Lateral Hypothalamic (LH) orexin (OX/HT) neurons respond on approach to food objects, modulated by metabolic state.

(A) Orexin-Cre mice received LH-directed injections of DIO-GCaMP6f and implantation of a fiber optic probe. (B) Microscopy images showing orexin-Cre dependent GCaMP6f expression in the LH, together with immunolabelling of orexin-A positive cells, and a merged image. (C) Proportion of GCaMP6+ cells that are Orexin-A+ (97% ±0.9%, n = 4 mice), and Orexin-A+ cells that are GCaMP6+ (53.5% ±3.6%). (D) Experimental groups, with their pretreatments and test configurations. (E) The latency to approach the food pellet was significantly different between groups. Post-hoc comparisons revealed that each group had a significantly lower approach latency compared to Ad Libitum (p’s < 0.05). (F) Example z-scored dF/F photometry traces for the Ad Libitum and Food Restriction conditions. (G) Mean z-dF/F traces and shaded standard error of the mean (SEM) region for each condition, centered around the time of first approach to food. (H) The peak amplitude of the z-dF/F signal during the approach period varied significantly by treatment condition with post-hoc comparisons showing a significant difference between Ad Libitum and Food Restriction (p = 0.048). (I) Area under the curve (AUC) analysis of the approach period also significantly varied by treatment condition, with post-hoc comparisons showing significant differences between Ad Libitum and Food Restriction (p = 0.039), and Ad Libitum and Palatable conditions (p = 0.021). Error bars = +SEM; scale bars: 25μm; *p<0.05; **p<0.01.

The ventral pallidum and the nucleus accumbens lateral shell make monosynaptic inhibitory projections to lateral hypothalamic (LH) orexin (OX/HT) neurons.

(A) Cre-dependent ChR2 was expressed in the ventral pallidum (VP) of VGAT-Cre mice and TdTomato was expressed in LH OX/HT neurons. (B) Example image of TdTomato expressing OX/HT neurons (red), orexin neurons labelled with Alexa 488 (green), and the merged image. (C) Example traces of optically evoked inhibitory postsynaptic currents (IPSCs) onto LH OX/HT neurons during baseline, during TTX (1μM) application and then during TTX + 4-AP (100μM) application. (D) A comparison of monosynaptic and polysynaptic current amplitude (n/N = 12 cells/7 mice). Only one polysynaptic current was detected suggesting that VP projections to LH OX/HT neurons are primarily monosynaptic. (E) Latency from optical stimulation to oIPSC peak. (F) We tested if it was possible to evoke optical EPSCs from VP GABAergic terminals onto LH OX/HT neurons. Example traces of optically evoked EPSCs onto LH OX/HT neurons during the baseline, TTX, and TTX + 4-AP application. (G) A comparison of monosynaptic (n/N = 5 cells/4 mice) and polysynaptic current (n/N = 2 cells/2 mice) amplitude. (H) Latency from optical stimulation to oEPSC peak. (I) Cre-dependent ChR2 was expressed in the lateral shell of the nucleus accumbens of VGAT-Cre mice and TdTomato was expressed in LH OX/HT neurons. Example traces of optically evoked IPSCs onto LH OX/HT neurons during baseline, TTX, and TTX + 4-AP application. (J) A comparison of monosynaptic (n/N = 8 cells/5 mice) and polysynaptic current (n/N = 3 cells/3 mice) amplitude. (K) Latency from optical stimulation to oIPSC peak. The latency of polysynaptic currents was significantly longer compared to monosynaptic currents (p = 0.033). Error bars = +SEM