Fig. 1: SD or REMD reduces MD-induced spine elimination in V1.

a Repeated imaging of dendritic spines on apical tuft dendrites of layer 5 pyramidal neurons in MD and non-deprived control mice. White arrowheads indicate eliminated spines, asterisks indicate filopodia. Scale bar: 3 μm. b Dendritic spine elimination was significantly higher in MD mice than in control mice over 4–24 h (control: n = 7, 6 and 6 mice for 4, 8, and 24 h, respectively; MD: n = 8, 8 and 5 mice for 4, 8, and 24 h, respectively; P = 0.0003, = 0.0013 and = 0.0043 for 4, 8, and 24 h, respectively, Wilcoxon–Mann–Whitney test, two-sided). c No significant difference in dendritic spine formation between MD and control mice (control: n = 7, 6 and 6 mice for 4 h, 8 h and 24 h, respectively; MD: n = 8, 8 and 5 mice for 4, 8, and 24 h, respectively; P = 0.817, = 1.000, and = 0.361 for 4, 8, and 24 h, respectively, Wilcoxon–Mann–Whitney test, two-sided). d Schematic of experimental design to assess the effect of sleep on MD-induced spine remodeling. e No significant difference in spine elimination (P = 0.487) or formation (P = 0.908) was observed in mice with or without sleep during the 0–4 h MD period (Sleep: n = 8 mice, SD: n = 7 mice, Wilcoxon–Mann–Whitney test, two-sided). f Schematic of experimental design. After the initial 4-h MD, mice were subjected to an additional 4-h MD with or without sleep/REM sleep to assess the effect of sleep. g SD or REMD during the second 4 h significantly reduced MD-induced dendritic spine elimination but not spine formation as compared to that in sleep or NREM-d mice, respectively (SD: n = 6 mice, Sleep: n = 8 mice, REMD: n = 5 mice, NREMD-d: n = 5 mice, P = 0.0005, Kruskal–Wallis test followed by multiple comparisons test). **P < 0.01, ***P < 0.001. Data are presented as mean ± s.e.m. Source data for (b), (c), (e), and (g) are provided as a Source Data file.