Drosophila FIT is a protein-specific satiety hormone essential for feeding control

Protein homeostasis is critical for health and lifespan of animals. However, the mechanisms for controlling protein feeding remain poorly understood. Here we report that in Drosophila, protein intake-induced feeding inhibition (PIFI) is specific to protein-containing food, and this effect is mediated by a fat body (FB) peptide named female-specific independent of transformer (FIT). Upon consumption of protein food, FIT expression is greatly elevated. Secreted FIT peptide in the fly haemolymph conveys this metabolic message to the brain, thereby promoting the release of Drosophila insulin-like peptide 2 (DILP2) and suppressing further protein intake. Interestingly, Fit is a sexually dimorphic gene, and consequently protein consumption-induced insulin release, as well as protein feeding behaviour, are also dimorphic between sexes. Thus, our findings reveal a protein-specific satiety hormone, providing important insights into the complex regulation of feeding decision, as well as the sexual dimorphism in feeding behaviour.


Supplementary Figure 2|Protein intake up-regulates fit expression level.
(a) The expression levels of Rp49 are comparable among starvation group (Agar) and three feeding groups. Dp and Actin 5C were used as the internal controls, respectively. n=3. 3 (b) Fit expression levels in WT flies greatly increased after 30 min feeding of tryptone, when analyzed with the Rp49, Dp, or Actin 5C as the internal control. n = 3.
(c) Fit expression levels significantly increased after the consumption of normal food or tryptone, but not after sucrose or lipid intake. n = 3.
(d) Fit expression levels increased after 30 min feeding of tryptone in a concentration-dependent manner. n = 3.
(e) BCAAs promote Fit expression, while non-BCAAs have little effect. Sucrose served as the sweetener, showing no effect on Fit expression on its own. n = 3.
(f) Food consumption was decreased when DMSO was added into the food, while it did not further decrease when Rapamycin was added. Sucrose-tryptone-mixed food was used. (a,,b) Compared to WT control flies, Fit KO flies showed normal basal feeding (a, n = 20-28), but exhibited increased food intake after starvation (b, n = 9-19).
(c) In the modified two-choice feeding assay, Fit KO and w 1118 flies showed similar preference for the tryptone-sucrose mix versus sucrose only. Left panel shows the Choice Ratio (CR) for two nutrients. Right panel shows the Preference Index calculated as CR S+T -CR S . n = 6. Data were analyzed by unpaired Student's t test in a, c and by one-way ANOVA, LSD's post hoc test in b. *, p < 0.05; ***, p < 0.001. n.s. indicates no statistical significance. Figure 5|Feeding behavior of two WT fly strains in the pre-feeding assay.

Supplementary
In pre-feeding assay (a), WT flies of wCS (b, c) and CS (d, e) displayed significant differences between sexes in tryptone pre-feeding groups, but little sexual difference in normal food or sucrose pre-feeding groups. n = 8-10.

Supplementary
(b) When tryptone-sucrose mixed food (T+S mixed food) was used instead of normal food in test feeding phase, WT flies also exhibited significant sexual differences in the tryptone pre-feeding groups. n = 8. Data were analyzed by one-way ANOVA, LSD's post hoc test in Food Consumption experiments. For Suppression Index analysis, P (Food*Sex) =1.31E-9 (two-way ANOVA, Bonferroni test).
(c) After pre-feeding of tryptone at different concentration, flies exhibited gradually decreased tryptone intake in the test feeding. n = 6. One-way ANOVA, LSD's post hoc test.
(e) The difference between overexpression groups and their parental controls showing as ΔCR for Tryptone = (CR Overexpression -CR control )/CR control . Overexpression of FIT resulted in a significant decease in the CR for tryptone, while overexpression of InR DN or FIT together with InR DN did not affect this choice ratio.

Supplementary Figure 10|Quantification of DILP2 immunostaining signals.
(a) Compared to that in agar (A) group, the overall mean of DILP2 signals were reduced in female flies after feeding of normal food (N) or tryptone (T), but not after sucrose (S) feeding, while it was not reduced in male flies after tryptone feeding. The total volume remained unchanged in all groups. n = 7-49. P (Food*Sex) = 0.0494 in Overall Mean Intensity of DILP2 and 0.0737 in Total Volume of DILP2 Signal (two-way ANOVA, Bonferroni test).
(b) The overall mean of DILP2 signals did not reduce after tryptone feeding in female flies with the secretion of IPCs blocked. The total volume remained unchanged. n = 12-16. P (Food*Genotype) = 3.49E-5 in Overall Mean Intensity of DILP2 and 0.8331 in Total Volume of DILP2 Signal (two-way ANOVA, Bonferroni test).
(c) The overall mean of DILP2 signals did not reduce after tryptone feeding in (d) The overall mean of DILP2 signals significantly reduced in brains incubated with FITbut not FITΔSP-conditioned medium. The total volume remained unchanged. n = 22-47.
One-way ANOVA, LSD's post hoc test.