Detour Ahead: Incretin Hormone Signaling Alters Its Intracellular Path as β-Cell Failure Progresses During Diabetes

The early events leading to b-cell failure during the pathogenesis of type 2 diabetes mellitus (T2DM) remain poorly understood (1). Upon glucose stimulation, insulin is released from b-cells in a biphasic manner with an initial first peak of insulin release (first phase), which – if the glucose stimulus persists is followed by a second prolonged phase of insulin release (2–5). Defective firstphase insulin release is among the earliest markers that predict the development of T2DM (2–4). This defect persists in patients with T2DM and is also found in first-degree relatives of patients with T2DM (4). However, the molecular underpinnings of this particular defect have largely remained elusive. A recent report by Odouri et al. (6) now provides important new insight that may help understand the early stages in the development of b-cell dysfunction of T2DM. In this commentary we place these important findings (6) in the broader context of incretin hormone signaling and the early defects that occur in incretin action in b-cells during diabetes pathogenesis. We close by outlining new avenues for scientific inquiry, which the findings by Oudori et al. have opened with their observations.


INTRODUCTION
The early events leading to b-cell failure during the pathogenesis of type 2 diabetes mellitus (T2DM) remain poorly understood (1). Upon glucose stimulation, insulin is released from b-cells in a biphasic manner with an initial first peak of insulin release (first phase), whichif the glucose stimulus persists -is followed by a second prolonged phase of insulin release (2)(3)(4)(5). Defective firstphase insulin release is among the earliest markers that predict the development of T2DM (2)(3)(4). This defect persists in patients with T2DM and is also found in first-degree relatives of patients with T2DM (4). However, the molecular underpinnings of this particular defect have largely remained elusive. A recent report by Odouri et al. (6) now provides important new insight that may help understand the early stages in the development of b-cell dysfunction of T2DM. In this commentary we place these important findings (6) in the broader context of incretin hormone signaling and the early defects that occur in incretin action in b-cells during diabetes pathogenesis. We close by outlining new avenues for scientific inquiry, which the findings by Oudori et al. have opened with their observations.

Gas or Gaq Signaling-Dependent Mechanisms
Glucose-stimulated insulin secretion (GSIS) occurs through a cascade of precisely orchestrated electro-physiologic events in pancreatic b-cells (7,8). Glucose transport into b-cells and enhanced glucose metabolism increase the cellular ATP/ADP ratio, leading to closing of ATP-sensitive K + (K ATP ) channels. Restricted K + exit through K ATP channels depolarizes the b-cell plasma membrane, which in turn is followed by activation and opening of voltage dependent Ca 2+ channels (VDCC). Calcium influx and thus increased cytoplasmic Ca 2+ concentration in the bcell, triggers insulin vesicle exocytosis (9)(10)(11)(12)(13). GSIS is amplified by hormones and neurotransmitters, of which many function by activation of G-protein coupled receptors via trimeric G-proteins containing Gas or Gaq, which signal, respectively, via cAMP and phospholipase C (PLC)-diacylglycerol (DAG)/-inositol 1, 4, 5-triposphate (IP 3 ) (9,14). Although both Gas-cAMP and Gaq-PLC-DAG)/IP 3 are clearly recognized to amplify b-cell GSIS, the exact mechanisms of their signaling pathways, how they differ and where they overlap remain incompletely understood. Both Gas-cAMP (via PKA-CREB activation) and Gaq-DAG-protein kinase C (PKC; via ERK1/2) activation increase IRS2 levels in b-cells, which is a central mediator in maintaining expression of PDX1 and other transcription factors that form a gene-regulatory network to maintain b-cell identity and maturity (15,16). Gas-cAMP and Gaq-PLC-dependent signaling pathways both regulate inositol tri-phosphate (IP 3 ) receptor activity to mobilize calcium stored in the endoplasmic reticulum (ER) into the cytoplasm; and both PKA and PKC phosphorylate select SNARE (=soluble Nethylmaleimide sensitive factor attachment protein receptor) complex proteins (i.e. SNAP25, MUNC18) to promote insulin vesicle exocytosis (17,18). Thus, Gas-PKA and Gaq-DAG-PKC signaling pathways converge on common targets that regulate GSIS amplification. Nevertheless, these two signaling systems fulfill distinct and non-redundant functions as b-cell-selective ablation of either Gas (19) or Gaq (20) results in defective b-cell function or survival.

THE CENTRAL ROLE FOR INCRETIN HORMONES IN REGULATING INSULIN SECRETION
Critical for preventing postprandial hyperglycemia are the amplifying effects of the incretin hormones glucagon-like peptide-1 (GLP-1) and glucose dependent insulinotropic peptide (GIP). Under physiologic conditions, the actions of these two incretin hormones are responsible for approximately 50% of insulin secretion after meal intake (21)(22)(23) GLP-1 and GIP are released from enteroendocrine L-and Kcells, respectively, upon nutrient stimulus in the intestine. Through the circulation they reach the pancreatic b-cells, where by activating their cognate G-protein coupled receptors, they potentiate glucose stimulated insulin release (21). The receptors for both GLP-1 (GLP-1R) and GIP (GIPR)-both abundantly expressed in b-cells, belong to the class B (secretin family) G-protein coupled receptors (GPCR); and both couple primarily to the G-protein Gas (21). Consequently, in the case of both hormones, binding and activation of their respective receptors stimulates intracellular cAMP synthesis and downstream signaling that is mediated by protein kinase A (PKA) and by the guanyl nucleotide exchange factor exchange protein activated by cAMP 2A (EPAC2A) (24). Cyclic AMP-PKA activation, via phosphorylation of PKA targets, stimulates multiple pathways within the b-cell that promote b-cell proliferation, survival, changes in gene expression as well as GSIS amplification. These are extensively reviewed elsewhere to which we refer the reader (21, 25) EPAC2A in b-cells, upon activation by cAMP translocates to the cytoplasmic membrane where it regulates exocytosis and mobilize intracellular calcium, thereby also amplifying GSIS (24,26).
In this context, is important to point out that in intestinal Lcells GLP-1 is produced through differential processing of proglucagon. More recently, there is increasing experimental evidence that GLP-1 may not only be produced in intestinal Lcells but also through differential posttranslational processing of proglucagon in pancreatic endocrine a-cells, where proglucagon is primarily processed to glucagon (27,28). As such, GLP-1 produced in a-cells may also act on b-cells in a paracrine manner (27,28). In certain experimental conditions, paracrine a−to b cell signaling of proglucagon products appears to be critical for normal b-cell function Furthermore, glucagon-the principal proglucagon-derived product in a-cells-binds to and activates both the glucagon receptor (GCGR) and GLP-1R (albeit with markedly reduced affinity as compared to GLP-1) on b-cells (29)(30)(31)(32). Like the incretin hormone receptors, GCGR belongs to the class B secretin family of GPCRs, but couples to both Gas and Gaq to stimulate, respectively cAMP-and DAG)/(IP 3 )dependent signaling (33)(34)(35). GCGR-dependent signaling has been primarily described in hepatocytes (33)(34)(35) and -absent clinching experimental data -it remains unclear whether GCGR normally also signals via both Gas and Gaq in b-cells.
Importantly, the loss of incretin hormone action in amplifying GSIS is an early characteristic of T2DM (23,(36)(37)(38). In individuals with defective first-phase insulin release and future risk of developing frank T2DM, pharmacologic GLP-1 receptor agonist treatment restores first-phase insulin secretion (39). However, quite early after the discovery of GLP-1 and GIP a remarkable difference was observed in their actions on b-cells of individuals with T2DM (22,38). While treatment with GLP-1 potently amplifies GSIS in patients with T2DM, GIP treatment fails to do so; despite the fact that receptors of both of these hormones couple to Gas to stimulate intracellular cAMPdependent signaling. As a consequence of these observations, GLP-1 receptor agonists have quickly become a focus of scientific inquiry as well as of drug development for T2DM. Although clearly a key characteristic of b-cell failure in T2DM, the difference in action between GLP-1 and GIP in islets of T2DM has at the molecular level remained thus far unexplained.

LINKING DEFECTIVE K ATP CHANNEL ACTIVITY AND DEFECTIVE INCRETIN HORMONE ACTION IN b CELLS
The carefully conducted studies reported by Odouri et al. (6) provide a potential molecular explanation for the differences in GLP-1R and GIPR signaling in healthy b-cells versus those in T2DM. The initial focus of Odouri et al. was on the role of K ATP channels in b-cells (6,40,41). Partial loss of K ATP channels increases electrical excitability and insulin secretion, resulting in hyperinsulinemia in humans and in mice (42,43). Complete loss of K ATP channels causes permanent depolarization and chronically elevated intracellular Ca 2+ concentrations at all glucose levels. However paradoxically and through unknown mechanisms, the absence of K ATP channels results in a downregulation of insulin secretion and defective GSIS (42,43).
Mice generated by Odouri et al. lack the pore-forming Kir6.2 component of the K ATP channel specifically in b-cells (b-DKcnj11 mice), and confirming prior observations are actually glucose intolerant (6,42,43). Importantly however, GLP-1 effects on amplifying GSIS was partially retained while the effects of GIP were practically absent (6). These observations are remarkably similar to findings in humans who develop T2DM (see above).

Signaling Detour for GLP-1R but Not for GIPR
A key observation by Odouri et al. is that in these "hyper-excited" K ATP channel-defective mouse islets, cAMP synthesis through GLP-1R or GIPR activation is markedly diminished. Importantly, while GIPR-dependent signaling is as a consequence practically silenced, the GLP-1R switches coupling from Gas to Gaq, thereby engaging an alternate signaling pathway and allowing GLP-1 to potentiate GSIS in bcell (6). In contrast GIPR fails to switch its coupling from Gas to Gaq. Importantly, Odouri et al. show that GLP-1R does not couple through Gaq in healthy control islets, but only in those with down-regulated K ATP channel activity (6).
In additional studies Odouri et al. examined the KK-Ay mouse, which spontaneously develops a human T2DM-like phenotype. While b-cells from non-diabetic control KK mice showed normal electro-physiologic activity and patterns of GSIS, b-cells from KK-Ay mice were chronically depolarized and lacked GSIS. Treatment with GLP-1 but not with GIP amplified GSIS in KK-Ay islets. And GLP-1 effects were inhibited with the Gaq antagonist YM-254890 (It would help had the authors also examined the effects of the more specific Gaq antagonist FR900359) (6). Thus, as in T2DM, in KK-Ay mice, b-cell GLP-1R but not GIPR signals via Gaq while Gasdependent signaling becomes defective.
The potential importance of these findings to understand bcell failure in a broader context of T2DM comes with additional studies, in which Odouri et al. (6) expose both mouse and human islets to elevated glucose levels (as found in T2DM) for 3 to 5 days-a maneuver that causes chronic K ATP channel closure (6,44). Remarkably, when such treated islets were returned to lower (i.e. normoglycemic) glucose levels, their K ATP channels remained suppressed and GSIS remained defective. However, similar to the observations made in b-DKcnj11 mice, GLP-1induced GSIS potentiation was maintained, whereas GIP treatment had no such effect (6). And again, GLP-1 receptor signaling had switched from Gas to Gaq coupling, whereas GIPR signaling had not made that switch.
These important observations by Odouri et al. provide a molecular explanation for the long known "incretin bias" in T2DM and link this particular phenomenon to a b-cell autonomous change in K ATP channel activity (6, 44)-for which the underlying pathogenic mechanisms remain incompletely understood.
While the observations made by Odouri et al. provide experimental evidence that chronic (3-5 days) exposure to elevated glucose levels can lead to a down-regulation of K ATP channel activity, it remains unclear whether other influences to which the b-cell is exposed in early T2DM pathogenesis (e.g., altered circulating lipid profiles, low level inflammation and likely additional as yet unrecognized factors) that precede hyperglycemia may also cause a down-regulation in K ATP channel activity.

DISCUSSION
As all good science, the studies reported by Oduori et al. (6) have not only significantly advanced the field but also point in which direction to look further. The remarkable finding that in b-cells GLP-1R coupling switches during diabetes pathogenesis adds impetus to understand in more detail the mechanism and pathways of Gas-and Gaq-dependent signaling in b-cell. Odouri et al. findings raise many questions related to incretin biology and to b-cell (dys-)function in T2DM pathogenesis as well as questions related to the treatment of b-cell dysfunction. Among these will undoubtedly be the following:  Whether other participants in T2DM pathogenesis also promote this switch remains unclear. 7. Which therapeutic maneuvers will allow GLP-1R signaling of the failing b-cell to couple back with Gas and also reactivate GIPR signaling in b-cells?
The following three questions pertain to therapeutic approaches of b-cell dysfunction in light of the new findings by Odouri et al.
1. Would GLP-1R agonists that bias coupling through Gaq rather than Gas be more effective in treating b-cell failure? 2. Will pharmacologic targeting of primarily Gaq-coupled GPCR, such as muscarinic M3 acetylcholine receptors (15,45,46) be more effective than GLP-1R agonists in preventing or reversing b-cell failure? 3. Would simultaneous stimulation of Gas-and Gaqdependent signaling pathways in promoting b-cell function and survival be superior to stimulation of either pathway alone?
While this article was under peer-review, a newly published study indicates a role for GIP-GIPR in amplifying amino-acidinduced glucagon secretion from islet alpha-cells as an additional mechanism of incretin hormone action (47).

AUTHOR CONTRIBUTIONS
All authors contributed to the article and approved the submitted version.