Current understanding of sex chromosome evolution is largely dependent on species with highly degenerated, heteromorphic sex chromosomes, but by studying species with recently evolved or morphologically indistinct sex chromosomes we can greatly increase our understanding of sex chromosome origins, degeneration, and turnover. Here, we examine sex chromosome evolution and stability in the gecko genus Aristelliger. We used RADseq to identify sex-specific markers and show that four Aristelliger species, spanning the phylogenetic breadth of the genus, share a conserved ZZ/ZW system syntenic with avian chromosome two. These conserved sex chromosomes contrast with many other gecko sex chromosome systems by showing a degree of stability among a group known for its dynamic sex determining mechanisms. Cytogenetic data from A. expectatus revealed homomorphic sex chromosomes with an accumulation of repetitive elements on the W chromosome. Taken together, the large number of female-specific A. praesignis RAD markers and the accumulation of repetitive DNA on the A. expectatus W karyotype suggests that the Z and W chromosomes are highly differentiated despite their overall morphological similarity. We discuss this paradoxical situation and suggest that it may, in fact, be common in many animal species.

We generated single-digest RADseq libraries for 10 males and 10 females of Aristelliger praesignis using a modified protocol from Etter et al. (2012) as described in Gamble et al. (2015). Libraries were sequenced using paired-end 125 bp reads on an Illumina HiSeq2500 at the Medical College of Wisconsin. We analyzed the RADseq data using a previously described bioinformatic pipeline (Gamble et al., 2015). Raw Illumina reads were demultiplexed, trimmed, and filtered using the process_radtags function in STACKS (1.41, Catchen, Amores, Hohenlohe, Cresko, & Postlethwait, 2011). We used RADtools (1.2.4, Baxter et al., 2011) to generate RADtags for each individual and identified candidate loci and alleles from the forward reads. We then used a custom python script (Gamble et al., 2015) to identify putative sex-specific markers from the RADtools output, i.e. markers found in one sex but not the other. The script also generated a list of “confirmed” sex-specific RAD marker that excluded any sex-specific markers found in the original read files of the opposite sex. Finally, we used Geneious (R11, Kearse et al., 2012) to assemble the forward and reverse reads of “confirmed” sex-specific RAD markers. 

We PCR amplified and Sanger sequenced the mitochondrial gene NADH dehydrogenase subunit 2 (ND2) and adjacent tRNAs with the primers L4437b (Macey, Larson, Ananjeva, & Papenfuss, 1997), L5005 (Jennings, Pianka, & Donnellan, 2003), and H5934a (Arevalo, Davis, & Sites Jr, 1994). We aligned these new sequences with previously published sequences (Gamble, Greenbaum, Jackman, Russell, & Bauer, 2012) in Geneious (R9.1.6, Kearse et al. 2012) using MUSCLE (Edgar, 2004) and inspected the resulting alignment by eye for errors. We constructed a maximum likelihood tree using RAxML (V8.2, Stamatakis, 2014) with a GTR+GAMMA model. Branch support was generated using 1000 bootstrap replicates. We included Quedenfeldtia trachyblepharus and Quedenfeldtia moerens as outgroups. Finally, we calculated between group genetic p-distance for putative species using MEGA X (V10.1, Kumar, Stecher, Li, Knyaz, & Tamura, 2018).