Insulin receptor (InsR) deletion from renal tubule reduces kidney size and upregulates gluconeogenic capacity of the proximal tubule in mice

Abstract

Insulin has been demonstrated to have a number of important actions in the kidney, including activation of transport, growth, and inhibition of proximal tubule (PT) gluconeogenesis. Metabolic syndrome (MetS) is associated with down-regulation of the abundance of insulin receptor (InsR) and early signaling. Previously, we found evidence of elevated renal gluconeogenesis in PT-select (non-inducible, deleted from birth) InsR KO mice. Whether development of the kidney was affected from birth in those mice was uncertain. In this study, we determine whether deletion of the InsR (in adult mice) from the whole-renal tubule using Paired Box 8 (PAX-8) targeting of tetracycline-inducible cre-recombinase (Cre) in floxed InsR-harboring mice, affects glucose homeostasis. Male (M) and female (F) Cre+ and Cre- mice (8-months old) were treated with doxycycline (DOX) in the diet for 2 weeks (to KO the InsR in the Cre+ mice, n = 9-12 mice/group). Fasting (18-hr) and refed (2-hr) glucose were obtained in the baseline, 10, and 45 days post-DOX treatment. Ten- and 45-day refed blood glucoses were significantly higher in M, but not F, KO (relative to their, respective WT). No differences in fasting glucose were observed. At 8-weeks (post-DOX), mice were fasted overnight (to upregulate gluconeogenesis), then provided a meal 3 hours prior to euthanization to raise endogenous insulin. No genotype effect on final body weight was found; however, kidney weights were lower (p = 0.01 for genotype) in the KO, especially in males (g, mean ± sem): 0.189±0.003 (WTM); 0.165 ± 0.006 (KOM); 0.142 ± 0.002 (WTF); 0.137 ± 0.006 (KOF). We developed an approach to measure glucose production in the medium of PT-suspensions harvested upon euthanasia. Suspended PT were aliquoted to receive 1 of 3 treatments for a 1 hour incubation: 1) vehicle (0 glucose containing Dulbecco's modified medium); 2) vehicle containing cyclic AMP (cAMP) and dexamethasone (DEXA); and 3) cAMP+DEXA+ 5000 pM insulin. Glucose in the media was measured by Amplex Red. Three-way ANOVA revealed significant effects of genotype (p = 0.033), sex (p = 0.048), and PT treatment (p = 0.0047) on glucose production. PT harvested from KO males produced the most glucose, and insulin reduced glucose production in all groups (% decrease versus vehicle): 43, 18, 29, 48 in WTM, KOM, WTF, and KOF, respectively. Surprisingly, DEXA and cAMP did not stimulate glucose production, as expected, which may represent differential regulation relative to liver. Remaining sensitivity of KO PT to insulin may be the result of insulin binding to insulin-like-growth factor receptors, which has been observed in humans with InsR mutations. Overall, these results demonstrate that deletion of the InsR from kidney tubular epithelium in adulthood reduces the size of the kidney, results in post-prandial relative hyperglycemia, and elevated glucose producing capacity in PT. This supports a non-redundant role for PT-InsR signaling in whole-body glucose homeostasis.