Loss of males from mixed-sex societies in termites

Abstract

Background: Sexual reproduction is the norm in almost all animal species, and in many advanced animal societies, both males and females participate in social activities. To date, the complete loss of males from advanced social animal lineages has been reported only in ants and honey bees (Hymenoptera), whose workers are always female and whose males display no helping behaviors even in normal sexual species. Asexuality has not previously been observed in colonies of another major group of social insects, the termites, where the ubiquitous presence of both male and female workers and soldiers indicate that males play a critical role beyond that of reproduction.

Results: Here, we report asexual societies in a lineage of the termite Glyptotermes nakajimai. We investigated the composition of mature colonies from ten distinct populations in Japan, finding six asexual populations characterized by a lack of any males in the reproductive, soldier, and worker castes of their colonies, an absence of sperm in the spermathecae of their queens, and the development of unfertilized eggs at a level comparable to that for the development of fertilized eggs in sexual populations of this species. Phylogenetic analyses indicated a single evolutionary origin of the asexual populations, with divergence from sampled sexual populations occurring about 14 million years ago. Asexual colonies differ from sexual colonies in having a more uniform head size in their all-female soldier caste, and fewer soldiers in proportion to other individuals, suggesting increased defensive efficiencies arising from uniform soldier morphology. Such efficiencies may have contributed to the persistence and spread of the asexual lineage. Cooperative colony foundation by multiple queens, the single-site nesting life history common to both the asexual and sexual lineages, and the occasional development of eggs without fertilization even in the sexual lineage are traits likely to have been present in the ancestors of the asexual lineage that may have facilitated the transition to asexuality.

Conclusions: Our findings demonstrate that completely asexual social lineages can evolve from mixed-sex termite societies, providing evidence that males are dispensable for the maintenance of advanced animal societies in which they previously played an active social role.

Keywords: Advanced social animals; All-female asexual societies; Asexual social lineages; Sexual reproduction; Social insects; Thelytokous parthenogenesis.

Fig. 1

The caste-developmental pathway in Kalotermitidae. L1–L5, first- to fifth-instar larva; N1–N2, first- to second-instar nymph

Fig. 2

Asexual and sexual populations of the termite Glyptotermes nakajimai. a Termite royals in mature colonies of asexual (top left) and sexual (top right) populations, and spermathecae of egg-laying queens without sperm in an asexual population (bottom left) and with sperm in a sexual population (bottom right). Spermathecae were stained by propidium iodide and observed under a confocal fluorescence microscope. Q, queen; K, king. Scale bars, 2 mm (top); 20 μm (bottom). b Geographical distribution of asexual and sexual populations. Each population may include more than one collection site if they are located less than 50 km apart. The number of colonies sampled in each population is shown in parentheses

Fig. 3

Increased hatching success of unfertilized eggs in asexual populations. Comparison of the percentage of eggs hatched within 100 days after colony foundation among unfertilized eggs of an asexual population (n = 134), unfertilized eggs of a sexual population (n = 193), and fertilized eggs of a sexual population (n = 127). Different letters on the bars indicate significant differences (P < 0.0001, Fisher’s exact probability test with Bonferroni correction). For raw data, see Additional file 7

Fig. 4

Evolutionary relationships among asexual and sexual populations of Glyptotermes nakajimai. Phylogenetic trees were obtained by Bayesian analyses of mitochondrial COII (left) and nuclear ITS2 (right) sequences of G. nakajimai individuals representing each of the collection sites. The asexual lineage is highlighted in red, and the sexual lineage is highlighted in blue. Posterior probabilities (≥ 0.70) are shown at each node. The horizontal bar represents a distance of 0.1 substitutions per site. Multiple Glyptotermes spp. as well as Cryptotermes domesticus and Kalotermes flavicollis were used as outgroups. GenBank accession numbers are shown in parentheses. The topologies shown were very similar to those derived from maximum likelihood analyses, with some minor differences (see Additional file 2: Figure S2)

Fig. 5

Chronogram showing divergence times among Kalotermitidae, including the asexual and sexual lineages of Glyptotermes nakajimai. The tree was inferred based on an alignment of mitochondrial COII sequences, using the program BEAST v1.8.2. The asexual lineage is highlighted in red, and the sexual lineage is highlighted in blue. Branch lengths are drawn to a time scale given in millions of years. Bars represent 95% confidence intervals for estimates of node times. Four fossil termites provided minimum age constraints for calibration of the molecular clock at the nodes denoted by asterisks (see “Methods” for further details concerning these fossils)

Fig. 6

Proposed karyotype of Glyptotermes nakajimai. Top: mitotic chromosomes of a female of the asexual lineage (left), a female of the sexual lineage (middle), and a male of the sexual lineage (right). MC, Mitotic chromosome. Bottom: mitotic karyotypes of a female of the asexual lineage (2n = 35) (left), a female of the sexual lineage (2n = 34) (middle), and a male of the sexual lineage (2n = 34) (right)

Fig. 7

Defensive advantage of colonies in the asexual lineage. a Cross section showing the structure of a Glyptotermes nakajimai nest. Small tunnels connecting chambers are indicated by arrows. (inset, right) A tunnel-blocking soldier (phragmotic defense). Scale bars, 2 mm. b Comparison of the within-colony coefficient of variation (CV) of soldier head width between the asexual and sexual lineages. Values are mean ± SEM (n = 5). Individual data points are represented by open circles. *, P < 0.05 (Mann–Whitney U test). For raw data, see Additional file 7. c Comparison of the proportion of soldiers to other individuals of mature field colonies between the asexual and sexual lineages. Values are mean ± SEM (n = 37). Individual data points are represented by open circles. ****, P < 0.0001 (GLM). For raw data, see Additional file 7