Oviposition preference for spherical surfaces is shared among multiple Drosophila species except D. melanogaster (Diptera: Drosophilidae)

Oviposition preference for spherical substrates has been reported in some insects but not in Drosophila species until the recent finding that Drosophila suzukii preferentially lays eggs on spherical surfaces with a smaller radius, whereas D. melanogaster does not. This finding raised two questions: (i) Was this trait specifically acquired in D. suzukii or lost in D. melanogaster? (ii) In the latter case, is it due to the long‐term laboratory culture using oviposition substrates with flat surfaces? To answer these questions, we examined the oviposition preference of three Drosophila species using the stocks recently established from wild individuals. As with D. suzukii, D. simulans and D. takahashii showed significant preference for spherical surfaces with a smaller radius, suggesting that this trait is shared by multiple Drosophila species. In contrast, D. melanogaster did not show any preference for either smaller or larger radii, showing that the preference already has been lost in the natural population of D. melanogaster. It may be possible that the loss of oviposition preference for spherical surfaces is involved in the evolutionary process of D. melanogaster becoming a human commensal.

Since the pioneering work by Ishii (1952) reporting that the adzuki bean beetle, Callosobruchus chinensis (L.), preferentially laid eggs on smaller glass beads, the effect of substrate size on oviposition behavior has been studied in a few groups of insects (Charnov et al. 1981;Katsoyannos & Pittara 1983;McDonald & McInnis 1985;Pittara & Katsoyannos 1992;Greenberg et al. 2004;Showler 2005).Among them, the oviposition preference for curvature of substrates was confirmed only in two species, the adzuki bean beetle and the Indianmeal moth Plodia interpunctella (Hübner) (Avidov et al. 1965;Sambaraju & Phillips 2008).Recently we found that Drosophila suzukii (Matsumura) preferentially lays eggs on spherical surfaces with a smaller radius in choice tests, whereas D. melanogaster (Meigen) did not discriminate between smaller and larger radii (Akutsu & Matsuo 2022).This finding left a question about the evolutionary history of this trait; that is, which was the ancestral state in the evolution of the D. melanogaster species group.Furthermore, because the Canton-S strain of D. melanogaster, which was established before 1916 (Bridges 1916) and has been maintained in the laboratory condition since then, was used in our previous study (Akutsu & Matsuo 2022), it remained possible that the long-term culture using artificial diet with a flat surface had selected against the oviposition preference for spherical surfaces.
To answer this question, we examined the oviposition preference for curvature in three Drosophila species, D. melanogaster, D. simulans (Sturtevant), and D. takahashii (Sturtevant) (Fig. 1), using the strains established from recently collected wild individuals (Table 1).The strains were maintained on a standard Drosophila culture medium with a flat surface made of corn meal, glucose, and dry yeast at 25 C with a 16 h light : 8 h dark cycle.Oviposition assay was carried out within eight generations after the collection of wild individuals (original individuals = first generation).Oviposition substrate was made of 1% agar solution (Seakem LE Agarose; Lonza, Basel, Switzerland) cast in silicone molds for ultraviolet resin-crafting hobbies, which have hemispherical holes of various sizes.In this study, we used the holes with four sizes (4.8, 5.7, 7.7, and 9.6 mm radius), and the volume of each substrate was fixed to 200 μL regardless of its radius (substrates were thinner than a hemisphere for larger radii).The top surface area was 1.31, 1.36, 1.49, and 1.63 cm 2 for the substrates with 4.8, 5.7, 7.7, and 9.6 mm radius, respectively.Ten females at the age of 7 to 9 days after eclosion were introduced into a petri dish (9 cm diameter), in which 10 oviposition substrates were placed on a wet cotton pad (3 Â 6 cm).Among 10 oviposition substrates (2 Â 5 rows), 5 were 4.8 mm radius and the other 5 were either of the larger radii (arranged in a checkered pattern), making three types of two choice assay (4.8-5.7,4.8-7.7,4.8-9.6 mm).The assay started 6 h before the transition to the dark phase.After 24 h of oviposition at 25 C, the number of eggs laid on each substrate was counted.Each female was used only once.Ten replications were made for each combination of the substrate radius.When the total number of eggs in a dish was less than 10, the corresponding data were excluded from the analysis, and additional replications were made.Deviation from neutrality (no preference) was examined using Wilcoxon signed rank test.
Figure 2 shows the results of two choice assay.D. simulans and D. takahashii showed significant preference for the smaller radius in the choice between 4.8 versus 7.7 or 9.6 mm, which was comparable to that of D. suzukii (Akutsu & Matsuo 2022).This result suggests that the oviposition preference for the substrate curvature is an ancestral trait shared among multiple Drosophila species.Our separate analysis using D. suzukii revealed that the oviposition preference for curvature overrode that for the hardness of substrates in the cross-modal competition assay (Akutsu & Matsuo 2023).In addition, we observed in our preliminary experiments that the preference for curvature overrode that for glucose (J.Akutsu, T. Matsuo, unpubl. data, 2023).These results collectively suggest that the preference for surface curvature has been maintained in many Drosophila species because it functions as a predominant oviposition cue.
In contrast, D. melanogaster did not show significant preference between any combinations of the surface radius, proving that the preference for curvature already had been lost in the wild population (Fig. 2).Although the natural history of D. melanogaster is surprisingly unknown, comparisons with D. simulans have suggested differences in several points related to human commensalism (Capy & Gibert 2004).First, D. melanogaster is more tolerant to alcohol than D. simulans, being capable of using fermented materials produced by human activity (David et al. 2004).Second, D. melanogaster has a tendency to enter the structures of human habitation, which is a trait absent from D. simulans (Capy & Gibert 2004;David et al. 2004).A recent study proposed that ancient D. melanogaster was a specialist of wild fruit, and its human commensalism is a derived trait acquired 1 million years ago, before it became a cosmopolitan generalist (Mansourian et al. 2018).Because the current "wild" African woodland D. melanogaster population showed a tendency to enter an ancient human habitation site as well, there should be other unknown key traits transforming D. melanogaster from a specialist in forests to a human commensal (Mansourian et al. 2018).Loss of oviposition preference for spherical surface might be involved in this process.The shape of substrates is a major difference between wild fruits and processed materials resulting from human activities.Losing curvature preference should have been adaptive as a generalist human commensal.Life history-related traits have been reported to be  Figure 2 Results of two choice oviposition assay between the substrates with different surface curvature.For D. takahashii, two independent strains established from single females captured on different days were used.Preference index was calculated as (N small ÀN large )/(N small + N large ) where N small and N large are the number of eggs laid on the substrates with the smaller and larger radii, respectively.Deviation from neutrality (no preference) was examined using Wilcoxon signed rank test.*P < 0.05, **P < 0.01, ***P < 0.001.N.S., not significant.

Figure 1
Figure1Phylogenetic relationship between the species used in this study.The estimated divergence time was acquired from TimeTree 5(Kumar et al. 2022).

Table 1
List of the species used in this study.African woodland population that is a specialist of marula fruit.On the other hand, the ecological role of the curvature preference in the other Drosophila species might be for avoiding over-decayed formless fruit that is inadequate for larval growth.It is necessary to test the curvature preference in other Drosophila species to confirm if this is a widely conserved trait, as well as asking if there are other species that have lost this trait.