Animals. AgRP-IRES-Cre male mice (strain AgRPtm1(cre)Lowl/J, Jackson laboratories Stock number:012899) were crossed with ROSA-CAG-LSL-Synaptophysin-tdTomato-WPRE (Ai34D flox/flox) female mice (Strain B6;129S-Gt (RO-SA)26Sortm34.1 (CAG-Syp/tdTomato)Hze/J, Jackson Laboratories, Stock number: 012570) to obtain AgRP-IRES-Cre;Syn-TdTomato flox/wt mice used in our experiments. Animals were group housed (2–3 per cage) with a 12-h light/dark cycle at 22 ± 2°C and 50–70% humidity with ad libitum access to water and control STD (Ssniff # V1534-300; kCals metabolizable energy: 67% carbohydrates; 24% proteins and 9% fat; with 5.3% of total weight coming from crude sugar) unless otherwise stated. Mice were randomly assigned to either STD control or HFD for 48h (HFD introduced at start of the dark phase) (Ssniff #EF acc. D12492 (I) mod; kCals metabolizable energy: 22% carbohydrates, 24% protein and 60% fat; with 9.4% of total weight coming from crude sugar). All animal experiments and procedures were approved by the animal welfare committee (Landesamt für Arbeitsschutz, Verbraucherschutz und Gesundheit; animal ethics application number 2347-36-2021). Experiments were conducted in accordance with the ARRIVE guidelines and the European Directive 2010/63/EU.
Metabolic measurements. All measurements were performed during the dark phase at ZT 12. Body weight and food intake were measured at onset of the dark cycle on the first day of the experiment, 24h and 48h later before the sacrifice. Percentage body weight change was calculated as: body weight (g) at 48h x 100/body weight (g) at start of the experiment. Glycemia was assessed at sacrifice by blood collection directly from the right atrium and measured with a glucometer (Contour Care, Ascensia).
Immunostaining. Mice were deeply anesthetised with an intra-peritoneal injection of Pentobarbital (400mg/kg) and sacrificed by transcardial perfusion with, first, 60mL of ice cold 1X phosphate-buffered saline (PBS), then 4% paraformaldehyde (PFA, pH 9.5, 3.8% Borate) after onset of the dark phase ZT 12–15. Brains were post fixed in 4% PFA + Borax for 4h then switched to a 20% sucrose solution overnight before being stored at -80°C until usage. Fixed brains were sliced coronally using a freezing sledge microtome (Slide 4004 M, pfm medical, Cat. 400410) to obtain 30µm frozen brain slices that were stored in anti-freeze solution (10% 10X PBS, 20% Ethyl glycol and 20% Glycerol) at -20° C until further use. The day of the staining, slices were washed 3x10min times with 0.02M KPBS, 5min in 0.3% glycine in PBS and 10min 0.03% SDS in PBS followed by a 1h incubation in blocking and permeabilization buffer containing 3% donkey serum in 0.25% Triton-X in PBS. Slices were then incubated with primary antibodies Rabbit anti-AgRP (Phoenix; Cat# H-003-57; dilution 1:4,000) and Goat anti-TdTomato (SICGEN, Cat# AB8181-200; dilution 1:4,000) mixed in SignalStain® Antibody Diluent (CellsSignal; Cat# 8112) for 48h at 4°C plus 2h at room temperature. After washing with 0.02M KPBS (1x5min and 3x10min), slices were incubated with the secondary antibodies Donkey anti-Rabbit (1:500) Alexa Fluor 488-conjugated and Donkey anti-Goat (1:500) Alexa Fluor 546-conjugated diluted in 0.25% Triton-X in 1x KPBS for 1h at room temperature then washed with 0.02M KPBS (1x5min and 3x10min). Brain slices were mounted on Superfrost Plus microscope slides and cover slipped with VECTASHIELD Antifade Mounting Medium with DAPI (Biozol; Cat# VEC-H-1200). Images of the PVHant (distance from Bregma − 0.83mm) and PVHmid (distance from Bregma − 0.95mm), selected based on the mouse brain atlas [27], were collected using a confocal microscope (Multiphoton Laser Scanning Microscope LSM 780, Zeiss) and Zen (Zeiss) software. 3D Z-stack (z = 17 slices/image, interval = 1.5µm) images were acquired. Two mice were excluded from further image processing and analysis due to obvious germline deletion of the transgene STOP cassette resulting in ubiquitous expression of Synaptophysin-TdTomato protein.
Image Processing and Analysis. Images were processed and analyzed with a custom-written ImageJ macro and performed by an experimenter blinded to the experimental conditions. First, a region of interest (ROI) defining the outlines of the PVH was generated (Figs. 2 and 3a). This ROI was then used in the following two methods:
-
Maximum intensity projection: For each channel, Z-stack images were converted to a 2D image using a maximum intensity projection, and saved opened as individual files. A threshold value was automatically applied to these images using the “Moments” algorithm and a binary image created. The ROI was then overlayed on to the binary image. The workflow is represented in Fig. 4.
-
Sum slices: For each channel, a sum-slice intensity projection was created: this creates a 2D image by adding together the intensity of each pixel of each slice of the Z-stack at every x-y pixel coordinate in the 2D image space. The middle slice (number 9 out of 17) of the Z-stack was used as the reference slice and the sum-slice image, thresholded using “Moments” algorithm same this was applied to all 17 slices of the Z-stack image and a binary 2D image was saved. The same ROI used for the maximum intensity projection method was applied here. The workflow is represented in Fig. 5.
For both methods, the size of the area, Raw Integrated Density (= sum of the pixels), and mean gray value (= sum of the gray values of all pixels divided by the number of pixels) were measured. Threshold values of the PVH ROI, from both analysis methods, were compared for differences across sexes and diet. All ImageJ scripts are supplied in the Supplementary Methods. All ROIs and confocal images used for the maximum intensity projection and sum slices analyses can be shared upon request.
Statistical analysis. All data are represented as mean +/- standard deviation (SD) with each single data point representing an individual mouse. GraphPad Prism 9 was used for all statistical analysis and graphs. An unpaired two-tailed t-test was conducted to analyse the body weight, the percentage change of body weight and blood glucose levels. AgRP projections were analysed with a two-way ANOVA to assess the sex, diet and Interaction of sex and diet. When appropriate, the two-way ANOVA was followed by Bonferroni post-hoc comparison. Threshold values were analyzed using a Mann Whitney comparison. Significant differences were considered when p-value < 0.05.