Preproglucagon Products and Their Respective Roles Regulating Insulin Secretion

Abstract

Historically, intracellular function and metabolic adaptation within the α-cell has been understudied, with most of the attention being placed on the insulin-producing β-cells due to their role in the pathophysiology of type 2 diabetes mellitus. However, there is a growing interest in understanding the function of other endocrine cell types within the islet and their paracrine role in regulating insulin secretion. For example, there is greater appreciation for α-cell products and their contributions to overall glucose homeostasis. Several recent studies have addressed a paracrine role for α-cell–derived glucagon-like peptide-1 (GLP-1) in regulating glucose homeostasis and responses to metabolic stress. Further, other studies have demonstrated the ability of glucagon to impact insulin secretion by acting through the GLP-1 receptor. These studies challenge the central dogma surrounding α-cell biology describing glucagon’s primary role in glucose counterregulation to one where glucagon is critical in regulating both hyper- and hypoglycemic responses. Herein, this review will update the current understanding of the role of glucagon and α-cell–derived GLP-1, placing emphasis on their roles in regulating glucose homeostasis, insulin secretion, and β-cell mass.

When a healthy individual consumes a meal, blood glucose concentrations rise, eliciting a series of events aimed at restoring homeostatic glucose levels. One critical response is that β cells, within the islets of the endocrine pancreas, respond to nutrient levels as well as several other endocrine and neural signals by secreting insulin, the only hormone capable of reducing blood glucose levels. In individuals with type 1 diabetes (T1D) or type 2 diabetes (T2D), glucose-stimulated insulin secretion is severely impaired due to lack of insulin and/or insulin resistance. A significant amount of attention has been placed on understanding the regulation of the intracellular processes within β-cells that dictate β-cell mass and function as well as the pathophysiology associated with the destruction of β-cells by the immune system in T1D or the progressive deterioration of β-cell function in T2D. Since its discovery 100 years ago, the understanding of what regulates insulin secretion has drastically advanced.

Another endocrine cell within the islet, the α-cell, has traditionally been thought of as a key cell that responds to low glucose levels by releasing glucagon, a critical counterregulatory hormone that has been thought to function in opposition of insulin action. Although previously suggested by pharmacological in vitro studies, recent evidence with current technical and genetic tools has further demonstrated that the α-cell, which is physically and functionally linked to the insulin-producing β-cells, also has a role to play in responding to hyperglycemia by stimulating insulin secretion. Here, we briefly review the traditional role of the α-cell but focus on the more recent data outlining the various contributions of α-cell products, glucagon-like peptide-1 (GLP-1) and glucagon, and their respective action on β-cell GLP-1 receptors (GLP-1R) to regulate insulin secretion.

α and β Cells in Type 2 Diabetes Mellitus: From Adaptation to Dysfunction

Obesity and the development of insulin resistance impacts metabolic function of most organ systems. However, when we think of hyperglycemia and T2D, we largely focus on insulin resistance and defective β-cell function (ie, insulin deficiency). This is because the β-cell is often seen as the link between insulin resistance and consequent hyperglycemia. Insulin resistance in insulin-responsive tissues such as liver, skeletal muscle, and even fat requires increased circulating insulin for normal nutrient uptake into these organs (1-4). This requires more insulin output from the β-cell and thus increased insulin synthesis and β-cell mass (Fig. 1). In individuals in which β-cell “adaptation” to insulin resistance fails, both the insulin response to nutrients and β-cell mass are reduced, leading to hyperglycemia (Fig. 1). The field of islet physiology has mostly focused on both intracellular mechanisms leading to β-cell death and decreased insulin production/secretion. However, another important component in the regulation of glucose homeostasis by the islet lies within the α-cell. Notably, T2D patients have significantly elevated glucagon levels, yet the mechanism behind this hyperglucagonemia remains unknown (5). Consequently, hyperglucagonemia has been postulated to contribute to the hyperglycemia of T2D (6). As we will discuss in the following sections, glucagon is an insulinotropic hormone; therefore, rationalizing the impaired insulin secretion in T2D along with hyperglucagonemia seems paradoxical at first glance.

Figure 1.

The β-cells adapt to insulin resistance by increasing β-cell mass and function. When β-cells fail to adapt appropriately to the increase in insulin resistance, β-cell failure characterized by β-cell death and decreased insulin secretion leads to the development of type 2 diabetes and hyperglycemia. Figure created with BioRender.com

Preproglucagon Gene and Prohormone Convertases

The α-cell, like other islet cell types, is known by the main hormone it secretes, glucagon. Glucagon is a peptide encoded by the preproglucagon gene (Gcg), which contains the sequences of glucagon and 2 glucagon-like peptides, GLP-1 and GLP-2. In addition to the islet α-cells of the pancreas, Gcg is also expressed in the intestinal enteroendocrine L-cells of the gut and in specialized neurons within the nucleus of the solitary tract of the brain (7,8). Tissue-specific posttranslational processing of Gcg by prohormone convertases (PC) yield various proglucagon products. For instance, PC2, which is predominately expressed in the islet α-cell, produces glucagon (9). Conversely, PC1/3, expressed in the intestine and brain, cleaves proglucagon to yield GLP-1, GLP-2, and oxyntomodulin (10,11). Gcg encodes additional products such as major proglucagon fragment, glicentin-related pancreatic polypeptide, and glicentin, yet their functional significance remains unknown (12). However, some PC1/3 expression has been demonstrated in α-cells as well, suggesting that α-cells can produce GLP-1 in addition to glucagon (13). If both PC1/3 and PC2 are present within an α-cell, and given that the preproglucagon gene is constitutively expressed, then this would allow for production of both glucagon and GLP-1 within 1 cell (14). In the following sections, we will address the controversy surrounding α-cell production of glucagon and GLP-1 and detail the contributions of these peptides to glucose homeostasis.

Glucagon- and GLP-1–induced Regulation of β-cell Function

Glucagon

Despite being seen as a counterregulatory hormone that opposes the glucoregulatory action of insulin, glucagon has long been shown to stimulate insulin secretion (15,16). Intravenous administration of glucagon in humans results in a rise in insulin levels (15). In this clinical study and in isolated pancreatic sections, the impact of glucagon on insulin secretion was glucose-dependent (15,16). In fact, it has been elegantly shown that the glycemic set point of a species is based on islet glucagon content (17). Islet grafts from mouse, human, or monkey were implanted in nude mice with streptozotocin-induced β-cell ablation. The resultant glycemia reflected the glycemic set point of the species graft and its corresponding glucagon content. Further, administration of a human-specific glucagon receptor (GCGR) antagonist increased glucose levels in the nude mice that received human islet grafts (17). Altogether, these early studies demonstrated that glucagon could regulate insulin and, consequently, glucose levels.

Recent studies have also shown that the dose of glucagon (18) and duration of high glucagon exposure (19) are critical in the maintenance of glycemia and in the role of glucagon to stimulate insulin secretion. In fed mice, low dose of glucagon (20 μg/kg) increases blood glucose levels, independent of changes in insulin, while a high dose of glucagon (1 mg/kg) reduced blood glucose and increased insulin levels (18). Importantly, the nutritional status at the time of exogenous glucagon administration was the determining factor in the glycemic response. If the mice were fasted overnight and then challenged with exogenous glucagon, blood glucose levels increased without significant response in insulin levels. However, if glucagon was administered along with glucose, that rise in blood glucose levels was diminished by the concurrent rise in insulin. The dogma that glucagon always increases blood glucose levels has also been challenged by dissecting the short vs prolonged exposure to hyperglucagonemia in mice. Mice with constitutive activation of mammalian target of rapamycin complex 1 signaling (by deletion of tuberous sclerosis complex 2 in α-cells) have mild chronic hyperglucagonemia resulting from increased glucagon production and secretion from the α-cell, with no changes in systemic or α-cell GLP-1 levels. Despite the chronic increase in glucagon levels, these mice show normoglycemia and improved intraperitoneal glucose tolerance under chow and high-fat diet conditions (19). Moreover, when these hyperglucagonemic islets were transplanted into the anterior chamber of the eye of nude mice, glycemia increased transiently before returning to normal levels. The authors attributed these effects to the decreased hepatic GCGR expression that were likely driven by the chronically elevated glucagon levels. Therefore, it is important to point out the fine line of glucagon’s role in regulating insulin secretion vs its impact on stimulating hepatic glucose production.

An even more interesting aspect of this role for glucagon in stimulating insulin secretion is not through its GCGR (20,21). Independent studies showed that in isolated islets, glucagon increases insulin secretion under high glucose conditions, and these effects are mediated primarily by β-cell GLP-1R (20-23). Several groups have shown that the GCGR-knockout (KO) mice are protected against hyperglycemia even after total loss of β-cells (24-26). While one interpretation of these data is that glucagon’s hyperglycemic effect is mediated through the GCGR and its hypoglycemic effects are mediated by the GLP-1R, GCGR-KO mice have a compensatory increase in plasma GLP-1 suggesting it may be GLP-1R signaling, rather than loss of GCGR activation per se, that attenuates the development of hyperglycemia even during insulinopenic conditions (27).

A critical question is what the clinical implications of glucagon’s role as an insulin secretagogue are. Many studies have shown that glucagon is elevated and hyperglycemia does not suppress glucagon secretion in patients with T2D as it does in nondiabetic subjects (5,28,29). The cause of this has been speculated to be related to the impact of β-cell regulation of the α-cell. Pancreatic α-cells are exposed to high levels of insulin secreted from the β-cells, and insulin is a potent inhibitor of glucagon secretion and gene transcription (30-34). Thus, the hyperglucagonemia with T2D is suggested to be due to the diminished insulin release during hyperglycemia that paradoxically stimulates the release of glucagon or prevents glucose-induced suppression of glucagon (28,35,36). If we pull this together with the role of glucagon on stimulating insulin secretion as noted in the previous discussion, we could further speculate that glucagon’s hyperglycemic effect is predominating when β-cell function is generally compromised.

Glucagon-like Peptide-1

The classic paradigm of GLP-1 physiology is that nutrient-stimulated GLP-1 secreted from L-cells in the intestinal tract acts as a hormone binding to its G-coupled protein receptor (GLP-1R) on the β-cells within the pancreas to stimulate insulin secretion, an effect not seen at low blood glucose levels (37). In fact, this GLP-1-driven enhancement in insulin secretion is only exhibited postprandially and is known as the incretin effect (38). However, questions about this hormonal model have persisted as GLP-1 is rapidly degraded by the presence of dipeptidyl peptidase-4 (DPP4), a protease that renders the half-life of GLP-1 to less than 2 min (39). This has led to several questions about GLP-1 physiology across the years. One recent idea that has been proposed is that the α-cell provides an additional “local” source of GLP-1 that is critical for regulating glucose homeostasis.

Under normal conditions and as early as embryonic day 13.5, the rodent pancreatic α-cell contains both glucagon and GLP-1 (Fig. 2). These α-cells (also known as pro-α cells) express PC1/3 as well as PC2 and are thought to be immature and undifferentiated (40,41). However, several studies have demonstrated that α-cells isolated from human islets are capable of processing and secreting GLP-1 in culture (42,43). Despite these studies, skepticism remains in terms of whether pancreatic GLP-1 is biologically active. Critics have argued that the antibodies and immunoassays used for detection often lack the resolution to distinguish active GLP-1 from inactive GLP-1 (1-36 amide/37) (44). Other groups have used mass spectrometry and found that little active GLP-1 within the pancreatic islets (45). However, in studies where mice with expression of Gcg only in the pancreas were administered exendin (9-39), a potent and specific GLP-1R antagonist, both oral and intraperitoneal glucose tolerance was impaired (46). Even more, reactivation of intestinal Gcg, which restored circulating GLP-1 levels, did not restore the antagonist effects of exendin (46). Additionally, islets isolated from a mouse model of α-cell PC1/3 deficiency demonstrated impaired glucose-stimulated insulin secretion (47). Together these data suggest that the pancreas, but not the intestine, is the source of GLP-1 required for GLP-1R–mediated insulin secretion.

Figure 2.

Glucagon and derived glucagon-like peptide-1 (GLP-1) are present in pancreatic α-cells in early development. Glucagon (green) and GLP-1 (7-37) (red) in wild-type mouse pancreas from embryonic day 13.5 and 16.5, postnatal day 1/newborn and adult. Scale = 50 μm. Antibodies: glucagon (Abcam Cat# ab10988, RRID:AB_297642) and GLP-1 (Peninsula Laboratories Cat# T-4363.0050, RRID:AB_518978).

The pancreatic α-cell has been revealed to be highly flexible in determining the ratio of preproglucagon peptides, glucagon and GLP-1 under various metabolic challenges such as pregnancy and hyperglycemia/T2D (48-51). In fact, in a streptozotocin (a β-cell toxin) rat model of diabetes, α-cell PC1/3 was significantly increased leading to an increase in α-cell GLP-1 production (52). Additionally, knockout of PC1/3 in α-cells using an inducible CRE resulted in reduced expression of GLP-1 with severe impairments in glucose tolerance and insulin secretion when treated with streptozotocin (47). These data suggest that the increases in α-cell GLP-1 provides protective effects in instances of metabolic stress. It also suggests that the α-cell alters glucagon/GLP1 production based on the demands of the neighboring β-cell. However, emerging data suggest that like β-cells, there is heterogeneity among α-cells. One study in particular found that 70% vs 50% of all glucagon-expressing α-cells coexpress GLP-1 in healthy human and rodent islets, respectively (53). This brings into question whether production of these 2 peptides within the islet α-cell is in part due to a shift in PC2 to PC1/3 or is it a consequence of heterogeneity within α-cell populations?

In addition to its role in increasing β-cell insulin secretion, GLP-1 has long been established as an inhibitor of α-cell glucagon secretion (54). This is interesting considering our previous discussion that glucagon potentiates postprandial insulin secretion through β-cell GLP-1R (18,22,23), a condition whereby circulating GLP-1 levels are increased. So how do we reconcile these competing conditions of nutrient-stimulated but hormonally inhibited glucagon secretion? Given that glucagon seems to have opposing glucoregulatory action depending on whether it is acting upon hepatic GCGR or β-cell GLP-1R (21), 1 theory could be that local glucagon levels are restrained postprandially to maximize insulin secretion through activation of the β-cell GLP-1R by both glucagon and GLP-1.

The recent data suggesting α-cell GLP-1 and glucagon regulate insulin secretion and consequently glucose homeostasis have led to an argument as to which peptide is the critical physiological glucoregulatory α-cell product. The GCGR-KO mouse has improved glycemia, but some of this has been attributed to compensatory increases in plasma GLP-1 (27). α-Cell ablation using a variety of Gcg-targeted tools, has very little impact on insulin levels and glucose tolerance in vivo (46,55-57). However, nearly every mouse model of α-cell ablation report increased insulin sensitivity (46,47), which could mask the demand for insulin secretion. In support of this, mice lacking GCGR/GLP-1R in β-cells demonstrated no changes in glucose tolerance until the mice were challenged with obesogenic diet (58). However, this mouse still is devoid of both GLP-1 and GCGR signaling limiting our ability to determine their relative necessity for glucoregulation. At present, limitations in current genetic tools prevent the ability to address this question.

Glucagon and GLP-1 and Regulation of α and β Cell Mass

Glucagon

Interestingly, both glucagon and GLP-1 have demonstrated effects on regulating β-cell mass, but only glucagon has been found to regulate α-cell mass. For instance, GCGR-KO mice present with α-cell hyperplasia (59,60). These α-cells are thought to be immature due to the expression of the GLP-1R and GLP-1 (61-63). Similarly, blockade of glucagon signaling through use of glucagon-neutralizing antibodies (64,65) all resulted in α-cell hyperplasia along with improved glucose tolerance. Recently, a study inhibited the GCGR using a monoclonal antibody (mAb) in an inducible model of β-cell apoptosis, a murine model of T1D (66). In this model of T1D, treatment with the GCGR mAb led to glycemic improvements through increased β-cell mass (66). Surprisingly, GCGR mAb treatment resulted in α-cell hyperplasia and stimulated α- to β-cell conversion. This suggests that inhibition of the glucagon receptor potentially enhances β-cell mass. Interestingly, elimination of GCGR in the liver also results in α-cell hyperplasia establishing the liver as a critical organ for increasing the proliferation of α-cells (67). Based on metabolic profiling of GCGR-KO mice, it is now known that an increase in specific species of amino acids is what drives the α-cell hyperplasia (68,69). However, further investigation into the actions of glucagon on β-cell mass is needed.

Glucagon-like Peptide-1

GLP-1 has been shown to contribute to β-cell regeneration, differentiation, and neogenesis (70-72). Most of the beneficial effects of GLP-1 on β-cell mass are due to 3′,5′-cyclic adenosine 5′-monophosphate (cAMP)-mediated signaling, activation of the prosurvival cAMP responsive element binding as well as through activation of epidermal growth factor receptor 1, a tyrosine kinase receptor (73,74). cAMP responsive element binding and epidermal growth factor receptor pathways initiate increases in the expression of antiapoptotic genes (75) as well as functioning to reduce endoplasmic reticulum (ER) stress (76). Chronic (77) but not acute (78) treatment with liraglutide, a GLP-1R agonist, in diabetic mice led to increased β-cell proliferation and reduced β-cell apoptosis. ER stress, impaired proliferation, and increases in β-cell apoptosis have all been thought to contribute to islet β-cell dysfunction in T2D. That being said, it has been demonstrated that like GCGR mAb, treatment with GLP-1 increases β-cell mass by promoting α- to β-cell transdifferentiation (79). In this study, the authors found that α-cell mass was increased by GLP-1 treatment and that the observed increase in β-cell mass was in part due to preexisting α-cells that produce insulin (79). Overall, these data demonstrate the impact of glucagon and GLP-1 signaling on β-cell mass and support that pharmacological targeting these receptors may offer β-cell protection. Furthermore, manipulation of GLP-1 or glucagon signaling through use of genetic mouse models or receptor antagonists has led to enhancements in glycemic control by α-cell hyperplasia or α- to β-cell conversion (79,80). What exactly does this mean in terms of creating therapeutics for T2D? It suggests that α-cells can serve as a source for β-cell regeneration. In the islet biology field, there has been a push toward identifying methods of inducing transdifferentiation to generate more functional β-cells. For example, several studies have demonstrated similar α- to β-cell conversion through manipulating various transcription factors (ie, Pax4, MafA, Pdx1, Arx, and Dnmt1) (81-85). Therefore, pharmacological targeting of GLP-1R and GCGR agonism may have added benefits to slowing the progression of T2D by preserving β-cell mass.

While certainly GLP-1R agonists are beneficial for improving glucose control in T2D, the jury is still out as to whether the targeting of GLP-1R and/or GCGR will provide similar benefits for islet mass in humans as it does in mice. It is well known that β-cell turnover is minimal in humans (86), which suggests that incretin therapies like GLP-1R and GCGR agonists will not deliver the increase in β-cell mass as exhibited in rodents. However, the data that suggest poor β-cell turnover in humans are based on islets procured from cadavers. These islets are heterogenous in terms of size and cell composition, which may confound the ability to quantify and thus determine whether GLP-1R and GCGR agonists can modulate human β-cell mass (66). Thus, additional studies are warranted to determine whether the pharmacological targeting GLP-1R and GCGR will increase β-cell mass in humans.

Pharmacological Applications of Glucagon and GLP-1

GLP-1 has been successfully targeted to treat T2D and long-acting agonists are successful in treating obesity as well. To date, there are 2 major drug classes for GLP-1 associated drugs, they are DPP4 inhibitors (DPP4i) and GLP-1R agonists (GLP-1RA). DPP4i increase active GLP-1 by preventing its degradation in the circulation (87). GLP-1RAs, on the other hand, are drugs based on the peptide sequence of GLP-1 and work through mimicking the actions of endogenous GLP-1 but with a much longer circulatory half-life. GLP-1RAs are designed to have longer half-lives of endogenous GLP-1 so that injections are once a day, and recent formulations have extended this to once weekly injections (88). GLP-1RAs have been demonstrated to cause weight loss and interestingly have a greater weight loss effect in non-T2D vs T2D patients (89). Preclinical work demonstrates that the central nervous system is required for the antiobesity effects exhibited by GLP-1RAs (90,91). A major limiting side effect of GLP-1RAs is nausea, and distinguishing the neural circuitry that drive mechanisms that elicit anorexigenic effects vs those that cause nausea is a critical future direction for enhancing the utilization of these drugs.

On one hand, the pharmacological targeting of glucagon signaling has proved to be essential in acute treatment of hypoglycemia in T1D and T2D patients. Based on the normoglycemia in rodent diabetic models with abrogated GCGR activity (26,92) and the hyperglucagonemia in T2D (5,6,93), glucagon antagonists have also been developed but have been limited by their hyperlipidemic effect.

Conversely, several recent studies have begun to explore the potential role for glucagon agonists in obesity treatment. This was due to data demonstrating that central glucagon administration in rodents inhibits food intake (94) as well as hepatic glucose production (95). This demonstrates the duality of the glucagon hormone. Central glucagon administration can decrease hepatic glucose production, ultimately preventing the overproduction of glucose, a stark contrast to the direct actions of glucagon on the liver (95). Additionally, glucagon increases energy expenditure and lipid oxidation in both humans and rodents (96), suggesting potential weight loss effects. Indeed, chronic treatment with a long-acting GCGR agonist in both db/db and dietary-induced obese mice stimulated weight loss through reductions of food intake and increases in energy expenditure (97,98). It should be noted that in these studies GCGR agonism also induced hyperglycemia, prompting further investigations on downstream pathways, mediating the GCGR agonist-induced weight loss. To this end, Kim et al found that farnesoid × receptor, a bile acid receptor in the liver is required for the antiobesity effects of GCGR agonists (98). These studies shed new light on targeting glucagon biology for obesity treatment.

Of course, a potential negative side effect of using glucagon agonists is hyperglycemia. A strategy that has been used to combat this is to combine a glucagon agonist with a hypoglycemia agent. Dual agonists such as the glucagon/GLP-1 combination have been shown to reduce body weight and fat mass along with improve glucose tolerance in rodent models, and this occurs to a greater extent than GLP-1R agonism alone (99,100). In addition to GCGR/GLP-1R agonists, there has been interest in creating coagonists with another incretin, gastric inhibitory peptide (GIP). Dual agonism of GLP-1R/GIP receptor, like GCGR/GLP-1R coagonism results in significant weight loss and improved glucose tolerance (101,102). Triagonism of GCGR/GLP-1R/GIP receptor in dietary-induced obese mice has also demonstrated significant weight loss and enhanced glucose tolerance (103). Even more remarkably, in unbalanced ratios where there is 3 times more GCGR than both GLP-1R and GCGR, the incretin agonists are adequate to combat the hyperglycemic effects of glucagon action while maximizing weight loss. Thus, it is clear that α-cell–derived agonists demonstrate great promise in enhancing glycemic control as well as aiding in weight loss.

Concluding Remarks

Recent advances in the field have shown that the α-cell and its preproglucagon products are important mediators of β-cell mass and function and thereby are important players in the development and treatment of T2D. Numerous studies have suggested that glucagon plays a critical role in T2D progression. T2D patients have an impaired glucagon responses to hypoglycemic conditions, elevated fasting glucagon levels, and postprandial hyperglucagonemia, the latter 2 being a particular problem in the face of inadequate insulin responses (93,104,105). However, there are many factors aside from hyperglucagonemia that contribute to the hyperglycemia (106). As we have demonstrated in the previous sections, hyperglucagonemia in T2D could be a compensatory response, rather than a driving factor of, metabolic distress. In this instance, we hypothesize that the increase in glucagon levels lead to the secretion of additional preproglucagon peptides such as GLP-1 that are able to combat the hyperglycemic effects of glucagon. Ultimately, the α-cell’s involvement in T2D progression is not fully understood for several reasons: (1) glucagon has been described as inducing hyperglycemia for decades, which would imply that glucagon and GLP-1 have opposing functions; (2) α-cells can produce both glucagon and GLP-1 during various metabolic states; and (3) glucagon acts on the GLP-1R to regulate insulin secretion, and separating the relative importance of each Gcg peptide is currently technically limited.

Finally, the traditional paradigm where the α-cell is notoriously known for being the villain and the β-cell is the hero in terms of T2D progression is rapidly evolving as emerging evidence has demonstrated the newfound importance of α-cell–derived Gcg products. Instead of the classical view whereby α- and β-cells are on opposing sides of a seesaw to regulate glucose levels, we should think of their relationship in terms of a merry-go-round, circular. In this scenario, α-cells produce glucagon and GLP-1 act in a synergistic fashion on the GLP-1R to regulate insulin secretion.

Additional Information

Disclosure Summary: The authors have nothing to disclose.

Data Availability

Data sharing is not applicable to this article as no data sets were generated or analyzed during the current study.