A Drosophila insulin-like peptide promotes growth during nonfeeding states

Abstract

In metazoans, tissue growth relies on the availability of nutrients--stored internally or obtained from the environment--and the resulting activation of insulin/IGF signaling (IIS). In Drosophila, growth is mediated by seven Drosophila insulin-like peptides (Dilps), acting through a canonical IIS pathway. During the larval period, animals feed and Dilps produced by the brain couple nutrient uptake with systemic growth. We show here that, during metamorphosis, when feeding stops, a specific DILP (Dilp6) is produced by the fat body and relays the growth signal. Expression of DILP6 during pupal development is controlled by the steroid hormone ecdysone. Remarkably, DILP6 expression is also induced upon starvation, and both its developmental and environmental expression require the Drosophila FoxO transcription factor. This study reveals a specific class of ILPs induced upon metabolic stress that promotes growth in conditions of nutritional deprivation or following developmentally induced cessation of feeding.

Figure 1

An Expression Time Course of DILP Genes during Development (A) At pupal stage, organismal growth requires insulin signaling. Temperature shift-up experiment (18°C–29°C) was carried out at the onset of the pupal stage (i.e., from 120 hr after egg deposition [AED] until adult eclosion) with tubGal80^ts; act > dInR RNAi. tubGal80^ts; act > w under the same temperature-shift program and tubGal80^ts; act > dInR RNAi grown at restrictive temperature (18°C, normalized to W) were used as controls. Graph shows adult mass of animals in which InR was silenced during pupal stage compared to controls. Means ± SD are presented (n ≥ 50; ^∗∗p < 0.01). (B) Expression of DILP genes during development. For each profile, fold changes are calculated relative to the minimal level. No cross-quantification is provided between the different DILP genes. Error bars represent SD.

Figure 2

DILP6 Is Required for Growth, but not for Carbohydrate and Lipid Metabolism (A) Schematic drawing of the genomic region around the DILP6 gene. Two DILP6 deletion alleles were generated by P element KG004792 jump. Deletion allele DILP6^#41 removes the first exon and part of the first intron of the dilp6 gene. Deletion allele DILP6^#68 removes DILP6 and four adjacent nonannotated genes: CG33218, CG2854, CG14050, and CG34052. (B) DILP6 loss of function were analyzed using either the two DILP6 mutant alleles, DILP6^#41 and DILP6^#68, at 25°C, or by silencing of DILP6 ubiquitously (act >), in the fat body (cg >) and in the gut (mia >) at 29°C. DILP6 overexpression in fat body was performed at 29°C. Graph represents means ± SD (n ≥ 50; ^∗∗p < 0.01). (C) Relative expression of DILP genes assessed by qRT-PCR analysis in the fat body compared to total larva at wandering stage (110 hr AED). Graph represents means ± SD (n = 3). (D) Circulating carbohydrate levels in hemolymph of DILP6 mutants compared to wild-type animals. Graph represents means ± SD (n ≥ 8). (E) TAG content of DILP6 mutant compared to wild-type larvae. Graph represents means ± SD (n ≥ 6).

Figure 3

DILP6 Activates Growth from Postfeeding Larval Development until Adult Emergence (A–C) Changes in DILP6 expression affect animal growth during late larval and pupal stages, but not during early larval development. tub-Gal80^ts, act > DILP6 or tub-Gal80^ts, act >, DILP6-RNAi flies were kept at restrictive temperature (29°C) until 96 hr AED in order to silence or overexpress DILP6 during early larval development. A shift-down to permissive temperature (18°C) at 96 hr AED has no effect on adult mass (A). When the temperature shift occurs at the larval-pupal transition (120 hr AED), changing DILP6 expression induces significant effects on animal mass (B). Temperature shift-up from 120 hr AED until adult emergence affects animal mass (C). Adult mass is compared to tub-Gal80^ts; act > w as control exposed to identical temperature-shift programs. Graph represents means ± SD (n ≥ 100; ^∗∗p < 0.01). (D) Measurement of starvation resistance of newly emerged flies where DILP6 was overexpressed or silenced from 120 hr AED until adult emergence. Starvation was performed at 18°C to inhibit Gal4 activity (error bars represent SD; n ≥ 84). (E and F) Measurement of TAG (E) and glycogen (F) contents of newly emerged flies where DILP6 was overexpressed or silenced from 120 hr AED until adult emergence. Graph represents means ± SD (n ≥ 3); ^∗p < 0.05, ^∗∗p < 0.01.

Figure 4

DILP6 Expression Is Developmentally Controlled by Ecdysone (A) Overexpression of TSC1/2 in the ring gland (P0206 > TSC1/2) delays the ecdysone peak: E74B, a direct target of ecdysone, is expressed with 24–30 hr delay compared with control larvae (P0206 > w) (see Layalle et al., 2008 for details). (B) Under the same experimental conditions, expression of DILP6 is also delayed. mRNA levels are relative to control at 110 hr AED; error bars represent SD. (C) Measurement of DILP6 expression by qRT-PCR in early (72 hr AED), mid (96 hr AED), and late (110 hr AED) third instar larva and in pupa. Control and EcR silencing in fat body cell (cg > EcR RNAi) conditions are shown. Fold changes are relative to control at 110 hr AED; error bars represent SD. (D) Measurement by qRT-PCR of E74B and DILP6 expressions in dissected fat bodies incubated with 0.2% ethanol (+EtOH) or 2 μM 20E in 0.2% ethanol (+20E). Fold changes are relative to ethanol-treated control; error bars represent SD; ^∗p < 0.05, ^∗∗p < 0.01.

Figure 5

DILP6 Regulates Growth during Starvation (A) Measurement of DILP6 expression by qRT-PCR in total larva or dissected organs from animals reared either in fed or starved conditions. Fold changes are relative to fed conditions; error bars represent SD. (B) Quantification of DILP2, DILP3, DILP5, and DILP6 expression relative to RP49 in fed and starved larvae. Fold changes are relative to DILP3 expression in fed conditions. Larvae were starved at 72 hr AED for 16 hr on PBS 1% sucrose; error bars represent SD. (C) Measurement of adult mass of control animals and DILP6 mutants (DILP6^#41 and DILP6^#68) under fed and starved conditions. DILP6 mutants display an aggravated loss of mass upon starvation compared with controls. Graph represents means ± SD (n ≥ 80; ^∗∗p < 0.01).

Figure 6

dFoxO Controls DILP6 Expression upon Starvation (A) Immunostainings showing dFoxO protein (green) accumulating in the nuclei of fat body cells at 110 hr AED compared to 72 hr AED (top panels). At 72 hr AED, starvation also provokes an increase in dFoxO accumulation in fat body cell nuclei (bottom panels). (B) Measurement of DILP6 expression by qRT-PCR in fed and starved conditions of wild-type, dFoxO mutant, and fat body-specific dFoxO knockdown larvae. Fold changes are relative to fed controls; error bars represent SD. (C) Chromatin immunoprecipitation (ChIP) performed on starved animals using either anti-dFoxO antibody or preimmune serum as a mock ChIP. Using two set of primers (gray and black arrowheads), qPCR shows that dFoxO binds to the DILP6 promoter. Primers for the 4EBP promoter and for a nonrelated genomic area were used as positive and negative controls, respectively. (D) Developmental DILP6 expression in dFoxO mutant animals. Although basal levels are reduced, DILP6 expression is still induced at the larval/pupal transition. Fold changes are relative to control at 110 hr AED; error bars represent SD. (E) Measurement of DILP6 expression by qRT-PCR in dissected fat bodies from control and cg > dFoxO RNAi animals incubated with 2 μM 20E (+20E) compared to controls treated with 0.2% ethanol (+EtOH). Fold changes are relative to ethanol-treated control; error bars represent SD.