Among the glucagon-secretion modulators, a few gastrointestinal hormones deserve special mention. The first is GLP-1, a peptide of its own family primarily produced in intestinal L cells. A very small portion of GLP-1 is also produced and locally released within the pancreatic islets by alpha cells (13). The effect of GLP-1 on glucagon secretion is controversial; the presence of GLP-1 on isolated alpha cell cultures leads to enhanced glucagon secretion; however, many authors have proposed an inhibitory effect in isolated pancreas and GLP-1-infused patients. This finding suggests an indirect mechanism mediated by an intermediate factor. Supporting relationship, the GLP-1 receptor (GLP-1R) has been detected on the surface of pancreatic delta cells, and perfusion of somatostatin-14 (SST-14) in isolated rat pancreas led to diminished glucagon release by alpha cells (14).

GIP is another gastrointestinal incretin hormone produced by K cells, primarily in the duodenum. GIP is able to enhance glucagon secretion of alpha cells in rats, healthy people or patients with T2DM both in the hyperglycemic and euglycemic state. This effect appears to be direct and due to the binding of GIP to its receptor (GIPR) on the alpha cell surface (15) and resulting increases in intracellular calcium concentrations, leading to glucagon vesicle release (16). Some authors propose that loss of beta cell GIPR function due to cell stress in T2DM patients may reduce the inhibitory effect of insulin on glucagon secretion, leading to elevated GIP-stimulated glucagon secretion even during hyperglycemia (17), but this hypothesis is controversial.

The gastrointestinal peptides that act on glucagon secretion also include ghrelin, a 28-amino-acid peptide primarily produced by X/A-like cells in the oxyntic glands of the gastric fundus. A small amount of ghrelin is also produced in the pancreatic islets by epsilon cells (18). Ghrelin receptor (GHSR) has been detected in alpha cell populations, and a direct stimulating effect of acyl ghrelin on glucagon secretion was observed in a pancreatic alpha cell line clone in isolated islets (19). The exact mechanism of ghrelin-enhanced glucagon secretion remains unknown.

GLP-2 is another enterohormone encoded by the Gcg gene and processed by PC1/3 in intestinal L cells. GLP-2 has also been described inside the pancreatic islets, and a glucagonotropic effect occurs after intravenous administration of GLP-2 to patients (20). Thus, GLP-2 may act directly by binding to its receptors on alpha cells, leading to glucagon release. However, cross-activation of GLP-1R by other members of the preproglucagon family has also been described. Therefore, we cannot rule out activation of GLP-2R on the surface of alpha cells by GLP-1 or by the released glucagon, thereby serving as a mechanism to amplify glucagon secretion.

Another interesting peptide is oxyntomodulin (OXM), a peptide related to GLP-1 and GLP-2 obtained from proglucagon processing by PC1/3 in L cells along the intestine but also at the caudal part of the nucleus of the solitary tract in the Central Nervous System (CNS) (21). OXM stimulates glucagon secretion by binding to glucagon receptors (GCGR) with low affinity, as has been demonstrated in vitro and in animal models (22). In practice, OXM is rapidly degraded by the DPP-4 enzyme, leading to a small amount of circulating OXM and negligible activity in carbohydrate metabolism.

PYY is a molecule co-secreted with GLP-1 by intestinal L cells. Its mechanism of action on alpha cells remains unknown, but PYY has been shown to stimulate glucagon production in cultured islets from diabetic GK rats. More recently, PYY expression inside pancreatic islets from rodents and humans and GLP-1–mediated paracrine activity in islets have been documented. However, the sensitivity of PYY to degradation by DPP-4 activity and the limited quantities in which it is expressed makes studying PYY difficult (23).

The regulation exerted by the insulin–glucagon pair on carbohydrate metabolism is well known. After a meal, high blood glucose levels stimulate insulin release by pancreatic beta cells, leading to glucose uptake and glycogen production in the liver and muscles in a process called glycogenesis, and insulin also inhibits glucagon release.

Glucagon is secreted by the pancreatic alpha cell population in situations of hypoglycemia or stress, reaching blood concentrations of up to 1 ng/mL in such situations. When glucagon reaches the target tissues, it binds to glucagon receptor (GCGR), a G-protein-coupled receptor with seven transmembrane domains on the cell surface.

The effects of glucagon in the liver also suggest paradoxical situations, such as those related to protein synthesis in fasting periods through transient protein kinase mechanistic target of rapamycin 1/2 (mTORC1/2) activation thanks to the GTPase activity of RAP-1 aided by exchange protein activated by cAMP (EPAC). The increase in hepatic sensitivity to insulin induced by glucagon due to phosphorylation of AKT at Ser 473 is similar to the effect of insulin on mTORC2, which has been observed in primary cultures of hepatocytes and is of special therapeutic interest (24).

After a meal, high blood glucose levels stimulate insulin release from the pancreatic beta cell population to the bloodstream, inhibit the pancreatic secretion of glucagon and enhance glucose uptake and glycogen storage, primarily in the liver and muscles. These effects lead to decreased blood glucose levels, facilitating euglycemia. In contrast, during fasting periods of 2 to 12 hours, blood glucose level depletion is prevented by the action of glucagon secreted by the pancreatic alpha cell population. As mentioned above, this initially induces glycogenolysis in peripheral tissues and the liver, but when glycogen reserves are depleted, glucagon also induces hepatic gluconeogenesis, ensuring maintenance of proper blood glucose levels. However, the role of glucagon in hepatic gluconeogenesis is complex; a recent study proposed that the mechanism of gluconeogenesis is more dependent on the availability of gluconeogenic amino acids than on the action of glucagon itself (24). Glucagon increases the availability of gluconeogenic amino acids. In this sense, glucagon regulates hepatic glucose metabolism more by the glycogenolysis pathway than by gluconeogenesis. This relationship is supported by the poor hepatic glucose production after glucagon administration in patients with low hepatic glycogen reserves subjected to a low carbohydrate diet. Therefore, glucagon may lead to immediate regulation mediated by glycogenolysis, or medium-term regulation mediated by gluconeogenesis ( Figure 2 ).

Patients with type 2 diabetes mellitus (T2DM) exhibit high plasma glucagon levels in the fasting periods and lack the ability to inhibit its secretion after a meal. Historically, steps have been taken to ameliorate the disease. Diabetic patients were administered somatostatin, which limited hyperglycemia in most patients (25). Later, glucagon receptor (Gcgr -/-) knockout mice treated with streptozotocin also exhibited euglycemia without hypoglycemic episodes in an interesting study (26). However, in this model, worse glycemic control may develop if GCGR ablation is induced after the onset of STZ-induced diabetes. Moreover, this partial control of hyperglycemia was lost when the animals were treated with a GLP-1R antagonist. These findings lead us to consider the establishment of an alternative pathway to GCGR after its ablation, the role of GLP-1R in this alternative pathway and the need for time to establish the pathway. Since that time, several studies have tested the GCGR antagonist effect on glucose metabolism in animals and humans with T2DM, confirming beneficial antidiabetogenic effects (27). However, the use of GCGR antagonists as pharmaceuticals has drawbacks, such as a notable increase in hepatic transaminases and LDL cholesterol (28). Other ways of limiting the effect of glucagon on glucose metabolism that have been studied include the use of anti-GCGR antibodies, which is efficient but also exhibited many problems in preclinical trials, such as alpha cell hyperplasia in treated animals, a common problem for any agents that disrupts glucagon receptor signaling (29). Similar beneficial effects on glucose metabolism have been demonstrated by Gcgr-antisense oligonucleotide administration, which leads to inhibited Gcgr expression; only mild and local side effects have been observed when these oligonucleotides are used (30).

Paradoxically, glucagon stimulates dose-dependent insulin release via GLP-1R in beta cells in rodents (31). Considering the therapeutic role of GLP-1R agonists such as exenatide, liraglutide, or lixisenatide on glucose tolerance (32), it is easy to understand why the combined effect of glucagon/GLP-1 dual co-agonists on glucose and lipid metabolism is being widely studied (33) and why they might exhibit a superior and more complete effect than single agonists. Therefore, several glucagon/GLP-1R co-agonists have been developed with different affinities for the two receptors and varying effects on glucose tolerance in rats and mice, likely due to different glucagon receptor tissue distributions in each receptor. Despite this difficulty, a handful of these co-agonists, as LY2944876/TT-401, LY3305677 (Eli Lilly), SAR425899 (Sanofi-Aventis), Efinopegdutide (Merk), BI 456906 (Boehringer Ingelheim), Efocipegtrutide (Hanmi Pharmaceutical), JNJ-54728518 (Jannsen) or cotadutide (AstraZeneca), have moved to different phases of clinical trials (33–36) oriented toward the treatment of obesity rather than diabetes. Finally, a triple agonist of GIPR, GLP1R and GCGR called Retratutide has also recently been tested in a phase 2 trial. This showed a great ability to substantially reduce body weight. However, it also presented a series of adverse gastrointestinal events, although limited to patients treated with high doses (37).

The induction of postprandial hyperglucagonemia in patients undergoing RYGB has long been observed (39). However, other bariatric techniques that involve duodenal exclusion, such as the duodenal-endoluminal sleeve (DES), seem to stimulate hyperglucagonemia (40). This is a paradoxical situation given that RYGB also stimulates insulin secretion and limits high glucose levels. However, supporting these observations, long-term increased alpha cell mass has been observed in Wistar rodent models after RYGB (41), and elevated plasma glucagon levels have been observed after the oral glucose tolerance test (OGTT) but not after intravenous glucose administration in T2DM patients after RYGB (39). Furthermore, glucagon gene expression and glucagon expression are increased in patients after RYGB, as determined by small intestine biopsy, indicating new sources of alternative glucagon secretion. This finding may be explained by modification of the intestinal mucosa and glucagon gene processing by L cells due to early contact with food derived from RYGB surgery. However, the number of patients studied is small, and recent studies combining liquid chromatography and mass spectrometry techniques did not observe increased fasting or postprandial glucagon levels in T2DM patients one year after RYGB (42). These findings therefore raise suspicions of cross-detection of other proglucagon-like molecules, such as glicentin or OXM (43). It is necessary to establish the sensitivity and specificity of these techniques and determine which is the most suitable for this type of study. In this sense, a recent study has established a close correlation between glucagon levels measured using sandwich enzyme-linked immunosorbent assay (ELISA) in pancreatectomized patients and glucagon-related hormone levels such as OXM or glicentin. These findings lead us to consider a cross-reaction between molecules of the same family. Therefore, it is necessary to confirm the hyperglucagonemia detected after RYGB using other techniques, preferably LC-MS. Meanwhile, all data referring to the presence and amount of glucagon announced in the studies should be taken with great caution.

The important role that glucagon could play after RYGB is also not clear. Episodes of severe postprandial hypoglycemia are one of the most important and frequent complications following RYGB surgery. Some authors attribute this phenomenon to an excessive and inappropriate insulin response after the meal. However, the published data are inconclusive. . In addition, there is a C-terminal 19-29 mini-glucagon from the fragmentation of glucagon during its processing and released together with it. Mini-glucagon 19-29 has a potent limiting effect on insulin secretion (44, 45) and it could happen that RYGB modifies the released ratios of glucagon/mini-glucagon 19-29, which could explain the episodes of postprandial hypoglycemia.

At the same time, many authors have proposed that the high postprandial plasma glucagon level is the key to avoiding severe hypoglycemic episodes after meals in RYGB patients (46). However, Tharakan et al. proposed that the combination of early elevated plasma glucagon and GLP-1 levels after meals may imply an insulinotropic effect triggering postprandial hypoglycemia (47), building on previous studies in which these hypoglycemic states were limited using GLP-1R antagonists. Therefore, some GLP-1R antagonists, such as exendin (9–39), may be of special therapeutic interest. Thus, postprandial glucagon has a dual role. It acts first as a GLP-1 agonist, leading to an insulinotropic effect through GLP-1R, and subsequently stimulates the liver to avoid hypoglycemia as a safety mechanism. In this sense, some experiments performed in diabetic rodents suggest that the response to GLP-1R agonists prior to surgery may act as a predictor of the future effect of RYGB on glucose metabolism and are of significant therapeutic interest. Thus, the response to GLP-1 R prior to surgery would indicate the degree of sensitivity of the patient´s GLP-1R to GLP-1 and other similar molecules such as glucagon. Giving us an idea about the efficiency of surgery to improve glucose metabolism but also about the possibilities of postprandial hypoglycemia episodes after surgery (48).

Another possible explanation for postprandial hypoglycemia is glucagon resistance. Hepatic steatosis and certain diets have been related to decreased hepatic GCGR expression in animals, probably due to GCGR internalization (49). However, given the presence of liver fat in animals and the consumption of a high-fat diet by patients after RYGB, this situation does not persist and postprandial hypoglycemia can be seen even years after bariatric/metabolic surgery. RYGB also appears to increase circulating levels of GIP in diabetic-GK rats. This is of great interest considering the possible synergistic insulinotropic effects established with GLP-1 in cultured diabetic and nondiabetic human islets (50) and its demonstrated insulinotropic effect in T2DM obese patients. However, the role of GIP on glucagon secretion is not entirely clear. Many authors propose a GIP-stimulated glucagon secretion even during hyperglycemia, but this hypothesis is controversial since in Type 1 diabetes patients without insulin secretion, GIP infusion cannot stimulate glucagon secretion during hyperglycemia (51). Therefore, more research is needed in this field to clarify these events.

Previous studies by our group have demonstrated the fundamental role of PYY as a regulatory molecule for the increased secretion of both GLP-1 and GIP after RYGB in diabetic-GK rats when PYY exhibits secretion patterns prior to those of GLP-1 and GIP after ingestion (52).

Within the glucagon family of molecules, GLP-2 also appears to be increased after RYGB (53), which is consistent with high postprandial glucagon levels given the glucagonotropic effect of GLP-2. This relationship suggests a dual role for GLP-2 in maintaining glucose homeostasis indirectly by stimulating glucagon secretion and facilitating the growth and adaptive response of the intestinal absorptive loop (54).

Today, SG is the most common restrictive surgical procedure performed in the USA, with undeniable postsurgical benefits for glucose metabolism. Many studies have been conducted to elucidate how SG improves T2DM, implying modified bile acid circulation, elevated amino acid levels, and modified incretin expression (55). We will focus on those that relate changes in glucose metabolism after surgery with glucagon.

Elevated glucagon: insulin ratios have been reported in diabetic-GK rats after SG, and an early increase in serum glucagon levels has been reported in patients. These findings are consistent with what was observed in previous studies by our group reporting a long-term increase in pancreatic alpha cell mass in healthy rodents after SG with potent glucagon secretion capable of buffering exogenous insulin administration (56). This finding could explain the improved insulin secretion reported after SG because beta cell-expressing GCGR is activated by glucagon in rodents. However, beta cell-specific Gcgr knockout animals exhibit normoglycemia, suggesting that the insulinotropic activity of glucagon is mediated mainly through GLP-1R present in beta cell populations (57). Therefore, the elevated plasma GLP-1 levels observed after SG suggest a joint effect of both on insulin release after SG. However, T2DM patients’ serum glucagon levels decrease approximately 6 months after surgery, but the improved glucose metabolism persists (58). Could this phenomenon be explained by the paracrine activity of these hormones in the pancreas rather than by their circulating levels? Indeed, SG has been shown to modify the cellular composition and activity of pancreatic islets, even increasing the residual pancreatic ghrelin cell-producing population (59). Furthermore, a ghrelin-positive effect on glucagon secretion has been reported using cultured mouse islets and alphaTC1 and InR1G9 alpha cell lines. However, ghrelin Knockout and Ghsr(-/-) Knockout mice exhibit no changes in plasma glucose patterns after OGTT or IPTT, suggesting a limited paracrine effect of ghrelin on pancreatic islets. It is possible that this paracrine effect also occurs in the case of glucagon secretion. Nonetheless, it is difficult to investigate these results in humans, and further efforts are needed.

Finally, SG has also been correlated to increases in the circulating levels of GIP in diabetic rodents (60). However, this relationship has not been confirmed in patients. The effect can be understood based on the idea that the anatomical rearrangements of the digestive tube due to SG surgery do not affect the duodenum. Other authors have reported increases in GIP in the immediate postoperative period. However, these levels normalized 12 months later (61). This change indicates a probable adaptive phenomenon in the digestive tube. Together, these data rule out the possible insulinotropic synergistic role of GIP as proposed in the case of RYGB.