Neurocircuitry underlying the actions of glucagon-like peptide 1 and peptide YY 3 – 36 in the suppression of food, drug-seeking, and anxiogenesis

circuits targeted by PYY 3 – 36 when administered independently or in combination with GLP-1 to curb the appetite for food or drugs of abuse like opiates, alcohol, and nicotine. Additionally, we also reviewed animal and human studies to assess the impact, if any, for GLP-1 and/or PYY 3 – 36 on mood-related behaviors in relation to anxiety and depression. As dual agonists targeting GLP-1 and PYY 3 – 36 may produce synergistic effects, they can be effective at lower doses and offer an alternative approach for therapeutic benefits while mitigating undesirable side effects.


Introduction
Obesity is a growing global epidemic affecting both men and women.Currently, approximately 2 billion people worldwide are overweight, of which 650 million people have obesity, as defined by a body mass index >25 kg/m 2 or >30 kg/m 2 , respectively (Boutari and Mantzoros, 2022).Obesity is a significant risk factor for other diseases including diabetes, cardiovascular disease, and many cancers (Mokdad et al., 2003;van Dis et al., 2009), thus effective weight management strategies to combat obesity may also mitigate the development and progression of other diseases (Gregg et al., 2012;Shah et al., 2009).Diet and exercise are the main mechanisms to manage body weight, and adopting healthy eating habits and physical activity are lifestyle choices needed to maintain a stable body weight.However, in lieu of medication, reducing caloric intake is most effective for weight loss, but the increased prevalence of inexpensive, ultra-processed foods (Zobel et al., 2016) and sedentary lifestyles (Healy et al., 2008) have made it harder to lose excess weight.Additionally, studies have found that individuals who manage to lose weight begin to regain it within nine months and completely regain the weight within one year (Machado et al., 2022).Therefore, in the modern obesogenic environment, medication that can facilitate and maintain weight loss is now in high demand.
Recent pharmaceutical successes have focused on gut hormones, especially those that are released when the stomach is full to slow gastric emptying and signal fullness or satiety to the brain.Glucagon-like peptide 1 (GLP-1) is the most notable gut peptide targeted for weight loss.While GLP-1 analogues were initially approved for treating Type 2 diabetes, they also produced weight loss.To date, two GLP-1 receptor (GLP-1R) agonists, liraglutide (branded as Saxenda, Novo Nordisk) and semaglutide (branded as Wegovy, Novo Nordisk) can now be prescribed for weight loss.Chronic subcutaneous administration of GLP-1R agonists like semaglutide (Garvey et al., 2022;Ghusn et al., 2022;Wilding et al., 2021) or liraglutide (Pi-Sunyer et al., 2015) for 20 weeks (Astrup et al., 2009) up to two years (Astrup et al., 2012) produced and maintained weight loss, but weight regain is imminent once the drug treatment is stopped (Rubino et al., 2021;Wilding et al., 2022).Semaglutide can elicit upwards of 15% weight loss after 6 months (Ghusn et al., 2022) for up to two years (Garvey et al., 2022;Wilding et al., 2021), while liraglutide resulted in up to 20% weight loss within 20 weeks (Astrup et al., 2009) for up to two years (Astrup et al., 2012;Pi-Sunyer et al., 2015).Interestingly, the magnitude of weight loss was related with comorbid diabetes, as weight loss was less prominent in individuals living diabetes (8-10% within 20 weeks) (Davies et al., 2021;Ghusn et al., 2022).
Recent pharmaceutical developments have focused on dual agonist moleculessingle molecules targeting two receptorsthat activate pathways by GLP-1 and that of another gut peptide to achieve synergistic effects that are greater than additive effects from the sum of effects when two analogues are co-administered.For example, the dual agonist tirzepatide used in brand name drug Zepbound (Eli Lilly) targets GLP-1R and glucose-dependent insulinotropic polypeptide (GIP) receptor and is more potent for weight loss as a dose of 15 mg of tirzepatide resulted in 5% more weight loss compared to 2.4 mg of semaglutide within 72 weeks in participants with (Azuri et al., 2023) and without (le Roux et al., 2023) diabetes.
In recognition of recent advances shared at the 13th International NPY-PYY-PP Research Conference and the robust impact of gut hormones on body weight regulation, this review will assess the capacity for another gut peptide, Peptide YY 3-36  ), to act as a candidate target for dual agonism with GLP-1.PYY is a member of the Neuropeptide Y (NPY) polypeptide family that is released by the same gut cells as GLP-1, and there is emerging preclinical research focusing on the development of a dual agonist for GLP-1/ PYY receptors to target obesity (Chepurny et al., 2018;Chichura et al., 2023;Milliken et al., 2021;Østergaard et al., 2021).
The current availability of effective weight loss drugs is exciting, but despite their efficacy at producing weight loss, their safety over the long term is still relatively unknown.Individuals taking semaglutide overwhelmingly report a loss of interest in food, particularly for energydense meals (Blundell et al., 2017) thus suggesting an overall state of demotivation.Given that food may elicit dependency through similar neurochemical mechanisms, we compared the impact of GLP-1, PYY 3-36 , or their combination in promoting demotivation towards food and drugs of abuse.However, while losing interest in food would be effective for weight loss, it can have undesirable impacts on quality of life as individuals lose pleasure in eating or cooking (Reiley, 2023), which may exacerbate underlying mental health links with obesity (Fulton et al., 2022).Furthermore, preapproved prescription drugs continue to be monitored for medical risks that may emerge only as the drugs are accessed by a larger and more diverse patient population (Ball et al., 2016).This includes ongoing monitoring of GLP-1R agonists after receiving voluntary reports of suicidal ideation (U.S. Food and Drug Administration, 2024), which had been associated with weight loss drugs (Christensen et al., 2007).We reviewed available preclinical data to determine whether demotivation may also impact mood-related disorders or their progression.Both GLP-1 and PYY may be synthesized within the brain (Daniels and Mietlicki-Baase, 2018), but in this review, we aimed to assess the implications of GLP-1-and PYY 3-36 -based medications on food intake (Section 2), substance use (Section 3), and mental health outcomes (Section 4), therefore we honed in on findings where these compounds were peripherally administered.

Systemic GLP-1R agonism stimulates vagal sensory afferents and acts in the brain to suppress food intake
Endogenous GLP-1 exists in two equally active forms: GLP-1 7-36 and GLP-1 7-37 (Schmidt et al., 1985), collectively referred to here as GLP-1.It is released from L-cells that extend from the duodenum to the colon (Jorsal et al., 2018) within minutes of nutrient ingestion (Sonne et al., 2014).Upon release, GLP-1 has a prominent role in reducing food intake and promoting satiety.GLP-1 (Schirra et al., 2000) and their analogues (Nakatani et al., 2017) can slow post-prandial gastric emptying by inhibiting peristaltic contraction from the antrum or distal stomach to the pyloris and then duodenum.The vagal afferent pathway via the dorsal motor nucleus mediates the inhibition of antro-pyloro-duodenal motility since vagal afferent denervation abolished the effect of GLP-1 on gastric emptying (Imeryüz et al., 1997).
In humans, acute intravenous administration of synthetic GLP-1 reduced meal size in normal-weight individuals (Flint et al., 1998) and those with obesity (Flint et al., 2001;Näslund et al., 1999;Verdich et al., 2001).However, GLP-1R agonists are most commonly administered subcutaneously, which revealed that GLP-1R agonists suppressed food intake in a dose-dependent manner (Verdich et al., 2001).The vagus nerve is a conduit between the gut and brain and plays a very important role in the anorectic effect of GLP-1, as intravenously administered GLP-1 administration did not suppress food intake in vagotomized rats (Abbott et al., 2005) or humans (Plamboeck et al., 2013).
GLP-1 also acts directly within the brain to regulate feeding (Fig. 1A).Intracerebroventricular administration of GLP-1 into the lateral ventricle (Turton et al., 1996) or third ventricle (Meeran et al., 1999;Schick et al., 2003) inhibited feeding in rats, but when GLP-1 was co-administered with a GLP-1R antagonist exendin-9, its anorectic effect was abolished (Meeran et al., 1999).GLP-1 might have a broad range of targets, as GLP-1R expression is widespread in the brain, including at brain structures implicated in the hedonic feeding circuit like the nucleus accumbens (NAc), dorsal striatum, and ventral tegmental area (VTA), as well as at structures in the homeostatic feeding circuit like the arcuate nucleus of the hypothalamus (ARC) (Baggio and Drucker, 2014).The dorsal vagal complex in the brainstem, including the nucleus of the solitary tract (NTS), area postrema, and dorsal motor nucleus of the vagus nerve (DMX) were also activated by peripherally-administered liraglutide (Washington et al., 2010) (Fig. 1).GLP-1R agonists that are delivered peripherally exert central effects by binding to GLP-1R at circumventricular organs like the area postrema that lack a blood-brain barrier (Rüttimann et al., 2009).GLP-1 can also bind directly to GLP-1R along medial hypothalamic structures like the arcuate nucleus (ARC) and paraventricular nucleus (PVN).While these hypothalamic nuclei are localized behind the blood-brain barrier, they have high capillary density to allow direct binding of some gut-derived peptides (Secher et al., 2014).

Peripherally administered GLP-1R agonists target the arcuate nucleus
The anorectic effect of GLP-1 is largely ascribed to GLP-1R expression in the ARC, which comprises two first-order neuronal populations vital for central energy balance.Namely, cells expressing proopiomelanocortin (POMC) signal satiety and suppress feeding while those coexpressing Neuropeptide Y (NPY) and agouti-related peptide (AgRP) promote feeding.Electrophysiological studies showed that GLP-1 directly activate POMC neurons and indirectly inhibit NPY/AgRP neurons by enhancing spontaneous GABAergic input to NPY/AgRP cells (He et al., 2019) (Fig. 1B).Intracranial infusion of a GLP-1R antagonist into the ARC abolished the anorectic effect of liraglutide that was administered peripherally (Secher et al., 2014).
The PVN was another hypothalamic region implicated in the anorectic effects of GLP-1 action (Secher et al., 2014), but the anorectic Y. Dumiaty et al. effects of peripherally administered liraglutide persisted despite the blockade of GLP-1R or when the PVN region was lesioned (Secher et al., 2014).Therefore, unlike the ARC and vagus nerve, the PVN may not be essential for the anorectic effects of GLP-1 treatment.

GLP-1R activation decreases the motivation to seek food
GLP-1Rs are also located in brain regions implicated in hedonic feeding and that encode the reward and motivation for food.Exenatide reduced food intake via GLP-1R independently of other hormonal or metabolic factors (van Bloemendaal et al., 2014).Participants receiving liraglutide and semaglutide consistently report feelings of reduced hunger and reduced incentive to eat thereby highlighting the significant impact of these drugs on appetite (Brindisi et al., 2019;Flint et al., 2001;Hanssen et al., 2021;Näslund et al., 1999).The striatum, orbitofrontal cortex (OFC), and insula are usually activated in response to viewing pictures of palatable foods, but brain activity at these structures were significantly reduced with exenatide administration (Grill, 2020;van Bloemendaal et al., 2014).
Rodent studies have consistently shown that GLP-1R-mediated reduction in food intake is in part due to a decrease in the motivation to seek food.Infusion of exendin-4 into the lateral and third ventricles decreased conditioned place preference for chocolate pellets and the motivation for sucrose under a progressive ratio schedule of reinforcement (Dickson et al., 2012).In the same study, infusions specifically into brain regions that are part of the mesolimbic reward pathway, including the VTA and NAc, recapitulated the effects of exendin-4 that diminished the motivation to work for food rewards (Dickson et al., 2012).GLP-1R activation in the NTS and brainstem can also decrease the motivation to work for food, but the influence of brainstem GLP-1R activation on food intake is linked to the mesolimbic pathway, which includes the VTA and NAc (Alhadeff and Grill, 2014;Richard et al., 2015).GLP-1R activation in the NTS influences dopamine (DA)-related gene expression in the VTA, specifically increasing the expression of the dopamine 2 receptor (D2R), which is known to decrease the activity of dopaminergic neurons.The upregulation of this receptor will then reduce the activation of dopaminergic neurons projecting to NAc in response to rewarding stimuli thus resulting in a suppression of food-related reward (Richard et al., 2015).
Palatable foods that are highly processed and calorie-rich can hijack homeostatic feeding signals to promote excess feeding.Some studies have suggested that GLP-1R agonists acting on reward pathways reduce food intake by shifting preference away from highly palatable and energy-dense foods.In rodents, GLP-1R agonists infusions into the VTA reduced the consumption of high-fat food but not of chow (Alhadeff et al., 2012).This was also seen in human patients taking GLP-1R agonists, as they report enhanced sensitivity and decreased pleasure for sweet and fatty foods (Brindisi et al., 2019).
Peripheral injection of PYY 3-36 decreased food intake and body weight in rodents (Batterham et al., 2002;Challis et al., 2003).Such anorectic effects were also reported in humans.Intravenous administration of PYY 3-36 reduced food intake by 30% in participants with obesity (Batterham et al., 2003;Degen et al., 2005), and this drop in food intake was accompanied by reports of increased fullness and reduced hunger (Sloth et al., 2007).Peripherally administered PYY 3-36 can bind directly to Y2 receptors in the circumventricular organs, area postrema, and NTS (Blevins et al., 2008) and lesioning these brain structures reduced the effectiveness of PYY (Baraboi et al., 2010).The vagus nerve is also important for the effect of PYY 3-36 on food intake since vagal ligation in rats abolished the anorectic effect of PYY 3-36 (Baraboi et al., 2010).
However, despite its promising effects on curbing food intake and body weight, PYY 3-36 administration has been associated with nausea and vomiting in human participants (Degen et al., 2005;Sloth et al., 2007).Additionally, PYY 3-36 is prone to proteolytic degradation, resulting in reduced effectiveness and shorter duration of action (Toräng et al., 2016).These challenges have hindered the development of obesity treatments using PYY 3-36 , but long-acting PYY 3-36 analogues that could be tolerated without nausea while reducing food intake in nonhuman primates (Rangwala et al., 2019) and mice (Jones et al., 2019) are in Fig. 1.Convergence of GLP-1 and PYY 3-36 pathways at the parabrachial nucleus suppressed appetite.A, GLP-1 and PYY are secreted from the distal intestine but act in the brain, especially at the hypothalamus and dorsal vagal complex, to signal satiety and attenuate food intake.B, The arcuate nucleus is the key hypothalamic region targeted by GLP-1 and PYY 3-36 .GLP-1 receptor activation stimulates anorexigenic POMC cells and also upstream GABAergic interneurons to dampen the excitability of orexigenic NPY cells, which are also inhibited by PYY 3-36 .GLP-1 and PYY 3-36 co-administration leads to the activation of multiple cell groups within the dorsal vagal complex, including the NTS, AP, and DMX.While GLP-1 receptor agonism can directly stimulate these cells, PYY 3-36 may disinhibit the NTS, AP, or DMV via an upstream GABAergic interneuron.C, Reduced GABAergic efferents and increased glutamatergic output from the arcuate combined with the increased glutamatergic drive from the dorsal vagal complex converge and drive the activation of the parabrachial nucleus (PBN), which projects to the amygdala to dampen appetitive behaviors.AP, area postrema; ARC, arcuate nucleus; DMX, dorsal motor nucleus of the vagus; DVC, dorsal vagal complex; GLP-1, glucagon-like peptide 1; PYY 3-36 , Peptide YY 3-36 ; NPY, Neuropeptide Y; NTS, nucleus tractus solitarius; PBN, parabrachial nucleus; POMC, proopiomelanocortin.

PYY 3-36 disinhibited ARC POMC cells to suppress food intake
In rodents, peripheral administration of PYY decreased Npy mRNA and increased Pomc mRNA expression in the ARC.PYY action increased the expression of the neuronal activity marker c-fos at POMC neurons (Challis et al., 2003) thus suggesting that PYY 3-36 elicited anorectic effects via POMC cells.Indeed, PYY 3-36 and PYY 3-36 analogue administration led to hyperpolarization of NPY neurons and increased spike firing at POMC cells, which was a result of disinhibition that was driven by the loss of GABAergic input from neighboring NPY cells (Batterham et al., 2002;Jones et al., 2019) (Fig. 1B).
It is perhaps important to note that the effect of PYY 3-36 on food intake may be dependant on its target site in the brain.Peripheral (Batterham et al., 2002) and intra-ARC (Batterham et al., 2002;Teubner and Bartness, 2013) PYY 3-36 administration is anorexigenic, but when administered into the third ventricle or PVN, PYY administration elicited orexigenic effects (Morley et al., 1985;Stanley et al., 1985).The orexigenic actions of PYY 3-36 are attributed to the activation of orexigenic Y1 and Y5 receptors, albeit with a lower binding affinity (Kanatani et al., 2000).This heterogeneity in the PYY 3-36 response adds to the complexity of PYY 3-36 actions and highlights the importance for PYY 3-36 -based targets to prioritize Y2 receptor activation.

PYY 3-36 activated hedonic brain circuits to promote satiety
Intravenous PYY 3-36 infusion reduced food intake and altered brain activity in several hedonic brain regions in human participants (Batterham et al., 2007).Batterham et al. (2007) assessed the change in human brain activity by functional magnetic resonance imaging (fMRI) in relation to caloric intake of fasting participants following PYY 3-36 infusion.When infused with PYY 3-36 , participants ate less calories when given a meal after their imaging session, and there was greater activation of the orbitofrontal cortex (OFC), which is well known for its role in reward processing, specifically to encode the value of a reward stimulus.Meanwhile, when participants were infused with saline, the change in brain activity related to feeding was primarily within the hypothalamus.Differences in hypothalamic to OFC activation following PYY 3-36 infusion indicated that PYY 3-36 can act as a satiety signal that switches the regulation of food intake from homeostatic to hedonic brain centers.The increased activity in the OFC was also correlated with a decrease in meal pleasantness thus showing a direct link between hedonic activation and a change in eating behavior and preference (Batterham et al., 2007).
In addition to the OFC, there are additional brain regions associating PYY 3-36 with reward-related neural circuits.Using blood-oxygen-leveldependent (BOLD) fMRI, fasted participants that were shown pictures of food elicited brain activity changes in regions associated with reward, including the OFC, amygdala, caudate nucleus, insula, NAc, and putamen (De Silva et al., 2011).Following PYY 3-36 administration, there was a reduction in BOLD signal across all six regions, with the most prominent changes in the OFC, insula, and left NAc (De Silva et al., 2011).Taken together, results from imaging studies showed that PYY 3-36 can confer hedonic control to suppress hunger and the motivation to eat.

GLP-1 and PYY 3-36 co-administration further decreased food intake and preference for high fat foods
GLP-1 and PYY 3-36 bear many similarities in suppressing food intake and weight gain.They both act in the ARC and can have indirect actions at hedonic brain regions to suppress food-motivated behaviors.Interestingly, the release of these two gut peptides was elevated after a meal in humans (le Roux et al., 2006) and rodents (Strader et al., 2005) that underwent Roux-en-Y gastric bypass (RYGB) surgery to treat obesity, thus they contribute to the efficacy of RYGB surgery for weight reduction.Obese rats that underwent RYGB surgery to induce weight loss showed enhanced GLP-1 and PYY 3-36 release after meal ingestion and had reduced consumption of previously-preferred high fat foods.While separate GLP-1 and Y2 receptor antagonists increased the preference for high fat foods, when co-administered together there was a further increase in the preference for fat (Dischinger et al., 2019;Haerting et al., 2023).This suggested that GLP-1 and PYY 3-36 co-administration act on their independent pathways to promote a negative energy balance.
GLP-1R agonists and PYY 3-36 analogues have also been coadministered in preclinical studies showing additive effects on feeding and body weight in humans (De Silva et al., 2011;Neary et al., 2005) and rodents (Neary et al., 2005).In some cases, the co-administration of GLP-1R and PYY 3-36 analogues have also resulted in synergistic effects that exceeded the additive effect of each agonist (Boland et al., 2021;Talsania et al., 2005).Such enhanced reduction in food intake seen with the co-administration of GLP-1 and PYY 3-36 could not be achieved with one agonist alone even when the agonist is applied at a much greater dose (Kjaergaard et al., 2019).
The co-administration of GLP-1 and PYY 3-36 analogues has been tested in human studies with promising anorectic outcomes of reduced hunger and increased feelings of fullness that together may lead to a loss of interest in eating (De Silva et al., 2011;Neary et al., 2005;Schmidt et al., 2014).GLP-1 or PYY 3-36 treatment in fMRI studies each showed decreased brain activity in brain regions known for their role in reward encoding, but interestingly, co-administering GLP-1 and PYY 3-36 produced an even greater reduction in brain activity at these reward-related regions.There was a lack of obvious differential activation pattern between PYY 3-36 and GLP-1, which suggested that these agonists acted on the reward centers through a common neural circuit (De Silva et al., 2011).
GLP-1 and PYY 3-36 may act synergistically to reduce appetite through independent mechanisms.At low doses, exendin-4 reduced food intake by acting on vagal sensory afferents whereas PYY 3-36 acts through the Y2 receptor (Talsania et al., 2005).However, the synergistic actions of GLP-1 and PYY 3-36 may also converge on neural pathways linking the brainstem and hypothalamus with the PBN to curb appetite.
The PBN is uniquely activated only by GLP-1 and PYY 3-36 coadministration but not by either GLP-1 or PYY 3-36 (Kjaergaard et al., 2019), therefore it is a putative site amplifying the actions of both peptides.PBN activation, especially at cells producing calcitonin generelated peptide (CGRP), can suppress appetite through projections to the amygdala (Campos et al., 2016;Mumphrey et al., 2016).The PBN can also receive projections from the area postrema (Ferguson, 1991) and NTS (Roman et al., 2016) to curb appetite, and these brainstem regions are responsive to GLP-1 and mediate its anorexigenic effects (Baraboi et al., 2011).Furthermore, the PBN integrates hypothalamic input and is innervated by GABAergic NPY/AgRP cells in the ARC (Campos et al., 2016), and this hypothalamic pathway to the PBN can be sensitive to both GLP-1 and PYY 3-36 .Both GLP-1 (He et al., 2019) and PYY 3-36 (Batterham et al., 2002) treatment can suppress the activity of GABAergic NPY/AgRP cells, thus their co-administration can disinhibit the PBN to increase its activity and c-fos expression (Fig. 1C).
Given the successes of GLP-1R analogues like Wegovy and Ozempic for weight loss (Wilding et al., 2021) and the potential synergistic effects seen by the co-administration of GLP-1 and PYY 3-36 or their analogues, the Y2 receptor is a plausible co-target for dual agonism with GLP-1R.Indeed, two recent dual agonists targeting GLP-1 and PYY receptors have been developed and show enhanced reduction in food intake compared to their mono-agonists (Østergaard et al., 2021;Milliken et al., 2021).

Opioids
Over 600,000 deaths every year are related to drug use, with 80% linked to opioid use and 25% to opioid overdose (World Health Organization, 2023).There are three main pharmacological treatments approved for opioid use disorder (OUD) to reduce opioid use, namely buprenorphine, methadone, and naltrexone (Substance Abuse and Mental Health Services Administration, 2021) but these treatments are associated with a high relapse rate and non-compliance, so new treatments to combat OUD are much needed.
Opioids bind to μ-opioid receptors on GABAergic interneurons in the VTA (Johnson and North, 1992;Ostrowski et al., 1982) to disinhibit DA VTA cells and increase DA to the NAc (Kosten and George, 2002).This increased DA release is associated with a euphoric and pleasurable feeling (Nestler, 2005).After chronic use of opioids, the μ-opioid receptors become less responsive to opioid stimulation and require higher levels of opiates to produce the same effect.Additionally, while the mesolimbic DA system is heavily involved in opioid addiction, other brain regions, especially the locus coeruleus and periaqueductal gray (Al-Hasani and Bruchas, 2011; Dang and Christie, 2012) are implicated to elicit withdrawal-like symptoms.Since the locus coeruleus and periaqueductal gray also regulate pain signaling, an individual undergoing opioid withdrawal can experience excruciating pain, especially as chronic opioid use had suppressed their pain signaling for a long time.For this reason, treatments that diminish withdrawal symptoms can be effective to prevent relapse (Kosten and Baxter, 2019).
Peripheral GLP-1R agonist treatment also decreased motivated seeking-like behaviors for oxycodone.Specifically, exendin-4 treatment suppressed opioid-induced hyperlocomotion compared to vehicletreated rats and decreased the number of lever presses for opioid selfadministration through operant conditioning (Zheng et al., 2019).In animal models of opioid relapse, GLP-1R activation attenuates the reinstatement of opioid-seeking behavior (Urbanik et al., 2022;Zhang et al., 2021Zhang et al., , 2020)).This suggests that the activation of GLP-1R could potentiate a multifaced treatment that promotes the cessation of opioid use and recovery from OUD.
OUD is mainly manifested by the hyperactivation of the DA VTA to NAc pathway.The NAc expresses GLP-1R (Merchenthaler et al., 1999), and NAc GLP-1R activation suppresses opioid-seeking by mimicking the output from DA VTA projections, as GLP-1R agonism in the NAc can suppress oxycodone self-administration (Zhang et al., 2020).At the VTA, opiates disinhibit DA VTA cells, and GLP-1 is thought to stimulate GABAergic input to DA VTA cells to circumvent opiate-induced disinhibition (Fig. 2A).One source of GABAergic afferents arises from the laterodorsal tegmental nucleus (LDTg), and GLP-1R activation at LDTg cells projecting to the VTA can attenuate cocaine-seeking (Hernandez et al., 2021).While the mechanisms of cocaine and opiates to drive DA VTA activity may vary, these findings delineate available circuits whereby GLP-1R activation can regulate DA VTA cells (Fig. 2A).

PYY 3-36 acts in the accumbens to attenuate opioid reinstatement
Opioid intake increases DA release from the VTA and increases locomotor activity (Kosten and George, 2002), which can be used to determine the efficacy of opioid administration (Santos et al., 2022).PYY 3-36 can decrease this hyperlocomotor activity induced by opiates like fentanyl (Santos et al., 2022) and apomorphine (Stadlbauer et al., 2014).Systemic PYY 3-36 administration did not attenuate voluntary, self-administered fentanyl intake, but it attenuated fentanyl reinstatement, as determined by lever pressing activity to seek fentanyl during a period of abstinence (Caffrey et al., 2023).The systemic effect of PYY 3-36 was mimicked by PYY 3-36 infusion directly into the VTA thus indicating that PYY 3-36 attenuates fentanyl reinstatement via the VTA (Caffrey et al., 2023).
PYY 3-36 can increase brain activity signatures in the VTA (Batterham et al., 2007), but while PYY does not stimulate the activity of DA VTA cells (Stadlbauer et al., 2014;West and Roseberry, 2017), it can regulate dopaminergic output at the NAc (Adewale et al., 2007) and lead to the activation of NAc cells (Stadlbauer et al., 2014).PYY 3-36 may act in the same way as NPY via the Y1 receptor to lower the reward threshold, which indirectly potentiates the rewarding effect of morphine at the NAc (Desai et al., 2013).Additionally, opioid self-administration can increase endogenous NPY release (Desai et al., 2013), and Y2 receptor expression (Adewale et al., 2007;Shaw et al., 2003) and activation at the NAc can further promote DA synthesis and release in the NAc (Adewale et al., 2007).As NAc cells can innervate NAc-projecting VTA cells that are not dopaminergic (Scofield et al., 2016;Xia et al., 2011), it is possible that PYY 3-36 acts via Y2 (and/or Y1) receptors at VTAprojecting NAc cells to lower the reward threshold or to attenuate reinstatement (Fig. 2B).

Alcohol
About 10% of the global population is diagnosed with alcohol use disorder (AUD) (Jerlhag, 2020), but the prevalence of AUD is projected to be much higher since not all individuals with AUD are diagnosed (The Centre for Addiction and Mental Health, 2023).There are three current pharmacological treatments approved in Europe (Stokłosa et al., 2023) and North America (Kranzler and Hartwell, 2023) and include naltrexone, acamprosate, and disulfiram.These pharmacological treatments significantly reduce alcohol intake, but their efficacy between patients is variable and may elicit adverse side effects like nausea, vomiting, dizziness, and fatigue that increase non-compliance leading to relapse (Maisel et al., 2013;Skinner et al., 2014;Stokłosa et al., 2023).
Alcohol intake elicits positive and negative reinforcement.The positive reinforcing properties of alcohol result in a reward stimulus and euphoric state and stem from GABA A receptor activation in the VTA leading to increased DA release at D1R NAc cells (Scofield et al., 2016).Furthermore, alcohol causes the release of β-endorphin (Hermann et al., 2017) to disinhibit DA VTA cells (Acquas et al., 1993;Koob et al., 1998;Spanagel et al., 1992) (Fig. 3A).The negative reinforcing properties elicit somatic and psychological withdrawal symptoms (Merlo Pich et al., 1995;Richter and Weiss, 1999) and stem from increased corticotropin-releasing factor (CRF) release and CRF receptor 1 (CRFR1) activation at the central amygdala (CEA) (Bajo et al., 2008;Cruz et al., 2012).Chronic alcohol intake sensitizes CEA neurons to CRF and exacerbates withdrawal-like symptoms (Roberto et al., 2010), which is the leading cause of alcohol dependence because it creates the need to selfmedicate and alleviate withdrawal symptoms (Logrip et al., 2011).
The CEA is a critical brain structure underlying alcohol withdrawallike symptoms or relapse (Roberto et al., 2021), and with the onset of alcohol-dependence or chronic alcohol intake, GABAergic transmission is elevated at the CEA (Roberto et al., 2004) such as via CRF signaling (Roberto et al., 2010).Within the CEA, ethanol increases glutamatergic transmission at the lateral part of the CEA (CEAl) (Melkumyan and Silberman, 2022;Silberman et al., 2015).As the CEAl projects to the medial CEA (CEAm), which then sends efferent output to downstream targets like the VTA and extended amygdala, the net actions of ethanol dampen CEAm output.Therefore, one key element in promoting recovery from alcohol use may be to reverse the elevation in GABA tone at the CEA to restore its function (Roberto et al., 2021).

Y2 receptor activation blocked ethanol-induced dysfunction at the central amygdala
Circulating serum levels of PYY are unaffected by alcohol consumption (Calissendorff et al., 2006), and specific actions of PYY 3-36 on alcohol intake is not known.However, it has been shown that NPY can act at the CEA to diminish alcohol intake (Gilpin and Koob, 2008), and Y2 receptor activation at presynaptic GABA terminals in the CEA can block the ethanol-induced GABAergic transmission in the CEA seen following chronic alcohol intake (Gilpin and Roberto, 2012).Therefore, it is possible that therapeutic targets at the Y2 receptor, such as on GABAergic CEAl presynaptic terminals (Fig. 3B) may be helpful against chronic alcohol use.

Nicotine
Despite a steady decline in nicotine use since the1960s, nicotine has regained popularity in part due to the marketing of e-cigarettes (Cummings and Proctor, 2014).In 2022, 48% of young adults reported to have tried vaping in the past 30 days compared to just 15% in 2019  (Statistics Canada, 2023).Inhaled vapor rapidly delivers nicotine to the brain, but as nicotine has a relatively short half-life of two hours (Benowitz, 2009), the time frame for users to crave their next hit is quickened and renders nicotine a very addictive substance.
Like other drugs of abuse, nicotine also implicates the mesolimbic DA pathway to promote DA release from VTA projections in the NAc (Exley et al., 2008).Nicotine binds to nicotinic acetylcholine (nACh) receptors and promotes the activity of the DA VTA to NAc projection by stimulating glutamatergic afferents to DA VTA cells or direct stimulation of DA VTA cells (Liu et al., 2012;Nisell et al., 1997).With chronic nicotine use, the expression of nACh receptors are upregulated to compensate for their desensitization (Fenster et al., 1999), and the upregulation of nACh receptors forms the root cause of nicotine dependence (Zhao-Shea et al., 2013).

GLP-1R agonists suppress nicotine seeking and promote nicotine avoidance
When mice were trained to lever press for a nicotine reward, peripheral exendin-4 administration decreased nicotine intake (Tuesta et al., 2017).GLP-1R agonists can also attenuate nicotine-mediated expression of condition place preference, and as Glp1r deletion can increase lever pressing for nicotine (Tuesta et al., 2017), these findings suggest that the GLP-1R pathway is targeted in the development of nicotine addiction (Egecioglu et al., 2013a).Furthermore, GLP-1R agonists can also prevent withdrawal-like anxiety behaviors (Brynildsen et al., 2018) and decrease nicotine-mediated hyperactivity (Egecioglu et al., 2013a), which is consistent with the effects of GLP-1 on cocaine (Klausen et al., 2022) and amphetamine (Eren-Yazicioglu et al., 2021b).
The basis of nicotine-induced addiction stems from the potentiation of DA release from VTA projections into the NAc (Buchta and Riegel, 2015;Deutch and Roth, 1990), where GLP-1 can attenuate nicotineseeking by suppressing nicotine-induced DA release (Falk et al., 2023;Egecioglu et al., 2013a).While it is known that the VTA expresses GLP-1R (Cork et al., 2015;Merchenthaler et al., 1999), the specific identities of GLP-1R VTA cell have not been resolved.As GLP-1R agonism can attenuate cocaine-seeking by GLP-1R activation at GABAergic interneurons upstream of DA VTA cells (Hernandez et al., 2021), we can speculate that GLP-1R expression and activation of intra-VTA GABAergic interneurons would dampen the excitability of DA VTA cells (Fig. 4A).Furthermore, GLP-1 can prevent nicotine relapse by promoting avoidance in part by stimulating GABAergic cells in the interpeduncular nucleus (IPN).The IPN promotes nicotine avoidance when stimulated (Fowler et al., 2011), thus GLP-1R expression (Göke et al., 1995) can directly activate IPN cells to promote nicotine avoidance.In parallel, GLP-1 can also stimulate glutamatergic NTS afferents to enhance excitatory input in the IPN (Tuesta et al., 2017) (Fig. 4B).

PYY 3-36 diminished negative nicotine withdrawal-like symptoms
PYY 3-36 can be an effective therapeutic companion to curb nicotine use by minimizing negative symptoms that arise during nicotine withdrawal.Acute introduction of nicotine had no effect on circulating PYY levels in the rat (Shankar et al., 2024) and human (Driva et al., 2022).However, chronic nicotine intake increases baseline PYY levels, which remain elevated even in ex-smokers compared to individuals who have never smoked before (Bodén et al., 2023).Paradoxically, increased PYY levels are associated with reduced urges for nicotine during withdrawal thus reducing relapse rates in humans (Al'absi et al., 2014(Al'absi et al., , 2021;;Lemieux and Al'absi, 2018).
High PYY levels are associated with increased positive affect by promoting positive emotional states such as joy and cheerfulness during withdrawal periods (Al'absi et al., 2014).High PYY levels may also diminish negative withdrawal-like symptoms, which may include paw tremors and body shakes, reduced locomotor activity in an open field, increased anxiety-like behavior, and deficits in social behavior (Al'absi et al., 2014).Nicotine can elicit withdrawal-like symptoms after just one exposure (Harris et al., 2013), and PYY 3-36 administration may buffer the intense negative side effects seen within the first 24 h of nicotine withdrawal, which comprises the most vulnerable period for relapse (Zelle et al., 2017).The protective effects of PYY 3-36 may be attributed to Y1 receptor activation, which can reduce the somatic signs of nicotine withdrawal, including abdominal constipation, eye blinks, and ptosis (Rylkova et al., 2008).
While the neural circuitry underlying the effects of PYY 3-36 is not known, we can speculate that the Y1 receptor is a key mechanism underlying the protective effects of PYY 3-36 .Y1 receptor activation inhibits DA VTA cells (West and Roseberry, 2017), which is associated with attenuated nicotine-seeking, therefore PYY 3-36 may act directly at the VTA to curb nicotine intake (Fig. 4A).Moreover, in addition to promoting aversion or avoidance (Wolfman et al., 2018), upregulation of nACh receptors (Hilario et al., 2012) at somatostatin IPN cells can also drive anxiety-like behaviors during nicotine withdrawal (Zhao-Shea et al., 2013).The IPN is rich in Y1 receptor expression (Migita et al., 2001), thus PYY 3-36 may act directly at somatostatin IPN cells to diminish the expression of negative symptoms during nicotine withdrawal (Fig. 4B).

Dual receptor agonists may decrease drug-seeking without inducing malaise
There are limited studies currently available to assess whether a dual agonist targeting GLP-1 and PYY 3-36 to suppress drug-seeking behavior or to mitigate negative withdrawal-like symptoms could amplify the benefits of GLP-1 or PYY 3-36 individually.However, a recent low dose dual agonist, GEP44, targeting GLP-1/ PYY 3-36 holds promising therapeutic potential for treating substance use disorders.GEP44 is a dual agonist for GLP-1R and Y2 receptors, and peripheral administration of GEP44 intraperitoneally decreased opioid-seeking behavior without promoting malaise (Zhang et al., 2021).Compared to vehicle treatment, GEP44 decreased active lever presses for fentanyl in an operant Fig. 4. GLP-1 and PYY 3-36 suppressed nicotine-seeking and withdrawal while promoting nicotine avoidance.(A) Nicotine intake activates nicotinic acetylcholine receptors at DA VTA cells to increase DA release at the NAc.GLP-1 receptor activation stimulates intra-VTA GABA input to DA VTA cells to dampen the stimulatory actions of nACh receptor activation.PYY 3-36 acts via Y1 receptor to directly inhibit DA VTA cells to dampen nicotine-induced DA release at the NAc.(B) Chronic nicotine intake upregulates nAChR expression at somatostatin (SST) cells in the interpeduncular nucleus (IPN) to promote nicotine withdrawal, but PYY 3-36 -mediated Y1 receptor activation can diminish the negative effects of withdrawal.GLP-1 receptor agonists act primarily at the IPN to promote avoidance but may also increase glutamatergic input, such as from the NTS, to the IPN.DA, dopamine; GLP-1, glucagon like peptide 1; IPN, interpeduncular nulcues; nAChR, nicotinic acetylcholine receptor; NAc, nucleus accumbens; NTS, nucleus tractus solitarius; PYY 3-36 , Peptide YY 3-36 ; SST, somatostatin; VTA, ventral tegmental area; Y, Neuropeptide Y receptor.
conditioning task so that rats were less motivated to work for the opioid reward.GEP44 also suppressed locomotor activity during the operant self-administration task and exhibited less opioid-seeking behavior (Zhang et al., 2021).

Mental health implications
There is a bidirectional link between obesity (Fulton et al., 2022;Luppino et al., 2010;Mannan et al., 2016;Pan et al., 2012) and/or diabetes (Harding et al., 2019;Smith et al., 2013) and the risk for depression and anxiety.In addition to shared biological mechanisms underlie the link between metabolic and psychiatric disease (Fulton et al., 2022) the internalization of weight bias, which is characterized as having self-derogating and negative thoughts related to body image, is also an important risk factor for associated depressive, anxious, and suicidal thoughts (Brochu, 2020;Carels et al., 2013;Durso et al., 2016;O'Brien et al., 2016).These underlying links can be further complicated by adverse psychiatric effects previously reported with weight loss drugs.For instance, rimonabant, which reduced appetite by blocking cannabinoid receptor 1 which has widespread expression in the central and peripheral nervous system (Mackie, 2005), was approved as an antiobesity drug but was later withdrawn due to concerns over serious psychiatric side effects, including increased risk of depression and suicidal ideation (Akbas et al., 2009;Sam et al., 2011;Van Gaal et al., 2008).
As GLP-1R agonists are still a relatively new class of approved weight loss drugs and anxiety, depression, or suicidal ideation after been recently reported with the intake of GLP-1R agonists (Arillotta et al., 2023), it is important to continue monitoring their safety for long-term use.Human patients taking Wegovy or Ozempic, another GLP-1R agonist often report feeling demotivated or losing interest or pleasure in lifestyle activities they used to enjoy.Demotivation and loss of interest is symptomatic of major depressive disorder and schizophrenia (Fervaha et al., 2015(Fervaha et al., , 2016)), thus mood-related impacts are important considerations for the increasing popularity and long-term use of GLP-1R agonists as new weight loss drugs.
Interestingly, a retrospective study compared the incidence of suicidal ideation between patients taking semaglutide to those taking non-GLP-1R agonists as an anti-obesity medication.Their analyses showed lower incidences of suicidal ideation in both patients without, as well as those with, a prior history of suicidal thoughts following six months of semaglutide treatment (Wang et al., 2024).These results do not indicate that medications based on GLP-1R agonism are risk-free, but it is at least promising that the risk for suicidal ideation is lower than other antiobesity medications (Wang et al., 2024).However, this study took place over a 6-month follow-up period and only queried suicidal ideation.As psychiatric outcomes may develop or advance over a much longer time course, the long-term safety of such weight loss agents must still be monitored.
As of September 2023, the FDA Adverse Events Reporting System has received approximately 500 reports of mental health side effects including feelings of anxiety, depression, and suicidal thoughts after GLP-1R agonist administration (Lupkin, 2023), and this number is expected to climb as it has shown a steady and progressive increase since 2018 (Chiappini et al., 2023).Therefore, we reviewed the literature based on rodent and human studies to assess the potential impact of GLP-1 and/or PYY 3-36 on anxiety-and depressive-like outcomes.

Prolonged treatment with GLP-1R agonists reduced anxiogenesis
Studies assessing GLP-1R agonists on mood-related effects have revealed conflicting findings, as some studies report anxiogenic while others report anxiolytic effects.Interestingly, this difference may be explained by the duration of GLP-1 administration.Acute treatments (i.e., <2 weeks in animals; <3-6 months in humans) are associated with anxiogenic effects (Eren-Yazicioglu et al., 2021a;Kamble et al., 2016).
By contrast, chronic treatments (i.e., 2+ weeks in animals; 6+ months in humans) lead to anxiolytic effects (Tsai et al., 2022).Such durationdependent factors are also seen with selective serotonin reuptake inhibitors used to treat depression and anxiety (Cassano et al., 2002), as patients may experience anxiety over the first days or weeks before anxiolytic and/or anti-depressive effects emerge (Anderberg et al., 2016).
Reduced risks for depression or anxiety by chronic GLP-1-mediated treatment emerged through the analysis of large-scale health record databases over an extended treatment period.Human patients treated with GLP-1 for 1-3 months reported an increase in depressive symptoms and perceived stress levels, and this was seen in individuals with or without previous history of depression (Eren-Yazicioglu et al., 2021a;Li et al., 2023).However, as the duration of GLP-1-mediated intervention increases, the incidence of depression or anxiety decreases.Patients with diabetes taking GLP-1R agonists like liraglutide, dulaglutide, and exenatide report comparable incidences of depression or anxiety.However, within 3 months of treatment, their incidence of depression or anxiety was lower than among patients prescribed either placebo or alternative treatments like insulin to manage their diabetes (Tsai et al., 2022).After 6 months, the incidence of anxiety was much lower in patients treated with GLP-1R agonists though the risk reduction for incidences of depression was moderate (Tsai et al., 2022).The anti-depressant properties of GLP-1R agonists may evolve over a longer duration.A subsequent meta-analysis analyzed the contributions of GLP-1R agonists in Type 2 diabetes patients with and without obesity who do not have depression.This study revealed that over a treatment period of 6-12 months GLP-1R agonists have anti-depressant properties (Chen et al., 2024).
The mechanisms underlying the anxiolytic or anti-depressant effects of GLP-1R agonists are not clear, but basic research using animal models may provide insights to the implicated brain structures.The two classical paradigms to assess anxiety-like behavior in rodents is via the open field test and elevated plus maze, which comprises two open and two closed arms.Rodents inherently avoid open spaces to avoid predators and they are afraid of heights, so a rodent that spends less time in the center of an open field or open arm, respectively, is said to exhibit more anxiety-like behavior.Peripheral administration of liraglutide in rats acutely reduced entries to the center of an open field test or entries to the open arms of an elevated plus maze thus indicating an increase in their anxiety-like behavior (Kamble et al., 2016).However, chronic GLP1-R agonist administration over 2-3 weeks decreased anxiety-like behaviors, as mice spent more time in the center of an open field (Celikyurt et al., 2014;Weina et al., 2018).Similarly, chronic GLP-1R agonist administration also minimized depressive-like behaviors, as GLP-1R agonist-treated mice increased the amount of time spent escaping when hung by the tail or staying afloat in a water column in the tail suspension and forced swim test, respectively (Celikyurt et al., 2014;Isacson et al., 2011;Seo et al., 2023;Weina et al., 2018).The performance of these mice on the behavioral tests was comparable to mice treated with the antidepressant fluoxetine (Weina et al., 2018).
The amygdala and extended amygdala are brain structures implicated in the mood-related functions of GLP-1R agonists.Intracerebroventricular administration of GLP-1 acutely induced anxiogenic effects in the open field and elevated plus maze (Gulec et al., 2010) and is associated with neuronal activation in the central amygdala (CEA), which is well known to regulate emotion-related behavior (Möller et al., 2002) and promote anxiety-like behavior when activated (Gilpin et al., 2015).The CEA expresses GLP-1R (Zeng et al., 2021), and acute GLP-1 administration directly into the CEA of mice can also mimic anxiety-like behavior (Kinzig et al., 2003) thus implicating the CEA as a target site for GLP-1-mediated anxiety (Fig. 5A).The CEA receives strong projections from the bed nucleus of the stria terminalis (BNST) (Gungor et al., 2015), which is also a critical structure regulating anxiety and stress responses (Daniel and Rainnie, 2016).GABAergic BNST cells express GLP-1R, and knockdown of Glp1r mRNA in the anterior BNST reduced Y.Dumiaty et al. anxiety-like behavior (Zheng et al., 2019).These animal studies point to the amygdala as a key structure underlying the acute anxiogenic actions of GLP-1, but it is not known whether these structures also support the anxiolytic actions of chronic GLP-1 treatment.Chronic GLP-1 exposure can result in GLP-1R internalization and desensitization (Baggio et al., 2004;Shaaban et al., 2016), but while it has not yet been shown in vivo, reduced GLP-1R activation in the CEA may explain the anxioloytic actions of chronic GLP-1 administration (Fig. 5B).

PYY 3-36 acutely elicited apathy and demotivation but suppressed anxiogenesis over time
PYY 3-36 -treated mice showed deficits in social interaction and sensorimotor gating that are traits common in animal models of schizophrenia (Kjaergaard et al., 2019).These traits are related to Y2 receptor activation, as mice with a central deletion of the Y2 receptor spent more time in active social interaction with another mouse when placed in the same cage (Karl et al., 2010).Peripheral PYY 3-36 administration in mice also led to decreased motivation and elicited apathylike symptoms like decreased nesting behavior, and these effects were abolished by a Y2 receptor antagonist or a DA reuptake inhibitor.These findings suggested that Y2 receptor activation suppressed DA release and implicated DA-linked neural circuits in the expression of PYY 3-36mediated apathy or demotivation (Yamada et al., 2018).Taken together, these studies indicated that PYY 3-36 treatment can broadly impact motivation beyond food and may render animals in a general state of disinterest.
The net effect on anxiety in the basolateral amygdala (BLA) was dose-dependant, as low dose agonists stimulated the Y2 receptor to produce anxiogenesis, while higher doses mediated anxiolysis via Y5 receptors (Sajdyk et al., 2002).It may thus be possible to induce anxiolytic effects following PYY 3-36 administration at higher doses, but as PYY 3-36 availability in the brain may be lower following peripheral PYY 3-36 administration, its anxiolytic potential is masked.The anxietyrelated actions of PYY 3-36 may converge at the BLA where Y2 receptor activation mediates an anxiogenic phenotype while Y1 and Y5 receptor activation mediates anxiolytic outcomes (Michaelson et al., 2020;Molosh et al., 2013;Sajdyk et al., 2002).Y1 receptor activation suppresses glutamatergic input and increases GABAergic input to BLA cells leading to the net inhibition of glutamatergic pyramidal BLA neurons (Molosh et al., 2013).The BLA sends excitatory projections to the CEA, which in turn projects to downstream regions like the periaqueductal gray and BNST to promote anxiogenesis.Therefore, PYY 3-36 -mediated inhibition of the BLA can suppress anxiogenesis (Babaev et al., 2018) (Fig. 5B).

No anxiogenesis upon GLP-1 and PYY 3-36 co-administration
Few studies have investigated the impact of GLP-1 and PYY 3-36 coadministration on mood-related behaviors in rodents.Their coadministration did not affect the duration of time that rats spent in the center or open arms of the open field or elevated plus maze, respectively (Dischinger et al., 2021).There was also no effect on locomotion, grooming, or rearing that can be indicators for anxiety-and depressionlike phenotypes in rodents (Neary et al., 2005).While the full scope of mood-related effects from GLP-1 and PYY 3-36 co-administration is in its early stages, we did not find evidence of anxiogenesis such as that seen with GLP-1 treatment acutely.

Conclusion
GLP-1 and PYY 3-36 can elicit a state of demotivation to suppress food intake and drug-seeking behavior.Thus far, studies co-administering GLP-1 and PYY 3-36 analogues have revealed promising outcomes for weight loss.Furthermore, emerging GLP-1/ PYY 3-36 dual agonists also show promising outcomes that suppress opioid use and withdrawal.There is some concern that such weight loss medication can increase the risks for mood-related disorders or suicidal ideation, but to date, there is no evidence to support the concerns around such risks.In fact, medications based on GLP-1R agonism may decrease the risk for anxiety and suicidal ideation compared to other weight loss medications.In general, the therapeutic potential of a medication can be maximized at high(er) doses, but such aggressive treatment is often accompanied by nausea or malaise that decreases compliance.Since dual agonists aimed at multiple targets can enhance the effects of targeting each system individually, they are promising alternatives to provide effective treatment while minimizing unwanted side effects.

Declaration of competing interest
None.

Fig. 5 .
Fig. 5. Anxiolytic effect of GLP-1 and PYY 3-36 treatment is duration-dependent.(A) Acute administration of PYY 3-36 promoted net anxiolytic effects via Y1 and Y5 receptors to decrease glutamatergic input and reduce GABAergic output from the CEA while activating Y2 receptors at GABAergic BLA interneurons.Meanwhile, acute GLP-1 treatment activated GABAergic CEA cells to promote anxiogenesis.Therefore, the co-administration of GLP-1 and PYY 3-36 may not have acute effects on anxiety-like behavior.(B) Chronic administration of PYY 3-36 preferentially activates anxiolytic Y5 receptors to dampen excitatory projections from the BLA to CEA.Meanwhile, chronic GLP-1 receptor activation can desensitize and internalize to diminish CEA GABA output.Taken together, chronic co-administration of GLP-1 and PYY 3-36 analogues would elicit anxiolytic effects.BLA, basolateral amygdala; CEA central amygdala; GLP-1, glucagon-like peptide 1; Glu, glutamatergic cell; Y, Neuropeptide Y receptor.