Female semiochemicals stimulate male courtship but dampen female sexual receptivity

Significance

Bactrocera dorsalis is a peculiar example of Tephritid with both males and females capable of releasing pheromones. Nevertheless, the identities of female-derived pheromones in this species remain unclear. In this study, we unveil the synthesis of four semiochemicals, ethyl laurate, ethyl myristate, ethyl cis-9-hexadecenoate, and ethyl palmitate, by female B. dorsalis as signals of sexual maturity, inciting male courtship. Remarkably, after mating, males acquire these female-specific compounds as mating history indicators, reducing female receptivity. We demonstrated the role of doublesex in regulating the expression of elongase11, orchestrating the biosynthesis of these semiochemicals exclusively in females. The dual functionalities of these semiochemicals in conferring female attractiveness and indicating male mating status expand our understanding of chemical communication in B. dorsalis.

Abstract

Chemical communication plays a vital role in mate attraction and discrimination among many insect species. Here, we document a unique example of semiochemical parsimony, where four chemicals act as both aphrodisiacs and anti-aphrodisiacs in different contexts in Bactrocera dorsalis. Specifically, we identified four female-specific semiochemicals, ethyl laurate, ethyl myristate, ethyl cis-9-hexadecenoate, and ethyl palmitate, which serve as aphrodisiacs to attract male flies and arouse male courtship. Interestingly, these semiochemicals, when sexually transferred to males during mating, can function as anti-aphrodisiacs, inhibiting the receptivity of subsequent female mates. We further showed that the expression of elongase11, a key enzyme involved in the biosynthesis of these semiochemicals, is under the control of doublesex, facilitating the exclusive biosynthesis of these four semiochemicals in females and guaranteeing effective chemical communication. The dual roles of these semiochemicals not only ensure the attractiveness of mature females but also provide a simple yet reliable mechanism for female mate discrimination. These findings provide insights into chemical communication in B. dorsalis and add elements for the design of pest control programs.

The oriental fruit fly, Bactrocera dorsalis (Hendel), is ranked as one of the most damaging and aggressive fruit pests due to its invasive potential and high reproductive capacity (1). Achieving successful reproduction necessitates efficient chemical signaling for the purpose of courtship and making choices on mating (2–4). The production of pheromones is, therefore, one of the most important steps to bring together sexually mature males and females for mating (4, 5). As a polyphagous species, the spatiotemporal presence of female B. dorsalis is less predictable due to the high spatial and temporal variability of ovipositional sites, resulting in the evolution of a non-resource-based mating system (6, 7). Male B. dorsalis aggregate in a mating lek, producing a characteristic high-pitched buzzing sound (8) and releasing male-borne pheromones (9), which could attract the female to move toward the lek site. The female can then choose to accept or reject the male attempts based on her perception of male pheromones, acoustic signals, and her own readiness to mate. The rejection of copulation by female B. dorsalis is frequently observed in natural leks (5, 10). Females can repeatedly resist male attempts by bending their abdomen or displaying downward ovipositor extrusion (11). In addition, in the wild lek site, there is a high frequency of noncalling males, which approach females without performing obvious courtship displays (10). Thus, it seems essential for females to evaluate the social status of males and female choice seems to be a key determinant of mating success in B. dorsalis. The key role of female B. dorsalis in mediating mating success and the degree of host damage underscores the urgent necessity for a more comprehensive understanding of its biology and ecology so that more efficient control measures can be developed (12). Despite decades of intensive studies, the identities of female-borne pheromones and the mechanisms underlying mating discrimination in B. dorsalis have not been fully unveiled. Whether female B. dorsalis can attract males based on their maturity, and whether and how sexual signals in female B. dorsalis coordinate female attractiveness and female choice are questions which warrant further scrutinization. In tandem with our constrained cognizance of female-derived sexual signals, the molecular mechanism underlying how sex determination pathway orchestrates female-specific biosynthesis in this species also remains remarkably scarce. The sex determination pathway is commonly believed to be a conserved mechanism governing an array of downstream genes, which in turn guides the intricate manifestation of sexually dimorphic traits (13–15). The signaling cascades that transform sex differences into alternative splicing of sex differentiation genes could intertwine various genes involved in multiple biosynthetic pathways, generating sex-specific compounds (16, 17). Unraveling the key nodes at which sex-determining signals connect with the biosynthetic pathway is crucial to understand the genetic basis of the chemical communication in B. dorsalis and to improve the monitoring and control systems of this economically important invasive species.

In this study, we provide evidence that B. dorsalis has evolved elegant semiochemical parsimony by employing four female-specific semiochemicals as both aphrodisiacs and anti-aphrodisiacs in different contexts. Four prominent semiochemicals exclusively produced in female B. dorsalis were identified, including ethyl laurate (EL), ethyl myristate (EM), ethyl cis-9-hexadecenoate (cEH), and ethyl palmitate (EP). These four chemicals act as aphrodisiacs by arousing male courtship behaviors, and their concentrations progressively amplify with female maturity. After mating, these female-borne semiochemicals could be sexually transferred to their male partners, acting as anti-aphrodisiacs, which would reduce female receptivity and defer subsequent mating of the recently mated males. We further unveiled the crucial role of doublesex (DSX) in regulating the expression of elongase11 (elo11), leading to the exclusive biosynthesis of these semiochemicals by females. Collectively, our data demonstrate the collective role of four semiochemicals in coordinating female attractiveness and mate discrimination, exemplifying the remarkable functional parsimony that underlies chemical communication in B. dorsalis.

Results

Identifications of Active Compounds.

We initially compared the chemical profiles of virgin female and male B. dorsalis (10 d postemergence) by gas chromatography–mass spectrometry (GC–MS) analyses and found that four compounds, ethyl laurate (EL), ethyl myristate (EM), ethyl cis-9-hexadecenoate (cEH), and ethyl palmitate (EP), were exclusively present in the female body extract (SI Appendix, Figs. S1 and S2 and Table S1). In addition, male antennae (10 d postemergence) responded robustly to these four compounds from the extract of virgin females, as revealed by gas chromatography–electroantennographic detection (GC-EAD) (Fig. 1A). We speculated that these four compounds were cuticular lipids, as they could be detected in a 10-min extract (4, 18). We analyzed the components of the hexane extracts from abdominal epidermal tissues and rectal glands in virgin females (10 d postemergence). The results from GC–MS revealed that these four compounds were only detected in abdominal epidermal tissues, which excluded the possibility of them being glandular products (SI Appendix, Fig. S3 A–C). We also noted that these four compounds could be identified on females regardless of mating status and were present on mated males (immediately following copulation) but not virgin males (10 d postemergence) (Fig. 1B). Through quantitative analyses, we determined that one virgin female B. dorsalis (10 d postemergence) contained approximately 2.19 ± 0.096 μg of EL, 1.43 ± 0.048 μg of EM, 1.41 ± 0.047 μg of cEH, and 0.87 ± 0.009 μg of EP (Fig. 1 C–F). The levels of these compounds significantly dropped by 18.86 ± 1.95% for EL, 16.14 ± 3.62% for EM, 19.69 ± 6.61% for cEH, and 8.94 ± 2.20% for EP in the mated female B. dorsalis (immediately extracted following copulation) relative to the virgin females (Fig. 1 C–F). The approximately concurrent increases of these four compounds on the mated males indicated a possibility of transference of these compounds from females to males. Based on these results, we suggested that during mating, female B. dorsalis could transfer four active semiochemicals onto males, which were originally absent in virgin males.

Fig. 1.

Identification of sex-specific compounds from B. dorsalis. (A) Responses of male antennae (10 d postemergence) to the hexane extract of virgin female B. dorsalis (10 d postemergence) by gas chromatography–electroantennographic detection (GC-EAD). GC corresponds to the responses from flame ionization detector (FID) on the GC. ♂ EAD corresponds to the male antennal responses. Y-scale for GC = 100 mV/div; Y-scale for EAD = 1 mV/div. Ethyl laurate (EL), ethyl myristate (EM), ethyl cis-9-hexadecenoate (cEH), and ethyl palmitate (EP). (B) Gas chromatography–mass spectrometry (GC-MS) profiles from male and female B. dorsalis under different mating status (10 d postemergence), and internal standard (IS, n-C18) was indicated (C) Quantification of the amount of EL under different mating status. (D) Quantification of the amount of EM under different mating status. (E) Quantification of the amount of cEH under different mating status. (F) Quantification of the amount of EP under different mating status. Data analyses were based on one-way ANOVA; different letters indicated significant difference, P < 0.05.

Female-Borne Semiochemicals Attract Males and Arouse Male Courtship.

To examine the levels of these four semiochemicals at different developmental stages of female B. dorsalis, we performed extensive GC–MS analyses of the female body extracts. Continuous increases of these four semiochemicals in female B. dorsalis were observed after the third day postemergence, and the four semiochemicals persisted at high levels from the 8th to 15th day postemergence (Fig. 2A and SI Appendix, Fig. S4). Female-specific semiochemicals typically indicate sexual maturity and can act as aphrodisiacs on males (19). We first compared the copulation success rates of female flies at different ages (1 to 15 d postemergence) with those of mature males (10 d postemergence) by single-pair mating assays. Females at the first or second day postemergence (in the absence of four semiochemicals) did not engage in copulation. Low copulation success rates were recorded for the females from the third to sixth day postemergence (when low amounts of the four semiochemicals were already detected). The copulation success rate sharply increased to 60% on the eighth day and remained consistently high until the 15th day postemergence. This increase was accompanied by a higher number of male copulation attempts (Fig. 2 B and C). These results indicate a correlation between copulation success rate and female maturity, prompting further investigation into the aphrodisiac roles of these four semiochemicals.

Fig. 2.

The roles of the four-semiochemical blend in attracting male flies and arousing male courtship. (A) Quantification of the amounts of ethyl laurate (EL), ethyl myristate (EM), ethyl cis-9-hexadecenoate (cEH), and ethyl palmitate (EP) from female B. dorsalis under different developmental stages (female of 1 to 15 d postemergence). (B) Copulation success rate between a virgin male fly (10 d postemergence) and a virgin female fly at different developmental stages (1 to 15 d postemergence). (C) Number of copulation attempts by virgin males (10 d postemergence) toward virgin females at different developmental stages (female of 1 to 15 d postemergence). (D) Top: Schematic of the mating arena where a treated virgin female [a 5-d virgin female was perfumed with hexane (5 d ♀ + H), extract of the 10-d female (5 d ♀ + E) or a semiochemical blend, (5 d ♀ + S)] was introduced to an untreated virgin male (10 d postemergence, 10 d ♂). Below: Copulation success rate. Statistical comparisons were conducted between each treated group with the control group (10 d ♀ × 10 ♂). (E) Copulation success rate between an untreated virgin male fly (10 d postemergence) and a virgin female (5 d postemergence) perfumed with different concentrations of the semiochemical blend. Data in (B, D, and E) were analyzed by Fisher’s exact tests for differences in copulation success rates with Bonferroni-corrected P values above each bar when significant (P < 0.05). Each treatment in (B), (D), and (E) contains 50 parallel mating pairs. (F) Number of copulation attempts by virgin males (10 d postemergence) toward virgin females (5 d postemergence) perfumed with different concentrations of the semiochemical blend. n above each bar in (C) and (F) represented the number of observed couples. (G) Field trap assays of the semiochemical blend (S) in Jiangxi, China. Methyl eugenol (ME, 5 mg/mL) was served as a positive control. C stands for hexane negative control. (H) Field trap assays of the semiochemical blend in Yunnan, China. Ten replicates were conducted for each experiment in (G) and (H). Data in (A), (C), (F), (G), and (H) were analyzed by one-way ANOVA; different letters indicated significant difference, P < 0.05.

Next, we performed single-pair mating assays to monitor copulation success rates after perfuming females with different chemicals following a previous study (20). Since females typically have low receptivity on the third day postemergence, we used virgin females at the fifth day for the mating trials. One hour following perfuming, compounds remaining on the body surface of female B. dorsalis were confirmed by GC–MS (SI Appendix, Fig. S5). Our results demonstrated that 58% and 62% of virgin males (10 d postemergence) could successfully court the immature females (5 d postemergence) when those females were perfumed with a four-semiochemical blend (equivalent to the amount carried by a 10-d female) or the extract of the 10-d female, and the copulation success rates were comparable to those of the 10-d mature females (10-d males and 10-d females, 66%), while only a few males (8%) engaged in copulation with the 5-d immature females perfumed with the hexane vehicle (Fig. 2D and SI Appendix, Fig. S5). Males are able to display different intensities of courtship toward females depending on the sexual signals present in females (21). We conducted a comprehensive analysis quantifying both the duration of male courtship (the time spent pursuing a given female), and the frequency of courtship behaviors, including orientation, following, wing-fanning, antennal touching, licking, tussling, and copulation attempts (SI Appendix, Fig. S6A) (5, 11, 22). We found that perfuming 5-d immature females with a semiochemical blend or the 10-d female extract resulted in significant changes in several courtship parameters of males, including faster locating of females, reduced time spent in following and pausing, and prolonged duration of wing-fanning and tussling, ultimately leading to decreased copulation latency, compared to the hexane control (SI Appendix, Fig. S6 A and B). The male became more excited and tended to court more vigorously toward the immature female fly (5 d postemergence) perfumed with a semiochemical blend compared with the male responses toward the female perfumed with hexane (Movie S1). Additionally, there was a positive correlation between the amount of semiochemical blend on the female body surface and both the copulation success rate and the number of male copulation attempts (Fig. 2 E and F), indicating that these four semiochemicals could arouse male courtship in a concentration-dependent manner and making them potential female-derived aphrodisiacs for male B. dorsalis. We further explored the attractiveness of the four-semiochemical blend (the amount expressed by a 5-d or 10-d female) to virgin male flies (10 d postemergence) by trap assays and demonstrated that male flies exhibited a robust preference for the semiochemical blend odors (SI Appendix, Fig. S7). However, there was a reduced attraction toward individual chemicals (SI Appendix, Fig. S7). The aforementioned experiments were conducted using long-term laboratory-adapted B. dorsalis. We next conducted two field assays by trapping field B. dorsalis from Jiangxi (June 2022) and Yunnan (July 2022) Province in China. The four-semiochemical blend showed potential in trapping male B. dorsalis with a negligible number of females (0 ± 0/65 ± 6 in Jiangxi and 1 ± 0/58 ± 4 in Yunnan). The trap rate was slightly lower than that of the commercial trap agent, methyl eugenol (ME), a highly attractive male lure of B. dorsalis (23) (142 ± 12 and 126 ± 10) (Fig. 2 G and H). We thus concluded that these four female-borne semiochemicals could be considered active sex attractant pheromones for B. dorsalis by exhibiting pivotal aphrodisiac and attractive activities, thus eliciting strong responses from male B. dorsalis.

Female-Borne Semiochemicals on the Male Body Surface Discouraged Copulation.

Since males might acquire these four semiochemicals from their female partners during mating (Fig. 1B), we hypothesized that these semiochemicals might function as anti-aphrodisiacs to impede copulation. To address this hypothesis, we quantified these four semiochemical concentrations on mated males at several time points: immediately after the end of copulation (AEC, referred to as 0 h AEC), and ~ 1 h to 7 h AEC. The levels of these four semiochemicals declined over time on the mated males, and these four semiochemicals were undetectable on males at 6 h AEC, probably due to evaporation (Fig. 3A and SI Appendix, Fig. S8). We next examined the copulation success rates between males at different mating status (10 d postemergence) and virgin females (10 d postemergence). As expected, the copulation success rate of couples with 6 h AEC males and virgin females was comparable to that of virgin couples (virgin male and virgin female), while the copulation success rates declined to 4% and 12% in males at 0 and 2 h AEC (Fig. 3B). These data suggest that the decline in the copulation success rates immediately after copulation coincided with the high levels of four female-specific semiochemicals on the male body surface. It could be argued that the mating desire of these recently mated males would drop significantly after completing the exhausting mating process. However, we note that at 5 h AEC, the copulation latency and the number of male copulation attempts (toward virgin females at 10 d postemergence) had already recovered to a level comparable to that of virgin males (SI Appendix, Fig. S9), indicating that the reduced copulation success rate at 5 h AEC is probably due to the low level of female receptivity rather than the absence of male attempts. We then analyzed the female responses to male courtship by manually annotating the durations of different kinds of ovipositor extrusion to calculate the ovipositor extrusion index, as described in a previous study (24). By annotating female behaviors from the initiation of male courtship until copulation (or up to 60 min), we found that females displayed different patterns of ovipositor extrusion toward virgin and recently mated males and that a high level of full ovipositor extrusion might impede copulation and repel male courtship (Fig. 3C). Virgin females (10 d postemergence) preferentially displayed partial ovipositor extrusion in response to the courtship of virgin males or males at 6 h AEC to allow copulation, while virgin females exhibited high levels of full ovipositor extrusion toward these recently mated males (0 to 5 h AEC, 10 d postemergence) (Fig. 3D), indicating that females are less receptive to the copulation attempts of recently mated males.

Fig. 3.

Females showed a high level of full ovipositor extrusion to impede copulation when the male was marked by female-borne semiochemicals. (A) Quantification of the amounts of ethyl laurate (EL), ethyl myristate (EM), ethyl cis-9-hexadecenoate (cEH), and ethyl palmitate (EP) from male B. dorsalis at 0 to 7 h AEC (after the end of copulation). (B) Copulation success rate between an untreated virgin female fly and a male under different mating status. Virgin, 0, 2, 5, 6 stand for a virgin, 0 h, 2 h, 5 h, or 6 h AEC male (10 d postemergence) to mate with a virgin female (10 d postemergence). (C) Images showing the dorsal view of female abdomens either not extruding (NO Ovi. Ext.) or performing partial ovipositor extrusion (Partial Ovi. Ext.) or full ovipositor extrusion (Full Ovi. Ext.) (Left). Full ovipositor extrusion would impede copulation (Right). (D) Ovipositor extrusion index displayed by the virgin females (10 d postemergence) in response to the courtship exhibited by males (10 d postemergence) under different mating status. Each droplet represented the time spent with certain type of ovipositor extrusion during courtship over the total time of courtship (or up to 60 min). (E) Single-pair of copulation success rate between an untreated virgin female fly (10 d postemergence) and a virgin male (10 d postemergence) perfumed with different compounds. ♂ + H, E, or S stand for the virgin males were perfumed with hexane, extract from the male at 0 h AEC, or a semiochemical blend equivalent to the amount in a male at 0 h AEC to mate with a virgin female. Each treatment in (B) and (E) contains 50 parallel mating matches. Data in (B) and (E) were analyzed by Fisher’s exact tests, with Bonferroni-corrected P values above each bar to indicate the significance (P < 0.05). n above each bar in (B) and (E) represented the number of fly couples engaged in successful copulation. (F) Ovipositor extrusion index displayed by virgin females (10 d postemergence) in response to the courtship exhibited by virgin males (10 d postemergence) perfumed with different compounds. n above each bar in (D) and (F) represented the number of observed couples. Statistical comparisons in (A), (D), and (F) were based on one-way ANOVA; different letters indicated significant difference, P < 0.05. (G) Competition between two males (10 d postemergence), virgin or 2 h (6 h) AEC, to copulate with a virgin female (10 d postemergence). (H) Competition between two virgin males (10 d postemergence), one male perfumed with a semiochemical blend (S) (or extract from the male at 0 AEC, E) and another male perfumed with solvent hexane (H), to copulate with a virgin female (10 d postemergence). Pie charts represented copulation success of the rival males. Statistical comparisons in (G) and (H) were based on the chi-square test, *P < 0.05; **P < 0.01; and ***P < 0.001.

To directly test whether the presence of these four semiochemicals on males triggers alterations in female ovipositor extrusion responses, we perfumed virgin male (10 d postemergence) with a semiochemical blend (equivalent to the amount carried by a male at 0 h AEC) or with the extract from the male at 0 h AEC (SI Appendix, Fig. S10). The results showed that virgin females (10 d postemergence) displayed higher levels of full ovipositor extrusion toward virgin males perfumed with the semiochemical blend or extract from the male at 0 h AEC, and the copulation success rates also declined in these fly pairs compared with the hexane-treated control (hexane-treated virgin male and untreated virgin female) (Fig. 3 E and F). We extended this analysis by performing a competition assay, in which a virgin female (10 d postemergence) was allowed to choose between virgin and 2 h (or 6 h) AEC males (10 d postemergence). We found that virgin females exhibited a stronger preference for virgin males over 2 h AEC males, but showed a similar preference for 6 h AEC males as for virgin males (Fig. 3G and Movie S2). By introducing semiochemical blend- or extract-perfumed virgin males (10 d postemergence) into the mating arena alongside hexane-perfumed virgin males, the hexane-perfumed virgin males (10 d postemergence) exhibited a clear copulation advantage over semiochemical- or male extract-perfumed virgin males (Fig. 3H and Movie S3). Taken together, our results suggest that the presence of these four female-specific semiochemicals on males could act as anti-aphrodisiacs, reducing female receptivity by eliciting a high level of full ovipositor extrusion.

Elo11 Is Required for the Production of These Four Semiochemicals.

To identify the candidate genes involved in the synthesis of these four semiochemicals, we performed comparative RNA sequencing (RNA-seq) analysis on abdominal integument samples from virgin females and males (10 d postemergence) (Accession no. PRJNA901648), the possible synthetic site for these four semiochemicals (SI Appendix, Fig. S3). Since these four semiochemicals are ethyl esters of linear-chain fatty acids, we hypothesized that certain elongases (elos), which act as the key rate-limiting enzymes for linear-chain fatty acid biosynthesis (25), could be preferentially expressed in females. Compared to male samples, several elo transcripts were highly expressed in the female abdominal integument, of which elo5 and elo11 exhibited the highest upregulation in female flies (SI Appendix, Fig. S11A and Dataset S1). Sequence alignment revealed that different ELO proteins showed high homology and contained conserved HXXHH and YXYY motifs (SI Appendix, Fig. S12), which are believed to be involved in the elongation steps (26). Quantitative reverse-transcription polymerase chain reaction (qRT-PCR) verified the high expression levels of elo5 and elo11 in females (SI Appendix, Fig. S11B). By analyzing the transcript levels of these two genes in different tissues, we found that elo5 and elo11 were mostly abundant in the female abdominal integument and head (Fig. 4A). Moreover, the expression levels of elo5 and elo11 in females increased with developmental stage (Fig. 4B), resembling the increasing pattern of these four semiochemicals. We then applied RNA interference (RNAi) to validate the roles of candidate elos in the production of these four semiochemicals. Injection of dselo5 or dselo11 into females significantly decreased the mRNA levels of the target genes (SI Appendix, Fig. S13). GC–MS analysis showed that the knockdown of elo11 caused significant decreases in the levels of these four semiochemicals in females (54.80 ± 10.19%, 50.76 ± 5.46%, 52.44 ± 6.23%, and 55.38 ± 3.58% decreases for EL, EM, cEH, and EP, respectively, compared to levels in the dsGFP-injected flies), but no significant changes were observed following the knockdown of elo5 (Fig. 4C and SI Appendix, Fig. S13). Thus, elo11 may function as one of the key enzymes for the synthesis of these four semiochemicals in female flies. As expected, elo11-knockdown virgin females showed a reduced copulation success rate compared with the control females (received equal amounts of dsGFP) when paired with untreated virgin males (10 d postemergence) (Fig. 4D), and there were significant increases in the durations of orienting and following but decreases in antennae touching and wing-fanning in these pairs (untreated males and dselo11-injected females, SI Appendix, Fig. S14). We did not observe significant changes in the copulation rates and courtship patterns after the females were injected with dselo5 (Fig. 4D and SI Appendix, Fig. S14).

Fig. 4.

The expression of elongase11 (elo11) is required for these four semiochemical biosynthesis in female flies. (A) Tissue distribution of elo5 and elo11 mRNA in the virgin females (10 d postemergence). H, P, F, I, MG, MT, and O stand for head, pronotum, fat body, integument, midgut, malpighian tube, and ovary. (B) Temporal expressions of elo5 and elo11 in virgin female abdominal integuments under different developmental stages (female of 1 to 10 d postemergence). (C) Effects of elo5 and elo11 knockdown on the four semiochemical productions in the virgin females (10 d postemergence). Statistical comparisons in (A–C) were based on one-way ANOVA; different letters indicated significant difference, P < 0.05. (D) Copulation success rates of virgin male (10 d postemergence) and dsRNA-injected virgin female (10 d postemergence) couples. Each treatment contains 50 parallel mating matches. n above each bar in (D) represented the number of fly couples engaged in successful copulation. Data in (D) were analyzed by Fisher’s exact tests, with Bonferroni-corrected P values above each bar to indicate the difference (P < 0.05). (E) Representative and truncated GC–MS profiles from yeast cells transformed with pYES2 or pYES2-ELO11 with adding precursor compound, ED (ethyl decanoate). Ethyl laurate (EL), ethyl myristate (EM), ethyl palmitate (EP), and internal standard (IS, n-C18) were indicated.

To confirm the elongation activity of elo11, we expressed B. dorsalis ELO11 protein in the Saccharomyces cerevisiae strain INVSc1 under the control of an inducible GAL1 promoter. Following incubation with 2% galactose and 1% raffinose, hexane extracts from yeast cells were subjected to GC–MS analyses, which showed that forced ELO11 expression led to the production of EL, EM, and EP in the presence of ethyl decanoate (ED) as the precursor, whereas yeast cells transformed with the empty vector did not generate any target products (Fig. 4E and SI Appendix, Fig. S15). Due to the lack of proper desaturases to catalyze the double bonds at the specific position, we failed to generate cEH in our expression system. In conclusion, these results demonstrate that the expression of elo11 was closely associated with the biosynthesis of these four semiochemicals, which could regulate the copulation success rate in B. dorsalis. However, we cannot rule out the possible roles of other genes in the biosynthesis of these four semiochemicals.

Doublesex (DSX) Controls elo11 Expression and the Synthesis of Four Semiochemicals.

The synthesis of these four female-specific semiochemicals is associated with the expression of elo11, we then wondered whether elo11 expression was controlled by the key transcription factor in the sex determining pathway. To investigate the role of DSX in elo11 expression and the synthesis of these four semiochemicals, 1,000 eggs (laid within 2 h, the period when dsx initiates sex-dimorphic expression) were injected with dsdsx^F, and the other 1,000 eggs were injected with dsdsx^M, targeting the regions specific to female and male isoforms (SI Appendix, Table S2 and Fig. S16). The embryos injected with dsdsx^M resulted in 138 emerged flies comprising 110 females, 24 males, and 4 intersexual individuals, yielding a female-to-male sex ratio of 4.6:1.0 (SI Appendix, Table S3). In addition, 1,000 embryos injected with dsdsx^F resulted in 130 emerged flies with a female-to-male ratio of 1.0:7.3 (SI Appendix, Table S3). In the dsx^F-knockdown females, the female-specific dsx transcript was observed at a lower level than that of dsGFP-injected females, and the male dsx transcript was also observed in these masculinized females (SI Appendix, Fig. S16). Likewise, in the dsx^M-knockdown male group, we detected a lower level of dsx^M transcript than that in the control group, and the female dsx transcript could also be detected in these dsdsx^M-injected males (SI Appendix, Fig. S16). We then performed morphometric measurements in these adults to examine the effects of dsx knockdown. Five intersexual individuals exhibited female dorsal phenotypes with bristles on the side of their terga in the dsdsx^F-injected pools, and four intersexual individuals exhibited male dorsal phenotypes but without bristles on the side of terga in the dsdsx^M-injected pools (Fig. 5A). Observation of the internal reproductive organs of the same set of insects revealed immature ovaries and unstructured testes (Fig. 5A). The four female-specific semiochemicals could be detected in the dsdsx^M-injected males but not in the dsGFP-injected males. In contrast, the levels of EL, EM, cEH, and EP were dramatically reduced in RNAi females compared with those in the dsGFP-injected females (Fig. 5A and SI Appendix Fig. S17). qRT–PCR showed a detectable increase in elo11 expression in these dsdsx^M-feminized males. The decreases in these four female semiochemical levels were also paralleled with the remarkable decrease in elo11 expression in the dsdsx^F-treated females (Fig. 5B).

Fig. 5.

Doublesex (DSX) directly binds to the upstream regulatory regions of elongase11 (elo11) to regulate its expression. (A) The effects of dsx^M or dsx^F knockdown on the production of these four semiochemicals, with the representative images of the abdominal phenotypes and internal reproductive organs (10 d postemergence, virgin). Images showed dorsal views and internal reproductive organs of representative flies. The scale bar of testis and ovaries in male and female represented 300 μm and 650 μm, and the scale bar of abdomens in male and female represented 0.1 cm. Ethyl laurate (EL), ethyl myristate (EM), ethyl cis-9-hexadecenoate (cEH), ethyl palmitate (EP), and internal standard (IS, n-C18) were indicated. (B) Temporal changes of elo11 by qRT-PCR in virgin females and males (10 d postemergence) after dsRNA injection. (C) Elo11 promoter activities in response to dsx^F and dsx^M overexpression. Different letters denote the significant differences calculated by one-way ANOVA (P < 0.05). (D) Copulation success rate between a dsdsx^M (or dsGFP)-treated virgin male and an untreated virgin female (10 d postemergence). (E) Ovipositor extrusion index displayed by untreated virgin females (10 d postemergence) in response to the courtship exhibited by dsdsx^M-treated or dsGFP-treated virgin males (10 d postemergence). (F) Copulation success rate between a dsdsx^F (or dsGFP) treated virgin female (10 d postemergence) and an untreated virgin male (10 d postemergence). Each treatment in (D) and (F) contains 50 parallel mating matches. n above each bar in (D) and (F) represented the number of fly couples engaged in successful copulation in each group. Data in (D) and (F) were analyzed by Fisher’s exact tests for differences in copulation success rate with Bonferroni-corrected P values above each bar when significant (P < 0.05). (G) Number of copulation attempts by untreated virgin males (10 d postemergence) toward virgin females injected with dsdsx^F or dsGFP (10 d postemergence). n above each bar in (E) and (G) represented the number of observed couples. Statistical comparisons in (B), (E), and (G) were based on Student’s t test. The level of significance for the results was set at *P < 0.05; **P < 0.01; and ***P < 0.001.

The presence of a putative DSX-binding site in elo11 (−117) suggests that elo may be directly regulated by DSX. An electrophoresis mobility shift assay (EMSA) demonstrated that the DSX-DNA-binding domain specifically and efficiently bound the wild-type site of the elo11 promoter region, which was abolished when the DSX-DNA-binding site was mutated (ACAATG to GAGCAG) (SI Appendix, Table S2 and Fig. S18). To confirm the role of DSX in regulating elo11 expression, the elo11 promoter-reporter was cotransfected with dsx^F or dsx^M plasmid into Sf9 cells. The luciferase reporter assay showed that the forced overexpression of dsx^F increased elo11-reporter activity, while a suppressive effect of dsx^M was observed (Fig. 5C). Collectively, these results indicate that DSX can specifically bind to the upstream regulatory regions of elo11, and dsx^M inhibits elo11 expression in males, whereas dsx^F promotes elo11 expression in females.

We next investigated whether changes in dsx expressions could lead to modifications in mating behaviors. Untreated virgin females (10 d postemergence) showed a higher level of full ovipositor extrusion (lower level of receptivity) toward courting males who received dsdsx^M, consequently resulting in a decreased copulation success rate compared to males injected with dsGFP (Fig. 5 D and E and Movie S4). The number of copulation attempts by untreated virgin males toward dsdsx^F-injected virgin females was significantly decreased compared to females treated with dsGFP (10 d postemergence), and the copulation success rate was significantly lower in the former group (Fig. 5 F and G and Movie S5). Collectively, these findings indicate that dsx^F activates elo11 expression and dsx^M represses elo11 expression to ensure female-specific production of these four semiochemicals and to elicit a specialized courtship drive in males toward females.

Discussion

It is generally acknowledged that lekking male B. dorsalis initiate sexual behaviors by releasing male-specific pheromones to attract females to the lek sites (9, 22). Our study addresses a long-standing mystery regarding the functional role of female-borne semiochemicals in B. dorsalis and provides direct evidence that female B. dorsalis also employ semiochemicals as attractants and aphrodisiacs to attract males and arouse male courtship. Our data suggest that the age-dependent production of these four semiochemicals in female B. dorsalis might impart a sexual signal to guarantee that these mature females are attractive to males and are ready to mate. Field trials in previous studies indicated that the number of mature female B. dorsalis present on a particular tree remained relatively constant throughout the late afternoon; in contrast, the number of males on the same tree increased across census periods, peaking at the 18:30 census (the time of high sexual activity) (27, 28). Thus, it is reasonable to hypothesize that females might employ these four semiochemicals to convey their maturity to males. This, in turn, could indicate that the site with mature females nearby presumably constitutes a suitable site for male calling, resulting in male aggregating and initiation of male courtship (29). In addition, after copulation, these mated males are marked for a specific period (6 h, as four semiochemicals disappeared from males 6 h after copulation), to advertise the recent mating status, which serves as a mechanism to reduce the receptivity of other females. It seems to be very important for female B. dorsalis to be able to discriminate against these recently mated males, as previous studies have indicated that males must undergo a complementary nutrition stage prior to another mating activity and a lower level of egg hatchability has been observed when a female was housed with mated males compared with virgin males (30, 31). However, we do not have enough evidence to confirm whether there has been depletion of reproductive reserves in these recently mated male B. dorsalis, nor can we determine whether the duration of 6 h is sufficient for the full replenishment of the reproductive reserves. In addition, further exploration is needed to determine whether wild females also utilize these four semiochemicals to discriminate lekking males, establishing a comprehensive understanding of mate selection in this species. Nevertheless, it is likely that female B. dorsalis could recognize these recently mated males through their own signals, which can be communicated to other females through a less sophisticated information processing system. This system allows other females to simply avoid males that emit familiar scents. To summarize, these four semiochemicals can serve as both aphrodisiacs and anti-aphrodisiacs, allowing female B. dorsalis to communicate their sexual maturity and discriminate against recently mated males. It is important to note that our focus on these four compounds may limit our understanding, as the 10-min extraction method may overlook other crucial elements (32), which could potentially influence the mating behaviors of B. dorsalis.

A series of saturated/unsaturated fatty acid esters are commonly found in female tephritid flies. EL, EM, and EP have been found in Bactrocera correcta (33), Bactrocera oleae (34), Bactrocera frauenfeldi (35), Bactrocera musae (36), Bactrocera bryoniae (37), and Bactrocera tryoni (38) female extracts. In addition, EL, EM, and EP were found in the cuticle of mature female B. tryoni, Bactrocera carambolae, Bactrocera papaya, Bactrocera philippinensis and Bactrocera invadens, with varying relative abundances among different species (39, 40), suggesting that these compounds might have roles in chemical recognition. Notably, EL, EM, and EP elicited electrophysiological responses in male and female B. bryoniae (37) and B. oleae (34), but only EL elicited responses in male B. musae (36), which reveals the functional divergence of these fatty acid esters in different tephritid flies. However, the functional roles of these electrophysiologically active female-specific esters have not yet been fully elucidated in tephritid flies, despite their potential profound implications for female biology, considering the large quantity of these compounds produced. The current study addresses this knowledge gap by demonstrating that EL, EM, cEH, and EP function as both aphrodisiacs and anti-aphrodisiacs for B. dorsalis, shedding light on the pheromonal roles of these fatty acid esters in tephritid flies. Notably, some recent studies also found these similar fatty acid esters serving as brood pheromone for Varroa destructor (41), aggregation pheromones for Drosophila (42), and attractant pheromone for Glossina morsitans (4). These identified pheromones have been employed parsimoniously for multiple purposes in different scenarios, known as pheromonal parsimony (43). In Drosophila melanogaster, cis-vaccenyl acetate (cVA) presents a typical example of a single molecule leading to different behaviors in varying social contexts (44). The presence of cVA in courting male stimulates female reproductive motivation and increases the probability of copulation success, while cVA suppresses the courtship motivation of other males (45–47). Phenylacetonitrile has been proposed as both an aggregation pheromone and a male courtship inhibitor in Schistocerca gregaria (48–50). This particular compound has been reported as a chemical defense for deterring cannibalism in Locusta migratoria (51). The characterizations of the parsimonious usage and the functional roles of these compounds hold the key to the complex and integrated insect chemical communication system, which can process considerable quantities of context-dependent information with a limited pheromonal repertoire.

Furthermore, we identified elo11 as a key enzyme for the biosynthesis of these four semiochemicals, and the binding of DSX at the upstream sequence of the elo11 locus dictates the expression of elo11. The binding of dsx^F activates the transcription of elo11, whereas male-specific dsx^M shows negative regulatory activity on elo11 expression. When dsRNA injections targeting dsx^F and dsx^M were administered to the embryos respectively, we observed a reduction in the production of these four semiochemicals in masculinized females, and a discernible presence of these semiochemicals in feminized males, which correlated with the changes in the copulation success rates. We surmise that the exclusive production of these semiochemicals in females might impede the same-sex sexual behaviors since they function as anti-aphrodisiacs to females and attractants to males. However, it remains unclear whether males marked by these semiochemicals can attract other males and initiate male-to-male courtship, as reported in the case of Blattella germanica (16).

Collectively, we identified four female-specific semiochemicals as aphrodisiacs that advertise female maturity to attract males and stimulate male courtship. After mating, these semiochemicals could be transferred onto males, acting as anti-aphrodisiacs to reduce the receptivity of other female encounters. The expression of elo11, a key enzyme involved in the biosynthesis of these semiochemicals, is under the control of DSX to facilitate these four semiochemical production only in females. Taken together, our findings highlight a unique example of semiochemical parsimony in which four female-specific semiochemicals act as both aphrodisiacs and anti-aphrodisiacs, eliciting different behavioral reactions depending on the social context, to facilitate effective chemical communication in B. dorsalis.

Materials and Methods

Insects and Cell Lines.

B. dorsalis were obtained from a laboratory-reared stock colony (Key Laboratory of Pesticide and Chemical Biology, South China Agricultural University, Guangzhou, China) and maintained at 28 °C with 70% relative humidity under a 14:10 (light:dark) h photoperiod. The artificial diets for adult flies consisted of yeast extract (25%) and sucrose (75%) (52). To prevent copulation, we separated male and female flies on the first day postemergence and placed male and female flies in separate chambers. Each chamber (30 x 30 x 30 cm) consisted of 500 same-sex flies.

Sf9 cells were derived from the pupal ovarian tissue of Spodoptera frugiperda, which were routinely maintained at 28 °C (53). A total of 5 × 10⁵ cells were inoculated in a total volume of 3 mL antibiotic-free Grace’s medium (M4001-01, Elite-Media), which contained 7% fetal bovine serum (FBS), 0.3% yeast extract, and 0.3% lactalbumin hydrolysate, and grown in a 25-cm² plastic tissue culture flask.

The virgin and mated flies, the rectal glands (54), and the abdominal epidermis were prepared, and the identification of compounds was conducted following the methodology from a previous study with slight modifications (55). The copulation success rate was defined as the percentage of fly couples engaged in successful copulation during a 1-h observation (56), and the number of copulation attempts by males and copulation latency were recorded. We followed established protocols described in previous studies for various assays, including heterologous expression (17), trap assay (57), competitive mating assay (58), chemical application (20, 59), gene knockdown of dsx (60, 61), and electrophoresis mobility shift assay (62).

Detailed materials and methods used in this study are described in SI Appendix, Materials and Methods, including Identification of Compounds, Gas Chromatography–electroantennographic Detection, Behavioral Assays, Chemical Application, RNA-seq and Assembly, RNA Extraction, QRT-PCR and Semiquantitative RT-PCR, RNAi, Heterologous Expression, Recombinant Protein Expression and Purification, Electrophoresis Mobility Shift Assay, and Dual Luciferase Reporter Assay.

Supplementary Material

Appendix 01 (PDF)

Dataset S01 (XLSX)

Dataset S02 (XLSX)

Movie S1.

Video clip showing typical courtship behavior of untreated virgin males (10 d postemergence, 10 d ♂) toward hexane- (5 d ♀ + H) or semiochemical blend-perfumed (5 d ♀ + S) virgin female (5 d postemergence).

Movie S2.

Video clip showing competitive assays associating with a male at 2 AEC (or 6 AEC) and a virgin male to mate with a given virgin female. The flies used in the competitive assay were at 10 d postemergence. The rival flies (2 h AEC or 6 AEC) were marked by UV-fluorescent powder. AEC stands for after the end of copulation.

Movie S3.

Video clip showing competitive assays associating with a semiochemical blend-perfumed virgin male (10 d ♂ + S) (or extract from the male at 0 h AEC-perfumed virgin male, 10 d ♂ + E) and a hexane-perfumed virgin male (10 d ♂ + H) to mate with a given virgin female (10 d ♀). The flies used in the competitive assay were virgin females and males at 10 d postemergence. The rival flies (semiochemical blend-perfumed or extract-perfumed male flies) were marked by UV-fluorescent powder. AEC stands for after the end of copulation.

Movie S4.

Video clip showing the responses of untreated virgin females (10 d postemergence) to the courtship of dsdsx^M-treated (or dsGFP-treated) virgin males (10 d postemergence).

Video clip showing typical courtship behaviors exhibited by untreated virgin males (10 d postemergence) toward dsdsx^F-treated or dsGFP-treated virgin females (10 d postemergence).

Data, Materials, and Software Availability

Sequencing data have been deposited in the bioproject PRJNA901648 in the NCBI (63). Data for qualification and quantification of four semiochemicals were available on Zenodo (DOI: 10.5281/zenodo.8355098) (64). All study data are included in the article and/or supporting information.