Coevolutionary dynamics and geographic mosaics in the Social Parasite Harpagoxenus sublaevis and its two Host Species

Social parasites such as bees, wasps and ants parasitize complete insect societies. They take advantage of the brood care behaviour of other social insect species, and thus avoid the costs of parental care similar to avian brood parasites such as cuckoos and cowbirds. The European social parasite Harpagoxenus sublaevis is an obligate slavemaking ant species that exploits mainly two closely related host species of the genus Leptothorax. To found a new colony, a slavemaking queen invades a host colony, kills the resident queen and workers. The inseminated queen raises the alien brood and the later emerging host workers accept the parasite queen as their own and become slaves that carry out all necessary colony tasks. A year later, slavemaking workers emerge, which conduct regular slave raids on neighbouring host colonies for worker brood to replenish the labour force of the slavemaker nest. These slave raids can impose severe selection pressure on the hosts, as slavemaking colonies attack several host nests per year and raided host colonies often perish as a consequence of the attack. According to the geographical mosaic theory of coevolution, differences in the advance or trajectory of the coevolutionray process between local communities are predicted due to their composition and the strength of ecological selection pressures through competition and resource availability. In our study system, investigations of the impact of the slave making ant H. sublaevis at several geographic distant sites allow general conclusions on the virulence, the degree of reciprocal adaptation and specialization of the species, and the evolutionary trajectories within this host-parasite system. The European slave making ant H. sublaevis and its host species are good examples as parasites and hosts are widely distributed throughout Eurasia whereas other social parasites use host species with small or patchy populations, e.g. Myrmoxenus or Chalepoxenus, where selection should be strong to decrease their virulence. Furthermore H. sublaevis produces a large army of slave making workers indicating that this species remains highly virulent. In accordance with the assumption of a geographic mosaic in the interaction of H. sublaevis and its hosts, these studies have shown that parasite prevalence is a good predictor of the strength of reciprocal adaptation in different communities. In our genetic, chemical and behavioural studies we could show that H. sublaevis prefers the smaller host L. muscorum, which is more limited in dispersal than its larger competitor L. acervorum. Both hosts showed differences in defense strategies of which L. acervorum is the more aggressive host, while L. muscorum tend to flee when getting into contact with its parasite. Moreover for the genetic more variable parasite the chemical profile of L. muscorum may be easier to imitate as this host is more limited in gene flow than its counter part. Further explanation of the better resemblance of the parasite to its smaller host could be the easier acquisition of the more volatile shorter hydrocarbons. Also in the field manipulation study both host species showed different responses to the parasite pressure of H. sublaevis following the two strategies, investment in sexuals or in workforce. Moreover our crossfostering experiment indicated that the local parasite showed a greater impact on its hosts than the allopatric one. This led to the conclusion that coevolutionary trajectories differ between communities, asumed by different historical processes, community context and ecological conditons at each site which confirms the geographic mosaic theory of coevolution