5-HT2C Receptor Stimulation in Obesity Treatment: Orthosteric Agonists vs. Allosteric Modulators

Obesity is a substantial health and economic issue, and serotonin (5-hydroxytryptamine, 5-HT) is an important neurotransmitter system involved in the regulation of body weight. The 5-HT2C receptors (5-HT2CRs), one of 16 of the 5-HT receptor (5-HTRs) subtypes, play a significant role in food intake and body weight control. In this review, we focused on the 5-HTR agonists, such as fenfluramines, sibutramine, and lorcaserin, which act directly or indirectly at 5-HT2CRs and have been introduced into the clinic as antiobesity medications. Due to their unwanted effects, they were withdrawn from the market. The 5-HT2CR positive allosteric modulators (PAMs) can be potentially safer active drugs than 5-HT2CR agonists. However, more in vivo validation of PAMs is required to fully determine if these drugs will be effective in obesity prevention and antiobesity pharmacology treatment. Methodology strategy: This review focuses on the role of 5-HT2CR agonism in obesity treatment, such as food intake regulation and weight gain. The literature was reviewed according to the review topic. We searched the PubMed and Scopus databases and Multidisciplinary Digital Publishing Institute open-access scientific journals using the following keyword search strategy depending on the chapter phrases: (1) “5-HT2C receptor” AND “food intake”, and (2) “5-HT2C receptor” AND “obesity” AND “respective agonists”, and (3) “5-HT2C receptor” AND “PAM”. We included preclinical studies (only present the weight loss effects) and double-blind, placebo-controlled, randomized clinical trials published since the 1975s (mostly related to antiobesity treatment), and excluded the pay-walled articles. After the search process, the authors selected, carefully screened, and reviewed appropriate papers. In total, 136 articles were included in this review.


Introduction
Overweight and obesity are global epidemics affecting more than 2.5 billion people worldwide. Overweight and obesity are related to a body mass index (BMI) >25 kg/m 2 and >30 kg/m 2 , respectively, and result in substantial health and economic problems. Importantly, obesity is associated with several comorbidities, including chronic diseases, such as diabetes (type 2), cardiovascular pathologies (atherosclerosis, hypertension, and myocardial infarction), stroke, and different kinds of cancers. Thus, obesity is responsible for approximately 15% of human mortality [1]. Therefore, respective medications for obesity are important and are still being investigated in several laboratories.
Serotonin (5-hydroxytryptamine, 5-HT) is a key monoaminergic neurotransmitter in the central nervous system. Mammalian 5-HT is synthesized from the amino acid Ltryptophan by tryptophan hydroxylase to 5-hydroxy-L-tryptophan, and finally by aromatic amino acid decarboxylase enzymes. The 5-HT controls numerous physiological functions, such as thermoregulation, motor activity, wakefulness, cognitive function, and mood [2]. The 5-HT plays a crucial role in food intake modulation and body weight control. In

5-HT 2C Receptor Action in Food Intake Regulation
The hypothalamus is the most important brain region involved in the central control of feeding and energy expenditure. Within the hypothalamus, the arcuate nucleus controls food intake and metabolism [13]. Functionally, the arcuate nucleus neurons are divided into two antagonistic groups that express specific neuropeptides and elicit disparate physiological actions. The first group consists of orexigenic neurons that express neuropeptide Y and agouti-related peptide to stimulate appetite. The second, anorexigenic group of neurons suppresses appetite with pro-opiomelanocortin (POMC) [14,15]. The modulation of feeding behavior via local 5-HT signaling acts mainly by an anorexigenic effect on POMC neurons. They are controlled by excitatory projections from the raphe nuclei 5-HT-reactive neurons stimulate the POMC-expressing arcuate nucleus neurons [16].
Approximately 25% of adult mouse POMC neurons express the 5-HT 2C Rs, but activation of POMC neurons via their receptor has been shown to account for the anorexigenic effects of several 5-HT drugs [17]. It has been demonstrated that 5-HT 2C Rs expressed in hypothalamic POMC/cocaine amphetamine-regulated transcript neurons are required to control energy and glucose homeostasis, implicating POMC neurons as the target for the effect of 5-HT 2C R agonists on weight-loss induction and improved glycemic control [6,18]. Moreover, 5-HT 2C R activation in POMC neurons both induces the Pomc mRNA expression and increases POMC neuronal activity [19]. Notably, the 5-HT 2C R knockout phenotype in POMC neurons provokes hyperphagia, sensitizes mice to diet-induced obesity, and directly dysregulates glucose homeostasis [18]. Moreover, selectively reintroducing the 5-HT 2C R to POMC neurons in whole-body 5-HT 2C R knockout mice is sufficient to completely negate the knockout phenotype [20]. Finally, 5-HT 2C Rs re-expression solely in POMC mice neurons is sufficient to mediate the effects of 5-HT drugs on food intake [21].
Food intake is also regulated by the reward brain system (the mesocorticolimbic pathway) structures, such as the ventral tegmental area, nucleus accumbens, prefrontal cortex, hippocampus, and amygdala [22,23]. Changes in reward processing are hypothesized to play a crucial role in the onset and maintenance of binge eating. Binge eating disorder is related to recurrent episodes of frequently consuming unusually large amounts of food and feeling unable to stop eating. Binge eating disorder is also related to its frequent comorbidity, obesity, as well as complications of being overweight [24]. Similar to addictive substances, such as cocaine, food activates the ventral tegmental area-nucleus accumbens pathways and stimulates pleasure by increasing the release of dopamine [25,26]. Moreover, a subset of dopamine neurons of the ventral tegmental area has recently been demonstrated as a key stimulator of binge-like eating behavior in mice [27]. At the same time, 5-HT cell bodies, projecting from the dorsal raphe nuclei to the ventral tegmental area and the nucleus accumbens, and a subset of dopamine and gamma-aminobutyric acid neurons in the ventral tegmental area express the 5-HT 2C Rs [28,29]. Generally, the activation of 5-HT 2C Rs plays an inhibitory role in the regulation of reward-related behavior by inhibiting release in certain areas of the brain [30,31]. Activation of ventral tegmental area 5-HT 2C R-expressing neurons significantly reduces homeostatic feeding in mice [32].

5-HT 2C Receptor Agonists in Control over Food Intake Preclinical Research
Functionally, the 5-HTR agonists are drugs that orthosterically bind to one (=selective agonist) or more (=nonselective agonist) subtypes of 5-HTR. Direct or indirect (through additional mechanisms) 5-HTR activation by agonists provides a similar response to the intended activation by endogenous 5-HT (Table 1). In rodents, the 5-HT 2C R agonist drugs reduce food intake in a manner consistent with an enhancement of satiety by action in the 5-HT 2C R binding site in the brain hypothalamus. Table 1. Description of antiobesity drugs affinity to 5-HT 1-2 receptor (5-HT 1-2 R) subtypes, type of binding and mechanisms of 5-HT 2C receptor (5-HT 2C R) activation.

Piperazine Derivatives
The first reports showing that the stimulation of 5-HT 2C Rs is responsible for the reduction of food intake were based on experiments performed with piperazine derivatives (m-chlorophenyl piperazine; m-CPP and tri-fluoromethylphenylpiperazine; TFMPP), which are nonselective 5-HT receptor agonists (Table 1). These compounds reduced food intake in rodents, and such an effect did not appear in mice lacking 5-HT 2C Rs [7] or in rats pretreated with 5-HT 2C R antagonists [37]. It has also been shown that piperazine 5-HT receptor agonists (especially m-CPP) reduce hepatic glucose production and insulin tolerance, and both effects are independent of the reduction in food intake [38]. Other authors presented evidence that the effect of m-CPP and other compounds acting via 5-HT 2C Rs increases satiety, reduces food intake, does not affect energy expenditure, and decreases body weight [39]. It should also be emphasized that hypophagia, an effect depending on the activation of 5-HT 2C Rs, is also related to an increase in POMC release in the nucleus arcuatus of the hypothalamus and stimulation of melanocortin receptors in the paraventricular nucleus (PVH) [38,40]. In contrast to the above-mentioned piperazine derivatives, several compounds with different chemical structures that have been previously described are selective 5-HT 2C R agonists that reduce food intake in rodents. In fact, Org 37684, Ro 60-175, PNU 22394, VER 3323, YM 348, and WAY 163909 have been found to induce hypophagia, which is reduced or blocked by 5-HT 2C R antagonists [37,[41][42][43][44]. Interestingly, the hypophagia induced by 5-HT 2C R agonists seems to be selective. There is a lack of effect on water intake, and tolerance is not induced after chronic administration. Furthermore, some of these compounds (e.g., PNU 22394) appeared to be clinically active [37].

Fenfluramines
Acting at the presynaptic level, fenfluramine and d-fenfluramine are indirect agonists of 5-HT receptors (Table 1), among which the 5-HT 2C R are mostly required for the antiobesity effect. Accordingly, it has been demonstrated that, in 5-HT 2C R knockout mice, the anorexic effect of d-fenfluramine was not induced [21,45]. Furthermore, the hypophagic effect of d-fenfluramine was blocked by 5-HT 2C R antagonists [39,46]. Preclinical studies show that low (2.5 or 3 mg/kg) and high (10 mg/kg) doses of fenfluramine or d-fenfluramine reduced food intake and decreased body weight in rat models [47][48][49]. Moreover, in the case of cue-induced relapse, d-fenfluramine (3 mg/kg) blocked the reinstatement of lever pressing to food-seeking behavior in male rats [50].

Lorcaserin
A series of new 5-HT 2C R agonists, as derivatives of 3-benzazepine, were synthesized by Smith et al. [60]. The most potent and selective compound was (1R)-8-chloro-2,3,4,5tetrahydro-1-methyl-1H-3-benzazepine (APD 356, lorcaserin) ( Table 1). The authors performed in vitro (effect on turnover of [3H]phosphoinositol) and in vivo (effect on food intake and body weight after acute or chronic per os administration of lorcaserin) experiments. The results demonstrated its 5-HT 2C R selectivity (versus 5-HT 2A and 5-HT 2B sharing sequence homology) as well as a reduction in food intake after acute administration and reductions in both food intake and body weight after 28 days of treatment. The effect of lorcaserin on food intake has been examined in different animal models of obesity. Some authors used a regular diet [60,61], and other investigators evaluated lorcaserin in rats maintained on a high-fat diet [36,62,63]. Lorcaserin treatment rats (1-2 mg/kg) reduced the percentage of body weight by selective reduction of body fat mass [36,60]. Moreover, Xu et al. performed their experiments using binge-like eating and hunger-driven feeding in mice, where lorcaserin reduced food intake [27]. Lorcaserin also reversed the binge-like feeding observed following stimulation of the nucleus accumbens µ-opioid receptors and blocked nucleus accumbens µ-opioid enhancement of fat intake [64]. The simple antiobesity effect of lorcaserin may be related to the 5-HT 2C Rs location in brain structures associated with food intake control. It has been shown that the stimulation of such receptors expressed in POMC neurons induces anorexia [20,65], and the stimulation of 5-HT 2C Rs located on the brain patterning transcription factor single-minded 1 expressing neurons in the PVH increases food intake in mice [66,67].
Other studies showed a reduction in glucose consumption [61] and decreased binge food intake in rats [63] after lorcaserin (1 or 3 mg/kg) treatment. In addition, the abovementioned antiobesity activity of 5-HT 2C R agonists is related to increased satiety and decreased food intake. Such effects of lorcaserin can also be connected with its inhibitory effect on motivation and impulsivity [68]. In some experiments, the effect of lorcaserin on glycemic control has been shown. It reduced the production of glucose and increased the sensitivity to insulin. However, it did not affect the secretion of insulin. The above effects were independent of weight loss [69]. Importantly, single-dose pharmacokinetic studies have shown rapid absorption, high oral bioavailability, and a moderate half-life of lorcaserin [60].

Fenfluramines
At the beginning of 1970, fenfluramine and d-fenfluramine (an active dextro-rotatory stereoisomer of fenfluramine) were the first antiobesity drugs acting via the 5-HT system to be introduced on the market. In clinical trials, d-fenfluramine showed dose-and time-dependent weight loss (Table 2). In short-term treatment (3-6 months), a higher dose (30-60 mg) of d-fenfluramine led to greater weight loss than low doses (10-15 mg), compared to placebo patients. Long-term treatment (12 months or more) showed greater weight loss in d-fenfluramine patients compared to short-term treatment. Despite the efficacy of fenfluramine or d-fenfluramine in anorexic action, undesirable effects have been observed in obese patients. Table 2. Efficacy of 5-HT 2C receptor agonist properties in double-blind, placebo-controlled, randomized clinical obesity trials. The patient's inclusion criteria are increased body mass index (BMI; >25 kg/m 2 ) or ideal body weight (IBW; 120-180%). The mean weight loss (kg or %) of the drug group (D) is compared to the placebo group (P). Multicenter studies were marked by an asterisk (*). QD-once-a-day treatment; BID-two times-a-day treatment.

Drug
Dosage (  In several clinical trials, cardiovascular and heart valve disorders were observed. Echocardiography performed 12 months after the initiation of fenfluramine-phentermine therapy in twenty-four women demonstrated unusual heart valvular morphology and regurgitation in all patients. Moreover, 30% of them also had newly documented pulmonary hypertension [92]. Another study reported that, of 233 obese patients who took appetite suppressants containing d-fenfluramine, 12.7% of them had a prevalence of cardiac valvular insufficiency [93]. Moreover, after the echocardiography of 200 patients with a history of exposure to anorectic medications, significant aortic valve regurgitation was observed in 14% of patients exposed to d-fenfluramine [94]. Other authors showed occurrences of multivalvular disease and pulmonary hypertension after fenfluramine treatment [95][96][97][98]. Nine controlled studies of fenfluramine showed mild or greater aortic valve regurgitation in 9.6% of treated almost 3300 patients compared with 3.9% of 2000 control subjects [99]. Besides the cardiovascular problems, the most common events reported during d-fenfluramine (15 mg twice daily) treatment were diarrhea, asthenia, dry mouth, headache, and tiredness. Side effects were reported significantly more frequently in patients receiving dexfenfluramine 30 mg twice daily than in those receiving 15 mg twice daily [100].
In 1997, fenfluramines were withdrawn from clinical use due to adverse side effects, such as cardiac valvulopathy and pulmonary hypertension. The most likely explanation for fenfluramine-associated valvulopathy is the activation of 5-HT 2B Rs by norfenfluramine, being a fenfluramine metabolite. Fenfluramines bind weakly to 5-HT 2A Rs, 5-HT 2B Rs, and 5-HT 2C Rs. Norfenfluramine exhibited a high affinity for 5-HT 2B Rs, and 5-HT 2C Rs, and a more moderate affinity for 5-HT 2A Rs [33]. In cells expressing recombinant 5-HT 2B Rs, norfenfluramine potently stimulated the hydrolysis of inositol phosphates, increased intracellular Ca 2+ , and activated the mitogen-activated protein kinase cascade, the latter of which has been linked to mitogenic actions of the 5-HT 2B Rs. This 5-HT 2B Rs action is important for heart valvulopathy occurrence. Moreover, the level of 5-HT 2B Rs and 5-HT 2A Rs transcripts in heart valves was at least 300-fold higher than the levels of the 5-HT 2C R transcript, which were barely detectable [101].

Sibutramine
Sibutramine was approved by the U.S. Food and Drug Administration (FDA) in 1997 for weight loss in the short-and long-term treatment of obesity. Sibutramine has doseand time-dependent antiobesity effectiveness, as evidenced in clinical studies ( Table 2). Compared to placebo-treated patients, higher doses (15-30 mg) of sibutramine result in greater weight loss than low doses (10 mg). Moreover, sibutramine efficacy in reducing body weight in a long-term period (over one year) was greater than in short-term treatment (six months). However, the Sibutramine Cardiovascular Outcomes (SCOUT) trial showed that, at 12 months but not six months of sibutramine treatment, patients with marked lower baseline blood pressure had the tendency to increase [102]. In some other clinical studies, sibutramine did not increase blood pressure in obese patients [103,104]. Besides, patients taking sibutramine experienced a significant increase in heart rate [83,[105][106][107][108][109], but without difference in the overall status of the cardiac valves compared to placebo subjects [105,[110][111][112]. Otherwise, sibutramine's most frequently encountered adverse effects were constipation, dry mouth, headache, insomnia, and nausea.
In 2010, the FDA decided to withdraw sibutramine from the market as a consequence of increased primary outcome events (POEs), such as nonfatal myocardial infarction, nonfatal stroke, resuscitation after cardiac arrest, or cardiovascular death. Indeed, the risk of POEs was increased by 16% in the sibutramine treatment group as compared with the placebo [80]. Nevertheless, other results from SCOUTs did not show an overall deleterious effect of treatment but an increase in POEs in subjects with a history of cardiovascular disease and type 2 diabetes [113,114]. Unexpectedly, sibutramine therapy in obese or overweight high-risk patients induced significant mean reductions for all blood lipids [115]. Although sibutramine and its metabolites exhibit low affinity for 5-HT 1A-B Rs and 5-HT 2A Rs, these receptors are reported as a possible target for the causation of unwanted cardiovascular effects [116].

Lorcaserin
In 2012, the FDA approved lorcaserin on the market as a weight loss medication. The results of several clinical studies have shown its antiobesity effect (Table 2). Despite the time of treatment, lorcaserin (10 mg once and twice daily) caused greater weight loss than with a placebo. The most commonly reported adverse effects were headache, nausea, dizziness, and fatigue, but generally, lorcaserin was well-tolerated. In contrast, the unwanted effects of fenfluramines and sibutramine, in randomized, double-blind, placebocontrolled, multinational clinical trials of lorcaserin treatment, were not associated with any increase in cardiovascular events among patients with high cardiovascular risk [88]. Interestingly, lorcaserin treatment for 6-13 months showed an improvement in metabolic and cardiovascular parameters in obese patients [91,117,118]. It has also been reported that this drug decreases the risk of developing diabetes [119]. At the same time, compared to a placebo, lorcaserin did not affect blood pressure, heart rate, or heart valves [120][121][122].
However, the US Food and Drug Administration Adverse Event Reporting System (FAERS) database scrutinization identified that lorcaserin medication was associated with a significantly greater valvular disorder [123]. Only in overweight and obese patients with renal dysfunction, lorcaserin treatment barely increased cardiovascular disease risk [124]. At the same time, the results of several clinical studies have shown its modest antiobesity effects, and lorcaserin's position on the market slowly declined. Moreover, patients treated with lorcaserin were diagnosed with more frequent occurrences of cancer. Indeed, incidences of malignancies during lorcaserin pharmacotherapy (though not different from the respective placebo groups) were the reason that it was withdrawn as an antiobesity medication in 2020 [125][126][127].

Positive-Allosteric Modulators of the 5-HT 2C R
The 5-HT 2C R is not only an antiobesity drug target, but also a binding site for positiveallosteric modulators (PAMs). PAMs are groups of substances that increase the affinity and/or efficacy of respective receptor ligands (Figure 1). A potential PAM therapeutic role is the ability to potentiate the effect of an endogenous cognate ligand or other probes interacting orthosterically by binding at a distinct, allosteric, receptor recognition site [128]. Thereafter, several new molecules related to 5-HT 2C R PAMs were pharmacologically evaluated. Recently, there have been a few papers showing 5-HT 2C R PAM to be potentially safer antiobesity drugs. Garcia-Carceles et al. [129] screened a chemical library from Vivia Biotech using the ExviTech platform and validated and synthesized analogues of Compound 5 (VA240). Among 35 synthesized akins, the most interesting was Compound 11 (PAM 11: (N-[(1-benzyl-1H-indol-3-yl)methyl]pyridin-3-amine, VA012). This compound dose-dependently enhanced 5-HT efficacy (highest potentiation-by 35% at 10 µM-of 5-HT-induced inositol monophosphate release in 5-HT 2C R HeLa cells), displayed low binding competition with 5-HT or other orthosteric ligands, and did not exhibit significant off-target activities. Importantly, PAM 11 and WAY 161503 (an agonist of 5-HT 2C Rs used as a reference drug) were administered acutely to reduce food intake in rats, and PAM 11 acted with higher efficacy and with more prolonged action. Moreover, PAM 11-evoked food intake inhibition was not eliminated after pretreatment with the 5-HT 2A R antagonist ketanserin. These findings indicate that the effect of PAM 11 was not related to the activation of 5-HT 2A Rs and suggest that it has no direct action at the orthosteric site of the 5-HT 2C R. Furthermore, PAM 11 administered subchronically (seven days) in a restricted food access model also reduced both food intake and body weight gain in rats. Finally, after combined treatment with PAM 11 and sertraline (a 5-HT reuptake inhibitor) administered in a dose-induced mild feeding suppression, potentiation of feeding reduction was demonstrated. This is clear evidence for the allosteric potentiation of 5-HT-induced anorexia, especially 5-HT 2C R-induced anorexia [129]. Another study synthesized a series of N-heterocycle imidazole-linked phenyl cyclopropyl methanone molecules. One of three active 5-HT 2C R allosteric modulator molecules, Compound 58, showed unique and beneficial dual characteristics. Compound 58 selectively exhibited PAM response at 5-HT 2C Rs and negative-allosteric modulator at 5-HT 2B Rs, which is rarely reported for allosteric modulators. From the point of view of 5-HT 2C R agonist's unwanted cardiovascular effects (depending on 5-HT 2B Rs activation), the negative 5-HT 2B Rs modulation is a unique property. Besides, at the same dose, both Compound 58 and lorcaserin reduced food intake in rats [130].

Conclusions and Perspectives
This paper reviews the preclinical and clinical efficacy of 5-HT 2C R agonists in obesity treatment. Moreover, we showed evidence of new therapeutic molecules, based on 5-HT 2C R targets, as potential antiobesity drugs.
In preclinical studies, nonselective or more selective 5-HT 2C R agonists showed effectiveness in decreasing body weight, reducing food intake, and inhibiting food seekingbehavior. Consequently, 5-HT 2C R drugs, such as fenfluramines, sibutramine, and lorcaserin, were approved in the trade as obesity pharmacology medications. Clinical trials confirmed 5-HT 2C R agonist's action in reduced food intake and decreased body weight in patients with BMI >25 kg/m 2 or an ideal body weight of 120-180%. However, after a few years, the 5-HT 2C R agonist's position on the market slowly declined, until being withdrawn from the market due to unwanted effects (higher risk of cardiac valvular abnormalities) occurrence. Nonetheless, the above-mentioned drugs have contributed to understanding the mechanisms of 5-HTR action that permitted the design of new selective molecules. Herein, we showed that 5-HT 2C R PAM intensifies 5-HT action for reduced food intake and decreased weight loss in a few in vivo studies. Moreover, we presented selective 5-HT 2C R PAM without 5-HT 2B R-linked cardiac valve affinity that offers a new angle into the pharmacological potential safer in obesity treatment.
Nowadays, more attention is being paid to precision medicine that offers new grounds for obesity prevention and targeting correct treatment based on genetic, pharmacogenomic, and environmental factors [131][132][133]. Recent human studies underline the connectivity between 5-HT availability and BMI as a predictor of obesity treatment success [134]. Furthermore, the loss of 5-HT 2C R function due to mutation can predispose humans to obesity. Thus, sequencing genes encoding the 5-HT 2C R should be included in diagnostic panels for obesity [135]. Currently, a series of highly specific 5-HT 2C R PAM selectively increase in vitro 5-HT efficacy [136]. This elevates the possibility of 5-HT 2C R PAM being used in a more directed way as precision medicine.
In conclusion, we underline the role of 5-HT 2C R as a crucial therapeutic target in obesity treatment. Further, 5-HT 2C R PAM discovery and syntheses are recommended. Furthermore, 5-HT 2C R PAM optimization and clinical weight loss validation are necessary for successful antiobesity treatment and to reveal the full therapeutic potential of PAM.

Conflicts of Interest:
The authors declare no conflict of interest.