The Drosophila TRPA channel, Painless, regulates sexual receptivity in virgin females

Authors


T. Sakai, Department of Biological Sciences, Tokyo Metropolitan University, Hachioji, Tokyo 192-0397, Japan. E-mail:sakai-takaomi@tmu.ac.jp

Abstract

Transient receptor potential (TRP) channels play crucial roles in sensory perception. Expression of the Drosophila painless (pain) gene, a homolog of the mammalian TRPA1/ANKTM1 gene, in the peripheral nervous system is required for avoidance behavior of noxious heat or wasabi. In this study, we report a novel role of the Pain TRP channel expressed in the nervous system in the sexual receptivity in Drosophila virgin females. Compared with wild-type females, pain mutant females copulated with wild-type males significantly earlier. Wild-type males showed comparable courtship latency and courtship index toward wild-type and pain mutant females. Therefore, the early copulation observed in wild-type male and pain mutant female pairs is the result of enhanced sexual receptivity in pain mutant females. Involvement of pain in enhanced female sexual receptivity was confirmed by rescue experiments in which expression of a pain transgene in a pain mutant background restored the female sexual receptivity to the wild-type level. Targeted expression of pain RNA interference (RNAi) in putative cholinergic or GABAergic neurons phenocopied the mutant phenotype of pain females. However, target expression of pain RNAi in dopaminergic neurons did not affect female sexual receptivity. In addition, conditional suppression of neurotransmission in putative GABAergic neurons resulted in a similar enhanced sexual receptivity. Our results suggest that Pain TRP channels expressed in cholinergic and/or GABAergic neurons are involved in female sexual receptivity.

Genetic analysis can be used to identify genes required for specific behaviors and can lead to the identification of neural circuits and molecular mechanisms underlying the behavior. The fruit fly, Drosophila melanogaster, is an ideal model organism for investigating the molecular mechanisms or circuits underlying male and female sexual behaviors owing to the availability of vast genetic information and clearly defined behaviors. Genetic studies on D. melanogaster have resulted in the identification of genes regulating male courtship behaviors (Emmons & Lipton 2003; Hall 1994), and sites in the central nervous system (CNS) relevant to male courtship have been mapped (Broughton et al.2004; Demir & Dickson 2005; Kimura et al.2008; Manoli et al.2005; Sakai & Kitamoto 2006). However, brain mechanisms that control sexual behavior in virgin females remain largely unknown.

In Drosophila, female sex pheromones evoke male courtship behavior (Greenspan & Ferveur 2000; Hall 1994; Jallon 1984; Tompkins 1984), and male courtship songs produced by wing vibrations stimulate the sexual receptivity in virgin females (Kyriacou & Hall 1982; Manning 1967; Tomaru et al.2000; von Schilcher 1976). Receptive virgin females decide to accept courting males and copulate with them. In contrast, non-receptive females show a variety of rejection responses (Ewing 1983; Hall 1994; Spieth 1952). Several mutations [e.g. paralytic, apterous, spinster and icebox (ibx)] result in the low sexual receptivity in virgin females of D. melanogaster (Hall 1994; Kerr et al.1997; Ringo et al.1991; Suzuki et al.1997; Tompkins et al.1982), whereas the induction of a repressor isoform of Drosophila CREB, dCREB2-b, enhances the sexual receptivity in virgin females (Sakai & Kidokoro 2002). Thus, the sexual receptivity in virgin females can be enhanced or inhibited by various genetic manipulations.

Transient receptor potential (TRP) channels are evolutionarily conserved Ca2+-permeable cation channels that are expressed in a variety of cell types in invertebrates and vertebrates (Pedersen et al.2005; Vriens et al.2004). Cation influx through TRP channels depolarizes the neuronal membrane potential and increases the intracellular Ca2+ concentration (Ramsey et al.2006). The Drosophila Painless (Pain) TRP channel is a homolog of the mammalian wasabi receptor, TRPA1/ANKTM1 (Corey 2003). The pain gene was first found to play an essential role in noxious thermal and mechanical perception in larval sensory neurons (Tracey et al.2003), and Xu et al.(2006) have reported that pain is also required for thermal nociception in adult flies. A recent report indicates that the Pain TRP channel is activated by noxious heat in vitro (Sokabe et al.2008). Furthermore, painGAL4 containing a GAL4-enhancer trap insertion in the first exon of pain drives green fluorescent protein (GFP) reporter expression in adult taste receptor neurons, and pain is required for the behavioral response to wasabi in Drosophila (Al-Anzi et al.2006). In this article, we show that the pain gene plays a critical role in the regulation of sexual receptivity in Drosophila virgin females, showing for the first time its involvement in mating behavior.

Materials and methods

Fly stocks

Drosophila melanogaster wild-type Canton-S (CS), pain mutants (pain1, pain3 and painGAL4), UAS- pain, two UAS- pain RNA interference (RNAi) lines (1A and 2A), several neural GAL4 lines [choline acetyltransferase (Cha)-GAL4-9A and Cha-GAL4-19B; tyrosine hydroxylase (TH)-GAL4; dopa decarboxylase (Ddc)-GAL4; glutamate decarboxylase 1 (Gad)-GAL4-2 and Gad-GAL4-5], Cha-GAL80 and Gad-GAL80 were raised on glucose–yeast–cornmeal medium at 24.5 ± 0.5°C in a 12-h light/dark cycle. All pain mutants, Cha-GAL4-19B, TH-GAL4, Ddc-GAL4 and UAS-pain, were outcrossed for five or six generations to white flies with the CS genetic background.

For Gad-GAL4 and Gad-GAL80 transformants, the 4.5 kb BamHI/BamHI Gad1 5-flanking DNA was fused to GAL4 (Brand & Perrimon 1993) and GAL80 (Lee & Luo 1999) cDNAs, respectively. The resultant fusion genes were inserted to the pCaSpeR2 vector. For Cha-GAL80, the lacZ portion of pCaSpeR-7.4 kb-lacZ (Kitamoto et al.1992) was replaced with a GAL80 cDNA. The UAS-pain RNAi construct was created using a vector developed by Lee and Carthew (2003) with a transgene containing inverted repeats of partial pain cDNA (nucleotide positions 2280–2780) (Tracey et al.2003). Transformed fly lines were established by injecting these DNA constructs in white mutant embryos using standard procedure (Spradling & Rubin 1982). When the genetic background of the white mutants used for injection is not known, transformants were backcrossed to white flies with the CS genetic background for at least six generations.

Virgin males or females were collected without anesthesia within 6 h of eclosion and maintained in vials until experiments. All the experiments except for the temperature shift experiments were carried out during daytime between Zeitgeber time (ZT) 0 and ZT5 at 24.5 ± 0.5°C in 50–60% relative humidity.

Observation of mating behavior

A pair of male and female flies was placed in an acrylic plastic observation chamber (15-mm diameter × 3-mm depth) using a manual aspirator. We observed mating behaviors of 10 pairs at the same time for 20 min, and at least 40 pairs were observed for each set of experiments. We measured the mating success rate, defined as the percentage of pairs that copulated during the 20-min period after placing male and female flies together in the observation chamber. Furthermore, we calculated the mean courtship latency and mean time to copulation (TC). The courtship latency is defined as the period between the moment the flies are placed in the chamber and the first courtship, and TC is defined as the time for male/female pairs to copulate after the male flies initiate courtship behavior. CS males were used in all the observations. All the flies used in the experiment were 4–6 days old.

Analysis of male courtship

A pair of male and female flies was placed in an observation chamber as described above. We then observed male courtship behavior for 10 min or until the moment of copulation and calculated the courtship index (CI). The CI is defined as the percentage of time spent courting in a given observation period. For each group, 15–20 pairs were observed. CS females, pain mutant females and CS males were used in the experiment. All the flies used in the experiment were 4–6 days old.

Quantification of locomotor activity by video-tracking

Intact single females were allowed to walk freely in an acrylic plastic observation chamber (15-mm diameter × 3-mm depth). Five-day-old virgin females were used. Four individuals were simultaneously videotaped at a capture rate of 30 frames per second for 10 min. We used total distance moved (mm) as an index of locomotor activity. Traces were generated and total distance moved (mm) was calculated using Move-tr/2D 7.0 (Library Co., Tokyo, Japan).

Temperature shift experiments

UAS-shits1/Gad-GAL4-2, UAS- shits1/+ and +/Gad-GAL4-2 females were grown at 23.0 ± 0.5°C (permissive temperature), and collected virgin females were also kept at the permissive temperature until 30–60 min before observation of mating behavior, at which point they were shifted to 30°C (restrictive temperature) until the end of the experiments. As a control, observation of mating behavior was carried out at 23.0 ± 0.5°C. Five-day-old virgin females of these genetic constructs and 5-day-old CS males were used in the experiments.

Microscopy

GFP fluorescence was observed using a fluorescence microscope (Leica MZ 16F; Leica Microsystems K.K. (Japan), Tokyo, Japan) and a confocal microscope (Nikon D-ECLIPSE C1 Si; Nikon, Tokyo, Japan). For confocal microscopy, Z sections were collected at 1-μm intervals and processed to construct projections through an extended depth of focus.

Statistical analysis

In most cases, the courtship latency, TC and CI were not distributed normally. Thus, we carried out a log transformation of courtship latency and TC, and an arcsine transformation of CI. However, the transformed values did not show normal distribution. Thus, we used non-parametric tests (Kruskall–Wallis test and Mann–Whitney U-test). Because total walking distances were distributed normally and homoscedasticity was evident, they were analyzed by the one-way analysis of variance (anova). We used computer software (SPSS 10.0J Base System for Windows) in these tests (SPSS Japan inc., Shibuya-ku, Tokyo, Japan). The G test with Williams's correction was used for comparisons of mating success rates (Sokal & Rohlf 1995).

Results

pain mutant females copulate earlier than wild-type females

We observed mating behavior of wild-type CS males and painless (pain) mutant females. We used mutant alleles pain1, pain3 and painGAL4. Both pain1 and pain3 have a P-element insertion in the 5 untranslated exon of the pain gene. painGAL4, which contains a GAL4-enhancer trap insertion, was generated from pain1 by P-element replacement (Al-Anzi et al.2006; Tracey et al.2003). The mating success rate, defined as the percentage of pairs that copulated, was determined every 1 min (Fig. 1). The mating success rates of females homozygous for pain1, pain3 and painGAL4 were higher than those of CS females during the entire observation period (Fig. 1). An analysis of the 20-min data showed significant differences between the accumulated mating success rates of pain mutant females and those of CS females (Fig. 1).

Figure 1.

Cumulative mating success rate (%) in pain mutant and wild-type females.pain1 (crosses), pain3 (squares), painGAL4 (triangles), and CS (circles) females and CS males were used. Observation period was 20 min. Forty pairs were observed for each genotype. *P < 0.05; **P < 0.01.

Next, we measured TC defined as the time for male/female pairs to copulate after males initiate courtship behavior. As shown in Fig. 2a, TC of pain mutant females was significantly shorter than that of CS females. In contrast to the shortened TC in pain mutant females, no significant difference was detected in the courtship latency (Fig. 2b) and CI (Fig. 2c) between CS and pain mutant females. These results show that pain mutant females do not elicit increased male courtship behavior than CS females, and that the rapid copulation of virgin pain mutant females is because of their enhanced sexual receptivity.

Figure 2.

Time to copulation, courtship latency, courtship index, and locomotion in pain mutant and wild-type females. (a) Time to copulation (sec), (b) male courtship latency (sec), (c) male courtship index (%), (d) total walking distance (mm). In all experiments, pain1, pain3, painGAL4, and CS females and CS males were used. In each box plot, the box encompasses the interquartile range, a line is drawn at median and error bars correspond to the 10th and 90th percentiles. Each square within a box is the mean. n, sample size; ***P < 0.001.

In D. melanogaster, receptive females spent less time in locomotion before mating, and this female immobility is important for a successful copulation attempt (Markow & Hanson 1981). To examine whether the enhanced sexual receptivity in pain mutant females results from their reduced general locomotion, we measured the total walking distance within 10 min, as an indicator of locomotion, using the video-tracking system. No significant difference was found between CS and pain mutant females (Fig. 2d), suggesting that the enhanced sexual receptivity in pain mutant females does not simply result from general sluggishness.

pain expression in the pain mutant background restores the female sexual receptivity to the wild-type level

We used the GAL4/UAS system (Brand & Perrimon 1993) to examine whether expression of wild-type pain rescues the enhanced sexual receptivity in pain mutant females. In transgenic flies with painGAL4 on the second chromosome and UAS-pain on the third chromosome (painGAL4; UAS- pain), wild-type pain should be expressed in a pain mutant background with an expression pattern similar to that of pain in wild-type flies. UAS- pain females without pain mutations were used as a control. Time to copulation of females homozygous for painGAL4; UAS- pain was significantly longer than that of painGAL4 mutant females, and no significant difference was detected between females homozygous for UAS- pain and painGAL4; UAS- pain (Fig. 3a). This result confirmed that the enhanced sexual receptivity in pain mutant females is indeed caused by mutations in the pain gene. In addition, this result indicates that the GAL4-positive neurons in painGAL4 are relevant to female sexual receptivity.

Figure 3.

pain expression in painGAL4-positive neurons is involved in the regulation of female sexual receptivity. (a) Time to copulation (sec) in painGAL4, UAS- pain and painGAL4; UAS- pain females. CS males were used. The observation period was 20 min. n, sample size; NS, not significant; *P < 0.05; **P < 0.01. (b) Time to copulation (seconds) in painGAL4/+, UAS- pain RNAi (2A)/+ and painGAL4/UAS- pain RNAi (2A) females. Canton-S males were used. The observation period was 20 min. n, sample size; NS, not significant; *P < 0.05; **P < 0.01.

Induction of pain RNAi in painGAL4-positive neurons enhances female sexual receptivity

Next, we further explored whether pain expression plays a key role in female sexual receptivity using RNAi to knock down the expression of pain. A UAS- pain RNAi line (2A) was used in combination with the GAL4/UAS binary expression system. In F1 females between painGAL4 and UAS- pain RNAi (painGAL4/UAS- pain RNAi), pain expression should be repressed in GAL4-positive neurons in painGAL4. Females heterozygous for painGAL4 (painGAL4/+) and UAS- pain RNAi (+/UAS- pain RNAi) were used as controls. Time to copulation of painGAL4/UAS- pain RNAi females was significantly shorter than that of control females, whereas no significant difference was detected between control females (painGAL4/+ and +/UAS- pain RNAi) (Fig. 3b). This result indicates that the induction of pain RNAi in GAL4-positive neurons in painGAL4 enhances female sexual receptivity.

Induction of pain RNAi in neuronal subsets enhances sexual receptivity

Several neural GAL4 drivers (Cha-GAL4-9A, Cha-GAL4-19B, TH-GAL4, Ddc-GAL4, Gad-GAL4-2 and Gad-GAL4-5) were used in combination with two UAS-pain RNAi lines (1A and 2A) to investigate the possible involvement of particular neurotransmitter systems in the sexual receptivity controlled by pain. Cha-GAL4, TH-GAL4, Ddc-GAL4 and Gad-GAL4 transgenes are composed of GAL4 cDNA fused to the regulatory region of the neurotransmitter-synthesizing enzymes for acetylcholine (Cha), dopamine (TH), dopamine and serotonin (Ddc), and GABA (Gad), respectively.

We found that the enhanced sexual receptivity in pain mutant females was phenocopied when pain RNAi was expressed in wild-type virgin females using two independent Cha-GAL4 lines, Cha-GAL4-9A and Cha-GAL4-19B (Fig. 4). Time to copulations of 1A/Cha-GAL4-9A, 2A/Cha-GAL4-9A and 1A/Cha-GAL4-19B females were significantly shorter than that of +/Cha-GAL4-9A or +/Cha-GAL4-19B control females (Fig. 4a,b, gray boxes), whereas no significant difference was detected between GAL4 and UAS control females (Fig. 4a,b, white boxes). In contrast to Cha-GAL4, expression of pain RNAi using TH-GAL4 did not affect the female sexual receptivity (Fig. 5a). No significant difference in TC was detected between 1A/Ddc-GAL4 and +/Ddc-GAL4 females (Fig. 5b), suggesting that the induction of pain RNAi driven by Ddc-GAL4 does not affect female sexual receptivity. However, TC of +/Ddc-GAL4 females was significantly shorter than that of 1A/+ females. Ddc-GAL4 line has been backcrossed to white line with CS genetic background six generations. Thus, the increased female receptivity observed in the Ddc-GAL4 line is likely because of the P-element insertion itself. F1 females between Gad-GAL4 and UAS-pain RNAi showed enhanced sexual receptivity (Fig. 6a,b, gray boxes), whereas control females did not (Fig. 6a,b, white boxes). Thus, the induction of pain RNAi driven by Cha-GAL4 or Gad-GAL4 enhanced sexual receptivity.

Figure 4.

Induction of pain RNAi driven by Cha-GAL4 enhances sexual receptivity. Two UAS- pain RNAi lines (1A and 2A), Cha-GAL4-9A (a) and Cha-GAL4-19B (b) were used. The observation period was 20 min. White boxes indicate control females, and gray boxes indicate F1 females between Cha-GAL4 and UAS- pain RNAi. n, sample size; NS, not significant; *P < 0.05; **P < 0.01; ***P < 0.001.

Figure 5.

Induction of pain RNAi driven by TH-GAL4 or Ddc-GAL4 does not affect female sexual receptivity. Two UAS- pain RNAi lines (1A and 2A), TH-GAL4 (a) and Ddc-GAL4 (b) were used. The observation period was 20 min. White boxes indicate control females, and gray boxes indicate F1 females between GAL4 lines and UAS- pain RNAi. n, sample size; NS, not significant; **P < 0.01.

Figure 6.

Induction of pain RNAi driven by Gad-GAL4 enhances sexual receptivity. Two UAS- pain RNAi lines, 1A and 2A, Gad-GAL4-2 (a) and Gad-GAL4-5 (b) were used. The observation period was 20 min. White boxes indicate control females, and gray boxes indicate F1 females between Gad-GAL4 and UAS- pain RNAi. n, sample size; NS, not significant; *P < 0.05; ***P < 0.001.

painGAL4 and Cha-GAL4 drive GFP reporter gene expression in the sensory neurons

Expression of wild-type pain in painGAL4-positive neurons rescued the enhanced sexual receptivity in painGAL4 mutant females (Fig. 3a), and induction of pain RNAi in painGAL4-positive neurons enhanced female sexual receptivity (Fig. 3b), indicating that GAL4-positive neurons in painGAL4 are responsible for the regulation of female sexual receptivity. Because olfactory (Gailey et al.1986; Kurtovic et al.2007; Savarit et al.1999; Tompkins et al.1980) and auditory (Kyriacou & Hall 1982; Manning 1967; Tomaru et al.2000; von Schilcher 1976) inputs are critically involved in female sexual receptivity, we examined whether painGAL4 drives GFP reporter gene expression in sensory neurons of the adult head. We generated transgenic flies (painGAL4 UAS- GFP) with both painGAL4 and UAS- GFP constructs in the second chromosome. In females homozygous for painGAL4 UAS-GFP, GFP signal was detected in the sensory neurons of Johnston's organ (the fly's auditory organ) located within the second antennal segment, the maxillary palp (one of the Drosophila olfactory organs), taste neurons and the reproductive tract (Figs 7a–c and S1a). A similar expression pattern of GFP reporter gene was also observed in females heterozygous for painGAL4 UAS- GFP (Fig. 7d). GFP signal was not detected in the olfactory organs within the third antennal segment in females homozygous and heterozygous for painGAL4 UAS- GFP (Fig. 7a,d) in agreement with Al-Anzi et al.(2006).

Figure 7.

painGAL4 drives GFP reporter expression in the sensory organs. We observed GFP reporter expression using females homozygous for painGAL4 UAS- GFP (a–c), females heterozygous for painGAL4 UAS- GFP (d), Cha-GAL80/painGAL4 UAS- GFP females (e,f) and Gad-GAL80/painGAL4 UAS-GFP females (g–i). (a–i) are confocal microscope images. (a) and (d) show adult head. (b) shows the second segment of antenna. (c) and (h) show the maxillary palp. (e) and (g) show the second and third antennal segments. (f) shows the maxillary palp and taste neurons. (i) shows taste neurons. Arrowheads show the second antennal segment. Asterisks show the third antennal segment. Arrows show the maxillary palp. Triangles show taste neurons.

In F1 females between Cha-GAL4 and UAS- GFP, GFP reporter gene expression was detected in the second and third antennal segments, maxillary palp, taste neurons (Fig. 8a,b) as reported by Salvaterra and Kitamoto (2001) and reproductive tract (Fig. S1b). However, in Gad-GAL4/UAS- GFP females, GFP reporter gene expression was not detected in the second and third antennal segments, maxillary palp, taste neurons in the proboscis (Fig. 8c,d), legs and wings (data not shown), and reproductive tract (Fig. S1c), suggesting that the enhanced sexual receptivity in F1 females between UAS- pain RNAi and Gad-GAL4 does not result in the dysfunction of auditory and/or chemosensory reception.

Figure 8.

GFP reporter expression in Cha-GAL4 and Gad-GAL4 lines. We observed GFP reporter expression using F1 females between GAL4 lines and UAS- GFP. All pictures are confocal microscope images. (a) Cha-GAL4-9A/UAS- GFP females. (b) Cha-GAL4-19B/UAS- GFP females. (c) Gad-GAL4-2/UAS- GFP females. (d) Gad-GAL4-5/UAS- GFP females. Arrowheads show the second antennal segment. Asterisks show the third antennal segment. Arrows show the maxillary palp. Triangles show taste neurons.

To determine whether GAL4-expressing cells in the sensory neurons between painGAL4 and Cha-GAL4 overlap, we used a yeast transcriptional regulator, GAL80. GAL80 blocks GAL4 activity by binding to the C-terminal activation domain of GAL4 (Ma & Ptashne 1987). Thus, Cha-GAL80 should suppress GAL4 expression in cholinergic neurons. In Cha-GAL80/painGAL4 UAS- GFP females, GFP reporter gene expression in the second antennal segment (Fig. 7e) and maxillary palp (Fig. 7f) was notably suppressed, suggesting that GAL4-expressing cells in these two sensory organs between painGAL4 and Cha-GAL4 overlap. However, Gad-GAL80 in females heterozygous for painGAL4 UAS- GFP did not affect GFP reporter gene expression (Fig. 7g–i), suggesting that GFP-positive sensory neurons in painGAL4 UAS- GFP females are not GABAergic.

pain mutant females seldom copulate with wild-type males without wings

To examine whether an auditory system is required for enhanced sexual receptivity in pain mutant females, we cut off wings of CS males and measured mating success rates (Fig. S2). CS females or pain1 mutant females were paired with CS males with or without wings (wingless or winged males, respectively) for 20 min. pain1 or CS females seldom copulated with wingless males (Fig. S2), suggesting that pain mutant females can detect these courtship songs produced by courting males, and detection of courtship song triggers high mating success rate in pain mutant females.

Conditional disruption of neurotransmission in Gad-expressing neurons enhances female sexual receptivity

Because TRP channels are calcium-permeable cation channels that can modulate neuronal activities and affect neurotransmitter release, the enhancement of female sexual receptivity by Gad-GAL4-induced pain RNAi expression is possibly because of reduced synaptic transmission from Gad-expressing neurons. The temperature-sensitive Dynamin mutation, shibirets1 (shits1), can be used in combination with the GAL4/UAS binary expression system to deplete neurotransmitters in a cell-type-specific and temperature-dependent manner (Kitamoto 2001, 2002). To determine whether the suppression of neurotransmission in Gad-expressing neurons also affects the female sexual receptivity, shits1 was expressed by Gad-GAL4 and the effects of conditional disruption of Dynamin function on female sexual receptivity were examined by comparing TC at 23°C and 30°C (Fig. 9a). In UAS- shits1/Gad-GAL4-2 females, TC at 30°C was significantly shorter than that at 23°C, whereas no significant difference was detected in +/Gad-GAL4-2 or UAS- shits1/+ (Fig. 9b). These data further show that GAL4-positive neurons in Gad-GAL4 play a key role in female sexual receptivity.

Figure 9.

Disruption of Dynamin function in Gad-GAL4-positive neurons enhances sexual receptivity. (a) Experimental paradigms of temperature shift are indicated. (b) UAS- shits1/+, +/Gad-GAL4-2 and UAS-shits1/Gad-GAL4-2 females and CS males were used. n, sample size; NS, not significant; *P < 0.05.

Although it is possible that the reduced TC in UAS- pain RNAi/Cha-GAL4 females is also caused by the suppression of neurotransmission in pain-expressing cholinergic neurons, we could not use UAS- shits1/Cha-GAL4 females to test this possibility because they are paralyzed at 30°C as reported by Kitamoto (2001).

Discussion

pain mutations enhance sexual receptivity in Drosophila virgin females

In this study, we found a new role for the Drosophila Pain TRP channel in the regulation of female sexual receptivity. pain mutant females showed a high mating success rate and enhanced sexual receptivity (Figs 1 and 2a). There are at least two possible factors that contribute to induction of the shortened TC: (1) pain mutant females elicit more vigorous courtship behaviors in males than CS females and (2) pain mutant females accept males more quickly. Because there was no significant difference in the courtship latency and CI between CS females and pain mutant females when CS males were used as sexual partner (Fig. 2b,c), we concluded that pain mutant females did not elicit more vigorous courtship behaviors in males than CS females. Hence, males are not the major factor for the enhanced sexual receptivity observed in the sexual behaviors of CS males and pain mutant females. Rather, the observed short TC is mainly attributable to pain mutant females. Because the expression of the wild-type pain gene in a pain mutant background restores female sexual receptivity to the wild-type level (Fig. 3a), the phenotype of the enhanced sexual receptivity is indeed caused by the pain mutations. Furthermore, we confirmed that knockdown of pain by RNAi in painGAL4-positive neurons also enhances female sexual receptivity (Fig. 3b), indicating that the reduced level of wild-type pain expression in painGAL4-positive neurons induces the hyper-receptive phenotype. In contrast to females, we confirmed that pain mutant males are fertile and their courtship does not show any major defects.

In vivo analysis of the Pain TRP channel showed that Pain is required for thermal and mechanical nociception (Tracey et al.2003) as well as for avoidance of wasabi (Al-Anzi et al.2006). Recently, Sokabe et al.(2008) found that Pain is directly activated by heat stimulation in vitro. Unexpectedly, however, Pain is completely insensitive to direct touch or the main ingredient of wasabi, allyl isothiocyanate (AITC), in vitro (Sokabe et al.2008). Endogenous ligands of Pain relevant to the regulation of female sexual receptivity remain elusive. It will be very interesting to investigate the physiological mechanisms by which the activity of the Pain TRP channel in the nervous system of females is regulated in response to their interaction with courting males.

Several mutations induce a low sexual receptivity in virgin females (Hall 1994; Kerr et al.1997; Ringo et al.1991; Suzuki et al.1997; Tompkins et al.1982), whereas little is known about the mutations that lead to the enhanced sexual receptivity. Sakai and Kidokoro (2002) have reported that the induction of a repressor isoform of dCREB2, dCREB2-b, enhances female sexual receptivity as observed in pain mutants. dCREB2-b functions as a repressor of Drosophila CREB (Yin et al.1995). Because the Pain TRP channel is permeable to Ca2+ (Sokabe et al.2008) and CREB can function as a Ca2+-inducible transcription factor (Lonze & Ginty 2002), it is possible that dCREB2-mediated gene expression is under the control of the Pain TRP channels and plays a role in maintaining the wild-type level of female sexual receptivity.

Female sexual receptivity is regulated by the Pain TRP channel expressed in neurons with particular neurotransmitter phenotypes

Induction of pain RNAi driven by Cha-GAL4 and Gad-GAL4 enhanced female sexual receptivity (Figs 4 and 6), suggesting that downregulation of pain expression in cholinergic and/or GABAnergic neurons leads to the enhanced sexual receptivity observed in pain mutant females. However, targeted expression of pain RNAi in dopaminergic and/or serotonergic neurons had no effect on female sexual receptivity (Fig. 5). In Drosophila, dopamine-depleted females are significantly less receptive to males than control females (Neckameyer 1998a), indicating the positive regulatory role of dopaminerginc neurons in the regulation of female sexual receptivity. However, our results suggest that the dopaminergic regulation of female sexual receptivity is not mediated by the Pain TRP channel.

Results of two independent experiments using UAS- pain RNAi and UAS- shits1 in combination with Gad-GAL4 suggest that the downregulation of pain in Gad-GAL4-positive neurons leads to the suppression of GABAergic synaptic transmission, resulting in enhanced female receptivity. Sokabe et al.(2008) have recently reported that Pain is a cation channel with extremely high Ca2+ permeability. Hence, it is possible that pain mutations or knockdown of pain expression causes a decrease in the level of intracellular Ca2+, which consequently reduces synaptic transmission in GAL4-expresssing neurons in the Gad-GAL4 line and results in the enhanced sexual receptivity in females.

Possible anatomical sites in the sensory neurons and CNS responsible for Pain-mediated regulation of female sexual receptivity

We showed that induction of pain RNAi in Cha-GAL4-positive neurons enhances female sexual receptivity (Fig. 4), suggesting that a decrease in the level of intracellular Ca2+ in pain-expressing cholinergic neurons also enhances female sexual receptivity. However, owing to the widespread gene expression in the nervous system directed by Cha-GAL4 (Salvaterra & Kitamoto 2001), the neural circuits or the sites in Cha-expressing cells relevant to pain-mediated regulation of female sexual receptivity are not yet determined. Possible anatomical sites that are important for pain-mediated regulation of female receptivity include sensory neurons involved in detecting signals that are transmitted from male flies during courtship. Because the courtship song generated by courting males stimulates female sexual receptivity (Kyriacou & Hall 1982; Manning 1967; Tomaru et al.2000; von Schilcher 1976), the auditory neurons in the antennal second segment are the most likely candidate. Al-Anzi et al.(2006) have observed the pain-reporter gene expression in Johnston's organs in the second antennal segment. We also confirmed that both painGAL4 and Cha-GAL4 drive the GFP expression in those sensory neurons (Figs 7a,d and 8a,b). Furthermore, Cha-GAL80 suppresses GFP expression in Johnston's organs in painGAL4 females (Fig. 7e). These data suggest the expression of Pain in cholinergic auditory neurons. pain1 or CS females seldom copulated with wingless males (Fig. S2). Thus, it is possible that the enhanced female sexual receptivity is caused by altered neural properties of the cholinergic auditory system in pain mutants. Alternatively, Cha-GAL4-positive neurons in the maxillary palp and/or adult CNS may also be responsible for the pain-mediated regulation of female sexual receptivity.

The ibx gene encodes the L1-type cell adhesion molecule, Neuroglian. ibx mutant females show a low sexual receptivity and have defects in central brain structures including the mushroom bodies (MBs) that are composed of the higher order neurons of olfaction (Carhan et al.2005; Kerr et al.1997). Carhan et al.(2005) have reported that the expression of the wild-type ibx gene in the ibx mutant background partially rescues the female sexual receptivity without rescuing the brain structure phenotype (Carhan et al.2005). Furthermore, Neckameyer (1998b) has reported that female sexual receptivity is unaffected by the ablation of MBs. Thus, MBs may not be the important sites for female sexual receptivity, although it has been suggested that Pain TRP channels are expressed in MBs (Al-Anzi et al.2006; Xu et al.2006).

GFP reporter gene expression was not detected in the sensory neurons of Gad-GAL4/UAS- GFP female head (Fig. 8c,d), and Gad-GAL80 did not suppress the reporter gene expression in the sensory neurons of painGAL4 females (Fig. 7g–i). In addition, we confirmed that Gad-GAL4 drives GFP expression in the adult brain (Fig. S3) as reported by Mehren and Griffith (2006). Taken together, our results indicate that Gad-GAL4 drives gene expression mainly in the CNS. Thus, it is possible that GABAergic painGAL4 neurons in the adult brain play a key role in the regulation of female sexual receptivity. Although the neural circuits or the regions of the brain relevant to pain-mediated regulation of sexual receptivity are still unknown in Drosophila virgin females, our results suggest that the Pain TRP channels not only in sensory neurons but also in the brain are involved in the manifestation of the wild-type level of female sexual receptivity in Drosophila.

Acknowledgments

We thank Seymour Benzer and W. Daniel Tracey Jr for all pain mutants and UAS-pain; Yoshiaki Kidokoro, Junjiro Horiuchi, Takaaki Sokabe and Azusa Kamikouchi for critically reading the manuscript and useful comments; and Minoru Saitoe and Kohei Ueno for useful discussion. This work was supported by a Selective Research Fund of Tokyo Metropolitan University (to T.S.) and an NIH grant MH62684 (to T.K.).

Ancillary