Metformin Can Attenuate Beta-Cell Hypersecretion-Implications for Treatment of Children with Obesity

Abstract

In children with obesity, insulin hypersecretion is proposed to precede insulin resistance. We investigated if metformin could be used to attenuate insulin secretion from palmitate-treated isolated islets and its implication for children with obesity. Human islets were exposed to palmitate for 0.5 or 1 day, when metformin was introduced. After culture, glucose-stimulated insulin secretion (GSIS) was measured. Children with obesity, who had received metformin for over six months (n = 21, age 13.9 ± 1.8), were retrospectively evaluated. Children were classified as either "reducing" or "increasing" based on the difference between AUC_0-120 of insulin during OGTT before and after metformin treatment. In human islets, GSIS increased after culture in palmitate for up to 1 day but declined with continued palmitate exposure. Whereas adding metformin after 1 day of palmitate exposure increased GSIS, adding metformin after 0.5 days reduced GSIS. In children with "reducing" insulin AUC_0-120 (n = 9), 2 h glucose and triglycerides decreased after metformin treatment, which was not observed in patients with "increasing" insulin AUC_0-120 (n = 12). In isolated islets, metformin attenuated insulin hypersecretion if introduced when islet secretory capacity was maintained. In children with obesity, improved glycemic and lipid levels were accompanied by reduced insulin levels during OGTT after metformin treatment.

Keywords: OGTT; childhood obesity; free fatty acids; glucose-stimulated insulin secretion; human islets; hyperinsulinemia; metformin.

Figure 1

Metformin increases insulin secretion from isolated human islets with declining secretory function. Islets were cultured in the absence (C) or presence of palmitate for 1 (P1), 2 (P2), or 3 (P3) days (Panel (A), top panel). Metformin was added to islets cultured in the presence of palmitate for 1 day and culture continued for 1 (P2M1) or 2 (P3M2) days in the continued presence of metformin and palmitate (Panel (A), bottom panel). After culture, insulin secretion was monitored shown with representative graphs of dynamic insulin secretion during 20 min in the presence of 5.5 followed, by 20 min in the presence of 11 mmol/L glucose (Panel (A)). AUCs of insulin secretion during 20 min (grey bars) in the presence of 11 mmol/L glucose (Panel (B)) and during the initial 10 min (black bars) and latter 10 min (white bars) in the presence of 11 mmol/L glucose (Panel (C)) are shown. Values are presented as mean ± SEM (n = 5). # p < 0.05 compared with control, * p < 0.05 compared with P3.

Figure 2

Metformin reduces insulin hypersecretion from isolated human islets with maintained secretory function (Panel (A), top panel). Islets were cultured in the absence (C) or presence of palmitate for 0.5 (P0.5) or 1 (P1) day. Metformin was added to islets cultured in the presence of palmitate for 0.5 days and culture continued for 0.5 days (P1M0.5) in the continued presence of metformin and palmitate (Panel (A), bottom panel). After culture, insulin secretion was monitored shown with representative graphs of dynamic insulin secretion during 20 min in the presence of 5.5, followed by 20 min in the presence of 11 mmol/L glucose (Panel (A)). AUCs of insulin secretion during 20 min (grey bars) in the presence of 11 mmol/L glucose (Panel (B)) and during the initial 10 min (black bars) and latter 10 min (white bars) in the presence of 11 mmol/L glucose (Panel (C)). Values are presented as mean ± SEM (n = 7). # p < 0.05 compared with control, * p < 0.05 compared with P1.

Figure 3

Insulin and glucose concentrations in children with obesity during OGTT before (open symbols) and after (closed symbols) metformin treatment. Insulin (Panel (A,B)) and glucose (Panel (C,D)) concentrations during OGTT of the “reducing” (Panel (A,C)) and “increasing” (Panel (B,D)) patient groups are shown. Values are presented as means ± SEM for n = 9 (Panel (A,C)) and n = 12 (Panel (B,D)). Y-axis starts from 5 (Panel (C,D)). * p < 0.05, ** p < 0.01 compared with before treatment.

Figure 4

Correlation of changes in insulin amounts during OGTT before and after metformin treatment and BMI-SDS. Change in insulin AUC_0–30 min (Panel (A)) and AUC_30–120 min (Panel (B)) during OGTT before and after treatment was positively correlated to BMI-SDS before treatment (r² = 0.3, p = 0.01 and r² = 0.23, p = 0.03, respectively). Positive correlation was found between changes in insulin AUC_30–120 min and BMI-SDS before and after treatment (r² = 0.21, p = 0.03) (Panel (C)). Correlation relationship between variables were assessed by simple liner regression. X-axis starts from 2.5 (Panel (A,B)).