Skip to main content
. 2023 Mar 17;14:1486. doi: 10.1038/s41467-023-37044-4

Fig. 1. LHLepR neurons are the food-specific subpopulation of LHGABA neurons.

Fig. 1

a, b Schematic of micro-endoscopic calcium imaging (left, middle), and image of GCaMP6s expression (right) in the LH from Vgat-cre (a) and LepR-cre mice (b). The experiment was repeated 6 times (a) or 4 times (b) independently with similar results. fx, fornix; 3 V, the 3rd ventricle. c, d Spatial map of raw data (left), accepted cells using CNMFe (middle), and cells that only respond to food-related behaviour (right) from LHGABA neurons (c) and LHLepR neurons (d). Cells are coloured according to the maximum Z-score. Scale bar: 50 μm. e, f Schematic of the multi-phase test 2, consummatory behaviour test 1, consummatory behaviour test 2 (food and non-food) (top). Heatmap depicting calcium signals aligned to the onset of feeding behaviours (running to food, rearing to food, contact with food, contact with edible object) (below). Four populations are discriminated: food-specific responsive (yellow), non-specific responsive (grey), non-food-specific responsive (blue), and non-responsive (white) cells. (LHGABA neurons 218 cells, 6 mice (e), LHLepR neurons 48 cells, 4 mice (f)). g, m Representative traces of four populations from LHGABA neurons (g) and LHLepR neurons (m). The dotted line separates each behavioural experiment. h, k Venn diagram of food responsive and non-food responsive neurons. Percentage of food-responsive neurons are as follows (LHGABA neurons 8% (18/218 cells) (h), LHLepR neurons 63% (30/48 cells) (k)). i, l Proportion of food-specific responsive (yellow), non-specific responsive (grey), non-food-specific responsive (blue), and non-responsive (white) cells from LHGABA neurons (i) and LHLepR neurons (l). j Venn diagram simulating the number of LHLepR positive (yellow) and food-specific (grey) neurons when the total number of LH GABA neurons is simulated as 1000. Source data are provided as a Source data file. The schematics in a, b, e and f were created using BioRender.