Unexpected Lethal Consequences of Disrupting Angiotensinogen in Agouti‐Related Peptide (AgRP)‐Expressing Cells

Abstract

Genetic disruption of angiotensin (ANG) type 1A receptors (AT1A) in cells which express Agouti‐related peptide (AgRP) abolishes thermogenic sympathetic nerve activity (SNA) and resting metabolic rate (RMR) responses to leptin and high fat diet (HFD). Although it is known that glia secrete angiotensinogen (AGT), in silico reanalysis of published single‐cell RNAseq datasets describing transcriptomes of cells from mouse hypothalamus (GSE74672) and RNAscope fluorescent in situ hybridization independently confirm that AgRP neurons of the hypothalamic arcuate nucleus (ARC) in adult wildtype mice express AGT mRNA. We therefore sought to explore the functional significance of this expression pattern. Why, if AgRP neurons are bathed in AGT in the interstitial space, would they synthesize AGT? Mice were generated which lack AGT specifically in AgRP cells (AGTAgRP‐KO mice) using the Cre‐Lox approach (sire: AgRP‐Cre+, AGTF/WT x dam: AgRP‐Cre−, AGTF/F). Of the first 116 offspring, this resulted in expected overall distributions across sex (53% female), Cre (47% Cre+) and AGT (46% AGTF/F) genotypes, but an underrepresentation of targeted AGTAgRP‐KO mice at weaning (n=5 females + 1 male; Χ2=52.0, df=3, p<0.001), suggesting pre‐weaning or in utero lethality of the genotype. Before and at 8 weeks of age, female AGTAgRP‐KO mice exhibited normal body mass (n=5, 17.3±0.4 vs n=12 littermates 17.3±0.3 g); 5 weeks of 45% HFD caused significant weight gain in all mice (p<0.05), but AGTAgRP‐KO mice gained less (body mass: 18.3±0.4 vs 19.7±0.3 g, p<0.05, and fat mass by NMR: 1.6±0.4 vs 2.3±0.2 g, p<0.05), which was not due to suppression of food intake (p=NS) or digestive efficiency as assessed by bomb calorimetry (p=NS). Histological examination of the kidney suggests the presence of interstitial fibrosis, inflammation, glomerulosclerosis and tubular dilation in AGTAgRP‐KO mice. Interestingly, Cre activity is present within the medulla of both the adrenal gland and kidney but not liver of AgRP‐Cre mice. Preliminary tests suggest reduced total aldosterone (n=5, 3.6±1.7 vs n=9 littermates 6.9±1.1 ng/d) and corticosterone (n=6, 0.4±0.1 pg vs n=9 littermates 0.9±0.2 ng/d) in the urine, but no changes in AGT gene expression in the liver or kidney (n=3,4 p=NS) of AGTAgRP‐KO mice. We conclude that (i) AGT is expressed by AgRP cells of the ARC, (ii) the AgRP‐Cre construct drives Cre activity in these cells, but also in layers of the adrenal gland and kidney, (iii) disruption of AGT in cells targeted by AgRP‐Cre unexpectedly causes a developmental lethality that is more penetrant in males, and (iv) in surviving females, the disruption of AGT in AgRP‐expressing cells causes resistance to weight gain through increased energy expenditure.This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.