Introduction

A group of intracellular bacteria belonging to the genus Wolbachia (Class Alpha-proteobacteria) have been isolated from a wide variety of arthropods and filarial nematodes (Werren, 1997; Bourtzis and O'Neill, 1998; Stouthamer et al. 1999). Being transmitted exclusively from mothers to offspring, Wolbachia increases its fitness by altering the reproductive systems of the host through various means such as cytoplasmic incompatibility, male killing, feminization, induction of parthenogenesis, and others (Werren, 1997; Stouthamer et al. 1999; Charlat et al. 2003).

In the adzuki bean borer, Ostrinia scapulalis (Lepidoptera: Crambidae), naturally-occurring Wolbachia selectively kill male progeny during embryonic and larval development (Kageyama and Traut, 2004). Interestingly, when Wolbachia-infected female moths were treated with tetracycline prior to oviposition, they often produced genetically male moths with sexually mosaic phenotypes (Kageyama et al. 2003a; Kageyama and Traut, 2004). This finding demonstrates that the Wolbachia strain in O. scapulalis has a feminizing effect on genetic males. Kageyama and Traut (2004) proposed that the feminizing effect underlies the death of Wolbachia-infected males, and sexual mosaics are produced when this effect is incompletely exerted.

It is known that in various insect species reproductive manipulation such as cytoplasmic incompatibility, male killing, and parthenogenesis caused by Wolbachia can be suppressed by high temperatures, either through direct suppression of the activity of Wolbachia, or through the suppression of proliferation of Wolbachia in host cells (Trpis et al. 1981; Hoffman et al. 1986; Pintureau et al. 1999; Hurst et al. 2000, 2001). Here we examined whether high temperature treatments of O. scapulalis female adults infected with Wolbachia can induce the production of sexually mosaic moths.

Materials and Methods

Results

The matriline MD0014 exhibited the all-female trait, which was shown to be due to the Wolbachia-induced male killing (Kageyama and Traut, 2004). In matrilines MD2161 and MD2164, sex ratios did not significantly deviate from 1:1.

The survival rates of females after heat treatment decreased as temperature rose (Table 1). No significant difference in survival rates were detected between infected and uninfected females at any temperature examined (P > 0.05, 2 × 2χ² test).

Table 1.

Survival rates of female Ostrinia scapulalis after heat treatment

No effect of high temperature treatments on the sex ratio or sex differentiation was observed in the progeny of uninfected mothers; they consistently developed to normal males and females at nearly 1:1 ratio (Table 2). In contrast, progeny of the infected mothers treated at 25°C and 34°C developed to only female adults (Table 2). Among the progeny of infected mothers treated at 36°C, seven of 97 individuals examined showed sexually mosaic phenotypes at the pupal stage (Table 2; Figure 1A). The seven mosaic pupae had male-like mid-tibia and femalelike morphology in the last abdominal segment (Figure 1A). All seven mosaic pupae were derived from eggs laid within 24 h after the completion of heat treatment. Two of the mosaic pupae developed to adults, and the other five died. The wings of the two emerged moths, although curled, exhibited wing mark/color patterns characteristic of the sexual mosaics (Figure 1D). The curled wings were observed only in sexual mosaics. Mosaic patterns were also found in the coloration of the dorsal abdomen of the two individuals (Figure 1D). Except for these sexual mosaics, all progeny produced by Wolbachia-infected females consisted of normal females. Almost all the offspring of infected mothers maintained at 25°C (control) developed to adults, > 0.97 in most cases.

Table 2.

Sex ratio (female: male) of the progeny of Wolbacha-infected and -uninfected Ostrinia scapulalis collected daily as eggs after heat treatment.

Discussion

In the present study, sexually mosaic progeny were produced from Wolbachia-infected O. scapulalis females treated with a high temperature (36°C) prior to oviposition. Thus, two different treatments of Wolbachia-infected O. scapulalis female moths, i.e., high temperature and oral administration of antibiotics (Kageyama et al. 2003a), have been shown to induce sexually mosaic moths. The present finding confirmed that Wolbachia in O. scapulalis has the potential to interfere with the host's sex determination pathway.

Heat and antibiotic treatments of infected O. scapulalis females showed a clear difference in the induction pattern of sexually mosaic moths, as follows. In antibiotic treatment, sexual mosaics developed from eggs laid ≥ 4 days after treatment in O. scapulalis (Kageyama et al. 2003a; Kageyama and Traut, 2004), and ≥ 3 days after treatment in the sister species O. furnacalis (Sakamoto et al. 2007). Meanwhile, the effect of heat treatment appeared promptly and transiently; sexual mosaics developed from eggs laid < 24 h after treatment, but no sexual mosaics developed from eggs laid > 24 h after treatment. It is possible that orally administered antibiotics take longer to show an effect because they must be absorbed through the gut and then transported to the reproductive tissue. The transience of the effect of high temperature suggests that the occurrence of sexual mosaics is a consequence of direct suppression of the activity of Wolbachia, rather than the consequence of an indirect effect such as a decrease in Wolbachia density.

Regarding the generation of sexually mosaic moths, we hypothesize that Wolbachia interferes with key step(s) of sex determination during the egg/embryonic development of individuals with the male genotype (ZZ). Considering the promptness and transience of the effect, heat treatment would be more useful than antibiotic treatment for spotting crucial time events when Wolbachia interferes with the sex determination step(s) during development. Heat treatments at various times during egg/embryonic development would clarify the crucial timing of Wolbachia action against normal sex determination in genetic males of O. scapulalis. Such knowledge may give insights into the mechanism of Wolbachia-induced feminization in arthropods.

Finally, only two moths successfully emerged from seven sexually mosaic pupae. The death of five mosaic pupae might be attributable to the delayed expression of male killing; heat treatment probably decreased the Wolbachia density to a level at which it could not kill male hosts during the larval stage, but Wolbachia density may have subsequently recovered to a level sufficient to kill male hosts at the pupal stage. It is known in the butterfly Hypolimnas bolina that tetracycline treatment of females leads to the attenuated and delayed expression of male killing (Mitsuhashi et al. 2004; Charlat et al. 2007). In H. bolina, Wolbachia has the ability to kill males during the larval stage through bacterial activity (Charlat et al. 2007).