A fatty acid- dependent step is critically important for both glucose- and non-glucose-stimulated insulin secretion.

Lowering of the plasma FFA level in intact fasted rats by infusion of nicotinic acid (NA) caused essentially complete ablation of insulin secretion (IS) in response to a subsequent intravenous bolus of arginine, leucine, or glibenclamide (as previously found using glucose as the beta-cell stimulus). However, in all cases, IS became supranormal when a high FFA level was maintained by co-infusion of lard oil plus heparin. Each of these secretagogues elicited little, if any, IS from the isolated, perfused "fasted" pancreas when tested simply on the background of 3 mM glucose, but all became extremely potent when 0.5 mM palmitate was also included in the medium. Similarly, IS from the perfused pancreas, in response to depolarizing concentrations of KCl, was markedly potentiated by palmitate. As was the case with intravenous glucose administration, fed animals produced an equally robust insulin response to glibenclamide regardless of whether their low basal FFA concentration was further reduced by NA. In the fasted state, arginine-induced glucagon secretion appeared to be independent of the prevailing FFA concentration. The findings establish that the essential role of circulating FFA for glucose-stimulated IS after food deprivation also applies in the case of nonglucose secretagogues. In addition, they imply that (i) a fatty acid-derived lipid moiety, which plays a pivotal role in IS, is lost from the pancreatic beta-cell during fasting; (ii) in the fasted state, the elevated level of plasma FFA compensates for this deficit; and (iii) the lipid factor acts at a late step in the insulin secretory pathway that is common to the action of a wide variety of secretagogues.

The insulinotropic potency of fatty acids is influenced profoundly by their chain length and degree of saturation.

Lowering of the elevated plasma FFA concentration in 18- 24-h fasted rats with nicotinic acid (NA) caused complete ablation of subsequent glucose-stimulated insulin secretion (GSIS). Although the effect of NA was reversed when the fasting level of total FFA was maintained by coinfusion of soybean oil or lard oil (plus heparin), the more saturated animal fat proved to be far more potent in enhancing GSIS. We therefore examined the influence of individual fatty acids on insulin secretion in the perfused rat pancreas. When present in the perfusion fluid at 0.5 mM (in the context of 1% albumin), the fold stimulation of insulin release from the fasted pancreas in response to 12.5 mM glucose was as follows: octanoate (C8:0), 3.4; linoleate (C18:2 cis/cis), 5.3; oleate (C18:1 cis), 9.4; palmitate (C16:0), 16. 2; and stearate (C18:0), 21.0. The equivalent value for palmitoleate (C16:1 cis) was 3.1. A cis--> trans switch of the double bond in the C16:1 and C18:1 fatty acids had only a modest, if any, impact on their potency. A similar profile emerged with regard to basal insulin secretion (3 mM glucose). When a subset of these fatty acids was tested in pancreases from fed animals, the same rank order of effectiveness at both basal and stimulatory levels of glucose was seen. The findings reaffirm the essentiality of an elevated plasma FFA concentration for GSIS in the fasted rat. They also show, however, that the insulinotropic effect of individual fatty acids spans a remarkably broad range, increasing and decreasing dramatically with chain length and degree of unsaturation, respectively. Thus, for any given level of glucose, insulin secretion will be influenced greatly not only by the combined concentration of all circulating (unbound) FFA, but also by the makeup of this FFA pool. Both factors will likely be important considerations in understanding the complex interplay between the nature of dietary fat and whole body insulin, glucose, and lipid dynamics.

Essentiality of circulating fatty acids for glucose-stimulated insulin secretion in the fasted rat.

We asked whether the well known starvation-induced impairment of glucose-stimulated insulin secretion (GSIS) seen in isolated rat pancreas preparations also applies in vivo. Accordingly, fed and 18-24-h-fasted rats were subjected to an intravenous glucose challenge followed by a hyperglycemic clamp protocol, during which the plasma-insulin concentration was measured. Surprisingly, the acute (5 min) insulin response was equally robust in the two groups. However, after infusion of the antilipolytic agent, nicotinic acid, to ensure low levels of plasma FFA before the glucose load, GSIS was essentially ablated in fasted rats, but unaffected in fed animals. Maintenance of a high plasma FFA concentration by coadministration of Intralipid plus heparin to nicotinic acid-treated rats (fed or fasted), or further elevation of the endogenous FFA level in nonnicotinic acid-treated fasted animals by infusion of etomoxir (to block hepatic fatty acid oxidation), resulted in supranormal GSIS. The in vivo findings were reproduced in studies with the perfused pancreas from fed and fasted rats in which GSIS was examined in the absence and presence of palmitate. The results establish that in the rat, the high circulating concentration of FFA that accompanies food deprivation is a sine qua non for efficient GSIS when a fast is terminated. They also serve to underscore the powerful interaction between glucose and fatty acids in normal beta cell function and raise the possibility that imbalances between the two fuels in vivo could have pathological consequences.