Extended Data Fig. 6: ArcAgRP→MPOA neuromediator identification and behavioural effects of Npy knockdown and HCN blockade.
From: Integration of hunger and hormonal state gates infant-directed aggression
a, g, Recordings from MPOA neurons without (Pre) and with bath application of 100 µM AgRP (a) or 100 µM NPY (g). b–f and g–l, biophysical parameters of MPOA neurons at baseline (Pre) and with addition of AgRP or NPY: baseline firing frequency (b, n = 167 (Pre), 15 (AgRP) cells, N = 48 (Pre), 5 (AgRP) mice; h, n = 167 (Pre), 11 (NPY) cells, N = 48 (Pre), 4 (NPY) mice), percentage of silent neurons at resting membrane potential (c, n = 167 (Pre), 15 (AgRP) cells, N = 48 (Pre), 5 (AgRP) mice; i, n = 144 (Pre), 11 (NPY) cells, N = 48 (Pre), 4 (NPY) mice), resting membrane potential (d, n = 166 (Pre), 15 (AgRP) cells, N = 48 (Pre), 5 (AgRP) mice; j, n = 166 (Pre), 11 (NPY) cells, N = 48 (Pre), 4 (NPY) mice), input resistance (e, n = 142 (Pre), 15 (AgRP) cells, N = 46 (Pre), 5 (AgRP) mice; k, n = 142 (Pre), 11 (NPY) cells, N = 46 (Pre), 4 (NPY) mice), and voltage sag amplitude (f, n = 142 (Pre), 15 (AgRP) cells, N = 45 (Pre), 5 (AgRP) mice; l, n = 142 (Pre), 11 (NPY) cells, N = 45 (Pre), 4 (NPY) mice). m, Channelrhodopsin-assisted circuit mapping (CRACM) between ArcAgRP and MPOA neurons. Whole-cell recordings from MPOA neurons, and 450 nm widefield stimulation of ChR2+ ArcAgRP axons (see Methods). n, Example recording trace from MPOA neuron with sIPSCs. Note absence of light-evoked IPSCs in response to a train of 9 × 3-ms light pulses. o, Synaptic response pattern of MPOA neurons to acute ArcAgRP terminal activation (n = 11 cells, N = 3 mice). p, Representative example of action potentials in a ChR2-positive ArcAgRP neuron evoked by single 3-ms light pulses. q, Agrp knockdown (KD-A) in ArcAgRP→MPOA projections, and scrambled control (ctrl). r, Example images of Agrp KD and control. Scale bar, 20 µm. s, Agrp KD efficiency (n = 4 (KD-A), 4 (ctrl) mice). t, Voltage sag amplitude (n = 19 (Agg+), 26 (KD-A) cells, N = 6 (Agg+), 4 (KD-A) mice). u, Baseline firing (n = 20 (Agg+), 30 (KD-A) cells, N = 6 (Agg+), 4 (KD-A) mice). v, Percentage of silent neurons at resting membrane potential (n = 20 (Agg+), 31 (KD-A) cells, N = 6 (Agg+), 4 (KD-A) mice). w, Action potentials per injected current (n = 17 (Agg+), 16 (KD-A) cells, N = 6 (Agg+), 3 (KD-A) mice). x, Input resistance (n = 19 (Agg+), 26 (KD-A) cells, N = 6 (Agg+), 4 (KD-A) mice). y, Resting membrane potential (n = 20 (Agg+), 30 (KD-A) cells, N = 6 (Agg+), 4 (KD-A) mice). z, Cumulative incidence of aggression (n = 11 (KD-A), 13 (ctrl) mice). Shaded areas are confidence intervals. aa, Correlation between aggression latency and number of ArcAgRP→MPOA neurons transduced with KD-Y construct (linear regression, R2 = 0.450; P = 0.034; n = 10 mice). ab, 1-h food intake of animals with knockdown of Npy (KD-Y) or Agrp (KD-A) in ArcAgRP→MPOA projections, and control (n (left to right) = 8, 5, 13 mice). ac, Attack latency of animals with bilateral infusion of HCN blocker into MPOA (ZD, n = 4 mice) or after food deprivation (FD, n = 39 mice). ad, 1-h food intake of animals with bilateral infusion of ZD or vehicle (ctrl) into MPOA (n = 3 (ZD), 5 (ctrl) mice). Statistics: U test (two-sided) in b,d–f,h,j–l,t–u,x,y,ab,ac,ad. Fisher’s exact test (two-sided) in c,i,v. Unpaired t-test (two-sided) in s. Mixed linear model with cell ID as random effect in w. Log-rank test (one-sided) in z. Data are mean ± s.e.m. Box plots: median (line), interquartile range (box), whiskers, 1.5× IQR. Shaded areas in z represent 95% CI. *P < 0.05, **P < 0.01. See Supplementary Table 3 for further details of the statistical analyses.
