Female-specific Ilp7 neuropeptide-expressing motoneurons (FS-Ilp7 motoneurons) are required in Drosophila for oviduct function in egg laying. Here, we uncover cellular and genetic mechanisms underlying their female-specific generation. We demonstrate that programmed cell death (PCD) eliminates FS-Ilp7 motoneurons in males, and that this requires male-specific splicing of the sex-determination gene fruitless (fru) into the FruMC isoform. However, in females, fru alleles that only generate FruM isoforms failed to kill FS-Ilp7 motoneurons. This blockade of FruM-dependent PCD was not attributable to doublesex gene function but to a non-canonical role for transformer (tra), a gene encoding the RNA splicing activator that regulates female-specific splicing of fru and dsx transcripts. In both sexes, we show that Tra prevents PCD even when the FruM isoform is expressed. In addition, we found that FruMC eliminated FS-Ilp7 motoneurons in both sexes, but only when Tra was absent. Thus, FruMC-dependent PCD eliminates female-specific neurons in males, and Tra plays a double-assurance function in females to establish and reinforce the decision to generate female-specific neurons.

Males and females of most species exhibit distinct behavioral repertoires that are genetically hardwired within dimorphic neuronal circuits. Studies in Drosophila have provided important insight into the genetic pathways that underlie the construction of sexually dimorphic neural circuits (Manoli et al., 2013; Villella and Hall, 2008). Drosophila sexual identity arises from sensing the number of X chromosomes (Erickson and Quintero, 2007). In females, this leads to female-specific expression of functional RNA splicing factor proteins, Sex lethal (Sxl) and Transformer (Tra) (Salz and Erickson, 2010). Tra drives alternative pre-mRNA splicing of transcripts for two transcription factor genes, fruitless (fru) and doublesex (dsx), leading to the expression of the DsxF isoform and a fruF transcript encoding a premature stop codon. The absence of Tra protein in males allows for default male-specific splicing of these targets, leading to male-specific expression of functional FruM and DsxM isoforms (Burtis and Baker, 1989; Ryner et al., 1996). These male and female isoforms are expressed in partially overlapping neuronal lineages and postmitotic neurons, and function as effectors that direct dimorphic neuronal development in the male and female nervous systems (Burtis and Baker, 1989; Heinrichs et al., 1998; Hoshijima et al., 1991; Nagoshi et al., 1988; Ryner et al., 1996; Villella and Hall, 2008).

How FruM and DsxM isoforms construct male-specific circuitry has been intensely studied. FruM and DsxM have mostly non-overlapping roles in generating male-specific neurons, and in neuronal function, morphology and connectivity (Billeter et al., 2006; Cachero et al., 2010; Kimura et al., 2008; Rideout et al., 2007, 2010; Sanders and Arbeitman, 2008; Yu et al., 2010). The specific activities of FruM are determined by which of the four isoforms (FruMA-MD) are expressed, which differ only in their C-terminal DNA-binding zinc-finger domains. FruMA, FruMB and FruMC are expressed in partially overlapping domains in the nervous system, and have unique activities in shaping male nervous system development, whereas FruMD is not thought to be expressed in the CNS (Billeter et al., 2006; Dalton et al., 2013; Ito et al., 1996, 2016; Meissner et al., 2016; Neville et al., 2014; Nojima et al., 2014; Ryner et al., 1996; Soller et al., 2006; Usui-Aoki et al., 2000; von Philipsborn et al., 2014).

Recently, attention has increasingly turned to the study of female-specific behaviors and physiology. Numerous underlying circuits have now been identified (Bussell et al., 2014; Castellanos et al., 2013; Feng et al., 2014; Ferveur, 2010; Gligorov et al., 2013; Kapelnikov et al., 2008; Kimura et al., 2015; Laturney and Billeter, 2014; Rezával et al., 2014, 2012; Yang et al., 2009; Zhou et al., 2014). Yet, an understanding of how female-specific neurons and circuits are generated lags behind that of males. For example, female-specific neurons have been identified (Castellanos et al., 2013; Feng et al., 2014; Kimura et al., 2015; Rezával et al., 2014; Zhou et al., 2014), but the mechanisms leading to the generation of female-specific neuronal populations are only starting to be defined (Kimura et al., 2015). Here, we undertake a genetic analysis to determine the cellular and genetic mechanisms that generate female-specific Insulin-like peptide 7-expressing (FS-Ilp7) oviduct motoneurons, which are required for egg-laying and female fertility (Fig. 1) (Castellanos et al., 2013).

Pertinent lessons in how female-specific neurons may arise can be drawn from studies in males. First, differences in marker expression can be perceived as differences in neuronal number. For example, Lgr3 is directly downregulated by FruM, and differentially regulated by DsxF in discrete neuronal populations resulting in sex-specific marker expression (Meissner et al., 2016). Second, male-specific neurons have been shown to be generated by virtue of female-specific programmed cell death (PCD) or enhanced proliferation of neuronal progenitors in males. For example, in certain neuronal lineages, DsxF can trigger PCD in females and DsxM can trigger progenitor proliferation in males (Birkholz et al., 2013; Sanders and Arbeitman, 2008; Taylor and Truman, 1992). In addition, DsxM and FruM can both block PCD in certain neuronal lineages in males (Kimura et al., 2005; Sanders and Arbeitman, 2008). Applying such lessons, we postulate that either DsxF promotes female-specific progenitor proliferation or that DsxM might trigger male-specific PCD. Indeed, a recent report shows that DsxM triggers PCD to reduce the number of pMN2 neurons in males relative to that in females (Kimura et al., 2015). However, roles for Tra and Fru have not been explored in relation to the generation of female-specific neurons.

We reported that Tra protein expression is necessary in females and sufficient in males for the presence of FS-Ilp7 motoneurons (Castellanos et al., 2013). Examining sex determination gene expression in these motoneurons, we found that they do not express DsxF protein; using dsx-GAL4 to immortalize lineage marker expression, we found that FS-Ilp7 motoneurons never became marked by this lineage marker. Thus, the dsx gene was postulated to not contribute to FS-Ilp7 motoneuron generation (Castellanos et al., 2013). In contrast, the sex-specifically spliced fru transcript (from the P1 promoter) is expressed in FS-Ilp7 motoneurons. In testing a role for FruM, we found that FruM protein expression was required for the absence of FS-Ilp7 motoneuron in males. However, in females expressing FruM, generated from the fruM allele that only generates FruM isoforms, we did not observe any loss of FS-Ilp7 motoneurons. This left the precise function of fru unresolved and suggested that an additional factor is required. Moreover, it was unclear whether the apparent absence of FS-Ilp7 motoneurons in males was merely an artifact of absent Ilp7 protein expression in those neurons, or indeed whether these neurons are unique to females by virtue of enhanced neuroblast proliferation in females or programmed cell death in males.

Here, we have performed cellular and genetic analyses to define roles for tra and fru in female-specific FS-Ilp7 motoneuron generation. We find that the dimorphic emergence of FS-Ilp7 motoneurons in females is due to selective PCD in males. Moreover, we find that the FruMC isoform is required for PCD. By examining the emergence of FS-Ilp7 motoneurons in a fru allelic series in both males and females, we confirm that FruM only triggers FS-Ilp7 motoneuron elimination in males. In exploring this insufficiency in females, we found that Tra can prevent PCD of FS-Ilp7 motoneurons, even in the presence of fru alleles that can only generate FruM isoforms. Further genetic analysis indicated that this function is not likely mediated through Tra-dependent DsxF activity. These data provide evidence that Tra antagonizes the function of FruM in a pathway running parallel or downstream to its role in fru splicing, in an apparent double-assurance function to ensure the absence of FruM activity in subsets of female neurons.

Male-specific PCD of FS-Ilp7 motoneurons

Previously, we identified a set of female-specific (FS) Ilp7 motoneurons that are generated by post-embryonic neurogenesis. These motoneurons cluster around a set of embryonically born Ilp7-expressing neurons that innervate the gut (Castellanos et al., 2013) (Fig. 1A-D). These two neuronal populations can be discriminated by their differential expression of fork head (fkh) and fru. FS-Ilp7 motoneurons do not express the transcription factor, Fkh, but do express fruP1-GAL4 (referred to as fruP1>), a reporter for the sex-specifically spliced fru transcript transcribed from the P1 promoter (Stockinger et al., 2005). In contrast, embryonic Ilp7-neurons express Fkh but not fruP1>GFP (Castellanos et al., 2013) (Fig. 1B,C).

Fig. 1.

A subset of Insulin-like peptide 7 (Ilp7)-expressing motoneurons is female specific. (A) Cartoon of the adult Drosophila central nervous system (CNS), showing the abdominal ganglion (Abg) (highlighted, red box) and the female reproductive tract. The female-specific Ilp7 motoneurons (FS-Ilp7 motoneurons) reside in the Abg and provide motor input to the oviduct (green). (B,C) FS-Ilp7 motoneurons are present in females (C, arrowheads) but not in males (B). FS-Ilp7 motoneurons are uniquely identifiable in females by the coincidence of Ilp7 immunoreactivity (α-Ilp7; magenta) and fruP1-GAL4, UAS-GFP (fruP1>GFP; green) reporter activity. Embryonic Ilp7 neurons are marked by α-Ilp7 and α-Fork head immunoreactivity (α-Fkh, blue). (D) Cartoon summary and marker profile of embryonic (blue) and post-embryonic Ilp7 neurons (green, FS-Ilp7 motoneurons; yellow, dorsal Ilp7-motoneurons) in males and females. V, ventral; D, dorsal; A, anterior; P, posterior.

Fig. 1.

A subset of Insulin-like peptide 7 (Ilp7)-expressing motoneurons is female specific. (A) Cartoon of the adult Drosophila central nervous system (CNS), showing the abdominal ganglion (Abg) (highlighted, red box) and the female reproductive tract. The female-specific Ilp7 motoneurons (FS-Ilp7 motoneurons) reside in the Abg and provide motor input to the oviduct (green). (B,C) FS-Ilp7 motoneurons are present in females (C, arrowheads) but not in males (B). FS-Ilp7 motoneurons are uniquely identifiable in females by the coincidence of Ilp7 immunoreactivity (α-Ilp7; magenta) and fruP1-GAL4, UAS-GFP (fruP1>GFP; green) reporter activity. Embryonic Ilp7 neurons are marked by α-Ilp7 and α-Fork head immunoreactivity (α-Fkh, blue). (D) Cartoon summary and marker profile of embryonic (blue) and post-embryonic Ilp7 neurons (green, FS-Ilp7 motoneurons; yellow, dorsal Ilp7-motoneurons) in males and females. V, ventral; D, dorsal; A, anterior; P, posterior.

We examined whether male-specific programmed cell death (PCD) explains the absence of FS-Ilp7 motoneurons in males, as opposed to an absence of anti-Ilp7 staining in the male counterparts of the motoneurons or enhanced proliferation in females. We expressed the anti-apoptotic caspase inhibitor, baculovirus p35 (UAS-p35) (Hay et al., 1994) in fruP1-expressing cells. In females, fruP1>+ controls had one to five FS-Ilp7 motoneurons per fly (3.0±0.4 motoneurons per fly) and this was not significantly changed in fruP1>p35 females (3.4±0.3 motoneurons per fly) (Fig. 2B,D). In contrast, male fruP1>+ controls had zero FS-Ilp7 motoneurons per fly (Fig. 2A,C), but in fruP1>p35 males we observed frequent rescue of one to five FS-Ilp7 motoneurons per VNC (2.7±0.4 per fly) (Fig. 2A,C). Thus, male-specific PCD within fruP1-expressing cells accounts for the absence of FS-Ilp7 motoneurons in males and that their generation is female specific.

Fig. 2.

FS-Ilp7 motoneurons are eliminated by programmed cell death in males. (A-D) We expressed the baculovirus p35 caspase blocker (UAS-p35), which blocks programmed cell death (PCD), in cells expressing fruP1-GAL4, Ilp7-GAL4 or the pan-neuronal GAL4 drivers elavC155-GAL4 and nsyb-GAL4. We quantified FS-Ilp7 motoneuron numbers in both sexes. (A,B) Representative z-projections of FS-Ilp7 motoneurons in males (A) and females (B), in control (fruP1>+, left panels) and fruP1>p35 (right panels). FS-Ilp7 motoneurons were not observed in wild-type males (fruP1>+) but were always observed in wild-type females (fruP1>+) (arrows, Ilp7+/Fkh-/fruP1+). In the fruP1>p35 genotype, FS-Ilp7 motoneurons were generated in males (A, right panel, arrowheads), but were not produced at increased numbers in females (B, right panel, arrowheads). (C,D) Quantification of FS-Ilp7 motoneuron numbers per fly (each point in scatter plot) of each sex and of each genotype (shown along the x-axis). FS-Ilp7 motoneurons were generated only in males of the fruP1>p35 genotype. No difference in the number of FS-Ilp7 motoneurons was observed in females of the genotypes tested. Data are mean±s.e.m. ****P<0.0001 compared with fruP1>+ control.

Fig. 2.

FS-Ilp7 motoneurons are eliminated by programmed cell death in males. (A-D) We expressed the baculovirus p35 caspase blocker (UAS-p35), which blocks programmed cell death (PCD), in cells expressing fruP1-GAL4, Ilp7-GAL4 or the pan-neuronal GAL4 drivers elavC155-GAL4 and nsyb-GAL4. We quantified FS-Ilp7 motoneuron numbers in both sexes. (A,B) Representative z-projections of FS-Ilp7 motoneurons in males (A) and females (B), in control (fruP1>+, left panels) and fruP1>p35 (right panels). FS-Ilp7 motoneurons were not observed in wild-type males (fruP1>+) but were always observed in wild-type females (fruP1>+) (arrows, Ilp7+/Fkh-/fruP1+). In the fruP1>p35 genotype, FS-Ilp7 motoneurons were generated in males (A, right panel, arrowheads), but were not produced at increased numbers in females (B, right panel, arrowheads). (C,D) Quantification of FS-Ilp7 motoneuron numbers per fly (each point in scatter plot) of each sex and of each genotype (shown along the x-axis). FS-Ilp7 motoneurons were generated only in males of the fruP1>p35 genotype. No difference in the number of FS-Ilp7 motoneurons was observed in females of the genotypes tested. Data are mean±s.e.m. ****P<0.0001 compared with fruP1>+ control.

We then examined whether PCD occurs prior to the onset of Ilp7 expression. We have not identified other discriminatory markers for FS-Ilp7 motoneurons or their progenitors; therefore, the onset of Ilp7 expression provides the earliest unambiguous marker for these motoneurons. We generated a transgenic reporter for Ilp7 expression, Ilp7-nls.tdTomato (see Materials and Methods), that provided the earliest robust marker for Ilp7 expression. We verified that Ilp7-nls.tdTomato is expressed in all abdominal Ilp7 motoneurons, using a published anti-Ilp7 antibody (Yang et al., 2008) and a newly generated anti-Ilp7 antibody (see Materials and Methods) (Fig. S1). Using Ilp7-nls.tdTomato, we first detected FS-Ilp7 motoneurons between 41 and 46 h after puparium formation (APF), in cells that express fruP1>GFP at a high level (Fig. S2A-C′). In males, Ilp7-nls.tdTomato was only seen in the embryonically born Ilp7 neurons, and was never seen in any adjacent cells that could represent FS-Ilp7 motoneurons prior to PCD. In fruP1>p35 males, Ilp7-nls.tdTomato expression in surviving FS-Ilp7 motoneurons cells was first observed at 50-55 h APF (Fig. S2D,E). Thus, PCD occurs prior to the onset of Ilp7 expression in FS-Ilp7 motoneurons. In agreement, we found that driving UAS-p35 from Ilp7-GAL4 failed to prevent the elimination of FS-Ilp7 motoneurons in males (Fig. 2C). We also tested whether FS-Ilp7 motoneurons could be rescued from PCD by expressing UAS-p35 from early postmitotic pan-neuronal drivers elavC155-GAL4 (Berger et al., 2007; Lin and Goodman, 1994) and nsyb-GAL4 (Pauli et al., 2008) (Fig. 2C). In both cases, PCD was not blocked, indicating that PCD must occur before these GAL4 drivers are expressed or before they can drive sufficient levels of p35 to rescue cells from PCD. No significant difference in the number of FS-Ilp7 motoneurons was observed in any of these genotypes in females. We note that one FS-Ilp7 motoneuron survived in one elav>p35 male. This may provide some evidence that PCD occurs at a time of low level elavC155-GAL4 expression; however, this GAL4 line is expressed in subsets of neuronal and glial progenitors and then robustly in postmitotic motoneurons, so this does not fully resolve whether PCD occurs in postmitotic motoneurons (Berger et al., 2007). The identification of markers that allow us to image lineage progression will be required to precisely determine the developmental stage of PCD.

FruMC is necessary for cell death of FS-Ilp7 motoneurons in males

We previously reported that Tra was necessary in females and sufficient in males for FS-Ilp7 motoneuron survival. We also found that FruM was necessary for FS-Ilp7 motoneuron elimination in males, but that FruM in females (generated from a hemizygous fruM allele that constitutively splices fru in the male mode at the endogenous locus in fruP1-expressing cells) failed to eliminate FS-Ilp7 motoneurons in females (Castellanos et al., 2013). This discrepancy was intriguing, because in females the constitutive male-splicing fruM allele often masculinizes neuronal morphology and function (Cachero et al., 2010; Demir and Dickson, 2005).

As our previous conclusions regarding fru function in PCD came from the analysis of single fru heteroallelic genotypes in either sex, we wished to extend the genetic analysis of fru in FS-Ilp7 generation to a more extensive fru allelic series. We generated heteroallelic combinations of fru by combining one of four fru alleles that each reduces or eliminates FruM protein expression (fru3, fruF, fruP1-GAL4, fruSat15, see Materials and Methods for details), with either wild-type fru (+) or an engineered fru allele that only generates either functional male FruM isoforms (the fruM allele) or female-spliced transcripts that do not generate functional Fru protein (the fruF allele) (Demir and Dickson, 2005). The nature of these alleles is depicted in Fig. 3A. In males, we found that any genotype in which FruM is severely reduced or eliminated resulted in significant survival of FS-Ilp7 motoneurons; this averaged between 4.2 and 6.5 FS-Ilp7 motoneurons per fly (Fig. 3B), with the fruF/fru3 genotype being the highest at 6.5±0.5. In contrast, genotypes with a single copy of either fru+ or the fruM allele all resulted in a significant reduction in FS-Ilp7 motoneuron number (averaging 0.06 to 1.1 FS-Ilp7 motoneurons per fly) (Fig. 3B). Thus, by testing numerous fru heteroallelic combinations, we confirmed that FruM is required for PCD of FS-Ilp7 motoneurons in males. In a subset of the genotypes where only a single allele generates fru+ or fruM, we did observe occasional survival of FS-Ilp7 motoneurons in a few animals, and often slightly higher numbers in fruM than in fru+. We interpret this as being due to the escape of some cells from PCD when FruM levels are lowered.

Fig. 3.

FruMC is necessary to eliminate FS-Ilp7 motoneurons in males. (A) Schematic of the fru locus (not to scale) showing the fru alleles used (see also Materials and Methods), the fru promoters (white boxes), the exons (colored boxes), as well as splicing (solid lines) and alternate splicing (broken lines). P1 transcripts are sex-specifically spliced, as shown, so that male isoforms do not include an early stop codon, but female isoforms do include the stop codon. Four FruM protein isoforms are generated (FruMA, FruMB, FruMC and FruMD) by alternate use of exons A-D (colored boxes). (B,C) We tested a role for FruM and its isoforms in male FS-Ilp7 motoneuron PCD. (B) We placed alleles that generate only male-specific splicing (M, fruM), female-specific splicing (F, fruF) or a control chromosome (+, w1118) over a series of alleles that prevent/reduce FruM protein expression (fruF, fruP1>, fru3 and fruSat15). In these genotypes (shown along the x axis), we counted FS-Ilp7 motoneurons per fly and present these as mean±s.e.m. FS-Ilp7 motoneurons were rarely observed in males with one or more FruM-expressing alleles (filled circles). However, four to six FS-Ilp7 motoneurons were observed in genotypes that had no FruM protein expression (unfilled circles). (C) We tested which FruM isoforms are required for PCD. We placed nonsense FruM isoform mutants (fruΔA, fruΔB and fruΔC) over either fruF or fruP1> alleles, and counted the number of FS-Ilp7 motoneurons, represented as scatter plots and shown as mean±s.e.m. FS-Ilp7 motoneurons survived in fruΔC heteroallelic combinations in comparable numbers to wild-type females, showing that only FruMC is required for PCD. Significant differences are shown compared with pertinent controls (+); ****P<0.0001.

Fig. 3.

FruMC is necessary to eliminate FS-Ilp7 motoneurons in males. (A) Schematic of the fru locus (not to scale) showing the fru alleles used (see also Materials and Methods), the fru promoters (white boxes), the exons (colored boxes), as well as splicing (solid lines) and alternate splicing (broken lines). P1 transcripts are sex-specifically spliced, as shown, so that male isoforms do not include an early stop codon, but female isoforms do include the stop codon. Four FruM protein isoforms are generated (FruMA, FruMB, FruMC and FruMD) by alternate use of exons A-D (colored boxes). (B,C) We tested a role for FruM and its isoforms in male FS-Ilp7 motoneuron PCD. (B) We placed alleles that generate only male-specific splicing (M, fruM), female-specific splicing (F, fruF) or a control chromosome (+, w1118) over a series of alleles that prevent/reduce FruM protein expression (fruF, fruP1>, fru3 and fruSat15). In these genotypes (shown along the x axis), we counted FS-Ilp7 motoneurons per fly and present these as mean±s.e.m. FS-Ilp7 motoneurons were rarely observed in males with one or more FruM-expressing alleles (filled circles). However, four to six FS-Ilp7 motoneurons were observed in genotypes that had no FruM protein expression (unfilled circles). (C) We tested which FruM isoforms are required for PCD. We placed nonsense FruM isoform mutants (fruΔA, fruΔB and fruΔC) over either fruF or fruP1> alleles, and counted the number of FS-Ilp7 motoneurons, represented as scatter plots and shown as mean±s.e.m. FS-Ilp7 motoneurons survived in fruΔC heteroallelic combinations in comparable numbers to wild-type females, showing that only FruMC is required for PCD. Significant differences are shown compared with pertinent controls (+); ****P<0.0001.

Three FruM isoforms, FruMA, FruMB and FruMC, are expressed in the male nervous system. These differ in their C2H2 zinc-finger DNA-binding domains (Billeter et al., 2006; Dalton et al., 2013; Ito et al., 1996; Neville et al., 2014; Nojima et al., 2014; Ryner et al., 1996; Usui-Aoki et al., 2000; von Philipsborn et al., 2014). We wished to test which isoform(s) were necessary for male-specific PCD of FS-Ilp7 motoneurons. To first test which isoforms may be required for PCD, we blocked PCD with fruP1>p35 and assayed which isoforms are expressed in ‘undead’ FS-Ilp7 motoneurons. In these animals, we took advantage of the fruAmyc, fruBmyc and fruCmyc alleles, which express a functional Myc-tagged version of each FruM isoform (von Philipsborn et al., 2014). We found that ‘undead’ male FS-Ilp7 motoneurons expressed FruMB and FruMC, but not FruMA (Fig. S3).

We next tested which isoform is necessary for male PCD using isoform-specific mutants that contain a premature stop codon within one of the distinct 3′ exons, referred to as fruΔA, fruΔB and fruΔC (Fig. 3A,C) (Billeter et al., 2006; Neville et al., 2014). We combined these with the fruF or fruP1-GAL4 alleles that severely reduce FruM expression. In males, FS-Ilp7 motoneurons were mostly absent in fruΔA or fruΔB heteroallelic genotypes, confirming that FruMA and FruMB are not necessary for PCD. In fruΔC heteroallelic genotypes, however, we observed robust survival of FS-Ilp7 motoneurons (4.1±0.5 and 4.5±0.4 FS-Ilp7 motoneurons for fruΔC/fruF and fruΔC/fruP1-GAL4, respectively) (Fig. 3C). Therefore, the FruMC isoform is necessary for PCD.

FruM in females is insufficient to kill Ilp7 motoneurons

In numerous cases where FruM is necessary for a male-specific neuronal property or function, its expression in females often masculinizes corresponding neurons in females (Cachero et al., 2010; Demir and Dickson, 2005; Rideout et al., 2010; von Philipsborn et al., 2014). In contrast, we had shown that females hemizygous for the fruM allele failed to eliminate FS-Ilp7 motoneurons (Castellanos et al., 2013). Here, we re-examined this conclusion by testing an expanded fru allelic series. Importantly, we found that FS-Ilp7 motoneurons were not eliminated in any genotype that generates FruM protein in the female. Even in fruM/fruΔtra females where both alleles produce only FruM protein, irrespective of Tra activity (Demir and Dickson, 2005), we observed 3.0±0.3 FS-Ilp7 motoneurons, which was not significantly different from w1118 controls (Fig. 4A). Only in one case did we find a modest but significant FruM-driven reduction in FS-Ilp7 motoneurons, when comparing fruM/fruF (3.2±0.4 motoneurons per fly) with fruF/fruF (5.0±0.3 motoneurons per fly) (Fig. 4A). This could support the notion that the presence of FruM may subtly but incompletely increase the likelihood of PCD in females. Nevertheless, the weight of evidence demonstrates that FruM expression in females is not sufficient to eliminate FS-Ilp7 motoneurons in females, as it is in males.

Fig. 4.

Tra blocks PCD of FS-Ilp7 motoneurons genetically downstream of fru splicing. (A-C) We tested whether constitutive male-type of fru male mode is sufficient for PCD of FS-Ilp7 motoneurons in females, and tested genetic interactions between tra and fru that lead to FS-Ilp7 motoneuron survival in males and females. (A) We quantified FS-Ilp7 motoneurons per female in a fru allelic series (similar to Fig. 3) showing mean±s.e.m. per genotype (filled circles, FruM protein-expressing genotypes; unfilled circles, genotypes that cannot express FruM). There was no loss of FS-Ilp7 motoneurons in any genotype tested. (B) In males, ectopic Tra expression (fruP1>traF) led to survival of FS-Ilp7 motoneurons in most animals, whereas survival was rare in controls (fruP1>+ or UAS-traF). Introduction of a constitutive male-splicing allele fruM in this background (fruM/fruP1>traF) did not decrease FS-Ilp7 motoneuron number in comparison with fruP1>traF. (C) We knocked down Tra using RNAi (elav>Dcr2;tradsRNAi;fru+/+). This eliminated all FS-Ilp7 motoneurons. We then tested whether FruM protein expression is specifically required for PCD in the absence of Tra. Therefore, we prevented male splicing of fru in the tra RNAi background (elav>Dcr2;tradsRNAi) and observed a dose response of fru in killing FS-Ilp7 motoneurons. Expressing two copies of fru (fru+/M) that can each generate FruM protein expression eliminated FS-Ilp7 motoneurons in most animals, with only two animals retaining one or two FS-Ilp7 motoneurons, suggesting that the fruM allele may generate less FruM protein than the fru+ wild-type allele. One copy of fru available for FruM protein expression led to partial survival (fru+/F); no copies of fru available for FruM protein expression (fruF/F) led to full survival of FS-Ilp7 motoneurons. Preventing expression of the FruMC protein isoform (fruΔC/F) also led to full FS-Ilp7 motoneuron rescue. This demonstrated that FruMC protein expression alone is sufficient to elicit PCD in females when Tra is knocked down. Data are FS-Ilp7 motoneurons per fly in scatter plots (mean±s.e.m.). Significant differences within each experimental group are shown compared with the appropriate controls (+); *P<0.05, **P<0.01, ****P<0.0001.

Fig. 4.

Tra blocks PCD of FS-Ilp7 motoneurons genetically downstream of fru splicing. (A-C) We tested whether constitutive male-type of fru male mode is sufficient for PCD of FS-Ilp7 motoneurons in females, and tested genetic interactions between tra and fru that lead to FS-Ilp7 motoneuron survival in males and females. (A) We quantified FS-Ilp7 motoneurons per female in a fru allelic series (similar to Fig. 3) showing mean±s.e.m. per genotype (filled circles, FruM protein-expressing genotypes; unfilled circles, genotypes that cannot express FruM). There was no loss of FS-Ilp7 motoneurons in any genotype tested. (B) In males, ectopic Tra expression (fruP1>traF) led to survival of FS-Ilp7 motoneurons in most animals, whereas survival was rare in controls (fruP1>+ or UAS-traF). Introduction of a constitutive male-splicing allele fruM in this background (fruM/fruP1>traF) did not decrease FS-Ilp7 motoneuron number in comparison with fruP1>traF. (C) We knocked down Tra using RNAi (elav>Dcr2;tradsRNAi;fru+/+). This eliminated all FS-Ilp7 motoneurons. We then tested whether FruM protein expression is specifically required for PCD in the absence of Tra. Therefore, we prevented male splicing of fru in the tra RNAi background (elav>Dcr2;tradsRNAi) and observed a dose response of fru in killing FS-Ilp7 motoneurons. Expressing two copies of fru (fru+/M) that can each generate FruM protein expression eliminated FS-Ilp7 motoneurons in most animals, with only two animals retaining one or two FS-Ilp7 motoneurons, suggesting that the fruM allele may generate less FruM protein than the fru+ wild-type allele. One copy of fru available for FruM protein expression led to partial survival (fru+/F); no copies of fru available for FruM protein expression (fruF/F) led to full survival of FS-Ilp7 motoneurons. Preventing expression of the FruMC protein isoform (fruΔC/F) also led to full FS-Ilp7 motoneuron rescue. This demonstrated that FruMC protein expression alone is sufficient to elicit PCD in females when Tra is knocked down. Data are FS-Ilp7 motoneurons per fly in scatter plots (mean±s.e.m.). Significant differences within each experimental group are shown compared with the appropriate controls (+); *P<0.05, **P<0.01, ****P<0.0001.

Tra prevents FruM-dependent PCD to promote FS-Ilp7 motoneuron survival

We have previously reported that Tra knockdown in females recapitulates male-like elimination of FS-Ilp7 motoneurons (Castellanos et al., 2013). Therefore, Tra must have a function that is independent of fru splicing into female-specific transcripts and explains its necessity and sufficiency for FS-Ilp7 motoneuron survival. We revisited a possible role for dsx downstream of Tra in playing an antagonistic role to FruM activity in females, despite evidence suggesting that Dsx is not expressed in FS-Ilp7 motoneurons or their lineage (Castellanos et al., 2013). We tested dsx nulls and observed no significant change in the number of FS-Ilp7 motoneurons in either males or females (Fig. S4A,B). Thus, dsx is not required for either FS-Ilp7 motoneuron elimination in males (via DsxM expression) or survival in females (via DsxF expression). However, DsxF expression in females has been postulated to antagonistically repress the masculinizing effect of the fruM and fruΔtra alleles on courtship behavior (Shirangi et al., 2006). To test the hypothesis that DsxF may be required to antagonize FruM function in PCD leading to FS-Ilp7 motoneuron survival, we overexpressed UAS-DsxF in male and female fruP1-GAL4-expressing cells (Fig. S4C,D). This did not prevent FS-Ilp7 motoneuron elimination in the majority of males and did not significantly increase FS-Ilp7 motoneuron numbers in females. In one male, overexpressed DsxF did rescue the survival of two FS-Ilp7 motoneurons, which may suggest that excess DsxF has limited capacity to reduce FruM-dependent PCD, or may represent a rare case of survival in a fru heterozygotic genotype as observed in fruP1>+ in Fig. 3B. Regardless, our results do not support an essential role for DsxF as the crucial factor downstream of Tra that mediates Tra-dependent antagonism of FruM-dependent PCD of FS-Ilp7 motoneurons.

We tested a role for Tra itself as a modifier of FruM function, uncoupled from its role in fru transcript splicing. We ectopically expressed UAS-traF from the fruP1-GAL4 driver in males and observed significant survival of FS-Ilp7 motoneurons (2.1±0.4 motoneurons per fly) (Fig. 4B; fruP1>traF). This survival was expected, under the assumption that FruM expression is prevented or reduced in these males. We note that these males did have fewer FS-Ilp7 motoneurons compared with wild-type females (Fig. 4A) and with males that lack FruM expression (Fig. 3B). This could be due to residual FruM expression resulting from insufficient Tra overexpression around the time of PCD. We then tested whether the ability of Tra to prevent PCD in males was mediated through fru transcript splicing, preventing generation of functional FruM protein. We ectopically expressed UAS-TraF in the presence of constitutive FruM generation caused by the fruM allele. Notably, expression of FruM failed to restore PCD in the presence of Tra, as we observed the survival of 2.7±0.3 FS-Ilp7 motoneurons (Fig. 4B; fruP1>traF/fruM). To ensure that ectopic expression of Tra in males did not inadvertently rescue FS-Ilp7 motoneurons through the expression of DsxF, we co-expressed UAS-traF with an effective UAS-dsxRNAi transgene (Hudry et al., 2016) in males and counted the number of FS-Ilp7 motoneurons by their co-expression of Ilp7-nls.tdTomato, anti-Ilp7 and fruP1>nGFP (Fig. S4). We found that co-expressing UAS-Dcr2 with UAS-traF and UAS-dsxRNAi in males generated the same numbers of FS-Ilp7 motoneurons as did expression of UAS-Dcr2 with UAS-traF alone in males (average of 1.4±-0.2 and 1.5±0.4 FS-Ilp7 motoneurons per fly, respectively). These data suggest that Tra can override FruM-dependent PCD in a mechanism unrelated to fru and dsx splicing.

To determine whether the FruM-modifying activity of Tra in males is an artifact of Tra overexpression, we tested the same epistatic relationship in females. We repeated and confirmed previous results (Castellanos et al., 2013) that RNAi-mediated knockdown of Tra in neurons eliminated all FS-Ilp7 motoneurons, using elavGAL4-C155 to drive UAS-Dcr2 and UAS-tradsRNAi (Fig. 4C). We also tested tra null mutant females (traKO/traKO) and found that these also have zero FS-Ilp7 motoneurons (n=7), whereas traKO/TM6B females have 5.3±0.4 FS-Ilp7 motoneurons (n=8) (P<0.0001, unpaired student t-test). To confirm this effect is due to Tra-dependent splicing, we tested whether the obligatory splicing co-factor of Tra, Tra2, is necessary to eliminate FS-Ilp7 motoneurons in females. In tra2 mutant females and males [tra2[B]/Df(2L)trix], there were 0±0 FS-Ilp7 motoneurons (n=13 and n=14, respectively). These results demonstrate that Tra and its splicing co-factor Tra2 are both necessary for FS-Ilp7 motoneuron survival in females.

Next, we tested whether elimination of FS-Ilp7 motoneurons in elav>tradsRNAi females is due to the generation of FruM. To achieve this, we examined the number of FS-Ilp7 motoneurons in genotypes that progressively reduced a dose of FruM in a tra RNAi knockdown background. In elav>tradsRNAi females, 0±0 FS-Ilp7 motoneurons were generated in the presence of two copies of wild-type fru, and 0.2±0.1 FS-Ilp7 motoneurons were generated in the presence of a fru+/M genotype. Next, we found that, in elav>tradsRNAi females, a fru+/F genotype resulted in 1.3±0.2 FS-Ilp7 motoneurons and that a fruF/F genotype resulted in survival of 5.3±0.3 FS-Ilp7 motoneurons (Fig. 4C). These data show that FruM protein is indeed required for the elimination of FS-Ilp7 motoneurons in females, but it can function only in the absence of Tra.

Finally, we tested whether the FruMC isoform is necessary in females to trigger PCD in Tra-deficient females, as in males. In confirmation, we found that in elav >tradsRNAi females, the presence of the fruΔC mutant allele (fruΔC/F) led to survival of 4.0±0.5 FS-Ilp7 motoneurons (Fig. 4C). This demonstrates that FruMC is required for FS-Ilp7 motoneuron PCD in females when Tra is absent. Our results provide evidence for the ability of Tra to function downstream of fru splicing to block the function of FruM once it is generated.

Analysis of fru, dsx and tra function in the Drosophila nervous system has transformed our understanding of the construction of sexually dimorphic neuronal circuits (Yamamoto and Koganezawa, 2013). Owing to the elaborate stereotyped behaviors of males, studies have historically focused on construction of the male nervous system. In contrast, female nervous system construction is less well characterized. Here, we explore the cellular and genetic mechanisms that generate a population of female-specific motoneurons in Drosophila. In addressing this issue, we uncover functions for fru and tra that are of general significance to understanding the development of dimorphic nervous systems, and also the interpretation of genetic studies of these factors.

FruMC removes female-specific neurons from the male nervous system by PCD

Surveys of fruP1-expressing motoneurons have identified many male-specific populations (Cachero et al., 2010; Yu et al., 2010). The increased numbers of fruP1-positive motoneurons in males is largely attributed to sex-specific isoforms of dsx, through their control of PCD or proliferation in neuronal lineages (Kimura, 2011; Kimura et al., 2008; Rideout et al., 2007). FruM is generally considered to be a regulator of the neuronal morphology and function of these motoneurons after periods of PCD or proliferation have established neuronal numbers (Yamamoto and Koganezawa, 2013). However, there is now evidence to implicate specific FruM isoforms in the control of neuronal numbers leading to additional fruP1-expressing neurons in males (von Philipsborn et al., 2014). A mechanistic explanation comes from a study showing that FruM prevents PCD of mAL neurons selectively in males (Kimura et al., 2005). Our results now provide an expanded view of FruM function, showing that the FruMC isoform removes female neuronal components from the male nervous system via PCD (Fig. 5). Typically, between two and six FS-Ilp7 motoneurons are generated per wild-type female or mutant male although we observed genotypes that can generate 8 to 10 FS-Ilp7 motoneurons. The earliest marker available to unambiguously identify these neurons is the Ilp7 reporter itself, which we used to identify these neurons at 41-46 h APF in segments A6 and A7, as predicted by our previous analysis of Hox gene expression in FS-Ilp7 motoneurons (Castellanos et al., 2013). The identification of discriminatory markers earlier than Ilp7 expression to image earlier lineage and postmitotic stages will be important to determine the developmental stage at which fruP1 transcripts are first transcribed and when PCD occurs. In addition, it will be interesting to discover the mechanisms that give rise to the natural diversity of FS-Ilp7 neuronal numbers, and precisely how fru and tra genotypes affect this process. Our data are relevant in light of a recent examination of fruP1-expressing brain neuroblast lineages, showing that blockade of PCD using UAS-p35 increased neuronal number in both sexes; thus, PCD was also proposed to restrict neuronal number in male lineages (Ren et al., 2016). We believe that our observation that FruMC eliminates neurons in males through PCD provides a mechanism to account for neuronal loss in a number of those cases. These findings provide a novel framework that accounts for targeted elimination of neurons in males and the corresponding generation of female-specific neurons and circuitry.

Fig. 5.

Fru and Tra have novel opposing roles in constructing sexually dimorphic FS-Ilp7 motoneurons. In the male nervous system, FruM isoforms A-C are expressed and direct most male-specific differences in behavior and neuronal morphology, whereas in the female nervous system, Tra prevents FruM protein production via alternative splicing. (Left) In males, we found that FruMC is necessary for PCD of FS-Ilp7 motoneurons. (Right) In females, we found that Tra not only prevents fruM splicing, by generating fruF, but genetically acts in parallel or downstream of FruMC to prevent PCD of FS-Ilp7 motoneurons. We propose that this additional role for Tra outside of fru splicing acts as a failsafe to ensure the survival of FS-Ilp7 motoneurons, which are required for egg laying.

Fig. 5.

Fru and Tra have novel opposing roles in constructing sexually dimorphic FS-Ilp7 motoneurons. In the male nervous system, FruM isoforms A-C are expressed and direct most male-specific differences in behavior and neuronal morphology, whereas in the female nervous system, Tra prevents FruM protein production via alternative splicing. (Left) In males, we found that FruMC is necessary for PCD of FS-Ilp7 motoneurons. (Right) In females, we found that Tra not only prevents fruM splicing, by generating fruF, but genetically acts in parallel or downstream of FruMC to prevent PCD of FS-Ilp7 motoneurons. We propose that this additional role for Tra outside of fru splicing acts as a failsafe to ensure the survival of FS-Ilp7 motoneurons, which are required for egg laying.

A novel failsafe mechanism for tra that builds the female nervous system

Our genetic manipulation studies provide evidence that Tra can antagonize FruMC-dependent cell death in both males and females (Fig. 5). This provides a novel perspective for understanding the development of the female nervous system, and for studies in which forced male-specific splicing of fru is used to test the sufficiency of FruM protein activity in females. For example, there have been numerous reports that FruM expression in females is insufficient for full masculinization (Demir and Dickson, 2005; Kimura et al., 2008; Rideout et al., 2007). Males require FruM for the enlargement of numerous brain regions relative to females, but these are only partially enlarged in females with the fruM allele. In contrast, these regions are enlarged in tra mutant females to match males (Cachero et al., 2010). In addition, females with the fruM allele do not exhibit the full FruM-dependent behavioral repertoire of males (Demir and Dickson, 2005; Kimura et al., 2008; Rideout et al., 2007). In contrast, tra mutant females have near full behavioral masculinization in all these behaviors (Kimura et al., 2008; Kyriacou and Hall, 1980; Rideout et al., 2007; Sturtevant, 1945). In both cases, tra mutant females more closely resemble a fully masculinized phenotype than do fruM females. The other arm of the sex-determination cascade, which is regulated by dsx, can account for this discrepancy in numerous cases (Kimura et al., 2008; Rideout et al., 2007). However, our demonstration that Tra also antagonizes FruM functional activity in females independently of dsx gene function in certain cellular contexts offers a novel perspective for studies in which FruM is expressed in females.

What function might such a failsafe role for Tra play? Possibly, it may serve as a safeguard against incomplete splicing of fru (and perhaps also dsx) sex-specific transcripts. Indeed, RNA sequencing has shown that fruM RNA transcripts are in fact generated in wild-type females at a low level, and it is also possible that this may be exacerbated in stressful environments such as high temperature or hypoxia (Ferri et al., 2008; Graveley et al., 2011; Mohr and Hartmann, 2014; Usui-Aoki et al., 2000; Yamamoto, 2007; Yamamoto and Koganezawa, 2013). Tra and Tra2 binding have also been shown to repress reporter translation in S2 cells, suggesting a role for Tra in not only preventing DsxM and FruM expression by activating alternative splicing, but also in directly repressing their translation (Usui-Aoki et al., 2000). The use of such a failsafe in the sex-determination pathway is also seen in males, whereby mir-124 targets tra transcripts for degradation in the male nervous system to ensure the total elimination of Tra protein in males (Weng et al., 2013).

tra has been shown to act independently of fru/dsx to promote female-specific properties of tissues outside of the CNS (Evans and Cline, 2013; Hudry et al., 2016; Rideout et al., 2015). In the fat body, tra is necessary for the non-cell-autonomous increase in growth and body size of females relative to males, in a mechanism that is insensitive to dsx and fru (Rideout et al., 2015). In intestinal stem cells that do not express dsx or fru, tra acts to enhance cellular proliferation to expand tissue size (Hudry et al., 2016). Thus, these studies provide evidence for non-canonical roles of tra outside of the CNS. In this report, we show that tra plays a double assurance role to antagonize FruM isoform function in the female CNS. Thus, our findings add to a growing literature supporting a more expansive role for tra than had long been postulated. Overall, our results lend additional support to an emerging literature that tra and also sxl direct certain sexually dimorphic properties outside of a strictly linear sex-determination cascade that uses fru and dsx as sole effectors (Evans and Cline, 2013; Hudry et al., 2016; Rideout et al., 2015). It will be interesting to explore the mechanisms of action of such additional functions for tra and sxl, and determine the benefits that drove divergence from a strictly linear sex-determination cascade.

Fly stocks

Flies were maintained on standard cornmeal food at 70% humidity at 18°C or 25°C. Ilp7-GAL4 (Castellanos et al., 2013). Strains from Bloomington Drosophila Stock Center were: P{10xUAS-IVS-Syn21-GFP-p10} (referred to as UAS-GFP) (Pfeiffer et al., 2012); P{UAS-p35.H}BH1 (Hay et al., 1994); P{GawB}elavGal-C155 (referred to as elav> and elavC155-GAL4) (Lin and Goodman, 1994); UAS-Dicer2 (Dietzl et al., 2007); dsx1 (amorphic allele) (Ota et al., 1981); Dp(1;Y)BS;cn tra2Bbw1 (amorphic tra2 allele); Df(2L)trix (tra2 deletion) (Goralski et al., 1989); y1v1;P{TRiP.JF02256}attP2 (referred to as UAS-dsxRNAi); and w1118 (referred to as ‘+’). The following alleles were obtained as generous gifts: nsyb-GAL4 (Pauli et al., 2008); UAS-dsxF (Lee et al., 2002); and Df(3R)dsx15 (dsx deficiency) (Baker et al., 1991). tra alleles used were: UAS-traF (Ferveur et al., 1995); UAS-tradsRNAi (Chan and Kravitz, 2007); and traKO (amorphic allele) (Hudry et al., 2016). Putative or reported severe hypomorphs or nulls of fru P1 transcripts include: fruP1-GAL4 (Stockinger et al., 2005), P{PZ}fru3 and Df(3R)fruSat15. The engineered fru alleles that constitutively splice into female- or male-specific isoforms include fruF, fruΔtra and fruM (Demir and Dickson, 2005). The following are FruM isoform-specific nonsense mutants: fruΔA and fruΔB (Neville et al., 2014), and fruΔC (Billeter et al., 2006). The following are Myc-tagged FruM isoform-specific alleles: fruAmyc, fruBmyc and fruCmyc (von Philipsborn et al., 2014).

Generating the Ilp7-nls.tdTomato reporter

To generate the Ilp7-nls.tdTomato reporter, we PCR amplified −2964 to +424 (Ilp7 start codon) relative to the transcriptional start site of the Ilp7 gene. We overlapped the Ilp7 translational start site with a tdTomato ORF (Han et al., 2011) fused at the C-terminus to the Tra nuclear localization signal and SV40-pA sequences from the pHstinger vector (Barolo et al., 2000). This construct was inserted into the psD7-001 vector. Fly transformation by P-element insertion was performed by Best Gene. P-element insertions on the second chromosome were recovered and established as stable fly strains.

Tissue processing and immunohistochemistry

Verification of correct genotypes in adults was determined by loss of balancer chromosomes and/or by evidence of re-sexualization in appropriate genotypes (e.g. chaining behavior and/or changes in abdominal pigmentation and genitalia). Male and female adult VNCs were dissected within 24 h of eclosion. Standard protocols for immunohistochemistry were used (Eade and Allan, 2009). Primary antibodies used were rabbit anti-Ilp7 (1:1000; a gift from E. Hafen, ETH, Zurich, Switzerland); rabbit anti-Ilp7 (1:2000; this study, see below for details); guinea pig anti-Fork head (1:1000; a gift from H. Jäckle, Max Planck Institute, Göttingen, Germany); rat anti-Myc (1:1000; Abcam, JAC6); mouse anti-Elav (1:100; Developmental Studies Hybridoma Bank, University of Iowa). Secondary antibodies used were: anti-rabbit, anti-guinea pig, anti-mouse and anti-rat IgG (H+L) conjugated to DyLight 488, Cy3 or Cy5 (1:400, Jackson ImmunoResearch).

Ilp7 antibody generation

We generated a rabbit polyclonal antibody to the Ilp7 neuropeptide using the antibody generation services of GenScript USA. They synthesized the KLH-conjugated peptide CRSQSDWENVWHQETHS. This peptide sequence was chosen based on the sequence from Yang and colleagues (Yang et al., 2008) (NH2-RSQSDWENVWHQETHS-CONH2), but we added an N-terminal cysteine for KLH conjugation and did not block the C terminus from amidation. GenScript tested the antigen purity by mass spectrometry, immunized rabbits, affinity purified the polyclonal antibody against antigen and tested the antibody titer using ELISA. We tested the polyclonal antibody on Drosophila VNCs and found that it generated immunoreactivity consistent with the anti-Ilp7 antibody generated by Yang and colleagues, localizing to the cytosol of all nuclei labeled using the Ilp7-nls.tdTomato reporter (Fig. S1).

Image and statistical analysis

All images were acquired using an Olympus FV1000 confocal microscope. FS-Ilp7 motoneurons were manually counted using Fluoview Software (FV10-ASW). All representative images in figures were processed using Adobe Photoshop CS6 (identically for all images being compared), and figures were made in Adobe Illustrator CS6. For images collected from fru>GFP genotypes, we lowered the brightness of the green channel for the readers to easily observe the other channels; for Ilp7-Tdtomato genotypes we increased the brightness of the red channel to easily observe expression in FS-Ilp7 motoneurons. Where appropriate, images were false-colored for clarity, and colors were chosen for color-blind readers. All statistical analysis and graphing were performed using Prism 6 software (GraphPad Software). A minimum n=8 flies was used for each genotype studied. All data underwent D'Agostino and Pearson normality testing; data within graphs were compared using one-way ANOVA followed by Tukey post hoc analysis. Statistical differences are shown if P<0.05.

We are very grateful to Stephen Goodwin and Megan Neville (Oxford University, UK) for valuable advice and for the generous gift of fru alleles, to Barry Dickson (Janelia Farms, USA) for critical fru reagents, and to Elizabeth Rideout (University of British Columbia, Canada) for valuable advice. For discussion and/or reagents, we are also grateful to Dr Stefan Thor (Linkoping University, Sweden), Dr Vanessa Auld (University of British Columbia, Canada), Dr Bruce Baker (Janelia Farms, USA), Dr Michael Gordon (University of British Columbia, Canada), Dr Lyubov Veverytsa (University of British Columbia, Canada), Dr Ernst Hafen (ETH, Switzerland), Dr Heinz Jäckel (Max Planck Institute, Germany), Dr Ken-Ichi Kimura (Hokkaido University, Japan), Dr Edward Kravitz (Harvard University, USA), Dr Thomas Merritt (Laurentian University, Canada) and Dr Irene Miguel-Aliaga (Imperial College London, UK).

Author contributions

Conceptualization: S.R.C.G., M.C.C., D.W.A.; Methodology: S.R.C.G., T.L., D.W.A.; Validation: D.W.A.; Formal analysis: S.R.C.G., M.C.C., K.E.B., D.W.A.; Investigation: S.R.C.G., M.C.C., K.E.B., T.L., D.W.A.; Resources: D.W.A.; Data curation: S.R.C.G., M.C.C., K.E.B., D.W.A.; Writing - original draft: S.R.C.G., M.C.C., D.W.A.; Writing - review & editing: S.R.C.G., M.C.C., D.W.A.; Visualization: S.R.C.G., D.W.A.; Supervision: D.W.A.; Project administration: D.W.A.; Funding acquisition: S.R.C.G., M.C.C., D.W.A.

Funding

This work was funded by a Discovery Operating Grant and a Discovery Accelerator Operating Grant from The Natural Sciences And Engineering Research Council of Canada (RGPIN-2014-15095, and RGPAS-462170-14, respectively). M.C.C was a recipient of a Four Year Graduate Studentship from the University of British Columbia. S.R.C.G. was a recipient of the NSERC CGS-M Studentship.

Baker
,
B. S.
,
Hoff
,
G.
,
Kaufman
,
T. C.
,
Wolfner
,
M. F.
and
Hazelrigg
,
T.
(
1991
).
The doublesex locus of Drosophila melanogaster and its flanking regions: a cytogenetic analysis
.
Genetics
127
,
125
-
138
.
Barolo
,
S.
,
Carver
,
L. A.
and
Posakony
,
J. W.
(
2000
).
GFP and beta-galactosidase transformation vectors for promoter/enhancer analysis in Drosophila
.
BioTechniques
29
,
726, 728, 730, 732
.
Berger
,
C.
,
Renner
,
S.
,
Lüer
,
K.
and
Technau
,
G. M.
(
2007
).
The commonly used marker ELAV is transiently expressed in neuroblasts and glial cells in the Drosophila embryonic CNS
.
Dev. Dyn.
236
,
3562
-
3568
.
Billeter
,
J.-C.
,
Villella
,
A.
,
Allendorfer
,
J. B.
,
Dornan
,
A. J.
,
Richardson
,
M.
,
Gailey
,
D. A.
and
Goodwin
,
S. F.
(
2006
).
Isoform-specific control of male neuronal differentiation and behavior in Drosophila by the fruitless gene
.
Curr. Biol.
16
,
1063
-
1076
.
Birkholz
,
O.
,
Rickert
,
C.
,
Berger
,
C.
,
Urbach
,
R.
and
Technau
,
G. M.
(
2013
).
Neuroblast pattern and identity in the Drosophila tail region and role of doublesex in the survival of sex-specific precursors
.
Development
140
,
1830
-
1842
.
Burtis
,
K. C.
and
Baker
,
B. S.
(
1989
).
Drosophila doublesex gene controls somatic sexual differentiation by producing alternatively spliced mRNAs encoding related sex-specific polypeptides
.
Cell
56
,
997
-
1010
.
Bussell
,
J. J.
,
Yapici
,
N.
,
Zhang
,
S. X.
,
Dickson
,
B. J.
and
Vosshall
,
L. B.
(
2014
).
Abdominal-B neurons control Drosophila virgin female receptivity
.
Curr. Biol.
24
,
1584
-
1595
.
Cachero
,
S.
,
Ostrovsky
,
A. D.
,
Yu
,
J. Y.
,
Dickson
,
B. J.
and
Jefferis
,
G. S. X. E.
(
2010
).
Sexual dimorphism in the fly brain
.
Curr. Biol.
20
,
1589
-
1601
.
Castellanos
,
M. C.
,
Tang
,
J. C. Y.
and
Allan
,
D. W.
(
2013
).
Female-biased dimorphism underlies a female-specific role for post-embryonic Ilp7 neurons in Drosophila fertility
.
Development
140
,
3915
-
3926
.
Chan
,
Y.-B.
and
Kravitz
,
E. A.
(
2007
).
Specific subgroups of FruM neurons control sexually dimorphic patterns of aggression in Drosophila melanogaster
.
Proc. Natl. Acad. Sci. USA
104
,
19577
-
19582
.
Cline
,
T. W.
(
1993
).
The Drosophila sex determination signal: how do flies count to two?
Trends Genet.
9
,
385
-
390
.
Dalton
,
J. E.
,
Fear
,
J. M.
,
Knott
,
S.
,
Baker
,
B. S.
,
McIntyre
,
L. M.
and
Arbeitman
,
M. N.
(
2013
).
Male-specific Fruitless isoforms have different regulatory roles conferred by distinct zinc finger DNA binding domains
.
BMC Genomics
14
,
659
.
Demir
,
E.
and
Dickson
,
B. J.
(
2005
).
fruitless splicing specifies male courtship behavior in Drosophila
.
Cell
121
,
785
-
794
.
Dietzl
,
G.
,
Chen
,
D.
,
Schnorrer
,
F.
,
Su
,
K.-C.
,
Barinova
,
Y.
,
Fellner
,
M.
,
Gasser
,
B.
,
Kinsey
,
K.
,
Oppel
,
S.
,
Scheiblauer
,
S.
, et al.
(
2007
).
A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila
.
Nature
448
,
151
-
156
.
Eade
,
K. T.
and
Allan
,
D. W.
(
2009
).
Neuronal phenotype in the mature nervous system is maintained by persistent retrograde bone morphogenetic protein signaling
.
J. Neurosci.
29
,
3852
-
3864
.
Erickson
,
J. W.
and
Quintero
,
J. J.
(
2007
).
Indirect effects of ploidy suggest X chromosome dose, not the X:A ratio, signals sex in Drosophila
.
PLoS Biol.
5
,
e332
.
Evans
,
D. S.
and
Cline
,
T. W.
(
2013
).
Drosophila switch gene Sex-lethal can bypass its switch-gene target transformer to regulate aspects of female behavior
.
Proc. Natl. Acad. Sci. USA
110
,
E4474
-
E4481
.
Feng
,
K.
,
Palfreyman
,
M. T.
,
Hasemeyer
,
M.
,
Talsma
,
A.
and
Dickson
,
B. J.
(
2014
).
Ascending SAG neurons control sexual receptivity of Drosophila females
.
Neuron
83
,
135
-
148
.
Ferri
,
S. L.
,
Bohm
,
R. A.
,
Lincicome
,
H. E.
,
Hall
,
J. C.
and
Villella
,
A.
(
2008
).
fruitless Gene products truncated of their male-like qualities promote neural and behavioral maleness in Drosophila if these proteins are produced in the right places at the right times
.
J. Neurogenet.
22
,
17
-
55
.
Ferveur
,
J.-F.
(
2010
).
Drosophila female courtship and mating behaviors: sensory signals, genes, neural structures and evolution
.
Curr. Opin. Neurobiol.
20
,
764
-
769
.
Ferveur
,
J.-F.
,
Störtkuhl
,
K. F.
,
Stocker
,
R. F.
and
Greenspan
,
R. J.
(
1995
).
Genetic feminization of brain structures and changed sexual orientation in male Drosophila
.
Science
267
,
902
-
905
.
Gligorov
,
D.
,
Sitnik
,
J. L.
,
Maeda
,
R. K.
,
Wolfner
,
M. F.
and
Karch
,
F.
(
2013
).
A novel function for the Hox gene Abd-B in the male accessory gland regulates the long-term female post-mating response in Drosophila
.
PLoS Genet.
9
,
e1003395
.
Goralski
,
T. J.
,
Edström
,
J. E.
and
Baker
,
B. S.
(
1989
).
The sex determination locus transformer-2 of Drosophila encodes a polypeptide with similarity to RNA binding proteins
.
Cell
56
,
1011
-
1018
.
Graveley
,
B. R.
,
Brooks
,
A. N.
,
Carlson
,
J. W.
,
Duff
,
M. O.
,
Landolin
,
J. M.
,
Yang
,
L.
,
Artieri
,
C. G.
,
van Baren
,
M. J.
,
Boley
,
N.
,
Booth
,
B. W.
, et al.
(
2011
).
The developmental transcriptome of Drosophila melanogaster
.
Nature
471
,
473
-
479
.
Han
,
C.
,
Jan
,
L. Y.
and
Jan
,
Y.-N.
(
2011
).
Enhancer-driven membrane markers for analysis of nonautonomous mechanisms reveal neuron-glia interactions in Drosophila
.
Proc. Natl. Acad. Sci. USA
108
,
9673
-
9678
.
Hay
,
B. A.
,
Wolff
,
T.
and
Rubin
,
G. M.
(
1994
).
Expression of baculovirus P35 prevents cell death in Drosophila
.
Development
120
,
2121
-
2129
.
Heinrichs
,
V.
,
Ryner
,
L. C.
and
Baker
,
B. S.
(
1998
).
Regulation of sex-specific selection of fruitless 5’ splice sites by transformer and transformer-2
.
Mol. Cell. Biol.
18
,
450
-
458
.
Hoshijima
,
K.
,
Inoue
,
K.
,
Higuchi
,
I.
,
Sakamoto
,
H.
and
Shimura
,
Y.
(
1991
).
Control of doublesex alternative splicing by transformer and transformer-2 in Drosophila
.
Science
252
,
833
-
836
.
Hudry
,
B.
,
Khadayate
,
S.
and
Miguel-Aliaga
,
I.
(
2016
).
The sexual identity of adult intestinal stem cells controls organ size and plasticity
.
Nature
530
,
344
-
348
.
Ito
,
H.
,
Fujitani
,
K.
,
Usui
,
K.
,
Shimizu-Nishikawa
,
K.
,
Tanaka
,
S.
and
Yamamoto
,
D.
(
1996
).
Sexual orientation in Drosophila is altered by the satori mutation in the sex-determination gene fruitless that encodes a zinc finger protein with a BTB domain
.
Proc. Natl. Acad. Sci. USA
93
,
9687
-
9692
.
Ito
,
H.
,
Sato
,
K.
,
Kondo
,
S.
,
Ueda
,
R.
and
Yamamoto
,
D.
(
2016
).
Fruitless represses robo1 transcription to shape male-specific neural morphology and behavior in Drosophila
.
Curr. Biol.
26
,
1532
-
1542
.
Kapelnikov
,
A.
,
Rivlin
,
P. K.
,
Hoy
,
R. R.
and
Heifetz
,
Y.
(
2008
).
Tissue remodeling: a mating-induced differentiation program for the Drosophila oviduct
.
BMC Dev. Biol.
8
,
114
.
Kimura
,
K.-I.
(
2011
).
Role of cell death in the formation of sexual dimorphism in the Drosophila central nervous system
.
Dev. Growth Differ.
53
,
236
-
244
.
Kimura
,
K.-I.
,
Ote
,
M.
,
Tazawa
,
T.
and
Yamamoto
,
D.
(
2005
).
Fruitless specifies sexually dimorphic neural circuitry in the Drosophila brain
.
Nature
438
,
229
-
233
.
Kimura
,
K.-I.
,
Hachiya
,
T.
,
Koganezawa
,
M.
,
Tazawa
,
T.
and
Yamamoto
,
D.
(
2008
).
Fruitless and doublesex coordinate to generate male-specific neurons that can initiate courtship
.
Neuron
59
,
759
-
769
.
Kimura
,
K.-I.
,
Sato
,
C.
,
Koganezawa
,
M.
and
Yamamoto
,
D.
(
2015
).
Drosophila ovipositor extension in mating behavior and egg deposition involves distinct sets of brain interneurons
.
PLoS ONE
10
,
e0126445
.
Kyriacou
,
C. P.
and
Hall
,
J. C.
(
1980
).
Circadian rhythm mutations in Drosophila melanogaster affect short-term fluctuations in the male's courtship song
.
Proc. Natl. Acad. Sci. USA
77
,
6729
-
6733
.
Laturney
,
M.
and
Billeter
,
J. C.
(
2014
).
Neurogenetics of female reproductive behaviors in Drosophila melanogaster
.
Adv. Genet.
85
,
1
-
108
.
Lee
,
G.
,
Hall
,
J. C.
and
Park
,
J. H.
(
2002
).
Doublesex gene expression in the central nervous system of Drosophila melanogaster
.
J. Neurogenet.
16
,
229
-
248
.
Lin
,
D. M.
and
Goodman
,
C. S.
(
1994
).
Ectopic and increased expression of Fasciclin II alters motoneuron growth cone guidance
.
Neuron
13
,
507
-
523
.
Manoli
,
D. S.
,
Fan
,
P.
,
Fraser
,
E. J.
and
Shah
,
N. M.
(
2013
).
Neural control of sexually dimorphic behaviors
.
Curr. Opin. Neurobiol.
23
,
330
-
338
.
Meissner
,
G. W.
,
Luo
,
S. D.
,
Dias
,
B. G.
,
Texada
,
M. J.
and
Baker
,
B. S.
(
2016
).
Sex-specific regulation of Lgr3 in Drosophila neurons
.
Proc. Natl. Acad. Sci. USA
113
,
E1256
-
E1265
.
Mohr
,
C.
and
Hartmann
,
B.
(
2014
).
Alternative splicing in Drosophila neuronal development
.
J. Neurogenet.
28
,
199
-
215
.
Nagoshi
,
R. N.
,
McKeown
,
M.
,
Burtis
,
K. C.
,
Belote
,
J. M.
and
Baker
,
B. S.
(
1988
).
The control of alternative splicing at genes regulating sexual differentiation in D. melanogaster
.
Cell
53
,
229
-
236
.
Neville
,
M. C.
,
Nojima
,
T.
,
Ashley
,
E.
,
Parker
,
D. J.
,
Walker
,
J.
,
Southall
,
T.
,
Van de Sande
,
B.
,
Marques
,
A. C.
,
Fischer
,
B.
,
Brand
,
A. H.
, et al.
(
2014
).
Male-specific fruitless isoforms target neurodevelopmental genes to specify a sexually dimorphic nervous system
.
Curr. Biol.
24
,
229
-
241
.
Nojima
,
T.
,
Neville
,
M. C.
and
Goodwin
,
S. F.
(
2014
).
Fruitless isoforms and target genes specify the sexually dimorphic nervous system underlying Drosophila reproductive behavior
.
Fly (Austin)
8
,
95
-
100
.
Ota
,
T.
,
Fukunaga
,
A.
,
Kawabe
,
M.
and
Oishi
,
K.
(
1981
).
Interactions between sex-transformation mutants of Drosophila melanogaster
.
Genetics
99
,
429
-
441
.
Pauli
,
A.
,
Althoff
,
F.
,
Oliveira
,
R. A.
,
Heidmann
,
S.
,
Schuldiner
,
O.
,
Lehner
,
C. F.
,
Dickson
,
B. J.
and
Nasmyth
,
K.
(
2008
).
Cell-type-specific TEV protease cleavage reveals cohesin functions in Drosophila neurons
.
Dev. Cell
14
,
239
-
251
.
Pfeiffer
,
B. D.
,
Truman
,
J. W.
and
Rubin
,
G. M.
(
2012
).
Using translational enhancers to increase transgene expression in Drosophila
.
Proc. Natl. Acad. Sci. USA
109
,
6626
-
6631
.
Ren
,
Q.
,
Awasaki
,
T.
,
Huang
,
Y.-F.
,
Liu
,
Z.
and
Lee
,
T.
(
2016
).
Cell class-lineage analysis reveals sexually dimorphic lineage compositions in the drosophila brain
.
Curr. Biol.
26
,
2583
-
2593
.
Rezával
,
C.
,
Pavlou
,
H. J.
,
Dornan
,
A. J.
,
Chan
,
Y. B.
,
Kravitz
,
E. A.
and
Goodwin
,
S. F.
(
2012
).
Neural circuitry underlying drosophila female postmating behavioral responses
.
Curr. Biol.
22
,
1155
-
1165
.
Rezával
,
C.
,
Nojima
,
T.
,
Neville
,
M. C.
,
Lin
,
A. C.
and
Goodwin
,
S. F.
(
2014
).
Sexually dimorphic octopaminergic neurons modulate female postmating behaviors in Drosophila
.
Curr. Biol.
24
,
725
-
730
.
Rideout
,
E. J.
,
Billeter
,
J.-C.
and
Goodwin
,
S. F.
(
2007
).
The sex-determination genes fruitless and doublesex specify a neural substrate required for courtship song
.
Curr. Biol.
17
,
1473
-
1478
.
Rideout
,
E. J.
,
Dornan
,
A. J.
,
Neville
,
M. C.
,
Eadie
,
S.
and
Goodwin
,
S. F.
(
2010
).
Control of sexual differentiation and behavior by the doublesex gene in Drosophila melanogaster
.
Nat. Neurosci.
13
,
458
-
466
.
Rideout
,
E. J.
,
Narsaiya
,
M. S.
and
Grewal
,
S. S.
(
2015
).
The sex determination gene transformer regulates male-female differences in drosophila body size
.
PLoS Genet.
11
,
e1005683
.
Ryner
,
L. C.
,
Goodwin
,
S. F.
,
Castrillon
,
D. H.
,
Anand
,
A.
,
Villella
,
A.
,
Baker
,
B. S.
,
Hall
,
J. C.
,
Taylor
,
B. J.
and
Wasserman
,
S. A.
(
1996
).
Control of male sexual behavior and sexual orientation in Drosophila by the fruitless gene
.
Cell
87
,
1079
-
1089
.
Salz
,
H. K.
and
Erickson
,
J. W.
(
2010
).
Sex determination in Drosophila: the view from the top
.
Fly (Austin)
4
,
60
-
70
.
Sanders
,
L. E.
and
Arbeitman
,
M. N.
(
2008
).
Doublesex establishes sexual dimorphism in the Drosophila central nervous system in an isoform-dependent manner by directing cell number
.
Dev. Biol.
320
,
378
-
390
.
Shirangi
,
T. R.
,
Taylor
,
B. J.
and
McKeown
,
M.
(
2006
).
A double-switch system regulates male courtship behavior in male and female Drosophila melanogaster
.
Nat. Genet.
38
,
1435
-
1439
.
Soller
,
M.
,
Haussmann
,
I. U.
,
Hollmann
,
M.
,
Choffat
,
Y.
,
White
,
K.
,
Kubli
,
E.
and
Schäfer
,
M. A.
(
2006
).
Sex-peptide-regulated female sexual behavior requires a subset of ascending ventral nerve cord neurons
.
Curr. Biol.
16
,
1771
-
1782
.
Stockinger
,
P.
,
Kvitsiani
,
D.
,
Rotkopf
,
S.
,
Tirián
,
L.
and
Dickson
,
B. J.
(
2005
).
Neural circuitry that governs Drosophila male courtship behavior
.
Cell
121
,
795
-
807
.
Sturtevant
,
A. H.
(
1945
).
A gene in drosophila melanogaster that transforms females into males
.
Genetics
30
,
297
-
299
.
Taylor
,
B. J.
and
Truman
,
J. W.
(
1992
).
Commitment of abdominal neuroblasts in Drosophila to a male or female fate is dependent on genes of the sex-determining hierarchy
.
Development
114
,
625
-
642
.
Usui-Aoki
,
K.
,
Ito
,
H.
,
Ui-Tei
,
K.
,
Takahashi
,
K.
,
Lukacsovich
,
T.
,
Awano
,
W.
,
Nakata
,
H.
,
Piao
,
Z. F.
,
Nilsson
,
E. E.
,
Tomida
,
J.-Y.
, et al.
(
2000
).
Formation of the male-specific muscle in female Drosophila by ectopic fruitless expression
.
Nat. Cell Biol.
2
,
500
-
506
.
Villella
,
A.
and
Hall
,
J. C.
(
2008
).
Neurogenetics of courtship and mating in Drosophila
.
Adv. Genet.
62
,
67
-
184
.
von Philipsborn
,
A. C.
,
Jörchel
,
S.
,
Tirian
,
L.
,
Demir
,
E.
,
Morita
,
T.
,
Stern
,
D. L.
and
Dickson
,
B. J.
(
2014
).
Cellular and behavioral functions of fruitless isoforms in Drosophila courtship
.
Curr. Biol.
24
,
242
-
251
.
Weng
,
R.
,
Chin
,
J. S. R.
,
Yew
,
J. Y.
,
Bushati
,
N.
and
Cohen
,
S. M.
(
2013
).
miR-124 controls male reproductive success in Drosophila
.
Elife
2
,
e00640
.
Yamamoto
,
D.
(
2007
).
The neural and genetic substrates of sexual behavior in Drosophila
.
Adv. Genet.
59
,
39
-
66
.
Yamamoto
,
D.
and
Koganezawa
,
M.
(
2013
).
Genes and circuits of courtship behaviour in Drosophila males
.
Nat. Rev. Neurosci.
14
,
681
-
692
.
Yang
,
C.-H.
,
Belawat
,
P.
,
Hafen
,
E.
,
Jan
,
L. Y.
and
Jan
,
Y.-N.
(
2008
).
Drosophila egg-laying site selection as a system to study simple decision-making processes
.
Science
319
,
1679
-
1683
.
Yang
,
C.-H.
,
Rumpf
,
S.
,
Xiang
,
Y.
,
Gordon
,
M. D.
,
Song
,
W.
,
Jan
,
L. Y.
and
Jan
,
Y. N.
(
2009
).
Control of the postmating behavioral switch in Drosophila females by internal sensory neurons
.
Neuron
61
,
519
-
526
.
Yu
,
J. Y.
,
Kanai
,
M. I.
,
Demir
,
E.
,
Jefferis
,
G. S.
and
Dickson
,
B. J.
(
2010
).
Cellular organization of the neural circuit that drives Drosophila courtship behavior
.
Curr. Biol.
20
,
1602
-
1614
.
Zhou
,
C.
,
Pan
,
Y.
,
Robinett
,
C. C.
,
Meissner
,
G. W.
and
Baker
,
B. S.
(
2014
).
Central brain neurons expressing doublesex regulate female receptivity in Drosophila
.
Neuron
83
,
149
-
163
.

Competing interests

The authors declare no competing or financial interests.

Supplementary information