Odorant-binding proteins OBP57d and OBP57e affect taste perception and host-plant preference in Drosophila sechellia

Abstract

Despite its morphological similarity to the other species in the Drosophila melanogaster species complex, D. sechellia has evolved distinct physiological and behavioral adaptations to its host plant Morinda citrifolia, commonly known as Tahitian Noni. The odor of the ripe fruit of M. citrifolia originates from hexanoic and octanoic acid. D. sechellia is attracted to these two fatty acids, whereas the other species in the complex are repelled. Here, using interspecies hybrids between D. melanogaster deficiency mutants and D. sechellia, we showed that the Odorant-binding protein 57e (Obp57e) gene is involved in the behavioral difference between the species. D. melanogaster knock-out flies for Obp57e and Obp57d showed altered behavioral responses to hexanoic acid and octanoic acid. Furthermore, the introduction of Obp57d and Obp57e from D. simulans and D. sechellia shifted the oviposition site preference of D. melanogaster Obp57d/e(KO) flies to that of the original species, confirming the contribution of these genes to D. sechellia's specialization to M. citrifolia. Our finding of the genes involved in host-plant determination may lead to further understanding of mechanisms underlying taste perception, evolution of plant-herbivore interactions, and speciation.

Figure 1. The Locus Responsible for Interspecies Difference in HA Avoidance Is Mapped to Obp57e

(A) Behavioral screening of interspecies hybrids between D. melanogaster deficiency strains and D. sechellia. Response to HA was measured by the trap assay [17]. Response index (RI) = (N_h − N_w)/(N_h + N_w), where N_h is the number of individuals trapped in 0.5% HA solution, and N_w is that of individuals trapped in distilled water. Error bars indicate 95% confidence intervals determined by the binominal test of summed data from five replications of the assay with 100 females for each replication. (B) Determination of break points in deficiency chromosomes. A filled circle indicates that the deficiency-chromosome–derived sequence was detected, and an open circle indicates that the deficiency-chromosome–derived sequence was not detected at that position. (C) The Df(2R)AA21 chromosome has a 10-bp deletion in the first exon of the Obp57e gene. A genomic sequence of Df(2R)AA21 is aligned with that of the wild-type strain (CS). Predicted ORFs are boxed and capitalized. Arrows indicate the position and direction of translation start sites (ATG). (D) Comparison of Obp57e structure between D. melanogaster (mel), D. simulans (sim), and D. sechellia (sec). Predicted signal peptide sequence is boxed. Altered amino acid residues are highlighted.

Figure 2. Quantitative RT-PCR Analysis of Obp57d and Obp57e Transcripts

Heads and legs from 20 staged females were used for analysis. Transcript level relative to that of the ribosomal protein gene rp49 is shown. Each bar represents the mean of three replicates. Error bars indicate standard error.

Figure 3. GFP Reporter Assay of Obp57e

(A) Genomic structure around the Obp57e gene. Positions of the region used for the GFP reporter construct and 4-bp insertion in D. sechellia are indicated. Arrows on Obp57d and Obp57e indicate the position of the predicted translation-start sites (ATG), not that of the transcription start sites, which are unknown for these genes. (B–G) GFP reporter expression in tarsi. GFP driven by D. melanogaster Obp57e upstream sequence. (B) Dorsal view and (C) lateral view. GFP driven by (D) D. simulans and (E) D. sechellia Obp57e upstream sequence. (F and G) Removal of CCAT insertion from the sechellia > GFP construct restored GFP expression. (F) Dorsal view and (G) lateral view. (H) D. sechellia-specific 4-bp insertion in the upstream regions of Obp57e. Sequences from D. melanogaster (mel), D. simulans (sim), D. mauritiana (mau), and D. sechellia (sec) are aligned. Numbers indicate positions relative to the translation start site (ATG).

Figure 4. Generation of Obp57d/e Knock-Out Flies by Gene Targeting

(A) Targeted gene replacement by the ends-out method. A donor transgene integrated into the other chromosome by P element–based transformation was excised by the FLP recombination enzyme at FLP recognition target (FRT) sites. Resulting circular DNA was linearized by I-SceI, inducing a precise replacement of a target gene with a marker gene. Finally, a marker gene was excised by Cre recombinase that recognizes loxP sequences, leaving a single 34-bp loxP sequence. (B) Vector structures for Obp57d/e-targeted mutagenesis.

Figure 5. Olfactory Response of Obp57d/e Knock-Out Strains to HA in the Trap Assay

Response index (RI) = (N_h − N_w)/(N_h + N_w), where N_h is the number of individuals trapped in 1% HA solution, and N_w is that of individuals trapped in distilled water. At least 300 individuals were tested in five replications of the assay. Error bars indicate 95% confidence intervals determined by the binominal test.

Figure 6. Preferred Concentration of Acids in Oviposition Site–Preference Assay

Staged females were individually provided with four types of medium containing each acid at different concentrations. The concentrations were 0 mM, 10 mM, 20 mM, and 30 mM for acetic acid (AA), butyric acid (BA), and hexanoic acid (HA), and 0 mM, 2.5 mM, 5 mM, and 7.5 mM for octanoic acid (OA). The number of eggs laid on each medium was scored, and the weighted mean of acid concentration was calculated individually. Each bar represents a mean of 36 individuals from three replications. Error bars indicate standard error.