Introduction

Sexual plasticity, the ability to modify sexual state or reproductive behavior in response to changing social conditions, is observed in both vertebrates and invertebrates (1–5). In rodents, exposure to unfamiliar males often leads to the sudden termination of pregnancy, known as the Bruce effect. It is induced by male urinary peptides, such as MHC I peptides, activating the vomeronasal organ (6–8). This effect enhances reproductive fitness of both sexes, by eliminating the offspring of competing males and enabling females to select better mates even after conception. Many species also adapt their reproductive behaviors in response to the social sexual context change (SSCC), involving encounters with new sexual partners or competitors. Understanding the neuronal circuit mechanisms behind female responses to SSCC emerges as a central focus in neuroscience (9–12).

Drosophila melanogaster, the fruit fly, displays various social behaviors like aggregation, aggression, and sexual behavior (13–15). Similar to rodents, they primarily use the olfactory system to communicate socially through pheromones (16, 17). Some of these pheromones act as aphrodisiacs, while others regulate aggression or foster aggregation. For instance, cis-vaccenyl acetate (cVA) attracts females but repels males and promotes aggregation (14, 18, 19). 7-Tricosene (7-T), a cuticular hydrocarbon (CHC) present in males, is an aphrodisiac to females and affects social interactions between males (20, 21). On the other hand, 7,11-Heptacosadiene (7,11-HD), a related female-specific pheromone, functions as an aphrodisiac to males, triggering courtship behavior and involving species recognition (22, 23).

The fruit fly’s chemo-sensory organs, present in various body parts, detect these pheromones (24, 25). Olfactory receptor neurons (ORNs) in sensilla of the antenna and maxillary palps pick up long-range volatile pheromones like cVA, while short-range pheromones like 7-T are sensed by neurons on the forelegs and labellum (16, 17, 24).

The olfactory receptor Or47b, expressed ORNs located in at4 trichoid sensilla on the third antennal segment, is involved in several socio-sexual interactions, including male mating success, mating partner preference, and female aggression to mating pairs (12, 26–29). In males, Or47b senses fatty acid methyl ester and fatty acid that affect mating competition and copulation (30, 31). While Or47b’s role in female aggression is established (12), its involvement in female sexual behaviors is uncertain. For both sexes, the Or47b ORNs project to the VA1v glomeruli, where the VA1v projection neurons receive their signal and project to the mushroom body calyx and lateral horn. Male Or47b neurons link with neurons like aSP5, aSP8, and aSP9, which express a male-specific transcriptional factor Fru^M (32).

CHC pheromones that function as short-range pheromones are mainly detected by neurons on the forelegs and labellum that express gustatory receptors (GR), ionotropic receptors (IR), or the ppk/DEG-ENaC family of sodium channels (16, 17, 24). CHCs like 7-T and 7,11-HD are sensed by ppk23-expressing M and F cells in the tarsi (33). 7-T and cVA are sensed by M cells expressing ppk23, while 7,11-HD and 7,11-ND are sensed by F cells expressing ppk23, ppk25, and ppk29. In males 7-T or 7,11-HD affects the neuronal activity of the Fru^M-expressing P1 neurons (34–36). However, how these CHCs signal in the female brain remains unknown.

Sperm ejection is a process through which females can remove the male ejaculate or the mating plug after copulation. This phenomenon was observed in various animal species including feral fowl (37), blacklegged kittiwake (38), and dunnock (39). In the fruit fly, it typically occurs approximately 90 minutes after mating (40). This specific interval, referred to as ‘ejaculate holding period (EHP)’, likely affects sperm usage and fecundity (40, 41). The neurosecretory neurons in the brain pars intercerebralis (PI) that produce diuretic hormone 44 (Dh44), an insect orthologue of the corticotropin-releasing factor, regulate EHP (40). Drosophila females seem to signal the social sexual context through sperm ejection. For instance, when placed in the food patch with the male ejaculate deposited by other females, females exhibited increased likelihoods of egg-laying (42). However, it remains unknown whether the social sexual context influences sperm ejection and EHP.

The pC1 neurons in females, expressing a specific transcriptional factor Dsx^F, integrate olfactory and auditory cues associated with male courtship (43, 44). The female pC1 neurons in flies, their male counterparts (i.e., P1 neurons), and the ventrolateral subdivision of ventromedial hypothalamus (VMHvl) neurons in mice share conserved circuit configurations and demonstrate functional similarity in coordinating social and sexual behaviors (45, 46). There are Dsx positive 14 pC1 neurons in each brain hemisphere of the brain, responsive to the male sex-pheromone cVA and courtship songs (44, 47).

Connectome analyses identified 10 pC1 neurons, categorized into five subtypes, with pC1a-c associated with mating behaviors and pC1d and e subtypes linked to aggression (47–52). Although direct evidence connecting pC1 neurons to sperm ejection is limited, they are promising candidates for regulating sperm ejection or EHP, because sperm ejection enables females to eliminate the mating plug and male ejaculate, restoring sexual attractiveness (53).

In this study, we demonstrate that two male pheromones, 2-methyltetracosane (2MC) and 7-T, significantly shorten EHP through Or47b neurons and ppk23 neurons, respectively. These pheromonal pathways converge onto pC1 neurons, increasing their cAMP levels. The elevated cAMP in pC1 neurons reduced EHP, similar to the effects of the male pheromones. Elevated cAMP also enhances pC1 neuron excitability, rendering them more responsive to both olfactory and auditory male courtship cues and promoting further mating after earlier removal of the mating plug. Our study introduces a novel behavioral paradigm, shedding light on the intricate molecular and neuronal pathways underlying female sexual plasticity.

Results

Male-induced EHP shortening (MIES) is dependent on olfaction

To investigate the impact of changes in social sexual context on EHP, we conducted a comparison between the EHP of post-mating females isolated from any male presence and those exposed to actively courting, naive wild-type Canton-S (CS) males immediately following copulation (Fig. 1A). Notably, the EHP of females incubated with naive males was approximately 30 minutes shorter than that of females left in isolation after mating (Fig. 1A, 1B). We refer to this phenomenon as ‘male-induced EHP shortening (MIES)’. In contrast, there was little difference in EHP observed between females incubated with virgin females and those isolated after mating (Fig. 1C).

Fig. 1.

Male fruit flies employ various sensory signals to attract females during courtship (15). To assess the role of the visual signal in MIES, we examined MIES under dim red lighting conditions and observed that limited illumination had a marginal impact on MIES (Fig. 1D). Next, we examined MIES in post-mating females incubated with decapitated CS males. These males could serve as a source for olfactory or gustatory signals, but not for auditory or visual signals. Once again, we observed no reduction in MIES (Fig. 1E). This strongly suggests that olfactory or gustatory cues are the key signals responsible for MIES. This is further supported by the observation that females with a deficiency in the odorant receptor coreceptor (Orco¹) exhibited no MIES (Fig. 1F). Thus, it is highly likely that male odorant(s), especially those detected by olfactory receptors (Or), are inducing MIES.

MIES is dependent on Or47b receptor and Or47b-expressing ORNs

In fruit fly antenna, the trichoid sensilla and their associated olfactory receptor neurons are known to detect sex pheromones (54). To explore the contribution of ORNs located in the trichoid sensilla to MIES, we silenced 11 different ORN groups found in the trichoid and intermediate sensilla (55, 56) by expressing either the active or inactive form of Tetanus toxin light chain (TNT) (57). Our findings revealed that silencing ORNs expressing Or13a, Or19a, Or23a, Or47b, Or65c, Or67d or Or88a significantly impacted MIES (Fig. S1A).

We then focused on analyzing Or47b-positive ORNs (Fig. 2A), which, unlike the others, exhibited nearly complete abolition of MIES when silenced. Activation of these neurons with the thermogenetic activator dTRPA1 (58) resulted in a significant EHP shortening, even in the absence of male exposure (Fig. 2B). Subsequently, we addressed whether restoring Orco expression in Or47b ORNs in Orco-deficient females would restore MIES. Our results confirmed that indeed it did (Fig. 2C). To ascertain the necessity of the Or47b receptor gene for MIES, we examined Or47b-deficient females (Or47b²/Or47b³) and observed a complete absence of MIES, whereas heterozygous controls exhibited normal MIES (Fig. 2D). Furthermore, the reintroduction of Or47b expression in Or47b ORNs of Or47b-deficient females nearly completely restored MIES (Fig. 2E). With strength of these observations, we concluded that MIES hinges on the Or47b receptor gene and Or47b-expressing ORNs.

Fig. 2.

2-Methyltetracosane (2MC) induces MIES through Or47b and Or47b ORNs

Previous studies have shown that methyl laurate (ML) and palmitoleic acid (PA) can activate Or47b ORNs only in the presence of a functional Or47b gene (30, 31). However, in our investigation, none of these odorant ligands for the Or47b receptor induced significant EHP shortening, even up to 1440 ng (Fig. S2). This prompted us to seek a new pheromone capable of activating Or47b ORNs and therefore shortening EHP.

Oenocytes are known to produce a significant portion, albeit not all, of cuticular hydrocarbons or pheromones. To ascertain whether the male pheromone responsible for MIES is produced by oenocytes, we conducted experiments to determine whether incubation with females possessing masculinized oenocytes would lead to EHP shortening (Fig. S3A). Indeed, our findings confirmed that females with male oenocytes significantly shortens EHP, strongly suggesting that male oenocytes serve as a source for the MIES pheromone. Unexpectedly, however, incubation with males possessing feminized oenocytes also resulted in significant EHP shortening (Fig. S3A). This raises the possibility that oenocytes may not be the sole source of the MIES pheromone, implying the involvement of more than one pheromone, for instance one from oenocytes and another from an alternate source, in MIES.

The genus Drosophila exhibits distinct CHC profiles, with certain CHC components shared among closely related species (59). We found that incubation with males from other closely related species, such as D. similans, D. sechellia, and D. erecta, also induced EHP shortening, whereas incubation with D. yakuba males did not (Fig. S3B). On a search for a male-specific pheromone present in D. melanogaster, D. similans, D. sechellia, and D. erecta, but not in D. yakuba (30, 60), we identified 2MC, which shortens EHP within a physiologically relevant and narrow concentration range (Fig. 3A; Fig. S4). Moreover, EHP shortening induced by 2MC was not observed in females deficient in Orco or Or47b (Fig. 3B, 3C), but it was restored when Orco expression is reinstated in Or47b ORNs in Orco-deficient mutants (Fig. 3D). 2MC was found mainly in males, but not in virgin females (30). Our behavioral observations strongly indicate that 2MC acts as an odor ligand for Or47b and shortens EHP through this receptor.

Fig. 3.

7-Tricosene (7-T) shortens EHP through ppk23 neurons

In contrast to incubation with virgin females, incubation with mated females resulted in a significant shortening in EHP (Fig. 4A). Unlike virgin females, mated females carry male pheromones, including 7-T and cVA, transferred during copulation (53). This raised the possibility that that these male pheromones might also induce EHP shortening. Indeed, our experiments revealed that incubation with a piece of filter paper infused with 150 or 375 ng of 7-T significantly shortened EHP. Conversely, incubation with cVA and 7-pentacosene, a related CHC, did not produce the same effect (Fig. 4B, 4C; Fig. S5A-5B). The concentrations of 7-T capable of inducing EHP shortening appear to be physiologically relevant. 7-T was found in quantities of 432 ng in males (61), 25 ng in virgin females, and 150 ng in mated females (53).

Fig. 4.

Although the receptors for 7-T remain unknown, ppk23-expressing tarsal neurons have been shown to sense these compounds and regulate sexual behaviors in males and females (23, 62, 63). Subsequently, we silenced ppk23 neurons, and as a result, MIES is almost completely abolished, underscoring the pivotal role of 7-T in MIES (Fig. 4D). However, DEG/ENac channel genes expressed in ppk23 neurons, including ppk23 and ppk29, were found to be dispensable for MIES (Fig. S5C-5E). This aligns with previous observations that neither ppk23 deficiency nor ppk28 deficiency replicates the sexual behavioral defects caused by silencing ppk23 neurons (64).

A pC1 neuron subset regulates EHP and MIES

The neuropeptide Dh44 determines timing of sperm ejection or EHP (40). The same study found that Dh44 receptor neurons involved in EHP regulation also express a sexually dimorphic transcriptional factor gene doublesex (dsx). A recent study has revealed that pC1 neurons, a specific subgroup of dsx-expressing central neurons in the female brain, indeed express Dh44 receptors (47). With these findings, we set out to investigate the roles of pC1 neurons in the regulation of EHP and MIES. The pC1 neurons comprise five distinct subtypes. Among these, the pC1a, b, and c subtypes have been associated with mating receptivity (47, 48), while the remaining pC1d and e subtypes have been linked to female aggression (48, 52). To investigate the impact of these subtypes on EHP, we employed GtACR1, an anion channel activated by blue light in the presence of all-trans-retinal (ATR), to silenced specific pC1 subtypes immediately after mating. Our experiments revealed that silencing the pC1 subset comprising pC1a, b and c subtype with GtACR1 led to an increase in EHP (Fig. 5A), whereas silencing the pC1d and e subtypes had a limited effect on EHP (Fig. 5B). Furthermore, we dissected the roles of pC1b and c along with pC1a separately.

Fig. 5.

We generated a subtype-specific split-Gal4 for pC1a and found that, as expected, silencing pC1a using this split-Gal4 nearly completely suppressed mating receptivity (Fig. S6). However, the silencing pC1a alone did not result in increased EHP, suggesting a marginal role of the pC1a subtype in EHP regulation (Fig. 5C). In contrast, concomitant silencing of both pC1b and pC1c significantly increased EHP by 56 + 6.9 minutes (Fig. 5D). Currently, we lack the genetic tools required to further distinguish the roles of pC1b and pC1c subtypes individually.

Our recent research has uncovered that pC1 neurons exhibit elevated cAMP activity during sexual maturation, with this increase in cAMP being closely linked to heightened excitability of the pC1 neurons (47). The same study also revealed that a mating signal (i.e., sex peptide in the male seminal fluid) reduces cAMP activity in pC1 neurons. Thus, we hypothesized that male odorants responsible for inducing MIES, such as 2MC or 7-T, would elevate cAMP activity in pC1b and pC1c neurons in freshly mated females. This, in turn, would lead to increased excitability of pC1 neurons and, as a consequence, a reduction in the EHP. To monitor the cAMP activity in these neurons, we prepared females that selectively produce the CRE-Luciferase reporter gene in pC1b and pC1c neurons. Indeed, when exposed to 2MC or 7-T, pC1b and pC1c neurons exhibited a significant increase in CRE-luciferase activity, indicating that these neurons produce higher levels of cAMP in response to these odorants (Fig. 5E). Notably, the CRE-Luciferase activity appeared to reach its peak at specific odorant concentrations that induced a significant shortening of EHP (Fig. S7).

In contrast, when examining other pC1 subsets, such as pC1a, and pC1d and e, we detected no sign of increased CRE-Luciferase reporter activity upon exposure to 2MC or 7-T treatment (Fig. 5E). It is worth noting that CRE-luciferase reporter activity in the pC1a neurons appears to be dependent on the mating status, as it reaches levels similar to those of pC1b and pC1c neurons in virgin females (Fig. S8). This observation aligns well with connectome data, which indicates that SAG neurons, responsible for relaying SP-dependent mating signals, primarily establishes synapses into pC1a subtype and to a much lesser extent into other pC1 subtypes (49).

Having shown that MIES-inducing male odorants, 2MC or 7-T, increase cAMP activity in pC1b and pC1c neurons from mated females, we next asked whether this induced elevation of cAMP levels in pC1b and pC1c would shorten EHP, leading to MIES. We employed the photoactivatable adenylate cyclase (PhotoAC), which augments cellular cAMP levels upon exposure to light. Indeed, the induced elevation of cAMP activity in pC1b and pC1c significantly shortened EHP, whereas the same treatment applied to pC1a or pC1d and pC1e had no such effect (Fig. 5F). This further underscores the pivotal role of pC1b and pC1c in EHP regulation.

Next, we asked whether the expression of Dh44R1 and Dh44R2, GPCRs that increase cellular cAMP in response to their ligand Dh44, in pC1b and pC1c neurons is necessary for MIES. However, double knockdown of Dh44R1 and Dh44R2 in pC1 neurons seemed to have a limited impact on MIES (Fig. S9). This suggests that Dh44R signaling in pC1 neurons is not essential for the regulation of EHP or MIES, opening up the possibility that other GPCRs may be involved in up-regulating cAMP levels in pC1 neurons in response to 2MC or 7-T.

Lastly, we investigated how the increased cAMP activity affects physiological activity of pC1 neurons. pC1 neurons from virgin females exhibit robust Ca²⁺ transients in response to male courtship cues, such as the male pheromone cVA and courtship pulse song (44). In contrast, those from mated females display significantly diminished Ca²⁺ transients (47). Shortly after mating, a decrease in pC1 responsiveness to cVA was observed. However, immediately following the activation of PhotoAC in pC1 neurons, pC1 neurons in freshly mated females became more excitable and exhibited stronger Ca²⁺ transients in response to cVA (Fig. 5G). It is important to note that this PhotoAC-induced increase in pC1 excitability is transient and diminishes rapidly within 10 minutes (Fig. 5G). Together, these findings suggest that the increased cAMP levels in pC1 neurons would not only promote MIES but also facilitate re-mating in post-mating females, which typically engage in additional mating at a low frequency. To test this hypothesis, we examined the re-mating frequency of freshly mated females paired with naive males while inducing a cAMP increase in pC1 neurons. As expected, PhotoAC activation in pC1b and c neurons substantially increased the re-mating rate compared to the control group (Fig. 5H). Therefore, we concluded that male odorants, which stimulate cAMP elevation in pC1 neurons. expedite the removal of mating plug, consequently leading to increased instances of re-mating.

Discussion

Males employ a diverse range of strategies to enhance their reproductive fitness. One such strategy involves the formation of a ‘mating plug’, a mechanism that prevents females from engaging in further mating and consequently increases fertilization success rates (65–67). As a means of intra-sexual competition, rival males often promote the removal or precocious expulsion of the mating plug. This behavior is driven by the fact that polyandrous females often eliminate the mating plug to engage in additional mating with males possessing superior traits or higher social status than their previous partners (37, 68). In the dunnock (Prunella modularis), a small European passerine bird, the male often engages in cloacal pecking of copulated females, inducing the expulsion of prior mate’s sperm and mating plug, thereby increasing their chance of successful mating (39). In this study, we discovered that in D. melanogaster, when kept with actively courting males, freshly mated females exhibit an earlier removal of the mating plugs or a shorter EHP. This behavior is primarily induced by the stimulation of females via male sex pheromones. Furthermore, our study has uncovered the conserved neural circuitry that processes male courtship cues and governs mating decisions play an important role in regulating this behavior. By delving into the molecular and neuronal mechanisms underlying MIES, our study provides valuable insights into the broader aspect of behaviors induced by social sexual context changes.

Our findings highlight the involvement of the Or47b receptor and Or47b ORNs in MIES. These OR and ORNs have been linked to a range of social and sexual behaviors in both males and female fruit flies (12, 26–31). Methyl laurate and trans-palmitoleic acid are odor ligands for Or47b, which explain many of these functions particularly in males (30, 31). In this study, we provide compelling evidence that 2MC induces EHP shortening via both Or47b receptor and Or47b ORNs, suggesting that 2MC functions as an odor ligand for Or47b. Notably, a gas chromatography–mass spectrometry (GC-MS) analysis of cuticular hydrocarbons of 4-day-old wild type D. melanogaster indicated the presence of 2MC solely in males, excluding females (30). Surprisingly, however, unlike 2MC, neither methyl laurate nor trans-palmitoleic acid influences EHP. The reason of this paradoxical results remains unclear. Plausible interpretation is that the EHP shortening induced by 2MC may require not only Or47b but also other as-yet-unidentified ORs.

With the establishment of a behavioral and cellular assessment for 2MC activity, the search for additional odorant receptors responsive to 2MC is now feasible. An other important avenue for further research is whether 2MC can also elicit behaviors that were previously associated with methyl laurate or transpalmitoleic acid, such as the promotion of male copulation and courtship (30, 31).

We observed that both 2MC and 7-T exhibit both cellular and behavioral activity within a specific concentration range (Fig. S4, S5, S7). This observation is of particularly interest, given the multitude of environmental and biological factors that influence the levels of 2MC and 7-T, potentially affecting the capacity of males to induce MIES. For instance, exposure to low temperatures during development has been linked to increased production of both 2MC and 7-T (69). Similarly, the mutation of the desiccation stress gene CG9186, which encodes a protein associated with lipid droplet, has been found to impact 2MC levels (70). Furthermore, 2MC levels rise with the aging of males (71). Thus, we propose that the levels of 2MC and possibly 7-T may serve as indicators of male’s age and their resilience against environmental stresses in a complex and non-linear manner.

In mated females, treatment with 2MC or 7-T increases cAMP activity in pC1b,c neurons but not in pC1a neurons. In contrast, pC1a neurons in virgin females are fully responsive to both male pheromones, showing a cAMP activity that is similar to that of pC1b,c neurons (Fig. S8). The absence of cAMP activity in pC1a neurons in mated females likely results from the mating signal (i.e., sex peptide) silencing pC1a neurons. Connectome and electrophysiology data support this interpretation, as SAG neurons, which relay sex peptide signals, exhibit the strongest synaptic connection with pC1a subtype among five pC1 subtypes (49). However, SAG neuron activity may also influence pC1c neurons, as they also have substantial synaptic connections with pC1c neurons in hemibrain (72). Notably, induced activation of SAG neurons significantly shortens EHP (Fig. S10), suggesting they regulate EHP through pC1c neurons.

We found that increased cAMP levels causes pC1b,c neurons in mated females, which typically do not respond to male courtship cues like cVA and pulse song, to become responsive, showing strong Ca²⁺ transients. Since pC1b,c neurons play a role in generating sexual drive and increasing female receptivity to male courtship, the 2MC or 7-T-induced increases in cAMP likely govern the removement of the mating plug and engagement in further mating of mated females. This observation well aligns with the previous report that mating reduces sensitivity of Or47b ORNs, which we find responsive to 2MC, leading to the increased preference for pheromone-rich males after mating (29). Moreover, the finding that 2MC and 7-T induce cAMP activity in pC1b,c neurons in virgin females suggests that virgin females may also use 2MC and 7-T as odorant cues to assess male quality during their first mating. Indeed, females seem to evaluate male quality with the amount of 7-T, as increased 7-T promotes mating receptivity and shortens mating latency (20).

Physiological factors like the nutritional status of females before mating and the nutritional status of their mating partners have been shown to influence EHP (73), and therefore potentially MIES. Hence, it is highly probable that MIES is regulated by additional central neurons such as Dh44-PI neurons that regulate these processes (40). However, it remains unclear whether and how Dh44-PI neurons and pC1 neurons interact to modulate EHP and MIES. The observation that double knockdown of Dh44R1 and Dh44R2 has only a marginal effect on MIES suggests that Dh44-PI neurons may also function independently of pC1 neurons, suggesting the possibility that multiple independent central circuits may contribute to the production of MIES.

Our initial screening of ORNs responsible for MIES revealed the involvement of Or47b ORNs, as well as several other ORNs. In addition to 2MC, which operates through Or47b expressing ORNs, our findings indicate that 7-T and ppk23 neurons capable of detecting 7-T also play a role in MIES induction. In D. melanogaster and other related species, food odors typically serve as long-range signals that attract both males and females (74, 75), implying that certain food odors might also influence EHP (42). The involvement of multiple odorants and ORNs in EHP regulation implicates that pC1 neurons can process various cues, not restricted to those associated with mating behaviors, including food odors. Future studies will explore the full spectrum of odorants processed by pC1 neurons in EHP regulation.

In conclusion we have identified a circuit that, via the detection of a novel male pheromone, potentially signals male quality and governs the female decision to remove the mating plug of its last mate and mate again.

Materials and Methods

Fly care

Flies were cultured on a standard medium composed of dextrose, cornmeal, and yeast, under room temperature in a 12hr : 12hr light:dark cycle (40, 73). Behavioral assays were performed at 25 °C, except for thermal activation experiment with dTRPA1. Virgin males and females were collected immediately after eclosion. Males were individually aged for 4-6 days, while females were aged in groups of 15–20. For EHP and mating assays, females were aged for 3-4 days. Assays were performed at Zeitgeber time (ZT) 3:00– 11:00, and were repeated at least on three separate days.

Fly stocks

The following stocks are from the Bloomington Drosophila Stock Center (BDSC), the Vienna Drosophila Resource Center (VDRC): Canton S (CS) (RRID: BDSC_64349), w¹¹¹⁸ (VDRC #60000), R71G01 (pC1-Gal4) (RRID: BDSC_39599), Orco1 (RRID: BDSC_23129), Or13a-Gal4 (RRID: BDSC_9946), Or19a-Gal4 (RRID: BDSC_9948), Or23a-Gal4 (RRID: BDSC_9955), Or43a-Gal4 (RRID: BDSC_9974), Or47b-Gal4 (RRID: BDSC_9983), Or47b-Gal4 (RRID: BDSC_9984), Or65a-Gal4 (RRID: BDSC_9993), Or65b-Gal4 (RRID: BDSC_23901), Or65c-Gal4 (RRID: BDSC_23903), Or67d-Gal4 (RRID: BDSC_9998), Or83c-Gal4 (RRID: BDSC_23131), Or88a-Gal4 (RRID: BDSC_23137), UAS-Or47b (RRID: BDSC_76045), Or47b2/2 (RRID: BDSC_51306), Or47b3/3 (RRID: BDSC_51307), UAS-TNT active (RRID: BDSC_28837), UAS-TNT inactive (RRID: BDSC_28839), UAS-dTRPA1 (RRID: BDSC_26263), UAS-CsChrimson (RRID: BDSC_55135), UAS-GCaMP6m (RRID: BDSC_42748), R52G04-AD (RRID: BDSC_71085), SAG-Gal4 (VT50405) (RRID:Flybase_FBst0489354, VDRC #200652), UAS-Dh44R1-RNAi (RRID:Flybase_FBst0482273, VDRC #110708), UAS-Dh44R2-RNAi (RRID:Flybase_FBst0465025, VDRC #43314), UAS-Dicer2 (VDRC #60007). The following stocks are reported previously: PromE(800)-Gal4 (59), UAS-FLP, CRE-F-luc (76), LexAop-FLP (77), UAS-CsChrimson (78), UAS-GtACR1 (79), UAS-PhotoAC (PACα) (80), pC1-A (48), pC1-S (48), Dh44-pC1-Gal4 (47), ppk23-Gal4, ppk23-, ppk28-, ppk29-(62), and Orco-Gal4, UAS-EGFP-Orco (81). pC1a-split-Gal4 is generated by combining R52G04-AD (RRID: BDSC_71085) and dsx-DBD (49). Drosophila subgroups species are obtained from the EHIME-Fly Drosophila stock center. To enhance knock-down efficiency, RNAi experiments were performed using flies carrying UAS-Dicer2 (VDRC #60007).

Chemical information

All trans-retinal (Cat# R2500), methyl laurate (Cat# W271500), and Triton™ X-100 (Cat# X100) were obtained from Sigma-Aldrich (St. Louis, MO, USA). The following chemicals are from the Cayman chemical (Ann Arbor, MI, USA): 7(Z)-Tricosene (CAS No. 52078-42-9, Cat# 9000313), 7(Z)-Pentacosene (CAS No. 63623-49-4, Cat# 9000530), trans-palmitoleic acid (CAS No. 10030-73-6, Cat# 9001798), 11-cis-vaccenyl acetate (cVA) dissolved in EtOH (CAS No. 6186-98-7, Cat# 10010101). 2-methyltetracosane (>98%, purity) was custom-synthesized by KIP (Daejeon, Korea).

Behavior assays

For mating behavior assays, we followed the procedures described previously (82). Individual virgin females and naive CS males were paired in chambers with a diameter of 10 mm and were recorded using a digital camcorder (SONY, HDR-CX405 or Xiaomi, Redmi Note 10) for either 30 min or 1 hour for the mating assay and 6 hours for the re-mating assay. In the re-mating assay, females that completed their initial mating within 30 min were subsequently paired with naive CS males.

To measure EHP, which is defined as the time elapsed between the end of copulation and sperm ejection, we used the following procedure: Virgin females were paired individually with CS males in 10-mm diameter chambers. Following copulation, females were transferred to new chambers, either with or without a CS male or a piece of filter paper treated with pheromones, and their behavior was recorded using a digital camcorder (SONY, HDR-CX405). Typically, females that completed copulation within 30 minutes were used for analysis. To present the pheromone, females were individually kept in 10-mm diameter chambers containing a piece of Whatman filter paper (2 mm x 2 mm) that had been treated with 0.5 μl of the pheromone solution and left to air dry for 1 minute. For thermal activation experiments, females were incubated at the specified temperatures immediately after copulation ended. For light-activation experiments, a custom-made light activation setup utilizing a ring of 104 multi-channel LED lights (NeoPixel, Cat# WS2812; Red light, 620-625 nm, 390-420 mcd; Green light, 522-525 nm, 660-720 mcd; Blue light, 465-467 nm, 180-200 mcd) was employed. Females were individually placed in 10-mm diameter chambers, and the chamber was illuminated with light at an intensity of 1100 lux across the chamber, as measured by a lightmeter HS1010, during the assay. Flies used in these experiments were prepared by culturing them in food containing vehicle (EtOH) or 1 mM all-trans-retinal (ATR) immediately after eclosion. They were kept in complete darkness for 3-4 days until the assay was conducted.

Calcium imaging

We followed the procedures described previously (47, 83). Following copulation, freshly mated female fly was temporally immobilized using ice anesthesia, and its head was attached to a custom made thin metal plate with a 1mm diameter hole using photo-curable UV glue (ThreeBond, A16A01). An opening in the fly’s head was generated using a syringe needle under saline (108 mM NaCl, 5 mM KCl, 2 mM CaCl₂, 8.2 mM MgCl₂, 4 mM NaHCO₃, 1 mM NaH₂PO₄, 5 mM trehalose, 10 mM sucrose, 5 mM HEPES pH 7.5). Imaging was performed using Zeiss Axio Examiner A1 microscope equipped with an electron multiplying CCD camera (Andor Technology, Luca^EM R 604M) and a LED light-source (CoolLED, Precis Excite). Imaging analysis was used Metamorph software (Molecular Devices, RRID:SCR_002368). To deliver the male pheromone using an airflow, we used the Syntech Stimulus Controller (Type CS-55). 2 μl of pheromone solution was applied to a piece of Whatman filter paper (2 mm x 1 mm), which was then inserted into a glass Pasteur pipette after solvent evaporation.

Luciferase assay

We followed the procedures described previously (47, 76). 3-day-old virgin females or freshly mated females were used for assay. A group of three fly heads, kept at -80°C, was homogenized using cold homogenization buffer (15 mM HEPES, 10 mM KCl, 5 mM MgCl₂, 0.1 mM EDTA, 0.5 mM EGTA).

Luciferase activity was measured by using beetle Luciferin, potassium Salt (Promega, Cat# E1603) and a microplate Luminometer (Berthold technologies, Centro XS³ LB 960), following manufacturer’s instructions. To present the odorant, flies were placed in 10-mm diameter chambers containing a piece of Whatman filter paper (4 mm x 6 mm) that had been treated with 1 μl of the pheromone solution and left to air dry for 1 minute.

Immunohistochemistry

3–5-day-old virgin female flies were dissected in phosphate-buffered saline (PBS) and fixed for 30 minutes at room temperature in 4% paraformaldehyde in PBS. Following fixation, the brains were thoroughly washed in PBST (0.1% Triton™ X-100 in PBS) and then blocked using 5% normal goat serum in PBST. After blocking, the brains were incubated with the primary antibody in PBST for 48 hours at 4 °C, washed by PBST and then incubated with the secondary antibody in PBST for 24 hours at 4 °C. The samples were washed three times with PBST and once with PBS before being mounted in Vectashield (Vector Laboratories, Cat# H-1000). Antibodies used were rabbit anti-GFP (1:1000; Thermo Fisher Scientific, Cat# A-11122, RRID:AB_221569), mouse anti-nc82 (1:50; Developmental Studies Hybridoma Bank, Cat# Nc82; RRID: AB_2314866), Alexa 488-conjugated goat anti-rabbit (1:1000; Thermo Fisher Scientific, Cat# A-11008, RRID:AB_143165), Alexa 568-conjugated goat anti-mouse (1:1000; Thermo Fisher Scientific, Cat# A-11004, RRID:AB_2534072). Brain images were acquired using Zeiss LSM 700/Axiovert 200M (Zeiss) and processed using Fiji (https://imagej.net/software/fiji/downloads, RRID:SCR_002285)

Color-depth MIP based anatomy analysis

A stack of confocal images from pC1a-split-Gal4>UAS-myr-EGFP adult female brains stained with anti-GFP and anti-nc82 were used. Images were registered onto the JRC2018 unisex brain template (84). using the Computational Morphometry Toolkit (CMTK, https://github.com/jefferis/fiji-cmtk-gui). Color depth MIP masks of pC1a-split-Gal4 neurons and pC1a (ID, 5813046951) in Hemibrain (72) (Fig. S6A) were generated using the ColorMIP_Mask_Search plugin (85) for Fiji (https://github.com/JaneliaSciComp/ColorMIP_Mask_Search) and NeuronBridge (86) (https://neuronbridge.janelia.org/). Similarity score and rank were calculated using NeuronBridge.

Statistical analysis

Statistical analysis was conducted using GraphPad Prism 9 (Graphpad, RRID:SCR_002798), with specific details on each statistical method provided in the figure legends.

Acknowledgements

We thank S. Kang, J-H. Yoon, B. Lee for excellent technical assistance. Fly stocks were obtained from the Bloomington Drosophila Stock Center (NIH P40OD018537), the Vienna Drosophila Resource Center (VDRC), the Kyoto Stock Center, EHIME-Fly Drosophila species stock center, KYORIN-Fly Drosophila species stock center, and the Korea Drosophila Resource Center (NRF-2022M3H9A1085169). This work was supported by National Research Foundation of Korea grants NRF-2022R1A2C3008091 (Y-J.K.), 2022M3E5E8081194 (Y-J.K.), NRF-2019R1A4A1029724 (Y-J.K.), GIST Research Institute (GRI) GIST-MIT research collaboration grant funded by the GIST in 2022 (Y-J.K.), 2023 GIST grant (Y-J.K.), 2017R1A6A3A11027866 (D-H.K.), NRF-2021R1I1A1A01060304 (D-H.K.), 2022 AI-based GIST Research Scientist Project (D-H.K.) and 2023 AI-based GIST Research Scientist Project (D-H.K.).

Author contributions

Conceptualization: M.Y., Y-J.K., Methodology: K-M.L., M.K., B.S.H., Investigation: M.Y., D-H.K., T.S.H., E.P., Visualization: M.Y., Supervision: Y-J.K., Writing—original draft: M.Y., Y-J.K., Writing—review & editing: M.Y., M.K., B.S.H., Y-J.K.

Competing interests

The authors declare that they have no competing interest.