Beadex Function in the Motor Neurons Is Essential for Female Reproduction in Drosophila melanogaster

Drosophila melanogaster has served as an excellent model system for understanding the neuronal circuits and molecular mechanisms regulating complex behaviors. The Drosophila female reproductive circuits, in particular, are well studied and can be used as a tool to understand the role of novel genes in neuronal function in general and female reproduction in particular. In the present study, the role of Beadex, a transcription co-activator, in Drosophila female reproduction was assessed by generation of mutant and knock down studies. Null allele of Beadex was generated by transposase induced excision of P-element present within an intron of Beadex gene. The mutant showed highly compromised reproductive abilities as evaluated by reduced fecundity and fertility, abnormal oviposition and more importantly, the failure of sperm release from storage organs. However, no defect was found in the overall ovariole development. Tissue specific, targeted knock down of Beadex indicated that its function in neurons is important for efficient female reproduction, since its neuronal knock down led to compromised female reproductive abilities, similar to Beadex null females. Further, different neuronal class specific knock down studies revealed that Beadex function is required in motor neurons for normal fecundity and fertility of females. Thus, the present study attributes a novel and essential role for Beadex in female reproduction through neurons.


Introduction
The nervous system plays an important role in modulating several physiological processes and complex behaviors in multicellular animals. Drosophila has served as an excellent model to unravel the neuronal regulation of multiple complex behaviors like memory and learning, aggression, courtship and female reproduction [1][2][3]. The neuronal regulation of female reproduction in particular has been studied extensively and the multiple circuits which play a major role have been identified. Several studies have shown that octopaminergic neurons from the central nervous system regulate multiple female reproductive behaviors like ovulation, egg laying and also sperm release [4][5][6][7][8]. An ex vivo study has demonstrated the direct role of octopamine in the contraction of the Drosophila female reproductive tract [9]. Glutamatergic neurons also modulate the contraction of oviduct by acting in conjunction with octopamine during egg laying [10]. A recent study has revealed that octopamine brings about the contraction/relaxation of oviduct through CamKII mediated signaling in the epithelial cells of oviduct [7]. Apart from neuronal circuits intrinsic to the female, seminal proteins transferred from the male during copulation are also essential for inducing post mating behaviors in female through these neuronal circuits [5]. For example, Sex peptides (SP) transferred from male into the female reproductive tract during copulation bind to receptors on the sensory neurons of the female reproductive tract, that project on to the central nervous system regions (known to send projections on to female reproductive tract), and bring about post mating reproductive behaviors like increased ovulation, egg laying and reduced receptivity [11,12]. With such well studied neuronal circuits regulating the female reproduction and simple reproductive assays, the Drosophila female reproductive system serves as a good model to understand the functions of uncharacterized genes implicated in neuronal functions.
Beadex (Bx) was isolated as a spontaneous mutant with beaded wings (long, narrow and excised along both margins) in Drosophila (http://flybase.org, version 13 FB2008_09) [13]. An elegant study by Lifschytz and Green showed that the mutant wings' phenotype was caused by over production of the Bx gene product (hypermorph) [14]. Zhu and co-workers in 1995 identified a homolog of human LIM domain oncogene in Drosophila [15]. Later three independent studies reported that Bx itself codes for Drosophila LIM only (dLMO) domain containing protein [16][17][18]. Any disruption in the Bx coding sequence leads to Bx loss of function and is evident as held-up wings phenotype. However, abrogation in the 39UTR sequence, which harbors negative regulatory elements, leads to gain of function alleles of Bx [17].
The LIM domain consists of two tandemly organized zinc finger motifs with the conserved cysteine-histidine residues that co-ordinate zinc binding. LIM domain mediates its function through protein-protein interactions [19]. In Bx hypermorphs, increased Bx levels disrupt the functional complex of Apterous-Chip (Ap-Chip) due to the competitive binding of Bx to LIM-binding domain (LBD) of Chip [18,20]. Bx also plays an important role in the development of dorsal macrochaetae (bristles on the dorsal side of thorax) by activating the transcription of Acheate and Scute through its positive interaction with Pannier (Pnr) in the sensory organ precursor (SOP) specification [21,22]. Lmo4, one of the vertebrate homologues of Bx, suppresses the formation of excitatory glutamatergic inter-neurons and promotes the development of GABAergic inter-neurons in mice [23]. Apart from regulating the above mentioned developmental processes, Bx also affects Drosophila behavior in response to cocaine and ethanol [24][25][26]. Bx affects ethanol mediated sedation through repression of anaplastic lymphoma kinase (Alk) in the brain of Drosophila [25]. The modulation of activity of Bx homologues in mice, Lmo4 and Lmo3, also affect the behavior of mice in response to cocaine and ethanol, through mechanisms similar to those in Drosophila [26,27].
Though several studies have reported Bx role in multiple behaviors through neurons, the molecular mechanisms and the targets through which Bx effects its function are not clearly understood. In this study, we attribute a novel and essential function for Bx in Drosophila female reproduction. We show that multiple, post mating reproductive processes like ovulation, egg laying and sperm release are compromised in the Bx null females. We further provide evidence that Bx regulates these functions through the motor neurons that innervate the female reproductive tracts.

Fly stocks and Genetics
All the flies were maintained at 2362uC, under 12 hrs Light/ Dark (L/D) cycles on cornmeal-sucrose-yeast agar media. Either w 1118 females or female progeny of Canton-S males crossed with w 1118 females were used as controls, unless specified. Bx hpd-G14-1 was a kind gift from Pascal Heitzler (CNRS, France, described in [21]). New Bx alleles were generated by mobilizing P{GawB}Bx hdp-G14-1 element using P{D2-3} as source for transposase (Bloomington Stock (BS)#4368) [28]. Hop-outs from F1 were balanced over FM7a followed by screening of F2 progenies for white eye (excision of P-element results in white eye). These white eyed females were crossed with wild type males and resulting F3 males were screened for held-up wings (since, as reported earlier, Bx loss of function alleles show held-up wings phenotype) [17]. Isolated hop-out alleles which showed held-up wings were maintained over duplication of Bx locus on Y chromosome, Dp(1;Y)W39 [29]. Bx RNAi line was obtained from Vienna Drosophila RNAi Centre (VDRC Transformant ID#2917). All the Gal4 lines and GFP tagged reporter protein construct fly lines were procured either from Bloomington Drosophila stock centre, Indiana or National Center for Biological Sciences (NCBS) stock centre, Bangalore, India.

Molecular characterization of Bx 7 allele
RNA from whole body of 1-2 days old w 1118 and Bx hopout flies was isolated using trizol (Sigma) following manufacturer's instructions. From the RNA, 2 mg was taken for making cDNA using first strand cDNA kit following manufacturer's protocol (Thermo Scientific, India). Primers specific for Bx-RA and Bx-RB transcript, were designed as follows-Bx-RAFP 59 -CTAATTGAGTCGAGTGTGCGTG -39, Bx-RARP 59 -AAGGAGGTTGGTTGTCGTCGTC -39, Bx-RBFP 59 -ATGGAGTACCTCTACAACGCTA -39 and Bx-RBFP 59 -TTATTTCGGGACCCGTAC -39. House-keeping gene rp49 was used as control using following primers rp49FP 59 -TTCTACCAGCTTCAAGATGAC -39 and rp49RP 59 -GTGTATTCCGACCACGTTACA -39. In order to characterize the molecular lesion in hop-out Bx allele, genomic DNA was isolated from wild type (w 1118 ) and isolated Bx mutant (Bx 7 ). Primers were designed on either side of the p{GawB} Bx hdp-G14-1 insertion site (assessed from [21]). The sequences of primers used for PCR amplification of Bx gene region are Bx-FP 59 -GGCTCGTTGGTCTAGAGGTA -39 and Bx-RP 59 -CATAATGGCATCTCCGCAAG -39. The location of the primers in the Bx gene is represented in Figure 1A. PCR amplification of Bx gene region was performed with ExTaq (TaKaRa). The resultant PCR products were analyzed through agarose gel electrophoresis and documented using Alpha DigiDoc RT2 (JH BIO Innovations, India). Further, the PCR product from the Bx mutant was cloned into TA vector (InsTAclone PCR cloning kit, Thermo scientific, India) and sequenced (Amnion, India). The resultant sequences were BLAST aligned against Drosophila genome (http://flybase. org/blast/) to determine the molecular lesion in the Bx mutant.

Knock down studies
For efficient knock down studies, UAS-Bx RNAi construct was brought together with UAS-Dcr2 (BS#24646) [33]. Similarly, to perform developmental stage specific knock down, tub-Gal80 ts (drives temperature sensitive inhibitor of Gal4 activity, Gal80 ts , under alpha-tubulin84B promoter, [34]) was brought along with UAS-Bx RNAi . Virgin females of UAS-Dcr2; tub-Gal80 ts ; UAS-Bx RNAi or UAS-Dcr2; +; UAS-Bx RNAi were crossed either with different Gal4 construct males or with wild type males (for heterozygote controls) and grown at appropriate temperatures (for knock down with Dmef2-Gal4, Bx RNAi without UAS-Dcr2 was used, since knock down along with Dcr2 caused larval lethality). For knock down across all the stages, progeny were grown from the early embryonic stage up to the test age (3-4 days old adult flies) at 29uC. For the pupal stage knock down, larvae were grown till wandering third instar stage at 18uC and were then shifted to 29uC till the adult flies eclosed. The eclosed flies were then kept at 18uC for the next 4-5 days. For the adult stage knock down of Bx, larvae and pupae were allowed to develop at 18uC and the freshly eclosed adult flies were kept at 29uC for next the 4-5 days.

Reproductive ability assays -Fertility, Fecundity, Ovulation and Oviposition
Virgin females (3-4 days old) of interested genotype and 2-3 days old Canton-S males (grown at 25uC) were isolated for crosses. Single pair matings were set up in the media vials. After (E) Bx 7 females have highly reduced fecundity (F, ***, p,0.0001; **, p = 0.0011) and fertility (G, ***, p,0.0001) (no of eggs tested.800 (w 1118 and controls), ,200 (Bx 7 )) (Students unpaired t-test with Welch's correction) compared to wild type and other controls. (H) almost 100% of the eggs are laid on the surface of the media by Bx 7 mutant females compared to control females which deposit upto 90% of the eggs into the media (schematic of eggs laid by control and Bx 7 mutant females is shown below the X-axis of the graph) (***, p,0.001, Two way anova with Bonferroni post tests). doi:10.1371/journal.pone.0113003.g001 the visual confirmation of mating and recording the duration of copulation, males were separated and the females were transferred into egg laying media vials (1% Sucrose, 1% Agar-Agar and 0.2% Propionic acid) supplemented with yeast paste and kept in 25uC or 29uC incubator with 12 hrs L/D cycle. The females were removed 24 hrs later and the number of eggs laid was counted to assess the fecundity. Fecundity was plotted as the average number of eggs laid per female in 24 hrs. Further, same set of vials were incubated at 25uC and checked 24 hrs later to count the hatched eggs for the fertility assay. Fertility was plotted as percent hatched eggs. For assessing the ovulation ability of the females, 24 hrs post mating (hpm) with Canton-S males, reproductive tract was dissected and the percent of females showing the presence of egg in the uterus was recorded. Minimum of 15 females per genotype were used for all the experiments and the experiments were independently repeated at least three times. For oviposition assay, mated females were transferred on to petri plates with egg laying media and kept at 25uC or 29uC incubator for 6 hrs. The females were then removed and the percent of total eggs deposited on the media was recorded. Data for all the above assays were plotted as mean with standard deviation.

Ovariole development
Ovaries of the females mated with wild type males were dissected and the number of developing ovarioles was counted in each ovary. Following this, the same samples were fixed in 4% PF for 30 minutes and washed for 15 minutes (365 minutes) with PBTx. The washed samples were then incubated with Hoechst (1 mg/ml, Sigma) for 30 minutes for nuclear staining, washed again in PBTx and mounted in Vectashield mounting media (Vector labs). The mounted samples were visualized and imaged using Zeiss fluorescent microscope (Zeiss Axio Observer Z1).

Sperm storage assay
To study the sperm storage and releasing ability, females of the interested genotype were mated with 2-3 days old dj-GFP males, which have GFP labeled sperm tails (BS#5417) [35]. At 6 hpm and 10 days post mating (dpm), the female reproductive tract was dissected, fixed in 4% PF and washed for 15 minutes in PBTx. Following this, the samples were stained with phalloidin-TRITC (Sigma) for visualizing seminal receptacle. The presence or absence of sperms was monitored by looking for GFP signal in the seminal receptacle. Imaging was performed using Zeiss LSM510 Meta confocal microscopy and the images were processed with LSM software (version 3.2.0.115, Zeiss).

Data plotting and Statistical analysis
Data was plotted, graphs were made and statistical test analysis was performed using Graph pad Prism 5 (Graphpad).

Generation of Bx hypomorphic alleles
LIM domain of Bx can bind to proteins containing LBD and can thus regulate the function of several proteins. In order to study whether Bx regulates the female reproduction or not, generation of loss of function or hypomorphic alleles of Bx was essential. New Bx alleles were generated by mobilizing P{GawB} element (Bx hdp-G14-1 ) inserted near 39 end of the first intron ( Figure 1A, and described in [21]). Out of several individual hop-outs obtained, twelve lines were revertants and three showed held-up wings phenotype. Of these three, only Bx 7 , which showed severe held-up wings defect, was character-ized and used in the present study ( Figure 1C & C'). cDNA prepared from the Bx 7 mutant flies revealed complete absence of transcript products of Bx-RA and Bx-RB isoforms (Figure 1D). Furthermore, PCR amplification of genomic DNA of the hop-out revealed deletion of ,2 kb of Bx gene sequence ( Figure 1E) which harbors a Bx-RB specific exon (1), two constitutive exons (2 & 3) and about 70 bp of the 59UTR ( Figure 1B). Thus, Bx 7 is a null allele of the Bx.

Bx null females show reduced fecundity and fertility
In order to find whether Bx regulates the female reproduction, Bx 7 females were mated with Canton-S males. Bx 7 females showed normal receptivity (100% receptivity-data not shown) and normal mating duration ( Figure S1). However, despite showing normal mating duration, mutant females showed highly reduced fecundity and fertility (egg hatching percentage) compared to wild type (w 1118 ) and other controls (Bx 7 /+, Bx hdp-G14-1 / Bx hdp-G14-1 , and Bx 7 /Bx hdp-G14-1 ) ( Figure 1F and 1G). It is interesting to note that homozygous Bx hdp-G14-1 also showed reduced fecundity compared to controls. This indicated that insertion of P{GawB} element also affects the Bx function. The unhatched eggs were white in color but not brown indicating a fertilization defect. Another evident phenotype of the Bx 7 females was abnormal oviposition, wherein all the eggs laid by the mutant females were deposited on the surface of the media in contrast to the control females which deposited eggs into the media ( Figure 1H).

Bx null females show normal ovary development but defective post mating responses
Ovaries of Bx 7 and control females were analyzed to see if the reduced fecundity was due to abnormal ovary development. At gross level, it could be observed that the number of ovarioles was less in the Bx 7 mutant females compared to those in the control females (Figure 2A). Nuclear staining of Bx 7 ovarioles with Hoechst showed normal development of the ovarioles across various stages similar to that of the controls, indicating normal oogenesis in Bx 7 mutant females ( Figure 2D and 2E). Mating induces several physiological and behavioral changes in the females which include increased egg laying (fecundity), ovulation and oogenesis [36][37][38]. Since the mutant females showed reduced fecundity, we checked if the ovulation was defective in the mutant females. Indeed, the Bx 7 mutant females showed highly reduced ovulation compared to the controls ( Figure 2B), indicating that the reduced fecundity could be the result of reduced ovulation rather than a defect in the development of ovarioles. However, we can't completely rule out developmental role of Bx in ovaries during ovariole development, as ovariole numbers have been shown to be influenced by cell differentiation and numbers established during metamorphosis stage [39]. For an efficient post mating response, sperms and the accessory gland proteins bound to sperms must be stored in the female sperm storage organs-spermathecae and seminal receptacle [40,41]. Moreover, only 10% of the eggs laid by the mutant females hatched ( Figure 1G). So we looked at the sperm storing ability of the mutant females and the control females by crossing them with the dj-GFP males. As shown in the Figure 2F and F', within 6 hpm dj-GFP labeled sperms accumulated in seminal receptacle in both the control (w 1118 ) and Bx 7 mutant females. Thus, Bx 7 mutant females have normal sperm storing ability. Over a period of 10-15 dpm, sperms stored in the storage organs are released for fertilizing eggs, thus depleting sperms from the storage organs [42]. As seen in Figure 2G, dj-GFP sperms from the seminal receptacle of the control flies (w 1118 ) were totally depleted by 10 dpm. However, the seminal receptacle of Bx 7 mutant females retained dj-GFP sperms even after 10 dpm ( Figure 2C and 2G'). This could be one possible reason for the reduced hatching of the eggs laid by the mutant females.
Bx function in the neurons, but not in the muscles, is required for the female reproductive success A detailed analysis of multiple reproductive behaviors of the Bx 7 null females indicated that the defective phenotypes could result from reduced motor activity of the reproductive tract. Motor activity of the reproductive tract is modulated through the muscles and neurons. In order to test which of the tissues was affected in the Bx 7 mutant females, Bx was knocked down either in the muscles or neurons separately and the flies were assessed for their fecundity and fertility. When Bx was knocked down in the ovarian muscles using Dmef2-Gal4 (which showed strong expression in all the female reproductive tract muscles) ( Figure 3A), neither fecundity nor fertility of the females was affected ( Figure 3C and 3D). Elav-Gal4, a pan neuronal Gal4, showed strong expression in neurons innervating more or less all the muscles of the female reproductive tract ( Figure 3B and Figure S2). Knock down of Bx in the neurons with Elav-Gal4 led to highly reduced fecundity and fertility which was comparable to the Bx 7 mutant females ( Figure 3C and 3D). Even in this case the unhatched eggs were white in color, clearly indicating a defect in the fertilization of the eggs. Knock down of Bx in the neurons not only affected fecundity and fertility, but also phenocopied all the reproductive defects observed in the Bx 7 mutant females like reduced ovulation ( Figure 3E), reduced sperm release from the seminal receptacle ( Figure 3F) and defective egg deposition ( Figure S3) with the exception that unlike the Bx 7 mutant females, the neuronal knock down of Bx led to more accumulation of mature eggs in the ovary ( Figure S4).

Developmental and post developmental function of Bx is essential for the female reproductive ability
Bx acts as a transcription co-activator in several developmental processes like wing disc morphogenesis and bristle development [18,[20][21][22]. In order to understand if Bx function in the neurons was required during development for female reproduction, stage specific knock down of Bx in the neurons was carried out. Bx knock down during the pupal development or in the adult stage separately did not cause any reproductive defects ( Figure 4A, A' and 4B, B'). However, knock down of Bx during both pupal and adult stages together showed significantly reduced fecundity and fertility of the females ( Figure 4C, C'). It is possible that Bx loss of function during the pupal stage and associated developmental defects are compensated by the Bx expressed during the adult stage. However, such compensation would not happen with the adult stage specific knock down of Bx. Thus, it is possible that Bx function in the neurons is more critical during development of the female reproductive tract. Bx knock down in glutamatergic, but not aminergic neurons, partially affects the female reproductive abilities Several genetic studies have revealed that octopaminergic and glutamatergic classes of neurons play critical roles in female reproductive behaviors like ovulation, fecundity and sperm storage [4][5][6][7][8]. Moreover, these classes of neurons show innervations onto various parts of the female reproductive tract [4,10] (Figure S5). Thus, to find if Bx modulated the female reproduction through these classes of neurons, Bx was knocked down in octopaminergic and glutamatergic neurons using dTdc2-Gal4 and VGlut-Gal4 respectively. dTdc2-Gal4 neurons showed innervations onto the oviduct and uterine muscles of the female reproductive tract (Figure S5 C, D), while VGlut-Gal4 neurons showed projections only onto the uterine muscles ( Figure S5 A, B). Knock down of Bx in the octopaminergic neurons did not affect either fecundity or fertility of the females ( Figure 5A and 5B). However, females in which Bx was knocked down in the glutamatergic neurons showed a partial reduction in fertility without affecting fecundity ( Figure 5B). Moreover, knock down of Bx in the glutamatergic neurons led to oviposition defect in the females similar to that of Bx 7 mutant females, though to a lesser extent ( Figure S6). One of the possible reasons why knock down of Bx either in the octopaminergic neurons or glutamatergic neurons separately did not significantly affect female reproduction could be due to the innervation of these classes of neurons onto fewer female reproductive tract muscles.

Bx function in the motor neurons is important for female reproduction
As mentioned earlier, the reproductive defects in Bx 7 mutants could be due to reduced motor activity caused by the abnormal neuronal function. In order to test this hypothesis, Bx was knocked down in the motor neurons innervating the female reproductive tract muscles using motor neuron specific Gal4 line C380 [43]. UAS-SyteGFP reporter under C380-Gal4 revealed extensive innervations of this class of neurons onto the oviduct, common oviduct, uterine and seminal receptacle muscles ( Figure 6A and 6B). Knock down of Bx in these motor neurons in the females resulted in significant reduction in fecundity ( Figure 6C) and fertility ( Figure 6D) compared to the controls. Moreover, the egg deposition defect was also recapitulated when Bx was knocked down in the C380 motor neurons (data not shown). This indicated that Bx function in the motor neurons is essential for female reproduction.

Discussion
In this study, we report a novel and essential role of the Bx in Drosophila female reproduction. The Bx null females generated in this work showed normal ovariole development but with highly reduced fecundity and ovulation. Reduced fecundity correlates well with reduced ovulation. However, unlike earlier reported mutants which showed reduced fecundity and ovulation [4], accumulation of mature eggs in the ovaries is not seen in the Bx 7 mutant females. One possible reason for this could be the highly reduced abdominal cavity caused by the distended crop ( Figure S7). This is supported by a recent study which identified Bx in a screen for defective gastric emptying, thereby causing bulged crop [44] (unpublished data from lab). Other reason could be due to the function of Bx in the ovariole development, which was not investigated in this study. Highly reduced fertility of Bx 7 mutants could be attributed to the failure of sperm release from the storage organs (seminal receptacle) during ovulation (Figure 2G'). There is no direct evidence as to what stimulates the release of sperms from the storage organs in the females. One of the proposed stimuli is egg movement/presence in the uterus which activates stretch receptors in the uterus and induces release of sperm from storage organs [42]. It is thus possible that the reduced sperm release observed in the Bx 7 null females is due to reduced ovulation.
The overall defects like reduced fecundity, sperm release from storage organs and abnormal oviposition (Figure 1 and 2) could be attributed to the reduced motor activity of the female reproductive tract. Motor activity of the female reproductive tract is brought about by neuromuscular activities. Neuronal knock down of Bx in the females phenocopied all the reproductive defects of the Bx null females (Figure 3). This clearly indicates that reproductive defects observed in the Bx 7 mutant are due to loss of Bx function in the neurons. However, it needs to be noted that the neuronal knock down of Bx leads to accumulation of mature eggs in the ovary unlike Bx null ( Figure S4). This could be because there is no crop distension in flies with Bx knocked down in the neurons unlike the Bx 7 mutant females (data not shown). In the females where Bx is knocked down neuronally, it is also possible that Bx functions in the ovarioles are unaltered, which might lead to accumulation of mature eggs in the ovarioles.
Bx plays an important role during the development of wing disc and SOP [18,[20][21][22]. Similarly our study indicates that, for an  efficient female reproduction, Bx function is essential during the development of the reproductive tract (pupal stage). During development, Bx might transcriptionally regulate the genes which affect the neuronal circuits innervating the female reproductive tract. However, Bx might work independent of Pnr, a known transcription factor that interacts with Bx, in regulating the female reproduction, since pnr knock down in the neurons does not affect fecundity or fertility (data not shown). Lmo4, a vertebrate homolog of Bx is known to affect the number of neurite outgrowths and their length in human SH-SY5Y neuroblastoma cells [45]. Similar to Lmo4, Bx also might regulate the projection of neurons on to target muscles during development. Though the present study does not show any evidence for such functions, it offers a look at the neuro-circuits regulating reproduction for dissecting the molecular mechanism of Bx function.
Octopaminergic circuits innervating the female reproductive tract are major players regulating fecundity, ovulation and sperm release [4][5][6][7][8]. However, knock down of Bx in these circuits did not cause any reproductive defects ( Figure 5). One of the possible reasons could be that Bx might work through other circuits independently or in unison with the octopaminergic neurons for modulating female reproduction. For instance, glutamatergic neurons also play a vital role in oviduct contraction along with the octopaminergic neurons [10]. But, knock down of Bx in the glutamatergic neurons did not reduce fecundity, but reduced fertility only partially, and caused abnormal egg deposition on the surface of the media. This could be due to the strong innervations of glutamatergic neurons onto the uterine muscle ( Figure 5, S5 and S6). Since Bx knock down in the octopaminergic neurons did not cause any defect and knock down in the glutamatergic neurons only gave a partial defect, it can be speculated that Bx functions through multiple class of neurons for regulating female reproduction. Bx regulates the development of peripheral sensory organs (dorsal macrochaete) [21,22]. Female reproductive tract harbors sensory neurons which are stimulated by sex peptides from male ejaculum [11,12]. To see if Bx affects the female reproduction through these sensory neurons, Bx was knocked down in these sensory neurons using ppk-Gal4 (BS#32078, expresses Gal4 in class IV dendritic arborization neurons and also in sensory neurons in female reproductive tract) [12]. However, Bx knock down in these sensory neurons did not cause any defect in fecundity or fertility ( Figure S8).
Reproductive defects in the Bx null females might be caused by the reduced motor activity of the reproductive tract. This is supported by the fact that knock down of Bx in the motor neurons reduced fecundity and fertility of females significantly. This clearly shows that Bx activity in the motor neurons is essential for the female reproductive behavior. One possible explanation could be that Bx affects the neurotransmission activity thereby causing reduced female reproductive tract contractions. However, this may not be the case since Bx knock down in the neurons only in adult stage did not cause any reproductive defects in the females ( Figure 4B). Drosophila Islet, a LIM-homeo domain protein, is essential in deciding the neurotransmitter identity and axon path finding in the motor neurons during embryonic development [46]. More recent studies also report the modulation of voltage-dependent fast K + channel (encoded by shaker, sh) expression levels by Islet along with Lim3, another LIM-homeo domain protein [47,48]. It is possible that in the Bx 7 mutants and in the neuronal knock down of Bx, the functions of Islet and Lim3 are modulated (either increased or decreased activity) in the neurons due to their reduced interaction with Bx and this might result in altered neuronal function and the described phenotypes. This further emphasizes that Bx function is crucial during the female reproductive tract development. Drosophila Protein Interaction Map (DPiM) project has identified almost 100 proteins which interact with Bx in S2 cell lines [49]. Of these only 18 proteins show high expression in the adult brains (Flybase, version FB2014_04). The function of these interacting partners could be affected in the neurons of Bx 7 mutant flies and also in the neurons with Bx knock down, there by affecting the female reproduction.

Conclusion
We conclude that Bx is essential for Drosophila female reproduction where it regulates ovulation, sperm release from the storage organs and oviposition. All the reproductive defects of Bx null females can be phenocopied by knocking down Bx in the neurons. More specifically, Bx plays an important role in female reproduction through the motor neurons. This work shows the importance of utilizing the female reproductive tract circuits in understanding the molecular pathways through which Bx might modulate neuronal functions. Figure S1 Copulation duration of Bx 7 females is similar to that of wild type females. Both wild type and Bx 7 mutant females showed normal duration of copulation when mated with wild type Canton-S males.