 12: The Zan tree placing Gorilla with Nomascus is highly irregular, and breaks the monophyly of Hominidae. As overwhelming evidence suggest the current consensus of Hominidae tree (Homo+Pan+Gorilla clade), almost certainly the Zan tree is "wrong" here. This could put doubt in accuracy of the Zan tree in general. This should be talked about in the discussion near L840.

Response: Agreed. We believe the spurious grouping of Gorilla with Nomascus reflects the availability of Zan sequence from only one member of the Family Hylobatidae. This is now addressed in the text.

L629: All but two trees (2~54)?

Response: We clarify that all trees (#2-55) but one (#1) were shown to be equally likely.

Major comments:

(1) The main problem of using a single gene is of course incomplete lineage sorting (ILS). The idea of using a reproductive recognition gene is that this gene would evolve quickly during a speciation event and keep mostly unchanged without speciation events. The problem is, speciation events do not have to occur "one-by-one". For example, suppose there is one species with very wide geographic distribution, and speciation occurs at two edges of distribution at similar time point (e.g., due to a glacial event splitting populations). The three resulting species could still be an unresolvable polytomy, and ILS could cause different genes (even ones like Zan) to produce different gene trees.

Response: We agree that these potential concerns must be considered. Yes, different genes tell different stories, reflecting their different functions in the adaptive evolution of incipient species, and thus giving rise to the ILS phenomenon. But because the function of Zan is to promote reproductive isolation regardless of whether speciation occurs sympatrically, peripatrically, allopatrically, etc., its divergence should necessarily reflect speciation independent of other selective pressures that also drive species divergence. Indeed, in the reviewer’s example the widely dispersed ancestral population would, because of rapid Zan evolution, likely harbor an uneven distribution of Zan alleles consistent with an ongoing contribution of Zan evolution to incipient speciation events. When geographic isolation (e.g. from the postulated glacial event) split the ancestral population, the pre-existing Zan alleles would sort randomly into the newly separated populations, with different allelic frequencies. Then, within those geographically isolated populations, Zan would continue to evolve as a speciation gene concordant with the further divergence of the populations into new species. Thus, we believe that by continually driving prezygotic reproductive isolation, Zan sequence divergence would reflect onset and progression of speciation irrespective of other contributing factors, and thus would be expected to sort true to lineage.

.

(2) Similarly, not all speciation events results in clear reproductive barrier. In mammals there are many genera within which interspecific hybridization is possible and introgression has been observed. That would likely mean that these speciation events do not cause a directional divergence in reproduction recognition genes like Zan. This should be mentioned in the paper.

Response: Yes, we agree that not all speciation events always generate complete reproductive isolation and that many mammal genera exhibit hybridization and introgression. We have noted this in the paper.

(3) Please split Discussion into subsections. 300 lines of continuous text is not very readable.

Response: We split Discussion into distinct subsections as suggested.

(4) In general, the Discussion section has much of its content describing the Orders one by one, with some content a bit redundant with Results. Maybe the authors should focus here more on comparison of the tree topologies here with the literature, particularly when there is a clear incongruence with commonly accepted phylogeny (see: Canidae at base of Carnivora, placement of Gorilla etc). There should be also more discussion on methodology, particularly how to minimize effects of ILS and introgression.

Response: We moved much of the detailed Results to the SI to remove some redundancy. We agree that gene tree heterogeneity is something to consider, which can occur due to many biological processes, such as incomplete lineage sorting and hybridization. The unique caveat about Zan is that it is unique because it accounts for ILS and hybridization due to its direct contribution in speciation? We did not minimize effects of those processes because Zan can track lineages as they speciate and even predict hybridization events using Zan sequence divergence (high sequence identity predicts recent or still occurring hybridization event).

Reviewer 3 Report

The article "Gamete Recognition Gene Divergence Yields a Robust Eutherian Phylogeny Across Taxonomic Levels" represents a contemporary and pertinent addition to the field of mammalian phylogenetics. The study incorporates state-of-the-art molecular methodologies, offering a novel perspective through the use of the Zan gene. This approach not only reaffirms established taxonomic relationships but also introduces new insights, particularly concerning taxa such as Dermoptera, Primates, Rodentia, and Lagomorpha, which have posed historical challenges in phylogenetic studies.

The identification of these persistently enigmatic taxa underscores the complex nature of phylogenetic research and underscores the necessity for further investigation. The article rightfully highlights the importance of continued exploration, which can extend to other taxonomic groups, such as Ungulates.

The outcomes of this study bear significance for mammalian taxonomy and systematics, providing a robust foundation for future research. The conclusions drawn from the utilization of the Zan gene exhibit the potential to refine the classification of these taxa. This, in turn, contributes to a deeper comprehension of reproductive isolation, hybridization, and biodiversity dynamics within eutherian mammals. In summary, this article stands as a timely and valuable scholarly contribution to the domain of mammalian phylogenetics.

Author Response

Reviewer #3

The article "Gamete Recognition Gene Divergence Yields a Robust Eutherian Phylogeny Across Taxonomic Levels" represents a contemporary and pertinent addition to the field of mammalian phylogenetics. The study incorporates state-of-the-art molecular methodologies, offering a novel perspective through the use of the Zan gene. This approach not only reaffirms established taxonomic relationships but also introduces new insights, particularly concerning taxa such as Dermoptera, Primates, Rodentia, and Lagomorpha, which have posed historical challenges in phylogenetic studies.

The identification of these persistently enigmatic taxa underscores the complex nature of phylogenetic research and underscores the necessity for further investigation. The article rightfully highlights the importance of continued exploration, which can extend to other taxonomic groups, such as Ungulates.

The outcomes of this study bear significance for mammalian taxonomy and systematics, providing a robust foundation for future research. The conclusions drawn from the utilization of the Zan gene exhibit the potential to refine the classification of these taxa. This, in turn, contributes to a deeper comprehension of reproductive isolation, hybridization, and biodiversity dynamics within eutherian mammals. In summary, this article stands as a timely and valuable scholarly contribution to the domain of mammalian phylogenetics.

Response: Thank you very much for the above comments. We have incorporated some of these ideas into the manuscript.

Reviewer 4 Report

I am impressed by the idea and soundness of the paper, grat job and fantastic ms!

I am happy I was a reviewer of a such well written and sound manuscript.

 I have only five comments that may improve the manuscript and make it highly citable.

1. would it be possible you provide primers to amplify zonadhesin (ZAN) gene? This could be just a citation.Please notice primer sequences to amplfy ZAN gene if provided in this paper, they  may be often cited/used.

2. Since ZAN gene is a speciation gene, could you please provied examples of amino acid alignments of selected taxa or groups? I do not mean to extent significantly this large manuscript by adding protein trees, I just would be happy to see whether the ZAN as a speciation gene has AA substitutions - indeed, it has to! - could you please rovide 1-3 AA alignments in supp material for selected taxa/groups?

3. Trees for Primates (fig. 12)  could be wrong (both, ZAN and supertree). Iam not an expert on this gropup, but threes form your ms are strinkingly different form trees form the paper by Peelman et al 2011 (https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1001342) - please check it  to avoid confusion or explain inconsistiencies. I mean: doreally Pan and Homo pair group with a clade including Trachypithecus and others? and not Gorilla?

4. (minor comment): page 25, line 953 and 954: Toward this end, it may be that Zan is the preeminent molecular marker that defines genetic [please add here: mammalian] species.

5. This is rather a question than suggestion to the ms: is it possible to use ZAN sequences to infer intraspecific mammalian phylogenies, eg, subspecies, something similar to mitochondrial DNA lineages, etc? If you have any results of intraspecific mammalian phylogenies, it would be great to have a very small paragraph on it  - this may have a great value for readers  as ongoing speciation process could be indeed initated by ZAN polymorphism as you mention in Introdution. Thus, one should expect some ZAN divergence among genetically distinct subspecies within a given mammalian species. This is however out of a scope of this manuscript.

Good luck with the paper!

- best regards,

Author Response

Reviewer #4

I am impressed by the idea and soundness of the paper, grat job and fantastic ms!

I am happy I was a reviewer of a such well written and sound manuscript.

 I have only five comments that may improve the manuscript and make it highly citable.

Thank you for the below comments. We believe that they helped to strengthen the manuscript.

1. would it be possible you provide primers to amplify zonadhesin (ZAN) gene? This could be just a citation.Please notice primer sequences to amplfy ZAN gene if provided in this paper, they  may be often cited/used.

Response: Thank you for that suggestion. The Zan DNA sequences used for the tree spanned 15+ exons, so primers for those sequences are not available. In addition, Zan is rapidly evolving so its gene sequences are diverging fast enough that it’s difficult if not impossible to design primers that work on all species.

2. Since ZAN gene is a speciation gene, could you please provied examples of amino acid alignments of selected taxa or groups? I do not mean to extent significantly this large manuscript by adding protein trees, I just would be happy to see whether the ZAN as a speciation gene has AA substitutions - indeed, it has to! - could you please rovide 1-3 AA alignments in supp material for selected taxa/groups?

Response: We performed a Zan protein alignment (Figure S2) including a representative species from each placental mammalian Order with duck-billed platypus ZanL sequence as the outgroup.

3. Trees for Primates (fig. 12)  could be wrong (both, ZAN and supertree). Iam not an expert on this gropup, but threes form your ms are strinkingly different form trees form the paper by Peelman et al 2011 (https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1001342) - please check it  to avoid confusion or explain inconsistiencies. I mean: doreally Pan and Homo pair group with a clade including Trachypithecus and others? and not Gorilla?

Response: We believe the spurious grouping of Gorilla with Nomascus reflects the availability of Zan sequence from only one member of the Family Hylobatidae. This is now addressed in the text.

4. (minor comment): page 25, line 953 and 954: Toward this end, it may be that Zan is the preeminent molecular marker that defines genetic [please add here: mammalian] species.

Response: Added.

5. This is rather a question than suggestion to the ms: is it possible to use ZAN sequences to infer intraspecific mammalian phylogenies, eg, subspecies, something similar to mitochondrial DNA lineages, etc? If you have any results of intraspecific mammalian phylogenies, it would be great to have a very small paragraph on it  - this may have a great value for readers  as ongoing speciation process could be indeed initated by ZAN polymorphism as you mention in Introdution. Thus, one should expect some ZAN divergence among genetically distinct subspecies within a given mammalian species. This is however out of a scope of this manuscript.

Response: Yes, Gasper and Swanson (2006), doi: https://doi.org/10.1086/508473, previously characterized Zan diversity in primate species, including polymorphisms among 48 humans. The study identified numerous human allelic variants consistent with the gene’s rapid evolution by positive selection. Furthermore, we are currently completing a study of human ZAN RNA splice variation (Ashton et al. unpublished), that also contributes to individual variation in expression of ZAN transcripts. All of these processes likely promote diversification of Zan gene products that in turn drive speciation.

Good luck with the paper!

Response: Thank you!

- best regards,