Da-Rong Yang

The complete mitochondrial genomes of two ghost moths, Thitarodes renzhiensis and Thitarodes yunnanensis: the ancestral gene arrangement in Lepidoptera

BackgroundLepidoptera encompasses more than 160,000 described species that have been classified into 45–48 superfamilies. The previously determined Lepidoptera mitochondrial genomes (mitogenomes) are limited to six superfamilies of the lineage Ditrysia. Compared with the ancestral insect gene order, these mitogenomes all contain a tRNA rearrangement. To gain new insights into Lepidoptera mitogenome evolution, we sequenced the mitogenomes of two ghost moths that belong to the non-ditrysian lineage Hepialoidea and conducted a comparative mitogenomic analysis across Lepidoptera.ResultsThe mitogenomes of Thitarodes renzhiensis and T. yunnanensis are 16,173 bp and 15,816 bp long with an A + T content of 81.28 % and 82.34 %, respectively. Both mitogenomes include 13 protein-coding genes, 22 transfer RNA genes, 2 ribosomal RNA genes, and the A + T-rich region. Different tandem repeats in the A + T-rich region mainly account for the size difference between the two mitogenomes. All the protein-coding genes start with typical mitochondrial initiation codons, except for cox1 (CGA) and nad1 (TTG) in both mitogenomes. The anticodon of trnS(AGN) in T. renzhiensis and T. yunnanensis is UCU instead of the mostly used GCU in other sequenced Lepidoptera mitogenomes. The 1,584-bp sequence from rrnS to nad2 was also determined for an unspecified ghost moth (Thitarodes sp.), which has no repetitive sequence in the A + T-rich region. All three Thitarodes species possess the ancestral gene order with trnI-trnQ-trnM located between the A + T-rich region and nad2, which is different from the gene order trnM-trnI-trnQ in all previously sequenced Lepidoptera species. The formerly identified conserved elements of Lepidoptera mitogenomes (i.e. the motif ‘ATAGA’ and poly-T stretch in the A + T-rich region and the long intergenic spacer upstream of nad2) are absent in the Thitarodes mitogenomes.ConclusionThe mitogenomes of T. renzhiensis and T. yunnanensis exhibit unusual features compared with the previously determined Lepidoptera mitogenomes. Their ancestral gene order indicates that the tRNA rearrangement event(s) likely occurred after Hepialoidea diverged from other lepidopteran lineages. Characterization of the two ghost moth mitogenomes has enriched our knowledge of Lepidoptera mitogenomes and contributed to our understanding of the mechanisms underlying mitogenome evolution, especially gene rearrangements.

The reproductive success of Ficus altissima and its pollinator in a strongly seasonal environment: Xishuangbanna, Southwestern China

Fig trees (Ficus spp.) are of great ecological significance, producing fruits that are fed on by more birds and mammals than any other plants in the tropics. They are pollinated by host-specific pollinator fig wasps (Hymenoptera, Agaonidae), and their fruit phenology and reproductive success are, therefore, modulated by symbiotic fig wasps. However, there are few studies focusing on the variation of Ficus reproductive success in strongly seasonal environments. We examined the phenology and reproductive success of Ficus altissima growing in a highly seasonal climate towards the northern limit of the range of fig trees in Xishuangbanna, China. Leaf production occurred at irregular intervals throughout the year, with new leaves and syconia initiated together, producing between three and seven crops over a 3-year period. Syconia were produced in synchronous crops with asynchrony between trees. The syconia produced more seeds than pollinators, and those syconia with more seeds also produced more pollinators. Reproductive success (measured as the number of seeds and pollen-carrying agaonid females produced by each syconium) varied greatly between seasons. It was highest for crops that matured during the cooler, relatively dry periods from February to March and October to November, and was lowest during the summer months from April to August. This variation corresponded to small differences in the number of flowers in the syconia, but was mainly driven by large seasonal differences in the relative abundance of non-pollinating fig wasps.

The population dynamics of a non-pollinating fig wasp on Ficus auriculata at Xishuangbanna, China

All fig wasps are confined to figs as larvae, and their specialized diets are restricted to fig embryos, galled fig ovaries or other fig wasp larvae (Boucek 1988). Almost all of the ∼800 Ficus species (Moraceae) are involved in species-specific obligate mutualism with a pollinating fig wasp (Hymenoptera: Agaonidae), which also strictly depends on its host fig for reproduction (Berg 2003, Ramirez 1970, Wiebes 1979). More than half of all fig species are functionally dioecious, with male and female functions relegated to separate plants, called gall and seed figs (Kjellberg et al. 1987). Gall figs are functionally male because they foster the pollinator larvae that disperse the figs pollen as adults. Seed figs are functionally female and produce only seeds. The styles are too long for pollinator ovipositors to reach the ovules, and hence they cannot lay eggs. Gall fig styles are short (Ganeshaiah et al. 1995, Weiblen et al. 1995).

Secondary galling: a novel feeding strategy among ‘non-pollinating’ fig wasps from Ficus curtipes

The interaction between pollinator fig wasps (Agaonidae) and their host fig trees (Ficus) is a striking example of an obligate plant–insect mutualism, but figs also support numerous ‘parasites’ of the mutualism. Female agaonids (foundresses) lay their eggs in shorter‐styled flowers, whereas longer‐styled flowers produce seeds. A few ‘non‐pollinating’ fig wasps (NPFWs) can also enter figs to oviposit Fig wasp oviposition site choice and larval biology in figs of an Asian monoecious species, Ficus curtipes Corner, were recorded where two NPFW species oviposit inside the figs, such as the agaonid. Eupristina sp. agaonids chose flowers in proportion to their availability, rather than preferring to oviposit in shorter‐styled flowers. Diaziella yangi van Noort & Rasplus and Lipothymus sp. (Pteromalidae) foundresses followed Eupristina sp. into receptive figs and laid their eggs entirely in flowers that already contained pollinator eggs. This indicates that both NPFWs are inquilines under the widely‐used terminology in the fig wasp literature, because they utilise galls generated by the pollinators. However, their adult bodies and galls were larger than those of the pollinators, showing that they independently stimulate ovule growth. These species are better described as secondary gallers that modify galls previously generated by the pollinators and kill these primary gallers. Use of the term ‘inquiline’ among NPFWs inadequately and often inappropriately describes their biology. No known NPFWs are inquilines in the strict sense that they do not harm their hosts. ‘Primary gallers’, ‘secondary gallers’, ‘seed predators’, and ‘parasitoids’ describe their biology more accurately.

The incidence and pattern of copollinator diversification in dioecious and monoecious figs

Differences in breeding system are associated with correlated ecological and morphological changes in plants. In Ficus, dioecy and monoecy are strongly associated with different suites of traits (tree height, population density, fruiting frequency, pollinator dispersal ecology). Although approximately 30% of fig species are pollinated by multiple species of fig‐pollinating wasps, it has been suggested that copollinators are rare in dioecious figs. Here, we test whether there is a connection between the fig breeding system and copollinator incidence and diversification by conducting a meta‐analysis of molecular data from pollinators of 119 fig species that includes new data from 15 Asian fig species. We find that the incidence of copollinators is not significantly different between monoecious and dioecious Ficus. Surprisingly, while all copollinators in dioecious figs are sister taxa, only 32.1% in monoecious figs are sister taxa. We present hypotheses to explain those patterns and discuss their consequences on the evolution of this mutualism.

Body size in a pollinating fig wasp and implications for stability in a fig‐pollinator mutualism

The fig–fig pollinator association is a classic case of an obligate mutualism. Fig‐pollinating wasps often have to fly long distances from their natal syconia to a receptive syconium and then must enter the narrow ostiole of the syconium to reproduce. Large wasps are expected to have a greater chance of reaching a receptive syconium. In this study, we tested this hypothesis and then examined whether the ostiole selectively prevented larger pollinators from entering the syconial cavity. In Xishuangbanna, China, Ceratosolen solmsi marchali Mayr (Hymenoptera: Agaonidae) pollinates the dioecious syconia of Ficus hispida L. (Moraceae). The body size of newly emerged wasps and wasps arriving at receptive syconia were compared. Wasps arriving at receptive syconia were significantly larger than newly emerged wasps. We also compared the size of wasps trapped in the ostiole with those in the cavity. Wasps trapped in the ostiole were significantly larger than those in the syconial cavity. Thus, in the case of F. hispida, large wasps were more likely to reach receptive syconia, but the ostiole limited maximum fig wasp size. This indicates that the ostiole, as a selective filter to pollinators, stabilizes pollinator size. Hence, it helps to maintain stability in the fig–fig pollinator mutualism.

Costs of inflorescence longevity for an Asian fig tree and its pollinator

In ecological situations where interactions between two species are to their mutual benefit, traits are expected to evolve to maximise the value of the timing of their encounters. Plants that depend on animals for pollination vary in the longevity of their flowers and also in how the quality of the rewards they offer varies in flowers of different ages. However, costs of floral longevity are rarely studied. Using field experiments with Ficus semicordata, a SE Asian dioecious fig tree, we examined sexual differences in syconium ageing, how the reproductive success of the plant and its pollinator change with syconium age and whether these changes are reflected in pollinator preferences. Un-pollinated syconia remained receptive to their host-specific pollinators for long periods, but eventually abort. Compared with male syconia, un-pollinated female syconia aborted more quickly and lost their ability to attract pollinators more quickly. Older pollinated female syconia were also more likely to abort. Further, declines in productivity with syconium age were also more apparent in female syconia, though older male syconia also produced fewer, smaller wasp offspring. The longevity costs are reflected in pollinator preferences. This suggests that sexual differences in duration of receptivity may be adaptive and a component of the reproductive strategies. It also indicates that placing fig wasps onto older syconia over-estimates their likelihood of being pollinated under natural conditions and prolonged receptivity increases the likelihood of pollination at the cost of reduced productivity with syconium age. This opens interesting perspectives on the co-evolution of this inter-specific interaction.

‘Push’ and ‘pull’ responses by fig wasps to volatiles released by their host figs

In the specific mutualism between fig trees (Ficus) and their obligate pollinating fig wasps (Agaonidae), it is crucial that fig wasps can recognize the developmental stages of their host figs. However, the responses of fig wasps to volatiles released from figs during their developmental phases are less clearly understood and are the focus of this study. We extracted and identified the volatiles released from the figs of Ficus curtipes throughout their development. Using Y-tube choice experiments, we also compared the behavioural responses of the tree’s pollinator (Eupristina sp.) to figs at different developmental stages, and compared these results to those obtained by trapping fig wasps as they arrived at a tree with a developing fig crop. The chemical composition of the fig volatiles changed during fig development with the blends exhibiting clear segregation among figs at different developmental phases. Male phase figs had the most distinct blend. Fig wasp females were preferentially attracted to receptive figs, but figs at most other developmental phases were also attractive. Conversely, male phase figs had a repellent effect. These results were supported by the behaviour of the wasps under natural conditions, with small numbers of fig wasps arriving at the tree before and after receptive figs were present. These results indicate a more complex relationship between fig volatiles and fig wasp behaviour than previously realized, with volatiles mediating both the initial meeting of the mutualists to achieve pollination and egg laying and the subsequent departure of the next generation of fig wasps. This offers an explanation for the specialization and long-term coexistence of figs and fig wasps.

Host pollination mode and mutualist pollinator presence: net effect of internally ovipositing parasite in the fig-wasp mutualism

The Ficus–their specific pollinating fig wasps (Chalcidoidea, Agaonidae) interaction presents a striking example of mutualism. Figs also shelter numerous non-pollinating fig wasps (NPFW) that exploit the fig–pollinator mutualism. Only a few NPFW species can enter figs to oviposit, they do not belong to the pollinating lineage Agaonidae. The internally ovipositing non-agaonid fig wasps can efficiently pollinate the Ficus species that were passively pollinated. However, there is no study to focus on the net effect of these internally ovipositing non-agaonid wasps in actively pollinated Ficus species. By collecting the data of fig wasp community and conducting controlled experiments, our results showed that internally ovipositing Diaziella bizarrea cannot effectively pollinate Ficus glaberrima, an actively pollinated monoecious fig tree. Furthermore, D. bizarrea failed to reproduce if they were introduced into figs without Eupristina sp., the regular pollinator, as all the figs aborted. Furthermore, although D. bizarrea had no effect on seed production in shared figs, it significantly reduced the number of Eupristina sp. progeny emerging from them. Thus, our experimental evidence shows that reproduction in Diaziella depends on the presence of agaonid pollinators, and whether internally ovipositing parasites can act as pollinators depends on the host fig’s pollination mode (active or passive). Overall, this study and others suggest a relatively limited mutualistic role for internally ovipositing fig wasps from non-pollinator (non-Agaonidae) lineages.

Ultrastructure of antennal sensilla of female Ceratosolen solmsi marchali (Hymenoptera: Chalcidoidea: Agaonidae: Agaoninae)

Abstract Fig-pollinating wasps are phytophagous wasps that mainly use olfaction to locate their fig (Ficus L., Moraceae) hosts. To provide a morphological framework for studying agaonid olfaction, we examined the antennal sensilla of female Ceratosolen solmsi marchali Mayr by scanning and transmission electron microscopy. We identified and characterized (ultrastructure, distribution, abundance, and position) 13 types of sensilla: multiporous placoid sensilla (types 1 and 2), basiconic sensilla (types 1 and 2), basiconic capitate peg sensilla, sensilla chaetica (types 1–3), sensilla trichodea, sensilla coeloconica (types 1–3), and one specialized sensillum regarded as a sensillum obscurum. We suggest that five types are chemoreceptors because they are porous and innervated by multiple sensory neurons. Sensilla coeloconica type 1 may also function as chemoreceptors, based on external morphology. Other sensilla may be involved in mechanoreception, thermo- and (or) hygro-reception, or pressure detection. We discuss our results in relation to the lifestyle of C. solmsi marchali.