Soldier Ants and Caste Evolution

            In many species of the ant genus Pheidole, there are three different castes: queens, soldiers, and minor workers. These three castes are determined by two juvenile hormone (JH) mediated switches in response to environmental cues.  At the first switch the queen and the two other castes are divided, and at the second switch the soldier and the worker castes are separated (see Fig. 1). Queens have four wings, whereas soldiers and minor workers do not have any. Soldiers are bigger than minor workers and have one set of vestigial wing discs, whereas minor workers lack them (see Fig. 1).

Fig. 1: Wing polyphenism in P. morrisi: the ability of a single genome to produce (A) winged queens and wingless (B) soldiers and (C) minor workers. Caste determination occurs at two JH-mediated switch points in response to environmental cues. (D) Wing discs in queen larvae showing conserved hinge and pouch expression of sal. (E) Vestigial wing discs in soldier larvae showing a soldier-specific pattern of sal expression, where it is conserved in the hinge but down-regulated in the pouch. Asterisks represent the absence of visible wing discs and sal expression in (E) soldier and (F) minor worker larvae. Scale bars indicate the relative sizes of queen, soldier, and minor worker larvae and adults. From Rajakumar et al. 2012.

           

In a recent Science paper, Rajakumar et al. (1) showed that the formation of wings of queens is generated by the expression of the sal gene in the wing hinges and pouches (Fig. 1), whereas in soldiers sal is expressed only in the hinges of vestigial discs and minor workers show no gene expression. Interestingly, some species have a subclass called supersoldiers, which are bigger than regular soldiers and have two sets of vestigial wing discs, where the gene sal is expressed. It is known that this type of supersoldiers occasionally appear as mutant forms in the regular three-caste species. Rajakumar et al. then conducted experiments to study whether supersoldiers can be generated in a three-caste species when methoprene (a JH analog) was applied after soldiers and minor workers are separated. They then showed that the application of methoprene induces supersoldiers.

This observation suggests that all species of Pheidole has the potential for producing a subclass of supersoldiers. They constructed a phylogenetic tree for the eleven species examined and inferred the evolutionary changes of the occurrence of supersoldiers (Fig. 2). Their conclusion was that the ancestral species of this group of ants probably had the gene sal but this gene was lost in some descendant lineage. However, the developmental pathway leading to supersoldiers was retained in the genome. Therefore, when a backward mutation occurred at the sal locus in one of the descendant species (P. obtusospinosa), the subcaste of supersoldiers was restored (Fig. 2). They tested this hypothesis by applying methoprene to several three-caste species and found that supersoldiers are indeed inducible by the reactivation of the sal gene.



Fig. 2: Evolutionary history of ancestral developmental potential and phenotypic expression of supersoldiers (XSDs). MYA, million years ago. Purple represents the pattern of sal expression; asterisks indicate the absence of vestigial wing discs and sal expression. Green arrows and boxes represent the induction of XSD potential. From Rajakumar et al. 2012.



This study suggested that apparently independent evolution of a caste system can be due to mutations of regulatory genes controlling the downstream developmental pathways. If this is the case, the various forms of caste systems existing in Hymenoptera could be due to different signal proteins generated by mutations. Kamakura (2) showed that the first switch gene in honeybees is royalactin, but different switch genes may be used in different groups of hymenopteran species. Furthermore, different numbers of switch genes may be involved in different groups. Ethologists seem to believe that different caste systems must have evolved by natural selection. However, if we consider that there are various types of sex determination in insects and some changes among them could be largely due to genetic drift (3), we may have to entertain the possibility of non-adaptive changes of caste systems.

Behavioral evolution is one of the most complex research areas of evolution, and I am hoping that modern molecular biology will be able to solve this problem.

References

1. Rajakumar R, San Mauro D, Dijkstra MB, Huang MH, Wheeler DE et al. 2012. Ancestral developmental potential facilitates parallel evolution in ants. Science 335:79-82.

2. Kamakura M. 2011. Royalactin induces queen differentiation in honeybees. Nature 473:478-483.

3. Gempe T and Beye M. 2010. Function and evolution of sex determination mechanisms, genes and pathways in insects. Bioessays 33:52-60.