Differences in the co-distribution of the Cannabinoid Receptor-1, FAAH, and MAGL in the human and mouse brain could be limiting clinical translation of endocannabinoid outcomes in mouse models of Alzheimer disease.

Endocannabinoid system (ECS) outcomes in mouse models of Alzheimer disease (AD) do not always align with the clinical disease course. We compared the distribution of the Cannabinoid Receptor-1 (CB1R) and the enzymes, FAAH and MAGL, in autopsy AD brain samples as well as in the J20 (hAPPSwe/Ind) mouse model of AD. Both sources revealed several anti-CB1R immunoreactive species, e.g. one at 47-kDa (corresponding to the protein-coding sequence) and a reported putative splice variant at 37 kDa. We did not observe any changes in the mean expression in CB1R, FAAH or MAGL in the human samples, but did observe sex- and genotype-specific changes in the mouse brain. Regression analysis revealed strong sex- and APOE {varepsilon}4-dependent associations among the CB1Rs as well as between CB1Rs and MAGL (but not FAAH) in human cortical (but not hippocampal) samples. In the J20 mouse, associations between CB1Rs were limited to hippocampal samples, whereas associations between CB1R and both MAGL and FAAH were observed in the cortex. A diagnosis of AD disrupted any associations in the human dataset, whereas in several instances, the associations were enhanced by the hAPPSwe/Ind transgene. This inferred species-dependent regulation of the ECS could impact the clinical translational of ECS outcomes in preclinical models of AD pathology.