Hypothalamic POMC Deficiency Improves Glucose Tolerance Despite Insulin Resistance by Increasing Glycosuria

Abstract

Hypothalamic proopiomelanocortin (POMC) is essential for the physiological regulation of energy balance; however, its role in glucose homeostasis remains less clear. We show that hypothalamic arcuate nucleus (Arc)POMC-deficient mice, which develop severe obesity and insulin resistance, unexpectedly exhibit improved glucose tolerance and remain protected from hyperglycemia. To explain these paradoxical phenotypes, we hypothesized that an insulin-independent pathway is responsible for the enhanced glucose tolerance. Indeed, the mutant mice demonstrated increased glucose effectiveness and exaggerated glycosuria relative to wild-type littermate controls at comparable blood glucose concentrations. Central administration of the melanocortin receptor agonist melanotan II in mutant mice reversed alterations in glucose tolerance and glycosuria, whereas, conversely, administration of the antagonist Agouti-related peptide (Agrp) to wild-type mice enhanced glucose tolerance. The glycosuria of ArcPOMC-deficient mice was due to decreased levels of renal GLUT 2 (rGLUT2) but not sodium-glucose cotransporter 2 and was associated with reduced renal catecholamine content. Epinephrine treatment abolished the genotype differences in glucose tolerance and rGLUT2 levels, suggesting that reduced renal sympathetic nervous system (SNS) activity is the underlying mechanism for the observed glycosuria and improved glucose tolerance in ArcPOMC-deficient mice. Therefore, the ArcPOMC-SNS-rGLUT2 axis is potentially an insulin-independent therapeutic target to control diabetes.

Figure 1

Improved glucose tolerance despite insulin resistance in C57BL/6J background 24-week-old obese ArcPOMC-deficient mice. A: Body weight. B: Fasting plasma insulin levels. C and D: ITT. E and F: OGTT. Bar graphs in C, D, E, and F represent the corresponding AUCs. Two-tailed Student t test was used for comparisons. *P < 0.05, **P < 0.01, ***P < 0.001 (n = 6). Error bars are mean ± SEM. KO, knockout; WT, wild type.

Figure 2

Evaluation of glucose homeostasis in 129S6/SvEvTac-background 24-week-old obese ArcPOMC-deficient mice. A: Body weight. B: Fasting plasma insulin levels. C and D: ITT. E and F: OGTT. Bar graphs in C, D, E, and F represent the corresponding AUCs. Two-tailed Student t test was used for comparisons. *P < 0.05, **P < 0.01, ***P < 0.001 (n = 6). Error bars are mean ± SEM. KO, knockout; WT, wild type.

Figure 3

Improved glucose tolerance in the presence of insulin resistance in 12-week-old weight-matched ArcPOMC-deficient mice. A: Body weight. B: Fasting plasma insulin levels. C and D: ITT in weight-matched ArcPOMC-deficient mice. E and F: OGTT in weight-matched ArcPOMC-deficient mice. Bar graphs in C, D, E, and F represent the corresponding AUCs. Two-tailed Student t test was used for comparisons. *P < 0.05 (n = 6). Error bars are mean ± SEM. KO, knockout; WT, wild type.

Figure 4

Improved glucose tolerance assessed by FSIVGTT in ArcPOMC-deficient mice. A and B: Blood glucose and corresponding AUCs of 0–10-min data points in 8-week-old ArcPOMC-deficient mice during FSIVGTT. C and D: Plasma insulin levels and glucose effectiveness in 8-week-old mice during FSIVGTT. Two-tailed Student t test was used for comparisons. *P < 0.05 (n = 6 or 7). Error bars are mean ± SEM. KO, knockout; WT, wild type.

Figure 5

Elevated glycosuria, normal liver glycogen levels, and normal gluconeogenesis in weight-matched or obese ArcPOMC-deficient mice. A: Twenty-four-hour urine glucose concentration in weight-matched 12-week-old mice before and after OGTT. B: Twenty-four-hour urine glucose concentration in weight-matched 12-week-old mice that were challenged with 10% glucose provided in their drinking water. C: Fasting hepatic glycogen levels at baseline and after glucose challenge in weight-matched 8-week-old mice (n = 6). D: PTT in 8-week-old weight-matched mice (n = 6). E: PTT in 16-week-old obese mutant mice (n = 6). Two-tailed Student t test was used for comparisons. *P < 0.05, **P < 0.01 (n = 6 or 7). Error bars are mean ± SEM. KO, knockout; NS, not significant at P > 0.05; WT, wild type.

Figure 6

MTII and Agrp treatment in female ArcPOMC-deficient and wild-type mice, respectively. A: OGTT in ArcPOMC-deficient and wild-type mice (bar graphs represent AUC). B: Twenty-four-hour body weight change after MTII or Agrp treatment. C: Twenty-four-hour food intake. Two-tailed Student paired t test was used for comparisons. **P < 0.01, ***P < 0.001 (n = 6). Error bars are mean ± SEM. KO, knockout; WT, wild type.

Figure 7

Reduced rGLUT2 but not rSGLT2 levels in ArcPOMC-deficient mice. A and B: Representative Western blot images of renal cortical GLUT2 and SGLT2, respectively (n = 8 per group for bar graphs that represent relative expression). Two-tailed Student t test was used for comparisons. **P < 0.01. Error bars are mean ± SEM. KO, knockout; WT, wild type.

Figure 8

ArcPOMC regulates rGLUT2 through the SNS. A: Whole-kidney epinephrine and norepinephrine levels (n = 6). B: Renal cortical GLUT2 expression in the presence of epinephrine treatment in ArcPOMC-deficient mice. C: Renal cortical GLUT2 expression in the presence of epinephrine treatment in wild-type mice (n = 8 per group for bar graphs that represent relative expression). D: OGTT in epinephrine- or saline-treated mice (n = 6 per group) [genotype effect: F(1,23) = 20.2, P = 0.0002; treatment effect: F(1,23) = 84.9, P < 0.0001]. E: Proposed insulin-independent mechanism by which hypothalamic POMC deficiency improves glucose tolerance. Two-tailed Student t test or two-way ANOVA followed by Bonferroni multiple comparisons test were used for comparisons. **P < 0.01. Error bars are mean ± SEM. E, epinephrine; KO, knockout; NE, norepinephrine; WT, wild type.