What the model had to fix
Getting an antipsychotic to make a mouse fat has been the hard part. Injection and drinking water produced only modest changes in food intake and body weight, so the authors put clozapine into the food at 50 mg/kg, in a diet matched to the control for macronutrient composition and energy density. Two numbers turn that into a dosing model. Plasma clozapine averaged 1.12 µg/mL after four weeks, inside the therapeutic range reported in patients. And mice offered both diets for 8 days ate them equally, which removes taste as the reason they ate more.[1]
The metabolic chambers say what kind of result this is. Energy expenditure went up on the clozapine diet, by 1.18 kcal per day, while intake went up by 11.75 kcal per day. The gap is the whole effect: the weight comes from eating, and the small rise in burn does nothing to cancel it.[1]
The comparisons that carry the claim
The weight curve alone would not carry the mechanism; the comparisons around it do. Ziprasidone, an antipsychotic with minimal propensity to make patients gain weight, went in at the same 3 mg/kg and left the MC4R neurons of the hypothalamus untouched. Olanzapine, which does put weight on patients, failed to open the potassium channel at all, and mice with Kir7.1 removed from those neurons still grew fat on an olanzapine diet. Each of those could have come out the other way and would have sunk the mechanism.[1]
Two independent handles land on the same channel. Deleting Kir7.1 from the MC4R neurons blunted the weight gain, and ML418, which shuts the channel with a drug instead of a gene, brought weight down after 10 days in mice already obese on the clozapine diet, along with fat mass and plasma leptin. Converging genetics and pharmacology is the strongest form this kind of evidence takes inside one species. It still leaves room for a shared step further downstream that neither experiment isolates.[1]
Where the result stops
The design also draws its own boundary. The mice are female C57BL/6, chosen because the metabolic effects of antipsychotic drugs are more pronounced in females, and the 12-week weight result rests on 18 animals; setmelanotide and ML418 are mouse results, and no one taking clozapine should read them as a treatment. What is worth watching next is the model itself. I expect the next published chronic clozapine mouse study to put the drug in the food and report a measured plasma level, and a methods section that returns to injection or drinking water would show the field passed on the point.[1]