Fly genetics and maintenance

w¹¹¹⁸ (iso31, BDSC#5905) was used as wt, 2xBRP, and background control. A brp null mutation (c04298, BDSC#85966) was utilized to reduce brp gene copy [33,40]. A genomic brp P[acman] construct [87], which was mapped to be integrated into the 5′ UTR of CG11357 (S3L and S3M Fig) [88], was used to increase brp gene copy number from 2xBRP to 3xBRP and 4xBRP [33] or to rescue the sleep phenotypes of 1xBRP flies (gBRP). A line with a P-element inserted at the 5′ UTR of CG11357 (EY12484, BDSC#20838) was acquired to mimic the effects of a P-element at CG11357 5′ UTR, which was largely comparable to wt (S3N–S3V Fig). Short sleep mutants wake (EY02219, BDSC#15858) and inc (f00285, BDSC#18307) and autophagy atg7 mutant (d06996, BDSC#19257) were described previously [56,57,89]. Alzheimer’s disease model flies were generated by pan-neuronal elav-Gal4-driven expression of Aβ-arctic (BDSC#458 and #33774). UAS-mCD8.GFP (BDSC#32188) was expressed by elav-Gal4 as control for Alzheimer’s disease model flies or by R23E10-Gal4 for patch-clamp electrophysiology. CaLexA (BDSC#66542) was expressed pan-neuronally by elav-Gal4 or in either dFB neurons marked by R23E10-Gal4 [53] or R5 neurons driven by R58H05-Gal4 [51]. All fly strains were backcrossed to w¹¹¹⁸ wt background for at least six generations to remove potential genetic modifiers.

Flies were raised under standard laboratory conditions on semidefined medium (Bloomington recipe) under 12/12h light/dark cycles with 65% humidity at 25°C. Unless specifically stated, female flies at certain ages were used except aversive olfactory memory experiments in which mixed populations of both sexes were used.