Adaptation to new nutritional environments: larval performance, foraging decisions, and adult oviposition choices in Drosophila suzukii

Abstract

Background: Understanding how species adapt to new niches is a central issue in evolutionary ecology. Nutrition is vital for the survival of all organisms and impacts species fitness and distribution. While most Drosophila species exploit rotting plant parts, some species have diversified to use ripe fruit, allowing earlier colonization. The decomposition of plant material is facilitated by yeast colonization and proliferation. These yeasts serve as the main protein source for Drosophila larvae. This dynamic rotting process entails changes in the nutritional composition of the food and other properties, and animals feeding on material at different stages of decay are expected to have behavioural and nutritional adaptations.

Results: We compared larval performance, feeding behaviour and adult oviposition site choice between the ripe fruit colonizer and invasive pest Drosophila suzukii, and a closely-related rotting fruit colonizer, Drosophila biarmipes. Through the manipulation of protein:carbohydrate ratios in artificial diets, we found that D. suzukii larvae perform better at lower protein concentrations and consume less protein rich diets relative to D. biarmipes. For adult oviposition, these species differed in preference for substrate hardness, but not for the substrate nutritional composition.

Conclusions: Our findings highlight that rather than being an exclusive specialist on ripe fruit, D. suzukii's adaptation to use ripening fruit allow it to colonize a wider range of food substrates than D. biarmipes, which is limited to soft foods with higher protein concentrations. Our results underscore the importance of nutritional performance and feeding behaviours in the colonization of new food niches.

Keywords: Drosophila biarmipes; Drosophila suzukii; Foraging; Niche; Nutrition; Nutritional geometry.

Fig. 1

The effects of protein and carbohydrate content of the larval diet on four larval life-history traits of D. suzukii (left column) and D. biarmipes (right column). The fitted response surfaces of the effects of 24 different diets varying in protein, carbohydrate, and caloric composition for: (first row) proportion of larvae surviving from first instar larvae to pupae; (second row) developmental time from first instar larvae to pupae; (third row) female pharate weight; and (fourth row) total number of ovarioles of adult females. Dashed black lines represent the P:C ratios. We replicated four times each block of 24 diets per species using 30 first instar larvae per diet. Filled black circles represent the respective nutritional coordinates of each of the 24 diets used (if a dot is absent not enough larvae survived that treatment to measure the trait)

Fig. 2

D. suzukii and D. biarmipes differ in the quantity of food eaten depending on the diet. A Amount of protein and carbohydrate ingested in the no-choice assay. Twenty larvae were offered one of six P:C ratios and were able to forage for 2 or 4 h. Each dot represents the mean value of 10 replicates and the error bars are 95% confidence intervals of the means. B Differences between protein and carbohydrates offsets in D. biarmipes and D. suzukii. Each condition was replicated ten times (both time points were pooled). Dashed line represents the normalized median for each macronutrient (0 = no macronutrient offset). Differences in font type (regular versus italic) between letters represent significant differences across least squared trends for a protein and b carbohydrates between the two species. C Amount of protein and carbohydrate ingested in the two-choice assay. Each dot represents the mean value of 10 replicates, except the triangles, which represent the average intake target for both species calculated from the pooled data of all treatments/time points. Error bars are 95% confidence intervals of the means. Twenty larvae were offered a choice between two protein to carbohydrate (P:C) ratios in three different combinations. Larvae were able to forage for 2 or 4 h and each time point was replicated ten times. The quantity of food in the larval gut was determined by spectrophotometer. D The figures show the percentage of the total amount of food ingested that corresponded to the higher protein food (1:1 for the first and second food pairs, and 1.5:1 for the third food pair) found in the guts of D. suzukii and D. biarmipes larvae in the two-choice assay. Black asterisks represent significant differences to no choice (50%-dashed black line—see Additional file 13: Table S10) and grey asterisks represent significant differences between species for the same diet (Least squared means comparison, see Additional file 14: Table S11)

Fig. 3

D. suzukii and D. biarmipes have the same oviposition preference for P:C ratios, but differ for substrate hardness. The oviposition site preference was estimated by the percentage of eggs laid in each diet. The letters (red for D. suzukii, blue italics for D. biarmipes) indicate significant differences in the proportion of eggs laid between different diets within a species, with significant differences marked by different letters, as determined by least squared means assuming no-choice value of 33% (dashed line in all panels—see Additional file 14: Table S11, Additional file 15: S12). Each treatment was replicated 25 times. Stars represent significant differences from the null hypothesis of 33%. A Females were offered a choice between three diets differing in their P:C ratios for oviposition. B Females were offered a choice between three diets differing in their hardness (agar concentration) for oviposition