Peripheral Mechanisms Mediating the Sustained Antidiabetic Action of FGF1 in the Brain

Abstract

We recently reported that in rodent models of type 2 diabetes (T2D), a single intracerebroventricular (icv) injection of fibroblast growth factor 1 (FGF1) induces remission of hyperglycemia that is sustained for weeks. To clarify the peripheral mechanisms underlying this effect, we used the Zucker diabetic fatty fa/fa rat model of T2D, which, like human T2D, is characterized by progressive deterioration of pancreatic β-cell function after hyperglycemia onset. We report that although icv FGF1 injection delays the onset of β-cell dysfunction in these animals, it has no effect on either glucose-induced insulin secretion or insulin sensitivity. These observations suggest that FGF1 acts in the brain to stimulate insulin-independent glucose clearance. On the basis of our finding that icv FGF1 treatment increases hepatic glucokinase gene expression, we considered the possibility that increased hepatic glucose uptake (HGU) contributes to the insulin-independent glucose-lowering effect of icv FGF1. Consistent with this possibility, we report that icv FGF1 injection increases liver glucokinase activity by approximately twofold. We conclude that sustained remission of hyperglycemia induced by the central action of FGF1 involves both preservation of β-cell function and stimulation of HGU through increased hepatic glucokinase activity.

Figure 1

Glucose-lowering effect of a single icv FGF1 injection in ZDF and ZDL rats. Daily BG (A), FI (B), and BW (C) values from ad libitum–fed ZDF rats after a single icv injection of either Veh (n = 9) or FGF1 (3 μg; n = 9). Daily BG (D), FI (E), and BW (F) values from ad libitum–fed ZDL rats after a single icv injection of either Veh (n = 7) or FGF1 (3 μg; n = 8). Data are mean ± SEM. *P < 0.05 vs. icv Veh.

Figure 2

Contribution of transient anorexia to the glucose-lowering effect of icv FGF1 in ZDF rats. Daily BG (A), FI (B), and BW (C) values from pair-fed ZDF rats after a single icv injection of either Veh (n = 9) or FGF1 (3 μg; n = 9). Data are mean ± SEM. ***P < 0.001 vs. icv Veh.

Figure 3

Preserved basal insulin secretion by icv FGF1 in ZDF rats. Weekly plasma insulin (A), glucagon (B), FFA (C), and TG (D) from ad libitum–fed ZDF rats after a single icv injection of either Veh (n = 9) or FGF1 (3 μg; n = 9). Data are mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 vs. icv Veh.

Figure 4

Time course of the effect of icv FGF1 on β-cell mass in ZDF rats. Representative images of staining of pancreatic islet sections from ZDF rats for insulin (A), Nkx6.1 (B), and merged images (C). Pancreatic β-cell mass at day 0 (n = 7), 3 weeks (n = 8 icv Veh vs. n = 10 icv FGF1), and 7 weeks (n = 9/group) in ad libitum–fed ZDF rats after a single icv injection of either Veh or FGF1 (3 μg) (D). Data are mean ± SEM. **P < 0.01 vs. icv Veh at day 0 and 3 weeks; ***P < 0.001 vs. icv Veh at day 0, 3 weeks, and 7 weeks; ###P < 0.001 vs. icv FGF1 at 7 weeks.

Figure 5

Effect of icv FGF1 on determinants of glucose tolerance in ZDF rats. Plasma glucose (A), Δ plasma glucose (correcting for differences in basal glucose) (B), Δ plasma AUCglucose (Glc AUC) (C), plasma insulin (D), Δ plasma insulin (correcting for differences in basal insulin) (E), Δ plasma insulin AUC (Ins AUC) (F), S_G (G), AIR_G (H), and S_I (I) during an FSIGT in ZDF rats 2 weeks after receiving a single icv injection of either Veh (n = 7) or FGF1 (3 μg; n = 7). BG (J) and percent Δ BG (K) during an IVITT in ZDF rats 3 weeks after receiving a single icv injection of either Veh (n = 7) or FGF1 (3 μg; n = 6). Data are mean ± SEM. *P < 0.05 vs. icv Veh.

Figure 6

Dependence of basal insulin secretion on antidiabetic effect of icv FGF1. ZDF rats underwent matched euglycemic clamp 14 days after a single icv injection of either Veh (n = 11) or FGF1 (3 μg; n = 11). Daily morning BG (A), FI (B), and BW (C) values from ad libitum–fed ZDF rats. Plasma glucose (D), plasma insulin (E), insulin infusion rate (F), and glucose infusion rate (GIR) (G) during the matched euglycemic clamp. Plasma glucagon (H) and plasma Cort (I) at the start (t = 0 min) and end (t = 120 min) of the matched euglycemic clamp. Data are mean ± SEM. *P < 0.05, ***P < 0.001 vs. icv Veh.

Figure 7

Effect of icv FGF1 on plasma lactate and hepatic GCK in ZDF rats. Plasma lactate (A), Δ plasma lactate (correcting for differences in basal lactate) (B), Δ plasma lactate AUC (C), and calculated hepatic K_GK (D) during an FSIGT in ZDF rats 2 weeks after receiving a single icv injection of either Veh (n = 7) or FGF1 (3 μg; n = 7). E: Hepatic mRNA expression of Gck, Pklr, Gys2, Pck1, and G6pc by real-time PCR in ZDF rats 3 weeks after receiving a single icv injection of either Veh (n = 11) or FGF1 (3 μg; n = 11). F: Hepatic GCK activity in ZDF rats at day 0 baseline (icv Veh, n = 6) or 3 weeks after receiving a single icv injection of either Veh (n = 8) or FGF1 (3 μg; n = 11). Data are mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 vs. icv Veh.