Fig 3. Loss of nowl in the nervous system causes sleep fragmentation.

(A-D) Quantification of average sleep-bout duration (A and B) and number (C and D) during day and night for males with pan-neuronal knockdown of nowl using the stronger (elav>) and weaker (nSyb>) GAL4 drivers and the nowl¹ insertional mutant. Neuronal knockdown of nowl using elav> (elav>nowl-RNA) and nSyb> (nSyb>nowl-RNAi) or mutation disrupting nowl (nowl¹/Y) result in decreased sleep-bout length and increased sleep-bout number, with a stronger effect during night-time, compared to controls (elav>+, nowl-RNAi/+, nSyb>/+, w¹¹¹⁸/Y, and +/Y, which does not carry any transgene or the w¹¹¹⁸ mutation and therefore is denoted by open circles). (E-F) Fraction of time male animals spent sleeping in short (5-49-min.), medium (50-149-min.) or long (150-720-min.) sleep bouts during night for controls (elav>+ and nowl-RNAi/+) compared to elav>nowl-RNAi animals and for nowl¹ (nowl¹/Y) mutants compared to controls (w¹¹¹⁸/Y and +/Y). Flies carrying a nowl mutation or in which nowl has been knocked down neuronally spend significantly less time in long consolidated sleep episodes, compared to controls. (G) Western-blot analysis shows increased levels of phospho-ERK (pERK) in nowl mutant males compared to controls (w¹¹¹⁸/Y), standardized to alpha-Tubulin. (H) Sleep pattern of Neurofibromin-1 (Nf1) knockdown males under a 12-hour light/dark (white and black bars) cycle in 30-minute intervals shows effects on sleep compared to controls (elav>+ and Nf1-RNAi/+) similar to animals lacking nowl, with decreased sleep mainly during the night. Graphs represent means with SEM (n = 32–92) of data pooled from one to three independent experiments. Significance was determined using Kruskal-Wallis test with Dunn's post-hoc testing (* p<0.05, ** p<0.01, *** p<0.001).