Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

RNAi-based screens uncover a potential new role for the orphan neuropeptide receptor Moody in Drosophila female germline stem cell maintenance

  • Tianlu Ma,

    Roles Conceptualization, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Department of Biochemistry and Molecular Biology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, United States of America

    ⨯
  • Shinya Matsuoka,

    Roles Formal analysis

    Current address: AstraZeneca K. K., Osaka, Japan

    Affiliation Department of Biochemistry and Molecular Biology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, United States of America

    ⨯
  • Daniela Drummond-Barbosa

    Roles Conceptualization, Funding acquisition, Project administration, Supervision, Writing – review & editing

    dbarbosa@jhu.edu

    Affiliation Department of Biochemistry and Molecular Biology, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland, United States of America

    ⨯

Abstract

Reproduction is highly sensitive to changes in physiology and the external environment. Neuropeptides are evolutionarily conserved signaling molecules that regulate multiple physiological processes. However, the potential reproductive roles of many neuropeptide signaling pathways remain underexplored. Here, we describe the results of RNAi-based screens in Drosophila melanogaster to identify neuropeptides/neuropeptide receptors with potential roles in oogenesis. The screen read-outs were either the number of eggs laid per female per day over time or fluorescence microscopy analysis of dissected ovaries. We found that the orphan neuropeptide receptor encoded by moody (homologous to mammalian melatonin receptors) is likely required in somatic cells for normal egg production and proper germline stem cell maintenance. However, the egg laying screens had low signal-to-noise ratio and did not lead to the identification of additional candidates. Thus, although egg count assays might be useful for large-scale screens to identify oogenesis regulators that result in dramatic changes in oogenesis, more labor-intensive microscopy-based screen are better applicable for identifying new physiological regulators of oogenesis with more subtle phenotypes.

Citation: Ma T, Matsuoka S, Drummond-Barbosa D (2020) RNAi-based screens uncover a potential new role for the orphan neuropeptide receptor Moody in Drosophila female germline stem cell maintenance. PLoS ONE 15(12): e0243756. https://doi.org/10.1371/journal.pone.0243756

Editor: Christian Wegener, Biocenter, Universität Würzburg, GERMANY

Received: October 13, 2020; Accepted: November 25, 2020; Published: December 11, 2020

Copyright: © 2020 Ma et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: This work was supported by National Institutes of Health (NIH) grants R01 GM069875 (D.D.-B) and R01 GM125121 (D.D.-B.). T.M. was supported by NIH training grant T32 CA009110. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Reproduction is highly responsive to changes in physiology and the external environment [1]. In mammals, many of these changes impinge on the hypothalamic-pituitary-gonadal axis, the central regulator of reproduction. Gonadotropin-releasing hormone (GnRH) is produced and secreted by neurosecretory cells in the hypothalamus and acts on the anterior pituitary gland to stimulate the release of follicle stimulating hormone (FSH) and luteinizing hormone (LH) [2]. Obesity and excessive exercise can both lead to reduced gonadotropin levels in humans [3], and psychological stress decreases LH and FSH levels in rodents and other mammals [4]. Neuropeptides are an important group of signaling molecules that lie at the intersection of the hypothalamic-pituitary-gonadal axis and physiology. For example, the neuropeptide kisspeptin, encoded by KISS1, is a key activator of GnRH secretion and is in turn regulated by additional neuropeptides such as neuropeptide Y (NPY) [5]. Kisspeptin is also regulated by systemic factors, including insulin and adipocyte-derived leptin [2]. Neuropeptides likely have more ancient roles in communicating physiological state from the brain to the gonad that are independent of the hypothalamic-pituitary-gonadal axis, considering that this axis is a more recent evolutionary addition to the physiology of reproduction.

Neuropeptides are evolutionarily conserved signaling molecules present from invertebrates to humans [6]. There are over 100 neuropeptides in humans [7], while the Drosophila melanogaster genome encodes about 50 neuropeptides, some of which have been implicated in development, behavior, and reproduction [8,9]. Drosophila is a powerful model system for studying how physiology and the environment impact oogenesis [9]. Each Drosophila ovary is composed of 16 to 20 individual units called ovarioles, where follicles develop through 14 recognizable stages, including stage 8, when vitellogenesis begins, and stage 14, when the mature oocyte has fully formed dorsal appendages (Fig 1A). Follicles are formed in the anterior germarium, which houses two to three germline stem cells (GSCs) within a niche composed primarily of cap cells (Fig 1B). GSCs divide to self-renew and give rise to cystoblasts, which undergo four synchronous rounds of incomplete division to form 16-cell cysts composed of 15 nurse cells and one oocyte. Sixteen-cell cysts are enveloped by somatic follicle cells to form a new follicle (or egg chamber) that buds off from the germarium [9]. Several neuropeptides are known to control female reproduction. Neural-derived insulin-like peptides (ILPs) promote follicle growth and follicle cell proliferation in response to a nutrient rich diet [10], and insulin signaling is also required for GSC proliferation and maintenance, early germline cyst survival, and vitellogenesis [10–12]. ILP7 is also involved in oviposition [13], while gut-derived neuropeptide F (NPF; the Drosophila ortholog for NPY) regulates female GSC proliferation in response to mating and sex peptide (SP) signaling [14,15]. Multiple neuropeptides also regulate courtship and mating behavior [8]. The roles of additional neuropeptides/neuropeptide receptors in regulating Drosophila oogenesis, however, remain largely underexplored.

thumbnail
Download:
Fig 1. RNAi-based screens to identify neuropeptides/neuropeptide receptors required for normal egg production.

(A) Diagram of Drosophila ovariole, which is composed of progressively more developed follicles (or egg chambers), ending with a mature stage 14 oocyte identifiable by its dorsal appendages. (B) Diagram of germarium, the anterior-most portion of the ovariole, which houses germline stem cells (GSCs) in a niche comprising of cap cells, terminal filament cells, and a subset of escort cells. GSCs divide to self-renew and give rise to cystoblasts, which undergo four synchronous divisions to form 16-cell cysts, each composed of 15 nurse cells and an oocyte. The 16-cell cysts become enveloped by follicle cells and bud off to form a new follicle. (C) Egg counts for pan-neuronal knockdown of the four candidate neuropeptide genes. nSyb-Gal4 was used to drive UAS-hairpin RNA lines. The number of eggs laid per female per day were counted on days five, 10, and 15. nSyb>GFPdsRNA.143 served as control. (D) Egg counts for ubiquitous somatic knockdown of the 16 candidate neuropeptide receptor genes. tub-Gal80ts; tub-Gal4 (tubts) was used to drive UAS-hairpin RNA lines. The number of eggs laid per female per day on days five and 10 are shown, with tubts>LucJF01355 as control. InR knockdown served as an internal control. Day 15 data are included in S1 File. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001, Student’s t-test. Data shown as mean±s.e.m.

https://doi.org/10.1371/journal.pone.0243756.g001

Here, we describe RNAi-based screens for neuropeptides/neuropeptide receptors regulating oogenesis. Three of these screens used egg counts as a read out, while a fourth smaller screen involved ovary dissection and microscopy analysis of specific oogenesis processes. The orphan neuropeptide receptor encoded by moody was identified as a new factor likely promoting egg production and GSC maintenance in both an egg count-based screen and the dissection-based screen. However, the egg count-based screens did not have sufficient signal-to-noise ratio to reliably identify additional novel regulators. Our results suggest that while egg counts can be valuable for screening large numbers of genes for major effects in oogenesis, the analysis of dissected ovaries is a more useful approach for identifying genes with more subtle physiological roles in specific steps of oogenesis.

Results

Egg count-based screens for neuropeptide/neuropeptide receptors with roles in oogenesis

To identify novel neuropeptide signaling pathways that might regulate oogenesis, we performed three separate screens using egg counts as a read-out: pan-neuronal neuropeptide knockdown using nSyb-Gal4 [16], ubiquitous somatic neuropeptide receptor knockdown using tub-Gal4 in combination with the temperature-sensitive Gal4 inhibitor Gal80ts (tubts) [17], and germline-specific neuropeptide receptor knockdown using the maternal triple driver MTD, which combines three germline drivers (otu-Gal4::VP16, nos-Gal4::VP16, and Gal4-nos.NGT) expressed in the germarium and throughout oogenesis [18]. Zero-to-two-day-old females were paired with y w males at 29°C on a nutrient-rich diet (wet yeast paste on molasses-agar plates), and the numbers of eggs laid in a 24-hour period on days five, 10, and 15 were counted. nSyb>GFPdsRNA.143 females (in which nSyb-Gal4 drives expression of a UAS-hairpin RNA targeting GFP as a control RNAi) laid around 80.7±6.4 eggs at day five, consistent with the number of eggs laid by wild-type and control females on a rich diet [19,20], and were used as controls for the neuronal neuropeptide screen. By contrast, tubts>GFPdsRNA.143 females laid only 15.1±4.7 eggs at day five of control RNAi induction, much lower than previously reported data [19,20], prompting us to test additional control hairpin RNA lines with tubts. tubts>LucJF01355 females (in which tubts-Gal4 drives expression of a UAS-hairpin RNA targeting Luciferase, Luc, as a control RNAi) laid more eggs at five and 10 days after transgene induction than females with tubts-Gal4 driving other control hairpin RNAs (S1A Fig) and were used as controls for the ubiquitous somatic neuropeptide receptor knockdown screen. We also tested multiple control UAS-hairpin RNA lines with MTD-Gal4 and used MTD>GFPdsRNA.142 females, which laid the most eggs at day five (S1B Fig), as controls for the germline neuropeptide receptor screen.

We initially screened 36 neuropeptides and 46 total neuropeptide receptors (45 of which were screened in somatic cells and 20 in the germline) based on the design above. Knockdown of four neuropeptide receptors in the germline resulted in a statistically significant decrease in egg production on two different timepoints with at least one UAS-hairpin RNA line (S2 Fig and S1 File). Candidates from the initial neuropeptide neuronal knockdown or somatic neuropeptide receptor knockdown screens were tested a second time, and this secondary screen included additional genes that had not been initially screened (due to the late arrival of the corresponding fly stocks) (S1 File). Knockdown of four neuropeptides from the neuronal RNAi screen and 16 neuropeptide receptors from the somatic RNAi screen resulted in statistically significant decreases in egg production (Fig 1C and 1D and S1 File). Ubiquitous somatic InR knockdown resulted in nearly zero eggs laid and served as a positive control, and, consistent with previously published results [21], somatic knockdown of ETHR led to decreased egg production with two out of three independent RNAi lines. However, although the egg count assay allowed us to screen a total of 82 genes and over 100 UAS-hairpin RNA lines, there was not only a large variability in egg production among control lines (see above), but also among different UAS-hairpin RNA lines targeting the same gene. [For example, the number of eggs laid on day five with ubiquitous somatic knockdown of torso with four different UAS-hairpin RNA lines, all from the TRiP collection (fgr.hms.harvard.edu), ranged from 55% to 163% of the number of eggs laid by LucJF01355 control.] These results indicated that the egg counting assay was excessively noisy and should ideally be validated with multiple UAS-hairpin RNA lines, genetic mutants, and additional ovarian analysis. Nevertheless, Diuretic hormone 31 (Dh31) and Diuretic hormone 31 Receptor (Dh31-R), which encode a ligand-receptor pair, emerged as candidates from the neuronal and somatic screens, respectively, leading us to focus next on these genes.

Dh31/Dh31-R do not regulate Drosophila oogenesis

To directly examine the potential roles of Dh31 and Dh31-R in oogenesis, we performed additional RNAi knockdown and genetic mutant analyses. We first determined knockdown efficiency of UAS-Dh31-R hairpin and UAS-Dh31 hairpin lines (Fig 2A and 2B) driven by tubts for seven days using RT-PCR. Dh31-RJF01945 did not induce knockdown of Dh31-R, but both Dh31-RGD3782 and Dh31-RKK108756 resulted in approximately 40% knockdown of Dh31-R (S3A Fig). Dh31HMS02354 resulted in an approximately 80% decrease in Dh31 mRNA levels, the three Dh31GD4601 lines led to a 20–30% decrease, but neither of the two Dh31GD16889 lines led to a strong reduction in Dh31 mRNA levels (S3B Fig). Ubiquitous somatic knockdown of Dh31-R led to a modest reduction in egg number compared to Luc knockdown control (S3C Fig), whereas neuronal knockdown of Dh31 with all hairpin RNA constructs led to a statistically significant decrease in egg laying on day five (S3D Fig). However, the severity of the phenotype did not correlate with the level of knockdown observed. For example, although Dh31HMS02354 resulted in the strongest Dh31 knockdown, nSyb>Dh31HMS02354 females laid more eggs than other females with less severe Dh31 knockdown (S3B and S3D Fig). Thus, the RNAi analysis was inconclusive.

thumbnail
Download:
Fig 2. Dh31 and Dh31-R do not appear to regulate Drosophila oogenesis.

(A) Schematic of the Dh31-R gene, showing three mRNA isoforms for Dh31-R, which differ in the length of protein coding sequences in the last exon, with isoform RC having the longest protein-coding sequence and isoform RB having the shortest protein-coding sequence. The transposable element insertions Dh31-Rf05546 and Dh31-Rf06589 map to the 5’ intronic region. The UAS-Dh31-RJF01945 hairpin targets the common 5’ UTR of Dh31-R transcripts, while UAS-Dh31-RGD3782 and UAS-Dh31-RKK108756 target protein-coding regions shared by Dh31-R transcripts. The Df(2R)Exel7124 deficiency uncovers the entire protein-coding region of Dh31-R. (B) Schematic of the Dh31 gene, showing two mRNA isoforms, which differ by three bases in the third exon. Dh31KG09001 is a strong hypomorphic allele resulting from a P element insertion in the third intron, and Dh3151 is a null allele created by imprecise excision of the KG09001 P element, resulting in a 735 bp deletion that removes most of the protein coding region. The UAS-Dh31HMS02354 hairpin targets the 5’ UTR, UAS-Dh31GD4601 targets the coding region, and UAS-Dh31GD16889 targets the coding region and the 3’ UTR. Deficiencies Df(2L)ED623 and Df(2L)Exel7038 uncover the entire Dh31 gene region. (C-F) RT-qPCR analysis of Dh31-R (C), Dh31 (D-E), and Pdf (F) transcript levels in seven-day-old female heads. Three biological replicates were analyzed for each genotype, with ten heads per biological replicate. (G,H) Egg counts for Dh31-R (G) and Dh31 (H) mutants at 25°C on days five, 10, and 15 after eclosion, showing no consistent differences between homozygous mutants and heterozygous controls. (I) Germarium at five days of knockdown by tubts-driven LucJF01355 (top) or Dh31-RGD3782 (bottom), both showing a dying germline cyst labeled with ApopTag (green). Dying cysts found in LucJF01355 control and Dh31-R knockdown germaria are visually similar based on ApopTag staining. DAPI (blue) labels nuclei. Scale bar, 10 μm. (J) Percentage of ApopTag-positive germaria in females with ubiquitous somatic knockdown of Dh31-R or Luc control at zero or five days of RNAi. (K-L) Percentage of ApopTag-positive germaria in five-day-old Dh31-R (K) and Dh31 (L) mutant females compared to heterozygous controls at 25°C. Three biological replicates were analyzed per genotype, with 100 germaria per biological replicate. *p<0.05; **p<0.01, Student’s t-test. Data shown as mean±s.e.m.

https://doi.org/10.1371/journal.pone.0243756.g002

Owing to the inconsistencies and modest phenotypes observed in Dh31 and Dh31-R knockdown females, we proceeded to analyze Dh31 and Dh31-R genetic mutants (Fig 2A and 2B). Using RT-qPCR, we confirmed that Dh31-Rf06589 and Dh31-Rf05546 [22,23] are hypomorphic alleles (Fig 2C), and that Dh3151 [24] and Dh31KG09001 [25,26] are null and hypomorphic alleles, respectively (Fig 2D and 2E). Additionally, because published genetic evidence suggests that the neuropeptide PDF may also signal through Dh31-R [23], we tested Dh31 and Pdf double mutants and confirmed that Pdf01 is a hypomorphic allele (Fig 2F) [27]. Using these validated genetic alleles, we did not observe any consistent decreases in egg production of Dh31-R or Dh31 mutant females compared to heterozygous controls at 25°C or 29°C (the temperature at which previous RNAi experiments had been performed) (Fig 2G and 2H and S4 Fig). There were no significant differences in egg laying between Dh31 and Pdf single and double mutants at 25°C (S5A Fig); however, Dh3151; Pdf01 double mutants lay significantly fewer eggs at 29°C compared to either of the single mutants (S5B Fig). Dissection of Dh3151; Pdf01 ovaries after five days at 29°C revealed that the vast majority of ovarioles in some ovaries have more than one mature stage 14 egg, identified by their dorsal appendages (S5C Fig), suggesting that the decrease in egg laying is due to egg retention. This may indicate a potential role for Dh31 and Pdf in ovulation.

In parallel to the egg count experiments above, we also performed some initial characterization of dissected ovaries to determine if specific steps of oogenesis might be disrupted by Dh31-R loss-of-function. While we did not notice any obvious changes in overall ovariole morphology, five days of ubiquitous somatic knockdown of Dh31-R using Dh31-RGD3782 (but not Dh31-RKK108756) increased the numbers of dying early germline cysts (Fig 2I and 2J). Adding to these inconsistent results, analysis of Dh31-R, Dh31, and Dh3151; Pdf01 mutants did not show any significant increase in the percent of germaria containing dying germline cysts at either 25°C or 29°C compared to controls (Fig 2K and 2L and S6 Fig). Taken together, these data suggest that Dh31 and Dh31-R do not regulate oogenesis.

moody is likely required in somatic cells for proper maintenance of GSCs

We next focused on the orphan G protein coupled receptor (GPCR) moody. Ubiquitous somatic knockdown of moody with two out of three different hairpin RNA lines resulted in decreased egg production on days five, 10, and 15 of RNAi induction (Fig 1D and S1 File). Importantly, moody was identified in a separate RNAi-based screen aimed at identifying GPCRs with roles in regulating GSCs. In this screen, we tested a small subset of GPCRs that included five neuropeptide receptors (Table 1). We used either the ubiquitous somatic tubts or the germline driver nanos-Gal4::VP16 (referred to as nos-Gal4) [28] to drive UAS-hairpin RNAs against individual GPCRs and analyzed dissected ovaries for GSC loss or proliferation phenotypes at 10 days of RNAi knockdown. We identified AstC-R1 and moody as potential regulators of GSC proliferation, based on the increased frequency of EdU-positive GSCs in females with germline-specific AstC-R1 knockdown or ubiquitous somatic moody knockdown relative to controls (Table 1). However, these results were not reproducible for either AstC-R1 or moody when EdU incorporation experiments were repeated using larger sample sizes and additional UAS-hairpin RNA lines with MTD-Gal4 or tubts, respectively (S7 Fig). Nevertheless, the screen also showed that ubiquitous somatic moody knockdown using two different UAS-moody hairpin lines resulted in dramatic increase in GSC loss compared to control RNAi (Table 1), suggesting a potential role for moody in GSC maintenance.

thumbnail
Download:
Table 1. Results from RNAi-based screen for candidate GPCRs regulating GSC number and/or proliferation at 10 days of RNAi.

https://doi.org/10.1371/journal.pone.0243756.t001

moody encodes an orphan neuropeptide receptor (homolog of human melatonin receptor 1A and 1B, MTNR1A and MTNR1B [29]) required for integrity of the blood-brain barrier [30,31] and with no known roles in oogenesis. To follow up on the screen results suggesting a role for moody in GSC maintenance, we first performed RT-qPCR (using RNA from whole flies) to measure knockdown efficiency of the three UAS-moody hairpin lines expressed in the soma, which target distinct regions of moody (Fig 3A). Ubiquitous knockdown of moody driven by tubts resulted in a 50–60% decrease in moody mRNA levels compared to LucJF01355 control (Fig 3B). KK UAS-hairpin RNA lines were created by site-specific insertion into the unannotated 30B site; however, a significant number of KK lines have an additional insertion at 40D, which causes ectopic expression of the transcription factor Tiptop and can lead to secondary effects [32,33], including loss of GSCs [34]. Therefore, we confirmed that the moodyKK100674 line has only the 30B insertion, and not the 40D insertion (Fig 3C), using previously described PCR methods [32].

thumbnail
Download:
Fig 3. moody appears to function in somatic cells to regulate GSC maintenance.

(A) Schematic showing moody gene and its six mRNA isoforms. The hairpin lines UAS-moodyHMC06237 and UAS-moodyGD709 target distinct portions of the protein coding regions shared by all transcripts, UAS-moodyKK100674 targets part of the protein coding region and part of the 3’ UTR, and UAS-moodyGL01050 (optimized for germline expression) targets a portion of the 3’ UTR sequences shared by all transcripts. (B) RT-qPCR analysis of moody from ovariectomized females at seven days of ubiquitous somatic moody or Luc control RNAi. Three biological replicates were analyzed, with 10 females per replicate. (C) Left: Schematic of expected PCR band sizes for KK lines. Right: PCR gel showing that moodyKK100674 is inserted in the non-annotated 30B site, and not in the annotated 40D site. (D) Germaria from females at five days of ubiquitous somatic knockdown of Luc control or moody (using moodyGD709). DAPI (blue) labels nuclei. LamC (green), nuclear lamina of cap cells; α-Spectrin (green), fusome. Cap cells, arrows; GSCs, solid outlines. Scale bar, 10 μm. (E) Line graph showing the average number of GSCs per germarium at zero, five, 10, and 15 days of ubiquitous somatic knockdown of moody or control Luc. (F) Bar graph showing the percentage of germaria with zero to one, two, or more than three GSCs on the left y-axis and the average number of GSCs per germaria on the right y-axis at zero, five, 10, and 15 days of ubiquitous somatic knockdown of moody or control Luc. Same data are shown in (E and F). (G) Line graph showing the average number of GSCs per germarium at zero, five, 10, and 15 days after eclosion for UAS-alone controls of moody or Luc hairpin RNA lines. (H) Line graph showing the average number of cap cells per germarium at zero, five, 10, and 15 days of ubiquitous somatic knockdown of moody or control Luc. (I) Line graph showing the average number of GSCs per germarium at zero, five, 10, and 15 days of hh-Gal4-driven knockdown of moody or control Luc. For (E-I), three biological replicates were analyzed for each genotype. The total number of germaria per genotype per timepoint for (E,F,H) are indicated in (F). For (G,I), a total of 300 germaria were analyzed per genotype per timepoint. *p<0.05; **p<0.01; ****p<0.0001, ANOVA: two-factor with replication. Data shown as mean±s.e.m.

https://doi.org/10.1371/journal.pone.0243756.g003

Using these validated RNAi lines, we showed that ubiquitous somatic moody knockdown using UAS-moodyGD709 and UAS-moodyKK100674, but not UAS-moodyHMC06237, dramatically increased the rate of GSC loss compared to LucJF01355 control (Fig 3D–3F). Incidentally, ubiquitous somatic moody knockdown using UAS-moodyGD709 and UAS-moodyKK100674 (but not UAS-moodyHMC06237) also decreased egg production (Fig 1D). It is possible that UAS-moodyHMC06237 driven by tubts does not cause GSC loss due to the presence of a moody suppressor in the genetic background. Alternatively, the UAS-moodyHMC06237 insertion might not be well expressed (to induce efficient knockdown of moody) in the specific tissue where moody is required for GSC maintenance. (See Materials and Methods for description of insertion sites of UAS-moody hairpin lines.) In fact, there is clear evidence that a given UAS transgene can be expressed at relatively different levels in distinct tissues depending on insertion site (e.g. [35]). On the other hand, it is unlikely that the GSC loss resulting from ubiquitous expression of UAS-moodyGD709 or UAS-moodyKK100674 represents off-target effects, as these hairpin lines target different regions of moody (Fig 3A). Moreover, the GSC loss phenotype was entirely dependent on induction of these hairpin lines by tub-Gal4 because no GSC loss occurred when we analyzed UAS-alone controls (Fig 3G), further supporting the conclusion that moody is likely required somatically for GSC maintenance.

We next asked in what somatic tissue moody might be required for GSC maintenance. Because moody controls the blood-brain barrier [36] and insulin signaling controls GSC maintenance (through effects on cap cell numbers [11]), we first wondered if misregulation of insulin-like peptides produced in the brain might underlie the somatic moody GSC loss phenotype. However, we found no difference in cap cell numbers between moody and Luc control knockdown females (Fig 3H), indicating that somatic moody controls GSC numbers independently of changes in niche size (which is controlled by insulin signaling [11]), and thus ruling out effects on insulin signaling as a relevant mechanism. moody is widely expressed in adult Drosophila females, including at low levels in the ovary [37]. Previous single cell sequencing analysis of the adult ovary has detected moody expression in stretched cells (which cover the nurse cells in later follicles [38]) and corpus luteum cells [39], although expression in other ovarian populations of cells (e.g. niche) remained possible. Although somatic moody knockdown causes GSC loss in the absence of changes in cap cell numbers, it is conceivable that moody acts in cap cells or other nearby somatic cells to regulate GSC numbers. To determine if moody is required in neighboring somatic cells to regulate GSC numbers, we knocked down moody or Luc control using the niche driver hh-Gal4 [40], which is expressed in terminal filament cells, cap cells, and escort cells [35], and measured GSC numbers at zero, five, 10, and 15 days. However, we did not observe any differences in the rate of GSC loss with moody knockdown compared to Luc RNAi control (Fig 3I), indicating that moody is not required in somatic cells in the germarium to regulate GSC numbers. Given that the ovary is regulated by extensive inter-organ communication [9], it is conceivable that Moody might control GSC number by functioning in another tissue/organ through intermediate systemic modulators.

Discussion

Neuropeptide signaling lies at the junction of reproduction and physiology [2,8,9]. In Drosophila females, insulin signaling has been shown to be critical for oogenesis [9], and SP/NPF signaling is responsible for increased GSC proliferation in response to mating [14,15]. Ecdysis-triggering hormone (ETH) persists in Inka cells in adults and regulates juvenile hormone (JH) synthesis and thus vitellogenesis [21]. Additionally, mating-induced changes in behavior are relayed by neuropeptide signaling [8]. For example, myoinhibiting peptide precursor (MIP) regulates female food preference upon mating [41], and enhanced Dh44 signaling delays sperm ejection in females [42]. In Drosophila males, corazonin (Crz) regulates ejaculation [43], and PDF/NPF signaling regulates mating duration in response to the presence of rival males [44]. In mammals, kisspeptin has emerged as a key regulator of GnRH production and secretion [45]. In addition to systemic signals such as leptin and insulin, multiple neuropeptides also regulate kisspeptin production itself [5]. Neurokinin B (NKB) activates kisspeptin-producing neurons [46], in addition to directly stimulating GnRH release [47]. NPY null mice have lower levels of Kiss1 mRNA [48], while intracerebroventricular injection of NPY in male rats significantly increases Kiss1 mRNA levels [49]. However, much remains to be learned about how this large family of signaling molecules and their receptors communicate physiological state and regulate reproduction. In this study, we found that the orphan neuropeptide receptor moody (which encodes the homolog of mammalian melatonin receptors) is likely required in the soma for GSC maintenance. Additionally, the results from our four screens show that while egg count assays are useful for screening large numbers of genes for severe oogenesis phenotypes, they do not adequately capture more subtle changes in oogenesis that can be detected through the detailed analysis of dissected ovaries.

Screening for regulators of reproductive physiology can be challenging

Physiology relies on the convergence of multiple inputs/signaling pathways that coordinately regulate cellular processes/organ function throughout the body, including oogenesis [9]. Unlike the case for developmental phenotypes, where individual mutations often lead to severe blocks in development, genetic manipulations in a single physiological signaling pathway can often lead to relatively small phenotypic changes (e.g. in rates of certain oogenesis processes), which reflect its partial contribution in the context of a multitude of other integrated physiological inputs. It is therefore challenging to screen for genes regulating the physiology of oogenesis/reproduction. This study used two different screening strategies aimed at identifying potential physiological regulators of oogenesis: egg counting and ovary dissection analysis.

Egg counting has been successfully used previously to identify regulators of fecundity and oogenesis. For example, egg count assays were used to show the negative effect of toxic chemicals such as cadmium and bisphenol A on egg production [50,51]. The effect of poor diet on egg production has also been well documented [19]. Egg counting has also been successfully used to identify amino acid transporters necessary in adipocytes and nuclear receptors required in the soma for egg production [17,20]. By contrast, our RNAi-based screens using egg counting as a readout indicate that this assay can have low signal-to-noise ratio and thus be unreliable. In particular, egg production was very sensitive to genetic background, as different control UAS-hairpin RNA lines resulted in vastly different numbers of eggs laid (S1 Fig). The variability with genetic background is not surprising (given the large number of inputs that affect oogenesis) and has been previously reported by others [52]. The low signal-to-noise ratio makes the use of multiple UAS-hairpin RNA lines (to rule out off-target effects and ensure that phenotype penetrance correlates with knockdown efficiency) and other available genetic tools essential when using egg counts as a screening assay. Notably, while egg counts can be very useful for identifying genes or conditions that have large effects in oogenesis (e.g. InR and ETHR), our screens showed that the noise in the assay can overwhelm signals from more mild changes in oogenesis. In accordance, a previous study from our group found that ubiquitous somatic knockdown of seven up (svp, which encodes a nuclear receptor) led to decreased egg laying with effects in GSC maintenance, germline cyst survival, and vitellogenesis [17]. Interestingly, svp knockdown in adipocytes led to increased GSC loss and early germline cyst death, but no measurable effect on egg production, whereas svp knockdown in oenocytes caused degeneration of vitellogenic follicles and a reduction in number of eggs laid [17], indicating that egg count assays may not adequately capture more subtle changes in earlier steps of oogenesis. By contrast, we found that ubiquitous somatic knockdown of moody has a very strong effect on GSC maintenance (Fig 3D–3F) with consistent decreases in egg laying (Fig 1D). Similarly, insulin signaling in both the soma and germline are important for many processes during oogenesis, including vitellogenesis [9], and, consistent with those roles, knockdown of InR led to a significant reduction in egg laying in our screens (Fig 1D and S2 Fig). This again demonstrates that egg count assays can capture more dramatic ovarian phenotypes, although not necessarily more subtle phenotypes, which require analysis of dissected ovaries.

Dissection-based screens, despite their high resolving power, bring their own challenges. The fact that disruptions in specific signaling pathways might only have subtle effects on the rate of oogenesis processes demands larger sample sizes to identify these effects. While relatively large effects such as those seen with ubiquitous somatic moody knockdown on GSC maintenance were readily detectable even with one biological replicate, that replicate still comprised 80 germaria analyzed per genotype per timepoint. However, for phenotypes such as GSC proliferation (which relies on the frequencies of proliferation markers within the populations of GSCs analyzed), often hundreds of GSCs need to be analyzed in each experimental replicate to unambiguously determine whether the effect is indeed present. In fact, in this study, while both AstC-R1 and moody seemed to regulate GSC proliferation based on our initial screen, we were unable to recapitulate the phenotype when using larger sample sizes, which included over 430 GSCs per genotype (S7 Fig). Thus, the large sample sizes needed to reliably screen for these phenotypes makes for very labor- and time-intensive screens. Despite these challenges, given how closely reproduction is tied to overall physiology, it remains vitally important to identify additional genes controlling the physiology of oogenesis. In particular, the importance of understanding how the brain sends physiological inputs to the ovary warrants further exploration of the complex roles of neuropeptide signaling in oogenesis. In future similar efforts to identify new ovarian regulators, dissection-based screens focusing on specific steps of oogenesis should ideally strike an optimal balance of labor and time invested in analyzing a moderate number of (rationally selected) candidate genes using sufficiently large sample sizes.

Complex roles of neuropeptide signaling in whole-body physiology and reproduction

Neuropeptides are evolutionarily conserved signaling molecules that regulate a wide variety of physiological processes in organisms ranging from C. elegans to Drosophila to humans [6,8,53]. In addition to a key role in regulating the hypothalamic-pituitary-gonadal axis and reproduction, mammalian neuropeptides also control circadian rhythm, water reabsorption, feeding behavior, stress, immunity, and even alcohol intake [54–58]. Activation of NPY signaling, for example, increases food intake and decreases stress and anxiety [54]. In Drosophila, NPF also regulates food intake, metabolism, and aggression, while other conserved neuropeptide orthologs, including Hugin (Neuromedin U homolog), SIFamide (gonadotropin inhibiting hormone GnIH homolog), and Drosulfakinin DSK (Cholecystokinin CCK homolog), regulate feeding behavior, taste and olfaction, learning and behavior, sleep, nociception, and alcohol tolerance [8]. However, only a handful of neuropeptide signaling pathways have been implicated in Drosophila oogenesis. Our egg count screens identified some potentially interesting candidates such as AstC-R1, AstC-R2, CCHa1-R, TrissinR, CG13995, hec, PK2-R2, CCKLR-17D3, CG33639, Lgr4 in somatic cells, and ETHR and Pdfr signaling in the germline. In particular, CCKLR-17D3 binds DSK [59], which regulates feeding behavior and satiety [60], and somatic knockdown of CCKLR-17D3 with four different RNAi lines led to decreased egg production compared to Luc RNAi control (Fig 1D). Therefore, it will be informative to test whether these preliminary findings can be reproduced using additional UAS-hairpin RNA lines and genetic mutants, and, if so, what steps of oogenesis are controlled by these signaling pathways.

Given the low signal-to-noise ratio in our egg count screens, it is also possible that we failed to identify some neuropeptide signaling pathways that might impact oogenesis. For instance, neuropeptides such as SIFa promote feeding and food intake [61] and, given the known effects of diet on oogenesis, disruptions in SIFa signaling pathways would be expected to affect egg production; however, our screen presumably was not sufficiently sensitive to detect these predicted effects. In addition to the roles of individual neuropeptide pathways in oogenesis, if and how neuropeptide signaling pathways crosstalk to regulate egg production should also be explored. For example, we found that Dh3151; Pdf01 double mutants, but not Dh3151 or Pdf01 single mutants, retain mature oocytes at 29°C, resulting in decreased egg laying (S5C Fig). Both Dh31 and Pdf regulate circadian rhythm [62], and Drosophila egg laying is circadian regulated [63]. Although ablation of PDF-producing neurons does not affect egg laying circadian rhythm [64], it is possible that DH31 and PDF may act in redundant pathways to regulate circadian-regulated egg-laying behavior. Finally, although we were specifically interested in how neuropeptides produced in neurons regulate oogenesis, neuropeptides can have additional sites of production besides the nervous system [8]; therefore, it would be sensible to use a ubiquitous somatic driver such as tub-Gal4 (instead of the neuronal-specific nSyb-Gal4 driver) to more broadly identify neuropeptides (originating from any cell type) with potential roles in oogenesis.

A potentially novel role for Moody/MTNR in oogenesis and regulation of stem cell number

Drosophila moody is required in glial cells to maintain the blood-brain barrier and mediate behavioral responses to cocaine [30,31]. Interestingly, moody is also required in the glia for proper male courtship behavior [65]. Nonetheless, moody is broadly expressed in adult Drosophila females [37]. Our results indicate that moody is not required in the niche for GSC maintenance (Fig 3I). It would be interesting to test whether moody is required in the glia for GSC maintenance and whether or not that involves the regulation of the blood-brain barrier. Additionally, moody is most strongly expressed in the spermatheca [37], and it is unknown what roles moody may be playing there to regulate oogenesis. There is currently no known ligand for moody. The closest human orthologs of moody are melatonin receptors 1A and 1B (MTNR1A and MTNR1B) [29], which bind melatonin, a key regulator of circadian rhythm in vertebrates [66]. Like moody, melatonin receptors are expressed throughout the body, including in the central nervous system and in peripheral tissues such as the intestine, adipocytes, immune cells, epithelial tissues, ovary/granulosa cells, and myometrium [67]. In women, exogenous melatonin suppresses LH secretion and blocks ovulation via regulation of the hypothalamic-pituitary-gonadal axis, while melatonin binding to melatonin receptors on granulosa cells increases LH mRNA levels [67]. In culture, melatonin can also promote the proliferation of spermatogonial stem cell (SSCs) and mesenchymal stem cells (MSCs) [68,69]. In addition, given the pleiotropic role of melatonin signaling in mammals, it would be important to find out if melatonin receptors are present in SSCs and MSCs and if melatonin signaling regulates stem cell populations in vivo. Similarly, it would be interesting to identify the Moody ligand in Drosophila and determine its cellular source and requirement for GSC maintenance; to determine if Moody (and its ligand) are required for the function/behavior of other stem cells; to investigate mechanisms downstream of Moody controlling stem cells; and to pinpoint what external or physiological conditions modulate its production/secretion.

Materials and methods

Drosophila strains, culture conditions, and RNAi-based screen

Stocks were maintained at room temperature (22–25°C) on standard medium consisting of cornmeal, molasses, yeast, and agar. Standard medium supplemented with wet yeast paste was used for all experiments, except for egg count assays (see below). S1 Table lists all mutant and transgenic Drosophila lines, including Gal4 drivers, used in the study. Dh31-Rf06589 and Dh31-Rf05546 were backcrossed to isogenized y w for four generations and balanced over CyO. The Dh31-R deficiency line w1118; Df(2R)Exel7124/CyO was not backcrossed because there is no visible eye marker to readily track the deletion [70]. There are two landing sites for the construct used in KK UAS-hairpin RNA lines. The majority of KK lines (75%) has a single insertion at the unannotated 30B site [32,33]. However, approximately 25% of the KK library has an additional insertion of the UAS-hairpin RNA construct at the 40D landing site, which can cause non-specific phenotypes due to ectopic expression of the Tiptop transcription factor [32,33]. We confirmed landing site occupancy of moodyKK100674 using the previously described PCR-based method [32] (Fig 3C). UAS-moodyGD709 is a P element based transgene randomly inserted on the second chromosome [71], while UAS-moodyKK100674 and UAS-moodyHMC06237 were site-specifically inserted on chromosome 2L and 3L, respectively [32,72]. Other genetic elements are described in FlyBase (www.flybase.org).

For pan-neuronal neuropeptide knockdown, control and experimental nSyb>hairpin RNA females were raised at 18°C to minimize Gal4 expression during development [73]. Ubiquitous somatic knockdown was achieved using tubts [17], which combines tub-Gal4 [74] with a temperature sensitive allele of the Gal4 inhibitor tub-Gal80ts [75]. To screen for neuropeptide receptors required in the soma for egg production, females carrying tubts and UAS-neuropeptide receptor hairpin RNA or UAS-LuciferaseJF01355 were raised at 25°C. For germline-specific neuropeptide receptor knockdown, MTD>neuropeptide receptor hairpin RNA or MTD>GFPdsRNA.142 females were raised at 25°C. Upon eclosion, zero- to two-day old females of all genotypes were paired with y w males and shifted to 29°C, to promote nSyb-Gal4, tub-Gal4, and MTD activity, for various lengths of time. In the dissection-based GPCR screen and somatic moody knockdown experiments, females with tubts or nos-Gal4::VP16; tub-Gal80ts and UAS-hairpin RNA were raised at 18°C, the Gal80ts permissive temperature, for inhibition of Gal4 activity during development. [tub-Gal80ts, however, has no effect over nos-Gal4::VP16 activity [76].] Zero-to-two-day old females were collected after eclosion and paired with y w males at 18°C for two-to-three days then shifted to 29°C, the Gal80ts restrictive temperature, for zero, five, 10, or 15 days before dissection. For niche-specific moody knockdown, hh>moody hairpin RNA or LucJF01355 females were raised at 25°C. Zero-to-one-day old females were paired with y w males and shifted to 29°C for zero, five, 10, or 15 days before dissection.

Egg count assay

Five female flies were paired with five y w male flies per biological replicate, with three (pan-neuronal neuropeptide screen) or five (all other egg count experiments) replicates per experiment. Food was provided in the form of wet yeast paste evenly spread on molasses agar plates, and changed once (pan-neuronal neuropeptide screen) or twice (all other egg count experiments) daily. The number of eggs laid in a 24-hour period was counted on days five, 10, and 15 (RNAi), or all 15 days of the experiment (mutants), and the average number of eggs laid per female per day was calculated. Student’s t-test (Microsoft Excel) was used to determine statistical significance.

Ovary dissection and immunostaining

Ovaries were dissected in Grace’s insect medium with L-glutamine (Caisson Labs). After teasing ovarioles apart, ovaries were fixed by nutating for 13 minutes at room temperature in 5.3% formaldehyde (Ted Pella) diluted in Grace’s medium. Ovaries were rinsed three times and washed three times for 15 minutes in PBSTx (PBS; 10 mM NaH2PO4/NaHPO4, 175 mM NaCl, pH 7.4, plus 0.1% Triton X-100) then blocked for three hours in blocking solution consisting of 5% normal goat serum (NGS, MP Biomedicals) and 5% bovine serum albumin (BSA, Sigma-Aldrich) in PBSTx. Samples were then incubated overnight at room temperature in mouse monoclonal anti-α-Spectrin (3A9) (DSHB; Developmental Studies Hybridoma Bank, 1:25) and mouse monoclonal anti-Lamin C (LC28.26) (DSHB, 1:25) in blocking solution. Ovaries were rinsed and washed three times in PBSTx before incubating for two hours at room temperature in Alexa Fluor 488- or 568-conjugated goat anti-mouse secondary antibodies (ThermoFisher Scientific; 1:400) in blocking solution. Following three more rinses and washes in PBSTx, ovaries were mounted in Vectashield with 1.5 μg/mL 4’,6-diamidino-2-phenylindole (DAPI) (Vector Laboratories). Samples were imaged using a Zeiss LSM700 confocal microscope or Zeiss AxioImager-A2 fluorescence microscope. Whole ovary images were obtained with a Zeiss Axiocam ERc 5s camera mounted on a Zeiss Stemi 2000-CS dissecting microscope.

For EdU incorporation, ovaries were dissected in room temperature Grace’s medium and incubated for one hour at room temperature in 100 μM EdU from the Click-iT EdU Alexa Fluor 594 Imaging Kit (ThermoFisher Scientific) in Grace’s medium. Ovarioles were then teased apart, fixed, washed, blocked, and incubated in primary antibody as described above. Samples were subjected to the Click-iT reaction according to manufacturer’s instructions, then rinsed four times and washed four times for 15 minutes before incubating in secondary antibodies, washed, and mounted in Vectashield with DAPI, as described above.

For ApopTag labeling, ovaries were dissected and fixed as described above, then washed for 30 minutes in PBSTx. ApopTag Fluorescein Direct In Situ Apoptosis Detection Kit (Millipore Sigma) was used according to the manufacturer’s instructions. Briefly, ovaries were washed twice for five minutes in equilibration buffer, then incubated for one hour at 37°C in TdT solution (reaction buffer plus TdT enzyme) with occasional resuspension by light tapping. Ovaries were washed in Stop/Wash buffer twice for five minutes, then rinsed and washed three times in PBSTx before mounting in Vectashield with DAPI. To quantify death of early germline cysts, ApopTag-positive germaria were counted as a percentage of all germaria analyzed. Statistical significance of differences in the percentage of ApopTag-positive germaria across three independent experiments (100 germaria per experiment) was determined using Student’s t-test (Microsoft Excel).

GSC and cap cell quantification

Cap cells were identified by their ovoid shape and Lamin C-positive staining of their nuclear lamina. GSCs were identified by their direct contact with cap cells and juxtaposition of their fusome (a specialized organelle labeled by α-Spectrin [77]) to the GSC-cap cell interface, as previously described [78]. Two-way ANOVA with replication (also known as two-way ANOVA with interaction) (Microsoft Excel) was used to determine statistical significance of differences in the rate of cap cell and GSC loss over time for three independent experiments as described [20]. Student’s t-test (Microsoft Excel) was used to determine statistical differences in the percentage of EdU-positive GSCs for each genotype for three independent experiments.

RT-PCR and qRT-PCR

Guts (S3A Fig), heads (Fig 2C–2F and S3B Fig), or ovarectomized females (Fig 3B) were dissected in RNAlater Stabilization Solution (ThermoFisher Scientific) and placed on ice for at least 30 minutes. Gut-derived RNA was used for experiments in S3A Fig to obtain a more robust signal as Dh31-R is most strongly expressed in the gut in adult flies (www.flybase.org). To extract RNA, 250 μL lysis buffer from the RNAqueous-4PCR Total RNA Isolation kit (ThermoFisher Scientific) was added to each sample, and a motorized pestle was used to homogenize tissues. RNA was purified from the homogenate following the manufacturer’s instructions. cDNA was synthesized from 500 ng of total RNA for each sample using SuperScript II Reverse Transcriptase (ThermoFisher Scientific) according to the manufacturer’s instructions. S2 Table lists all primers used in this study. Rp49 primers were used as control. To quantify Dh31 or Dh31-R band intensity, ImageJ was used to measure net band intensity (by subtracting background pixels from band pixels in a fixed-size box) and normalized to the corresponding Rp49 control band. Normalized Rp49 band intensities were set at one, and experimental sample band intensities were normalized to Rp49.

PowerUp SYBR Green Master Mix (ThermoFisher Scientific) was used for quantitative RT-PCR. The reactions were performed in triplicate using the QuantStudio 3 Real-Time PCR System (ThermoFisher Scientific). Examples of amplification and melt curves obtained are plotted in S8 Fig. Amplification fluorescence threshold was determined by QuantStudio 3 software, and ΔΔCt were calculated using Microsoft Excel. Rp49 transcript levels were used as reference. Fold change of transcript levels were calculated using the equation 2-ΔΔCt (Microsoft Excel). y w was used as control for analysis of mutant RNA levels; tubts>LucJF01355 was used as control for analysis of RNAi efficiency.

Supporting information

S1 Fig. Genetic background of control UAS-hairpin RNA lines influences egg production.

(A,B) Average number of eggs laid per female per day for females with tubts (A) or MTD-Gal4 (B) driving different control UAS-hairpin RNA transgenes, raised at 25°C, and switched to 29°C for five, 10, or 15 days. Data shown as mean±s.e.m.

https://doi.org/10.1371/journal.pone.0243756.s001

(TIFF)

S2 Fig. Germline-specific RNAi-based screen for neuropeptide receptors that regulate Drosophila oogenesis.

MTD was used to drive UAS-hairpin RNA against neuropeptide receptor genes, and the number of eggs laid per female per day was counted on days five, 10, and 15. MTD>GFPdsRNA.142 served as negative control. InR knockdown served as an internal control. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001, Student’s t-test. Data shown as mean±s.e.m.

https://doi.org/10.1371/journal.pone.0243756.s002

(TIFF)

S3 Fig. Pan-neuronal Dh31 RNAi and ubiquitous somatic Dh31-R RNAi lead to variable effects on egg production that are not consistent with knockdown efficiency.

(A) Representative gel (left) and quantification (right) of RT-PCR analysis of Dh31-R transcript levels in female guts at seven days of ubiquitous somatic knockdown of Dh31-R or Luc control. Rp49 was used as a control. For each genotype, the ratio of Dh31-R band intensity (1:1 dilution) relative to Rp49 intensity (1:10 dilution) was normalized to that of tubts>LucJF01355, which was arbitrarily set at one. Three biological replicates were used for each genotype, with 10 guts per biological replicate. Gut-derived RNA was used as Dh31-R is most highly expressed in the gut in adults (www.flybase.org). (B) RT-PCR analysis of Dh31 transcript levels in female heads at seven days of ubiquitous somatic knockdown of Dh31 or Luc control. Dh31 relative to Rp49 band intensity normalized to tubts>LucJF01355 control was calculated as described in (A). Three biological replicates were used for each genotype, with 10 heads per biological replicate. Data shown as mean±s.e.m. (C) Graph showing the average number of eggs laid per female per day at five, 10, and 15 days of ubiquitous somatic knockdown of Dh31-R or Luc control. (D) Graph showing the average number of eggs laid per female per day at five, 10, and 15 days of pan-neuronal RNAi of Dh31 or Luc control. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001, Student’s t-test. Data shown as mean±s.e.m.

https://doi.org/10.1371/journal.pone.0243756.s003

(TIFF)

S4 Fig. Dh31 and Dh31-R mutants have similar rates of egg production as heterozygous controls.

(A,B) Line graphs showing the average number of eggs laid per female per day at different days after eclosion for Dh31-R mutants and heterozygous controls at 25°C (A) or 29°C (B). Data from days five, 10, and 15 at 25°C are also shown in Fig 2G. Note that in (B), the two homozygous mutants and transheterozygous mutant lay fewer eggs at 29°C; this is likely due to linked background mutation as neither Dh31-Rf06589/Df(2R)Exel7124 nor Dh31-Rf05546/Df(2R)Exel7124 lay fewer eggs than heterozygous controls. (C,D) Line graphs showing the average number of eggs laid per female per day for Dh31 mutants and heterozygous controls at 25°C (C) or 29°C (D). Data from day five, 10, and 15 at 25°C are also shown in Fig 2H. Data shown as mean±s.e.m.

https://doi.org/10.1371/journal.pone.0243756.s004

(TIFF)

S5 Fig. Dh3151; Pdf01 double mutants lay fewer eggs at 29°C due to mature oocyte retention in their ovaries.

(A,B) Line graphs showing the average number of eggs laid per female per day at different days after eclosion for Dh3151 homozygous, Pdf01 homozygous, or Dh3151; Pdf01 double homozygous females at 25°C (A) or 29°C (B). Data shown as mean±s.e.m. (C) Examples of ovaries from Dh3151, Pdf01, and Dh3151; Pdf01 5-day-old females at 25°C (left) or 29°C (right). Arrows point to multiple dorsal appendages to indicate examples of accumulated mature oocytes.

https://doi.org/10.1371/journal.pone.0243756.s005

(TIFF)

S6 Fig. Dh31, Dh31-R, and Pdf mutants do not show increased levels of early germline cyst death relative to control females.

Percent of germaria showing ApopTag-positive germline cysts in five-day old Dh31-R mutant (A), Dh31 mutant (B), or Dh3151; Pdf01 females at 25°C (C) or 29°C (D). Three biological replicates per genotype, 100 germaria per replicate. **p<0.01, Student’s t-test. Data shown as mean±s.e.m.

https://doi.org/10.1371/journal.pone.0243756.s006

(TIFF)

S7 Fig. Somatic moody and germline AstC-R1 do not appear to regulate GSC proliferation.

(A) Example of germarium from MTD>GFPdsRNA.142 female at 7 days of GFP knockdown showing one EdU-positive GSC (arrowhead) and one EdU-negative GSC. DAPI (blue) labels nuclei. LamC (green), nuclear lamina of cap cells; α-Spectrin (green), fusome; EdU (red), S-phase marker. Cap cells, arrows; GSCs, solid outlines. Scale bar, 10 μm. (B,C) Graphs showing the average percentage of EdU-positive GSCs at zero or 10 days of somatic knockdown of moody or Luc control (B) or at seven days of germline knockdown of AstC-R1 or GFP (C). Two biological replicates for moodyHMC06237, GFPdsRNA.142, and all Astc-R1 RNAi genotypes, and three biological replicates for all other genotypes. The total number of GSCs analyzed are indicated inside bars.

https://doi.org/10.1371/journal.pone.0243756.s007

(TIFF)

S8 Fig. Examples of qPCR amplification and melt curves.

(A) Amplification curves for three biological replicates of Dh31-Rf05546/+, with three technical replicates per biological sample, plotting Rn versus cycle number, with amplification for Dh31-R shown in shades of blue and Rp49 in shades of gray. (B) Melt curves for the same sample of Dh31-Rf05546/+ plotting the derivative of fluorescence intensity versus temperature, with Dh31-R products in shades of blue and Rp49 products in shades of gray. Each line represents one technical replicate of a biological replicate. Biological replicates were collected at different times, and qPCR was performed separately.

https://doi.org/10.1371/journal.pone.0243756.s008

(TIFF)

S1 Table. Transgenic Drosophila lines used in this study.

https://doi.org/10.1371/journal.pone.0243756.s009

(PDF)

S2 Table. Sequences of primers used in this study.

https://doi.org/10.1371/journal.pone.0243756.s010

(PDF)

S1 File. Results from egg count screens.

https://doi.org/10.1371/journal.pone.0243756.s011

(XLSX)

Acknowledgments

T.M. and D.D.-B. designed experiments, analyzed and interpreted data, and wrote the manuscript. The initial dissection-based screen was performed by S.M.; T.M. performed subsequent analyses of moody and all other experiments. We thank Metabel Markwei for her technical assistance with egg count experiments shown in S3 Fig. We are grateful to the Developmental Studies Hybridoma Bank for antibodies (National Institutes of Health [NIH] N01HD073263-000), and to the Bloomington Stock Center (NIH P400D018537), Vienna Drosophila Stock Center, Mark Wu, Erika Matunis, and Fumika Hamada for Drosophila stocks. We thank Lesley N. Weaver and Rodrigo Dutra Nunes for critical reading of the manuscript.

References

  1. 1. Jasienska G, Bribiescas RG, Furberg AS, Helle S, Núñez-de la Mora A. Human reproduction and health: an evolutionary perspective. Lancet. 2017;390: 510–520. pmid:28792413
  2. 2. Kaprara A, Huhtaniemi IT. The hypothalamus-pituitary-gonad axis: Tales of mice and men. Metabolism. 2018;86: 3–17. pmid:29223677
  3. 3. Lainez NM, Coss D. Obesity, Neuroinflammation, and Reproductive Function. Endocrinology. 2019;160: 2719–2736. pmid:31513269
  4. 4. Acevedo-Rodriguez A, Kauffman AS, Cherrington BD, Borges CS, Roepke TA, Laconi M. Emerging insights into hypothalamic-pituitary-gonadal axis regulation and interaction with stress signalling. J Neuroendocrinol. 2018;30: e12590. pmid:29524268
  5. 5. Saedi S, Khoradmehr A, Mohammad Reza JS, Tamadon A. The role of neuropeptides and neurotransmitters on kisspeptin/kiss1r-signaling in female reproduction. J Chem Neuroanat. 2018;92: 71–82. pmid:30008384
  6. 6. Burbach JPH. What Are Neuropeptides? Methods in Molecular Biology. 2011. pp. 1–36. pmid:21922398
  7. 7. Russo AF. Overview of Neuropeptides: Awakening the Senses? Headache. 2017;57: 37–46. pmid:28485842
  8. 8. Nässel DR, Zandawala M. Recent advances in neuropeptide signaling in Drosophila, from genes to physiology and behavior. Prog Neurobiol. 2019;179: 101607. pmid:30905728
  9. 9. Local Drummond-Barbosa D. and physiological control of germline stem cell lineages in drosophila melanogaster. Genetics. 2019;213: 9–26. pmid:31488592
  10. 10. LaFever L, Drummond-Barbosa D. Direct control of germline stem cell division and cyst growth by neural insulin in Drosophila. Science (80-). 2005;309: 1071–1073. pmid:16099985
  11. 11. Hsu HJ, Drummond-Barbosa D. Insulin levels control female germline stem cell maintenance via the niche in Drosophila. Proc Natl Acad Sci U S A. 2009;106: 1117–1121. pmid:19136634
  12. 12. Armstrong AR, Drummond-Barbosa D. Insulin signaling acts in adult adipocytes via GSK-3β and independently of FOXO to control Drosophila female germline stem cell numbers. Dev Biol. 2018;440: 31–39. pmid:29729259
  13. 13. Yang CH, Belawat P, Hafen E, Jan LY, Jan YN. Drosophila egg-laying site selection as a system to study simple decision-making processes. Science (80-). 2008;319: 1679–1683. pmid:18356529
  14. 14. Ameku T, Yoshinari Y, Texada MJ, Kondo S, Amezawa K, Yoshizaki G, et al. Midgut-derived neuropeptide F controls germline stem cell proliferation in a mating-dependent manner. PLoS Biol. 2018;16: e2005004. pmid:30248087
  15. 15. Ameku T, Niwa R. Mating-Induced Increase in Germline Stem Cells via the Neuroendocrine System in Female Drosophila. PLoS Genet. 2016;12: e1006123. pmid:27310920
  16. 16. Pauli A, Althoff F, Oliveira RA, Heidmann S, Schuldiner O, Lehner CF, et al. Cell-Type-Specific TEV Protease Cleavage Reveals Cohesin Functions in Drosophila Neurons. Dev Cell. 2008;14: 239–251. pmid:18267092
  17. 17. Weaver LN, Drummond-Barbosa D. The nuclear receptor seven up functions in adipocytes and oenocytes to control distinct steps of Drosophila oogenesis. Dev Biol. 2019;456: 179–189. pmid:31470019
  18. 18. Mazzalupo S, Cooley L. Illuminating the role of caspases during Drosophila oogenesis. Cell Death Differ. 2006;13: 1950–1959. pmid:16528381
  19. 19. Drummond-Barbosa D, Spradling AC. Stem cells and their progeny respond to nutritional changes during Drosophila oogenesis. Dev Biol. 2001;231: 265–278. pmid:11180967
  20. 20. Armstrong AR, Laws KM, Drummond-Barbosa D. Adipocyte amino acid sensing controls adult germline stem cell number via the amino acid response pathway and independently of Target of Rapamycin signaling in Drosophila. Dev. 2014;141: 4479–4488. pmid:25359724
  21. 21. Meiselman M, Lee SS, Tran RT, Dai H, Ding Y, Rivera-Perez C, et al. Endocrine network essential for reproductive success in Drosophila melanogaster. Proc Natl Acad Sci U S A. 2017;114: E3849–E3858. pmid:28439025
  22. 22. Thibault ST, Singer MA, Miyazaki WY, Milash B, Dompe NA, Singh CM, et al. A complementary transposon tool kit for Drosophila melanogaster using P and piggyBac. Nat Genet. 2004;36: 283–287. pmid:14981521
  23. 23. Goda T, Doi M, Umezaki Y, Murai I, Shimatani H, Chu ML, et al. Calcitonin receptors are ancient modulators for rhythms of preferential temperature in insects and body temperature in mammals. Genes Dev. 2018;32: 140–155. pmid:29440246
  24. 24. Head LM, Tang X, Hayley SE, Goda T, Umezaki Y, Chang EC, et al. The influence of light on temperature preference in Drosophila. Curr Biol. 2015;25: 1063–1068. pmid:25866391
  25. 25. Bellen HJ, Levis RW, Liao G, He Y, Carlson JW, Tsang G, et al. The BDGP gene disruption project: Single transposon insertions associated with 40% of Drosophila genes. Genetics. 2004;167: 761–781. pmid:15238527
  26. 26. Lajeunesse DR, Johnson B, Presnell JS, Catignas KK, Zapotoczny G. Peristalsis in the junction region of the Drosophila larval midgut is modulated by DH31 expressing enteroendocrine cells. BMC Physiol. 2010;10: 14. pmid:20698983
  27. 27. Renn SCP, Park JH, Rosbash M, Hall JC, Taghert PH. A pdf neuropeptide gene mutation and ablation of PDF neurons each cause severe abnormalities of behavioral circadian rhythms in Drosophila. Cell. 1999;99: 791–802. pmid:10619432
  28. 28. Holtzman S, Miller D, Eisman R, Kuwayama H, Niimi T, Kaufman T. Transgenic tools for members of the genus drosophila with sequenced genomes. Fly (Austin). 2010;4: 349–362. pmid:20890112
  29. 29. Agapite J, Albou LP, Aleksander S, Argasinska J, Arnaboldi V, Attrill H, et al. Alliance of Genome Resources Portal: Unified model organism research platform. Nucleic Acids Res. 2020;48: D650–D658. pmid:31552413
  30. 30. Bainton RJ, Tsai LTY, Schwabe T, DeSalvo M, Gaul U, Heberlein U. moody encodes two GPCRs that regulate cocaine behaviors and blood-brain barrier permeability in Drosophila. Cell. 2005;123: 145–156. pmid:16213219
  31. 31. Schwabe T, Bainton RJ, Fetter RD, Heberlein U, Gaul U. GPCR signaling is required for blood-brain barrier formation in Drosophila. Cell. 2005;123: 133–144. pmid:16213218
  32. 32. Green EW, Fedele G, Giorgini F, Kyriacou CP. A Drosophila RNAi collection is subject to dominant phenotypic effects. Nat Methods. 2014;11: 222–223. pmid:24577271
  33. 33. Vissers JHA, Manning SA, Kulkarni A, Harvey KF. A Drosophila RNAi library modulates Hippo pathway-dependent tissue growth. Nat Commun. 2016;7: 10368. pmid:26758424
  34. 34. Matsuoka S, Armstrong AR, Sampson LL, Laws KM, Drummond-Barbosa D. Adipocyte metabolic pathways regulated by diet control the female germline stem cell lineage in Drosophila melanogaster. Genetics. 2017;206: 953–971. pmid:28396508
  35. 35. Weaver LN, Ma T, Drummond-Barbosa D. Analysis of Gal4 expression patterns in adult Drosophila females. G3 Genes, Genomes, Genet. 2020; pmid:32917721
  36. 36. Limmer S, Weiler A, Volkenhoff A, Babatz F, Klämbt C. The drosophila blood-brain barrier: Development and function of a glial endothelium. Front Neurosci. 2014;8: 365. pmid:25452710
  37. 37. Leader DP, Krause SA, Pandit A, Davies SA, Dow JAT. FlyAtlas 2: A new version of the Drosophila melanogaster expression atlas with RNA-Seq, miRNA-Seq and sex-specific data. Nucleic Acids Res. 2018;46: D809–D815. pmid:29069479
  38. 38. Wu X, Tanwar PS, Raftery LA. Drosophila follicle cells: Morphogenesis in an eggshell. Semin Cell Dev Biol. 2008;19: 271–282. pmid:18304845
  39. 39. Jevitt A, Chatterjee D, Xie G, Wang XF, Otwell T, Huang YC, et al. A single-cell atlas of adult Drosophila ovary identifies transcriptional programs and somatic cell lineage regulating oogenesis. PLoS Biol. 2020;18: e3000538. pmid:32339165
  40. 40. Jenett A, Rubin GM, Ngo TTB, Shepherd D, Murphy C, Dionne H, et al. A GAL4-Driver Line Resource for Drosophila Neurobiology. Cell Rep. 2012;2: 991–1001. pmid:23063364
  41. 41. Hussain A, Üçpunar HK, Zhang M, Loschek LF, Grunwald Kadow IC. Neuropeptides Modulate Female Chemosensory Processing upon Mating in Drosophila. PLoS Biol. 2016;14: e1002455. pmid:27145127
  42. 42. Lee KM, Daubnerová I, Isaac RE, Zhang C, Choi S, Chung J, et al. A neuronal pathway that controls sperm ejection and storage in female drosophila. Curr Biol. 2015;25: 790–797. pmid:25702579
  43. 43. Zer-Krispil S, Zak H, Shao L, Ben-Shaanan S, Tordjman L, Bentzur A, et al. Ejaculation Induced by the Activation of Crz Neurons Is Rewarding to Drosophila Males. Curr Biol. 2018;28: 1445–1452.e3. pmid:29681474
  44. 44. Kim WJ, Jan LY, Jan YN. A PDF/NPF neuropeptide signaling circuitry of male Drosophila melanogaster controls rival-induced prolonged mating. Neuron. 2013;80: 1190–1205. pmid:24314729
  45. 45. Harter CJL, Kavanagh GS, Smith JT. The role of kisspeptin neurons in reproduction and metabolism. J Endocrinol. 2018;238: R173–R183. pmid:30042117
  46. 46. Sakamoto K, Murata K, Wakabayashi Y, Yayou KI, Ohkura S, Takeuchi Y, et al. Central administration of neurokinin B activates kisspeptin/NKB neurons in the arcuate nucleus and stimulates luteinizing hormone secretion in ewes during the non-breeding season. J Reprod Dev. 2012;58: 700–706. pmid:22972185
  47. 47. Fergani C, Leon S, Padilla SL, Verstegen AMJ, Palmiter RD, Navarro VM. NKB signaling in the posterodorsal medial amygdala stimulates gonadotropin release in a kisspeptin-independent manner in female mice. Elife. 2018;7: e40476. pmid:30565563
  48. 48. Luque RM, Kineman RD, Tena-Sempere M. Regulation of hypothalamic expression of KiSS-1 and GPR54 genes by metabolic factors: Analyses using mouse models and a cell line. Endocrinology. 2007;148: 4601–4611. pmid:17595226
  49. 49. Azizi V, Oryan S, Khazali H, Hosseini A. Central injection of neuropeptide Y modulates sexual behavior in male rats: interaction with GnRH and kisspeptin/neurokinin B/dynorphin. Int J Neurosci. 2020. pmid:32303141
  50. 50. Vimal D, Saini S, Kristipati RR, Chowdhuri DK. Atrazine or bisphenol A mediated negative modulation of mismatch repair gene, mlh1 leads to defective oogenesis and reduced female fertility in Drosophila melanogaster. Chemosphere. 2019;225: 247–258. pmid:30877919
  51. 51. Hu X, Fu W, Yang X, Mu Y, Gu W, Zhang M. Effects of cadmium on fecundity and defence ability of Drosophila melanogaster. Ecotoxicol Environ Saf. 2019;171: 871–877. pmid:30665104
  52. 52. Evangelou A, Ignatiou A, Antoniou C, Kalanidou S, Chatzimatthaiou S, Shianiou G, et al. Unpredictable effects of the genetic background of transgenic lines in physiological quantitative traits. G3 Genes, Genomes, Genet. 2019;9: 3877–3890. pmid:31540975
  53. 53. Li C, Kim K. Neuropeptide gene families in Caenorhabditis elegans. Adv Exp Med Biol. 2010;692: 98–137. pmid:21189676
  54. 54. Comeras LB, Herzog H, Tasan RO. Neuropeptides at the crossroad of fear and hunger: A special focus on neuropeptide y. Ann N Y Acad Sci. 2019;1455: 59–80. pmid:31271235
  55. 55. Barson JR, Leibowitz SF. Hypothalamic neuropeptide signaling in alcohol addiction. Prog Neuro-Psychopharmacology Biol Psychiatry. 2016;65: 321–329. pmid:25689818
  56. 56. Gizowski C, Trudel E, Bourque CW. Central and peripheral roles of vasopressin in the circadian defense of body hydration. Best Pract Res Clin Endocrinol Metab. 2017;31: 535–546. pmid:29224666
  57. 57. Pinho-Ribeiro FA, Verri WA, Chiu IM. Nociceptor Sensory Neuron–Immune Interactions in Pain and Inflammation. Trends Immunol. 2017;38: 5–19. pmid:27793571
  58. 58. Kormos V, Gaszner B. Role of neuropeptides in anxiety, stress, and depression: From animals to humans. Neuropeptides. 2013;47: 401–419. pmid:24210138
  59. 59. Kubiak TM, Larsen MJ, Burton KJ, Bannow CA, Martin RA, Zantello MR, et al. Cloning and functional expression of the first Drosophila melanogaster sulfakinin receptor DSK-R1. Biochem Biophys Res Commun. 2002;291: 313–320. pmid:11846406
  60. 60. Söderberg JAE, Carlsson MA, Nässel DR. Insulin-producing cells in the Drosophila brain also express satiety-inducing cholecystokinin-like peptide, drosulfakinin. Front Endocrinol (Lausanne). 2012;3: pmid:22969751
  61. 61. Martelli C, Pech U, Kobbenbring S, Pauls D, Bahl B, Sommer MV, et al. SIFamide Translates Hunger Signals into Appetitive and Feeding Behavior in Drosophila. Cell Rep. 2017;20: 464–478. pmid:28700946
  62. 62. Hegazi S, Lowden C, Garcia JR, Cheng AH, Obrietan K, Levine JD, et al. A symphony of signals: Intercellular and intracellular signaling mechanisms underlying circadian timekeeping in mice and flies. Int J Mol Sci. 2019;20: 2363. pmid:31086044
  63. 63. Manjunatha T, Hari Dass S, Sharma VK. Egg-laying rhythm in Drosophila melanogaster. J Genet. 2008;87: 495–504. pmid:19147938
  64. 64. Howlader G, Paranjpe DA, Sharma VK. Non-ventral lateral neuron-based, non-PDF-mediated clocks control circadian egg-laying rhythm in Drosophila melanogaster. J Biol Rhythms. 2006;13: 871–888. pmid:16461981
  65. 65. Hoxha V, Lama C, Chang PL, Saurabh S, Patel N, Olate N, et al. Sex-Specific Signaling in the Blood-Brain Barrier Is Required for Male Courtship in Drosophila. PLoS Genet. 2013;9: e1003217. pmid:23359644
  66. 66. Hardeland R, Cardinali DP, Srinivasan V, Spence DW, Brown GM, Pandi-Perumal SR. Melatonin-A pleiotropic, orchestrating regulator molecule. Prog Neurobiol. 2011;93: 350–384. pmid:21193011
  67. 67. Ekmekcioglu C. Melatonin receptors in humans: Biological role and clinical relevance. Biomed Pharmacother. 2006;60: 97–108. pmid:16527442
  68. 68. Sun TC, Li HY, Li XY, Yu K, Deng SL, Tian L. Protective effects of melatonin on male fertility preservation and reproductive system. Cryobiology. 2020. pmid:32001217
  69. 69. Hu C, Li L. Melatonin plays critical role in mesenchymal stem cell-based regenerative medicine in vitro and in vivo. Stem Cell Res Ther. 2019;10. pmid:30635065
  70. 70. Parks AL, Cook KR, Belvin M, Dompe NA, Fawcett R, Huppert K, et al. Systematic generation of high-resolution deletion coverage of the Drosophila melanogaster genome. Nat Genet. 2004;36: 288–292. pmid:14981519
  71. 71. Dietzl G, Chen D, Schnorrer F, Su KC, Barinova Y, Fellner M, et al. A genome-wide transgenic RNAi library for conditional gene inactivation in Drosophila. Nature. 2007;448: 151–156. pmid:17625558
  72. 72. Perkins LA, Holderbaum L, Tao R, Hu Y, Sopko R, McCall K, et al. The transgenic RNAi project at Harvard medical school: Resources and validation. Genetics. 2015;201: 843–852. pmid:26320097
  73. 73. Duffy JB. GAL4 system in Drosophila: A fly geneticist’s Swiss army knife. Genesis. 2002;34: 1–15. pmid:12324939
  74. 74. Lee T, Luo L. Mosaic analysis with a repressible neurotechnique cell marker for studies of gene function in neuronal morphogenesis. Neuron. 1999;22: 451–461. pmid:10197526
  75. 75. McGuire SE, Le PT, Osborn AJ, Matsumoto K, Davis RL. Spatiotemporal Rescue of Memory Dysfunction in Drosophila. Science (80-). 2003;302: 1765–1768. pmid:14657498
  76. 76. Del Valle Rodríguez A, Didiano D, Desplan C. Power tools for gene expression and clonal analysis in Drosophila. Nat Methods. 2012;9: 47–55. pmid:22205518
  77. 77. Lin H, Yue L, Spradling AC. The Drosophila fusome, a germline-specific organelle, contains membrane skeletal proteins and functions in cyst formation. Development. 1994;120: 947–956. pmid:7600970
  78. 78. De Cuevas M, Spradling AC. Morphogenesis of the Drosophila fusome and its implications for oocyte specification. Development. 1998;125: 2781–2789. pmid:9655801