Skip to main content
NCBI home page
Howard Hughes Medical Institute Author Manuscripts logoLink to Howard Hughes Medical Institute Author Manuscripts
. Author manuscript; available in PMC: 2014 May 14.
Published in final edited form as: Cell Rep. 2013 Oct 31;5(3):678–686. doi: 10.1016/j.celrep.2013.09.039

Motor control of Drosophila courtship song

Troy R Shirangi 1,*, David L Stern 1, James W Truman 1
PMCID: PMC3926192  NIHMSID: NIHMS543888  PMID: 24183665

SUMMARY

Many animals utilize acoustic signals – or songs – to attract mates. During courtship, Drosophila melanogaster males vibrate a wing to produce trains of pulses and extended tone called pulse and sine song, respectively. Courtship songs in the genus Drosophila are exceedingly diverse and different song features appear to have evolved independently of each other. How the nervous system allows such diversity to evolve is not understood. Here, we identify a wing muscle in D. melanogaster (hg1) that is uniquely male-enlarged. The hg1 motoneuron and the sexually dimorphic development of the hg1 muscle are required specifically for the sine component of the male song. In contrast, the motoneuron innervating a sexually monomorphic wing muscle, ps1, is required specifically for a feature of pulse song. Thus, individual wing motor pathways can control separate aspects of courtship song and may provide a “modular” anatomical substrate for the evolution of diverse songs.

INTRODUCTION

Animal courtship often involves some of the most elaborate behaviors ever observed. Courting males of many insect species, for example, utilize complex songs (Gerhardt and Huber, 2002) and dances (Spieth, 1974) to enhance their chance of mating. These courtship behaviors are exceedingly diverse between species, most likely as a result of sexual selection (Andersson, 1994). How the neural circuits encoding these complex behaviors allow such diversity to evolve is not known. The modularity of development is understood to have enabled the evolution of complex morphological forms (Schlosser and Wagner, 2004). Likewise, complex behaviors may also be built from modules that can be quickly modified and rearranged during evolution (Weber et al., 2013). The courtship songs of the genus Drosophila display enormous diversity, and individual features of the song appear to have evolved independently of each other [reviewed in (Markow and O’Grady, 2005)]. This pattern of evolutionary change suggests that the neural control of courtship song may be modular, but where this modularity may lie in the neural circuit controlling song is not known.

Previous work has led to the identification of multiple classes of neurons active in the Drosophila melanogaster male courtship song circuit, from neurons in the brain that integrate social cues (Kimura et al., 2008; Kohatsu et al., 2011; Pan et al., 2012; von Philipsborn et al., 2011), to descending neurons that activate pulse song production (Kohatsu et al., 2011; von Philipsborn et al., 2011), and the thoracic interneurons that may contribute to the pattern generator(s) that compose the pulse song (Clyne and Miesenbock, 2008; von Philipsborn et al., 2011). Despite this progress, it is not understood how these circuits orchestrate the peripheral neuromuscular events that modulate the wing vibrations of courtship song. Over thirty years ago, Ewing showed that some wing muscles fire rhythmically during pulse or sine song, or both (Ewing, 1977, 1979). However, these studies lacked the resolution to determine the causal links between individual wing muscles and song components.

Here, we identify a thoracic wing muscle in D. melanogaster, hg1, which is uniquely enlarged in males. Courting males with inhibited hg1 motoneurons are unable to produce sine song, but sing pulse song normally. Feminization of the hg1 muscle reduces the size of hg1 in males and reduces the volume with which males sing sine song. In contrast, males with silenced motoneurons innervating a sexually monomorphic wing muscle, ps1, have normal sine song, but generate pulse song with a decreased carrier frequency and amplitude. These results demonstrate that the motor control of Drosophila courtship song is modular. Changes in individual motor pathways during evolution would thereby allow discrete components of the male song to change independently of others. Finally, we show that females are less willing to mate with males that either lack sine song or produce sine song with reduced volume, suggesting that a female may judge male quality in part by how loudly a male sings sine song.

RESULTS

A sexually dimorphic control wing muscle in Drosophila

In Drosophila, contractions of the large indirect wing muscles deform the thorax to power the wings during flight [reviewed in (Dickinson and Tu, 1997)] and during courtship song (Ewing, 1977). Additional small muscles located adjacent to the lateral thoracic wall—the control wing muscles—modulate these wing movements (Dickinson and Tu, 1997; Ewing, 1979) (Figure 1A). Most control muscles insert into sclerites at or near the wing hinge (Wisser and Nachtigall, 1984) and are thought to influence wing motion by altering the mechanical properties of the hinge (Dickinson and Tu, 1997). We observed that most wing muscles are larger in females than in males, consistent with the difference in overall body size between the sexes (Figure 1A; Figure S1). In contrast, the control wing muscle hg1 is larger in males than in females (Figure 1A, second panel, inset; Figure 1B). Hg1 inserts into the posterior notal wing process (Wisser and Nachtigall, 1984) and, in Calliphora, hg1 activity is associated with changes in wing stroke amplitude during a flight turn (Dickinson and Tu, 1997; Nachtigall and Wilson, 1967).

Figure 1. A sexual dimorphism in the direct wing muscles of Drosophila melanogaster.

Figure 1

Open in a new tab

(A) Single slices from a confocal stack of a hemi-thorax stained with phalloidin to show the locations of 13 of the 18 control wing muscles (Dickinson and Tu, 1997; Wisser and Nachtigall, 1984). The direct wing muscles are classified according to the sclerite in which they insert (basalare, first and third pterale, and the posterior notal wing process) (Wisser and Nachtigall, 1984). ‘Tension muscles’ (Wisser and Nachtigall, 1984) adjust wing movement indirectly by regulating the rigidity of the pleural wall. Inset shows the difference in size between the hg1 muscles of males versus females.

(B) The volumes of the control wing muscles in males and females. Individual points, and the mean and standard deviation are given for each. P-values were obtained using a standard student’s t-test.

(C) Sexually dimorphic development of hg1 is dependent upon dsx and not male-specific fru. Null allele combinations were used. Individual points, and the mean and the standard deviation are given for each muscle. P-values were obtained using a standard student’s t-test.

(D) Confocal sections showing that dsxGal4 (Robinett et al., 2010), which accurately recapitulates the endogenous expression of dsx, drives reporter expression only in the ps1 and hg1 muscles.

(E) Confocal sections showing that fruGal4 (Stockinger et al., 2005) drives reporter expression in motoneurons that innervate seven control wing muscles (ps1, hg1-3, I2, and d-, v-Tp). fruGal4 is also expressed in motoneurons that innervate the dorsal longitudinal muscle (not shown), and the dorsoventral muscles (dvms).

We tested whether sexually dimorphic development of hg1 requires the activity of the Drosophila sex differentiation genes doublesex (dsx) and fruitless (fru) [reviewed in (Billeter et al., 2006; Christiansen et al., 2002)]. Among thoracic wing muscles, dsx is expressed only in muscles hg1 and ps1 (Figure 1D), while male-specific fru is not expressed in muscles, but rather in several wing motoneurons [more than described previously (Rideout et al., 2007)], including a motoneuron that innervates hg1 (Figure 1E). Removal of dsx function abolished the dimorphism, reducing the size of hg1 in males relative to its size in females (Figure 1C). The loss of male-specific fru, however, did not alter sexually dimorphic development of hg1 (Figure 1C). Thus, the male- and female-specific isoforms of dsx promote and suppress hg1 muscle growth, respectively, while fru function does not influence the size dimorphism of hg1.

The hg1 motoneuron is required for sine song but not pulse song

To test whether the hg1 motoneuron contributes to courtship song, we identified two transgenic lines (Jenett et al., 2012; Pfeiffer et al., 2008), R21A01-Gal4 and R14B02-LexA::p65, that each drove reporter expression in the motoneuron that innervates hg1 and in additional neurons in the nervous system (not shown). To target the hg1 motoneuron specifically, R14B02-LexA::p65 was used to drive a Flp recombinase, which excised a transcriptional stop cassette from a UAS-GFP transgene driven by R21A01-Gal4. When genetically intersected in adult males, these lines drove reporter gene expression in only the hg1 motoneuron (Figure 2A). This motoneuron likely provides sole excitatory input to hg1, because all control muscles examined thus far in Drosophila are innervated by a single excitatory motoneuron (Trimarchi and Schneiderman, 1994). We used this intersection (R14B02 ∩ R21A01) to suppress the activity of the hg1 motoneuron by driving a GFP-tagged version of the inwardly rectifying K+ channel, Kir2.1 (Baines et al., 2001). R14B02 ∩ R21A01 drove kir2.1-GFP expression stochastically; males either lacked kir2.1-GFP expression (“non-expressers”), or expressed kir2.1-GFP in the hg1 motoneuron unilaterally or bilaterally (Figure S2A). This allowed us to compare the songs produced by genetically identical males that had functional hg1 motoneurons, or hg1 motoneurons that were silenced on one or both sides. We recorded courtship song (Arthur et al., 2013) from R14B02 ∩ R21A01 > kir2.1-GFP males and then dissected the ventral nerve cords of these males to determine which hg1 motoneurons expressed kir2.1-GFP.

Figure 2. Silencing the hg1 motoneuron in males specifically impairs their ability to generate sine song.

Figure 2

Open in a new tab

(A) The intersection between R14B02-LexA::p65 and R21A01-Gal4 targets the hg1 motoneuron in adult males with perfect specificity. Magenta = Neuroglian, Green = GFP. Scale bar = 50 μm. Inset: hg1 neuromuscular junction. Magenta = phalloidin, Green = GFP.

(B) Expression of kir2.1-GFP in the hg1 motoneuron selectively ablates sine song during male courtship behavior. Pulse and sine song events were detected (Arthur et al., 2013) in a five-minute clip.

(C) The proportion of sine song amount relative to pulse song amount of non-expressers, and uni- and bilateral expressers. To quantify the amount of pulse or sine song that a courting male sings, we measured the pulse and sine song index, which is the fraction of time the male spends singing pulse or sing song. Histograms show the distribution of pulse and sine song indices amongst the three classes.

(D–J) Males with silenced hg1 motoneurons generate pulse song normally. (D) Mode inter-pulse interval, (E) mean pulse carrier frequency, (F) mean RMS amplitude, (G) mean rescaled pulse shape, and (H) mean pulse train length are statistically equivalent between the three classes. Mean pulse shape ± standard deviation for all pulses collected from individuals of the three classes is shown in (G). Histograms in (H) show the distribution of pulse train lengths amongst the three classes. (I) Mean intra-bout pause length is greater in the bi- and unilateral males due to reduced sine song production. Replacing sine song with silence in the non-expresser males increases their mean intra-bout pause length to an amount comparable to bilateral expressers. (J) Mean inter-bout pause length is equivalent between the sets. Panels (D–F) and (H–J) show individual points, and the mean and the standard deviation. (n.s.) Significance measured using one-way ANOVA with Tukey-Kramer post-hoc test for multiple comparisons. The P-value between the bilateral or unilateral males and the non-expresser males is greater than 0.7.

Males lacking kir2.1-GFP expression in both hg1 motoneurons generated abundant pulse and sine song (Figure 2B, C). In contrast, most males expressing kir2.1-GFP in both hg1 motoneurons displayed a complete absence of sine song (Figure 2B, C). Males with kir2.1-GFP expressed in only one hg1 motoneuron produced a quantity of sine song intermediate between the non-expressers and bilateral expressers (Figure 2C). All three categories of males produced statistically indistinguishable quantities of pulse song (Figure 2C), including inter-pulse intervals, pulse carrier frequencies, pulse amplitudes, pulse shapes, and pulse train lengths (Figure 2D–H). Similar results were obtained using tetanus neurotoxin light chain (Eisel et al., 1986; Pfeiffer et al., 2010; Sweeney et al., 1995) as an alternate means to silence the hg1 motoneuron (Figure S2B–F).

These data indicate that the hg1 motoneuron is required for sine song, but is dispensable for pulse song. Nevertheless, it appears that the nervous system of a male with compromised hg1 motoneurons continues to send the command to produce sine song. Courtship song is normally arranged in bouts of song (i.e., concatenated trains of pulse or sine song separated by pauses of less than a second) separated by periods of silence. The duration of the silent periods between bouts of song was similar in males of the three classes (Figure 2J). The intra-bout pauses, however, were lengthened in males that expressed kir2.1-GFP in the hg1 motoneurons bilaterally compared to non-expressers (Figure 2I). This increased intra-bout pause duration most likely reflects the dropping out of trains of the sine song during song bouts, since we can mimic this duration in normal males by computationally replacing their sine song trains with silence (Figure 2I). Simultaneous video and audio recording revealed that courting males with bilaterally silenced hg1 motoneurons extend a wing during periods of silence that often precede or follow trains of pulse song (Movie S1), which we interpret as putative sine song trains. Moreover, the intermediate reduction in sine song production among males with a unilaterally silenced hg1 motoneuron reflects their ability to sing sine song with the contralateral wing but not the ipsilateral wing (Movie S1). Therefore, it appears that males with silenced hg1 motoneurons “think” they are singing sine song, despite being mechanically unable to do so.

Males with feminized hg1 muscles sing sine song quietly

To test the role of hg1’s sexual size dimorphism in the production of courtship song, we feminized hg1 (thereby, reducing its size) in otherwise normal males by driving female-specific transformer (traF) (McKeown et al., 1988) in the hg1 muscle. We identified a Gal4 line (Jenett et al., 2012; Pfeiffer et al., 2008) (R84G06-Gal4) that targeted all fibers of several control wing muscles, including the two dsx-expressing muscles, hg1 and ps1 (Figure 3A, B; Figure 1D), and several neurons in the adult nervous system (Figure S3A; and not shown). R84G06-Gal4 driving traF (R84G06-Gal4 > traF) reduced the hg1 muscle to the size observed in females (Figure 3C). These males were otherwise phenotypically normal in gross appearance; they courted females vigorously and generated most aspects of song normally relative to controls (Figure 3D–G). However, R84G06-Gal4 > traF males sang sine song with significantly reduced amplitude compared to controls, while pulse song amplitude was unaffected (Figure 3H). This phenotype was not due to traF expression in the nervous system because R84G06-Gal4 > traF males carrying a transgene [R57C10-Gal80-6 (Harris, 2012)] to suppress the neuronal expression of traF also sang sine song with reduced volume relative to controls (Figure S3A–F). Moreover, the change in sine song amplitude in R84G06-Gal4 > traF males did not result from traF expression in ps1. We identified a line, R40D04-Gal4, which targeted the ps1, but not the hg1 muscle (Figure S3G, H). R40D04-Gal4 > traF males generated song normally compared to controls (Figure S3I–N). Thus, the hg1 motoneuron is required to generate sine song, whereas the sexual dimorphism of hg1 is necessary for males to sing sine song at normal volume.

Figure 3. Sexually dimorphic development of hg1 is required for maximal sine song amplitude.

Figure 3

Open in a new tab

(A, B) R84G06-Gal4 drives reporter expression in all fibers that constitute the ps1, hg1, and hg2 wing muscles (and others not shown). (*) refers to an unidentified muscle associated with ps1. This muscle is not targeted by dsxGal4 (Fig. 1D). R84G06-Gal4 was crossed to pJFRC2-10XUAS-IVS-mCD8::GFP. Magenta = Phalloidin; Green = GFP.

(C) R84G06-Gal4 driving traF feminizes and reduces the size of the hg1 muscle in males. P-values were obtained using a standard student’s t-test.

(D–H) Feminization of the hg1 muscle in males does not affect (D) the relative amount of sine and pulse song produced, (E) the mode inter-pulse interval, (F) the mean pulse and sine song carrier frequencies, or (G) the mean re-scaled pulse shape relative to control genotypes. (H) Mean RMS amplitude of sine song, but not pulse song, is reduced in males with a feminized hg1 muscle relative to controls. Panels (A, C, E, F) show individual points, and the mean and the standard deviation. Significance in panels (E) and (F, H) were measured using one-way ANOVA and Kruskal-Wallis test, respectively, with Tukey-Kramer post-hoc test for multiple comparisons. pBPGal4.1Uw (Pfeiffer et al., 2008) strain carries the ‘empty vector’ Gal4 inserted into attP2.

The ps1 control wing motoneuron is required for a discrete feature of pulse song

It is noteworthy that manipulating hg1 function did not affect pulse song. This suggests that some wing motor pathways may influence courtship song in specific ways. To test this, we drove kir2.1-GFP specifically in the motoneuron innervating the ps1 control muscle by genetically intersecting two transgenic lines (R48F07-LexA::p65 and R73C03-Gal4) (Jenett et al., 2012; Pfeiffer et al., 2008) (Figure 4A). Ps1, like hg1, expresses dsx (Figure 1D) and is innervated by a fruM-expressing motoneuron (Figure 1E), but is not enlarged in males (Figure 1B). Males with inhibited ps1 motoneurons displayed a reduction in their pulse carrier frequency and pulse amplitude relative to two controls, but sang pulse and sine song otherwise normally (Figure 4B–F). These data are consistent with the hypothesis that the motor control of Drosophila courtship song is, at least in part, modular.

Figure 4. Inhibition of the ps1 motoneuron specifically reduces pulse carrier frequency and amplitude.

Figure 4

Open in a new tab

(A) The intersection between R48F07-LexA::p65 and R73C03-Gal4 targets the ps1 motoneuron in adult males. Arrowheads in “Ventral VNC” point to the ps1 motoneuron cell body. We note that this intersection stochastically labels 3 or 4 interneurons in the mesothoracic ganglion that appear to innervate the leg neuropil (not present in this preparation), but consistently labels the ps1 motoneuron. This intersection also weakly targets 4 or 5 neurons in the anterior brain (arrowhead points to cell bodies in “Anterior brain.”). Scale bar: 50 mm. Inset: ps1 neuromuscular junction. Magenta = phalloidin, Green = GFP.

(B–F) The proportion of sine song relative to pulse song (B), the mode inter-pulse interval (C), and the mean re-scaled pulse shape (E) are statistically equivalent between experimental (magenta triangles) and controls (orange squares and black circles). Mean pulse carrier frequency (D) and mean RMS amplitude (F) of pulse song, but not sine song, is reduced in the experimental males relative to controls. (n.s.) The P-value between the experimental class and each control is greater than 0.99.

Males that lack sine song are ineffective courters

Our ability to precisely manipulate sine song production allowed a novel test of the requirement for sine song during courtship. R14B02 ∩ R21A01 males were used to drive kir2.1-GFP or tetanus neurotoxin light chain in the hg1 motoneuron and these males were tested for their ability to court, to sing pulse and sine song, and to mate with wild-type females relative to two control genotypes. Males with silenced hg1 motoneurons courted females (Figure 5A) and sang pulse song (Figure 5B, C) at levels statistically indistinguishable from controls, but were strongly impaired in their ability to sing sine song (Figure 5B, C). These males mated at a significantly lower rate than did control males (Figure 5D). These mating deficits are most parsimoniously attributed to either the large reduction in sine song production or the presence of larger gaps within the bouts of song. However, at present we cannot exclude the possibility that some other defects unnoticed in our analyses account for the reduction in mating success.

Figure 5. Reduced production of sine song is associated with lower mating success.

Figure 5

Open in a new tab

(A) Males with silenced hg1 motoneurons court wild-type females as vigorously as do control males. Courtship index is the fraction of time a male spends engaged in any step of male courtship behavior during 10 min of observation. Individual points, and the mean and standard deviation are given for each. (n.s.). The P-value between the experimental class and each control is greater than 0.99.

(B–C) Sine song indices of males with silenced hg1 motoneurons are reduced relative to controls, while their pulse song indices are comparable. The difference in mean song index between the experimental and controls was significant for sine song, but not significant for pulse song, at the 0.05 level.

(D) Females are less receptive to males with inhibited hg1 motoneurons. The fraction of male and female pairs that mated over 30 min is shown. P-values were measured using a logrank test. pBDPGal4.1Uw and pBDPLexA::p65Uw strains carry the ‘empty vector’ Gal4 and LexA::p65 inserted into attP2 and attP40, respectively.

(E) R84G06-Gal4 males driving traF court wild-type females as vigorously as controls. Courtship index measured as in (A).

(F) Females are less receptive to R84G06-Gal4 > traF males relative to controls. The fraction of male and female pairs that mated over 30 min is shown. P-values were measured using a logrank test.

We also tested if sexually dimorphic development of hg1—and thus production of relatively loud sine—song is important for female receptivity. R84G06-Gal4 > traF males courted females robustly (Figure 5E), produced pulse song normally (Figure 3B–E), and generated sine song with lower amplitude compared to controls (Figure 3H). Females mated with R84G06-Gal4 > traF males at a lower rate than they mated with controls (Figure 5F). These mating deficits were not due to traF expression in the ps1 muscle, because R40D04-Gal4 > traF males mated with females as efficiently as did controls (Figure S3O). These results support the hypothesis that sine song produced at wild type volume contributes to the mating efficiency of D. melanogaster males. Females may use the volume of sine song as an indicator of male quality, as others have postulated (Rybak et al., 2002).

DISCUSSION

We have shown that the hg1 wing muscle and its sexually dimorphic development are required for the sine component of courtship song, while the ps1 wing muscle is required for a specific aspect of pulse song, but not sine song. The sexual size dimorphism in hg1 is analogous to the sexual differences in the size and physiology of the laryngeal muscles of singing Xenopus laevis frogs (Kelley and Tobias, 1999). Contraction of hg1 pulls the posterior notal wing process in an anterioventral direction (Dickinson and Tu, 1997; Wisser and Nachtigall, 1984), but how this event relates to the wing motions underlying sine song is not clear. Our observation that feminizing hg1 reduces the amplitude of sine song suggests that hg1 may provide power to the wing strokes that generate sine song. Although hg1 is essential for sine song, it obviously does not work alone and the performance of this song component involves the synergistic actions of other wing muscles (Ewing, 1977, 1979). Given the role of ps1 in linking the pleural and sternal apophyses (Dickinson and Tu, 1997; Pringle, 1957), our results further suggest that thoracic rigidity regulates pulse carrier frequency.

Our findings echo a burgeoning idea that complex behaviors are composed of “modules” that allow discrete aspects of a behavior to evolve independently of others (Weber et al., 2013). Our results demonstrate that pulse and sine song are produced in part by separate sets of wing muscles, suggesting that the wing periphery is to a certain extent modular. By “modular,” we mean that discrete features of the behavior can be functionally mapped to morphologically discrete subunits in the motor periphery. Given that the wing periphery consists of a relatively small number of muscles, the modularity we observe may be due to the biomechanical constraints intrinsic to the wing musculo-skeletal system. Species of the genus Drosophila display extensive diversity in courtship song, and different song features appear to evolve independently of each other (Hoikkala, 2005; Markow and O’Grady, 2005). The apparent specialization of wing motor pathways for different aspects of song may provide a modular anatomical template for the evolution of different components of courtship song.

EXPERIMENTAL PROCEDURES

Fly stocks

Flies were reared on standard cornmeal and molasses food at 25°C. The stocks used in this paper included the following: Canton S (CSA), dsxGAL4 (Robinett et al., 2010), fruGal4 (Stockinger et al., 2005), pJFRC2-10XUAS-IVS-mCD8::GFP (attP2) (Pfeiffer et al., 2010), pBDPGAL4.1Uw (attP2) (Pfeiffer et al., 2008), and pBDPLexA::p65Uw (attP40)(Pfeiffer et al., 2010). dsx1649–9625 (i.e., Df(3R)f01649-d09625)(Chatterjee et al., 2011) was a gift from C. Robinett (HHMI/JFRC). dsx1, fru440, and fruAJ96u3 were provided by B. Baker (HHMI/JFRC). R21A01-Gal4, R84G06-Gal4, R40D04-Gal4, and R73C03-Gal4 are from the Rubin GAL4 collection(Jenett et al., 2012; Pfeiffer et al., 2008). R14B02-LexA::p65 and R48F07-LexA::p65 (attP40) were a gift from G. Rubin (HHMI/JFRC). pJFRC79-8XLexAop2-FlpL (attP40), pJFRC41-10XUAS-FRT>STOP>myrGFP (attP2), pJFRC39-10X-FRT>STOP>FRT-E86tetLC (attP2), pJFRC56-10XUAS-FRT>STOP>FRT-kir2.1-gfp (attP2), and R57C10-Gal80-6 (su(Hw)attP8) (Harris, 2012) were gifts from B. Pfeiffer (HHMI/JFRC). A Flpd-OUT STOP cassette (Nern et al., 2011) was cloned in front of a 10XUAS vector(Pfeiffer et al., 2010) containing E86tetLC, the cloned bacterial tetanus toxin light chain gene as previously described (Eisel et al., 1986), containing a few minor base pair changes to the published sequence resulting in pJFRC39-10X-FRT>STOP>FRT-E86tetLC (B. Pfeiffer, personal communication). M. McKeown (Brown University) provided UAS-traF (P-element).

Immunohistochemistry

To visualize GFP reporter expression in wing muscles or their neuromuscular junctions (NMJs), hemi-thoraxes of adults aged for about five days were dissect in PBS and fixed in 4% formaldehyde (buffered in PBS) for approximately 50 min at room temperature. To effectively stain the control muscles that line the thoracic lateral wall, we removed the six large fibers of the dorsal longitudinal muscle after fixation. Fixed tissues were washed in PBS-TX (PBS with 1% Triton X-100) and incubated for 3–4 days at 4°C in PBS-TX containing rabbit anti-GFP (1:1000; Invitrogen). Tissues were washed at room temperature for several hours in PBS-TX, and incubated for 3–4 days at 4°C in PBS-TX containing AlexaFluor 488-conjugated donkey anti-rabbit (1:500; Invitrogen) and Texas Red-X phalloidin (1:50; Life Technologies). Tissues were washed all day in PBS-TX, placed onto poly-lysine-coated coverslips, dehydrated through an ethanol series, cleared in xylenes, and mounted in DPX (Sigma-Aldrich). Nervous systems were prepared and stained as above except mouse anti-Neuroglian (1:40; Developmental Studies Hybridoma Bank) and AlexaFluor 649-conjugated donkey anti-mouse were included in the primary and secondary antibody incubations, respectively (and without the addition of phalloidin). Tissues were imaged on a Zeiss LSM 510 confocal microscope at 10x (hemi-thorax) or 40x (CNS) with optical sections at 1μm (hemi-thorax) or 0.5μm (CNS) intervals.

Measurement of wing muscle volume

To visualize and measure the volumes of wing muscles, hemi-thoraxes from males or females were dissected, fixed, and washed in PBS-TX as described above, and placed in PBS-TX containing Texas Red-X phalloidin (1:50; Life Technologies) for 3–4 days at 4°C. Tissues were washed in PBS-TX all day at room temperature, and cleared and mounted as described above. Confocal stacks of phalloidin-stained hemi-thoraxes were imported into Amira (Visualization Sciences Group). Wing muscles were segmented and reconstructed by selecting and assigning pixels through the confocal series to labels of their respective wing muscle. Amira was used to measure muscle volume using the appropriate voxel dimensions (in μm).

LexA/Gal4 intersectional strategy

A subset of lines from the Rubin GAL4 collection(Jenett et al., 2012; Pfeiffer et al., 2008) was screened for reporter expression at the NMJ of wing muscles. R21A01 and R14B02 were found to target the hg1 motoneuron and additional non-overlapping wing motoneurons and neurons in the CNS. R48F07 and R73C03 were found to target the ps1 motoneuron and additional neurons that were largely non-overlapping. A LexA::p65 version of R14B02 and R48F07 (in attP40 on the second chromosome) was found to also target the hg1 and ps1 motoneurons, respectively. To intersect the Rubin LexA::p65 and Gal4 lines and specifically target the hg1 or ps1 motoneurons, R14B02- or R48F07-LexA::p65 was used to drive a Flp recombinase, which excised a transcriptional stop cassette from a GFP-, kir2.1-GFP-, or E86tetLC-expressing transgene driven by R21A01- or R73C03-Gal4. The observed stochasticity was a useful feature of the cross. Males from a stock carrying R14B02-LexA::p65 and R21A01-Gal4, or R48F07-LexA::p65 and R73C03-Gal4, were crossed to virgin females from stocks carrying pJFRC79-8XLexAop2-FlpL with pJFRC41-10XUAS-FRT>STOP>myrGFP, or pJFRC39-10X-FRT>STOP>FRT-E86tetLC, or pJFRC56-10XUAS-FRT>STOP>FRT-kir2.1-gfp.

Recording courtship song

Newly eclosed males were collected under CO2, individually housed for 4–7 days (unless otherwise noted) at 25°C and 30% humidity with a 12-hr light/dark cycle. Virgin Canton S females were group housed and aged under similar conditions. Courtship song was recorded as described (Arthur et al., 2013) for 10–15 minutes at 25°C within 2 hrs after the start of the subjective day using individual pairs of males and decapitated females. In experiments using R14B02-LexA::p65 ∩ R21A01-Gal4 and kir2.1-GFP, the VNC of each male was dissected immediately after recording and stained as described above to score kir2.1-GFP expression in the hg1 motoneurons. Individual song recordings were subsequently categorized according to expression (i.e., non-expresser, unilateral and bilateral expresser) and analyzed.

Recording courtship movies

We recorded audio and video simultaneously of courting pairs using a Flea®3 (USB 3.0) color camera (FL3-U3-32S2C-CS from Point Grey) shutter-triggered by an output from the DAQ used to collect audio signals from the microphone placed directly beneath the courting flies (Arthur et al., 2013). Synchronized video and audio were captured using a custom Matlab program, called omnivore, written by B. Arthur (https://github.com/bjarthur/omnivore.git). Data were visualized and movies were exported from a custom Matlab program, called tempo, written by F. Midgley and B. Arthur (https://github.com/frank-midgley/tempo.git).

Courtship song analyses

Recordings of courtship song were segmented and analyzed using Matlab R2011b as described (Arthur et al., 2013). The pulse song index was calculated by dividing the sum of a male’s inter-pulse intervals by the total recording time. The sine song index was calculated by dividing the sum of the lengths of a male’s sine song trains by the total recording time. The inter-pulse interval, pulse and sine carrier frequencies, models of pulse shape, pulse train lengths and bout pause lengths were measured as described (Arthur et al., 2013). Pulse and sine song amplitude was measured by calculating the square root of the mean of the squares (RMS) of all pulses or trains of sine song. The inter-pulse intervals in Fig. 5C were estimated independently of pulse carrier frequency by fitting an envelope to each pulse to estimate pulse duration and calculating the inter-pulse interval as the time from the end of one pulse to the center of the next pulse. We considered bouts of courtship song as concatenated trains of pulse or sine song separated by pauses of less than a second. Intra-bout pauses are pauses of shorter than one second between trains of pulse and sine song.

Courtship and mating assays

Males and Canton S virgin females were collected, housed and aged as described above. Males and females were aged for 4–10 and 5–8 days, respectively. Courtship and mating assays were done at 25°C, under white light within 2 hrs after the start of the subjective day using individual pairs of males and virgin Canton S females. Males and females were loaded into behavioral chambers (diameter: 1 cm; height: 2 mm) at room temperature, kept separated by a plastic sheet, and allowed to acclimate for 15 minutes in the behavioral incubator. The barrier was quickly removed and the pairs were video recorded for 30 minutes. The courtship index was measured by dividing the total amount of time the male spent engaged in any courtship step by the total observation time.

Statistics

All statistics were calculated in Matlab. In most cases, a randomization test was performed to determine if the experimental and control classes displayed significant heterogeneity. For each dataset, we performed Analysis of Variance on 10,000 randomly permuted datasets and used the resultant distribution of F statistics to estimate the significance of the F statistic from the original data. For comparisons yielding significant heterogeneity, we performed a one-way ANOVA or a Kruskal-Wallis test, followed by a Tukey-Kramer post-hoc comparisons test.

Supplementary Material

01
02
Download video file (20.2MB, mp4)
03

HIGHLIGHTS.

  • Discrete features of song can be functionally mapped to discrete wing muscles.

  • Wing motor pathways may provide a “modular” anatomical substrate for song evolution.

Acknowledgments

We thank B. Arthur, B. Baker, M. Dickinson, D. Mellert, and T. Morita for helpful discussions; B. Arthur for assembling the video-audio apparatus; B. Arthur and F. Midgley for providing data acquisition and visualization software; B. Pfeiffer for pJFRC39, pJFRC56 and R57C10-Gal80-6 flies; G. Rubin for R14B02- and R48F07-LexA::p65; B. Baker and C. Robinett for fly stocks; J-C. Kao and T. Laverty for fly husbandry; and A. Howard for administrative assistance. This work was supported by the Howard Hughes Medical Institute.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Andersson M. Sexual Selection. Princeton University Press; 1994. [Google Scholar]
  2. Arthur BJ, Sunayama-Morita T, Coen P, Murthy M, Stern DL. Multi-channel acoustic recording and automated analysis of Drosophila courtship songs. BMC biology. 2013;11:11. doi: 10.1186/1741-7007-11-11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Baines RA, Uhler JP, Thompson A, Sweeney ST, Bate M. Altered electrical properties in Drosophila neurons developing without synaptic transmission. J Neurosci. 2001;21:1523–1531. doi: 10.1523/JNEUROSCI.21-05-01523.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Billeter JC, Rideout EJ, Dornan AJ, Goodwin SF. Control of male sexual behavior in Drosophila by the sex determination pathway. Current biology : CB. 2006;16:R766–776. doi: 10.1016/j.cub.2006.08.025. [DOI] [PubMed] [Google Scholar]
  5. Chatterjee SS, Uppendahl LD, Chowdhury MA, Ip PL, Siegal ML. The female-specific doublesex isoform regulates pleiotropic transcription factors to pattern genital development in Drosophila. Development. 2011;138:1099–1109. doi: 10.1242/dev.055731. [DOI] [PubMed] [Google Scholar]
  6. Christiansen AE, Keisman EL, Ahmad SM, Baker BS. Sex comes in from the cold: the integration of sex and pattern. Trends in genetics : TIG. 2002;18:510–516. doi: 10.1016/s0168-9525(02)02769-5. [DOI] [PubMed] [Google Scholar]
  7. Clyne JD, Miesenbock G. Sex-specific control and tuning of the pattern generator for courtship song in Drosophila. Cell. 2008;133:354–363. doi: 10.1016/j.cell.2008.01.050. [DOI] [PubMed] [Google Scholar]
  8. Dickinson MH, Tu MS. The Function of Dipteran Flight Muscle. Comp Biochem Physiol. 1997;116A:223–238. [Google Scholar]
  9. Eisel U, Jarausch W, Goretzki K, Henschen A, Engels J, Weller U, Hudel M, Habermann E, Niemann H. Tetanus toxin: primary structure, expression in E. coli, and homology with botulinum toxins. The EMBO journal. 1986;5:2495–2502. doi: 10.1002/j.1460-2075.1986.tb04527.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Ewing AW. The Neuromuscular Basis of Courtship Song in Drosophila: The Role of the Indirect Flight Muscles. J Comp Physiol. 1977;119:249–265. [Google Scholar]
  11. Ewing AW. The Neuromuscular Basis of Courtship Song in Drosophila: The Role of the Direct and Axillary Wing Muscles. J Comp Physiol. 1979;130:87–93. [Google Scholar]
  12. Gerhardt HC, Huber F. Acoustic communication in insects and anurans : common problems and diverse solutions. Chicago: University of Chicago Press; 2002. [Google Scholar]
  13. Harris RM. Form and function of the secondary hemilineages in the adult Drosophila thoracic nervous system. Seattle: University of Washington; 2012. [Google Scholar]
  14. Hoikkala A. Inheritance of male sound characteristics in Drosophila species. Boca Raton, FL: Taylor & Francis; 2005. [Google Scholar]
  15. Jenett A, Rubin GM, Ngo TT, Shepherd D, Murphy C, Dionne H, Pfeiffer BD, Cavallaro A, Hall D, Jeter J, et al. A GAL4-driver line resource for Drosophila neurobiology. Cell Reports. 2012;2:991–1001. doi: 10.1016/j.celrep.2012.09.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kelley DB, Tobias ML. In: In The design of animal communication. Hauser M, Konishi M, editors. Cambridge: MIT Press; 1999. pp. 9–35. [Google Scholar]
  17. Kimura K, Hachiya T, Koganezawa M, Tazawa T, Yamamoto D. Fruitless and doublesex coordinate to generate male-specific neurons that can initiate courtship. Neuron. 2008;59:759–769. doi: 10.1016/j.neuron.2008.06.007. [DOI] [PubMed] [Google Scholar]
  18. Kohatsu S, Koganezawa M, Yamamoto D. Female contact activates male-specific interneurons that trigger stereotypic courtship behavior in Drosophila. Neuron. 2011;69:498–508. doi: 10.1016/j.neuron.2010.12.017. [DOI] [PubMed] [Google Scholar]
  19. Markow TA, O’Grady PM. Evolutionary genetics of reproductive behavior in Drosophila: connecting the dots. Annual review of genetics. 2005;39:263–291. doi: 10.1146/annurev.genet.39.073003.112454. [DOI] [PubMed] [Google Scholar]
  20. McKeown M, Belote JM, Boggs RT. Ectopic expression of the female transformer gene product leads to female differentiation of chromosomally male Drosophila. Cell. 1988;53:887–895. doi: 10.1016/s0092-8674(88)90369-8. [DOI] [PubMed] [Google Scholar]
  21. Nachtigall W, Wilson DM. Neuro-muscular control of dipteran flight. J Exp Biol. 1967;47:77–97. doi: 10.1242/jeb.47.1.77. [DOI] [PubMed] [Google Scholar]
  22. Nern A, Pfeiffer BD, Svoboda K, Rubin GM. Multiple new site-specific recombinases for use in manipulating animal genomes. Proc Natl Acad Sci, USA. 2011;108:14198–14203. doi: 10.1073/pnas.1111704108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Pan Y, Meissner GW, Baker BS. Joint control of Drosophila male courtship behavior by motion cues and activation of male-specific P1 neurons. Proc Natl Acad Sci, USA. 2012;109:10065–10070. doi: 10.1073/pnas.1207107109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Pfeiffer BD, Jenett A, Hammonds AS, Ngo TT, Misra S, Murphy C, Scully A, Carlson JW, Wan KH, Laverty TR, et al. Tools for neuroanatomy and neurogenetics in Drosophila. Proc Natl Acad Sci, USA. 2008;105:9715–9720. doi: 10.1073/pnas.0803697105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Pfeiffer BD, Ngo TT, Hibbard KL, Murphy C, Jenett A, Truman JW, Rubin GM. Refinement of tools for targeted gene expression in Drosophila. Genetics. 2010;186:735–755. doi: 10.1534/genetics.110.119917. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Pringle JWS. Insect Flight. Cambridge: Cambridge University Press; 1957. [Google Scholar]
  27. Rideout EJ, Billeter JC, Goodwin SF. The sex-determination genes fruitless and doublesex specify a neural substrate required for courtship song. Current biology : CB. 2007;17:1473–1478. doi: 10.1016/j.cub.2007.07.047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Robinett CC, Vaughan AG, Knapp JM, Baker BS. Sex and the single cell. II. There is a time and place for sex. PLoS biology. 2010;8:e1000365. doi: 10.1371/journal.pbio.1000365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Rybak F, Aubin T, Moulin B, Jallon JM. Acoustic communication in Drosophila melanogaster courtship: Are pulse- and sine-song frequencies important for courtship success? Can J Zool. 2002;80:987–996. [Google Scholar]
  30. Schlosser G, Wagner GP. Modularity in Development and Evolution. University of Chicago Press; 2004. [Google Scholar]
  31. Spieth HT. Courtship behavior in Drosophila. Annual review of entomology. 1974;19:385–405. doi: 10.1146/annurev.en.19.010174.002125. [DOI] [PubMed] [Google Scholar]
  32. Stockinger P, Kvitsiani D, Rotkopf S, Tirian L, Dickson BJ. Neural circuitry that governs Drosophila male courtship behavior. Cell. 2005;121:795–807. doi: 10.1016/j.cell.2005.04.026. [DOI] [PubMed] [Google Scholar]
  33. Sweeney ST, Broadie K, Keane J, Niemann H, O’Kane CJ. Targeted expression of tetanus toxin light chain in Drosophila specifically eliminates synaptic transmission and causes behavioral defects. Neuron. 1995;14:341–351. doi: 10.1016/0896-6273(95)90290-2. [DOI] [PubMed] [Google Scholar]
  34. Trimarchi JR, Schneiderman AM. The motor neurons innervating the direct flight muscles of Drosophila melanogaster are morphologically specialized. J Comp Neurol. 1994;340:427–443. doi: 10.1002/cne.903400311. [DOI] [PubMed] [Google Scholar]
  35. von Philipsborn AC, Liu T, Yu JY, Masser C, Bidaye SS, Dickson BJ. Neuronal control of Drosophila courtship song. Neuron. 2011;69:509–522. doi: 10.1016/j.neuron.2011.01.011. [DOI] [PubMed] [Google Scholar]
  36. Weber JN, Peterson BK, Hoekstra HE. Discrete genetic modules are responsible for complex burrow evolution in Peromyscus mice. Nature. 2013;493:402–405. doi: 10.1038/nature11816. [DOI] [PubMed] [Google Scholar]
  37. Wisser A, Nachtigall W. Functional-morphological investigations on the flight muscles and their insertion points in the blowfly Calliphora erthrocephala (Insecta, Diptera) Zoomorph. 1984;104:188–195. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS543888-supplement-01.pdf (1.5MB, pdf)
02
Download video file (20.2MB, mp4)
03
NIHMS543888-supplement-03.pdf (3.9MB, pdf)

RESOURCES