Approaches to the Pharmacological Treatment of Obesity

Victoria Salem; Stephen R Bloom


Expert Rev Clin Pharmacol. 2010;3(1):73-88. 

In This Article

Expert Commentary

Gut Hormones in the Treatment of Obesity

The discovery of leptin and gut hormones as major neuroendocrine regulators of bodyweight is leading the way to the development of attractive therapeutic approaches to the long-term manipulation of energy homeostasis in favor of appetite reduction and weight loss. It is hoped that this may be associated with a relative paucity of central or unexpected side effects. The rest of this article will concentrate on these therapeutic strategies.


Leptin is a circulating 167-amino acid peptide that is the product of the ob gene that is exclusively expressed in adipose tissue. Plasma leptin concentrations are highly correlated with adipose tissue mass.[51] Peripheral and intracerebroventricular injections of leptin reduce food intake and bodyweight in both wildtype and leptin-deficient (ob) mice, but not in leptin-receptor-deficient (db) mice.[52–54] The leptin receptor is expressed throughout the CNS, particularly in the hypothalamus.[55] It is now understood that by modulating neural activity in hypothalamic neurones that regulate energy balance, leptin acts in an important negative feedback loop to homeostatically maintain adiposity. Congenital leptin deficiency in humans is associated with massive early-onset obesity which is reversible with leptin replacement treatment.[56] However, the majority of cases of human obesity are associated with elevated circulating leptin levels and hence leptin-resistance, such that the utility of leptin as monotherapy for the treatment of obesity is limited.[57] More recent studies have investigated the synergistic use of leptin alongside other antiobesity agents to induce clinically significant weight loss in obese individuals.[58] Understanding the molecular mechanisms underpinning leptin resistance may open the way to other potential anti-obesity treatments. The leptin receptor signals via the JAK-kinase signal transducers and activators of transcription (JAK-STAT) pathway.[59] This type of signaling is limited by the activation of suppression of cytokine signaling (SOCS) proteins, which inhibit JAK kinase. Mice deficient in SOCS3 are resistant to diet-induced obesity and remain leptin sensitive[60] and in the future, chemical inhibitors of these proteins may have potential as anti-obesity drugs.

Peptide Tyrosine Tyrosine

Along with pancreatic polypeptide (PP) and NPY, PYY is a member of the PP-fold family of peptides – so named due to a common hair-pin fold motif necessary for receptor binding. PP-fold proteins mediate their effects through a set of G-protein coupled Y receptors – Y1, Y2, Y4, Y5 and Y6. The receptors are distributed widely in the periphery as well as the CNS, and there is a degree of cross-reactivity between all of the PP peptides at each of the receptors, producing an overlapping system that is difficult to untangle. However, there is evidence that each of the PP-fold peptides bind more selectively to specific receptor subtypes and have distinct effects on energy homeostasis.[61] PYY is released from L-cells of the gastrointestinal tract, in proportion to the number of calories ingested. Circulating levels rise approximately 15 min after a meal has been ingested and remain elevated for 2 h.[62] Fasted humans infused intravenously with PYY3–36 to mimic postprandial concentrations consume 35% fewer calories at a subsequent buffet meal.[63] Importantly, in contrast to the problem of leptin resistance, obese subjects remain just as sensitive to the anorectic effects of the hormone.[64] The prospective finding of a negative correlation between peak postprandial PYY concentrations and bodyweight suggests that altered PYY physiology has a role to play in the pathogenesis of obesity.[65,66] Furthermore, fasting levels of PYY are chronically elevated in a number of anorexia-driven weight-losing enteropathies in humans.[67,68] PYY knockout mice display hyperphagia and increased adiposity, which is reversed upon exogenous replacement of PYY3–36.[69] Some of the satiety effects of PYY are because it slows gastric emptying, the 'ileal brake' effect, leading to a more prolonged feeling of fullness. However, PYY released into the bloodstream more importantly exerts its appetite-regulatory effects in a neurohumoral manner along the gut–brain axis. The major site of action of circulating PYY3–36 is most likely to be the hypothalamus via the nearby median eminence. Studies reveal that c-fos expression (a marker of neuronal activation) is increased in the hypothalamus following the peripheral administration of PYY, which increases POMC and decreases NPY mRNA expression in the ARC.[64,70,71] Acute stress induces anorexia by downregulating arcuate NPYergic activity in a very similar manner.[72] Exploiting the anorectic action of PYY3–36 still holds promise for the development of novel anti-obesity therapies. Similar to the other satiety-inducing gut hormones, the most commonly reported adverse effect with PYY, which is dose related, is nausea. This probably represents part of the spectrum of the expected effects of these hormones, from a pleasant sensation of fullness through to feeling 'overfull' and eventually nauseous. Phase II clinical trials of nasal PYY (Nastech/Merck) taken with each meal were terminated after 24 weeks since it was no better than currently available sibutramine at reducing weight, and because there were significant problems with nausea.[208] However, in this study the maximum plasma PYY concentrations achieved of 105 pmol/l were much higher than normal postprandial levels of 50 pmol/l. Furthermore, the time taken to reach peak levels of 18–26 min was greatly shorter than the more gentle and sustained natural post-prandial rise.[73] Thus the high incidence of nausea and failure of nasal PYY to induce longer-term appetite reduction and weight loss may possibly have been due to the pharmacokinetics of this mode of delivery. Future studies investigating the chronic use of steady-state preparations of PYY are warranted. A more detailed understanding of the Y-receptor system and the highly complicated neuronal circuitry involved in the appetite regulatory effects of PYY is vital for the development of optimally targeted antiobesity drugs of this nature. At physiologic concentrations, PYY activates hypothalamic presynaptic (autoinhibitory) Y2 receptors, as evidenced by the lack of anorectic effect of PYY in Y2 receptor knockout mice, and the attenuation of the anorectic effect of peripherally administered PYY when a Y2 receptor antagonist is concomitantly given.[63] Recent functional magnetic resonance imaging studies in humans point to the fact that PYY may also interact with even higher brain centers, such as those involved in the reward processing of food, to alter meal preference and palatability.[74] In B6.V- Lepob/J obese mice under chronic stress (who have very high circulating NPY levels), there is evidence that pharmacological inhibition or fat-targeted knockdown of peripheral Y2 receptors is anti-adipogenic, reducing abdominal obesity and metabolic abnormalities,[75] highlighting the complicated interplay between the Y-receptor superfamily and the PP-fold peptides.

Pancreatic Polypeptide

Pancreatic polypeptide is synthesized and secreted from PP cells within the islets of Langerhans in the pancreas. PP is released postprandially largely under vagal control in proportion to calories ingested and levels remain elevated for 6 h.[76] Chronic administration of PP to rodents reduces food intake, increases energy expenditure and results in a loss of bodyweight.[77–79] PP-overexpressing mice display a lean and hypophagic phenotype, reversible with the administration of antibodies against PP.[80] In fasted healthy lean human volunteers, an infusion of PP mimicking normal postprandial levels reduces caloric intake over the subsequent 24 h period by 25%.[81] PP probably acts both via the vagus nerve and directly at the brainstem to produce its appetite regulatory effects via the Y4 receptor, for which it shows particularly high affinity.[61] The anorexigenic effect of intra-peritoneally administered PP in rodents is abolished by vagotomy.[78] Direct injection of PP into the dorsal vagal complex (DVC) of rats stimulates efferent vagal activity, suggesting that circulating PP also acts directly on the brainstem.[82] This occurs via the nearby area AP, which is rich in Y4 receptors.[83] The hypothalamus has also been demonstrated to play a role in PP's appetite-modulatory effects. The Y4 receptor is expressed in both the ARC and PVN of the hypothalamus. Peripheral injection of PP results in a downregulation of hypothalamic NPY expression.[78] This may occur in direct response to circulating PP or secondary to incoming nervous signals from the brainstem. These initial insights into the satiety-inducing affects of PP and PYY have driven the development of synthetic analogs as potential anti-obesity agents. 7TM Pharmaceuticals (Copenhagen) have developed two such analogs. TM-30339 is a selective Y4-receptor agonist which demonstrated significant weight loss when administered chronically to diet-induced obese mice. Phase I/IIa trials are currently underway in healthy obese individuals over a 28 day study period to assess safety, tolerability and the effective dose to induce weight loss.[209] TM-30338, named Obinepitide, is a dual Y2-Y4 receptor agonist already shown to be safe and well tolerated in humans during Phase I/II clinical trials. In obese volunteers given once daily subcutaneous injections, the compound significantly reduced food intake for up to 9 h after dosing and is scheduled for further Phase II testing.[210]


Amylin (islet amyloid polypeptide) is a 37-amino acid peptide coreleased from pancreatic β-cells in a 1:100 molar ratio with insulin. In a similar fashion to other gut hormones, amylin's release is rapidly stimulated by nutrient ingestion, peaking at 60 min and remaining elevated for 4 h.[84] Amylin's effects are to potently inhibit glucagon release[85] and to slow gastric emptying.[86] It has no identified specific receptor of its own, and amylin mediates its effects via the calcitonin receptor, with tissue-specific sensitivities accorded by the differential expression of calcitonin receptor activity-modifying proteins.[87] Chronically administered amylin reduces appetite in humans.[88] The weight of evidence suggests that it acts directly at the brainstem. In rats, direct injection of amylin into the AP is strongly anorectic, while the reduction in food intake seen after peripheral administration is abolished with either NTS lesioning or direct injection of the amylin antagonist AC187 into the AP.[89,90] The endogenous form of amylin has the tendency to form amyloid fibrils which, with long-term administration, could deposit in tissues and cause organ damage. Amylin Pharmaceuticals have developed a nonamyloidogenic, stable, subcutaneously delivered analog called pramlintide, which is already licensed in the USA as an adjunct to insulin for the treatment of both Type 1 and Type 2 diabetes. They have also studied the effects of pramlintide as a weight-loss treatment in nondiabetic obese subjects, replicating the weight-loss benefits seen in diabetic cohorts.[91] Their second generation amylin analog, davalintide, has now been entered into Phase II clinical trials to study its safety, tolerability and efficacy as a weight-loss agent in healthy overweight and obese volunteers. Amylin Pharmaceuticals' combination treatment of pramlintide with recombinant human leptin, metreleptin, was developed following the observation in rodent studies that amylin increases leptin responsiveness in diet-induced obesity.[92] Phase II data is available for placebo-controlled testing of pramlintide and metreleptin alone and in combination. A total of 177 obese patients were studied following a 4-week lead-in period with either low-dose or high-dose injections of pramlinitide alone alongside dietary restriction. Over the entire 24-week study period, subjects treated with pramlintide/metreleptin lost 11.5 kg from baseline, significantly more than the 7.4-kg lost by subjects treated with metreleptin alone or 7.9-kg loss with pramlintide alone.[58] Nausea led to four withdrawals from the study during the lead-in period and one withdrawal in the pramlintide/metreleptin arm. Wider-scale clinical testing of the combination therapy is anticipated.

Glucagon-like Peptide-1

Glucagon-like peptide-1 is secreted from intestinal L-cells along with PYY and oxyntomodulin (OXM). It is rapidly released postprandially and levels remain elevated for several hours. Alongside OXM and glucose-dependent insulinotropic polypeptide (GIP), it is an incretin hormone which stimulates postprandial insulin release. In addition to its pancreatic glucoregulatory effects, the active form of GLP-1, GLP7–36amide, also inhibits food intake in a number of species when given centrally or peripherally. Chronic peripheral administration causes significant weight loss in mice, which is inhibited by the coadministration of the specific GLP-1 receptor antagonist exendin9–39.[93] Unexpectedly, the GLP-1 receptor knockout mouse displays normal food intake and bodyweight (although it is glucose intolerant).[94] However, as is true of many neuropeptide and receptor knockout strains, the lack of an obese phenotype is probably owing to developmental compensatory mechanisms. In fasted wild-type rats, direct intracerebroventricular (ICV) administration of GLP-1 strongly inhibits feeding, an effect that is abolished by the coadministration of exendin9–39. Furthermore, ICV exendin9–39 on its own doubles food intake in satiated rats, suggesting that endogenous GLP-1 physiologically regulates appetite.[95] In normal-weight humans, GLP-1 dose-dependently reduces appetite and acutely reduces food intake by 12%. These appetite inhibitory effects are preserved in obese and diabetic human studies.[96,97] Circulating GLP-1 is likely to exert its appetite regulatory effects both via vagal afferents and direct interaction with the brainstem and hypothalamus. GLP-1 administered either directly into the CNS or peripherally strongly induces c-fos expression in the PVN.[93,98] Peripheral administration in addition activates the brainstem. GLP-1 receptor mRNA is found in high levels in the hypothalamic ARC, PVN and supraoptic nuclei. A further observation is that vagotomy and lesioning to disconnect the brainstem from the hypothalamus attenuates peripherally delivered GLP-1's anorectic effects.[99] Short plasma half-life is a feature shared by all peptide gut hormones. In the case of GLP-1, the half-life is 2 min. Complete deactivation occurs following N-terminal cleavage by the enzyme dipeptidyl-peptidase (DPP)-IV.[100] Exendin1–39amide, a peptide purified from the saliva of the Gila monster Heloderma suspectum, was developed by Amylin Pharmaceuticals, in conjunction with Eli Lilly, into the drug exenatide. Better known by its trade name Byetta, exenatide is a long-acting GLP-1 agonist licensed as an adjunct to oral treatments for Type 2 diabetes. In addition to improved glycemic control, meta-analysis of clinical trials reveals an average weight loss of 2.13 kg in exenatide treated groups above placebo, and a 4.76-kg weight loss compared with insulin.[101] Eli Lilly have reported results from a 24-week, double-blinded, randomized, control trial of high-dose exenatide in obese, nondiabetic subjects as an adjunct to dietary and exercise intervention. Individuals who received exenatide lost 5.06 kg compared with a 1.61-kg loss in the placebo group. Only exenatide-treated subjects (9.6%) lost more than 10% of their bodyweight. As with other peptide hormone treatments, Byetta needs to be delivered by twice-daily subcutaneous injection, a mode that will be unfamiliar and unattractive to many. Amylin and Eli Lilly are seeking approval for their once-weekly formulation of exenatide LAR, which was demonstrated to produce significantly improved glycemic control compared with the twice daily version with similar weight-loss effects in the long term.[102] In early 2009, Alta Therapeutics Corp granted Amylin and Eli Lilly exclusive worldwide rights to develop transdermal exenatide using its Passport Transdermal Delivery System. A once-daily transdermal exenatide patch is now in Phase I testing for Type 2 diabetes. The most common side effect with exenatide is nausea. In comparative studies, the incidence ranged from 33 to 57% in exenatide groups compared with 0.4–9% in patients treated with insulin analogs.[103] In general, this side effect can be ameliorated by early dose titration and eventually mitigates with continued use. There is evidence to support the notion that the drug continues to exert its therapeutic benefits on both glycaemic control and bodyweight after side effects have subsided.[104] In October 2007, the FDA issued an alert that acute pancreatitis should be included on exenatide's product label precautions section. This followed 30 postmarketing reports of acute pancreatitis in patients taking Byetta. Among these were 22 examples of symptomatic improvement following discontinuation of the drug, six cases where the symptoms started at the point of dose escalation, and three examples of a resurgence of symptoms after the drug was reintroduced. It has been noted that the obese, Type 2 diabetic population is at an already increased risk of developing gallstone or hypertriglyceridemia-related pancreatitis. In fact, the drug's manufacturers have reported retrospective data negating an increased incidence of pancreatitis in patients taking Byetta.[212]

A second GLP-1 analog has recently been introduced: liraglutide (Victoza; Novo Nordisk). Liraglutide is 97% identical to the native human peptide. It is coupled to a 16-carbon fatty acid side chain, which promotes binding to albumin, thereby lengthening its half-life in a similar fashion to insulin detemir. It is argued that its 97% homology to the native human peptide reduces the drug's immunogenicity (around 40% of patients taking exenatide produce antibodies to it, although in practice this is rarely of clinical consequence). Victoza is already approved in the EU for the treatment of Type 2 diabetes, with evidence of improved glycemic control compared with exenatide and a similar weight loss profile.[105] NovoNordisk have reported results from their 32-week open-label extension study of liraglutide in obese non-diabetic subjects following on from the 20-week, double-blind, placebo-controlled Phase II study comparing liraglutide with open-label orlistat. At 52 weeks, high-dose liraglutide led to a 5.5–6.0-kg placebo-adjusted weight loss. A total of 75% of subjects receiving high-dose liraglutide lost greater than 5% bodyweight, and 35% lost over 10%. A 10% drop-out rate owing to side effects, largely nausea, is comparable to that seen with exenatide.[213] There was no increased incidence of pancreatitis in trial participants compared with controls. However, a slightly increased incidence of medullary thyroid tumors has been reported in two nonhuman species in clinically relevant liraglutide doses, which is worrisome.[214]

Other long-acting GLP-1 analogs in the pipeline include lixisenatide by Zealand Pharma(currently undergoing Phase III testing[215]); CJC-1134-PC from ConjuChem Biotechnologies Inc., which is a conjugate of exendin-4 and recombinant human albumin; albiglutide by GSK and taspoglutide by Roche. In a separate development, Emisphere have formulated orally administered PYY and GLP-1 using a sodium N-[8-(2-hydroxybenzoyl) amino] caprylate (SNAC) carrier. Early studies revealed that peptides delivered orally in this way are rapidly absorbed from the gastrointestinal tract and reach concentrations several-fold higher than those seen naturally postprandially. Oral GLP-1 (2-mg tablet) alone and the combination of oral GLP-1 (2-mg tablet) plus PYY3–36 (1-mg tablet) induced a significant reduction in calorie intake.[216]


Oxyntomodulin (OXM) is a 37-amino acid cleavage product of preproglucagon. It is released from intestinal L-cells in proportion to calories ingested, with circulating levels peaking at 30 min and remaining elevated for several hours. OXM's gastrointestinal effects are to inhibit gastric secretion, gastric emptying and pancreatic exocrine secretion.[106] When administered chronically to rodents, OXM inhibits food intake and reduces bodyweight.[107,108] Intravenous infusion in lean human volunteers reduces appetite without increasing nausea or affecting food palatability.[109] Caloric intake at a subsequent buffet meal is reduced by 19%. Chronic administration over 4 weeks to healthy overweight and obese volunteers resulted in a 2.3-kg weight loss compared with 0.5-kg weight reduction in placebo-treated controls.[110] Part of this weight loss is mediated via an increase in energy expenditure. Thrice daily subcutaneous administration of OXM to obese volunteers induced a 26% rise in activity-related energy expenditure over 4 days.[111] This finding is supported by pair-fed rat feeding studies – where rats receiving OXM display a greater reduction in bodyweight gain than identically fed, saline-treated controls.[112] No specific receptor has been identified for OXM. Its biological effects are thought to be mediated by the GLP-1 receptor, for which it displays a 50-fold lower binding affinity compared with GLP-1.[113] The anorectic effects of OXM are abolished in the GLP-1 receptor knockout mouse. As expected, OXM is a less potent incretin than GLP-1, although it is equipotent at reducing food intake.[98] This raises the possibility of interaction with other receptors or the existence of a hitherto unidentified OXM receptor in the CNS. In support of this, manganese-enhanced MRI scanning of mouse brains following the peripheral administration of OXM reveals neuronal activation in the ARC, PVN and supraoptic hypothalamic nuclei, whereas GLP-1 activates the PVN and ventromedial hypothalamus.[114] Thiakis Ltd., a spin-out biotechnology company from Imperial College London UK, has developed TKS1225, a long-acting synthetic analog of OXM as a potential anti-obesity agent. This was recently acquired by Wyeth Pharmaceuticals, and is currently in Phase I development.[216]


Cholecystokinin (CCK) was the first gut hormone demonstrated to have appetite regulatory effects, acting primarily to promote meal termination rather than induce longer-term satiety, largely via the vagus nerve.[115–117] Acute administration in both lean and obese humans reduces food intake, however there is a tendency to tachyphylaxis (an attenuation of the hormone's appetite-inhibitory effects) with repeated doses. In rats, continuous intraperitoneal infusion produces tolerance after 24 h.[118] Acute studies in rats reveal that CCK acts to initiate early meal termination but that this is ultimately compensated for by an increase in meal frequency.[119] GSK abandoned its CCK-1 agonist (GI 181771X) program after Phase II trials failed to produce significant weight loss, most probably owing to the development of tolerance in the chronic setting. Recent studies have suggested a synergistic weight loss effect in combination with leptin,[120] which may reopen doors to the development of another anti-obesity therapy.


Ghrelin was discovered after a search for the endogenous ligand of the GH secretagogue receptor which is widely expressed in the CNS, gastrointestinal tract, liver, pancreas, kidney and adipose tissue. Ghrelin is a 28 amino acid peptide principally secreted from the X/A-like cells of gastric oxyntic glands.[121] A post-translational octanoylation of ghrelin's third serine residue by the enzyme ghrelin O-acyltransferase (GOAT) is essential for receptor binding.[122] Ghrelin is the only known gut hormone that is a potent stimulator of appetite. Ghrelin secretion from the stomach correlates with hunger. Levels rise prior to meals, ultimately acting as a meal initiator, and fall afterwards in proportion to calories ingested.[123–125] With regular meal times, ghrelin spikes become entrained. In general, plasma ghrelin levels correlate inversely with bodyweight. However, in obese humans an attenuated postprandial reduction in ghrelin has been reported.[126] By contrast, diet-induced weight loss results in an increase in ghrelin levels, which would theoretically counter further attempts to calorie restrict and may in part explain why sustained dieting is so frequently unsuccessful. In rats, both peripheral and direct CNS administration of ghrelin acutely and potently increases food intake, and chronic administration produces weight gain.[127,128] In healthy lean humans, intravenous infusion increases buffet meal energy intake by 28%.[129] Ghrelin causes an upregulation of NPY and AgRP expression in the ARC.[130] Ablation of the ARC eliminates ghrelin's orexigenic effects.[131] However, c-fos expression is also observed in the brainstem following peripheral injection of ghrelin and there is also evidence for vagal involvement, since vagotomy attenuates its appetite stimulatory effects as well.[132] Interestingly, functional MRI studies in humans suggest that ghrelin may also modulate neuronal activity in higher brain centers which are involved in the reward processing of food, such as the orbitofrontal cortex.[133] It remains unclear as to whether this is mediated via upward projections from the hypothalamus. In the arena of antiobesity pharmacotherapy, GHS-R1a (ghrelin receptor) antagonists have been developed which acutely reduce food intake in lean, diet-induced obese and ob/ob mice, and chronic administration results in weight loss in the ob/ob strains.[134,135] Several pharmaceutical companies have ghrelin antagonist programs, but not all have yielded expected results. As an example, Ipsen Group's full GHS-R1a antagonist BIM-28163 actually increased food intake and bodyweight in vivo.[136] Attempts to develop drugs based on the effects of ghrelin have taken a few other turns. NOXXON Pharma AG has developed spiegelmers, stable L-enantiomer RNA-based aptamers, which irreversibly bind to and deactivate octanoylated ghrelin. NOX-B11 reduced food intake in rats and the related NOX-B11–2 caused weight loss when administered chronically to diet-induced obese mice.[137,138] Pfizer has been granted exclusive licence to take forward development of NOX-B11. Cyto Biotechnology AG created a ghrelin 'vaccine' using ghrelin immunoconjugates to induce an immune response to the endogenous peptide. Promising results in animal studies unfortunately did not translate to any weight loss in human Phase I/II trials. In the future, GOAT inhibition may provide a useful therapeutic target, since ghrelin is the only known protein to require octanylation for receptor activity.[139]


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: