The incretin/glucagon system as a target for pharmacotherapy of obesity

Summary Obesity is a chronic, multifactorial, relapsing disease. Despite multicomponent lifestyle interventions, including pharmacotherapy, maintaining bodyweight loss is challenging for many people. The pathophysiology of obesity is complex, and currently approved pharmacotherapies only target a few of the many pathways involved; thus, single‐targeting agents have limited efficacy. Proglucagon‐derived peptides, glucagon, and the incretin hormones glucagon‐like peptide‐1 (GLP‐1) and glucose‐dependent insulinotropic polypeptide (GIP), represent attractive targets for managing obesity and metabolic disorders because they may have direct roles in multiple mechanisms including satiety, energy homeostasis, and lipolytic activity. Unimolecular dual and triple agonists targeting glucagon and incretin hormone receptors have been shown to promote bodyweight loss, lower glucose levels, and reduce food intake in animal models of obesity. Multiple dual receptor agonists are in clinical development for the treatment of obesity, including GLP‐1/GIP and GLP‐1/glucagon receptor agonists. The extent to which glucagon contributes to treatment effects remains to be understood, but it may promote bodyweight loss by reducing food intake, while concomitant GLP‐1 receptor agonism ensures normal glucose control. Further research is required to fully understand the molecular mechanisms of action and metabolic effects of both dual and triple receptor agonists.


| INTRODUCTION
The overwhelming increase in the prevalence of obesity and people who are overweight in recent years represents one of the greatest global threats to public health. Worldwide, the prevalence of obesity has tripled since 1975, with over 650 million adults affected in 2016. 1 Obesity is now recognized as a multifactorial disease, characterized by abnormal or excessive fat accumulation that presents a risk to human health. 2 Obesity (body mass index [BMI] ≥30 kg/m 2 ) and being overweight (BMI 25-29.9 kg/m 2 ) 3 are associated with several health conditions including diabetes, cardiovascular disease, some forms of cancer, musculoskeletal disorders (especially osteoarthritis), sleep apnea, asthma, gallstones, depression, and nonalcoholic steatohepatitis (NASH). 1,2,[4][5][6][7] Obesity is a complex, chronic, relapsing disease; weight gain can be progressive, occurring over many years, and weight loss is difficult to achieve and even more difficult to maintain. 2,8,9 In a meta-analysis of 29 studies, more than half of lost weight (56%) was regained within 2 years and 79% of lost weight was regained by Year 5. 9 Furthermore, some people with obesity do not consider themselves overweight, whereas others who do consider themselves overweight have no desire to lose weight. 10 Around one third of people with obesity would like to lose weight but have not tried to do so within the past year and half have tried to lose weight without consulting a healthcare professional. 10

| Current treatment landscape
Current guidelines for obesity management recommend determining the degree to which an individual is overweight or has obesity and, depending on the severity, applying multicomponent interventions. [11][12][13][14][15][16] Lifestyle modifications are recommended for all patients who require weight loss, whereas additional pharmacotherapy is advised for individuals in whom lifestyle interventions have failed. [11][12][13][14][15][16] Lifestyle modifications can include reduced energy intake (typically to achieve an energy deficit of ≥500 kcal/day), increased aerobic physical activity levels to ≥150 min/week, and behavioral change strategies to facilitate adherence to diet and physical activity (self-monitoring and reporting of dietary intake, physical activity, and weight measurements). [11][12][13][14][15] A variety of diets designed to reduce energy intake may successfully result in weight loss in adults who are overweight or affected by obesity. Meal plans including Mediterranean-style or vegetarian/vegan-style diets, which are higher in plant-based foods including olive oil (rich in monounsaturated oleic acid) and lower in processed food and meat than typical Western diets, may promote weight loss and cardiovascular benefits that are similar to those associated with low-fat diets (25%-30% of calorie intake from fat). 11,14 Notably, in the Dietary Intervention-Randomized Controlled Trial (DIRECT), a low-fat diet in people with type 2 diabetes (T2DM) elicited a lower mean weight loss (2.9 kg) compared with a Mediterranean (4.4 kg) or a low-carbohydrate (4.7 kg) diet. 17 Compared with the low-fat diet, the low-carbohydrate diet improved lipid profiles, whereas the Mediterranean diet decreased fasting plasma glucose levels in patients with diabetes. 17 A recent randomized controlled trial also showed that a 6-week low-carbohydrate diet, with high intake of protein and fat and energy intake adjustments to ensure weight stability, improved glycemic control and reduced liver fat content in patients with T2DM. 18 These observations suggest that it is not necessarily fat intake that is responsible for increased fat deposition. Intermittent fasting has also gained interest for the treatment of obesity and diabetes, and has been recommended to comprise regular periods of no or very limited calorie intake (<25% of calorie requirement); for example, a 16-h daily fast or a 24-h fast on alternate days or two nonconsecutive days in a week. 19 19 In studies of patients with concurrent obesity and T2DM, improved glycemic control was also reported with intermittent fasting. 19 With dietary interventions, most patients will reach a plateau in bodyweight loss at approximately 6-12 months, ranging from 3 to 12 kg, then will slowly regain weight over 2-5 years, with total weight loss reducing to 0 to 3-4 kg. 11,12 This pattern is most likely due to the progressive reduction of energy expenditure associated with bodyweight loss and the reduction of lean body mass. Therefore,  25 which was similar to that seen in trials of orlistat or the noradrenergic drug sibutramine. 26,27 Previous pharmacological agents approved for the treatment of obesity, including amphetamine derivatives, cannabinoid receptor blockers, and serotonin reuptake inhibitors, have been withdrawn due to their unfavorable adverse event (AE) profiles (Table 1). 22,28 Bariatric surgery is an option for individuals with a BMI ≥40 kg/m 2 or ≥35 kg/m 2 and with comorbidities for which appropriate nonsurgical methods have failed. [11][12][13][14][15][16]31 Roux-en-Y gastric bypass, often called gastric bypass, has traditionally been considered the gold standard bariatric procedure for weight loss. The underlying mechanisms are loss of appetite resulting in reduced food intake, most likely driven by the exaggerated secretion of gut hormones that occurs a few days after surgery. The increased secretion of these hormones, including glucagon-like peptide-1 (GLP-1) and peptide YY (PYY), is due to accelerated exposure and absorption of nutrients in the small intestine. [32][33][34] Changes in anatomy leading to mechanical restriction of food intake and malabsorption of macronutrients were originally thought to be responsible for weight loss following bariatric surgery.
However, these effects have since been found to be inappreciable, 34 except with less commonly used procedures such as jejunoileal bypass, biliopancreatic diversion, and duodenal switch, which dramatically reduce intestinal resorption of nutrients. The mode of action of gastric sleeve operations, now the most widely used procedure to treat obesity, 35 is not fully elucidated, but the accelerated passage of nutrients into the small intestine, which also leads to exaggerated gut hormone secretion, is thought to play a role. 36 Most surgical procedures are, in principle, irreversible and are not without complications 37 ; moreover, surgical intervention alone is unlikely to manage obesity in the majority of patients. Therefore, there is a large unmet medical need for a highly efficacious pharmacological agent with a favorable benefit-risk profile for the treatment of obesity, especially in chronically ill patients with concomitant disease (e.g., hypertension, T2DM, and chronic obstructive pulmonary disease).

| Rationale for targeting the incretin/glucagon system in obesity
Energy balance is maintained by an intricate network of interacting feedback mechanisms involving the hypothalamus, the brainstem, higher brain centers and, in the periphery, the stomach, gut, liver, thyroid, endocrine pancreas, and adipose (fat) tissue. 38 Hormones from peripheral tissues such as leptin, ghrelin, cholecystokinin, pancreatic polypeptide, PYY (PYY3-36), GLP-1, and oxyntomodulin have been shown to regulate appetite. [39][40][41][42][43][44][45][46][47] Resistance to the actions of some of these hormones appears to be associated with common obesity. For example, leptin is secreted by adipose tissue and is thought to be a key peptide in reducing food intake based on the extreme obesity that develops in the absence of leptin signaling. 38,48 However, people affected by obesity have chronically elevated leptin levels and are resistant to its anorexigenic effects 39,48 -this is thought to be caused, in part, by downregulation of a feedback loop by the high leptin levels. 49 Food intake is also regulated by the mesolimbic reward system and has been shown to activate some of the same circuits involved in drug addiction. 38,[50][51][52] T A B L E 1 Previous pharmacological agents approved for the treatment of obesity and the AEs resulting in their withdrawal [28][29][30]  The pathophysiology of obesity is complex and currently approved therapies for obesity only target a few of the many pathways involved; thus, single-targeting agents have limited efficacy. 22,53 An integrated approach to the treatment of obesity that targets multiple mechanisms such as feeding circuits, glucose metabolism, and energy expenditure is therefore assumed to be more effective than single-targeting agents. 53  However, because the peptide is also cleared by the kidneys, prolongation techniques have been developed to ensure lasting agonism.
For example, the GLP-1R agonist liraglutide is acylated and its acyl moiety (palmitic acid) binds to albumin, whereby the peptide survives in the circulation. 64 This agonist has been shown to effectively cause bodyweight loss in humans and experimental animals, in which sufficient levels of the natural peptide do not remain in the circulation to account for this effect. [65][66][67] Investigations using rat models demonstrate that liraglutide may cross the blood-brain barrier via the circumventricular organs (the area postrema, the subfornical organ, the choroid plexus, and the median eminence) and reach, for instance, the arcuate nucleus. 67 Here As GLP-1, GIP, and glucagon have related peptide sequences, it is possible to create analogues with agonist activity at more than one receptor type, for instance, combining GLP-1R agonist activity with the effects of glucagon and/or GIP. 61 Here, we discuss preclinical and clinical findings in obesity and other therapeutic areas of interest for glucagon, the endogenous incretin hormones GIP and GLP-1 and GLP-1R agonists, as well as their actions when combined as dual and triple agonists.

| GLUCAGON IN OBESITY
Glucagon is a pancreatic hormone, with receptors predominantly expressed in the liver. There also appear to be receptors expressed in the kidneys (although the localization is uncertain), while expression in the heart, adipose tissue, CNS, adrenal gland, and spleen is variable and may be species dependent (Figure 1). 66 Glucagon regulates amino acid metabolism and is released from alpha cells following amino acid stimulation as part of the liver-alpha cell axis. [69][70][71] In addition, glucagon has long been recognized to regulate glucose homeostasis, counteracting the actions of insulin by stimulating hepatic glucose production (glycogenolysis and gluconeogenesis). 61 Glucagon, at least at pharmacological doses, may regulate lipid metabolism, energy expenditure, and food intake in multiple species. 54,58,[72][73][74][75][76] In humans, hepatic fat synthesis is suppressed after glucagon administration. 54 Glucagon stimulates beta-oxidation of fatty acids and inhibits the formation of malonyl-coenzyme A, the first intermediate of fatty acid synthesis. 77 However, the extent to which glucagon influences whole-body lipid metabolism, particularly in individuals affected by obesity, remains controversial. 58,77 In rodents, glucagon has been shown to stimulate lipolysis in adipocytes 78-80 ; however, glucagon receptor expression has not been successfully demonstrated in human adipocytes. 77 The potential lipolytic effect of glucagon in humans has only been shown in vitro and at concentrations much higher than physiological levels in plasma. 77 Glucagon may also increase energy expenditure by inducing thermogenesis in brown adipose tissue (BAT), as shown in humans and in animal models. [81][82][83] This thermogenic effect is thought to be mediated through activity of the sympathetic nervous system, given that inhibiting β-adrenergic activity impairs the ability of glucagon to increase energy expenditure. 84 However, the contribution of thermogenesis to overall energy expenditure remains unknown, and this effect may be too small to result in bodyweight loss. 75 In animal models, glucagon reduces food intake when administered peripherally and into the CNS. 56,66,85,86 Because of the extremely short half-life of glucagon in rodents, 87 long-acting glucagon analogues are likely to be more effective.
Glucagon infused into the hepatic portal vein reduces spontaneous meal size in rats. 85 Conversely, infusion of anti-glucagon antibodies into the hepatic portal vein increases spontaneous meal size in rats. 85,88 These observations have led to the suggestion that glucagon may act in the liver to generate a satiety signal that is relayed to the brain via the hepatic branch of the vagus nerve. 85  Hepatic steatosis can lead to glucagon resistance, wherein glucagoninduced amino acid metabolism is impaired, causing elevated plasma levels amino acids and hence also glucagon. 71 Indeed, it may be that among patients with T2DM, those with nonalcoholic fatty liver disease (the vast majority) and hyperaminoacidemia also have hyperglucagonemia. 92 This disruption of the liver-alpha cell axis is mainly due to the accumulation of intrahepatic lipid, and may contribute to the development of T2DM, rather than being a consequence of it. 71 effect. 123 The incidence of treatment-emergent AEs was higher across all doses of cotadutide compared with placebo and liraglutide, with GI disorders being most commonly reported. 123 In overweight individuals without diabetes, dual GLP-1/glucagon infusion increased energy expenditure to a similar degree as glucagon alone; however, the addition of GLP-1 reduced the hyperglycemic effect of glucagon. 124 Dual GLP-1/glucagon infusion has been reported to significantly reduce food intake (À13%, p < 0.05) compared with similar doses of GLP-1 and glucagon administered separately, although patients reported postprandial nausea and some vomiting. 124 A trend towards increased pulse rate was also seen with dual GLP-1/glucagon infusion compared with placebo or GLP-1 alone, although no substantial change in blood pressure was recorded. 81 Thus, concomitant GCGR and GLP-1R activation provides the beneficial effects of glucagon (i.e., maintaining a significant reduction in food intake with little effect on plasma glucose levels; Figure 2). 81,124 6 | DUAL GLP-1R/GIPR AGONISTS Although the lipogenic potential of GIP alone is under debate, coactivation of GLP-1R and GIPR is an attractive prospect in the treatment of T2DM and perhaps obesity ( Figure 2). 125 For example, GIP analogues that do not alter bodyweight when administered alone to mice with diet-induced obesity were found to enhance GLP-1-induced weight loss, reduce food intake, and prevent fat mass accumulation 126,127 ; however, similar results have also been obtained with GIP antibodies. 128 The dual GLP-1R/GIPR agonist tirzepatide (LY3298176) has been shown to improve insulin sensitivity independently of GLP-1R-induced weight loss in Glp-1r-null mice (i.e., via GIPR antagonism), but whether this effect is present in humans remains to be seen. 129 Furthermore, a balanced unimolecular GLP-1R and GIPR agonist reduced bodyweight, food intake, and fat mass in mice with diet-induced obesity to a greater extent than liraglutide. 127 Although the exact mechanisms of GLP-1/GIP synergism are unclear, it has been hypothesized that GIP could act directly via the CNS by inhibiting food intake, enhancing the anorexigenic action of GLP-1, or increasing tolerability to GLP-1R agonists. 130 Dual GLP-1R/GIPR agonism has also shown efficacy in humans. In a Phase II trial of tirzepatide, more individuals with T2DM achieved weight loss of ≤5% and ≤10%, and glucose control with the dual GLP-1R/GIPR agonist than with a GLP-1R agonist (dulaglutide) alone. 131 Decreased appetite (although desirable) was the second most common AE, with doserelated GI events being the most common but the majority being transient and mild to moderate in severity. 131 In the Phase III SURPASS-2 trial, treatment with tirzepatide was superior to semaglutide at reducing bodyweight in patients with T2DM at all tested doses (5, 10, or 15 mg), with 34%-57% of patients receiving tirzepatide experiencing bodyweight reductions of ≥10%, compared with 24% of those receiving semaglutide (1 mg). 132 In the Phase III SURPASS-3 trial of tirzepatide (5, 10, or 15 mg) in individuals with T2DM (with or without metformin and/or an SGLT-2 inhibitor), bodyweight reduction ranged from À9.8 to À15.2 kg. 133 The most commonly reported AEs in the tirzepatide arms were GI related and generally mild to moderate in severity, with up to $11% of participants in the tirzepatide arms discontinuing treatment due to AEs. 134

| TRIPLE GLP-1R/GCGR/GIPR AGONISTS
The synergistic actions of glucagon to reduce food intake and increase energy expenditure, GLP-1 to reduce calorie intake, and GIP to potentiate bodyweight loss may aid in the treatment of obesity (Figure 2).
The addition of both incretin components to glucagon appear to better mitigate the hyperglycemic action of glucagon compared with the presence of GLP-1 or GIP alone, allowing for greater glucagon dosing and therefore greater potential for weight loss. 135 In animal models of obesity, balanced unimolecular triple agonism proved superior to existing dual agonists and best-in-class monoagonists in reducing bodyweight and enhancing glycemic control. 136 In a murine model of diet-induced NASH and fibrosis, the triple combination of GLP-1R, GCGR, and GIPR monoagonists increased bodyweight loss, reduced liver triglycerides, and improved histological NASH disease activity score; weight loss was similar to that obtained with liraglutide alone, but histological NASH disease activity score was significantly improved (p < 0.01). 137   Glucagon and related peptides have a multitude of hormonal and metabolic effects and not all are desirable when targeting the receptors therapeutically. 66 Some unwanted effects are usually classified as GI, although it is likely that they are mainly due to interactions with central receptors. Whereas delayed gastric emptying may be sensed as fullness, one consequence of the interaction with area postrema receptors triggered by GLP-1 and glucagon appears to be mild-to-moderate transient nausea, 46,124,131 which has also been reported in studies of single GLP-1 agonists in patients with diabetes. 25,104 Additional GI AEs (vomiting and diarrhea) have been observed in trials of GLP-1R/GCGR dual agonists. 122,131 Cardiovascular AEs are of potential concern, and a number of cardiovascular outcomes trials will be required as development continues, such as the ongoing SURPASS-CVOT of tirzepatide. 140 Completed trials of the GLP-1R agonists liraglutide, semaglutide, and dulaglutide have demonstrated superiority with respect to rates of adverse cardiac outcomes in comparison with placebo. 141-143

| CONCLUSIONS
Obesity is associated with a considerable and progressive disease burden, and an effective pharmacological intervention is lacking.
Glucagon is an attractive target for bodyweight management in individuals with obesity due to its ability to reduce food intake and stimulate energy expenditure, potentially without cardiovascular AEs.
However, its action may need to be counterbalanced by concomitant Further research is required to fully understand the molecular mechanisms of action that underpin the efficacy of both dual and triple receptor agonists and the respective metabolic effects.