Acute sulfonylurea therapy at disease onset can cause permanent remission of KATP-induced diabetes

Abstract

Objective: Neonatal diabetes mellitus (NDM) can be caused by gain-of-function ATP-sensitive K(+) (K(ATP)) channel mutations. This realization has led to sulfonylurea therapy replacing insulin injections in many patients. In a murine model of K(ATP)-dependent NDM, hyperglycemia and consequent loss of β-cells are both avoided by chronic sulfonylurea treatment. Interestingly, K(ATP) mutations may underlie remitting-relapsing, transient, or permanent forms of the disease in different patients, but the reason for the different outcomes is unknown.

Research design and methods: To gain further insight into disease progression and outcome, we examined the effects of very early intervention by injecting NDM mice with high-dose glibenclamide for only 6 days, at the beginning of disease onset, then after the subsequent progression with measurements of blood glucose, islet function, and insulin sensitivity.

Results: Although ∼70% of mice developed severe diabetes after treatment cessation, ∼30% were essentially cured, maintaining near-normal blood glucose until killed. Another group of NDM mice was initiated on oral glibenclamide (in the drinking water), and the dose was titrated daily, to maintain blood glucose <200 mg/dL. In this case, ∼30% were also essentially cured; they were weaned from the drug after ∼4 weeks and again subsequently maintained near-normal blood glucose. These cured mice maintain normal insulin content and were more sensitive to insulin than control mice, a compensatory mechanism that together with basal insulin secretion may be sufficient to maintain near-normal glucose levels.

Conclusions: At least in a subset of animals, early sulfonylurea treatment leads to permanent remission of NDM. These cured animals exhibit insulin-hypersensitivity. Although untreated NDM mice rapidly lose insulin content and progress to permanently extremely elevated blood glucose levels, early tight control of blood glucose may permit this insulin-hypersensitivity, in combination with maintained basal insulin secretion, to provide long-term remission.

FIG. 1.

Short period of intraperitoneal glibenclamide treatment of DTG mice early on can cause permanent remission of diabetes. A and B: Individual traces of fed blood glucose vs. time. Control (black) and SU untreated DTG mice (light blue; A) and DTG mice treated with the SU glibenclamide (0.5 g/kg; B) were injected intraperitoneally on days 0–5 during tamoxifen induction. Cured mice are shown in red, and noncured mice are shown in dark blue. C: The same data as in A (untreated DTG only; light blue), and cured (red) and noncured (dark blue) glibenclamide-injected mice, for the first 8 days after tamoxifen induction. D: Individual values of fed blood glucose and plasma insulin (day 60 after tamoxifen induction) from controls (black; n = 27 mice) and DTG animals (cured: red, n = 10 mice; and noncured: dark blue, n = 24 mice) acutely treated with glibenclamide. Significant differences: *P < 0.05 with respect to control.

FIG. 2.

Early oral glibenclamide treatment can prevent the development of diabetes and preserve insulin content. A, left: Representative daily glibenclamide dose (top) and corresponding fed blood glucose (bottom) from a single (DTG cured [red] and noncured [blue]) mouse. A, right: Average glibenclamide dose and individual fed glucose at day 56 in controls (black; n = 10) and in cured (red; n = 3) and noncured DTG mice (blue; n = 6). Glibenclamide was added to the drinking water, with an initial dose of 160 mg/L. The dose was reduced by half each day that blood glucose fell. All cured mice were eventually weaned from the drug, whereas noncured mice, even at the highest dose, did not improve blood glucose. B: GFP fluorescence in pancreatic islets from DTG cured and noncured mice. C: Blood glucose vs. time after injection of 1.5 g/kg glucose after overnight fast. Glucose tolerance test in control (black) and noncured DTG (blue) and cured DTG (red) mice treated with glibenclamide (day 60 after tamoxifen induction; n = 10–34 mice per group) is shown. *Significant difference of P < 0.05 with respect to control at each time point. D: Insulin secretion from isolated islets in response to glucose (white and light color), glibenclamide (medium color), and KCl (dark color). E: Islet insulin content from control mice (black) and DTG mice treated with glibenclamide (cured: red; and noncured: blue). Data in D and E are shown as mean ± SEM (n = 8–10 mice in each group). Significant difference: *P < 0.05 with respect to 1 mmol/L glucose or to control islets, respectively; glib, glibenclamide; glu, glucose; ns, nonsignificant differences are indicated. (A high-quality digital representation of this figure is available in the online issue.)

FIG. 3.

DTG cured mice are more insulin sensitive. A: Insulin tolerance test was performed at day 60 after tamoxifen induction in controls and DTG mice treated early with glibenclamide (n = 8–10 mice per group). B: Glucose (mg/dL; top) and glucose infusion rate (GIR; mg/kg/min; bottom) over time during hyperinsulinemic-euglycemic clamp on control (black) and cured DTG (red) mice. *Significant difference of P < 0.05 with respect to control mice at each time point.