On the evolution of Wolbachia-induced parthenogenesis in Trichogramma wasps

Organisms display a great variety of sex ratios (ratios of females vs. males), ranging from 100% females to a male bias. These sex ratios are not always only determined by the genes of the organism itself but may actually often be manipulated or distorted by "sex ratio distorters". One sex ratio distorter, the bacterium Wolbachia that lives in the cytoplasm of the cells of its host organism, has received much attention by biologists all over the world. This interest mainly arises from the fact that it manipulates arthropod or nematode reproduction in several ways - feminization, induction of cytoplasmic incompatibility, male-killing and parthenogenesis-induction - to enhance its own inheritance from mother to daughter. Because sperm cells do not contain enough cytoplasm, they cannot transmit Wolbachia and males are a dead end for the bacterium. Recent estimates of Wolbachia 's prevalence range from 17 to even 76% of the insect species.In many wasp, thrips and mite species Wolbachia has switched the mode of reproduction from sexuality to complete parthenogenesis (±100% females). However, in minute parasitoid wasps of the genus Trichogramma , which are used worldwide as natural enemies in biological control of lepidopteran pests, only a part of the females in a population is infected with Wolbachia and can therefore reproduce through parthenogenesis. My aim in this thesis is to gain more insight in the dynamics of parthenogenesis-inducing (PI) Wolbachia and to explain the coexistence of infected and uninfected forms in Trichogramma wasps. After first reviewing the literature on PI Wolbachia in <strong>chapter 2</strong> , I tried to further our understanding of the coexistence of the two reproductive forms in natural Trichogramma kaykai and T. deion populations by combining fieldwork, molecular techniques, behavioural- and crossing experiments with model studies.Vertical transmission of Wolbachia from mother to daughter has been viewed as the main mode of transmission but in <strong>chapter 3 & 4</strong> we show an unexpectedly frequent natural inter- and intraspecific horizontal transmission between and within Trichogrammakaykai and T. deion larvae sharing a common food source, a butterfly egg. Originally uninfected immature wasps could acquire Wolbachia inside the host egg but not all newly infected females exhibit parthenogenesis. In T. kaykai, intraspecific horizontal transfer was followed by complete parthenogenesis in future generations but when T. kaykai females received Wolbachia from T. deion , the infection tended to be lost several generations after interspecific horizontal transfer. <strong></strong> Our results largely explain the discordance between Wolbachia - and (Trichogrammatid) host phylogenies. Frequent horizontal transfer might select for high virulence in these bacteria .Because of a nuclear-cytoplasmic conflict between Wolbachia and the nuclear genes of Trichogramma and the previously described horizontal transfer of Wolbachia , the infection is most likely associated with fitness costs in populations where infected and uninfected individuals coexist. In <strong>chapter 5</strong> we show that infected T. kaykai suffer a reduced survival compared to uninfected conspecifics when they shared the same host. The survival rate of infected immatures was higher when they competed with other infected immatures from a different infected parent than in competition with uninfected immatures. This shows that PI Wolbachia -infected Trichogramma can suffer a substantial fitness cost. Because of this reduced competitive ability of infected larvae, horizontal transfer that occurs under the same superparasitism circumstances does not contribute much to an increase in the infection rate in the population.Previous work showed that the presence of another sex ratio distorter in males, a B chromosome called PSR (Paternal Sex Ratio) that destroys the paternal chromosomes after fertilization thereby causing an all-male or a male-biased offspring sex ratio, contributes to a low infection frequency in T. kaykai . In <strong>chapter 6</strong> we determined if a PSR factor causes low infection frequencies in other species as well. Therefore, we studied natural populations of three Trichogramma species - T. kaykai , T. deion and T. pratti - from the Mojave Desert. Our data showed that all the male-biased and all-male Trichogramma broods collected from the butterfly Apodemia mormo deserti that contained males expressing the PSR phenotype, belonged to T. kaykai. In laboratory tests, 71.4% of the T. kaykaiPSR males horizontally transmitted the PSR phenotype to T. deion . This percentage is comparable to the transmission rate of PSR to T. kaykai females, namely 81.6%. Consequently, the PSR can be transmitted to T. deion and we expect this to happen in the field because T. kaykai and T. deion sometimes emerge from the same butterfly egg. Despite this, we cannot find PSR in T. deion . Modeling shows that low Wolbachia infection frequencies can only be attained when the PSR rates are very high. Therefore, other factors should keep the PI Wolbachia -infection from spreading to fixation in this species, e.g. nuclear suppressor genes.The mating structure in the host population plays a major role in the dynamics of PI Wolbachia and PSR . A PSR factor prevents the Wolbachia infection from spreading to all the females in T. kaykai because uninfected T. kaykai females show a high level of sib (brother-sister) mating. Sib mating is a barrier against the destructive effect of mating with a PSR -carrying male. Infected females do not have this advantage. Using a population genetic model with microsatellites as genetic markers in <strong>chapter 7</strong> , we estimated high levels of sib-mating of 70% and an off-patch mating of 15%. Thirty-five percent of the patches were estimated to be parasitized by two T. kaykai females. Incorporating such levels of sib mating in a previously developed model describing the dynamics of PI Wolbachia and PSR in a Trichogramma population, resulted in stable low frequencies of infection, i.e., a coexistence between infected and uninfected individuals, and of the PSR chromosome. Our results show how mating structure allows the two sex ratio distorters to coexist in the population.<strong></strong>The main conclusion from this thesis is that, despite the high vertical transmission and regular horizontal transfer of Wolbachia , a PI Wolbachia -infection can be attained at low frequencies in Trichogramma , due to the presence of a non-mendelian suppressor, like the male-biasing PSR factor in T. kaykai, but also due to other factors. In T. deion, for example, PSR does not keep the infection frequency at low levels but a nuclear mendelian suppressor against the PI Wolbachia might have evolved.Next to their significance for the understanding of the evolutionary pathways of Wolbachia -host interactions, the results reported in this thesis may also have important implications for future use of natural enemies, and more specifically Trichogramma wasps, in inundative biological control. We may now have a good method to render wasps parthenogenetic, via super- or multiparasitism by infected and uninfected females, thereby increasing the efficacy of parasitoid releases against lepidopteran pests