Enhanced heterozygosity from male meiotic chromosome chains is superseded by hybrid female asexuality in termites

Abstract

Although males are a ubiquitous feature of animals, they have been lost repeatedly in diverse lineages. The tendency for obligate asexuality to evolve is thought to be reduced in animals whose males play a critical role beyond the contribution of gametes, for example, via care of offspring or provision of nuptial gifts. To our knowledge, the evolution of obligate asexuality in such species is unknown. In some species that undergo frequent inbreeding, males are hypothesized to play a key role in maintaining genetic heterozygosity through the possession of neo-sex chromosomes, although empirical evidence for this is lacking. Because inbreeding is a key feature of the life cycle of termites, we investigated the potential role of males in promoting heterozygosity within populations through karyotyping and genome-wide single-nucleotide polymorphism analyses of the drywood termite Glyptotermes nakajimai We showed that males possess up to 15 out of 17 of their chromosomes as sex-linked (sex and neo-sex) chromosomes and that they maintain significantly higher levels of heterozygosity than do females. Furthermore, we showed that two obligately asexual lineages of this species-representing the only known all-female termite populations-arose independently via intraspecific hybridization between sexual lineages with differing diploid chromosome numbers. Importantly, these asexual females have markedly higher heterozygosity than their conspecific males and appear to have replaced the sexual lineages in some populations. Our results indicate that asexuality has enabled females to supplant a key role of males.

Keywords: genetic heterozygosity; hybrid asexuality; inbreeding; neo-sex chromosomes.

Fig. 1.

Population genetic structure in G. nakajimai. (A) Map showing the sampling sites of six asexual (all-female) populations (Ashizuri [AS], Muroto [MR], Tokushima [TK], Sata [ST], Toi [TI], and Saiki [SK]) and four sexual populations (Kushimoto [KS], Amami-Oshima Island [AM], Okinawa Island [OK], Ogasawara Islands [OG]) across Japan. (B) PCoA of 84 individuals from field colonies of asexual (10 or 2 female workers from a field colony in each of 6 populations) and sexual (5 male and 5 female workers from a field colony in each of 4 populations) G. nakajimai based on genetic distance calculated using 4,191 SNPs, resulting in three distinct groups: asexual lineage 1 (AL1), asexual lineage 2 (AL2), and sexual lineage 1 (SL1). PC1 and PC2 are the first and second principal coordinates, respectively, and the numbers in parentheses refer to the proportion of variance explained by the principal coordinates. (C) Structure clustering of the six asexual and four sexual populations using 4,191 SNP markers obtained for K = 2 (Top) and K = 10 (Bottom).

Fig. 2.

Enhanced heterozygosity in males by male meiotic chromosome chain formation and markedly higher heterozygosity in asexual females than males and sexual females in G. nakajimai. (A) Mitotic (Left) and meiotic (Right) chromosomes of a male from the Okinawa Island population of the G. nakajimai sexual lineage 1 (SL1). A diploid chromosome complement of 2n = 34 is seen in members of this and other populations of SL1 (ref. 22). Meiotic chromosomes show the characteristic chain formation of a subset of chromosomes (arrow), as seen commonly in kalotermitid termites (refs. , , and 30). The male meiotic chromosome complement includes a chain of 30 chromosomes, which is predicted to comprise 15 Y and neo-Y chromosomes and 15 X and neo-X chromosomes, plus 2 bivalents. At the end of meiosis, all Y and neo-Y chromosomes are expected to be inherited together into one gamete, while all X and neo-X chromosomes are expected to be inherited together into a separate gamete. Each gamete also inherits one copy of each non–sex-linked chromosome in a random fashion. (B) Comparison of the percentage of heterozygous SNP loci between males of SL1 (n = 20), females of SL1 (n = 20), females of the G. nakajimai asexual lineage 1 (AL1) (n = 33), and females of the G. nakajimai asexual lineage 2 (AL2) (n = 11). Values are mean ± SEM. Different letters on the bars indicate significant differences (P < 0.001, Tukey’s HSD test following nested ANOVA [colony: F₁₂, ₆₈ = 27.58, P < 0.0001; subject: F₃, ₆₈ = 12,482, P < 0.0001; nested ANOVA with colonies nested within subjects]).

Fig. 3.

Contemporary gene flow and migration rates between populations of G. nakajimai estimated from the SNP data using BayesAss. Arrows indicate direction of gene flow among populations. Values are mean rates. Only gene flows significantly greater than zero are shown. Distribution of the lineages was estimated by SNP genotyping. AL1, the G. nakajimai asexual lineage 1; AL2, the G. nakajimai asexual lineage 2; SL1, the G. nakajimai sexual lineage 1.