Differential use of multiple genetic sex determination systems in divergent ecomorphs of an African crater lake cichlid
Curation statements for this article:-
Curated by eLife
Evaluation Summary:
This paper will be of interest to evolutionary biologists and geneticists, particularly those interested in the evolution of sex determination and sexual conflicts. It provides an unprecedented dataset that enables the authors to show convincingly the presence of three different Y-chromosomes segregating within a species, differential presence of the Ys among ecomorphs, and identifies candidate sex determination genes on the different Ys. Examination of the impact of genetic sex on a male fitness proxy in ecological context provides a compelling case study to explain the stable maintenance of multiple genetic sex determination systems in a species.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1, Reviewer #2 and Reviewer #3 agreed to share their name with the authors.)
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
Abstract
African cichlid fishes not only exhibit remarkably high rates of speciation but also have some of the fastest evolving sex determination systems in vertebrates. However, little is known empirically in cichlids about the genetic mechanisms generating new sex-determining variants, what forces dictate their fate, the demographic scales at which they evolve, and whether they are related to speciation. To address these questions, we looked for sex-associated loci in full genome data from 647 individuals of Astatotilapia calliptera from Lake Masoko, a small isolated crater lake in Tanzania, which contains two distinct ecomorphs of the species. We identified three separate XY systems on recombining chromosomes. Two Y alleles derive from mutations that increase expression of the gonadal soma-derived factor gene ( gsdf ) on chromosome 7; the first is a tandem duplication of the entire gene observed throughout much of the Lake Malawi haplochromine cichlid radiation to which A. calliptera belongs, and the second is a 5 kb insertion directly upstream of gsdf . Both the latter variant and another 700 bp insertion on chromosome 19 responsible for the third Y allele arose from transposable element insertions. Males belonging to the Masoko deep-water benthic ecomorph are determined exclusively by the gsdf duplication, whereas all three Y alleles are used in the Masoko littoral ecomorph, in which they appear to act antagonistically among males with different amounts of benthic admixture. This antagonism in the face of ongoing admixture may be important for sustaining multifactorial sex determination in Lake Masoko. In addition to identifying the molecular basis of three coexisting sex determining alleles, these results demonstrate that genetic interactions between Y alleles and genetic background can potentially affect fitness and adaptive evolution.
Article activity feed
-
Evaluation Summary:
This paper will be of interest to evolutionary biologists and geneticists, particularly those interested in the evolution of sex determination and sexual conflicts. It provides an unprecedented dataset that enables the authors to show convincingly the presence of three different Y-chromosomes segregating within a species, differential presence of the Ys among ecomorphs, and identifies candidate sex determination genes on the different Ys. Examination of the impact of genetic sex on a male fitness proxy in ecological context provides a compelling case study to explain the stable maintenance of multiple genetic sex determination systems in a species.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the …
Evaluation Summary:
This paper will be of interest to evolutionary biologists and geneticists, particularly those interested in the evolution of sex determination and sexual conflicts. It provides an unprecedented dataset that enables the authors to show convincingly the presence of three different Y-chromosomes segregating within a species, differential presence of the Ys among ecomorphs, and identifies candidate sex determination genes on the different Ys. Examination of the impact of genetic sex on a male fitness proxy in ecological context provides a compelling case study to explain the stable maintenance of multiple genetic sex determination systems in a species.
(This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1, Reviewer #2 and Reviewer #3 agreed to share their name with the authors.)
-
Reviewer #1 (Public Review):
Through sequencing the genomes of nearly 650 individuals of a single species of cichlid fish within a lake, this study provides an unprecedented molecular view of the surprising extent of variation in sex-determination systems that can exist within a single species. The authors further find evidence for a complex interaction between genetic background, location along the depth gradient and sex determination system on male body size. Although the authors speculate that these interactions are relevant for fitness in the different habitats and the maintenance of the polymorphism in sex determination within the species, these arguments are mostly indirect and somewhat weaker. Still, this is a unique study with an impressive dataset that provides the foundation for gaining further insights into how variation in …
Reviewer #1 (Public Review):
Through sequencing the genomes of nearly 650 individuals of a single species of cichlid fish within a lake, this study provides an unprecedented molecular view of the surprising extent of variation in sex-determination systems that can exist within a single species. The authors further find evidence for a complex interaction between genetic background, location along the depth gradient and sex determination system on male body size. Although the authors speculate that these interactions are relevant for fitness in the different habitats and the maintenance of the polymorphism in sex determination within the species, these arguments are mostly indirect and somewhat weaker. Still, this is a unique study with an impressive dataset that provides the foundation for gaining further insights into how variation in sex determination systems evolves and is maintained.
Here I will summarize the strength of the evidence that supports each of the major conclusions of the manuscript:
1. Multiple Y alleles determine sex in Lake Masoko: Here the data are very strong; the GWAS analyses is well-done, and the data presented to support the duplication of the gsdf locus in the most prevalent Y-allele is convincing.
2. Molecular nature of the three sex-determination loci: In all three cases, there is circumstantial evidence supporting the molecular nature of the sex determination gene; i.e. duplication of and insertion of a transposable element next to an excellent candidate sex determination gene (gsdf) and insertion of a transposable element next to a gene (id3) previously not associated with sex determination. Nonetheless, there is no functional data (beyond expression data for gsdf) showing that these mutations are causative, and so the authors should be more circumspect in their claims that they have identified the molecular nature of these alleles.
3. Differential use of Y alleles in Lake Masoko: It is clear that the different Y-alleles are present at different frequencies in the different "genetic clusters" but these are somewhat arbitrarily defined, and PC1 only explains 2.16% of the overall variation in genomic variation. The argument that the benthic and littoral populations are genetically distinct and reproductively isolated, with a bias in gene flow from the benthic into the littoral population, is not so clear from the data presented and seems to rely somewhat on data presented in Malinsky et al. 2015, but this previous evidence is not clearly summarized in the present manuscript.
4. Antagonism between Y alleles and admixture: These are interesting results but not so clear to interpret. Males in the admixture zone with high levels of benthic ancestry and the benthic Y allele (gsdf-dup) are smaller, particularly if caught in the deeper water. But, all of these conditions would seem to favor the gsdf-dup allele, which is fixed in the deeper water benthic population. Furthermore, the frequency of gsdf-dup males does not differ between low and middle PC1 males, suggesting that there is not a selective difference between them. The authors argue that patterns of LD on the other two alleles are suggestive of selection on them, but they do not show patterns of LD on the gsdf-dup allele, which could also be under selection in the benthic population. This hypothesis would be consistent with the results of Malinsky et al. 2015, suggesting that the benthic population is derived from the littoral population. So, one could also argue that the littoral population is segregating for ancestral variation, and there has been directional selection for the gsdf-dup allele in benthic population.
5. Distribution of sex-determining alleles across the Lake Malawi radiation: These are quite nice data supporting that these three alleles are widespread either across species (gsdf-dup) or geographically within A. callipterus (chr19-ins, chr7-ins).
-
Reviewer #2 (Public Review):
The manuscript by Munby et al. presents genomics studies from a crater lake population of the Eastern happy, Astatotilapia calliptera. The data support convincingly the presence of three different Y-chromosomes in the population and also show a differential presence of the Ys among the benthic and littoral ecomorphs. While previous studies showing multiple sex determination systems in African cichlid species were based on captive-breeding experiment, the work by Munby et al. now is the first investigation on how sex determination acts in a natural population. Thus, this manuscript provides novel information, which is not only relevant for understanding the evolutionary ecology of this important group of fishes, but which is also very interesting for all those working on the evolution of sex determination …
Reviewer #2 (Public Review):
The manuscript by Munby et al. presents genomics studies from a crater lake population of the Eastern happy, Astatotilapia calliptera. The data support convincingly the presence of three different Y-chromosomes in the population and also show a differential presence of the Ys among the benthic and littoral ecomorphs. While previous studies showing multiple sex determination systems in African cichlid species were based on captive-breeding experiment, the work by Munby et al. now is the first investigation on how sex determination acts in a natural population. Thus, this manuscript provides novel information, which is not only relevant for understanding the evolutionary ecology of this important group of fishes, but which is also very interesting for all those working on the evolution of sex determination mechanisms and sex chromosomes.
Mapping of the GWAS to the high-quality reference genome of A. calliptera uncovered candidate genes for being the master regulators of male sexual development and even potential mechanisms of action of the sex determining gene. However, here I see a problem with the current version of the manuscript and how this part of the work is presented. I fully concur with the authors that their data suggest these genes as likely candidates and also a way how they are activated. But much more work is needed to provide the full experimental evidence to call them bona-fide sex determining (SD) genes. What the authors have found is the genetic evidence for association of these genes with the sex phenotype, but this is only half of the story. A SD gene has to be shown to be expressed at the right time at the right place, and a molecular mechanism has to act accordingly. Functional studies by genome modifications then bring final clarity. As this is not available for the gsdf and id3 candidates in the current study, a much more careful wording is necessary - or the experimental proof provided by inclusion of additional experiments. -
Reviewer #3 (Public Review):
Munby and Linderoth et al. examine genetic sex determination in East African cichlids in four contexts: 1) Genome-wide association mapping in a study population of the species Astatotilapia calliptera from Lake Masoko that identified three putative Y alleles, two at the gsdf gene, and one at the id3 gene, 2) Analysis of expression of these two genes in various tissues by sex genotype, 3) Comparison of body size as a fitness measure by sex genotype and ecomorph, and 4) A survey of A. calliptera populations and numerous Lake Malawi cichlid species for a broader view of use of the Y alleles in the species radiation. Taken together, the findings seem sufficient to make the claim that the gsdf is the first sex determination gene identified for a non-tilapiine cichlid, noteworthy because cichlids have complex and …
Reviewer #3 (Public Review):
Munby and Linderoth et al. examine genetic sex determination in East African cichlids in four contexts: 1) Genome-wide association mapping in a study population of the species Astatotilapia calliptera from Lake Masoko that identified three putative Y alleles, two at the gsdf gene, and one at the id3 gene, 2) Analysis of expression of these two genes in various tissues by sex genotype, 3) Comparison of body size as a fitness measure by sex genotype and ecomorph, and 4) A survey of A. calliptera populations and numerous Lake Malawi cichlid species for a broader view of use of the Y alleles in the species radiation. Taken together, the findings seem sufficient to make the claim that the gsdf is the first sex determination gene identified for a non-tilapiine cichlid, noteworthy because cichlids have complex and rapidly evolving sex determination systems and thus provide a model for evolution of sex determination. Identification of this widespread sex determination gene is extremely valuable for future studies re-examining its impact in various contexts, including where it interacts with other sex determination systems. Accompanying gene expression analysis presented here demonstrates that Y alleles of gsdf produce higher levels of gsdf transcript in somatic tissues, but does not provide appropriate comparisons in gonadal tissue to provide direct support for a role for gsdf in sex determination. The lack of gonadal expression analysis does not seriously impact the claim that gsdf is the master sex determination gene however, given replicated association with the gsdf Y alleles and sex in other cichlid species surveyed here, and the known role of gsdf as a sex determiner in other fish species. Genetic mapping results are also used to compare inferred fitness of males of different sex genotypes in a natural population, with results suggesting antagonism between different Y alleles and genetic background. These latter findings are noteworthy as providing an empirical example of interactions between sex determination, genetic variation, and ecology that could support polygenic sex determination as an evolutionarily stable strategy. Below I provide strengths and weaknesses of each of the four contexts listed above.
Association mapping
Strengths: Association mapping used a sufficiently large population to allow a stepwise mapping strategy to identify three putative sex determination alleles. The authors provide a careful analysis of genetic variation at the two loci involved including analysis involving use of long-read technology and analysis of coverage to identify copy number variants and transposable element insertions associated with sex determination. The associations with gsdf fall within a previously mapped interval for an XY system in the species, so there is concordance with previous work.
Weaknesses: Potential sequence variation for the associated genes does not appear to be described. It is unclear if there are sequence variants in the coding or untranslated regions of gsdf or id3 associated with X vs. Y alleles, or between the two copies of gsdf in the case of the gsdf duplication allele. Such polymorphism could impact protein function, transcription and translation, or provide landmarks for additional gene expression or gene evolution analyses.
Gene expression
Strengths: Expression is assessed for gsdf in four different male genotypes in four somatic tissues, with results consistent with upregulation of gsdf on the two Y alleles identified on chr7. Their findings suggest that evolution of gsdf as the master sex determination gene resulted in the gene being upregulated in numerous tissues throughout the organism, which could have unknown but significant pleiotropic effects.
Weaknesses: Analysis of expression of gsdf and id3 in the gonad is lacking, though gonad is the primary tissue of interest for sex determination via gsdf. Only two females and two males of unknown sex genotype are assayed for adult gonadal expression of gsdf, which provides little information; sexually dimorphic gsdf expression in the gonad would be expected in any fish, regardless of their master sex determination gene. For the somatic tissue comparisons, gsdf expression levels for females are not included, which would importantly demonstrate if, and to what degree, gsdf expression is sexually dimorphic. Allele-specific expression analysis is not included, which would more definitively support claims of regulatory evolution. There is an in-text mention that no differences in id3 expression were found, but no data is provided and the comparisons made are unclear. Overall, the gene expression data supports links between the described Y-linked gsdf alleles and gsdf expression in somatic tissues, but does not by itself indicate a role for gsdf in gonadal sex determination.
Genetic sex and fitness
Strengths: Sex genotype is shown to impact body size (a proxy of male fitness) in nature in a manner that varies by ecology and genetic background, supporting a scenario where there are different optimal sex genotypes for males depending on genetic background associated with ecology (depth). These findings were only possible through careful and extensive sampling in a natural population, and contextualizing sex genotype and trait information within population structure, using whole genome analysis. Why and how polygenic sex determination systems are evolutionary stable remain standing questions, and the findings presented here could provide a straightforward example for how multiple XY systems can be maintained along an ecological gradient in nature.
Weaknesses: Body size differs by sex genotype when partitioned by genetic background, using a principle component score; however, it is unclear how the specific PC1 score was chosen to partition the littoral fish into low and middle groups, or how differences in that choice might impact results. Though male size is used as a standard proxy for fitness in cichlids, the actual impact of potential pleiotropic effects of sex genotype and/or body size on lifetime fitness in the study population is unclear; admittedly this is also not trivial to measure, to say the least.
Species survey
Strengths: The author provide a broad survey of the identified Y alleles in numerous populations and species, providing an important catalog of sex determination systems present across the Lake Malawi radiation, and showing evolutionary dynamics of sex determination in the species radiation. The apparent continued association of the Y alleles with phenotypic sex in numerous species helps provide confirmation of gsdf as a master sex determination gene.
Weaknesses: The results of the survey could be better presented and summarized to allow readers to understand the evolutionary dynamics of the Y alleles. It is difficult to extract much insight from the current supplemental table. For example, presenting all species surveyed in phylogenetic context with frequency of each Y by sex would provide an extremely valuable overview for readers, and reveal evolutionary patterns not clear in the current presentation. Also, some species are unexpectedly listed as having greater than four copies of gsdf, with as many as eleven copies; the authors do not discuss these results or their interpretation.
-
-
-
-