Wolbachia infections in native and introduced populations of fire ants (Solenopsis spp.).

Wolbachia are cytoplasmically inherited bacteria that induce a variety of effects with fitness consequences on host arthropods, including cytoplasmic incompatibility, parthenogenesis, male-killing and feminization. We report here the presence of Wolbachia in native South American populations of the fire ant Solenopsis invicta, but the apparent absence of the bacteria in introduced populations of this pest species in the USA. The Wolbachia strains in native S. invicta are of two divergent types (A and B), and the frequency of infection varies dramatically between geographical regions and social forms of this host. Survey data reveal that Wolbachia also are found in other native fire ant species within the Solenopsis saevissima species complex from South America, including S. richteri. This latter species also has been introduced in the USA, where it lacks Wolbachia. Sequence data reveal complete phylogenetic concordance between mtDNA haplotype in S. invicta and Wolbachia infection type (A or B). In addition, the mtDNA and associated group A Wolbachia strain in S. invicta are more closely related to the mtDNA and Wolbachia strain found in S. richteri than they are to the mtDNA and associated group B Wolbachia in S. invicta. These data are consistent with historical introgression of S. richteri cytoplasmic elements into S. invicta populations, resulting in enhanced infection and mtDNA polymorphisms in S. invicta. Wolbachia may have significant fitness effects on these hosts (either directly or by cytoplasmic incompatibility) and therefore these microbes potentially could be used in biological control programmes to suppress introduced fire ant populations.

Multilocus evolution in fire ants: effects of selection, gene flow and recombination.

The reproductive success of individual fire ant queens (Solenopsis invicta) previously has been shown to be strongly influenced by their genotype at a single enzyme-encoding gene, designated Pgm-3. This paper presents evidence that a second, tightly linked gene, designated Gp-9, is under similarly strong selection in these ants. Selection appears to act independently on the two genes and is detectable in only one of the two social forms of this species (the "polygyne" social form in which nests contain multiple fertile queens). Strong directional selection on Pgm-3 in this form, involves worker destruction of all queens with genotype Pgm-3AA before they reproduce. Selection on Gp-9 is more complex, involving both lethality of all Gp-9bb females and a strong or even complete survival advantage to reproductive queens with the heterozygous genotype Gp-9Bb. Pgm-3 and Gp-9 are tightly linked (rf = 0.0016) and exhibit strong gametic phase disequilibrium in introduced populations in the U.S. This disequilibrium seems not to have stemmed from the founder event associated with the introduction, because the same associations of alleles found in the U.S. apparently occur also in two native populations in Argentina. Rather, selection acting independently on Pgm-3 and Gp-9, in conjunction with gene flow from the alternate, "monogyne" social form (in which nests contain a single fertile queen), may explain the origin of disequilibrium between the two loci in polygyne fire ants.

Effect of a founder event on variation in the genetic sex-determining system of the fire ant Solenopsis invicta.

Effects of a recent founder event on genetic diversity in wild populations of the fire ant Solenopsis invicta were studied, with particular attention given to the genetic sex-determining system. Diploid males are far more common relative to haploid males in introduced populations than in native populations of fire ants, and queens that produce diploid males account for a significantly larger proportion of the mated queens in introduced than in native populations. Differences between native and introduced populations in attributes of the mating systems (i.e., queen mating frequency or level of inbreeding) can be excluded as factors contributing to these different levels of diploid male production. Thus, we conclude that diploid males have increased in frequency in introduced populations because of a loss of allelic diversity at the sex-determining locus (loci). This loss of sex alleles has generated a substantial increase in the estimated segregational genetic load associated with production of sterile diploid males in introduced populations over the load in native populations. The loss of allelic diversity in the sex-determining system in introduced S. invicta is paralleled by a loss of electrophoretically detectable rare alleles at protein-encoding loci. Such concordance between these different types of markers is predicted because each of the many sex alleles present in the native populations is expected to be rare. Estimates of expected heterozygosity (Hexp) based on 76 electrophoretic loci do not differ significantly between the native and introduced fire ant populations, illustrating the lack of sensitivity of this measure for detecting many types of bottlenecks.