Fatty Acids and Insulin Secretion: From FFAR and Near?

Nonesterified, long-chain fatty acids (LCFAs) potentiate glucose-stimulated insulin secretion (GSIS). In other words, LCFAs do not trigger insulin release by themselves but strongly amplify the stimulating effect of glucose (1). Yet, from a physiological standpoint it does not seem logical that an energetic substrate whose circulating levels are elevated when glucose levels are low would stimulate insulin release (2). This apparent contradiction can be resolved if the LCFAs that stimulate insulin secretion come from the local islet environment, where their concentrations are likely significantly higher than plasma levels, rather than from the circulation (3).

LCFA potentiation of GSIS occurs via at least two distinct mechanisms. LCFAs are transported across the plasma membrane, activated into long-chain acyl-CoA, and metabolized to generate lipid signaling molecules that promote insulin exocytosis when glucose levels are elevated (3). LCFAs also bind to and activate the G-protein–coupled receptor FFAR1 (GPR40), which is mainly coupled to Gα_q/11 and triggers a signaling cascade leading to the activation of protein kinase D1 and insulin release (4) (Fig. 1). Intracellular metabolism and FFAR1 activation each account for approximately 50% of the overall effect of exogenous LCFA on insulin release (5). The question of how much LCFA is available at the vicinity of the islet for uptake or FFAR1 activation is complex. First, the vast majority of LCFA in the circulation and interstitial space is bound to albumin, and the bioactive, non-BSA–bound fraction is extremely small (6). Second, lipoprotein lipase expressed in islet capillaries delivers LCFA from postprandial chylomicrons to the β-cell (7). Third, β-cells exhibit significant glucose-stimulated lipolytic activity (8,9), which results in LCFA release in the extracellular space (10). This later observation suggests a possible autocrine effect of LCFA on insulin secretion (2,3).

In this issue of Diabetes, Hauke et al. (11) used complementary approaches to examine the role of endogenous LCFA on insulin secretion in insulin-secreting MIN6 cells and mouse primary β-cells. First, the authors showed that stringent washing in a perifusion system decreased cytosolic Ca²⁺ oscillations in response to glucose, which were recovered by addition of preconditioned medium from MIN6 cells. Second, addition of fatty acid (FA)–free BSA to the incubation buffer also impaired glucose-induced Ca²⁺oscillations and insulin secretion, an effect that was not observed if the BSA was preloaded with LCFA or chemicals that occupy the LCFA binding sites. The recovery of Ca²⁺ oscillations and insulin secretion was prevented in the presence of the FFAR1 inhibitor GW1100. Mass spectrometry analyses confirmed the predominance of palmitic and stearic acids in both cell extracts and supernatants. Next, the authors used sulfo-caged stearic and oleic acid with a photo-labile caging group (12) to deliver LCFAs at the plasma membrane. Ultraviolet-mediated release of LCFA led to recovery of Ca²⁺ oscillations in cells previously depleted by FA-free BSA incubation. Accordingly, forced stimulation of LCFA release by recombinant phospholipase A2 or lipoprotein lipase also restored Ca²⁺ oscillations and insulin secretion. Overall, the authors concluded that endogenous LCFAs contribute to the normal regulation of GSIS and that provision of exogenous LCFAs restores GSIS through FFAR1 signaling.

These findings, using thorough and complementary approaches, elegantly corroborate the notion that endogenous LCFA signaling plays a key role in stimulus-secretion coupling in the β-cell. The results also strongly suggest, but do not formally prove, that LCFAs released by the β-cell signal back on the same or neighboring cells in an autocrine or paracrine manner (Fig. 1). The data using the FFAR1 inhibitor indicate that this may be the case but would benefit from additional confirmation in islets from FFAR1 knockout animals. In fact, the extent to which FFAR1 signaling is involved in the observed effects is not entirely clear. Hauke et al. (11) suggest that the rapid termination and fast recovery of β-cell activity upon addition and reprovision of FA-free BSA is due to direct action of LCFA on ion channel activity, whereas FFAR1 signaling is implicated in longer-term effects. Yet, as discussed by the authors, FFAR1 signaling has also been shown to modulate ion channel activity in β-cells (13–17).

Of note, the reduction of β-cell activity upon washing with FA-free BSA is in contradiction with an earlier report by Straub and Sharp (18), who observed a large increase in insulin secretion from rat islets following a 4- to 6-h preincubation with FA-free BSA, a discrepancy that may be explained by the different time frames of the two studies (19).

A precedent for an autocrine action of lipid derivatives in β-cells was recently established by Tunaru et al. (20), who showed that the eicosanoid metabolite of arachidonic acid, 20-hydroxyeicosatetraenoic acid (20-HETE), is released in response to glucose, binds to FFAR1, and promotes GSIS (Fig. 1). The notion that LCFAs or derivatives “secreted” by the β-cell in response to glucose signal through FFAR1 to potentiate GSIS does resolve the physiological contradiction mentioned above and is consistent with the fact that the circulating concentration of non-BSA–bound LCFA is likely lower that of the EC50 for FFAR1 (2). However, if this glucose-induced, feed-forward autocrine signaling loop is essential for GSIS, one would expect FFAR1 knockout mouse islets to exhibit lower GSIS, which is not consistently observed (5,20). This issue is intrinsically difficult to assess experimentally as any manipulation of intracellular LCFA levels to limit their release will also alter the generation of intracellular signaling molecules that contribute to GSIS (3).

In summary, the study by Hauke et al. (11) confirms the critical role of endogenous LCFA signaling in the control of GSIS and presents converging and convincing lines of evidence in support of LCFA being added to the list of molecules to contribute to autocrine regulation of β-cell function (reviewed in ref. 21). It remains to be determined whether such a mechanism is operative and functionally important in human islets. Interestingly, both FFAR1 expression (22) and glucose-induced 20-HETE formation (20) are reduced in islets from subjects with type 2 diabetes, suggesting the possibility that this autocrine feedback loop may indeed contribute to islet dysfunction in type 2 diabetes.