1. Developmental Biology
  2. Genetics and Genomics
Download icon

Female-biased upregulation of insulin pathway activity mediates the sex difference in Drosophila body size plasticity

  1. Jason W Millington
  2. George P Brownrigg
  3. Charlotte Chao
  4. Ziwei Sun
  5. Paige J Basner-Collins
  6. Lianna W Wat
  7. Bruno Hudry
  8. Irene Miguel-Aliaga
  9. Elizabeth J Rideout  Is a corresponding author
  1. Department of Cellular and Physiological Sciences, Life Sciences Institute, The University of British Columbia, Canada
  2. MRC London Institute of Medical Sciences, and Institute of Clinical Sciences, Faculty of Medicine, Imperial College London, United Kingdom
Research Article
  • Cited 0
  • Views 1,161
  • Annotations
Cite this article as: eLife 2021;10:e58341 doi: 10.7554/eLife.58341

Abstract

Nutrient-dependent body size plasticity differs between the sexes in most species, including mammals. Previous work in Drosophila showed that body size plasticity was higher in females, yet the mechanisms underlying increased female body size plasticity remain unclear. Here, we discover that a protein-rich diet augments body size in females and not males because of a female-biased increase in activity of the conserved insulin/insulin-like growth factor signaling pathway (IIS). This sex-biased upregulation of IIS activity was triggered by a diet-induced increase in stunted mRNA in females, and required Drosophila insulin-like peptide 2, illuminating new sex-specific roles for these genes. Importantly, we show that sex determination gene transformer promotes the diet-induced increase in stunted mRNA via transcriptional coactivator Spargel to regulate the male-female difference in body size plasticity. Together, these findings provide vital insight into conserved mechanisms underlying the sex difference in nutrient-dependent body size plasticity.

Introduction

In insects, the rate of growth during development is influenced by environmental factors such as nutrient availability (Boulan et al., 2015; Edgar, 2006; Hietakangas and Cohen, 2009; Nijhout, 2003; Nijhout et al., 2014). When nutrients are abundant, the growth rate is high and body size is large (Beadle et al., 1938; Edgar, 2006; Mirth and Shingleton, 2012; Nijhout, 2003; Robertson, 1963). When nutrients are scarce, the growth rate is lower and body size is smaller (Beadle et al., 1938; Edgar, 2006; Mirth and Riddiford, 2007; Mirth and Shingleton, 2012; Nijhout, 2003; Robertson, 1963). This ability of an organism or genotype to adjust its body size in line with nutrient availability is a form of phenotypic plasticity (Agrawal, 2001; Garland, 2006). While the capacity of individuals to display nutrient-dependent changes to body size depends on many factors, one important factor that affects phenotypic plasticity is whether an animal is male or female (Stillwell et al., 2010; Teder and Tammaru, 2005). For example, in Drosophila the magnitude of changes to wing cell size and cell number in a nutrient-poor diet were larger in females compared with males (Alpatov, 1930). Similarly, the magnitude of protein- and carbohydrate-induced changes to several morphological traits was larger in female flies (Shingleton et al., 2017). While these studies clearly establish a sex difference in nutrient-dependent phenotypic plasticity, the genetic and molecular mechanisms underlying this increased trait size plasticity in females remain unclear.

Clues into potential mechanisms underlying the increased nutrient-dependent phenotypic plasticity in female flies have emerged from over 20 years of studies on nutrient-dependent growth in Drosophila (Andersen et al., 2013; Boulan et al., 2015; Edgar, 2006; Koyama and Mirth, 2018; Mirth and Piper, 2017). In particular, these studies have identified the conserved insulin/insulin-like growth factor signaling pathway (IIS) as a key regulator of nutrient-dependent growth in Drosophila (Böhni et al., 1999; Britton et al., 2002; Chen et al., 1996; Fernandez et al., 1995; Grewal, 2009; Teleman, 2010). In nutrient-rich conditions, insulin-producing cells (IPCs) in the larval brain release Drosophila insulin-like peptides (Dilps) into the circulation (Brogiolo et al., 2001; Géminard et al., 2009; Ikeya et al., 2002; Rulifson et al., 2002). These Dilps bind the Insulin-like Receptor (InR; FBgn0283499) on target cells to induce receptor autophosphorylation and recruitment of adapter proteins (Almudi et al., 2013; Böhni et al., 1999; Chen et al., 1996; Poltilove et al., 2000; Werz et al., 2009). These adapter proteins enable the recruitment of the regulatory and catalytic subunits of the Drosophila homolog of phosphatidylinositol 3-kinase (Pi3K21B; FBgn0020622 and Pi3K92E; FBgn0015279, respectively), which catalyze the production of phosphatidylinositol (3,4,5)-trisphosphate (PIP3) from phosphatidylinositol (4,5)-bisphosphate (PIP2) (Leevers et al., 1996). Increased plasma membrane PIP3 recruits and activates signaling proteins such as phosphoinositide-dependent kinase 1 (Pdk1; FBgn0020386) and Akt (Akt; FBgn0010379), which influence diverse cellular processes to enhance cell, tissue, and organismal size (Cho et al., 2001; Grewal, 2009; Rintelen et al., 2001; Verdu et al., 1999).

In contrast, when nutrients are scarce, Dilp release from the IPCs is reduced (Géminard et al., 2009), and plasma membrane Pi3K recruitment, PIP3 levels, and Pdk1- and Akt-dependent signaling are all reduced (Britton et al., 2002; Nowak et al., 2013). Together, these changes diminish cell, tissue, and organismal size (Arquier et al., 2008; Britton et al., 2002; Géminard et al., 2009; Honegger et al., 2008; Okamoto et al., 2013; Rulifson et al., 2002; Zhang et al., 2009). Indeed, the potent growth-promoting ability of IIS activation is shown by the fact that increased IIS activity augments body size (Arquier et al., 2008; Goberdhan et al., 1999; Honegger et al., 2008; Ikeya et al., 2002; Nowak et al., 2013; Okamoto et al., 2013; Oldham et al., 2002), whereas reduced IIS activity limits cell, organ, and body size (Böhni et al., 1999; Brogiolo et al., 2001; Chen et al., 1996; Colombani et al., 2003; Gao et al., 2000; Grönke et al., 2010; Leevers et al., 1996; Murillo-Maldonado et al., 2011; Rulifson et al., 2002; Weinkove et al., 1999; Zhang et al., 2009). Because increased IIS activity bypasses the reduction in cell size in low-nutrient conditions (Britton et al., 2002; Géminard et al., 2009; Nowak et al., 2013), and mutations that blunt IIS pathway activity reduce size in nutrient-rich contexts (Böhni et al., 1999; Brogiolo et al., 2001; Chen et al., 1996; Leevers et al., 1996), Drosophila studies have established IIS as one key pathway that promotes organismal growth downstream of nutrient input. While this highlights the impact of Drosophila on our knowledge of how IIS couples nutrient input with growth, it is important to note that most studies used a mixed-sex population of larvae. Given that cell and body size differ significantly between male and female flies (Alpatov, 1930; Brown and King, 1961; Okamoto et al., 2013; Partridge et al., 1994; Rideout et al., 2015; Sawala and Gould, 2017; Testa et al., 2013), more knowledge is needed of nutrient-dependent changes to body size and IIS activity in each sex.

Recent studies have begun to make progress in this area by studying IIS regulation and function in both sexes in a single dietary context (reviewed in Millington and Rideout, 2018). For example, in late third instar larvae, there are sex differences in dilp mRNA levels, IIS activity, and Drosophila insulin-like peptide 2 (Dilp2; FBgn0036046) secretion from the IPCs (Rideout et al., 2015; McDonald et al., 2020). Similarly, transcriptomic studies have detected male-female differences in mRNA levels of genes associated with IIS function (Mathews et al., 2017; Rideout et al., 2015), and revealed links between IIS and the sex determination hierarchy gene regulatory network (Castellanos et al., 2013; Chang et al., 2011; Clough et al., 2014; Fear et al., 2015; Garner et al., 2018; Goldman and Arbeitman, 2007). As increasing evidence of sex-specific IIS regulation accumulates, several reports reveal sex-limited and sex-biased phenotypic effects caused by changes to IIS function. Changes to IIS activity in larvae show sex-biased effects on growth and final body size (Grönke et al., 2010; Rideout et al., 2015; Shingleton et al., 2005; Testa et al., 2013; Millington et al., 2021), and there are widespread sex-specific and sex-biased changes to gene expression in adult flies with altered diet and IIS activity (Camus et al., 2019; Graze et al., 2018). Further, sex differences exist in how changes to diet and IIS activity affect life span (Bjedov et al., 2010; Clancy, 2001; Giannakou et al., 2004; Grönke et al., 2010; Regan et al., 2016; Tatar et al., 2001; Woodling et al., 2020; Wu et al., 2020). Together, these studies illuminate the utility of Drosophila in revealing sex-specific IIS regulation and describing the physiological impact of this regulation. Yet, more studies are needed to discover the molecular mechanisms underlying sex-specific IIS regulation, and to extend these studies beyond a single nutritional context.

Additional insights into male-female differences in the regulation of cell, tissue, and body size arise from studies on sex determination genes. In Drosophila, sex is determined by the number of X chromosomes. In XX females, a functional splicing factor called Sex-lethal (Sxl; FBgn0264270) is produced (Bell et al., 1988; Bridges, 1921; Cline, 1978; Salz and Erickson, 2010). Sxl-dependent splicing of transformer (tra; FBgn0003741) pre-mRNA allows a functional Tra protein to be produced in females (Belote et al., 1989; Boggs et al., 1987; Inoue et al., 1990; Sosnowski et al., 1989). In XY males, the lack of a functional Sxl protein causes the default splicing of tra pre-mRNA, and no functional Tra protein is produced in males (Cline and Meyer, 1996; Salz and Erickson, 2010; Belote et al., 1989; Boggs et al., 1987; Inoue et al., 1990; Sosnowski et al., 1989). The presence of functional Sxl and Tra proteins in females account for most aspects of female sexual development, behavior, and physiology (Anand et al., 2001; Billeter et al., 2006; Brown and King, 1961; Camara et al., 2008; Christiansen et al., 2002; Clough et al., 2014; Dauwalder, 2011; Demir and Dickson, 2005; Goodwin et al., 2000; Hoshijima et al., 1991; Hudry et al., 2016; Hudry et al., 2019; Ito et al., 1996; Millington and Rideout, 2018; Neville et al., 2014; Nojima et al., 2014; Pavlou et al., 2016; Pomatto et al., 2017; Regan et al., 2016; Rezával et al., 2014; Rezával et al., 2016; Rideout et al., 2010; Ryner et al., 1996; Sturtevant, 1945; von Philipsborn et al., 2014). Recently, new roles for Sxl and Tra in regulating body size were also described. While female flies are normally larger than males, females lacking neuronal Sxl were smaller than control females, and not different in size from males (Sawala and Gould, 2017). Similarly, females lacking a functional Tra protein were smaller than control females; however, these tra mutant females were still larger than males (Brown and King, 1961; Mathews et al., 2017; Rideout et al., 2015). Together, these studies indicate that Tra and Sxl are required to promote a larger body size in females; however, much remains to be discovered about the mechanisms by which Sxl and Tra impact body size. Moreover, which sex determination genes contribute to the male-female difference in diet-induced trait size plasticity remains unknown, as studies on sex determination genes used a single diet.

In the present study, we aimed to improve knowledge of the genetic and molecular mechanisms that contribute to male-female differences in nutrient-dependent phenotypic plasticity in Drosophila. Our detailed examination of body size revealed increased phenotypic plasticity in females in response to a protein-rich diet, in line with studies on plasticity in other traits (Shingleton et al., 2017). We discovered that a female-biased upregulation of IIS activity was responsible for the larger body size of females raised on a protein-rich diet. Mechanistically, we show that the nutrient-dependent upregulation of stunted (sun; FBgn0014391) mRNA levels by transcriptional coactivator Spargel (Srl; FBgn0037248) in females triggers the diet-induced increase in IIS activity, as females with reduced sun do not augment IIS activity or body size in a protein-rich diet. Importantly, we show that sex determination gene tra is required for the nutrient-dependent increase in sun mRNA, IIS activity, and phenotypic plasticity in females, and that Srl represents a key link between Tra and regulation of sun mRNA levels. In males, ectopic Tra expression confers nutrient-dependent body size plasticity via Srl-mediated regulation of sun mRNA levels and IIS activity. Together, these results provide new insight into the molecular mechanisms that govern male-female differences in body size plasticity, and identify a previously unrecognized role for sex determination gene tra in regulating nutrient-dependent phenotypic plasticity.

Results

High levels of dietary protein are required for increased nutrient-dependent body size plasticity in females

Previous studies identified a sex difference in nutrient-dependent plasticity in several morphological traits (Shingleton et al., 2017; Stillwell et al., 2010; Teder and Tammaru, 2005). To determine whether sex differences in nutrient-dependent body size plasticity exist in Drosophila, we measured pupal volume, an established readout for Drosophila body size (Delanoue et al., 2010), in white1118 (w; FBgn0003996) males and females reared on diets of varying nutrient quantity. We found that pupal volume in w1118 female larvae raised on the two-acid diet (1X) (Lewis, 1960) was significantly larger than genotype-matched females raised on a diet with half the nutrient quantity (0.5X) (Figure 1—figure supplement 1A). In w1118 males, pupal volume was also significantly larger in larvae raised on the 1X diet compared with the 0.5X diet (Figure 1—figure supplement 1A). No significant sex-by-diet interaction was detected using a two-way analysis of variance (ANOVA) (sex:diet interaction p=0.7048; Supplementary file 1), suggesting that nutrient-dependent body size plasticity was not different between the sexes in this context. We next compared pupal volume in w1118 males and females raised on the 1X diet with larvae cultured on a diet with twice the nutrient content (2X). Pupal volume in w1118 females was significantly larger in larvae raised on the 2X diet compared with larvae cultured on the 1X diet (Figure 1—figure supplement 1A). In w1118 males, the magnitude of the nutrient-dependent increase in pupal volume was smaller compared with female larvae (Figure 1—figure supplement 1A; sex:diet interaction p<0.0001; Supplementary file 1). This suggests that in nutrient-rich conditions, there is a sex difference in phenotypic plasticity, where nutrient-dependent body size plasticity is higher in females. To represent the normal body size responses of each sex to nutrient quantity, we plotted reaction norms for pupal volume in w1118 males and females raised on different diets (Figure 1—figure supplement 1B). The body size response to increased nutrient quantity between 0.5X and 1X was not different between the sexes (Figure 1—figure supplement 1B); however, the body size response to increased nutrient quantity between 1X and 2X was larger in females than in males (Figure 1—figure supplement 1B). Importantly, these findings were not specific to pupal volume, as we reproduced our findings using adult weight as an additional readout for body size (Figure 1A,B). Thus, our findings demonstrate that while phenotypic plasticity is similar between the sexes in some nutritional contexts, body size plasticity is higher in females than in males in a nutrient-rich environment.

Figure 1 with 7 supplements see all
Upregulation of IIS activity is required for increased nutrient-dependent body size plasticity in females in a protein-rich diet.

(A) Adult weight was significantly higher in w1118 males and females cultured on 1X compared with flies raised on 0.5X (p<0.0001 for both sexes; two-way ANOVA followed by Tukey HSD test). The magnitude of this increase in adult weight was the same in both sexes (sex:diet interaction p=0.3197; two-way ANOVA followed by Tukey HSD test). Adult weight was significantly higher in w1118 females raised on 2X compared to flies cultured on 1X; however, male adult weight was not significantly increased (p<0.0001 and p=0.4015, respectively; two-way ANOVA followed by Tukey HSD test), where the diet-dependent increase in adult weight was higher in females (sex:diet interaction p=0.0003; two-way ANOVA followed by Tukey HSD test). (B) Reaction norms for adult weight in response to changes in nutrient quantity in w1118 females and males, plotted using the data presented in panel A. n = 6–11 groups of 10 flies. (C) Adult weight was significantly higher in females cultured on 2Y compared with flies raised on 1Y; however, male adult weight was not significantly higher in flies raised on 2Y compared with males cultured on 1Y (p<0.0001 and p=0.7199, respectively; two-way ANOVA followed by Tukey HSD test, sex:diet interaction p<0.0001). (D) Reaction norms for adult weight in w1118 females and males reared on either 1Y or 2Y, plotted using data from panel C. n = 7–11 groups of 10 flies. (E) In females, mRNA levels of Foxo targets (insulin receptor (InR), brummer (bmm), and eukaryotic initiation factor 4E-binding protein (4E-BP)), were significantly lower in larvae raised on a protein-rich diet (2Y) compared with larvae raised on a diet containing half the protein content (1Y) (p<0.0001; Student’s t test). n = 8 biological replicates. (F) Quantification of the ratio between cell surface membrane-associated green fluorescent protein (GFP) and cytoplasmic GFP (GFP ratio [M:C]) in a dissected fat body of female larvae from the GFP-PH strain. The ratio was significantly higher in female larvae cultured on 2Y compared with larvae raised on 1Y (p=0.001; Student’s t test). n = 18 biological replicates. (G) In males, there was no significant difference in mRNA levels of Foxo targets between larvae raised on 2Y compared with larvae cultured on 1Y (p=0.7323; Student’s t test). n = 6–7 biological replicates. (H) In males, the M:C ratio for GFP-PH was not significantly different between males cultured on 2Y compared with larvae raised on 1Y (p=0.0892; Student’s t test). n = 15–18 biological replicates. (I) Pupal volume was significantly higher in both w1118 females and InRE19/+ females reared on 2Y compared with genotype-matched females cultured on 1Y (p<0.0001 for both genotypes; two-way ANOVA followed by Tukey HSD test); however, the magnitude of the nutrient-dependent increase in pupal volume was lower in InRE19/+ females (genotype:diet interaction p<0.0001; two-way ANOVA followed by Tukey HSD test). n = 58–77 pupae. (J) Pupal volume was significantly higher in both w1118 males and InRE19/+ males reared on 2Y compared with genotype-matched males cultured on 1Y (p<0.0001 for both genotypes; two-way ANOVA followed by Tukey HSD test). While we observed a sex:diet interaction in the w1118 control genotype, there was no sex:diet interaction in the InRE19/+ genotype (p<0.0001 and p=0.7104, respectively; two-way ANOVA followed by Tukey HSD test). n = 47–76 pupae. For body size plasticity graphs, filled circles indicate mean body size, and dashed lines indicate 95% confidence interval. *** indicates p<0.001, **** indicates p<0.0001; ns indicates not significant; error bars indicate SEM.

To narrow down macronutrients that account for the increased body size plasticity in females, we changed individual food ingredients and measured body size in w1118 males and females. We first altered dietary yeast, as previous studies show that yeast is a key source of protein and an important determinant of larval growth (Britton et al., 2002; Géminard et al., 2009; Robertson, 1963). In w1118 females raised on a diet with yeast content that corresponds to the amount in the 2X diet (2Y diet), pupal volume was significantly larger than in females raised on a diet containing half the yeast content (1Y) (Figure 1—figure supplement 1C). It is important to note that the yeast and calorie content of the 1Y diet was within the range of standard diets used in many larval growth studies (22.65 g/L vs. 21–46 g/L and 586 calories/L vs 459–760 calories/L, respectively) (Ghosh et al., 2014; Koyama and Mirth, 2016; Marshall et al., 2012; Sawala and Gould, 2017), and therefore does not represent a nutrient-restricted diet. In w1118 males, the magnitude of the nutrient-dependent increase in pupal volume was smaller than in females (Figure 1—figure supplement 1C; sex:diet interaction p=0.0001; Supplementary file 1), suggesting that nutrient-dependent body size plasticity was higher in females in a yeast-rich context. Indeed, when we plotted reaction norms for pupal volume in both sexes, the magnitude of the yeast-dependent change in pupal volume (Figure 1—figure supplement 1D) and adult weight (Figure 1C,D) was larger in females than in males. This sex difference in phenotypic plasticity in a yeast-rich context was reproduced in Canton-S (CS), a wild-type strain (Figure 1—figure supplement 2A,B), and using wing length as an additional measure of size (Figure 1—figure supplement 3A). Thus, our findings indicate that the male-female difference in nutrient-dependent body size plasticity persists across multiple genetic backgrounds, and confirms that body size is a robust trait to monitor nutrient-dependent phenotypic plasticity.

Given the sex difference in body size plasticity in response to altered yeast content, we hypothesized that yeast may trigger increased nutrient-dependent body size plasticity in females. To test this, we raised larvae on diets with altered sugar (Figure 1—figure supplement 4A) or calorie content (Figure 1—figure supplement 4B). Because we observed no sex:diet interaction for either manipulation (sex:diet interaction p=0.6536 and p=0.3698, respectively; Supplementary file 1), this suggests dietary yeast mediates the sex difference in nutrient-dependent body size plasticity. To test whether protein is the macronutrient in yeast that enables sex-specific phenotypic plasticity, we pharmacologically limited protein breakdown by culturing larvae on the 2Y diet supplemented with either a broad-spectrum protease inhibitor (protease inhibitor cocktail; PIC) or a serine protease-specific inhibitor (4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride; AEBSF). Previous studies suggest that these inhibitors are specific, as the growth-inhibitory effect of these protease inhibitors was buffered by feeding larvae with bacteria that enhance intestinal protease mRNA levels and gut proteolytic activity (Erkosar et al., 2015). While we found a significant body size reduction in both sexes treated with protease inhibitors (Figure 1—figure supplement 5A,B), in line with previous studies (Erkosar et al., 2015), the magnitude of the inhibitor-induced decrease in pupal volume was larger in female larvae than in males (sex:treatment interaction p=0.0029 [PIC] and p<0.0001 [AEBSF]; Supplementary file 1). This indicates that yeast-derived dietary protein is the macronutrient that augments nutrient-dependent body size plasticity in females. While two potential explanations for the male-female difference in body size plasticity are a sex difference in food intake or length of the growth period, we found no differences in either phenotype between w1118 male and female larvae cultured on 1Y or 2Y (Figure 1—figure supplement 6A–C). Moreover, the larger body size of female larvae does not explain their increased nutrient-dependent body size plasticity, as a genetic manipulation that augments male body size did not enhance phenotypic plasticity (Figure 1—figure supplement 7A,B). Taken together, our data reveals female larvae have enhanced body size plasticity in a nutrient-rich context, and identifies abundant dietary protein as a prerequisite for females to maximize body size.

The nutrient-dependent upregulation of IIS activity in females is required to achieve a larger body size in a protein-rich context

In a mixed-sex population of Drosophila larvae, IIS activity is positively regulated by nutrient availability to promote growth (Böhni et al., 1999; Britton et al., 2002; Chen et al., 1996; Fernandez et al., 1995; Grewal, 2009; Teleman, 2010). We therefore examined nutrient-dependent changes to IIS activity in larvae raised on 1Y and 2Y (Figure 1E–H). Previous studies show that high levels of IIS activity repress mRNA levels of several genes via transcription factor Forkhead box, sub-group O (Foxo; FBgn0038197) (Alic et al., 2011; Jünger et al., 2003; Kang et al., 2017; Puig and Tjian, 2005; Zinke et al., 2002). We therefore assessed mRNA levels of known Foxo target genes InR, brummer (bmm, FBgn0036449), and eukaryotic initiation factor 4E-binding protein (4E-BP, FBgn0261560) together to quantify IIS activity in each sex and dietary context, an established approach to analyze coregulated genes (Blaschke et al., 2013; Hudry et al., 2019). In w1118 females, mRNA levels of Foxo target genes were significantly lower in larvae reared on 2Y than in larvae raised on 1Y (Figure 1E). This suggests IIS activity is significantly higher in females raised on 2Y than in females cultured on 1Y. To confirm this, we used the localization of a ubiquitously-expressed green fluorescent protein (GFP) fused to a pleckstrin homology (PH) domain (GFP-PH) as an additional readout of IIS activity. Because high levels of IIS activity raise plasma membrane PIP3, and PH domains bind specifically to PIP3, larvae with elevated IIS activity show increased membrane localization of GFP-PH (Britton et al., 2002). We observed a significantly higher membrane localization of GFP-PH in females cultured on 2Y than in female larvae raised on 1Y (Figure 1F). Together with increased Foxo target gene repression in 2Y, this GFP-PH data indicates that females reared on 2Y have higher IIS activity than females cultured on 1Y. In males, the magnitude of the nutrient-dependent change in Foxo target genes was smaller than in females (Figure 1G), as we detected a significant sex:diet interaction for Foxo target genes (p=0.0007; Supplementary file 1). Indeed, there was no significant increase in GFP-PH membrane localization between males raised on 2Y and males reared on 1Y (Figure 1H). Taken together, these results reveal a previously unrecognized female-biased upregulation of IIS activity in a protein-rich context.

To determine whether increased IIS activity is required in females for the ability to maximize body size on a protein-rich diet, we measured pupal volume in larvae heterozygous for a hypomorphic mutation in the InR gene (InRE19/+) that were raised in either 1Y or 2Y. Previous studies have shown that while overall growth is largely normal in InRE19/+ heterozygous animals, growth that requires high levels of IIS activity is blunted (Chen et al., 1996; Rideout et al., 2012; Rideout et al., 2015). In w1118 control females, larvae cultured on 2Y were significantly larger than larvae raised on 1Y (Figure 1I); however, the magnitude of this protein-dependent increase in pupal volume was smaller in InRE19/+ females (Figure 1I; genotype:diet interaction p<0.0001; Supplementary file 1). This suggests that nutrient-dependent body size plasticity was reduced in InRE19/+ females. Indeed, while we observed a sex difference in phenotypic plasticity in the w1118 control genotype (sex:diet interaction p<0.0001 Supplementary file 1), the sex difference in nutrient-dependent body size plasticity was abolished in the InRE19/+ genotype (Figure 1I,J; sex:diet interaction p=0.7104; Supplementary file 1). Together, these results indicate that the nutrient-dependent upregulation of IIS activity in females is required for them to achieve a larger body size in a protein-rich context, and that the sex difference in body size plasticity arises from the female-biased upregulation of IIS activity in a protein-rich context.

dilp2 is required for the nutrient-dependent upregulation of IIS activity and a larger body size in females raised on a protein-rich diet.

Previous studies have identified changes to the production and release of Dilps as important mechanisms underlying nutrient-dependent changes to IIS activity and body size (Colombani et al., 2003; Géminard et al., 2009; Zhang et al., 2009). For example, mRNA levels of Drosophila insulin-like peptide 3 (dilp3; FBgn0044050) and Drosophila insulin-like peptide 5 (dilp5; FBgn0044048), but not dilp2, decrease in response to nutrient withdrawal (Colombani et al., 2003; Géminard et al., 2009; Ikeya et al., 2002), and the release of Dilps 2, 3, and 5 from the IPCs is altered by changes in nutrient availability (Géminard et al., 2009; Kim and Neufeld, 2015). Levels of Dilp2 also fluctuate during larval development (Slaidina et al., 2009). Interestingly, a recent study suggests that late third-instar female larvae have increased Dilp2 secretion compared with age-matched males when the larvae were raised in a diet equivalent to 2Y (Rideout et al., 2015). Given that Dilp2 is an important growth-promoting Dilp (Grönke et al., 2010; Ikeya et al., 2002), we tested whether dilp2 was required in females for the nutrient-dependent upregulation of IIS activity. In control w1118 females, mRNA levels of Foxo target genes were significantly lower in larvae raised on 2Y than in larvae reared on 1Y (Figure 2A), suggesting a nutrient-dependent increase in IIS activity. In contrast, mRNA levels of Foxo target genes were not significantly lower in dilp2 mutant female larvae raised on 2Y compared with genotype-matched females cultured on 1Y (Figure 2A), suggesting that loss of dilp2 in females eliminated the nutrient-dependent increase in IIS activity. The magnitude of the nutrient-dependent decrease in Foxo target gene expression was smaller in w1118 males compared with w1118 females (Figure 2B, sex:diet interaction p=0.0511; Supplementary file 1), but not in dilp2 mutant males compared with genotype-matched females (sex:diet interaction p=0.6754; Supplementary file 1). This indicates that dilp2 loss blocks the female-biased upregulation of IIS activity in a protein-rich diet.

Figure 2 with 4 supplements see all
Drosophila insulin-like peptide 2 is required for the nutrient-dependent upregulation of insulin pathway activity and increased female body size plasticity.

(A) In control w1118 females, mRNA levels of Foxo targets (insulin receptor (InR), brummer (bmm), and eukaryotic initiation factor 4E-binding protein (4E-BP)), were significantly lower in larvae cultured on a protein-rich diet (2Y) compared with larvae raised on a diet containing half the protein (1Y) (p<0.0001; Student’s t test). In dilp2 mutant females, there was no significant difference in mRNA levels of Foxo targets in larvae cultured on 2Y compared with larvae raised on 1Y (p=0.2231 Student’s t test). n = 8 biological replicates. (B) In control w1118 and dilp2 mutant males, mRNA levels of Foxo targets were significantly lower in larvae cultured on 2Y compared with larvae raised on 1Y (p=0.0066 and p=0.0023 respectively; Student’s t test). n = 7–8 biological replicates; however, the magnitude of the reduction in Foxo target gene expression in w1118 males was smaller than in genotype-matched females. (C) Adult weight was significantly higher in w1118 females raised on 2Y compared with flies cultured on 1Y (p<0.0001; two-way ANOVA followed by Tukey HSD test); however, adult weight was not significantly different between dilp2 mutant females reared on 2Y versus 1Y (p=0.1263; two-way ANOVA followed by Tukey HSD test). n = 7–11 groups of 10 flies. (D) Adult weight in control w1118 and dilp2 mutant males was not significantly higher in flies reared on 2Y compared with males raised on 1Y (p=0.8366 and p=0.8817, respectively; two-way ANOVA followed by Tukey HSD test). There was a significant sex:diet interaction in the control w1118 genotype (p<0.0001), but not in the dilp2 mutant genotype (p=0.0827; two-way ANOVA followed by Tukey HSD test). n = 10–12 groups of 10 flies. (E) Pupal volume was significantly higher in w1118 females but not in dilp2 mutant females reared on 2Y compared with genotype-matched females cultured on 1Y (p<0.0001 and p=0.6486 respectively; two-way ANOVA followed by Tukey HSD test). The magnitude of the nutrient-dependent increase in pupal volume was higher in w1118 females (genotype:diet interaction p<0.0001; two-way ANOVA followed by Tukey HSD test). n = 74–171 pupae. (F) Pupal volume was significantly higher in w1118 males and dilp2 mutant males reared on 2Y compared with genotype-matched males cultured on 1Y (p<0.0001 for both genotypes; two-way ANOVA followed by Tukey HSD test). The magnitude of the nutrient-dependent increase in pupal volume was not different between genotypes (genotype:diet interaction p=0.6891; two-way ANOVA followed by Tukey HSD test). n = 110–135 pupae. (G) Pupal volume was significantly reduced in females upon RNAi-mediated knockdown of dilp2 in 2Y when compared to both control genotypes (p<0.0001 [da>+], and p=0.002 [+>UAS-dilp2-RNAi], respectively; two-way ANOVA followed by Tukey HSD test), but not in males in 2Y (p<0.0001 [da>+], and 0.9634 [+>UAS-dilp2-RNAi], respectively; two-way ANOVA followed by Tukey HSD test). The magnitude of the effect of RNAi-mediated knockdown of dilp2 on pupal volume was higher in females (sex:genotype interaction p=0.003; two-way ANOVA followed by Tukey HSD test). n = 44–59 pupae. For all body size plasticity graphs, filled circles indicate mean body size, and dashed lines indicate 95% confidence interval. ** indicates p<0.01, **** indicates p<0.0001; ns indicates not significant; error bars indicate SEM.

To determine whether the inability to augment IIS activity on 2Y affects the nutrient-dependent increase in female body size, we measured body size in w1118 and dilp2 mutant larvae cultured on either 1Y or 2Y. In w1118 control females, adult weight was significantly higher in flies cultured on 2Y compared with flies raised on 1Y (Figure 2C); however, this nutrient-dependent increase in adult weight was not observed in dilp2 mutant females (Figure 2C; genotype:diet interaction p=0.0024; Supplementary file 1). In w1118 control males and dilp2 mutant males, there was no significant increase in adult weight in flies raised on 2Y compared with genotype-matched flies cultured on 1Y (Figure 2D; genotype:diet interaction p=0.935; Supplementary file 1). Indeed, in contrast to the sex difference in nutrient-dependent body size plasticity in the w1118 genotype (sex:diet interaction p<0.0001; Supplementary file 1), the sex difference in phenotypic plasticity was abolished in the dilp2 mutant genotype (sex:diet interaction p=0.0827; Supplementary file 1). Importantly, we replicated all these findings using pupal volume (Figure 2E,F), reproduced the female-specific effects of dilp2 loss by globally overexpressing a UAS-dilp2-RNAi transgene (Figure 2G), and show that dilp2 loss does not alter feeding behavior (Figure 2—figure supplement 1A). While we did not determine a sex difference in circulating Dilp2 levels in larvae with an endogenously tagged dilp2 allele due to body size plasticity defects in this strain (Park et al., 2014; Figure 2—figure supplement 2A,B), an experiment that will be important to repeat in future using alternative ways of measuring circulating Dilp2, we show that changes to dilp mRNA levels in males and females lacking dilp2 (Figure 2—figure supplement 3A,B), and nutrient-dependent changes to dilp mRNA levels (Figure 2—figure supplement 4A,B), were similar in both sexes. Together, our data reveals a previously unrecognized female-specific requirement for dilp2 in triggering a nutrient-dependent increase in IIS activity and body size in a protein-rich context.

A nutrient-dependent increase in stunted mRNA levels is required for enhanced IIS activity and a larger body size in females cultured in a protein-rich diet 

Nutrient-dependent changes in Dilp secretion from the IPCs, and consequently IIS activity, are mediated by humoral factors that are regulated by dietary nutrients (Britton and Edgar, 1998; Delanoue et al., 2016; Koyama and Mirth, 2016; Rajan and Perrimon, 2012; Rodenfels et al., 2014; Sano et al., 2015). For example, in a mixed-sex population of larvae, dietary protein augments mRNA levels of Growth-blocking peptides 1 and 2 (Gbp1, FBgn0034199; Gbp2, FBgn0034200), CCHamide-2 (CCHa2; FBgn0038147), unpaired 2 (upd2; FBgn0030904), and sun (Delanoue et al., 2016; Koyama and Mirth, 2016; Rajan and Perrimon, 2012; Sano et al., 2015). Increased levels of these humoral factors promote the secretion of IPC-produced Dilps to enhance IIS activity and growth (Delanoue et al., 2016; Koyama and Mirth, 2016; Meschi et al., 2019; Rajan and Perrimon, 2012; Sano et al., 2015). To determine whether any humoral factors contribute to the sex-biased increase in IIS activity in a protein-rich diet, we examined mRNA levels of each factor in larvae of both sexes raised on either 1Y or 2Y. In w1118 females, sun mRNA levels in larvae reared on 2Y were significantly higher than in larvae cultured on 1Y (Figure 3A). In contrast, mRNA levels of Gbp1, Gbp2, CCHa2, and upd2 were not significantly higher in female larvae reared on 2Y compared with 1Y (Figure 3B). Thus, while previous studies have shown that mRNA levels of all humoral factors were severely reduced by a nutrient-restricted diet or nutrient withdrawal (Delanoue et al., 2016; Koyama and Mirth, 2016; Rajan and Perrimon, 2012; Sano et al., 2015), our study suggests that for most factors, augmenting dietary protein beyond a widely used level does not further enhance mRNA levels. In males, there was no significant increase in sun mRNA levels (Figure 3C), or any other humoral factors (Figure 3D), in larvae reared on 2Y compared with 1Y. Thus, there is a previously unrecognized sex difference in the regulation of sun mRNA levels in a protein-rich context, which we confirm leads to a sex difference in circulating Sun levels (Figure 3—figure supplement 1A).

Figure 3 with 8 supplements see all
stunted is required for the nutrient-dependent upregulation of insulin pathway activity and increased female body size plasticity.

(A) In females, mRNA levels of stunted (sun)RA, but not sunRB, were significantly higher in larvae cultured on a protein-rich diet (2Y) compared with larvae raised on a diet containing half the protein (1Y) (p=0.0055 and p=0.2327, respectively; Student’s t test). n = 8 biological replicates. (B) mRNA levels of Growth-blocking peptide 1 (Gbp1) were significantly different in females cultured on 2Y compared with females raised in 1Y (p=0.0245; Student’s t test); however, mRNA levels of Growth-blocking peptide 2 (Gbp2), CCHamide-2 (CCHa2), and unpaired 2 (upd2) were not significantly different between female larvae raised on 1Y and 2Y (p=0.0662, 0.1416, and 0.7171, respectively; Student’s t test). n = 7–8 biological replicates. (C) In males, mRNA levels of sunRA and sunRB were not significantly different in larvae raised on 2Y compared with larvae raised on 1Y (p=0.5832 and p=0.2017, respectively; Student’s t test). n = 7–8 biological replicates. (D) Levels of Gbp1 and upd2 were not significantly different between male larvae raised on 2Y compared with larvae reared on 1Y (p=0.1487, and p=0.1686, respectively; Student’s t test); whereas levels of Gbp2 and CCHa2 were significantly different between males raised in 2Y and 1Y (p=0.0214, and p=0.0272, respectively; Student’s t test). n = 7–8 biological replicates. (E) In control r4>+, and +>sun-RNAi females, mRNA levels of Foxo targets (insulin receptor (InR), brummer (bmm), and eukaryotic initiation factor 4E-binding protein (4E-BP)), were significantly lower in larvae cultured on 2Y compared with larvae raised on 1Y (p<0.0001, for both comparisons; Student’s t test). However, in r4>sun-RNAi females, there was no significant difference in Foxo target mRNA levels (p=0.2792; Student’s t test). n = 8 biological replicates. (F) In control r4>+, and +>sun-RNAi males, mRNA levels of Foxo targets were significantly lower in larvae cultured on 2Y compared with larvae raised on 1Y (p<0.0001 and p=0.0001, respectively; Student’s t test). While r4>sun-RNAi males showed no significant difference in Foxo target mRNA levels (p=0.2469; Student’s t test), there was no genotype:diet interaction among males (p=0.1068), suggesting that genotype had no impact on Foxo target genes. Importantly, there was a significant sex:diet interaction for Foxo target mRNA levels in both the r4>+ control (p=0.0166; two-way ANOVA followed by Tukey HSD test) and +>sun-RNAi control (p=0.0119; two-way ANOVA followed by Tukey HSD test), but not in r4>sun-RNAi larvae (p=0.1121; two-way ANOVA followed by Tukey HSD test). n = 7–8 biological replicates. (G) Adult weight was significantly higher in female flies raised in 2Y compared with females raised in 1Y in r4>+ and +>UAS-sun-RNAi controls (p<0.0001 for both genotypes; two-way ANOVA followed by Tukey HSD test); however, adult weight was not significantly different between r4>UAS-sun-RNAi females reared on 2Y compared with genotype-matched females raised on 1Y (p=0.5035; two-way ANOVA followed by Tukey HSD test). n = 7–10 groups of 10 flies. (H) Adult weight was not significantly higher in male flies reared in 2Y compared with males cultured in 1Y for r4>+ and +>UAS-sun-RNAi controls or r4>UAS-sun-RNAi males (p=0.8883, 0.6317, and 0.554, respectively; two-way ANOVA followed by Tukey HSD test). There was a significant sex:diet interaction in the r4>+ and +>UAS-sun-RNAi control genotypes (p=0.011 and p=0.0005, respectively; two-way ANOVA followed by Tukey HSD test), but no sex:diet interaction in the r4>UAS-sun-RNAi genotype (p=0.8749; two-way ANOVA followed by Tukey HSD test). n = 6–9 groups of 10 flies. For all body size plasticity graphs, filled circles indicate mean body size, and dashed lines indicate 95% confidence interval. * indicates p<0.05, ** indicates p<0.01, *** indicates p<0.001 **** indicates p<0.0001; ns indicates not significant; error bars indicate SEM.

Given that a comprehensive series of genetic, molecular, and organ co-culture experiments have established that Sun promotes IIS activity by enhancing Dilp2 secretion (Delanoue et al., 2016), we hypothesized that the female-specific increase in sun mRNA levels in 2Y triggers the nutrient-dependent upregulation of IIS activity in females. To test this, we overexpressed UAS-sun-RNAi in the larval fat body using r4-GAL4, and cultured the animals on either 1Y or 2Y. Importantly, overexpression of the UAS-sun-RNAi transgene significantly decreased sun mRNA levels in both sexes (Figure 3—figure supplement 2A,B), where GAL4 expression was similar between the sexes in 1Y and 2Y (Figure 3—figure supplement 2C). In control r4>+ and +>UAS-sun-RNAi females, we observed a significant decrease in Foxo target gene expression in larvae cultured on 2Y compared with genotype-matched larvae reared on 1Y (Figure 3E). In contrast, the nutrient-dependent decrease in Foxo target gene expression was absent in r4>UAS-sun-RNAi females (Figure 3E; diet:genotype interaction p<0.0001; Supplementary file 1), suggesting sun is required in females for the nutrient-dependent increase in IIS activity. In males, the magnitude of the nutrient-dependent decrease in Foxo target gene expression was smaller than in genotype-matched females for the r4>+ and +>UAS-sun-RNAi control strains (p=0.0166 [r4>+]; p=0.0119 [+>UAS-sun-RNAi]; Supplementary file 1), but not in the r4>UAS-sun-RNAi strain (Figure 3F) (sex:diet interaction p=0.1121 [r4>UAS-sun-RNAi]; Supplementary file 1). Importantly, the lack of a diet:genotype interaction among males indicates that there was no effect of genotype on Foxo target gene expression (p=0.1068; Supplementary file 1). Together, this data suggests that in females a protein-rich diet stimulates a nutrient-dependent increase in sun mRNA that promotes IIS activity. In males, the 2Y diet did not augment sun mRNA levels, suggesting one reason for the female-biased increase in IIS activity in a protein-rich diet.

We next asked whether the female-specific increase in sun mRNA and its impact on IIS activity contribute to the nutrient-dependent increase in female body size in a protein-rich context. In r4>+ and +>UAS-sun-RNAi control females, adult weight was significantly higher in flies cultured on 2Y compared with genotype-matched flies raised on 1Y (Figure 3G). In contrast, the nutrient-dependent increase in adult weight was abolished in r4>UAS-sun-RNAi females (Figure 3G; genotype:diet interaction p=0.0014; Supplementary file 1). This indicates r4>UAS-sun-RNAi females have reduced nutrient-dependent body size plasticity, a finding that cannot be explained by changes to feeding behavior (Figure 3—figure supplement 3A). In r4>+, +>UAS-sun-RNAi, and r4>UAS-sun-RNAi male flies raised on 2Y, adult weight was not significantly higher than in genotype-matched males raised on 1Y (Figure 3H; genotype:diet interaction p=0.9278; Supplementary file 1). Importantly, in contrast to the sex difference in nutrient-dependent body size plasticity we observed in the r4>+ and +>UAS-sun-RNAi control genotypes (sex:diet interaction p=0.011 and p=0.0005, respectively; Supplementary file 1), the sex difference in phenotypic plasticity was abolished in the r4>UAS-sun-RNAi genotype (sex:diet interaction p=0.8749; Supplementary file 1), findings we reproduced using pupal volume (Figure 3—figure supplement 4A,B). While we observed no phenotypic plasticity effects in larvae with whole-body, pan-neuronal, or IPC loss of Sun receptor methuselah (mth; Fbgn0023000; Delanoue et al., 2016; Figure 3—figure supplement 5A–F), likely due to use of different dilp2-GAL4 lines, minor variation in rearing conditions, and sex-specific plasticity defects in the dilp2-GAL4 strain, we reproduced the female-specific effects of sun knockdown on body size using an additional fat body GAL4 line (Figure 3—figure supplement 6A). Further, we show that this role for sun in mediating the nutrient-dependent increase in female body size in a protein-rich context is unique to sun, as no other humoral factors caused sex-specific effects on body size (Figure 3—figure supplement 6B,C).

Our data suggests a model in which the nutrient-dependent increase in sun mRNA levels is one important reason that females raised in a protein-rich context have a larger body size. To determine whether increased sun mRNA levels could augment body size, we overexpressed sun specifically in the fat body in larvae of each sex reared on 1Y and 2Y. We found that fat body sun overexpression was sufficient to increase body size in both sexes, in both the 1Y and 2Y diets (Figure 3—figure supplement 7A,B). This demonstrates that increased sun mRNA levels are sufficient to enhance body size in these contexts. While this finding contrasts with data from a previous study using a different diet and a mixed-sex experimental group (Delanoue et al., 2016), when we replicated their experimental conditions we found a significant increase in body size that was obscured by pooling data from males and females (Figure 3—figure supplement 8A,B). Together, this data supports a model in which increased fat body sun mRNA levels enhance body size in multiple nutritional contexts, an effect that was previously overlooked due to minor variation between lab diets and use of a mixed-sex experimental group. It is important to note, however, that despite the larger body size of sun-overexpressing males and females, phenotypic plasticity was not increased in the sun-overexpressing larvae (Figure 3—figure supplement 7A,B; diet:genotype interaction p=0.4959; and p=0.0895, respectively; Supplementary file 1). This is likely due to the fact that the nutrient-dependent increase in sun mRNA levels was still absent in the context of sun overexpression in males (Figure 3—figure supplement 8C), as our model suggests it is the ability to upregulate sun mRNA in response to dietary protein, rather than absolute sun mRNA levels, that allows females raised on a protein-rich diet to achieve a larger body size.

Sex determination gene transformer promotes nutrient-dependent body size plasticity in females

To gain a more complete understanding of the sex difference in phenotypic plasticity, we wanted to identify genetic factors in females that confer the ability to upregulate sun mRNA levels in response to dietary protein. One candidate was sex determination gene tra, as tra was previously found to impact IIS activity and body size in a diet equivalent to 2Y (Rideout et al., 2015; Mathews et al., 2017). Thus, we performed loss- and gain-of-function studies with tra and monitored changes to IIS activity, sun mRNA, and body size in both the 1Y and 2Y diets. In control w1118 females, Foxo target gene expression was significantly lower in larvae raised on 2Y compared with larvae cultured on 1Y (Figure 4A); however, this nutrient-dependent decrease in Foxo target gene expression was abolished in tra mutant females (tra1/Df(3L)st-j7) (Figure 4A; diet:genotype interaction p=0.0081; Supplementary file 1). Similarly, while sun mRNA levels in w1118 control females were significantly higher in larvae raised on 2Y compared with 1Y (Figure 4B), this nutrient-dependent increase in sun mRNA levels was absent in tra mutant females (Figure 4B). This indicates that tra is required in females for the nutrient-dependent increase in sun mRNA levels and IIS activity in a protein-rich context.

Figure 4 with 3 supplements see all
Sex determination gene transformer (tra) regulates increased nutrient-dependent body size plasticity in females.

(A) In control w1118 females, mRNA levels of Foxo targets (insulin receptor (InR), brummer (bmm), and eukaryotic initiation factor 4E-binding protein (4E-BP)), were significantly lower in larvae cultured on a protein-rich diet (2Y) compared with larvae raised on a diet containing half the protein (1Y) (p=0.0057; Student’s t test). In tra mutant (tra1/Df(3L)st-j7) females, there was no significant difference in mRNA levels of Foxo targets in larvae cultured on 2Y compared with larvae raised on 1Y (p=0.2291 Student’s t test). n = 8 biological replicates. (B) In control females, mRNA levels of sunRA were significantly higher in larvae cultured on 2Y compared with larvae raised on 1Y (p=0.0011; Student’s t test); however, in tra1/Df(3L)st-j7 females there was no significant difference in sunRA mRNA levels between larvae cultured on 2Y compared with larvae raised on 1Y (p=0.1644; Student’s t test). n = 8 biological replicates. (C) Adult weight was significantly higher in w1118 females raised on 2Y compared with females reared on 1Y (p<0.0001; two-way ANOVA followed by Tukey HSD test); however, there was no significant difference in adult weight between tra1/Df(3L)st-j7 females cultured on 2Y compared with genotype-matched females raised on 1Y (p=0.9617; two-way ANOVA followed by Tukey HSD test). n = 7–8 groups of 10 flies. (D) Adult weight was not significantly higher in either w1118 control or tra1/Df(3L)st-j7 mutant males in flies raised on 2Y compared with males reared on 1Y (p=0.7808 and p=0.9983, respectively; two-way ANOVA followed by Tukey HSD test). There was a significant sex:diet interaction in the w1118 control genotype (p<0.0001; two-way ANOVA followed by Tukey HSD test); however, there was no sex:diet interaction in the tra1/Df(3L)st-j7 genotype (p=0.6598; two-way ANOVA followed by Tukey HSD test). n = 6–8 groups of 10 flies. (E) In control da>+, and +>traF males, mRNA levels of Foxo targets were significantly higher in larvae cultured on 2Y compared with larvae raised on 1Y a diet containing half the protein content (1Y) (p=0.0108 and p<0.0001, respectively; Student’s t test). However, in da>traF males, there was a significant decrease in Foxo target mRNA levels (p<0.0001; Student’s t test). n = 8 biological replicates. Importantly, there was a significant sex:diet interaction for Foxo target mRNA levels in both the da>+ control (p=0.0004; two-way ANOVA followed by Tukey HSD test) and +>traF control (p<0.0001; two-way ANOVA followed by Tukey HSD test), but not in da>traF larvae (p=0.3095; two-way ANOVA followed by Tukey HSD test). n = 7–8 biological replicates. (F) In control da>+ and +>UAS-traF males, mRNA levels of sunRA were not significantly different between larvae cultured on 2Y compared with larvae raised on 1Y (p=0.2064 and p=0.0711, respectively; Student’s t test). In contrast, da>UAS-traF males showed a significant increase in mRNA levels of sunRA in larvae cultured on 2Y compared with males raised on 1Y (p=0.0013; Student’s t test). n = 6–8 biological replicates. (G) Adult weight was not significantly higher in da>+ and +>UAS-traF control males reared on 2Y compared with genotype-matched males flies cultured on 1Y (p=0.5186 and p=0.8858, respectively; two-way ANOVA followed by Tukey HSD test); however, there was a significant increase in adult weight between da>UAS-traF males cultured on 2Y compared with genotype-matched flies raised on 1Y (p<0.0001; two-way ANOVA followed by Tukey HSD test). n = 7–8 groups of 10 flies. (H) Adult weight was significantly higher in r4-GAL4 control males with traF K-IN, which express physiological levels of a functional Tra protein, when reared on 2Y compared with 1Y ((p<0.0001 [r4,Df(3L)st-j7/traF K-IN]); two-way ANOVA followed by Tukey HSD test). In contrast, the nutrient-dependent increase in adult weight was abolished upon fat body knockdown of sun in a traF K-IN male ((p=0.9915 [UAS-sun-RNAi/+;r4,Df(3L)st-j7/traF K-IN]); two-way ANOVA followed by Tukey HSD test). Adult weight was not different in tra mutant r4-GAL4 males (r4,Df(3L)st-j7/traKO) reared on 2Y compared with genotype-matched males cultured on 1Y (p=0.9980; two-way ANOVA followed by Tukey HSD test). Adult weight was not reduced in 1Y with fat body knockdown of sun in a tra mutant male ((UAS-sun-RNAi/+;r4,Df(3L)st-j7/traKO) (p=0.9998 [UAS-sun-RNAi/+;r4,Df(3L)st-j7/traKO v r4,Df(3L)st-j7/traKO]); two-way ANOVA followed by Tukey HSD test). n = 9–11 groups of 10 flies. For all body size plasticity graphs, filled circles indicate mean body size, and dashed lines indicate 95% confidence interval. * indicates p<0.05, ** indicates p<0.01, **** indicates p<0.0001; ns indicates not significant; error bars indicate SEM.

To determine whether lack of tra also impacts nutrient-dependent body size plasticity, we measured body size in w1118 controls and tra mutants raised in 1Y and 2Y. In control w1118 females, adult weight was significantly higher in flies raised on 2Y compared with flies cultured on 1Y (Figure 4C); however, this nutrient-dependent increase in adult weight was blocked in tra mutant females (Figure 4C; genotype:diet interaction p<0.0001; Supplementary file 1), a finding we reproduced using pupal volume (Figure 4—figure supplement 1A). Given that we confirmed this result using an additional tra mutant allele (traKO) (Hudry et al., 2016; Figure 4—figure supplement 1B), and that this finding cannot be explained by changes to food intake (Figure 4—figure supplement 1C), this indicates that tra mutant females have reduced nutrient-dependent body size plasticity compared with control females (genotype:diet interaction p<0.0001 [tra1/Df(3L)st-j7]; p<0.0001 [traKO]; Supplementary file 1). In control w1118 and tra mutant males, adult weight was not significantly higher in flies raised on 2Y compared with genotype-matched flies reared on 1Y (Figure 4D; genotype:diet interaction p=0.4507, Supplementary file 1). Given that we observed a sex difference in nutrient-dependent body size plasticity in the w1118 genotype (sex:diet interaction p<0.0001; Supplementary file 1), but not in the tra mutant strains (sex:diet interaction p=0.6598 [tra1/Df(3L)st-j7]; p=0.5068 [traKO]; Supplementary file 1), findings we replicated with pupal volume (Figure 4—figure supplement 1D,E), our data reveals a previously unrecognized requirement for tra in regulating the sex difference in nutrient-dependent phenotypic plasticity. To determine whether tra affects phenotypic plasticity via regulation of sun, we overexpressed sun in the fat body of tra mutant females. We found that the reduced body size of tra mutant females in 2Y was rescued by fat body sun overexpression (Figure 4—figure supplement 2A). This supports a model in which the smaller body size of tra mutant females reared in 2Y was due at least in part to lower sun mRNA levels.

To determine whether lack of a functional Tra protein in males explains their reduced nutrient-dependent body size plasticity, we overexpressed UAS-traF in all tissues using daughterless (da)-GAL4. We first asked whether Tra overexpression impacted the nutrient-dependent regulation of sun mRNA and IIS activity. In control da>+ and +>UAS-traF males, there was no significant decrease in Foxo target gene expression in larvae reared in 2Y compared with larvae raised in 1Y (Figure 4E). In da>UAS-traF males, however, there was a significant nutrient-dependent decrease in mRNA levels of Foxo target genes (Figure 4E). Because we observed a significant diet:genotype interaction (p<0.0001; Supplementary file 1), the magnitude of the nutrient-dependent increase in IIS activity in the da>UAS-traF genotype was larger than in control males. Similarly, while sun mRNA levels in control da>+ and +>UAS-traF males were not significantly different in larvae raised on 2Y compared with larvae reared on 1Y (Figure 4F), there was a nutrient-dependent increase in sun mRNA levels in da>UAS-traF males (Figure 4F). In da>+, +>UAS-traF, and da>UAS-traF females, we observed a significant decrease in Foxo target gene expression, and a significant increase in sun mRNA levels (Figure 4—figure supplement 3A,B). Thus, the presence of a functional Tra protein in males confers the ability to upregulate sun mRNA levels and IIS activity, revealing that the lack of Tra in normal males accounts for the lack of a nutrient-dependent increase in sun mRNA and IIS activity.

We next tested whether the presence of a functional Tra protein in males would augment nutrient-dependent body size plasticity. We observed a significant increase in adult weight between da>UAS-traF males reared on 2Y compared with genotype-matched males raised on 1Y (Figure 4G; genotype:diet interaction p=0.0038; Supplementary file 1). This nutrient-dependent increase was not present in either control da>+ or +>UAS-traF males (Figure 4G), a finding we reproduced using pupal volume (Figure 4—figure supplement 3C). Because one study suggested high levels of Tra expression may cause lethality (Siera and Cline, 2008), we repeated the experiment using males from a recently published strain of flies in which flies carry a cDNA encoding the female-specific Tra protein knocked into the tra locus (traF K-IN). These males express Tra at a physiological level (Hudry et al., 2019). As with da>UAS-traF males, we found traF K-IN males had increased nutrient-dependent body size plasticity compared with control w1118 males and traKO males (Figure 4—figure supplement 3D; genotype:diet interaction p<0.0001; Supplementary file 1). Thus, males expressing a functional Tra protein have increased phenotypic plasticity compared with control males, revealing a new role for tra in conferring the ability to adjust body size in response to a protein-rich diet. In females, we observed a significant increase in both adult weight and pupal volume in da>+, +>UAS-traF, and da>UAS-traF flies raised on the 2Y diet compared with genotype-matched females cultured on the 1Y diet (Figure 4—figure supplement 3E,F); however, lack of a significant genotype:diet interaction indicates that phenotypic plasticity in da>UAS-traF females was not different from controls (p=0.5912; Supplementary file 1), findings we reproduced with the traF K-IN allele (Figure 4—figure supplement 3G; genotype:diet interaction p<0.0001 Supplementary file 1). Importantly, the sex difference in nutrient-dependent body size plasticity that we observed in the w1118 control genotype (sex:diet interaction p<0.0001; Supplementary file 1) was abolished between traF K-IN males and their genotype-matched females (p=0.3168; Supplementary file 1). To determine whether the nutrient-dependent upregulation of sun mRNA was required for Tra to enhance male body size in a protein-rich context, we overexpressed the UAS-sun-RNAi transgene in the fat body of traF K-IN males. We found that the nutrient-dependent body size increase in traF K-IN males was blocked in males with fat body sun loss (Figure 4H), a finding we reproduced in traF K-IN females (Figure 4—figure supplement 3H). This indicates that the nutrient-dependent upregulation of sun mRNA in larvae with a functional Tra protein is required for phenotypic plasticity. Together, these data demonstrate a new role for Tra in regulating the sex difference in nutrient-dependent body size plasticity, and identify fat body sun as one downstream factor that mediates Tra’s effects on phenotypic plasticity.

Transcriptional coactivator spargel represents one link between transformer and regulation of sun mRNA levels

While sex determination gene tra impacts sexual differentiation via regulation of confirmed target genes doublesex (dsx; FBgn0000504) and fruitless (fru; FBgn0004652), neither dsx nor fru affect body size (Rideout et al., 2015). Given the key role of sun in mediating the nutrient-dependent increase in body size downstream of Tra, we wanted to identify the link between Tra and regulation of sun mRNA levels. Previous studies show that transcriptional coactivator spargel (srl, FBgn0037248), the Drosophila homolog of peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α) (Tiefenböck et al., 2010), coordinates sun mRNA levels with dietary protein (Delanoue et al., 2016). To test whether Srl mediates the sex difference in nutrient-dependent upregulation of sun mRNA levels, we examined mRNA levels of sun in female larvae heterozygous for a strong hypomorphic allele of srl (srl1/+) (Tiefenböck et al., 2010). In females, we found that the nutrient-dependent upregulation of sun mRNA levels in w1118 control larvae was blunted in srl1/+ larvae (Figure 5A; diet:genotype interaction p<0.0001; Supplementary file 1). To determine whether a smaller nutrient-dependent increase in sun mRNA levels affects the ability of srl1/+ larvae to achieve a larger body size in a protein-rich context, we raised srl1/+ larvae on 1Y and 2Y. While we confirmed that srl1/+ larvae have no generalized developmental defects, as there was no decrease in body size in srl1/+ female or male larvae reared on 1Y (Figure 5B,C), we showed that the nutrient-dependent increase in body size in srl1/+ females was eliminated (Figure 5B). Given that adult weight was significantly higher in control w1118 females raised on 2Y compared with genotype-matched females cultured on 1Y (Figure 5B), this indicates that srl1/+ females have reduced nutrient-dependent body size plasticity (genotype:diet interaction p<0.0001; Supplementary file 1). In control w1118 and srl1/+ males, adult weight was not significantly higher in flies raised on 2Y compared with genotype-matched flies reared on 1Y (Figure 5C; genotype:diet interaction p=0.8323). This result suggests that Srl mediates the nutrient-dependent upregulation of sun mRNA levels and increased nutrient-dependent body size plasticity in female larvae in a protein-rich context, where future studies will need to determine whether Srl also impacts the sex difference in circulating Sun. Indeed, while Sun is also regulated at the level of secretion by fat body Target-of-Rapamycin (TOR) signaling (Delanoue et al., 2016), we found no sex difference in fat body TOR activity in either 1Y or 2Y (Figure 5—figure supplement 1A–D). Given that TOR activity does not affect sun mRNA levels (Delanoue et al., 2016), which we confirm (Figure 5—figure supplement 1E), our data indicates that the sex difference in nutrient-dependent upregulation of sun mRNA levels is due to Srl, and not TOR. This aligns with our previous finding that treating larvae with TOR inhibitor Rapamycin did not cause sex-biased effects on larval growth (Rideout et al., 2015).

Figure 5 with 2 supplements see all
Sex determination gene transformer (tra) requires transcriptional coactivator spargel (srl) for increased nutrient-dependent body size plasticity in females.

(A) In control w1118 females and females with heterozygous loss of srl (srl1/+), mRNA levels of sunRA were significantly higher in larvae cultured on a protein-rich diet (2Y) compared with larvae raised on a diet with half the protein (1Y) (p<0.0001 and p=0.0301; Student’s t test); however, there was a significant genotype:diet interaction indicating that the protein-dependent upregulation of sunRA was blunted in srl1/+ females (p<0.0001; two-way ANOVA followed by Tukey HSD test). n = 7–8 biological replicates. (B) Adult weight was significantly higher in w1118 females raised on 2Y compared with females reared on 1Y (p<0.0001; two-way ANOVA followed by Tukey HSD test); however, there was no significant difference in adult weight between srl1/+ females cultured on 2Y compared with genotype-matched females raised on 1Y (p>0.9999; two-way ANOVA followed by Tukey HSD test). n = 5–7 groups of 10 flies. (C) Adult weight was not significantly higher in either w1118 control or srl1/+ mutant males in flies raised on 2Y compared with males reared on 1Y (p=0.9906 and p>0.9999, respectively; two-way ANOVA followed by Tukey HSD test). n = 4–5 groups of 10 flies. (D) mRNA levels of sunRA were not significantly different in da>traF males with heterozygous loss of srl (UAS-traF/+;da-GAL4/srl1) cultured on 1Y compared to genotype matched males cultured on 2Y (p=0.1405; Student’s t test). n = 8 biological replicates. (E) In control da>traF males with heterozygous loss of srl, mRNA levels of Foxo targets (insulin receptor (InR), brummer (bmm), and eukaryotic initiation factor 4E-binding protein (4E-BP)), were significantly higher in larvae cultured on 2Y compared with larvae raised on 1Y (p=0.0266; Student’s t test). n = 8 biological replicates. (F) Adult weight was higher in da>UAS-traF males raised on 2Y compared with da>UAS-traF males reared on 1Y (p<0.0001; two-way ANOVA followed by Tukey HSD test). In contrast, the nutrient-dependent increase in adult weight was abolished in da>UAS-traF males heterozygous for srl1 (p=0.2811; two-way ANOVA followed by Tukey HSD test). n = 6–8 groups of 10 flies. (G) Adult weight was higher in da>UAS-traF females raised on 2Y compared with da>UAS-traF females reared on 1Y (p<0.0001; two-way ANOVA followed by Tukey HSD test). In contrast, the nutrient-dependent increase in adult weight was absent in da>UAS-traF females heterozygous for srl1 (p=0.2927; two-way ANOVA followed by Tukey HSD test). n = 6–7 groups of 10 flies. For all body size plasticity graphs, filled circles indicate mean body size, and dashed lines indicate 95% confidence interval. * indicates p<0.05, **** indicates p<0.0001; ns indicates not significant; error bars indicate SEM.

To determine whether Srl mediates the Tra-dependent regulation of sun mRNA levels, we measured sun mRNA levels in males with ectopic Tra expression (da>UAS-traF). While da>UAS-traF males show a significant nutrient-dependent upregulation of sun mRNA levels compared with da>+ and +>UAS-traF control males (Figure 4F), we found that sun mRNA levels were no longer higher in da>UAS-traF males heterozygous for the srl1 allele raised on 2Y compared with genotype-matched males reared on 1Y (Figure 5D). Similarly, we observed no decrease in Foxo target genes between da>UAS-traF males heterozygous for the srl1 allele raised on 2Y compared with genotype-matched males reared on 1Y (Figure 5E), indicating the nutrient-dependent upregulation of IIS activity in da>UAS-traF males was abolished in the context of reduced Srl function. Given that we observed no Tra-dependent changes to TOR activity (Figure 5—figure supplement 1F,G), when taken together our data indicates that Srl function is required for the Tra-dependent increase in sun mRNA levels in a protein-rich context. Srl therefore represents an additional link between sex determination gene tra and the regulation of gene expression. Moreover, we show that the Srl-dependent regulation of sun downstream of Tra is significant for phenotypic plasticity, as the nutrient-dependent increase in body size was blocked in da>UAS-traF females and males heterozygous for the srl1 allele (Figure 5F,G; genotype:diet interaction p=0.0146 and p=0.0008, respectively). While we find that Srl targets other than sun were also regulated in a sex-specific manner by nutrients and Tra function (Figure 5—figure supplement 2A–D), other functionally similar Srl targets did not reproduce sex-specific changes to nutrient-dependent body size plasticity that we observed upon loss of fat body sun (Figure 5—figure supplement 2E–H). Although we cannot rule out all Srl targets, our data indicates a key role for sun among Srl targets in mediating the effects of Tra on nutrient-dependent body size plasticity. This reveals a previously unrecognized role for Srl in mediating sex-specific changes to gene expression, and identifies Srl as a new link between Tra and nutrient-dependent changes to gene expression.

Increased nutrient-dependent body size plasticity in females promotes fecundity in a protein-rich context

Previous studies have shown that plentiful nutrients during development maximize body size to promote fertility in Drosophila females (Bergland et al., 2008; Green and Extavour, 2014; Grönke et al., 2010; Hodin and Riddiford, 2000; Klepsatel et al., 2020; Mendes and Mirth, 2016; Robertson, 1957a; Robertson, 1957b; Sarikaya et al., 2012; Tu and Tatar, 2003), and that high levels of IIS activity are required for normal egg development, ovariole number, and fecundity (Green and Extavour, 2014; Grönke et al., 2010; Mendes and Mirth, 2016; Richard et al., 2005). In line with these findings, w1118 female flies reared on 2Y produced significantly more eggs compared with genotype-matched females cultured on 1Y (Figure 6A). This aligns with findings from many studies showing that increased nutrients promote fertility (Green and Extavour, 2014; Grönke et al., 2010; Mendes and Mirth, 2016; Richard et al., 2005) and suggests that the ability to augment IIS activity and body size in response to a protein-rich diet allows females to maximize fecundity in conditions where nutrients are plentiful. To test this, we measured the number of eggs produced by InRE19/+ females and w1118 controls raised in either 1Y or 2Y. In contrast to w1118 females, the nutrient-dependent increase in egg production was absent in InRE19/+ females (Figure 6A). Similarly, there was no diet-induced increase in egg production in dilp2 mutant females (Figure 6B). These findings suggest that the nutrient-dependent increase in IIS activity and body size are important to promote fecundity in a protein-rich context. This result aligns with findings from a previous study showing that lifetime fecundity was significantly lower in dilp2 mutants raised in a yeast-rich diet (Grönke et al., 2010). To extend our findings beyond dilp genes, we next examined fecundity in females with an RNAi-mediated reduction in sun. We found that the nutrient-dependent increase in egg production in r4>UAS-sun-RNAi females was eliminated, in contrast to the robust diet-induced increase in fecundity in r4>+ and +>UAS-sun-RNAi control females (Figure 6C). Together, this data suggests that dilp2 and fat body-derived sun play a role in maximizing IIS activity and body size to promote egg production in a protein-rich context. Future studies will need to determine which aspect of ovary development is affected by these genetic manipulations (Green and Extavour, 2014; Grönke et al., 2010; Mendes and Mirth, 2016; Richard et al., 2005), whether this phenotype is specific to dilp2, and whether the effects require InR function in the ovary or in other tissues.

Increased nutrient-dependent body size plasticity in females promotes fertility.

(A) In control w1118 females, there was a significant increase in the number of eggs laid by females raised on a protein-rich diet (2Y) compared with females reared on a diet with half the protein (1Y) (p=0.0009; Student’s t test); however, there was no significant difference in the number of eggs laid between InRE19/+ females cultured on 2Y compared with genotype-matched females raised on 1Y (p=0.617; Student’s t test). n = 19–20 biological replicates. (B) In control w1118 females, there was a significant increase in the number of eggs laid by females raised on 2Y compared with females cultured on 1Y (p<0.0001; Student’s t test); however, there was no significant difference in the number of eggs laid between dilp2 mutant females cultured on 2Y compared with females raised on 1Y (p=0.4105; Student’s t test). n = 28–30 biological replicates. (C) In control r4>+ and +>UAS-sun-RNAi females, there was a significant increase in the number of eggs laid by females raised on 2Y compared with control females cultured on 1Y (p<0.0001 for both genotypes; Student’s t test). In r4>UAS-sun-RNAi females, the number of eggs laid by females cultured on 2Y was lower than females raised on 1Y (p=0.0243; Student’s t test). n = 20 biological replicates. (D) In control w1118 males, there was no significant difference in the number of offspring produced between a 1Y and 2Y diet (p=0.3662; Student’s t test). There was also no significant difference in the number of offspring produced between control w1118 males and males heterozygous for a loss-of-function allele of phosphatase and tensin homolog (pten; genotype pten2L100/+) raised on 1Y (p=0.4003; Student’s t test). Unlike control males, pten2L100/+ males reared on 2Y produced significantly more offspring than genotype-matched males raised on 1Y (p=0.0137; Student’s t test). n = 11 biological replicates. (E) In control r4>+ and +>UAS-sun and r4>UAS-sun males, there was no significant effect on the number of offspring produced between a 1Y and 2Y diet (p=0.9222, 0.0595, and 0.32 respectively; Student’s t test). There was also no significant difference in the number of offspring produced between control r4>+, +>UAS-sun males and r4>UAS-sun males raised on 1Y (p=0.9723 and p=0.9969 respectively; one-way ANOVA followed by Tukey HSD test). n = 8–10 groups of 10 flies. * indicates p<0.05, ** indicates p<0.01, *** indicates p<0.001, **** indicates p<0.0001; ns indicates not significant; error bars indicate SEM.

In males, which have a reduced ability to augment body size in response to a protein-rich diet, we also investigated the relationship between nutrient content, body size, and fertility. When we compared fertility in w1118 males reared on 1Y compared with males raised on 2Y, we found no significant difference in the number of offspring produced (Figure 6D). Thus, neither male body size nor fertility were enhanced by rearing flies in a protein-rich environment. Given that previous studies suggest that a larger body size in males promotes reproductive success (Ewing, 1961; Partridge et al., 1987; Partridge and Farquhar, 1983), we next asked whether genetic manipulations that augment male body size also increased fertility. One way to augment male body size in 1Y is heterozygous loss of phosphatase and tensin homolog (pten, FBgn0026379; pten2L100/+) (Figure 1—figure supplement 7B). Interestingly, fertility was not significantly higher in pten2L100/+ males compared with w1118 controls raised in 1Y (Figure 6D), suggesting that a larger body size does not always augment fertility in males. Similarly, when we measured fertility in r4>UAS-sun males, which are larger than control males (Figure 3—figure supplement 7B), fertility was not significantly different from r4>+ and +>UAS-sun control males (Figure 6E). Interestingly, when we examined fertility in pten2L100/+ and r4>UAS-sun males in 2Y, fertility was significantly increased in pten2L100/+ males compared with genotype-matched controls cultured in 1Y (Figure 6D), an observation we did not repeat in r4>UAS-sun males (Figure 6E). Ultimately, this less robust and more complex relationship between body size and fertility in males suggests a possible explanation for their decreased nutrient-dependent body size plasticity compared with females.

Discussion

In many animals, body size plasticity in response to environmental factors such as nutrition differs between the sexes (Fairbairn, 1997). While past studies have identified mechanisms underlying nutrient-dependent growth in a mixed-sex population, and revealed factors that promote sex-specific growth in a single nutritional context, the mechanisms underlying the sex difference in nutrient-dependent body size plasticity remain unknown. In this study, we showed that females have higher phenotypic plasticity compared with males when reared on a protein-rich diet, and elucidated the molecular mechanisms underlying the sex difference in nutrient-dependent body size plasticity in this context. Our data suggests a model in which high levels of dietary protein augment female body size by stimulating an increase in IIS activity, where we identified a requirement for dilp2 and sun in promoting this nutrient-dependent increase in IIS activity. Importantly, we discovered tra as the factor responsible for stimulating sun mRNA levels and IIS activity in a protein-rich context, revealing a novel role for sex determination gene tra in regulating phenotypic plasticity. Mechanistically, tra enhanced sun mRNA levels and body size in protein-rich conditions via transcriptional coactivator Srl, identifying Srl as one link between tra and the nutrient-dependent regulation of gene expression. Together, these findings provide new insight into how Drosophila females achieve increased nutrient-dependent body size plasticity compared with males.

One key feature of this increased phenotypic plasticity in females was a female-biased increase in IIS activity in a protein-rich context. This reveals a previously unrecognized sex difference in the coupling between IIS activity and dietary protein. In females, there was tight coupling between increased nutrient input and enhanced IIS activity across a wide protein concentration range in all control genotypes. In males, this close coordination between dietary protein and IIS activity was weaker in a protein-rich context. Our data shows that sex-biased nutrient-dependent change to IIS activity during development is physiologically significant, as it supports an increased rate of growth and consequently larger body size in females but not in males raised on a protein-rich diet. In future studies, it will be important to determine whether the sex difference in coupling between nutrients and IIS activity exists in other contexts. For example, previous studies on the extension of life span by dietary restriction have shown that male and female flies differ in the concentration of nutrients that produces the maximum life span extension, and in the magnitude of life span extension produced by dietary restriction (Magwere et al., 2004; Regan et al., 2016). Similar sex-specific effects of dietary restriction and reduced IIS on life span have also been observed in mice (Holzenberger et al., 2003; Kane et al., 2018; reviewed in Regan and Partridge, 2013; Selman et al., 2008) and humans (Van Heemst et al., 2005). Future studies will be needed to determine whether a male-female difference in coupling between nutrients and IIS activity account for these sex-specific life span responses to dietary restriction. Indeed, given that sex differences have been reported in the risk of developing diseases associated with overnutrition and dysregulation of IIS activity such as obesity and type 2 diabetes (Kautzky-Willer et al., 2016; Mauvais-Jarvis, 2018; Tramunt et al., 2020), more detailed knowledge of the male-female difference in coupling between nutrients and IIS activity in other models may provide insights into this sex-biased risk of disease.

In addition to revealing a sex difference in the nutrient-dependent upregulation of IIS activity, our data identified a female-specific requirement for dilp2 and sun in mediating the diet-induced increase in IIS activity in a protein-rich context. While previous studies have shown that both dilp2 and sun positively regulate body size (Ikeya et al., 2002; Grönke et al., 2010; Delanoue et al., 2016), we describe new sex-specific roles for dilp2 and sun in nutrient-dependent phenotypic plasticity. Elegant studies have shown that sun is a secreted factor that stimulates Dilp2 release from the IPCs (Delanoue et al., 2016). Together with our data, this suggests a model in which females are able to achieve a larger body size in a protein-rich diet because they have the ability to upregulate sun mRNA levels, whereas males do not. Indeed, we show that higher sun mRNA levels are sufficient to augment body size. This model aligns well with findings from two previous studies on Dilp2 secretion in male and female larvae. The first study, which raised larvae on a protein-rich diet equivalent to the 2Y diet, found increased Dilp2 secretion in females compared to males (Rideout et al., 2015). The second study, which raised larvae on a diet equivalent to the 1Y diet, found no sex difference in Dilp2 secretion and no effects of dilp2 loss on body size (Sawala and Gould, 2017). Thus, while these previous studies differed in their initial findings on a sex difference in Dilp2 secretion, our data reconcile these minor differences by identifying context-dependent effects of dilp2 on body size. It is important to note that absolute confirmation of a sex difference in hemolymph Dilp2 levels will be needed in future studies because the body size plasticity defects in the dilp2-HF strain precluded its use as a tool to quantify circulating Dilp2 levels in our study. Future studies will also need to determine whether these sex-specific and context-dependent effects of dilp2 are observed in other phenotypes regulated by dilp2 and other dilp genes. For example, flies carrying mutations in dilp genes show changes to aging, metabolism, sleep, and immunity, among other phenotypes (Bai et al., 2012; Brown et al., 2020; Cong et al., 2015; Grönke et al., 2010; Liu et al., 2016; Nässel and Vanden Broeck, 2016; Okamoto et al., 2009; Okamoto and Nishimura, 2015; Post et al., 2018; Post et al., 2019; Slaidina et al., 2009; Stafford et al., 2012; Zhang et al., 2009; Brogiolo et al., 2001; Cognigni et al., 2011; Linneweber et al., 2014; Semaniuk et al., 2018; Suzawa et al., 2019; Ugrankar et al., 2018). Further, it will be interesting to determine whether the sex-specific regulation of sun is observed in any other contexts, and whether it will influence sex differences in phenotypes associated with altered IIS activity, such as life span.

While our findings on sun and dilp2 provide mechanistic insight into the molecular basis for the larger body size of females reared on a protein-rich diet, a key finding from our study was the identification of sex determination gene tra as the factor that confers plasticity to females. Normally, nutrient-dependent body size plasticity is higher in females than in males in a protein-rich context. In females lacking a functional Tra protein, however, this increased nutrient-dependent body size plasticity was abolished. In males, which normally lack a functional Tra protein, ectopic Tra expression conferred increased nutrient-dependent body size plasticity. While a previous study showed that on the 2Y diet Tra promotes Dilp2 secretion (Rideout et al., 2015), our current study extends this finding in two ways: by identifying sun as one link between Tra, Dilp2, and changes to IIS activity; and by showing that Tra regulates sun mRNA via conserved transcriptional coactivator Srl. While previous studies discovered Srl as the factor that promotes sun mRNA levels in response to dietary protein in a mixed-sex larval population (Delanoue et al., 2016), our findings reveal a previously unrecognized sex-specific role for Srl in regulating transcription. Because loss of Tra reduces Srl transcriptional activity, this new link between Tra and Srl suggests an additional way in which Tra may impact gene expression beyond its canonical downstream targets dsx and fru. While this builds on recent studies that reveal a number of additional Tra-regulated genes (Clough et al., 2014; Hudry et al., 2016; Hudry et al., 2019), it will be important to determine whether these additional Tra-regulated genes including sun represent direct targets of Tra/Srl. Future studies will also be needed to elucidate how Tra impacts Srl transcriptional activity in a context-dependent manner. However, uncovering a connection between a sex determination gene and a key regulator of genes involved in mitochondrial function suggests an additional mechanism that may contribute to sex differences in phenotypes affected by mitochondrial function (e.g. lifespan) (Tiefenböck et al., 2010; Cho et al., 2011; Tower, 2015; Tower, 2017). In addition, it will be critical to explore how the presence of Tra allows an individual to couple dietary protein with body size. Because the tra locus is regulated both by alternative splicing and transcription (Belote et al., 1989; Boggs et al., 1987; Grmai et al., 2018; Inoue et al., 1990; Sosnowski et al., 1989), and Tra protein is regulated by phosphorylation (Du et al., 1998), our study highlights the importance of additional studies on the regulation of the tra genomic locus and Tra protein throughout development to gain mechanistic insight into its effects on nutrient-dependent body size plasticity.

While the main outcome of our work was to reveal the molecular mechanisms that regulate the sex difference in nutrient-dependent body size plasticity, we also provide some insight into how genes that contribute to nutrient-dependent body size plasticity affect female fecundity and male fertility. Our findings align well with previous studies demonstrating that increased nutrient availability during development and a larger female body size confers increased ovariole number and fertility (Green and Extavour, 2014; Klepsatel et al., 2020; Mendes and Mirth, 2016; Robertson, 1957a; Robertson, 1957b), as females lacking either dilp2 or fat body-derived sun were unable to augment egg production in a protein-rich context. Given that previous studies demonstrate IIS activity influences germline stem cells in the ovary in adult flies (Hsu et al., 2008; Hsu and Drummond-Barbosa, 2009; Kao et al., 2015; LaFever and Drummond-Barbosa, 2005; Lin and Hsu, 2020; Su et al., 2018), there is a clear reproductive benefit that arises from the tight coupling between nutrient availability, IIS activity, and body size in females. In males, however, the relationship between fertility and body size remains less clear. While larger males are more reproductively successful both in the wild and in laboratory conditions (Ewing, 1961; Partridge and Farquhar, 1983), other studies revealed that medium-sized males were more fertile than both larger and smaller males (Lefranc and Bundgaard, 2000). Given that our study revealed no significant increase in the number of progeny produced by larger males, the fertility benefits that accompany a larger body size in males may be context-dependent. For example, a larger body size increases the ability of males to outcompete smaller males (Flatt, 2020; Partridge et al., 1987; Partridge and Farquhar, 1983). Thus, in crowded situations, a bigger body may provide significant fertility gains. On the other hand, in conditions where nutrients are limiting, an imbalance in the allocation of energy from food to growth rather than to reproduction may decrease fertility (Bass et al., 2007; Camus et al., 2017; Jensen et al., 2015; Wood et al., 2018). Future studies will need to resolve the relationship between body size and fertility in males, as this will suggest the ultimate reason(s) for the sex difference in nutrient-dependent body size plasticity.

Materials and methods

Key resources table
Reagent type (species) or resourceDesignationSource or referenceIdentifiersAdditional information
Genetic reagent (Drosophila melanogaster)Canton-SBloomington Drosophila stock centerBDSC: 64349
Genetic reagent (Drosophila melanogaster)w1118Bloomington Drosophila stock centerBDSC: 3605
Genetic reagent (Drosophila melanogaster)tra1Bloomington Drosophila stock centerBDSC: 675
Genetic reagent (Drosophila melanogaster)Df(3L)st-j7Bloomington Drosophila stock centerBDSC: 5416
Genetic reagent (Drosophila melanogaster)srl1Bloomington Drosophila stock centerBDSC: 14965
Genetic reagent (Drosophila melanogaster)InRE19Bloomington Drosophila stock centerBDSC: 9646
Genetic reagent (Drosophila melanogaster)TRiP ControlBloomington Drosophila stock centerBDSC: 36303
Genetic reagent (Drosophila melanogaster)UAS-dilp2-RNAiBloomington Drosophila stock centerBDSC: 32475
Genetic reagent (Drosophila melanogaster)UAS-upd2-RNAiBloomington Drosophila stock centerBDSC: 33949
Genetic reagent (Drosophila melanogaster)UAS-traFBloomington Drosophila stock centerBDSC: 4590
Genetic reagent (Drosophila melanogaster)UAS-rhebBloomington Drosophila stock centerBDSC: 9688
Genetic reagent (Drosophila melanogaster)UAS-cyt-c-p-RNAiBloomington Drosophila stock centerBDSC: 64898
Genetic reagent (Drosophila melanogaster)UAS-Idh-RNAiBloomington Drosophila stock centerBDSC: 41708
Genetic reagent (Drosophila melanogaster)mth1Bloomington Drosophila stock centerBDSC: 27896
Genetic reagent (Drosophila melanogaster)y1,w1Bloomington Drosophila stock centerBDSC: 1495
Genetic reagent (Drosophila melanogaster)UAS-sun-RNAiVienna Drosophila resource centerVDRC: GD23685
Genetic reagent (Drosophila melanogaster)UAS-Gbp1-RNAiVienna Drosophila resource centerVDRC: KK108755
Genetic reagent (Drosophila melanogaster)UAS-Gbp2-RNAiVienna Drosophila resource centerVDRC: GD16696
Genetic reagent (Drosophila melanogaster)UAS-CCHa2-RNAiVienna Drosophila resource centerVDRC: KK102257
Genetic reagent (Drosophila melanogaster)UAS-mth-RNAiVienna Drosophila resource centerVDRC: KK106399
Genetic reagent (Drosophila melanogaster)dilp2Grönke et al., 2010
Genetic reagent (Drosophila melanogaster)pten2L100Oldham et al., 2002
Genetic reagent (Drosophila melanogaster)UAS-sunDelanoue et al., 2016
Genetic reagent (Drosophila melanogaster)traKOHudry et al., 2016
Genetic reagent (Drosophila melanogaster)traF K-INHudry et al., 2019
Genetic reagent (Drosophila melanogaster)y,w;;ilp2HFPark et al., 2014
Genetic reagent (Drosophila melanogaster)tGPH (GFP-PH)Britton et al., 2002
Genetic reagent (Drosophila melanogaster)da-GAL4Bloomington Drosophila stock centerBDSC: 55849Note. Discontinued stock, equivalent stocks available
Genetic reagent (Drosophila melanogaster)r4-GAL4Bloomington Drosophila stock centerBDSC: 33832
Genetic reagent (Drosophila melanogaster)cg-GAL4Bloomington Drosophila stock centerBDSC: 7011
Genetic reagent (Drosophila melanogaster)elav-GAL4Bloomington Drosophila stock centerBDSC: 458
Genetic reagent (Drosophila melanogaster)dilp2-GAL4Rulifson et al., 2002
AntibodyAnti-sun guinea pig polyclonalDelanoue et al., 2016(1:50)
AntibodyAnti-Cv-d guinea pig polyclonalPalm et al., 2012(1:1000)
AntibodyAnti-pS6k rabbit polyclonalCell Signaling: 9209(1:1000)
AntibodyAnti-Actin mouse monoclonalSanta Cruz: 8432(1:1000)

Fly husbandry

Request a detailed protocol

Larvae were raised at a density of 50 animals per 10 ml food at 25°C on Drosophila growth medium consisting of: 0.5x: 5.125 g/L sucrose, 17.725 g/L D-glucose, 12.125 g/L cornmeal, 11.325 g/L yeast, 4.55 g/L agar, 0.5 g CaCl2•2H2O, 0.5 g MgSO4•7H2O, 11.77 mL acid mix (propionic acid/phosphoric acid). 1x: 10.25 g/L sucrose, 25.45 g/L D-glucose, 24.25 g/L cornmeal, 22.65 g/L yeast, 4.55 g/L agar, 0.5 g CaCl2•2H2O, 0.5 g MgSO4•7H2O, 11.77 mL acid mix (propionic acid/phosphoric acid). 2x: 20.5 g/L sucrose, 70.9 g/L D-glucose, 48.5 g/L cornmeal, 45.3 g/L yeast, 4.55 g/L agar, 0.5 g CaCl2•2H2O, 0.5 g MgSO4•7H2O, 11.77 mL acid mix (propionic acid/phosphoric acid). 1Y: 20.5 g/L sucrose, 70.9 g/L D-glucose, 48.5 g/L cornmeal, 22.65 g/L yeast, 4.55 g/L agar, 0.5 g CaCl2•2H2O, 0.5 g MgSO4•7H2O, 11.77 mL acid mix (propionic acid/phosphoric acid). 2Y: 20.5 g/L sucrose, 70.9 g/L D-glucose, 48.5 g/L cornmeal, 45.3 g/L yeast, 4.55 g/L agar, 0.5 g CaCl2•2H2O, 0.5 g MgSO4•7H2O, 11.77 mL acid mix (propionic acid/phosphoric acid). Details for diets manipulating dietary sugar (1S) and calorie content (2Y calories) are found in Supplementary file 3. Our diets were also deposited in the Drosophila Dietary Composition Calculator (DDCC) (Lesperance and Broderick, 2020). Animals were collected as indicated in figure legends, and sexed by gonad size. When gonad size could not be used to determine sex (e.g. tra mutants, da-GAL4>UAS-traF), chromosomal females were identified by the presence of an X-linked GFP. Adult flies were maintained at a density of 20 flies per vial in single-sex groups.

Fly strains

Request a detailed protocol

The following fly strains from the Bloomington Drosophila Stock Center were used: Canton-S (#64349), w1118 (#3605), tra1 (#675), Df(3L)st-j7 (#5416), srl1 (#14965), InRE19 (#9646), TRiP control (#36303), UAS-ilp2-RNAi (#32475), UAS-upd2-RNAi (#33949), UAS-traF (#4590), y,w (#1495), da-GAL4 (ubiquitous), r4-GAL4 (fat body), cg-GAL4 (fat body), dilp2-GAL4 (IPCs), elav-GAL4 (post-mitotic neurons), UAS-rheb (#9688), UAS-cyt-c-p-RNAi (#64898), UAS-Idh-RNAi (#41708), mth1 (#27896). The following fly strains from the Vienna Drosophila Resource Center were used in this study: UAS-sun-RNAi (GD23685), UAS-Gbp1-RNAi (KK108755) UAS-Gbp2-RNAi (GD16696), UAS-CCHa2-RNAi (KK102257), UAS-mth-RNAi (KK106399). Additional fly strains include: dilp2 (Grönke et al., 2010), pten2L100, UAS-sun, tGPH (GFP-PH), traKO (Hudry et al., 2016), traF K-IN(Hudry et al., 2019), y,w;;ilp2HF (Park et al., 2014). All genotypes used in the manuscript are listed in Supplementary file 4.

Body size

Request a detailed protocol

Pupal volume was measured in male and female pupae as previously described (Delanoue et al., 2010; Marshall et al., 2012; Rideout et al., 2012; Rideout et al., 2015). For adult weight, 5-day-old virgin male and female flies were weighed in groups of 10 in 1.5 ml microcentrifuge tubes on an analytical balance. Wing length was measured as previously described (Garelli et al., 2012).

Developmental timing

Request a detailed protocol

Larvae were placed into the experimental diet ±2 hr post-hatching. Percent pupation was calculated by comparing the number of pupae at 12 hr intervals to the total pupae in the vial after all animals pupated.

Feeding behavior

Request a detailed protocol

Feeding behavior was quantified in sexed larvae by counting mouth hook contractions for 30 s.

Protease feeding experiments

Request a detailed protocol

We treated larvae with a broad-spectrum protease inhibitor (PIC; Sigma-Aldrich #P2714) or a serine protease-specific inhibitor (AEBSF; Sigma-Aldrich #A8456) by adding the inhibitors to the food at final concentrations of 100 ml of 1x PIC per L, and 4 mM AEBSF as previously described (Erkosar et al., 2015).

RNA extraction and cDNA synthesis

Request a detailed protocol

One biological replicate represents ten larvae frozen on dry ice and stored at −80°C. Each experiment contained three to four biological replicates per sex, per genotype, and per diet, and each experiment was repeated twice. RNA was extracted using Trizol (Thermo Fisher Scientific; 15596018) according to manufacturer’s instructions, as previously described (Marshall et al., 2012; Rideout et al., 2012; Rideout et al., 2015; Wat et al., 2020). cDNA synthesis was performed using the QuantiTect Reverse Transcription Kit according to manufacturer’s instructions (Qiagen; 205314).

Quantitative real-time PCR (qPCR)

Request a detailed protocol

qPCR was performed as previously described (Rideout et al., 2012; Rideout et al., 2015; Wat et al., 2020). To determine changes in Foxo target gene expression, we plotted and analyzed the fold change in mRNA levels for each of three known Foxo target genes (InR, bmm, and 4E-BP) together to quantify IIS activity in each sex and dietary context, an established approach to analyze co-regulated genes (Blaschke et al., 2013; Hudry et al., 2019). A complete primer list is available in Supplementary file 5.

Preparation of protein extract

Request a detailed protocol

Dissected fat bodies were prepared for SDS-PAGE by homogenizing sets of ten larval fat bodies 108 hr after egg laying in an appropriate volume of lysis buffer (20 mM Hepes (pH 7.8), 450 mM NaCl, 25% glycerol, 50 mM NaF, 0.2 mM EDTA, 1 mM DTT, 1× Protease Inhibitor Cocktail (Roche, 04693124001), 1x Phosphatase Inhibitor Cocktail (Roche, 4906845001) using the Omni Bead Ruptor (VWR). Cellular fragments were pelleted, and supernatant collected by centrifugation for 5 min at 10,000 rpm at 4°C (Thermo Scientific, Heraeus Pico 21 centrifuge). Protein concentration was determined by Bradford assay (Bio-Rad #550–0205) prior to SDS-PAGE).

SDS-PAGE and Western blotting

Request a detailed protocol

A total of 20 μL of sample with 20 μg protein was loaded into each well. Proteins were separated using a 12% gel SDS-PAGE gel in SDS running buffer, and transferred to a nitrocellulose membrane (Bio-Rad) for 2 hr at 40 V on ice. Membranes were incubated for 1 hr in blocking buffer (5% milk or 5% BSA in TBST 0.1%) then incubated with primary antibodies overnight at 4°C. Membranes were washed (3 × 2 min) in TBST 0.1% then probed with secondary antibodies in blocking buffer for 1 hr at room temperature. After washes (3 × 2 min, 2 × 15 min, 1 × 5 min) in TBST 0.1%, membranes were treated with Pierce ECL (Thermo Scientific #32134) or Immobilon Forte (Millipore #WBLUF0100). Images were quantified using Image Studio (LI-COR). Primary antibodies: Anti-pS6K (#9209; Cell Signalling), and anti–actin (#8432; Santa Cruz), were used at 1:1000. HRP-conjugated secondary antibodies were used at 1:5000 for pS6k (anti-rabbit #65–6120; Invitrogen) and 1:3000 for actin (anti-mouse #7076; Cell Signalling).

Hemolymph Western blotting

Request a detailed protocol

Hemolymph Western blotting was performed as previously described (Delanoue et al., 2016). Briefly, hemolymph from 40 larvae was collected in 40 μL of PBS with protease and phosphatase inhibitors (Roche 04693124001, Roche 4906845001), and hemocytes were removed by centrifugation according to the published protocol (Delanoue et al., 2016). Antibody concentrations used to detect hemolymph proteins were 1:50 for anti-Sun and 1:1000 for anti-Cv-d. Anti-guinea pig HRP-conjugated secondary was used at 1:2000.

Fecundity and fertility

Request a detailed protocol

For female fecundity, single 6-day-old virgin female flies raised as indicated were crossed to three age-matched CS virgin males for a 24 hr mating period. Flies were transferred to fresh food vials with blue 2Y food to lay eggs. The number of eggs laid over 24 hr was quantified. For male fertility, single 6-day-old virgin males were paired with three 6-day-old virgin CS females to mate, and females were allowed to lay eggs for 24 hr. The number of progeny was quantified by counting viable pupae.

Microscopy

Request a detailed protocol

GFP-PH larvae were picked into 1Y or 2Y food. Larvae were dissected 108 hr after egg laying (AEL) and inverted carcasses were fixed for 30 min in 4% paraformaldehyde in phosphate buffered saline (PBS) at room temperature. Carcasses were rinsed twice with PBS, once in 0.1% Triton-X in PBS (PBST) for 5 min, then incubated with Hoechst (5 μg/mL, Life Technologies H3570), , and phalloidin fluor 647 (1:1000, Abcam ab176759) in PBST for 40 min. The stained carcasses were washed with PBS and mounted in SlowFade Diamond (Thermo Fisher Scientific S36972). Images were acquired with a Leica SP5 (20X). Mean GFP intensity was quantified at the cell membrane (marked by phalloidin) and in the cytoplasm using Fiji (Schindelin et al., 2012). Three cells per fat body were measured, and at least five fat bodies per sex and per diet were measured.

Statistics and data presentation

Request a detailed protocol

Statistical analyses and data presentation were carried out using Prism GraphPad 6 (GraphPad Prism version 8.4.3 for Mac OS X). Statistical tests are indicated in figure legends and all p-values are listed in Supplementary file 1.

References

    1. Anand A
    2. Villella A
    3. Ryner LC
    4. Carlo T
    5. Goodwin SF
    6. Song HJ
    7. Gailey DA
    8. Morales A
    9. Hall JC
    10. Baker BS
    11. Taylor BJ
    (2001)
    Molecular genetic dissection of the sex-specific and vital functions of the Drosophila melanogaster sex determination gene fruitless
    Genetics 158:1569–1595.
    1. Britton JS
    2. Edgar BA
    (1998)
    Environmental control of the cell cycle in Drosophila: Nutrition activates mitotic and endoreplicative cells by distinct mechanisms
    Development 125:2149–2158.
    1. Cline TW
    (1978)
    Two closely linked mutations in Drosophila melanogaster that are lethal to opposite sexes and interact with daughterless
    Genetics 90:683–698.
    1. Goodwin SF
    2. Taylor BJ
    3. Villella A
    4. Foss M
    5. Ryner LC
    6. Baker BS
    7. Hall JC
    (2000)
    Aberrant splicing and altered spatial expression patterns in fruitless mutants of Drosophila melanogaster
    Genetics 154:725–745.
    1. Lewis EB
    (1960)
    A new standard food medium
    Drosophila Information Service 34:117–118.
    1. Sturtevant AH
    (1945)
    A gene in Drosophila melanogaster That Transforms Females into Males
    Genetics 30:297–299.

Decision letter

  1. Jiwon Shim
    Reviewing Editor; Hanyang University, Republic of Korea
  2. Utpal Banerjee
    Senior Editor; University of California, Los Angeles, United States

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

The insulin pathway differentially affects sex-specific phenotypes in most species, yet the molecular mechanisms underlying male-female differences in its regulation remain unclear. In this study, Rideout and colleagues investigated how females grow bigger in a protein-rich diet and unraveled that the sex determination gene, transformer, promotes a diet-induced trigger in Insulin signaling via stunted and srl/PGC-1a. This finding provides interesting genetic evidence for sex differences in complex phenotypes of development, physiology, and diseases.

Decision letter after peer review:

Thank you for submitting your article "Female-specific upregulation of insulin pathway activity mediates the sex difference in Drosophila body size plasticity" for consideration by eLife. Your article has been reviewed by three peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by Utpal Banerjee as the Senior Editor. The reviewers have opted to remain anonymous.

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

We would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). Specifically, when editors judge that a submitted work as a whole belongs in eLife but that some conclusions require a modest amount of additional new data, as they do with your paper, we are asking that the manuscript be revised to either limit claims to those supported by data in hand, or to explicitly state that the relevant conclusions require additional supporting data.

Our expectation is that the authors will eventually carry out the additional experiments and report on how they affect the relevant conclusions either in a preprint on bioRxiv or medRxiv, or if appropriate, as a Research Advance in eLife, either of which would be linked to the original paper.

Summary:

This is an interesting study addressing how sex-specific differences in endocrine signaling influences nutrient-dependent body size plasticity. The authors demonstrated that females, but not males, increase growth when raised on a high protein diet and revealed that growth difference results for sex-specific regulation of the insulin signaling pathway. Further, the authors identified that stunted expression is required for nutrient-dependent body size plasticity. Finally, the authors showed that the sex determination gene transformer is required for the female-specific expression of stunted on the high protein diet.

All the reviewers found the study very interesting, well designed and performed, and clearly written. Reviewers also agreed that the main finding of this study is novel and will be of broad interest in the field. However, at the same time, the reviewers raised several concerns about relatively weak points that need to be adequately addressed.

Essential revisions:

1) Identify the link between Tra and sun (related to reviewer 1 concerns 1-4, reviewer 2 concerns 4-5).

It will be critical to show the levels of circulating sun in females 1Y/2Y and males 1Y/2Y and/or the level of TOR signaling in the fat body in the four conditions.

2) Verify spargel mutant phenotype (e.g. sun expression in the spargel mutant) (related to reviewer 3).

3) Explain/repeat variable expressions of the IIS markers in males and females (reviewer 2 concerns 1-2).

4) Explain the relevance of 4EBP expression in r4>sun RNAi (2Y) condition (reviewer 2 concern 3).

5) Statistical analysis.

Some of these points may require additional experimental data and analyses. I have attached the original comments to clearly deliver reviewers' specific points.

Reviewer #1:

In this study, Rideout and colleagues investigated novel mechanisms underlying the female-specific size plasticity upon high protein diet and identified dilp2, sun, and tra as key molecules for the size control. Overall, experiments are well-performed, analyzed, and presented clearly. Also, the manuscript is well-written.

In the senior author's previous study, the author already showed a novel role for tra in the female body size determination through dilp2 in the IPCs under normal diet conditions. The function sun in the female-size plasticity is first shown and is the novel point of this work, and therefore, mechanisms involving tra-sun and sun-dilp2 require additional verifications. The manuscript describes the functions of tra, sun, or dilp2 on its own without showing the genetic/biochemical relationships of the three.

1) Is sun a direct transcriptional target of tra? The key question that needs to be addressed is whether tra functions upstream of sun and tra plays a role in the sun transcription.

2) Related to the above question, is sun the only target of tra? The authors investigated possible roles of humoral factors; however, downstream targets of tra may not be necessarily humoral factors.

3) No proof for the genetic relationship between tra-sun and sun-dilp2 is shown. For example, would overexpression of sun in tra mutant females rescue the nutrient-dependent growth?

4) Is mthl in the IPCs involved in this pathway?

5) To separate the body size difference and body size plasticity, it would be better to show the weight or volume changed (2Y minus 1Y) in different conditions for the plasticity, and absolute weight/volume numbers for growth.

6) It is interesting that InR/sun also controls egg production. However, without providing a detailed mechanism underlying this phenotype, this part makes the paper more complicated. Is tra also involved? Is this phenotype solely due to dilp2 and subsequent activation of InR in the ovary?

Reviewer #2:

This is an interesting study addressing the molecular basis of nutrient-dependent body size plasticity in Drosophila. The authors re-evaluate the sex-difference in nutritional plasticity in Drosophila and show that yeast (amino acids?) is the main component that drives plasticity in females, while males remain insensitive to increased yeast content (at least within the range of the experimental conditions used in the present study). They further imply insulin/IGF signaling (IIS) in this control, which is somehow expected. The novelty comes from the elucidation of the role of Stunted (Sun), a fat body factor controlling the level of circulating Drosophila insulin-like peptides (dilps), in the phenotypic plasticity observed in females. Whereas sun expression is increased in females raised on rich versus poor medium, this is not true in males. By knocking down sun in the fat body of larvae, they demonstrate the need for sun in this sex-specific regulation, and link it to the function of the sex-determination factor Transformer (Tra).

The data globally fits with the conclusions and the paper is rather convincing. It definitely brings a novel molecular twist to the interesting question of sex-specific nutritional plasticity. However, there are several issues with the experimental aspects that need to be corrected before the paper is ready.

1) Figure 2A,B: the markers for IIS show no variation in males fed 1Y or 2Y. However, this is not the case in the same experiments presented in Figure 4—figure supplement 2A, where Inr and 4E-BP go down in 2Y condition, as in females (see r4/+ and +>sun-RNAi controls). This casts doubts on the reproducibility of such analysis.

2) Figure 4—figure supplement 1C,D: statistical significance should compare +>sun-RNAi and r4>sun-RNAi, since they correspond to lower control values.

3) What is the significance of 4E-BP levels being increased in 2Y in r4>sun-RNAi conditions (both in females and males)? This would mean that IIS is generally reduced, which does not make sense. Therefore, what is the value of measuring 4E-BP as a marker for IIS?

4) Sun is secreted in the hemolymph and its circulating levels are controlled by TORC1 activity in FB cells (Delanoue et al., 2016). Therefore, an evaluation of circulating levels of Sun should be provided to better characterize female and male physiological responses to 1Y vs 2Y. Indeed, the results presented in Figure 4—figure supplement 4AB suggest that dysregulation of sun at the transcriptional level does not alter nutritional response in females and males. Looking at this figure, it is questionable whether nutritional plasticity is different in males and females of the r4>sun genotype.

5) Again, concerning Sun regulation, what is the link between Tra activiy and sun expression in response to Y content in females? Can the authors relay the level of Sun in 1Y vs 2Y to a difference in TOR activity specifically in female FB cells?

Reviewer #3:

This very well written manuscript by Millington et al., uses the fruit fly Drosophila melanogaster to understand how sex-specific differences in endocrine signaling influences nutrient-dependent body size plasticity. Through a series of well-designed experiments, the authors demonstrate that females, but not males, exhibit increase growth when raised on a diet with twice the nutrient diet of standard fly food. Through a series of logical experiments, the authors reveal that growth difference results for sex-specific regulation of the insulin signaling pathway – female, but not male, flies express increased levels of the humeral factor stunted, which is known to promote dilp2 secretion from the IPCs. The authors further demonstrate that the manner by which female flies regulated stunted expression is required for nutrient-dependent body size plasticity. Finally, the authors demonstrate that the sex determination gene transformer is required for female-specific expression of stunted on the high nutrient diet.

Overall, I found this a very nice story that works its way from a simple observation to a molecular mechanism. The story will be of broad interest and highlights the importance of studying sex-specific differences in animal growth and development. I have a few suggested revisions, but overall enjoyed reading the manuscript.

1) The authors used animals that are heterozygous for mutations in the gene spargel as a substitute for analyzing stunted mutants. While I understand the necessity of this experiment, spargel mutants have a wide range of metabolic defects that are independent of stunted and I'm concerned that this experiment requires a leap of faith. At a minimum, I'd like to see verification that heterozygous spargel mutants exhibit significant changes in stunted gene expression.

https://doi.org/10.7554/eLife.58341.sa1

Author response

Essential revisions:

1) Identify the link between Tra and sun (related to reviewer 1 concerns 1-4, reviewer 2 concerns 4-5).

We were also interested in the link between Tra and sun. To address Reviewer concerns, we completed several experiments, which we summarize below.

a) What is the link between Tra and sun mRNA levels?

We added a significant amount of data to the revised manuscript to indicate that spargel (srl), the Drosophila homolog of PGC-1a, represents one key link between Tra and the nutrient-dependent regulation of sun mRNA levels. In females, we show that heterozygous loss of srl blocks the nutrient-dependent upregulation of sun mRNA levels in a protein-rich diet (Figure 5A). In Tra-expressing males, heterozygous loss of srl similarly blocks the nutrient-dependent increase in sun mRNA levels (Figure 5D). Together, these experiments indicate that Srl is one link between Tra and the nutrient-dependent regulation of sun mRNA levels. Of note, srl1 heterozygotes have no generalized growth defects (Figure 5B, C), suggesting these larvae do not have wide-ranging metabolic defects.

b) What is the genetic relationship between tra and sun?

In the revised version of our manuscript, we performed genetic epistasis experiments that suggest both Srl and sun lie downstream of Tra in regulating the sex difference in nutrient-dependent body size plasticity. For example, heterozygous loss of srl blocked the Tra-dependent upregulation of IIS activity in 2Y (Figure 5E) and eliminated the increased nutrient-dependent body size plasticity we observed in Tra-expressing males (Figure 5F).

Fat body-specific expression of a UAS-sun-RNAi transgene in Tra-expressing males similarly abolished the nutrient-dependent increase in body size (Figure 4H). Importantly, we rescued the smaller body size of tra mutant females simply by overexpressing sun in the fat body (Figure 4—figure supplement 2A). Together, these experiments support a model in which Tra influences phenotypic plasticity via Srl-mediated regulation of sun mRNA levels, which we communicate to readers by including a new figure summarizing our findings (Figure 7).

c) Is sun a direct Tra target?

Our discovery of a link between sex determination gene Tra and metabolic regulator Srl significantly advances our understanding of how Tra influences gene expression, as known Tra targets dsx and fru do not affect body size or several other Tra-regulated phenotypes (Rideout et al., 2015; Hudry et al., 2016; Garner et al., 2018). While we were not able to determine whether sun is a direct target of splicing factor Tra, we added text to the revised manuscript to highlight the importance of this question for future studies on Tra, Srl, and sun.

Importantly, while Srl targets other than sun were also regulated in a sex-specific and Tra-dependent manner (Figure 5—figure supplement 2A-D), two other Srl targets did not reproduce the effects of sun on phenotypic plasticity (Figure 5—figure supplement 2E-H). This strengthens our finding that sun is a key factor that impacts the sex difference in nutrient-dependent body size plasticity. However, to acknowledge that we cannot rule out all Srl targets in regulating phenotypic plasticity we added text to this effect in the revised manuscript (Discussion).

d) What is the role of sun mRNA levels in regulating body size?

Identifying Srl as the link between Tra and the regulation of sun mRNA levels suggests that an increase in sun mRNA levels should be able to increase body size. In our 1Y and 2Y diets, we found that sun overexpression in the fat body was sufficient to enhance body size in both sexes (Figure 3—figure supplement 7A, B). This finding indicates that increased sun mRNA levels are able to increase body size, and supports a model in which the nutrient-dependent increase in sun mRNA levels promotes growth to augment body size.

While this finding differs from results reported in a previous study (Delanoue et al., 2016), when we repeated their experiment using identical dietary conditions we found a male-specific increase in body size (Figure 3—figure supplement 8A). Interestingly, this increase in male body size was lost when we combined body size data from both sexes (Figure 3—figure supplement 8B). This indicates that the previous study failed to see a body size increase with sun overexpression due to a combination of dietary factors (we do not supplement food with live yeast to maintain tight control over nutrient content) and not analyzing body size data by sex. To clarify this point for readers, we added text to the revised manuscript (subsection “Increased nutrient-dependent body size plasticity in females promotes fecundity in a protein-rich context”).

e) Circulating Sun levels in males and females.

We were fortunate to receive enough anti-Sun antibody to perform one Western blot to measure hemolymph Sun levels in males and females (Figure 3—figure supplement 1A). Our quantification of the hemolymph Sun blot suggests that levels are nearly twice as high in females as they are in males reared in 2Y (Figure 3—figure supplement 1A). This finding supports a model in which higher levels of circulating Sun in females contribute to their larger body size.

f) Fat body TOR levels.

A previous study showed an important role for the Target-of-Rapamycin (TOR) pathway in mediating the nutrient-dependent secretion of Sun (Delanoue et al., 2016). When we measured fat body TOR activity using an antibody directed against a phosphorylated form of ribosomal protein S6 kinase (pS6k), a known TOR target, we saw no differences in pS6k levels between males and females in either the 1Y or the 2Y diet (Figure 5—figure supplement 1A-D). Similarly, when we examined changes to pS6k levels between control females and tra mutant females, there was no significant difference in fat body pS6k levels between females with and without Tra function on either diet (Figure 5—figure supplement 1F, G).

Therefore, while TOR activity undoubtedly plays a role in regulating Sun secretion, we did not find evidence of sex-specific or tra-dependent regulation of fat body TOR activity. This aligns with our previous finding that treating larvae with TOR inhibitor rapamycin does not have sex-biased effects on body size (Rideout et al., 2015). Because we show that changes to sun mRNA levels were sufficient to augment body size (Figure 3—figure supplement 7A, B), and that fat body TOR activity does not affect sun mRNA levels (Figure 5—figure supplement 1E), the data in our revised manuscript supports a role for the sex-specific regulation of sun mRNA levels via Srl as one mechanism underlying the male-female difference in body size plasticity.

2) Verify spargel mutant phenotype (e.g. sun expression in the spargel mutant) (related to reviewer 3).

Reviewer 3 raised important points about our use of srl1/+ larvae that we addressed in our revised manuscript. First, to ensure that heterogyzous loss of srl does not cause generalized developmental defects, we measured body size in 1Y. We found no body size reduction in srl1/+ larvae of either sex compared with w1118 control larvae (Figure 5B, C). Given that body size encompasses the activity of many metabolic genes and pathways (Boulan et al., 2015), this suggests that srl1/+ larvae do not have generalized metabolic defects.

Second, we verified that heterozygous loss of srl blocks the nutrient-dependent upregulation of sun mRNA levels. In w1118 females, we normally observe a nutrient-dependent upregulation of sun RNA in larvae raised on 2Y compared with larvae cultured on 1Y (Figure 5A); however, the nutrient-dependent increase in sun mRNA was blunted in srl1/+ females reared on 2Y compared with genotype-matched controls raised on 1Y (Figure 5A). Similarly, in Tra-expressing males (da>traF), the nutrient-dependent increase in sun mRNA levels was blocked when those males were heterozygous for srl1 (Figure 4F and Figure 5D).

Taken together, the data we present in our revised manuscript suggests that Srl plays a key role in the nutrient-dependent upregulation of sun mRNA levels in a protein-rich context, and that Srl mediates Tra’s effects on sun mRNA regulation.

3) Explain/repeat variable expressions of the IIS markers in males and females (reviewer 2 concerns 1-2).

We also noticed variability in levels of genes we used to quantify IIS activity between different male groups. In our original manuscript, we did not apply statistical tests to detect potential sex:diet and genotype:diet interactions in our gene expression data. In our revised manuscript, we improved our statistical analysis by applying these more rigorous tests to all of our gene expression data. Further, we identified an established way of displaying and analyzing co-regulated genes (e.g., Blaschke et al., 2013; Hudry et al., 2019) so that we no longer need to divide our panels displaying mRNA levels of Foxo target genes between the main and supplemental figures.

In our revised manuscript, we show that levels of Foxo target genes were significantly lower in all control females reared in 2Y compared with females reared in 1Y (Figure 1E, Figure 2A, Figure 3E, Figure 4A). Given that high levels of IIS activity repress Foxo target genes, this indicates higher IIS activity in females reared in 2Y. This improved statistical analysis also shows that the magnitude of the increase in IIS activity in dilp2 mutant females, females with fat body loss of sun, and tra mutant females was significantly smaller than in control females (Figure 2A, Figure 3E, Figure 4A). This indicates that females with reduced dilp2, fat body sun, and tra were not able to upregulate IIS activity as much as control females.

When we repeated several key experiments in males, we still observed some variation in IIS readouts: males had either no change in Foxo target gene expression (Figure 1G), or a small but significant decrease in Foxo target genes (Figure 2B, Figure 3F). This suggested to us that males normally have a small but significant nutrient-dependent increase in IIS activity. To acknowledge this fact, we changed all instances of “female-specific” to “female-biased” in our revised manuscript when we refer to IIS activity.

Despite these minor differences among male genotypes, however, the most important conclusion we reached in our revised manuscript was that the magnitude of any nutrient-dependent changes to Foxo target genes in control males was always smaller than in genotype-matched females (sex:diet interactions in Supplementary file 1). This consistent and reproducible female-biased upregulation of IIS activity across all genotypes therefore supports one main finding of our paper: that a sex difference exists in the nutrient-dependent upregulation of IIS activity.

To display the female-biased decrease in Foxo target gene expression in a protein-rich diet more clearly, we included the % change in Foxo target gene expression for each genotype. This makes it is easier for readers to appreciate the sex difference in magnitude of nutrient-dependent changes to Foxo target gene expression.

4) Explain the relevance of 4EBP expression in r4>sun RNAi (2Y) condition (reviewer 2 concern 3).

4E-BP was one Foxo target gene among three Foxo targets that we measured to quantify IIS activity. In our revised manuscript, we sought a more rigorous way of analyzing gene expression for all three Foxo target genes so that we would not have to draw conclusions based on individual genes (e.g. 4E-BP). One established way we found to analyze co-regulated genes was to examine the behaviour of the genes as a group (Blaschke et al., 2013; Hudry et al., 2019). This way of analyzing gene expression allowed us to display Foxo target genes in a single graph, and to perform better statistical tests to detect genotype:diet and sex:diet interactions.

Using this improved statistical analysis, we showed that in r4>+ and +>UAS-sun-RNAi females there was a significant decrease in Foxo target gene expression between 1Y and 2Y that was absent in r4>UAS-sun-RNAi females (Figure 3E). Given that there was a significant diet:genotype interaction (p < 0.0001), this suggests that the magnitude of change in Foxo target gene expression was different between r4>UAS-sun-RNAi females and r4>+ and +>UAS-sun-RNAi controls. In contrast, there was no significant diet:genotype interaction between r4>UAS-sun-RNAi males and control males (p = 0.1068), indicating that Foxo target gene expression was not different in r4>sun-RNAi males compared with r4>+ and +>UAS-sun-RNAi control males (Figure 3F).

When we examined sex:diet interactions, we found that the magnitude of the nutrient-dependent change to Foxo target genes was greater in females than males for the r4>+ and +>UAS-sun-RNAi genotype, but not the r4>UAS-sun-RNAi genotype (p = 0.0166, 0.0119, and 0.1121, respectively). Thus, our improved statistical analysis of gene expression data indicates that the loss of fat body sun blocks the nutrient-dependent increase in IIS activity in females, but not in males.

5) Statistical analysis.

We thank the reviewer for pointing out that our original figures did not clearly communicate the fact that differences were only indicated as significant if the experimental genotype (e.g. r4>sun-RNAi) was significantly different from all control genotypes (e.g. r4>+ and +>sun-RNAi). In our revised manuscript, we added lines to each graph to show all statistical comparisons that were made. The p-values for all multiple comparisons can be found in Supplementary file 1.

Reviewer #1:

In this study, Rideout and colleagues investigated novel mechanisms underlying the female-specific size plasticity upon high protein diet and identified dilp2, sun, and tra as key molecules for the size control. Overall, experiments are well-performed, analyzed, and presented clearly. Also, the manuscript is well-written.

In the senior author's previous study, the author already showed a novel role for tra in the female body size determination through dilp2 in the IPCs under normal diet conditions. The function sun in the female-size plasticity is first shown and is the novel point of this work, and therefore, mechanisms involving tra-sun and sun-dilp2 require additional verifications. The manuscript describes the functions of tra, sun, or dilp2 on its own without showing the genetic/biochemical relationships of the three.

We thank the reviewer for their thoughtful comments on our manuscript, and for their suggestion that we examine the relationships between tra, sun, and dilp2 in more detail. We describe specific experiments in detail below.

1) Is sun a direct transcriptional target of tra?

For the sake of clarity, we answered this question below our two-part comments on whether tra functions upstream of sun transcription.

The key question that needs to be addressed is whether tra functions upstream of sun

In our original paper, we showed that females lacking tra were unable to upregulate sun mRNA levels when raised in a protein-rich diet (Figure 4B), whereas Tra expression in males was sufficient to enable the nutrient-dependent increase in sun mRNA levels in the 2Y diet (Figure 4F). While this suggests that tra may lie upstream of sun, in our revised manuscript we used a genetic approach to strengthen our conclusions about the relationship between Tra and sun. Further, we identify Srl as one link between Tra and sun mRNA levels (next point).

In our revised manuscript, we show that fat body-specific sun knockdown blocked the nutrient-dependent increase in body size we observed in Tra-expressing males (Figure 4H). Further, we demonstrate that fat body sun overexpression was sufficient to restore the smaller body size of tra mutant females raised in a protein-rich diet (Figure 4—figure supplement 2A), suggesting that sun lies downstream of Tra in promoting body size in this context.

Together, these new data support a model in which Tra promotes increased body size in a protein-rich context by acting upstream of sun. To ensure that these findings are clearly communicated to the reader, we included these new data and a new summary figure to the revised manuscript (Figure 7).

and tra plays a role in the sun transcription.

We were also curious about the link between Tra and regulation of sun mRNA levels, as two transcription factors known to mediate Tra’s gene expression effects do not affect body size (doublesex [dsx] and fruitless [fru]; Rideout et al., 2015). In our revised manuscript, we identify spargel (srl), the Drosophila homolog of PGC-1a, as one link between Tra and regulation of sun mRNA levels.

A previous study showed that Srl regulates sun mRNA levels in response to dietary protein (Delanoue et al., 2016), which we confirmed by showing that female larvae heterozygous for srl1 were unable to augment sun mRNA levels in a protein-rich context (Figure 5A). This suggests that normal Srl function was required for the nutrient-dependent upregulation of sun mRNA in females. Importantly, we confirmed that the srl1/+ larvae show no obvious developmental defects (Figure 5B, C).

To determine whether Srl mediates Tra’s effects on sun mRNA levels in a protein-rich context, we monitored nutrient-dependent changes to sun mRNA levels in Tra-expressing males, and in Tra-expressing males carrying the srl1 allele. While males with ectopic Tra expression normally upregulate sun mRNA levels in a protein-rich context (Figure 4F), heterozygous loss of Srl function in these Tra-expressing males blocked the nutrient-dependent upregulation of sun mRNA (Figure 5D). This suggests that Srl function mediates Tra’s effects on sun mRNA levels, identifying a new mechanism by which Tra affects gene expression.

Because Tra-expressing males carrying the srl1 allele no longer increase IIS activity or body size in a protein-rich context (Figure 5E, F), in contrast to Tra-expressing males with normal Srl function (Figure 4E, G), the data in our revised manuscript suggests that the regulation of sun mRNA levels by Srl plays an important role in mediating Tra’s effects on phenotypic plasticity. To reflect the importance of this new data, we added these figures, a significant amount of new text, a new model figure (Figure 7), and a new section to the Discussion.

Is sun a direct transcriptional target of tra?

Much of our knowledge about how Tra regulates gene expression comes from studies on how Tra, a splicing regulator, controls the sex-specific splicing of pre-mRNA of transcription factors dsx and fru. Because Tra impacts body size independently of these two genes (Rideout et al., 2015), our discovery of transcriptional coactivator Srl as one link between Tra and regulation of sun mRNA levels reveals a new way in which splicing factor Tra impacts gene expression.

While we adjusted the text to reflect the fact that the precise biochemical interactions between Tra, Srl, and sun require further study, as Tra is a splicing factor and Srl is a transcriptional coactivator that partners with multiple transcription factors to influence transcript levels (Tiefenbock et al., 2010) (Discussion), we also added text to highlight how uncovering the Tra-Srl link advances our understanding of Tra-dependent changes to gene expression independently of dsx and fru (Discussion).

For example, there is a rapidly growing body of literature describing Tra-dependent but dsx- and fru-independent effects on lifespan, organ plasticity, and neural circuits (Rideout et al., 2015; Hudry et al., 2016; Regan et al., 2016; Castellanos et al., 2013; Garner et al., 2018); however, the molecular mechanisms underlying these dsx- and fru-independent effects remain unknown. Our findings will therefore help researchers studying sex differences in traits such as lifespan, organ plasticity, and neural circuits by providing new insight into the mechanisms underlying Tra-dependent but dsx- and fru-independent effects on gene expression.

2) Related to the above question, is sun the only target of tra? The authors investigated possible roles of humoral factors; however, downstream targets of tra may not be necessarily humoral factors.

We previously showed that tra’s only confirmed direct downstream targets dsx and fru do not affect body size (Rideout et al., 2015); however, given that Srl mediates the nutrient-dependent upregulation of sun mRNA levels downstream of Tra, we also looked at gene expression changes in other known Srl targets. Like sun, we found that other confirmed Srl targets were regulated in a sex-specific and nutrient-dependent manner (Figure 5—figure supplement 2A-D). Yet when we used RNAi to knock down levels of Srl targets other than sun, loss of two additional Srl target genes that are functionally similar to sun did not affect phenotypic plasticity (Figure 5—figure supplement 2E-H). Note: loss of fat body bellwether (blw; FBgn0011211) and cytochrome c oxidase subunit 5a (Cox5a; FBgn0019624) did not produce viable animals for our measurements.

While this data suggests that sun plays an important role among Srl targets in mediating sex-specific body size plasticity downstream of Tra, we adjusted the text in our revised manuscript to reflect the fact that we cannot rule out all possible Srl targets in regulating sex-specific or nutrient-dependent growth.

3) No proof for the genetic relationship between tra-sun and sun-dilp2 is shown. For example, would overexpression of sun in tra mutant females rescue the nutrient-dependent growth?

We include this important experiment in our revised manuscript, and confirm that the overexpression of fat body sun in a tra mutant female is able to fully rescue the smaller body size in these females when they are raised in a protein-rich diet (Figure 4—figure supplement 2A). Further, we show that loss of fat body sun in a Tra-expressing male blocks the increased phenotypic plasticity we normally observe in males with ectopic Tra expression (Figure 4H). Together, these results strengthen a model in which sun lies downstream of Tra in regulating nutrient-dependent changes to body size. To address the sun-Dilp2 relationship, in our revised manuscript we highlight the elegant and comprehensive work that was done in the Léopold lab to show that sun impacts body size via regulation of Dilp2 secretion to ensure that the reader is aware of this important body of work.

4) Is mthl in the IPCs involved in this pathway?

To address a role for mth in the insulin-producing cells (IPCs) in the brain in this pathway, we used dilp2-GAL4 to knock down mth levels using RNAi. Unlike all other control strains we examined in this study, the dilp2-GAL4>+ control strain did not show a sex difference in nutrient-dependent body size plasticity (sex:diet interaction, p = 0.9995). We therefore used two alternative approaches to investigate a potential role for mth function in phenotypic plasticity: RNAi-mediated knock down of mth in all neurons, and whole-body loss of mth. Neither pan-neuronal, nor global, loss of mth significantly affected phenotypic plasticity (Figure 3—figure supplement 5A-F). The apparent discrepancy with the previous study is most likely due to differences in lab diets, using different dilp2-GAL4 lines, and the fact that we are not reproducing the original experiment with IPC-specific mth knockdown. While we included these new results in the revised manuscript (Figure 3—figure supplement 5A-F), we added text accompanying these figures to ensure the reader is aware of the limitations of our data compared with the original findings in Delanoue et al., 2016.

5) To separate the body size difference and body size plasticity, it would be better to show the weight or volume changed (2Y minus 1Y) in different conditions for the plasticity, and absolute weight/volume numbers for growth.

We thank the reviewer for this suggestion. In order to ensure consistency between our data on pupal volume/adult weight data and past papers on Sun (Delanoue et al., 2016), we chose to maintain our current data presentation style. This will ensure that the audience is easily able to compare our data with past literature on Sun, and more generally within the larval growth field.

6) It is interesting that InR/sun also controls egg production. However, without providing a detailed mechanism underlying this phenotype, this part makes the paper more complicated. Is tra also involved? Is this phenotype solely due to dilp2 and subsequent activation of InR in the ovary?

We were also very interested in the link between nutrition, IIS and egg production. To ensure we acknowledge the body of literature in this area, we added text and citations to ensure that the reader is aware of past studies showing that increased circulating levels of Dilp proteins and nutrition can enhance egg production by increasing ovariole number. Further, we suggest future studies that should be completed in order to gain detailed insight into the fertility phenotypes that arise from changes to InR, sun, and dilp2.

Reviewer #2:

This is an interesting study addressing the molecular basis of nutrient-dependent body size plasticity in Drosophila. The authors re-evaluate the sex-difference in nutritional plasticity in Drosophila and show that yeast (amino acids?) is the main component that drives plasticity in females, while males remain insensitive to increased yeast content (at least within the range of the experimental conditions used in the present study). They further imply insulin/IGF signaling (IIS) in this control, which is somehow expected. The novelty comes from the elucidation of the role of Stunted (Sun), a fat body factor controlling the level of circulating Drosophila insulin-like peptides (dilps), in the phenotypic plasticity observed in females. Whereas sun expression is increased in females raised on rich versus poor medium, this is not true in males. By knocking down sun in the fat body of larvae, they demonstrate the need for sun in this sex-specific regulation, and link it to the function of the sex-determination factor Transformer (Tra).

The data globally fits with the conclusions and the paper is rather convincing. It definitely brings a novel molecular twist to the interesting question of sex-specific nutritional plasticity. However, there are several issues with the experimental aspects that need to be corrected before the paper is ready.

We thank the reviewer for their careful reading and analysis of the paper. We address specific points and suggestions for improvement below.

1) Figure 2A,B: the markers for IIS show no variation in males fed 1Y or 2Y. However, this is not the case in the same experiments presented in Figure 4—figure supplement 2A, where Inr and 4E-BP go down in 2Y condition, as in females (see r4/+ and +>sun-RNAi controls). This casts doubts on the reproducibility of such analysis.

In our original manuscript, we did not apply statistical tests to detect potential sex:diet and genotype:diet interactions in our gene expression data. In our revised manuscript, we improved our statistical analysis by applying these more rigorous tests to all of our gene expression data. Further, we identified an established way of displaying and analyzing co-regulated genes (e.g., Blaschke et al., 2013; Hudry et al., 2019) so that we no longer need to divide our panels displaying mRNA levels of Foxo target genes between the main and supplemental figures.

These changes allowed us to make more accurate conclusions about the behavior of Foxo target genes in response to variables such as sex, diet, and genotype. Also, we arranged the data in a better way to highlight the female-biased decrease in Foxo target gene expression in a protein-rich diet. Additional measures to help the reader appreciate the sex difference in the magnitude of gene expression changes include displaying the % change in Foxo target gene expression for each genotype.

In our revised manuscript we show that levels of Foxo target genes were significantly lower in all control females reared in 2Y compared with females reared in 1Y (Figure 1E, Figure 2A, Figure 3E, Figure 4A, Figure 4—figure supplement 3A). Given that high levels of IIS activity repress Foxo target genes, this indicates higher IIS activity in 2Y, as we previously showed. In males, when we repeated several key experiments, we still observed some variation in IIS readouts: we found either no change in Foxo target gene expression (Figure 1G), or a small but significant decrease in Foxo target genes (Figure 2B, Figure 3F). This suggests that males normally have a small but significant nutrient-dependent increase in IIS activity. To acknowledge this fact, we changed all instances of “female-specific” to “female-biased” in our revised manuscript when we refer to IIS activity.

Despite these minor differences among male genotypes, however, the most important conclusion we reached in our revised manuscript was that the magnitude of any nutrient-dependent change to Foxo target genes in control males was always smaller than in genotype-matched females (sex:diet interactions in Supplementary file 1). This reproducible female-biased upregulation of IIS activity across all genotypes therefore supports our finding that a sex difference exists in the nutrient-dependent upregulation of IIS activity.

With respect to the r4>UAS-sun-RNAi experiment, we show that r4>+ and +>sun-RNAi males have a significant decrease in mRNA levels of Foxo target genes and r4>sun-RNAi males males do not; however, there was no significant genotype:diet interaction among r4>+, +>sun-RNAi, and r4>sun-RNAi males (p = 0.1068). This indicates that there was no effect of genotype on Foxo target gene expression in males (Figure 3F), in contrast to the significant genotype:diet interaction we observed in females (p < 0.0001) (Figure 3E). Importantly, the magnitude of the reduction in Foxo target gene expression in control males was smaller than in genotype-matched females (sex:diet interactions p = 0.0166 and 0.0119, respectively), but not different between r4>sun-RNAi males and females (p = 0.1121).

Overall, we obtained a more accurate picture of nutrient-dependent changes to IIS activity in each sex and genotype, across two diets, by applying more rigorous statistical tests to our analysis of Foxo target genes. Together, we believe these changes support our conclusion that there is a female-biased increase in IIS activity in response to dietary protein, and that this nutrient-dependent increase in IIS activity requires dilp2, sun, and tra function.

2) Figure 4—figure supplement 1C,D: statistical significance should compare +>sun-RNAi and r4>sun-RNAi, since they correspond to lower control values.

We thank the Reviewer for highlighting that our graphs do not clearly indicate that our statistical analyses make multiple comparisons between all three genotypes included in this, and other, experiments. We have adjusted all of the graphs in the manuscript to show the statistical comparisons between all the genotypes for the sake of clarity.

3) What is the significance of 4E-BP levels being increased in 2Y in r4>sun-RNAi conditions (both in females and males)? This would mean that IIS is generally reduced, which does not make sense. Therefore, what is the value of measuring 4E-BP as a marker for IIS?

We also wondered about why 4E-BP for some genotypes does not correspond with the behavior of other Foxo target genes. We therefore sought a way of analyzing the Foxo target genes together to gain a better picture of Foxo activity than we would obtain by drawing conclusions based on the behaviour of single genes. One established way to analyze co-regulated genes is to examine the behavior of the genes as a group (Blaschke et al., 2013; Hudry et al., 2019), and applying rigorous statistical tests to detect genotype:diet and sex:diet interactions.

Using this improved statistical analysis, we showed that in r4>+ and +>UAS-sun-RNAi females there was a significant decrease in Foxo target gene expression between 1Y and 2Y that was absent in r4>UAS-sun-RNAi females (Figure 3E). Given that there was a significant diet:genotype interaction (p < 0.0001), this suggests that the magnitude of change to Foxo target gene expression was different between r4>UAS-sun-RNAi females and r4>+ and +>UAS-sun-RNAi controls. In contrast, there was no significant diet:genotype interaction between r4>UAS-sun-RNAi males and control males (p = 0.1068), indicating that Foxo target gene expression was not different in r4>sun-RNAi males compared with r4>+ and +>UAS-sun-RNAi control males (Figure 3F).

Importantly, sex:diet interactions showed that the magnitude of the nutrient-dependent change to Foxo target genes was greater in females than males for the r4>+ and +>UAS-sun-RNAi genotype, but not the r4>UAS-sun-RNAi genotype (p = 0.0166, 0.0119, and 0.1121, respectively). Thus, our improved statistical analysis of gene expression data for this experiment supports our conclusion that the loss of fat body sun blocks the nutrient-dependent increase in IIS activity in females, but not in males.

4) Sun is secreted in the hemolymph and its circulating levels are controlled by TORC1 activity in FB cells (Delanoue et al., 2016). Therefore, an evaluation of circulating levels of Sun should be provided to better characterize female and male physiological responses to 1Y vs 2Y.

We were also curious about circulating Sun levels, and a potential role for TOR in males and females on both diets. We first measured hemolymph Sun levels in males and females raised on 1Y and 2Y. We were fortunate to receive enough antibody to make 0.5 ml of primary antibody solution to perform this Western blot. We found that hemolymph levels of Sun were approximately twice as high in males as in females (Figure 3—figure supplement 1A), suggesting that there is a sex difference in circulating Sun.

To determine whether fat body TOR may affect the sex difference in circulating Sun, we measured TOR activity in fat bodies isolated from males and females reared in 1Y and 2Y. We found that there were no sex differences in phospho-S6k (pS6k) levels, an established readout for TOR signaling, in either diet (Figure 5—figure supplement 1A-D). Similarly, we found no significant changes to pS6k levels between fat bodies isolated from control and tra mutant females (Figure 5—figure supplement 1F, G).

This suggests that TOR activity does not normally differ between males and females, or between control females and tra mutant females, indicating that TOR may not be the sole determinant of circulating Sun levels. Given that we show fat body sun overexpression increases body size in multiple nutritional contexts when larvae are analyzed according to sex (Figure 3—figure supplement 7A, B, Figure 3—figure supplement 8A), it will be interesting to test whether increased sun mRNA promotes circulating Sun levels. Anti-Sun antibody quantities are too limiting for us to do this experiment at present; however, we added text to highlight the importance of this experiment (subsection “A nutrient-dependent increase in stunted mRNA levels is required for enhanced IIS activity and a larger body size plasticity in females cultured in a protein-rich context”).

Indeed, the results presented in Figure 4—figure supplement 4AB suggest that dysregulation of sun at the transcriptional level does not alter nutritional response in females and males. Looking at this figure, it is questionable whether nutritional plasticity is different in males and females of the r4>sun genotype.

Given that our data suggests it is the ability to augment sun mRNA levels, rather than absolute sun mRNA levels, that forms the basis of phenotypic plasticity, we measured sun mRNA levels in 1Y and 2Y in male larvae with fat body sun overexpression. We found no nutrient-dependent change in sun expression in these males (Figure 3—figure supplement 8C). Thus, the likely reason that sun overexpression does not enhance phenotypic plasticity in males was that the nutrient-dependent increase in sun mRNA levels was still absent in this overexpression context. To more clearly communicate these key points to the readers we have added text to this effect in the revised manuscript (subsection “A nutrient-dependent increase in stunted mRNA levels is required for enhanced IIS activity and a larger body size plasticity in females cultured in a protein-rich context”), and included a graphical abstract summarizing our model (Figure 7).

5) Again, concerning Sun regulation, what is the link between Tra activiy and sun expression in response to Y content in females?

We thank the reviewer for the opportunity to identify the link between Tra and sun mRNA levels. We added a significant amount of data in the revised manuscript to indicate that spargel (srl), the Drosophila homolog of PGC-1a, represents one link between Tra and the diet-dependent regulation of sun mRNA levels.

A previous study showed Srl mediates the nutrient-dependent upregulation of sun mRNA levels in a mixed-sex population of larvae (Delanoue et al., 2016). In our revised manuscript, we confirmed Srl function is required for the nutrient-dependent increase in sun mRNA levels in females raised on 2Y: the nutrient-dependent upregulation of sun mRNA was blocked in females heterozygous for the strong hypomorphic srl1 allele (Figure 5A).

Further, we show that heterozygous loss of srl blocks the nutrient-dependent upregulation of sun mRNA levels in Tra-expressing males (Figure 5D). Given that heterozygous loss of srl also blocks the nutrient-dependent upregulation of IIS activity (Figure 5E) and increased body size of Tra-expressing males raised in a protein-rich diet (Figure 5F), our data suggests that Srl represents one important link between Tra and nutrient-dependent changes to sun mRNA levels, IIS activity, and phenotypic plasticity. To ensure the reader is aware of this new data, we added several figures, text, and discussion (subsection “Transcriptional coactivator Spargel represents one link between Transformer and regulation of sun mRNA levels”) on this topic to the revised manuscript.

Can the authors relay the level of Sun in 1Y vs 2Y to a difference in TOR activity specifically in female FB cells?

TOR has been shown to play an important role in regulating Sun secretion, and we confirm in the revised version of our manuscript that there is a sex difference in circulating Sun (Figure 3—figure supplement 1A, B). To determine whether TOR plays a role in regulating Sun release, we measured fat body TOR activity by quantifying pS6k levels in males and females in 1Y and 2Y. We found that fat body pS6k levels did not differ between the sexes in 1Y or 2Y (Figure 5—figure supplement 1A-D). Similarly, we found no significant difference in pS6k levels between control and tra mutant females (Figure 5—figure supplement 1F, G).

Given that we detected no sex differences in fat body TOR activity, and that increased TOR does not alter sun mRNA levels (Figure 5—figure supplement 1E), our data supports a model in which the sex-specific and Tra-dependent regulation of sun occurs primarily through Srl rather than TOR. To clarify this model, we added data and text to the revised manuscript to ensure the reader understands our model of sex-specific body size plasticity (Figure 7; subsection “A nutrient-dependent increase in stunted mRNA levels is required for enhanced IIS activity and a larger body size plasticity in females cultured in a protein-rich context”). Further, we added text to discuss how this finding aligns with our previous finding that rapamycin feeding in larvae did not impact sexual size dimorphism (Rideout et al., 2015) (subsection “A nutrient-dependent increase in stunted mRNA levels is required for enhanced IIS activity and a larger body size plasticity in females cultured in a protein-rich context”).

Reviewer #3:

This very well written manuscript by Millington et al., uses the fruit fly Drosophila melanogaster to understand how sex-specific differences in endocrine signaling influences nutrient-dependent body size plasticity. Through a series of well-designed experiments, the authors demonstrate that females, but not males, exhibit increase growth when raised on a diet with twice the nutrient diet of standard fly food. Through a series of logical experiments, the authors reveal that growth difference results for sex-specific regulation of the insulin signaling pathway – female, but not male, flies express increased levels of the humeral factor stunted, which is known to promote dilp2 secretion from the IPCs. The authors further demonstrate that the manner by which female flies regulated stunted expression is required for nutrient-dependent body size plasticity. Finally, the authors demonstrate that the sex determination gene transformer is required for female-specific expression of stunted on the high nutrient diet.

Overall, I found this a very nice story that works its way from a simple observation to a molecular mechanism. The story will be of broad interest and highlights the importance of studying sex-specific differences in animal growth and development. I have a few suggested revisions, but overall enjoyed reading the manuscript.

We thank the reviewer for their positive assessment of our manuscript. We address each suggestion for improvement below.

1) The authors used animals that are heterozygous for mutations in the gene spargel as a substitute for analyzing stunted mutants. While I understand the necessity of this experiment, spargel mutants have a wide range of metabolic defects that are independent of stunted and I'm concerned that this experiment requires a leap of faith. At a minimum, I'd like to see verification that heterozygous spargel mutants exhibit significant changes in stunted gene expression.

Reviewer 3 makes very good points about verifying the effects of Srl on sun mRNA levels, and ensuring no broad metabolic defects exist in srl larvae. In our revised manuscript, we show that the nutrient-dependent increase in sun mRNA levels was blocked in female srl1/+ larvae raised a protein-rich diet (Figure 5A).

To determine whether heterozygous loss of Srl also impacts sun regulation in the context of Tra overexpression, we measured sun levels in Tra-expressing male larvae (da-GAL4>UAS-tra). Normally, these Tra-expressing males show nutrient-dependent upregulation of sun mRNA (Figure 4F); however, in our revised manuscript we show heterozygous loss of srl in these Tra-expressing males blocks their ability to augment sun mRNA in a protein-rich context (Figure 5D).

Importantly, the changes we observe are unlikely to be caused by generalized metabolic defects, as body size in srl1/+ larvae was not reduced compared with a control strain in the 1Y diet (Figure 5B, C). Together, the new data we present suggests that the ability of Tra to promote phenotypic plasticity depends on the regulation of sun mRNA levels by Srl, as loss of this regulation abolishes the ability of Tra-expressing males to upregulate sun, IIS activity, and body size in a protein-rich context (Figure 5D-F). Indeed, loss of sun in Tra-expressing males and females blocks phenotypic plasticity (Figure 4H, Figure 4—figure supplement 3H). To ensure readers are aware of all these important points, we include text and new data on this topic in the revised manuscript (Results).

https://doi.org/10.7554/eLife.58341.sa2

Article and author information

Author details

  1. Jason W Millington

    Department of Cellular and Physiological Sciences, Life Sciences Institute, The University of British Columbia, Vancouver, Canada
    Contribution
    Conceptualization, Formal analysis, Investigation, Visualization, Writing - original draft, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-4330-2431
  2. George P Brownrigg

    Department of Cellular and Physiological Sciences, Life Sciences Institute, The University of British Columbia, Vancouver, Canada
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared
  3. Charlotte Chao

    Department of Cellular and Physiological Sciences, Life Sciences Institute, The University of British Columbia, Vancouver, Canada
    Contribution
    Formal analysis, Investigation
    Competing interests
    No competing interests declared
  4. Ziwei Sun

    Department of Cellular and Physiological Sciences, Life Sciences Institute, The University of British Columbia, Vancouver, Canada
    Contribution
    Investigation
    Competing interests
    No competing interests declared
  5. Paige J Basner-Collins

    Department of Cellular and Physiological Sciences, Life Sciences Institute, The University of British Columbia, Vancouver, Canada
    Contribution
    Investigation
    Competing interests
    No competing interests declared
  6. Lianna W Wat

    Department of Cellular and Physiological Sciences, Life Sciences Institute, The University of British Columbia, Vancouver, Canada
    Contribution
    Investigation
    Competing interests
    No competing interests declared
  7. Bruno Hudry

    MRC London Institute of Medical Sciences, and Institute of Clinical Sciences, Faculty of Medicine, Imperial College London, London, United Kingdom
    Present address
    Institut de Biologie Valrose, Centre de Biochimie, Faculte des Sciences, Universite Nice Sophia Antipolis, Nice, France
    Contribution
    Resources
    Competing interests
    No competing interests declared
  8. Irene Miguel-Aliaga

    MRC London Institute of Medical Sciences, and Institute of Clinical Sciences, Faculty of Medicine, Imperial College London, London, United Kingdom
    Contribution
    Resources
    Competing interests
    No competing interests declared
  9. Elizabeth J Rideout

    Department of Cellular and Physiological Sciences, Life Sciences Institute, The University of British Columbia, Vancouver, Canada
    Contribution
    Conceptualization, Supervision, Funding acquisition, Validation, Writing - original draft, Project administration, Writing - review and editing
    For correspondence
    elizabeth.rideout@ubc.ca
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-0012-2828

Funding

Canadian Institutes of Health Research (PJT-153072)

  • Elizabeth J Rideout

Natural Sciences and Engineering Research Council of Canada (RGPIN-2016-04249)

  • Elizabeth J Rideout

Michael Smith Foundation for Health Research (16876)

  • Elizabeth J Rideout

Canada Foundation for Innovation (JELF-34879)

  • Elizabeth J Rideout

H2020 European Research Council (ERCAdG787470)

  • Irene Miguel-Aliaga

European Molecular Biology Organization (aALTF782-2015)

  • Bruno Hudry

University of British Columbia (CELL Fellowship)

  • Jason W Millington

University of British Columbia (British Columbia Graduate Scholarship Award)

  • Lianna W Wat

NSERC (Undergraduate Student Research Award)

  • Ziwei Sun

MRC Intramural funding

  • Irene Miguel-Aliaga

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

We thank Dr. William Ja for the UAS-sun strain (Delanoue et al., 2016), Dr. Pierre Léopold for the anti-Sun antibody (Delanoue et al., 2016), Dr. Suzanne Eaton and Dr. Natalie Dye for the anti-Cv-d antibody (Palm et al., 2012), Dr. Bruce Edgar for the GFP-PH reporter (tGPH), and Dr. Linda Partridge for sharing the dilp2 mutant strain. Stocks obtained from the Bloomington Drosophila Stock Center (NIH P40OD018537) were used in this study. We thank the TRiP at Harvard Medical School (NIH/NIGMS R01-GM084947) for providing transgenic RNAi fly stocks and/or plasmid vectors used in this study. Transgenic fly stocks and/or plasmids were also obtained from the Vienna Drosophila Resource Center (VDRC, http://www.vdrc.at). We acknowledge critical resources and information provided by FlyBase (Thurmond et al., 2019) FlyBase is supported by a grant from the National Human Genome Research Institute at the U.S. National Institutes of Health (U41 HG000739) and by the British Medical Research Council (MR/N030117/1). Funding for this study was provided by grants to EJR from the Canadian Institutes for Health Research (PJT-153072), Natural Sciences and Engineering Research Council of Canada (NSERC, RGPIN-2016–04249), Michael Smith Foundation for Health Research (16876), and the Canadian Foundation for Innovation (JELF-34879), and to IMA from the European Research Council (ERCAdG787470) and MRC Intramural Funding. JWM was supported by a 4 year CELL Fellowship from UBC, LWW was supported by a British Columbia Graduate Scholarship Award, ZS was supported by an NSERC Undergraduate Student Research Award, and BH was supported by an European Molecular Biology Organization Fellowship (aALTF782-2015). We would like to acknowledge that our research takes place on the traditional, ancestral, and unceded territory of the Musqueam people; a privilege for which we are grateful.

Senior Editor

  1. Utpal Banerjee, University of California, Los Angeles, United States

Reviewing Editor

  1. Jiwon Shim, Hanyang University, Republic of Korea

Publication history

  1. Received: April 28, 2020
  2. Accepted: January 11, 2021
  3. Accepted Manuscript published: January 15, 2021 (version 1)
  4. Version of Record published: February 5, 2021 (version 2)

Copyright

© 2021, Millington et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 1,161
    Page views
  • 180
    Downloads
  • 0
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

Further reading

    1. Developmental Biology
    Feng Wang et al.
    Research Article Updated

    The X-linked gene Rlim plays major roles in female mouse development and reproduction, where it is crucial for the maintenance of imprinted X chromosome inactivation in extraembryonic tissues of embryos. However, while females carrying a systemic Rlim knockout (KO) die around implantation, male Rlim KO mice appear healthy and are fertile. Here, we report an important role for Rlim in testis where it is highly expressed in post-meiotic round spermatids as well as in Sertoli cells. Systemic deletion of the Rlim gene results in lower numbers of mature sperm that contains excess cytoplasm, leading to decreased sperm motility and in vitro fertilization rates. Targeting the conditional Rlim cKO specifically to the spermatogenic cell lineage largely recapitulates this phenotype. These results reveal functions of Rlim in male reproduction specifically in round spermatids during spermiogenesis.

    1. Developmental Biology
    2. Neuroscience
    Hiroki Takechi et al.
    Research Article

    Transmembrane protein Golden goal (Gogo) interacts with atypical cadherin Flamingo to direct R8 photoreceptor axons in the Drosophila visual system. However, the precise mechanisms underlying Gogo regulation during columnar- and layer-specific R8 axon targeting are unknown. Our studies demonstrated that the insulin secreted from surface and cortex glia switches the phosphorylation status of Gogo, thereby regulating its two distinct functions. Non-phosphorylated Gogo mediates the initial recognition of the glial protrusion in the center of the medulla column, whereas phosphorylated Gogo suppresses radial filopodia extension by counteracting Flamingo to maintain a one axon to one column ratio. Later, Gogo expression ceases during the midpupal stage, thus allowing R8 filopodia to extend vertically into the M3 layer. These results demonstrate that the long- and short-range signaling between the glia and R8 axon growth cones regulates growth cone dynamics in a stepwise manner, and thus shape the entire organization of the visual system.