Figure 3. ndi1 expression in the intestine stimulates feeding behavior. (A) Analysis of total food consumption using a capillary feeding (CAFE) assay. Flies expressing ndi1 in ISCs/EBs (esgGAL4>ndi1) consume significantly more food relative to controls (esgGAL4>+). (*p<0.05, **p<0.01, t test, 10 replicates per condition, 10 flies per replicate). (B) Analysis of feeding proportion using a colorimetric dye-tracking assay. Flies that express ndi1 in ISCs/EBs (esgGAL4>ndi1) had a significantly greater proportion of flies that fed during the assay period relative to controls (esgGAL4>+). (***p<0.001, binomial test, approximately 90 flies per condition). (C) Analysis of meal size using a colorimetric dye-tracking assay (constitutive ndi1 expression). Of those flies that ate during the assay period in (B), meal size was significantly greater in flies that express ndi1 in ISCs/EBs (esgGAL4>ndi1) relative to controls (esgGAL4>+). (*p<0.05, ***p<0.001, t test, 50-95 flies that ate from B per condition). (D) Analysis of total food consumption using a CAFE assay in 5961GS>ndi1 flies with or without RU486-mediated transgene induction. Ten days of induced ndi1 expression in ISCs/EBs of adults by exposure to RU486 (0.5mg/l) increases total food consumption relative to uninduced controls. (*p<0.05, t test, 6 replicates per condition, 10 flies per replicate). (E) Analysis of meal size using a colorimetric dye-tracking assay in 5961GS>ndi1 flies with or without RU486-mediated transgene induction. Ten days of induced ndi1 expression in ISCs/EBs of adults by exposure to RU486 (0.5mg/l) increases meal size relative to uninduced controls. (*p<0.05, t test, approximately 85 flies that ate during the assay period in Figure S3A). (F) Analysis of intestinal function by assaying excreta shape. The proportion of oblong deposits (RODs, Figure S3E, arrow) in excreta is significantly lower in deposits from flies expressing ndi1 in ISCs/EBs (esgGAL4>ndi1) relative to controls (esgGAL4>+). (*p<0.05, ***p<0.001, binomial test, at least 180 deposits per condition). (G) Analysis of intestinal function by excreta pH. Flies expressing ndi1 in ISCs/EBs (“ndi1”, esgGAL4>ndi1) had more alkaline excreta relative to controls (“+”, esgGAL4>+) at day 10 of adulthood in colorimetric analyses of excreta pH after feeding on BPB medium. (**p<0.01, ***p<0.001, t test, at least 180 deposits per condition). (H) Weights of flies with constitutive ndi1 expression in ISCs/EBs as a function of age. Flies that express ndi1 in ISCs/EBs (“ndi1”, esgGAL4>ndi1) are heavierthan isogenic controls (“+”, esgGAL4>+) and maintain their weight through day 30 of adulthood. (**p<0.01, **p<0.01, ***p<0.001, t test, 12 replicates per condition, 5 flies per replicate). (I) Weights of 5961GS>ndi1 flies with or without RU486-mediated transgene induction. Ten days of adulthood only induction of ndi1 in ISCs/EBs by exposure to RU486 (0.5mg/l) is sufficient to significantly increase weight relative to uninduced controls. (**p<0.01, t test, 6 replicates per condition, 10 flies per replicate). (J) Protein content as a function of age. Protein content is significantly decreased in control flies (“+”, esgGAL4>+) at 30 days of adulthood, but is maintained in flies expressing ndi1 in ISCs/EBs (“ndi1”, esgGAL4>ndi1). (*p<0.05, **p<0.01, t test, 4 replicates per condition, 5 flies per replicate). (K) Triacylglyceride content as a function of age. Thin-layer chromatography (Figure S3G) and densitometry for triacylglyceride content show a significant decrease in control flies (“+”, esgGAL4>+) at 30 days of age whereas flies that express ndi1 in ISCs/EBs (“ndi1”, esgGAL4>ndi1) maintain triacylglyceride stores with age. (**p<0.01, t test, 5 replicates per condition, 5 flies per replicate).