Dopamine-inhibited POMCDrd2+ neurons in the ARC acutely regulate feeding and body temperature

Dopamine acts on neurons in the arcuate nucleus (ARC) of the hypothalamus, which controls homeostatic feeding responses. Here we demonstrate a differential enrichment of dopamine receptor 1 (Drd1) expression in food intake–promoting agouti related peptide (AgRP)/neuropeptide Y (NPY) neurons and a large proportion of Drd2-expressing anorexigenic proopiomelanocortin (POMC) neurons. Owing to the nature of these receptors, this translates into a predominant activation of AgRP/NPY neurons upon dopamine stimulation and a larger proportion of dopamine-inhibited POMC neurons. Employing intersectional targeting of Drd2-expressing POMC neurons, we reveal that dopamine-mediated POMC neuron inhibition is Drd2 dependent and that POMCDrd2+ neurons exhibit differential expression of neuropeptide signaling mediators compared with the global POMC neuron population, which manifests in enhanced somatostatin responsiveness of POMCDrd2+ neurons. Selective chemogenetic activation of POMCDrd2+ neurons uncovered their ability to acutely suppress feeding and to preserve body temperature in fasted mice. Collectively, the present study provides the molecular and functional characterization of POMCDrd2+ neurons and aids our understanding of dopamine-dependent control of homeostatic energy-regulatory neurocircuits.


Arcuate Nucleus micropunches
To obtain Arcuate Nucleus micropunch biopsies, 8 male WT C57BL/6N mice (Charles River, France) were decapited and brains quickly isolated. Fresh coronal brain sections of 300 μm thickness were cut in a stainless steel brain matrix (World Precision Instruments) and immediately frozen on dry ice. The mediobasal Arcuate Nucleus of the hypothalamus was consequently micro dissected from these frozen sections using a 1 mm diameter circular biopsy punch (Rainer Medizintechnik, Cat# KAI 101), samples snapfrozen in liquid nitrogen and stored at -80°C until RNA was isolated. hours at 4°C. Samples were then incubated in 20 % sucrose in PBS for 12-24 hours at 4°C, frozen and stored at -80°C until sectioning.

Brain sectioning
PFA-fixed, frozen brain tissues were cut in a cryostat (Leica, CM3050S) at a chamber temperature of -20°C. For immunohistochemical analyses sections were cut at a thickness of 30 μm and mounted onto polysine-coated glass slides (ThermoFisher, Cat# J2800AMNZ). For RNA in situ hybridization sections were cut at a thickness of 20 μm and mounted onto SuperFrost Plus Gold slides (ThermoFisher, Cat#11976299).
Slides were stored at -80°C until further processing.

Microscopy and image processing
Microscopic images of RNA in situ hybridization for Sst/Th overlap were obtained with an Olympus SLIDEVIEW VS200 digital slide scanner at a 20x magnification (objective: UPLXAPO 20x/0.8; software: VS200 ASW), all other microscopic images were acquired using a confocal Leica TCS SP-8-X microscope at 40x magnification and zstack size of 0.9 μm (objective: 40x/1.30oil, software: LASX V.3.5.7.23225). Laser intensities were kept constant throughout all related conditions and adjustments in brightness and contrast of all channels, as depicted in the representative microscopic images, were applied equally throughout all related conditions.

Immunofluorescent stainings
For immunofluorescent stainings 14-16 week old male and female POMC Dre Drd2 Cre R26-lx-rx-ZsGreen mice and 11-14 week old male and female POMC Dre Drd2 Cre R26lx-rx-EGFP-L10a mice were perfused and brain sections mounted as described above ('Transcardial perfusion' and 'Brain sectioning'). For immunohistochemical assays, incubated with the respective secondary antibody for 60 minutes in the dark at room temperature. All utilized secondary antibodies are listed in the 'Resources Table' and were applied to samples at a dilution of 1:500. Samples were washed as before in washing buffer and mounted using Vectashield Antifade Mounting Medium with DAPI (Vector Laboratories). Co-stainings that utilized two antibodies from the same host (ZsGreen and POMC in brain sections of POMC Dre Drd2 Cre R26-lx-rx-ZsGreen mice) were stained for each target protein on discrete days, separated by an additional 10 min fixation in 4 % PFA in PBS, PBS wash and 60 min incubation in blocking solution at room temperature after completed staining for first target protein including its secondary antibody. Specifically, in POMC Dre Drd2 Cre R26-lx-rx-ZsGreen mice staining against ZsGreen was performed one day prior to staining against POMC. Microscopic images of immunofluorescent stainings in POMC Dre Drd2 Cre R26-lx-rx-ZsGreen and POMC Dre Drd2 Cre R26-lx-rx-EGFP-L10a mice were acquired using a confocal Leica TCS SP-8-X microscope ('Microscopy and image processing') and images analyzed utilizing the image processing software ImageJ (ImageJ 1.53f51) and its Cell Counter plugin (by Kurt De Vos, University of Sheffield).
Mice were i.p. injected with 3 mg/kg in 0.9% saline at 24 hours, 6 hours or 1 hour prior to sacrifice. Mice were decapitated, blood collected and incubated for at least 30 min on ice to allow clotting. Samples were centrifuged for 20 min at 10,000 rpm and 4°C, supernatants snapfrozen in liquid nitrogen and stored at -80°C until further processing.
MSH levels were determined utilizing the mouse MSH ELISA Kit (Abbexa, Cat#abx254513) according to manufacturer's instructions with following specifications: samples were processed in 1:2 dilutions and TMB substrate incubation was set to 20 min. Corticosterone levels were determined utilizing the mouse Corticosterone ELISA Kit (Crystal Chem, Cat#80556) according to manufacturer's instructions without modifications. Optical densities of standards and samples were assessed using a FilterMax F5 Multi-Mode microplate reader and SoftMax Pro 6.3 software (Molecular Devices).
Immediately before and one hour after CNO injection blood glucose levels were assessed on blood from a microincision in the tail tip using a Contour glucometer (Bayer).