The Sex Determination Gene transformer Regulates Male-Female Differences in Drosophila Body Size

Abstract

Almost all animals show sex differences in body size. For example, in Drosophila, females are larger than males. Although Drosophila is widely used as a model to study growth, the mechanisms underlying this male-female difference in size remain unclear. Here, we describe a novel role for the sex determination gene transformer (tra) in promoting female body growth. Normally, Tra is expressed only in females. We find that loss of Tra in female larvae decreases body size, while ectopic Tra expression in males increases body size. Although we find that Tra exerts autonomous effects on cell size, we also discovered that Tra expression in the fat body augments female body size in a non cell-autonomous manner. These effects of Tra do not require its only known targets doublesex and fruitless. Instead, Tra expression in the female fat body promotes growth by stimulating the secretion of insulin-like peptides from insulin producing cells in the brain. Our data suggest a model of sex-specific growth in which body size is regulated by a previously unrecognized branch of the sex determination pathway, and identify Tra as a novel link between sex and the conserved insulin signaling pathway.

Fig 1. Sex determination gene transformer regulates sex differences in body size.

(A) Photograph of w ¹¹¹⁸ male and female pupae, approximately 48 hr after puparium formation. (B) Quantification of pupal volume in w ¹¹¹⁸ male and female pupae (n>60). Females are approximately 30% larger in size than males (p = 1.8 x 10⁻⁴³, Student’s t-test). (C) Photograph of control male and female pupae (tra ¹ /+) compared to a tra mutant female (tra ¹/Df(3L)st-j7). (D) Pupal volume in tra mutant females is significantly reduced compared to control females (p = 7.1 x 10⁻⁷;2.3 x 10⁻⁶, one-way ANOVA followed by Tukey HSD post-hoc test). Male pupal volume is unaffected (p = 0.91;0.87, one-way ANOVA followed by Tukey HSD post-hoc test). n>60 for all genotypes. (E) Photograph of male and female control pupae (da>+) compared to males with ubiquitous overexpression of Tra (da>tra). (F) da>tra males and females are significantly larger than control animals (p = 0;0 and 0;3.6 x 10⁻⁵, one-way ANOVA followed by Tukey HSD post-hoc test). (A, C, E) Scale bars = 1 mm. * indicates a significant difference from control genotypes. N.S. means not significantly different from both control genotypes. The p-values indicated are listed in the following order: difference between the GAL4/UAS genotype and the GAL4 control (or first genotype control);difference between the GAL4/UAS genotype and the UAS control (or second genotype control). A list of all p-values obtained from the Tukey HSD post-hoc test is provided in S1 Table.

Fig 2. tra regulates growth in a cell-autonomous, and a non cell-autonomous manner.

(A) Fat body cell size is significantly larger in wild-type females (p = 7.6 x 10⁻⁷; Student’s t-test). (B) Mosaic expression of a tra2-RNAi transgene in the fat body causes a significant reduction in cell size in females (p = 0.024; Student’s t-test). Male cell size is unaffected by expression of the tra2-RNAi transgene (p = 0.063; Student’s t-test). (C) Mosaic overexpression of Tra in the fat body stimulates cell growth to increase cell size in both females and males (p = 6.8 x 10⁻¹⁰ and 5.6 x 10⁻⁹, respectively; Student’s t-test). (D) Expression of a tra2-RNAi transgene in the posterior compartment of the wing using en-GAL4 caused a significant reduction in the ratio of posterior:anterior (p = 1.5 x 10⁻⁶; Student’s t-test). Male posterior:anterior ratio was unaffected (p = 0.062; Student’s t-test). (E) The number of wing hairs in a fixed area in females expressing the tra2-RNAi transgene in the posterior compartment of the wing is greater than in control females (p = 0.0034; Student’s t-test). (F) The total number of cells in the posterior compartment of the wing is not significantly different in females upon expression of the tra2-RNAi transgene compared to wild-type females (p = 0.063; Student’s t-test). (G) Tissue-specific expression of the tra2-RNAi transgene in the fat body significantly decreased pupal volume (p = 0;0.0006 (r4), 0.008;0.001 (cg), one-way ANOVA followed by Tukey HSD post-hoc test). (H) Ubiquitous, or fat body-specific, overexpression of UAS-tra rescues the decreased body size of tra mutant females (one-way ANOVA followed by Tukey HSD post-hoc test, see S1 Table for full set of statistical comparisons between genotypes). * indicates a significant difference from control genotypes. N.S. means not significantly different from both control genotypes. The p-values indicated are listed in the following order: difference between the GAL4/UAS genotype and the GAL4 control;difference between the GAL4/UAS genotype and the UAS control.

Fig 3. Sex determination genes fru and dsx do not affect body size.

(A) Pupal volume in dsx mutant males and females (dsx ¹ /Df(3R)dsx ¹⁵) is not significantly different from control males and females (p = 0.92;1 and 0.99;0.99, respectively, one-way ANOVA followed by Tukey HSD post-hoc test). (B) Mosaic expression of dsx-RNAi in fat cells causes a significant reduction in cell size in both females (p = 1.1 x 10⁻²⁴, Student’s t-test) and males (p = 2.8 x 10⁻¹⁰, Student’s t-test). (C) Expression of dsx-RNAi in the fat body does not alter body size in females or males (p = 0.95;0.97 and 0.24;0.28, one-way ANOVA followed by Tukey HSD post-hoc test). (D) Females lacking the Tra-binding site in fru P1 transcripts (fru ^Δtra) ectopically express Fru^M in the central nervous system (CNS). Pupal volume in fru ^Δtra /Df(3R)fru ^4-40 females was not significantly different than in control females (p = 0.46;0.99, one-way ANOVA followed by Tukey HSD post-hoc test). Male body size was also unaffected (p = 0.99;0.99, one-way ANOVA followed by Tukey HSD post-hoc test). (E) Males with the fru ^F allele of fru have constitutively female-specific splicing of fru P1 transcripts, and thus produce no Fru^M in the CNS. Pupal volume in fru ^F /Df(3R)fru ^4-40 males was not significantly different than control males (p = 1;0.63, one-way ANOVA followed by Tukey HSD post-hoc test). Female pupal volume was unaffected (p = 1;0.99, one-way ANOVA followed by Tukey HSD post-hoc test). (F) Ubiquitous overexpression of Tra increases body size in males and females even in a dsx mutant background. Thus, no significant difference in body size is present between females overexpressing Tra in a wild-type, or dsx mutant background (p = 0.436, one-way ANOVA followed by Tukey HSD post-hoc test). Similar results were observed in males (p = 0.39, one-way ANOVA followed by Tukey HSD post-hoc test). * indicates a significant difference from control genotypes. N.S. means not significantly different from both control genotypes. The p-values indicated are listed in the following order: difference between the GAL4/UAS genotype and the GAL4 control;difference between the GAL4/UAS genotype and the UAS control. A list of all p-values obtained from the Tukey HSD post-hoc test is provided in S1 Table.

Fig 4. tra function in the fat body regulates male-female differences in dILP2 secretion.

(A) Females are normally 30% larger than males in nutrient-rich (high IIS/TOR activity); however, in nutrient-poor food, sex differences in body size are abolished (p = 0.082, one-way ANOVA followed by Tukey HSD post-hoc test). (B) Male and female transcript levels of insulin-like peptides 2, 3 and 5 (dilp2, dilp3, dilp5) were analyzed in larval carcasses devoid of fat body at 120 hr AEL at 25°C. Levels of dilp2 and dilp5 were not different between males and females (p = 0.051, 0.19, Student’s t-test), but levels of dilp3 were significantly higher in males (p = 0.0004; Student’s t-test). (C) Males (C’) have significantly higher dILP2 retention in the insulin producing cells (IPCs) in the brain than females (C) (p = 0.001). (D-D”,E) Levels of dILP2 retention in females expressing a tra2-RNAi transgene specifically in the fat body (r4-GAL4) are increased compared to control females (p = 0.001;0.025, Student’s t-test). (F) Loss of tra significantly reduces body size in females. However, body size in tra mutant animals that also have loss of one copy of PTEN, a repressor of IIS, rescues the reduced size of tra mutant females, but has no effect on tra mutant males (see S1 Table for complete list of p-values from one-way ANOVA followed by Tukey HSD post-hoc test). (G) Ubiquitous expression of UAS-tra significantly increases body size in males and females (p = 0;0 and 0;0, one-way ANOVA followed by Tukey HSD post-hoc test). However, in animals heterozygous for loss of one copy of the insulin receptor gene (InR ^PZ, InR ^E19), the increased growth driven by UAS-tra overexpression is blocked (see S1 Table for complete list of p-values from one-way ANOVA followed by Tukey HSD post-hoc test). * indicates a significant difference from control genotypes. N.S. means not significantly different from both control genotypes. The p-values indicated are listed in the following order: difference between the GAL4/UAS genotype and the GAL4 control;difference between the GAL4/UAS genotype and the UAS control.