 |
The actinomycete bacteria
The chemical defence presented by the metapleural gland secretion, grooming, and weeding by workers are not enough to prevent Escovopsis (a specialized fungal pathogen of leaf-cutting ant fungi) from establishing and spreading (Currie and Stuart, 2001; Currie et al., 2002; Bot et al., in press). Instead, the defence against Escovopsis is predominantly through a mutualistic bacterium associated with leaf-cutting ants (Currie et al., 1999b). This bacterium produces as yet unknown antibiotic compounds that specifically inhibit the growth of Escovopsis (Currie et al., 1999b) and hence reduce the negative impact of the parasite on leaf-cutting ant fungus gardens (Currie et al., 2002). The bacterium is filamentous and gram-positive and is currently classified in the genus Streptomyces (Currie et al., 1999b). Bacteria of this genus are of great abundance and ecological importance and are known to produce secondary metabolites that often have specific antibacterial or antifungal properties (Waksman, 1959; Currie et al., 1999b and references therein; Hu et al., 2000; Ravel et al., 2000).
The mutualistic bacterium grows on the cuticle of the ants and the most prominent place for growth is genus specific (Currie et al., 1999a). Within colonies the abundance of the bacterium is additionally dependent on both ant age and caste (Currie et al., 1999b; Currie et al., 2002). The relative abundance of the bacterium on the ant cuticle is furthermore dependent on the position of the worker in the fungus garden, and is inducible under stressful conditions such as garden decline or infection with Escovopsis (Currie et al., 2002). These features imply that the ants actively control the abundance of the bacterium and that the bacterial growth is correlated with the relative importance of the bacterium at that particular time and/or place in the colony (Currie et al., 2002). Active control by the ants further suggests that energy may be actively allocated to the bacterium and, hence, that a metabolic cost to the ants exists (Poulsen et al, submitted manuscript). The source of this energy provision is currently unknown, however secretions from exocrine glands have been proposed possible candidates (Currie et al., 2002), and the potential role of the metapleural gland secretion in maintaining the bacterium is examined in Poulsen et al. (in review). Furthermore, a biotest experiment will soon examine the in vitro effect of individual metapleural gland compounds on the Streptomyces bacterium (M. Poulsen, C.R. Currie, A.N.M Bot and J.J. Boomsma). From the results of these in vitro experiments, the in vivo results presented in Poulsen et al (in review), and an experiment conducted by Ortius-Lechner et al. (unpublished manuscript), showing that the composition of metapleural gland secretion is age and caste dependent, we hope to establish an explanation for a more distinct connection between the metapleural gland secretion and the temporal development of the mutualistic bacterium.
The Streptomyces bacterium is presumably vertically transmitted between colonies, being present on the cuticle of the colony founding gyne (Currie et al., 1999b). This seems reasonable, considering the high vulnerability of a newly founded fungus garden to an infection with Escovopsis (Currie et al., 1999b). This mode of transmission furthermore ensures that the fitness of the bacterium and that of the ants is aligned, both being dependent upon the production of gynes (prospective queens) (Currie et al., 1999b). The vertical transmission mode produces, as is the case with the mutualistic fungus, a bottleneck since presumably only one clone is propagated, a feature that is proposed to act as a stabilizing factor of mutualistic relationships (Frank, 1996; Herre et al., 1999). However, this may on an evolutionary time scale pose a potential problem in the "evolutionary arms-race" with the virulent parasite Escovopsis, to which the mutualistic bacterium continuously has to adapt. Given that the Streptomyces bacterium in rare cases could be horizontally transmitted, possible mixing of genetically different strains of the mutualistic bacterium could occur. Theory predicts that such mixing may be problematic if it induces competition between strains of symbionts and reduces their overall level of service to the host (Frank, 1996), as is the case with the fungal symbiont (Bot et al., 2001). Although negative consequences of mixing cannot be excluded, assuming that the defence against Escovopsis follows Red Queen like characteristics, mixing and sex among Streptomyces bacteria may be beneficial for the ants if a complex of recombining bacterial genotypes would provide better protection against Escovopsis than genetically purer cultures. The exchange of genetic material between free-living Streptomyces strains is known to occur under natural conditions (Waksman, 1959 and references therein; Ravel et al., 2000; Hu et al., 2000). Similar exchanges between Streptomyces lineages associated with leaf-cutting ants may thus occasionally occur and be favoured if the costs of conflicts do not exceed the benefits obtained by the exchange. Genetic recombination may thus both have short-term advantages for the production of antibiotics against Escovopsis and long-term advantages in the continuing "arms race" between Escovopsis on one side and the ants and their Streptomyces ally on the other. The opportunity for such horizontal transfer of the mutualistic Streptomyces bacterium is examined in Poulsen et al (submitted).
Questions for further research:
-where do resources to the bacterium originate from?
-how tight is the relationship between the ants and the Streptomyces symbiont? I.e. are there multiple strains of bacteria per colony/per ant.
-how are the capacities of the mutualistic bacterium to evolve in parallel with the continuously evolving Escovopsis parasite?
Send E-Mail to: mpoulsen@zi.ku.dk
Copyright © 2006 Michael Poulsen. All Rights Reserved