Extended Data Fig. 4: Circadian variations of synaptic input contribute to AgRP-neuron activity rhythms.
a. Schematic of ex-vivo recording from GFP labeled ARCNPY neurons in Npy-gfp mice and culling times. b,c. Representative loose-seal traces (b, scale: 1 s) and average ARCNPY neuron activity (c) overlaid with average in vivo AgRP:GCaMP7s trace. n = 39-82 neurons for control, and n = 8-19 neurons for data with synaptic blockers. d,e. Representative whole-cell recordings from ARCNPY neurons showing synaptic events (d) and average (e) spontaneous IPSC and EPSC frequency across day overlaid with average in vivo AgRP:GCaMP7s activity. n = 8-17 neurons for sEPSC, and n = 6-22 neurons for sIPSC, two-tailed, unpaired t-test, *p < 0.05, **p < 0.01. Scale: 2 s, 20 pA (sEPSC), 50 pA (sIPSC) f. Schematic of in vivo fiber photometry recording of glutamate dynamics on AgRP-neurons and representative image showing AgRP cells expressing iGluSnFR, and the placement of the fiber in the ARC (Scale 100 μm). g-i. Representative AgRP- iGluSnFR raw trace (g) and averaged signal (h) recorded during access to food in fasted mice and quantification of changes in iGluSnFR signal in the first minutes of food access (two-tailed paired t-test, p = 0.008, n = 3 mice, i). j-l. Summary of changes in glutamate levels on AgRP-neurons over 24 h in AL (average of 3 animals, 3 days, j) and fasted (k) animals, and overlay of two rhythms in averaged in 4 h bins (l, two-tailed paired t-test, *p < 0.05, **p < 0.01). m. Comparison of pre-dark onset (ZT8-ZT12) and dark onset (ZT12-ZT16) levels of glutamate on AgRP-neurons in AL and fasted animals (two-tailed paired t-test, **p = 0.0024, *p = 0.013, n = 3 mice). n-u. Same as f-m, except iGABASnFR was examined (**p = 0.0025, q; **p = 0.0076, *p < 0.05, t; *p = 0.029, u, n = 4 ad lib, n = 5 fasted). Data are presented as mean ± SEM. For exact P-values, number of subjects/neurons, and statistics, see Extended Data Fig. 4 Source Data.