Deletion of insulin-degrading enzyme elicits antipodal, age-dependent effects on glucose and insulin tolerance

Abstract

Background: Insulin-degrading enzyme (IDE) is widely recognized as the principal protease responsible for the clearance and inactivation of insulin, but its role in glycemic control in vivo is poorly understood. We present here the first longitudinal characterization, to our knowledge, of glucose regulation in mice with pancellular deletion of the IDE gene (IDE-KO mice).

Methodology: IDE-KO mice and wild-type (WT) littermates were characterized at 2, 4, and 6 months of age in terms of body weight, basal glucose and insulin levels, and insulin and glucose tolerance. Consistent with a functional role for IDE in insulin clearance, fasting serum insulin levels in IDE-KO mice were found to be ∼3-fold higher than those in wild-type (WT) controls at all ages examined. In agreement with previous observations, 6-mo-old IDE-KO mice exhibited a severe diabetic phenotype characterized by increased body weight and pronounced glucose and insulin intolerance. In marked contrast, 2-mo-old IDE-KO mice exhibited multiple signs of improved glycemic control, including lower fasting glucose levels, lower body mass, and modestly enhanced insulin and glucose tolerance relative to WT controls. Biochemically, the emergence of the diabetic phenotype in IDE-KO mice correlated with age-dependent reductions in insulin receptor (IR) levels in muscle, adipose, and liver tissue. Primary adipocytes harvested from 6-mo-old IDE-KO mice also showed functional impairments in insulin-stimulated glucose uptake.

Conclusions: Our results indicate that the diabetic phenotype in IDE-KO mice is not a primary consequence of IDE deficiency, but is instead an emergent compensatory response to chronic hyperinsulinemia resulting from complete deletion of IDE in all tissues throughout life. Significantly, our findings provide new evidence to support the idea that partial and/or transient inhibition of IDE may constitute a valid approach to the treatment of diabetes.

Figure 1. Constitutive hyperinsulinemia in IDE-KO mice.

Insulin levels in serum of 2-, 4- and 6-mo-old wild-type (WT) and IDE-KO (KO) mice following overnight fasting. Data are mean ± SEM of 10–12 mice per group. *P<0.05 determined by 2-tailed Student's t test.

Figure 2. Age-dependent changes in basal blood glucose and body weight.

A, Blood glucose in 2-, 4- and 6-mo-old wild-type (WT) and IDE-KO (KO) mice following overnight fasting. Note that 2-mo-old IDE-KO mice exhibit significantly lower basal glucose levels relative to controls. B, Body weight of fasted 2-, 4- and 6-mo-old WT and IDE-KO mice. Note that IDE-KO mice weigh significantly less than wild-type controls mice at 2 months yet significantly more at 6 months of age. Data are mean ± SEM of 10–12 mice per group. *P<0.05 IDE-KO vs. WT as determined by 2-tailed Student's t test.

Figure 3. Age-dependent emergence of the diabetic phenotype in IDE-KO mice.

A–C, Glucose tolerance tests in 2- (A), 4-(B), and 6-mo-old (C) wild-type (WT) and IDE-KO (KO) mice. Note that 2-mo-old IDE-KO mice exhibit improved glucose tolerance at some time points. Data are mean ± s.d. of 10–12 mice per condition. *P<0.05. D–F, Insulin tolerance tests in 2- (D), 4-(E), and 6-mo-old (F) WT (WT) and IDE-KO (KO) mice. Note the trend towards improved insulin tolerance in 2-mo-old IDE-KO mice. Data are mean ± s.d. of 10–12 mice per condition. *P<0.05 IDE-KO vs. WT as determined by 2-tailed Student's t test.

Figure 4. IDE-KO mice show age-dependent reductions in IR levels and insulin-stimulated glucose uptake that correlate with the onset of the diabetic phenotype.

A–C, IR levels in muscle (A) adipose (B) and liver (C) tissue from wild-type (WT) and IDE-KO (KO) mice. Note that IR levels in IDE-KO mice are significantly decreased in all tissues at 6 months, but not at 2 months of age. Graphs show mean ± SEM of IR levels quantified by luminescent imaging (see Materials and Methods ) and normalized to IR levels in 2-mo-old WT mice. *P<0.05 relative to 6-mo-old WT mice; ^#P<0.01 relative to 2-mo-old KO mice and P<0.05 relative to 6-mo-old WT and KO mice; ^⧫P<0.05 relative to 2-mo-old KO mice, all determined by 2-tailed Student's t tests. D, Insulin-stimulated glucose uptake is significantly impaired in primary adipocytes isolated from 6-mo-old IDE-KO mice. Quantification was performed by measuring cellular uptake of ¹⁴C-deoxyglucose (¹⁴C-DOG) various times after addition of insulin (16.7 nM). Data are mean of 2 independent experiments utilizing epididymal adipose tissue collected from 4 mice per condition for each replication.