Parthenogenesis in weevils of the tribe Naupactini (Coleoptera, Curculionidae): a Wolbachia-density dependent trait?

The intracellular bacteria Wolbachia pipientis can manipulate host reproduction to enhance their vertical transmission. It has been reported an association between parthenogenesis and Wolbachia infection in weevils from the tribe Naupactini. A curing experiment suggested that a threshold density of Wolbachia is required for parthenogenesis to occur. The aim of this study was to analyze Wolbachia infection status in the bisexual species Naupactus xanthographus and Naupactus dissimulator. Wolbachia infection was detected in both species from some geographic locations, not being fixed. In all positive cases, faint PCR bands were observed. Quantification through real time PCR confirmed that Wolbachia loads in bisexual species were significantly lower than in parthenogenetic ones; this strengthens the hypothesis of a threshold level. Strain typing showed that both species carry wNau1, the most frequent in parthenogenetic Naupactini weevils. These infections seem to be recently acquired by horizontal transfer. Wolbachia was located throughout the whole body, which reinforce the idea of recent transmission. Moreover, we demonstrated that this strain carries the WO phage. Finally, the analysis of eubacterial 16S rRNA gene showed intense PCR bands for both bisexual species, suggesting –the presence of additional bacteria. Interspecific competition might explain why the parthenogenetic phenotype is not triggered.


166
Primers for gatB and ITS1 genes suitable for qPCR analyses were designed using Primer3Plus 167 software [42]. Primer sequences for gatB were based on the sequence of wNau1 strain described in 168 [17], whereas primer sequences for ITS1 were designed using conserved regions of ITS1 sequences 169 from the three weevil species retrieved from GenBank (NCBI, NIH) (Table 1).  178 PCR product size amplified with gatBqPCR-F/R is 157 bp.
179 PCR product size amplified with ITS1qPCR-F/R is 92 bp.

181
The qPCR reactions were conducted in a Step One Plus Real-Time PCR System (Thermo Fisher  202 Normality of the data set was evaluated using the Shapiro-Wilk test, while homoscedasticity was 203 evaluated graphically. In case of non-homoscedasticity, variance was modeled using varIdent. Both 204 the Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) were applied to 205 select the best-fitting model. Analyses were carried out in two separate groups, including: (i) the 206 whole dataset to evaluate the effect of the reproductive mode (n=24); (ii) N. xanthographus + N.
207 dissimulator data to test the effect of the species, sex and location (n=16).

208
For (i), a model with the ratio obtained from Equation   239 xanthographus (Fig 1) for all the genes assayed. In addition, it was found in individuals of N.

Naupactus xanthographus (squares). Colors indicate Wolbachia infection status at each
244 sampling point (red, infected; blue, uninfected). For interpretation of the references to geographic 245 locations in this figure, the reader is referred to the S1 Table. 246 247 In all cases, the bands observed in agarose gel were faint for both species, while parthenogenetic 248 species used as positive controls showed intense bands for Wolbachia genes. Bisexual individuals 249 of P. postfasciatus showed no Wolbachia DNA amplification (Fig 2A). On the other hand, the 250 analysis of 16S rRNA gene revealed the presence of intense bands for N. xanthographus and N.

266
WO phage presence was revealed in the three species. While in parthenogenetic P.
267 postfasciatus a single bright band in the gel was observed, both N. dissimulator and N.

. Wolbachia quantification in bisexual species
275 The relative quantification obtained for Wolbachia in bisexual species was significantly lower than 276 in parthenogenetic P. postfasciatus (Fig 3A). 287 The fold change due to reproductive mode was at least twice. Additionally, an elevated inter-288 individual variation in bacterial densities within bisexual species was detected.

289
When comparing within bisexual species, no significant effects were detected in Wolbachia 290 loads related with species, sex or geographic location (p>0.05) (Fig 3B-C).

292 3.3. Wolbachia tissue localization in bisexual species
293 Wolbachia was detected throughout the different tissues analyzed in all the species and sexes 294 surveyed. These results are summarized in Table 2.  325 Wolbachia dynamics in these species.

326
Additionally, we report for the first time the presence of the temperate phage WO in Naupactini 327 weevils: N. xanthographus, N. dissimulator and P. postfasciatus yielded a positive diagnosis.
328 Although the three of them share the same Wolbachia strain, their phages seems to be non-identical, 329 as the agarose gel revealed a different band pattern between parthenogenetic and bisexual ones.

369
Another interesting topic is why the parthenogenetic phenotype is not triggered in these species.
370 A possible explanation for the association between Wolbachia and parthenogenesis is the lesser 371 ability of parthenogenetic weevils to rid themselves of Wolbachia infections once these happen 372 [17,19]. Actually, it appears that no species is able to dispose of Wolbachia. Instead, as it was 373 demonstrated in the present work, bisexual species are able to maintain the infection at low densities.