TRMOME 1091 No. of Pages 10
Review
Metabolism and Mental Illness Matija Sestan-Pesa1 and Tamas L. Horvath1,* Over the past century, overwhelming evidence has emerged pointing to the hypothalamus of the central nervous system (CNS) as a crucial regulator of systemic control of metabolism, including appetite and feeding behavior. Appetite (or hunger) is a fundamental driver of survival, involving complex behaviors governed by various parts of the brain, including the cerebral cortex. Here, we provide an overview of basic metabolic principles affecting the CNS and discuss their relevance to physiological and pathological conditions of higher brain functions. These novel perspectives may well provide new insights into future research strategies to facilitate the development of novel therapies for treating mental illness. Introduction A major focus of modern medicine is on chronic disease, and, there are few that are as challenging as depression and obesity. Both affect a large proportion of the global population: 7.6% of the USA population, 12 years of age and older, report moderate to severe depressive symptoms [1]. Obesity is estimated to affect 2.1 billion people worldwide [2]. Even though there does not seem to be any apparent link between the two problems, evidence accumulated over the past few decades suggests that not only is there a link but there also seems to be a general connection between feeding disorders and mental illness [3]. The association between depression and feeding disorders is most apparent in the case of depression and obesity, as shown by the data from the National Health and Nutrition Examination Surveys: 43.2% of adults with depression also suffer from obesity, which is significantly greater than the 33% in the control group [4]. These data, as well as others, only show correlation and not causation. Often, the criticism is that this effect could be due to antidepression therapy since weight gain is one of the most common side effects, with 5–10% of patients gaining significant weight during long-term treatment [5]. However, anorexia and bulimia nervosa are feeding disorders and mental illnesses, thus demonstrating the inherent connection between the mechanisms that control feeding and higher cognition (affected in eating disorders). The comorbidity between depression and anorexia, for instance, is even greater than obesity, with major depression occurring in 63.9% to 74.5% in patients suffering from anorexia (anorexia nervosa restrictive and anorexia nervosa with bulimic symptoms, respectively) [6]. In this review article, we first provide a summary of the hypothalamic neurons involved in regulation of feeding and energy homeostasis, and some of their regulatory hormones, to describe the basic circuitry potentially connecting feeding behavior (and energy homeostasis) and mental illness. We focus on depression and eating disorders (specifically anorexia nervosa) to give insight into possible future directions regarding therapy and research. The core of this review is centered on the serotoninergic system since it is deeply involved in energy homeostasis and the aforementioned mental illnesses. Lastly, we present some novel data concerning the endocannabinoid system, which may have an important impact on future research related to feeding and depression.
Hypothalamic Circuitry and Feeding Behavior Feeding behavior is complex and many parts of our brain contribute to its control. However, one structure stands out in the central nervous system (CNS), the hypothalamus, which is considered to be the main regulatory organ for mammals, and regulates feeding behavior. Two subsets
Trends in Molecular Medicine, Month Year, Vol. xx, No. yy
Trends Hypothalamic neurons in the arcuate nucleus regulate feeding behavior and are modulated by a complex network of peripheral signals. Changes in feeding behavior and energy metabolism are common symptoms of mental illness and side effects of psychopharmacological treatment, indicating a possible connection between feeding and mental illness. Peripheral signals acting on hypothalamic neurons in the arcuate nucleus may help explain some of the molecular mechanisms by which feeding and mental health can be connected. In particular, serotonin signaling and metabolic pathways appear to be critical for such connections, as they are linked to both mental illness and feeding behavior.
1 Program in Integrative Cell Signaling and Neurobiology of Metabolism, Section of Comparative Medicine, Yale University School of Medicine, New Haven, CT 06520, USA
*Correspondence:
[email protected] (T.L. Horvath).
http://dx.doi.org/10.1016/j.molmed.2015.12.003 © 2015 Elsevier Ltd. All rights reserved.
1
TRMOME 1091 No. of Pages 10
of neurons are key players in feeding: the orexigenic (promoting feeding) and anorexigenic (suppressing feeding) within the arcuate nucleus, which is a cluster of neurons located in the mediobasal hypothalamus, lying on either side of the third ventricle, just above the median eminence. These two distinct groups of neurons communicate with many areas of the brain both within and outside of the hypothalamus. One anorexigenic population is situated more laterally in the arcuate nucleus expressing proopiomelanocortin (POMC), and when stimulated, suppresses feeding [7]. POMC neurons project to numerous areas involved with feeding and appetite control, such as the parabrachial nucleus (PBN), the paraventricular nucleus of the hypothalamus (PVH), the dorsal vagal complex (DVC), and the intermediolateral column of the spinal cord (IML) [8]. They produce /-melanocyte-stimulating hormone (/-MSH), which serves as an agonist for the anorectic melanocortin receptors (MC3R and MC4R) [9]. Through these Gs-protein-coupled receptors, melanocortins activate neurons by increasing cAMP levels [10]. The other population of neurons, coexpressing neuropeptide Y (NPY) and agouti-related protein (AgRP), is located more medially in the arcuate nucleus and upon stimulation increases feeding. The NPY/AgRP neurons project to the PVH, PBN, and to the POMC neurons [8]. While NPY produces it orexigenic effects via NPY receptors (NPY1R, NPY2R, NPY3R, NPY4R, NPY5R), AgRP functions as an inverse agonist for the melanocortin receptors, thus producing an opposite effect. NPY receptors are Gi/o-protein-coupled receptors that decrease cAMP levels inside the cell. Through this mechanism, NPY can inhibit MC4R-expressing PVH neurons via NPY1R [11], as well as suppress neuronal activity in the arcuate nucleus (POMC and GABAergic neurons) and the ventromedial hypothalamic nucleus (VMN) [12,13].
Leptin and Insulin: Regulators of Feeding and Obesity Leptin, from the adipose tissue, and insulin, from the pancreas, are considered long-term adiposity signals. They control energy balance and food intake, in part, by promoting or inhibiting hypothalamic neurons in the arcuate nucleus. The effects of leptin are thought to be mediated by hypothalamic KLF-4 (a zinc finger-type of transcription factor), as shown in rats overexpressing KLF-4, which induced food intake, increased body weight, and blunted leptin sensitivity via stimulation of AgRP [14]. By contrast, downregulation of KLF-4 inhibited food intake and dietrelated obesity [14]. Stimulating anorexigenic POMC-expressing neurons appears to be another mechanism of the inhibitory effects of leptin on food intake [15]. However, their role cannot be simplified to on/off mechanisms, as research has suggested that leptin receptor resistance plays an important role in obesity [16]. Studies on rats have shown that leptin directly influences serotonin metabolism, by increasing serum serotonin levels, while serotonin in brain tissue (hypothalamus and hippocampus) seems to be decreased after leptin administration [17]. Like leptin, insulin resistance has been proposed to contribute greatly to obesity and feeding behavior in general. Similar to leptin, insulin is thought to have central effects on feeding, although less effective; this is believed to be accomplished through the inhibition of NPY/AgRP neurons in the arcuate nucleus [18]. Insulin resistance contributes to many metabolic abnormalities including obesity and, as such, suggests its importance in feeding regulation. Pharmacological intervention with insulin sensitivity-improving drugs, such as metformin, has shown inhibition of food intake and weight loss in rats and humans, which is thought to be mediated via inhibition of NPY and AgRP in the hypothalamus [19,20]. In line with this, insulin receptors have been found in areas of the brain crucial to food intake regulation and also seem to be important in energy homeostasis [21].
Role of Gut Hormones in Feeding Feeding control is multifaceted and involves numerous other players, all of which act in concert to control appetite (Boxes 1 [26_TD$IF]and 2[27_TD$IF] and Figure 1). One such player is the gut, which is the largest endocrine organ in the human body. The gastrointestinal (GI) tract produces various peptide hormones, which are thought to be important in short-term energy intake control by playing a critical role in meal initiation and termination. The key peptides in question are cholecystokinin
2
Trends in Molecular Medicine, Month Year, Vol. xx, No. yy
TRMOME 1091 No. of Pages 10
Box 1. Cytokines, Depression, and Feeding: the Neuroendocrine–Immune Axis The administration of cytokines, such as interleukin 1 (IL-1), IL-6, tumor necrosis factor / (TNF-/), and interferon (IFN), has been shown to cause a reduction in food intake; these cytokines seem to mediate anorexia as part of a sickness response in mice that accompanies infection and inflammation [[7_TD$IF]72]. Connections between cytokine levels and hypothalamic circuitry are present in rats, with IL-1 and NPY being antagonists in their mechanisms of action [[8_TD$IF]73]. The anorectic effects of IL-1 are present initially but are not maintained when MC4R is blocked in rats [[9_TD$IF]74]. It has been suggested that these proinflammatory molecules exert their anorexigenic effects through serotonin signaling, at least in chronic diseases [41,42]. Since loss of appetite can be attributed to anhedonia behavior, a major symptom during depression, cytokines are potentially important in both feeding and depression [[10_TD$IF]75]. Indeed, increased levels of proinflammatory cytokines have been measured in depressed patients, and 50% of patients on immunotherapy (e.g., IFN) develop clinical depression [[1_TD$IF]76]. A demonstration that antidepressant treatment is effective in patients with IFN-induced depression further suggests a connection between depression and feeding circuitry [[12_TD$IF]77]. However, a few discrepancies in the literature suggest that further investigation into these relationships is certainly warranted [[13_TD$IF]78].
(CCK), ghrelin, pancreatic polypeptide (PP), peptide YY (PYY), glucagon-like peptide 1 (GLP-1), and oxyntomodulin (OXM) [21]. Ghrelin Ghrelin is the only orexigenic GI tract hormone known to date and has been proposed as a meal initiator since it was first described in 1999 [22,23]. Its signaling mechanism involves the growth hormone secretagogue receptor expressed by the NPY/AgRP neurons of the arcuate nucleus [23]. Direct pharmacological blockade of ghrelin decreases food intake and body weight in rodents [24]. Increase in body weight and food intake by ghrelin is achieved through separate mechanisms, with central ghrelin activity inducing feeding and peripheral activity in white adipose tissue, which affects lipid metabolism [25]. Other than its effects on feeding, ghrelin also appears to be important for stress responses in mice, attenuating anxious behavior and the activity of the hypothalamic–pituitary–adrenal (HPA) axis [26]. Intracerebroventricular administration of ghrelin in mice has been reported to increase the expression of depression-associated genes coding for Box 2. Additional Players of Feeding Behavior Cholecystokinin CCK was the first hormone discovered to have effects on appetite. It is mostly produced by neuroendocrine cells in the small intestine and has local regulatory effects such as gallbladder control and gastric emptying, while its central effects have been shown to be mediated by hypothalamic neurons [21]. Intravenous infusion of a CCK-A receptor antagonist in healthy human males induces a significant increase in calorie intake and subjective hunger ratings, suggesting that physiological CCK is important for satiety signals in humans [[14_TD$IF]79]. Peptide YY Similar to PP, PYY achieves its anorexigenic effect approximately 1 h after ingestion [[15_TD$IF]80] and does this via Y family Gprotein-coupled receptors [[16_TD$IF]81]. A study in young men showed a significant increase in PYY levels after 7 days of overfeeding, which did not correlate with BMI indicating that the effects of PYY on energy metabolism were independent of adiposity status [[17_TD$IF]82]. Pancreatic Polypeptide Acute changes in PP after exercise support its role as an anorexigenic modifier of appetite, since PP levels decrease similar to other anorexigenic gut hormones (and leptin), while ghrelin levels (orexigenic) increase [[18_TD$IF]83]. However, long-term changes in PP levels increase 1 year after weight loss, unlike other anorexigenic hormones [[19_TD$IF]84]. These contradictory results indicate a need for further investigation into the relationship among PP, diet, and exercise. Glucagon-Like Peptide 1 Produced and secreted in proportion to ingested calories by the small intestine and colon [[20_TD$IF]85], circulating GLP-1 demonstrates an inhibitory effect on eating through activation of GLP-1 receptors in the hindbrain of male rats [[21_TD$IF]86]. Oxyntomodulin OXM is produced by the small intestine after feeding in a calorie-dependent manner [[2_TD$IF]87]. In response to peripheral administration of OXM in humans, satiation is increased, food intake is reduced, and loss of body weight ensues with repeated injections [[23_TD$IF]88]. A novel GIP (gastric inhibitory polypeptide)–OXM hybrid peptide induces beneficial effects on glucose homeostasis, insulin secretion, and body weight, and these effects are potentially mediated via GIP, glucagon, and GLP-1 receptors [[24_TD$IF]89].
Trends in Molecular Medicine, Month Year, Vol. xx, No. yy
3
TRMOME 1091 No. of Pages 10
IL-6R
6( +) 5-H T (+ )
GHSR
5-H T( -
(-)
POMC
GABA
)
CCK1
NPY2R LepRb
LepRb
in (+) Ghrel
PP
NPY2R
NPY2R
L ep
+) in (
Insulin (-)
ul Ins n
(-) PYY
Le p n
(-
(+)
GI tract Pancreas
InsR
NPY/ AgRP
5-HT1B
5-HT2C InsR
IL-1R1
NPY4R
IL-1β (-)
IL-
IL-
IL-6R
-) 6(
Adipose ssue
) CCK (-)
Raphe nuceli
Immune system
Figure 1. A Complex Network of Peripheral Signals Operate in the Neuronal Modulation of Feeding Behavior. This cartoon visually portrays the many circuits that have been implicated in the regulation of feeding. Because it is a complex signaling network, many challenges lie ahead in carefully dissecting the contribution of each pathway in feeding behavior.[6_TD$IF] Abbreviations: Raphe nuclei, cluster of nuclei in the brainstem that release serotonin to the rest of the brain; POMC, proopiomelanocortin expressing neurons; NPY/AgRP, neuropeptide Y and agouti-related protein expressing neurons; GABA, g-amino butyric acid releasing neurons; GI, gastrointestinal; InsR, insulin receptor; LepRb, leptin receptor; NPY2R, neuropeptide Y receptor type 2; NPY4R, neuropeptide Y receptor type 4; 5-HT2C, serotonin(5-HT) receptor type 2C; 5-HT1B, serotonin(5-HT) receptor type 1B; CCK1, cholecystokinin A receptor; GHSR, growth hormone secretagogue receptor (ghrelin receptor); IL-1R1, interleukin 1 receptor type 1; IL-6R, interleukin 6 receptor; IL-1b, interleukin 1 beta; IL-6, interleukin 6; PYY, peptide YY; PP, pancreatic polypeptide; CCK, cholecystokinin.
the delta opioid receptor (DOR), the lutropin–choriogonadotropin hormone receptor (LHCGR), the serotonin transporter (SERT), and to reduce expression of interleukin 1 beta (il1b), a cytokine that has been connected to depression and anorexic behavior [27]. Serotonin Serotonin (5-HT) exerts a modulating effect on almost every function of the brain. Through inhibition or stimulation of GABAergic neurons, it regulates mood, sleep, appetite, circadian rhythm, neuroendocrine functions, body temperature, pain sensitivity, motor activity, and cognitive functions [28]. To understand the role of serotonin in appetite control, it is important to analyze the relationship between diet and 5-HT levels in the brain. The amino acid tryptophan, the precursor for 5-HT production, must compete with large neutral amino acids (LNAAs) for transport over the blood– brain barrier making the tryptophan/LNAA ratio a determining factor in the supply of tryptophan to the brain. Thus, diets that are rich in protein would provide large amounts of amino acids, lowering the tryptophan/LNAA ratio, which would in turn decrease the production of 5-HT. By contrast, a carbohydrate-rich diet would cause the release of insulin, which would in turn promote amino acid uptake in peripheral tissue, thus increasing the tryptophan/LNAA ratio and stimulating 5-HT synthesis [29]. In vivo experiments in rats have shown that food intake
4
Trends in Molecular Medicine, Month Year, Vol. xx, No. yy
TRMOME 1091 No. of Pages 10
increases 5-HT release [30], and upon closer inspection, it was shown that this increase was dependent on carbohydrates, while protein dependency showed the reverse effect [31]. The administration of pure tryptophan decreases food intake within the first hour of systemic administration, although this effect is most likely dose-dependent and responses vary based on experimental design [32,33]. Experiments using mice with genetically altered 5-HTT (serotonin reuptake transporter) have shown changes in animal weight: 5-HTT deficiency led to increased body weight [5-HTT knockout (KO) mice], while 5-HTT overexpression led to decreased body weight [34]. These data complement findings in humans showing that the SS variant genotype of the 5-HTTLPR (serotonin transporter gene promoter) polymorphism in the SLC6A4 gene is associated with an increase in body mass index (BMI) and obesity in non-elderly stroke patients [35]. Interestingly, this same polymorphism has been connected to anorexia nervosa, depression, impulsivity, and anxiety [36]. However, a meta-analysis of clinical trials investigating the efficiency of pharmacological intervention on the 5-HTT pathway using D-fenfluramine (a 5-HT release and reuptake inhibitor) had previously reported significant decreases in body weight of obese patients [37]. Further investigation in rats has shown that the hypophagic effects of D-fenfluramine (suppressed feeding) are not attributable to its interference with 5-HT synthesis but, rather, on the 5-HT2C receptor, which might explain the contradictory data from the above-mentioned studies [38]. This shifted the focus from 5-HT synthesis to the role of serotonin receptors, which are a diverse group of seven different types of receptors. While 5-HT2C agonists cause a reduction in food intake, they also activate the HPA axis, a common finding among patients with depression [39]. Further investigation on rats has shown that 5-HT stimulates POMC neurons via the 5-HT2C receptor. Conversely, through the stimulation of 5-HT1B receptors, serotonin hyperpolarizes NPY/AgRP neurons, thus reducing the GABAergic inhibitory effect on the POMC neurons [40]. Other serotonin receptors also seem to play a role in feeding behavior. Experiments using adenocarcinoma or lung tumor-bearing mice as cachexia models (weight loss) showed that an increase in serotonin signaling was correlated with decreased food intake [41,42]. In vitro experiments using hypothalamic cell lines showed that serotonin exposure decreased NPY secretion [28_TD$IF]and downregulated 5-HT2B receptor [41,42]. In mice lacking MC4R, the administration of a 5-HT2C agonist, BVT.X (which reduces food intake) did not induce hypophagia, suggesting that the MC4R pathway of food intake regulation is necessary for 5-HT2C agonist-induced hypophagia to occur [43]. It has also been proposed that MC3R may be important in this process as well [44].
Serotonin and Eating Disorders There is a significant disruption of serotonin pathways in eating disorders, but not those that one might expect. A general decrease in 5-HT tone is present in women with anorexia, which is indicated by decreased platelet binding of serotonin uptake inhibitors [45], blunted prolactin and cortisol responses to 5-HT agonists and partial agonists [46], and reduced levels of 5HIAA (5-HT metabolite) [47]. Despite the anorexigenic properties of serotonin, patients with anorexia have decreased 5-HT levels in the brain. After restoration of full weight in patients who have recovered from anorexia nervosa, a normal plasma PRL (prolactin) response to D-fenfluramine (used to assess serotoninergic function) has been observed [48]. This may be attributed to diet-induced reduction in the intake of tryptophan [49]. Positron emission tomography (PET) scans have shown an increase of cerebral spinal fluid (CSF) 5-HIAA and elevated 5-HT1A receptor binding in former anorexia nervosa patients, suggesting that anorexia may be related to a state of increased 5-HT tone, but which could be obscured by decreased 5-HT tone due to malnutrition [50]. In hospitalized patients who underwent refeeding, their low blood serotonin, plasma tryptophan, and LNAA and tryptophan/LNAA ratio improved in parallel with improvement of BMI, suggesting that the low serotonin tone in anorexia nervosa is most likely attributable to dietary restrictions. Symptoms of depression,
Trends in Molecular Medicine, Month Year, Vol. xx, No. yy
5
TRMOME 1091 No. of Pages 10
anxiety, and obsessive behavior (which often present themselves in anorexia) also improved with BMI improvement [51]. To further explain this paradoxical situation, it is relevant to address how eating disorders are distinguished. According to the Diagnostic and Statistical Manual of Mental Disorders (DSM) classification, there are two eating disorders: anorexia nervosa and bulimia nervosa. The avoidance of weight gain by controlling food intake is the primary symptom of anorexia nervosa, while in bulimia a person goes through episodes of binge eating followed by behaviors that compensate these episodes, such as exercising, vomiting, or using laxatives. This apparent contrast, where anorexia involves ‘control’ of food intake, and bulimia often presents itself as a ‘lack of control’ behavior, has produced a loosely defined classification into two new groups, ‘restrictors’ and ‘bingers/purgers’. On the one hand, restrictors are those who constantly restrict food intake without episodes of binging or purging, typically anorexic. On the other hand, patients who binge and/or purge can sometimes be anorexic, sometimes bulimic, and sometimes obese [49]. Another way to classify eating disorders has been by behavior trait characteristics. Thus there are distinct groups, patients with an eating disorder who are behaviorally and emotionally dysregulated, correlating more with a ‘binge/purge’ type of eating disorder, with a range of mild to severe symptoms. Other patients are behaviorally and emotionally constricted, which is more often associated with the ‘restrictor’ group [52]. The contradictory finding of a reduction in 5-HT tone in patients with anorexia may make more sense when seen through the above classifications. It appears that levels of 5-HIAA are lower in eating disorders characterized by binging/purging than in the restrictive form [47]. PET scans of recovered patients with anorexia nervosa show an increase in 5-HT1A binding in bulimic (binge/ purge) anorexia, but not in patients with restrictive anorexia. 5-HT1A binding has been associated with anxiety in patients after recovering from restrictive anorexia, suggesting that the psychobiological alterations might be trait-related [53]. Differences between binge/purge and restrictor anorexic patients have even been seen in gut hormone levels, with PYY and leptin being lower in the former than in the latter [54]. Collectively, these data indicate that different illnesses sharing symptoms and etiology may have different molecular profiles, which can reflect distinct traits and behavior. Underlying biochemical differences should be taken into account with regard to therapy for eating disorders. An example is the use of fluoxetine (e.g., Prozac) as therapeutic for anorexia nervosa, which has not shown much usefulness [55]. Because anorexia is incredibly difficult to treat, many combinations have been used and it has been suggested that a combination of selective serotonin reuptake inhibitors (SSRIs) and atypical antipsychotics might be of therapeutic use. For instance, unlike other atypical antipsychotics aripiprazole (e.g., Abilify) has proven to be more effective then olanzapine (e.g., Zyprexa) in this regard; this might be due to the fact that aripiprazole is both a partial agonist as well as an antagonist for dopamine and serotonin receptors, respectively [56]. Findings such as these indicate that changes in biochemistry in combination with behavioral traits should be taken into account for future therapeutic research as well as for current treatments.
Serotonin and Depression A large number of studies have shown serotoninergic abnormalities in depression, such as the recurrence of depression after tryptophan depletion in fully recovered patients, low CSF levels of 5-HIAA in depressed patients with a history of suicidal behavior, a decrease in serotonin uptake and transporter binding sites in the brain and in platelets from depressed individuals, blunted neuroendocrine responses to serotoninergic stimuli, as well as changes in the density of 5-HT receptors in depression and suicide [57]. Symptoms of depression seem to be relieved by treatment that increases serotoninergic activity, through inhibition of serotonin uptake or metabolism [58]. Although many SSRIs simulate 5-HT2C receptors by increasing levels of synaptic serotonin, the delayed antidepressant effects coincide with a downregulation of
6
Trends in Molecular Medicine, Month Year, Vol. xx, No. yy
TRMOME 1091 No. of Pages 10
the 5-HT2C receptors [59]. Molecular imaging studies on patients suffering from depression have shown a highly significant reduction in 5-HTT availability [60], which complements the data on the SS variant genotype of 5-HTTLPR being connected to depression and, furthermore, has even been associated with different responses to antidepressant medication. Patients with an SS genotype have shown a negative correlation between plasma paroxetine levels with improvement in MADRS (Montgomery–Asberg Depression Rating Scale) scores, while patients with the SL and LL variant genotypes showed a positive correlation between paroxetine plasma levels and improvement in MADRS scores [61]. This reinforces the concept of individualized therapeutics for depression and eating disorders.
Treatment of Mood Disorders SSRIs are the center of much controversy regarding the treatment of depression. While several studies report that tricyclic antidepressants (most of which are serotonin–norepinephrine reuptake inhibitors) may cause a weight gain of 0.57–1.4 kg per month with treatment [62], SSRIs and similar antidepressants (which base their action on serotoninergic mechanisms) appear to exhibit differential effects, depending on which antidepressant is in question; some antidepressants are associated with weight gain, while others seem to promote weight loss. Among the SSRIs, paroxetine (e.g., Paxil), appears to induce the greatest weight gain during long-term treatment, while the atypical antidepressants, bupropion (e.g., Wellbutrin) and nefazodone (e.g., Serzone®[25_TD$IF]), do not have this effect. In fact, while nefazodone seems to be weight gain neutral, bupropion is possibly associated with weight loss [63]. Another antidepressant, fluoxetine, has shown hypophagic effects on rats, and 5-HT1A and 5-HT2C receptors seem to be mediating this effect [64]. These inconsistencies could be caused by confounding factors in these studies, such as a lack of placebo-controlled trials and possible reoccurrence of depressive symptoms. There is also a peculiar paradox in the action of SSRIs, which have been shown to promote weight loss in the beginning of treatment, only to rebound and cause an eventual net weight gain after prolonged administration [5]. Because of inconsistencies in such data, an onslaught of side effects, as well as a lack of effective treatments, the pharmacological industry has been accused of unjustly supporting the serotonin hypothesis in the hopes of marketing SSRIs. It has been suggested that more effective and cheaper treatments are being marginalized to promote the success of SSRIs and similar antidepressant drugs [65]. One such drug is ketamine (e.g., Ketalar), which recent studies propose as more effective than SSRIs, and can even inhibit suicidal thoughts within 40 min to 24 h of administration. The fact that ketamine acts via the glutamatergic system only further tests the serotonin hypothesis of depression [66]. Some data actually even go as far as suggesting that SSRIs are not effective in mild to moderate depression and only show some improvement in severe depression [67]. Others have reported that 80% of the response to SSRIs can be attributed to a placebo effect [68]. Regardless, abnormalities in serotonin metabolism in depression do exist and need to be explained. In addition, data have emerged suggesting that the use of SSRIs for the treatment of depression has been wrongly viewed or pushed by pharmaceutical companies, and, that results actually suggest a consistent superiority of SSRIs relative to placebo [69]. These inconsistencies could provide a direction for future research in antidepressant therapy, especially that which concerns weight gain side effects, as it can cause considerable discouragement for taking medication. Future discernment of the role of serotonin in depression and many other mental illnesses may also provide valuable information on the effects it can exert on food intake and energy homeostasis.
The Endocannabinoid System and Mood Regulation Many times, research provides paradoxical and inconsistent data, which must be reconciled. A similar paradox to that of serotonin has been observed in studies of the endocannabinoid system, where an unexpected increase in the activity of POMC neurons (which promote satiety) was found in CB1R agonist-induced feeding in mice [70]. This finding warranted further
Trends in Molecular Medicine, Month Year, Vol. xx, No. yy
7
TRMOME 1091 No. of Pages 10
investigation, where inhibitory and stimulatory designer receptors exclusively activated by designer drugs (DREADD) containing viruses (gene delivery) were administered to mouse brains and used to demonstrate that POMC neuronal activity was responsible for CB1R-driven feeding [70]. The mechanism by which this occurred included CB1R-induced increased production of b-endorphin by hypothalamic POMC neurons [70]. Not only is the endocannabinoid system involved in feeding behavior but it also appears to contribute to mood states. For instance, the CB1R antagonist rimonabant was used to treat obesity, only to be taken off the market in 2008 due to the appearance of side effects that included depression and anxiety. These side effects may have been associated with the effect of rimonabant on emotional memory, where the drug showed negative bias on memory recognition and emotional word memory tasks in healthy human subjects [71]. Further investigation into the effects of endogenous cannabinoids [anandamide (AEA) and 2-arachidonoylglycerol (2-AG)] on mood disorders and mental illness are warranted, since it appears that they are involved in both energy metabolism and mood regulation.
Concluding Remarks The aforementioned examples illustrate the multiple means by which different mechanisms associated with the regulation of appetite are intertwined with biological processes controlling normal and impaired higher brain functions. Of course, these examples pertaining to hypothalamic circuits controlling appetite do not begin to cover all aspects of the remarkable relationship that exists between homeostatic needs driven by peripheral tissue functions and the brain (see Outstanding Questions). Nevertheless, they point to an important and underappreciated aspect of primitive control of complex behaviors with biochemical and clinical implications for mental disorders. These connections themselves do not necessarily mean that mental illnesses share etiology with disruptions in the regulation of feeding but, instead, they offer clues to new directions for future research, novel therapies, and better prognoses in response to psychopharmacological treatment. Insight into the different metabolic profiles and feeding behaviors of patients with various mental illnesses could potentially improve both diagnostic tools and treatment options. Ultimately, the energetic metabolism of cells in various tissues can have a significant influence on various aspects of mental well-being and behavior. Because hypothalamic feeding circuits are at the heart of systemic control of fuel availability and utilization for every cell, it is reasonable to assume that these circuits and brain areas play crucial, and previously unrecognized, roles in the etiology and propagation of mental health and disease. References 1. Pratt, L.A. and Brody, D.J. (2014) Depression in the U.S. household population, 2009-2012. NCHS Data Brief 172, 1–8 2. Ng, M. et al. (2014) Global, regional, and national prevalence of overweight and obesity in children and adults during 1980-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 384, 766–781
9. Xu, Y. et al. (2011) Central nervous control of energy and glucose balance: focus on the central melanocortin system. Ann. N. Y. Acad. Sci. 1243, 1–14 10. Smith, M.A. et al. (2007) Melanocortins and agouti-related protein modulate the excitability of two arcuate nucleus neuron populations by alteration of resting potassium conductances. J. Physiol. 578, 425–438
3. Kishi, T. and Elmquist, J.K. (2005) Body weight is regulated by the brain: a link between feeding and emotion. Mol. Psychiatry 10, 132–146
11. Yulyaningsih, E. et al. (2011) NPY receptors as potential targets for anti-obesity drug development. Br. J. Pharmacol. 163, 1170–1202
4. Pratt, L.A. and Brody, D.J. (2014) Depression and obesity in the U. S. adult household population, 2005-2010. NCHS Data Brief 167, 1–8
12. Roseberry, A.G. et al. (2004) Neuropeptide Y-mediated inhibition of proopiomelanocortin neurons in the arcuate nucleus shows enhanced desensitization in ob/ob mice. Neuron 41, 711–722
5. Aronne, L.J. and Segal, K.R. (2003) Weight gain in the treatment of mood disorders. J. Clin. Psychiatry 64 (Suppl. 8), 22–29
13. Chee, M.J. et al. (2010) Neuropeptide Y suppresses anorexigenic output from the ventromedial nucleus of the hypothalamus. J. Neurosci. 30, 3380–3390
6. Godart, N. et al. (2015) Mood disorders in eating disorder patients: prevalence and chronology of ONSET. J. Affect. Disord. 185, 115– 122 7. Meister, B. (2007) Neurotransmitters in key neurons of the hypothalamus that regulate feeding behavior and body weight. Physiol. Behav. 92, 263–271 8. Sohn, J.W. et al. (2013) Neuronal circuits that regulate feeding behavior and metabolism. Trends Neurosci. 36, 504–512
8
Trends in Molecular Medicine, Month Year, Vol. xx, No. yy
14. Imbernon, M. et al. (2014) Hypothalamic KLF4 mediates leptin's effects on food intake via AgRP. Mol. Metab. 3, 441–451 15. Chaudhri, O.B. et al. (2008) Can gut hormones control appetite and prevent obesity? Diabetes Care 31 (Suppl. 2), S284–S289 16. Enriori, P.J. et al. (2007) Diet-induced obesity causes severe but reversible leptin resistance in arcuate melanocortin neurons. Cell Metab. 5, 181–194
Outstanding Questions What are the roles of endogenous cannabinoids in feeding and energy metabolism? Manipulation of the endocannabinoid system has the ability to radically change the function of POMC neurons and influence the mood of a person. It is very likely that the endogenous cannabinoids, AEA and 2-AG, serve an important role in the regulation of feeding behavior and mood. Can differences in metabolic profiles of serotonin in patients suffering from anorexia be used for better diagnosis and treatment? Since these differences in serotoninergic tone can reflect differences in behavioral traits in anorexia, they could also be used to separate patients into groups that might have different etiologies and responses to therapy. Do changes in serotonin metabolism in patients with depression produce expected behavioral changes in feeding and energy homeostasis? Since some patients with depression gain weight while others lose weight, it would be useful to investigate whether such differences in energy homeostasis correlate with the state of a patient's serotonin metabolism. Can one observe expected effects on appetite-regulating neurons in the arcuate nucleus? How do chronic changes in 5-HT receptor expression influence feeding behavior and energy metabolism? Many SSRIs cause weight loss in patients with depression, only to rebound and cause eventual net weight gain. Downregulation of 5-HT receptors is thought to be the cause for the delayed effects of some antidepressants, but this phenomenon could also be important in mediating changes in feeding behavior and energy metabolism in an individual.
TRMOME 1091 No. of Pages 10
17. Haleem, D.J. et al. (2015) Behavioral, hormonal and central serotonin modulating effects of injected leptin. Peptides 74, 1–8 18. Garcia-San Frutos, M. et al. (2007) Impaired central insulin response in aged Wistar rats: role of adiposity. Endocrinology 148, 5238–5247 19. Lv, W.S. et al. (2012) The effect of metformin on food intake and its potential role in hypothalamic regulation in obese diabetic rats. Brain Res. 1444, 11–19 20. Malin, S.K. and Kashyap, S.R. (2014) Effects of metformin on weight loss: potential mechanisms. Curr. Opin. Endocrinol. Diabetes Obes. 21, 323–329 21. Perry, B. and Wang, Y. (2012) Appetite regulation and weight control: the role of gut hormones. Nutr. Diabetes 2, e26 22. Kojima, M. et al. (1999) Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402, 656–660 23. Tschop, M. et al. (2000) Ghrelin induces adiposity in rodents. Nature 407, 908–913 24. Foster-Schubert, K.E. and Cummings, D.E. (2006) Emerging therapeutic strategies for obesity. Endocr. Rev. 27, 779–793 25. Perez-Tilve, D. et al. (2011) Ghrelin-induced adiposity is independent of orexigenic effects. FASEB J. 25, 2814–2822 26. Spencer, S.J. et al. (2012) Ghrelin regulates the hypothalamic– pituitary–adrenal axis and restricts anxiety after acute stress. Biol. Psychiatry 72, 457–465 27. Poretti, M.B. et al. (2015) Ghrelin effects expression of several genes associated with depression-like behavior. Prog. Neuropsychopharmacol. Biol. Psychiatry 56, 227–234 28. Feijo Fde, M. et al. (2011) [Serotonin and hypothalamic control of hunger: a review]. Rev. Assoc. Med. Bras. 57, 74–77 29. Wurtman, R.J. et al. (1980) Precursor control of neurotransmitter synthesis. Pharmacol. Rev. 32, 315–335 30. Schwartz, D.H. et al. (1989) Feeding increases extracellular serotonin in the lateral hypothalamus of the rat as measured by microdialysis. Brain Res. 479, 349–354 31. Rouch, C. et al. (1999) Determination, using microdialysis, of hypothalamic serotonin variations in response to different macronutrients. Physiol. Behav. 65, 653–657
43. Lam, D.D. et al. (2008) Serotonin 5-HT2C receptor agonist promotes hypophagia via downstream activation of melanocortin 4 receptors. Endocrinology 149, 1323–1328 44. Rowland, N.E. et al. (2010) Effect of serotonergic anorectics on food intake and induction of Fos in brain of mice with disruption of melanocortin 3 and/or 4 receptors. Pharmacol. Biochem. Behav. 97, 107–111 45. Weizman, R. et al. (1986) Reduced 3H-imipramine binding but unaltered 3H-serotonin uptake in platelets of adolescent enuretics. Psychiatry Res. 19, 37–42 46. Monteleone, P. et al. (1998) Prolactin response to D-fenfluramine is blunted in people with anorexia nervosa. Br. J. Psychiatry 172, 439–442 47. Kaye, W.H. et al. (1984) Differences in brain serotonergic metabolism between nonbulimic and bulimic patients with anorexia nervosa. Am. J. Psychiatry 141, 1598–1601 48. Ward, A. et al. (1998) Neuroendocrine, appetitive and behavioural responses to D-fenfluramine in women recovered from anorexia nervosa. Br. J. Psychiatry 172, 351–358 49. Steiger, H. (2004) Eating disorders and the serotonin connection: state, trait and developmental effects. J. Psychiatry Neurosci. 29, 20–29 50. Kaye, W.H. et al. (1991) Altered serotonin activity in anorexia nervosa after long-term weight restoration Does elevated cerebrospinal fluid 5-hydroxyindoleacetic acid level correlate with rigid and obsessive behavior? Arch. Gen. Psychiatry 48, 556–562 51. Gauthier, C. et al. (2014) Symptoms of depression and anxiety in anorexia nervosa: links with plasma tryptophan and serotonin metabolism. Psychoneuroendocrinology 39, 170–178 52. Westen, D. and Harnden-Fischer, J. (2001) Personality profiles in eating disorders: rethinking the distinction between axis I and axis II. Am. J. Psychiatry 158, 547–562 53. Bailer, U.F. et al. (2005) Altered brain serotonin 5-HT1A receptor binding after recovery from anorexia nervosa measured by positron emission tomography and [carbonyl11C]WAY-100635. Arch. Gen. Psychiatry 62, 1032–1041 54. Eddy, K.T. et al. (2015) Appetite regulatory hormones in women with anorexia nervosa: binge-eating/purging versus restricting type. J. Clin. Psychiatry 76, 19–24
32. Latham, C.J. and Blundell, J.E. (1979) Evidence for the effect of tryptophan on the pattern of food consumption in free feeding and food deprived rats. Life Sci. 24, 1971–1978
55. Sebaaly, J.C. et al. (2013) Use of fluoxetine in anorexia nervosa before and after weight restoration. Ann. Pharmacother. 47, 1201–1205
33. Morris, P. et al. (1987) Food intake and selection after peripheral tryptophan. Physiol. Behav. 40, 155–163
56. Marzola, E. et al. (2015) Atypical antipsychotics as augmentation therapy in anorexia nervosa. PLoS ONE 10, e0125569
34. Line, S.J. et al. (2011) Opposing alterations in anxiety and speciestypical behaviours in serotonin transporter overexpressor and knockout mice. Eur. Neuropsychopharmacol. 21, 108–116
57. Owens, M.J. and Nemeroff, C.B. (1994) Role of serotonin in the pathophysiology of depression: focus on the serotonin transporter. Clin. Chem. 40, 288–295
35. Lan, M.Y. et al. (2009) Serotonin transporter gene promoter polymorphism is associated with body mass index and obesity in non-elderly stroke patients. J. Endocrinol. Invest. 32, 119– 122
58. Montgomery, S.A. and Kasper, S. (1995) Comparison of compliance between serotonin reuptake inhibitors and tricyclic antidepressants: a meta-analysis. Int. Clin. Psychopharmacol. 9 (Suppl. 4), 33–40
36. Hernandez-Munoz, S. and Camarena-Medellin, B. (2014) [Role of serotonin transporter gene in eating disorders]. Rev. Colomb. Psiquiatr. 43, 218–224 37. Carvajal, A. et al. (2000) Efficacy of fenfluramine and dexfenfluramine in the treatment of obesity: a meta-analysis. Methods Find. Exp. Clin. Pharmacol. 22, 285–290
59. Berg, K.A. et al. (2005) Physiological relevance of constitutive activity of 5-HT2A and 5-HT2C receptors. Trends Pharmacol. Sci. 26, 625–630 60. Gryglewski, G. et al. (2014) Meta-analysis of molecular imaging of serotonin transporters in major depression. J. Cereb. Blood Flow Metab. 34, 1096–1103
38. Vickers, S.P. et al. (2001) Evidence that hypophagia induced by Dfenfluramine and D-norfenfluramine in the rat is mediated by 5HT2C receptors. Neuropharmacology 41, 200–209
61. Tomita, T. et al. (2014) The influence of 5-HTTLPR genotype on the association between the plasma concentration and therapeutic effect of paroxetine in patients with major depressive disorder. PLoS ONE 9, e98099
39. Heisler, L.K. et al. (2007) Serotonin activates the hypothalamic– pituitary–adrenal axis via serotonin 2C receptor stimulation. J. Neurosci. 27, 6956–6964
62. Fava, M. (2000) Weight gain and antidepressants. J. Clin. Psychiatry 61 (Suppl. 11), 37–41
40. Garfield, A.S. and Heisler, L.K. (2009) Pharmacological targeting of the serotonergic system for the treatment of obesity. J. Physiol. 587, 49–60 41. Dwarkasing, J.T. et al. (2015) Differences in food intake of tumourbearing cachectic mice are associated with hypothalamic serotonin signalling. J. Cachexia Sarcopenia Muscle 6, 84–94 42. Dwarkasing, J.T. et al. (2015) Hypothalamic inflammation and food intake regulation during chronic illness. Peptides Published online July 6, 2015. http://dx.doi.org/10.1016/j.peptides.2015. 06.011
63. Sussman, N. et al. (2001) Effects of nefazodone on body weight: a pooled analysis of selective serotonin reuptake inhibitor- and imipramine-controlled trials. J. Clin. Psychiatry 62, 256–260 64. Stanquini, L.A. et al. (2015) Prelimbic cortex 5-HT1A and 5-HT2C receptors are involved in the hypophagic effects caused by fluoxetine in fasted rats. Pharmacol. Biochem. Behav. 136, 31–38 65. Healy, D. (2015) Serotonin and depression. BMJ 350, h1771 66. Murrough, J.W. (2012) Ketamine as a novel antidepressant: from synapse to behavior. Clin. Pharmacol. Ther. 91, 303–309 67. Fournier, J.C. et al. (2010) Antidepressant drug effects and depression severity: a patient-level meta-analysis. JAMA 303, 47–53
Trends in Molecular Medicine, Month Year, Vol. xx, No. yy
9
TRMOME 1091 No. of Pages 10
68. Kirsch, I. et al. (2008) Initial severity and antidepressant benefits: a meta-analysis of data submitted to the Food and Drug Administration. PLoS Med. 5, e45
79. Beglinger, C. et al. (2001) Loxiglumide, a CCK-A receptor antagonist, stimulates calorie intake and hunger feelings in humans. Am. J. Physiol. Regul. Integr. Comp. Physiol. 280, R1149–R1154
69. Hieronymus, F. et al. (2015) Consistent superiority of selective serotonin reuptake inhibitors over placebo in reducing depressed mood in patients with major depression. Mol. Psychiatry Published online April 28, 2015. http://dx.doi.org/10.1038/mp.2015.53
80. Adrian, T.E. et al. (1985) Human distribution and release of a putative new gut hormone, peptide YY. Gastroenterology 89, 1070–1077
70. Koch, M. et al. (2015) Hypothalamic POMC neurons promote cannabinoid-induced feeding. Nature 519, 45–50 71. Horder, J. et al. (2012) Effects of 7 days of treatment with the cannabinoid type 1 receptor antagonist, rimonabant, on emotional processing. J. Psychopharmacol. 26, 125–132 72. Dantzer, R. (2006) Cytokine, sickness behavior, and depression. Neurol. Clin. 24, 441–460 73. Plata-Salaman, C.R. (2000) Central nervous system mechanisms contributing to the cachexia–anorexia syndrome. Nutrition 16, 1009–1012 74. Whitaker, K.W. and Reyes, T.M. (2008) Central blockade of melanocortin receptors attenuates the metabolic and locomotor responses to peripheral interleukin-1beta administration. Neuropharmacology 54, 509–520 75. De La Garza, R.2nd (2005) Endotoxin- or pro-inflammatory cytokine-induced sickness behavior as an animal model of depression: focus on anhedonia. Neurosci. Biobehav. Rev. 29, 761–770 76. Dantzer, R. et al. (2008) From inflammation to sickness and depression: when the immune system subjugates the brain. Nat. Rev. Neurosci. 9, 46–56 77. Dantzer, R. (2001) Cytokine-induced sickness behavior: mechanisms and implications. Ann. N. Y. Acad. Sci. 933, 222–234 78. Janszky, I. et al. (2005) Self-rated health and vital exhaustion, but not depression, is related to inflammation in women with coronary heart disease. Brain Behav. Immun. 19, 555–563
10
Trends in Molecular Medicine, Month Year, Vol. xx, No. yy
81. Keire, D.A. et al. (2000) Primary structures of PYY [Pro(34)]PYY, and PYY-(3-36) confer different conformations and receptor selectivity. Am. J. Physiol. Gastrointest. Liver Physiol. 279, G126–G131 82. Cahill, F. et al. (2011) Serum peptide YY in response to short-term overfeeding in young men. Am. J. Clin. Nutr. 93, 741–747 83. Schubert, M.M. et al. (2014) Acute exercise and hormones related to appetite regulation: a meta-analysis. Sports Med. 44, 387–403 84. Sumithran, P. et al. (2011) Long-term persistence of hormonal adaptations to weight loss. N. Engl. J. Med. 365, 1597–1604 85. Herrmann, C. et al. (1995) Glucagon-like peptide-1 and glucosedependent insulin-releasing polypeptide plasma levels in response to nutrients. Digestion 56, 117–126 86. Punjabi, M. et al. (2014) Circulating glucagon-like peptide-1 (GLP1) inhibits eating in male rats by acting in the hindbrain and without inducing avoidance. Endocrinology 155, 1690–1699 87. Wynne, K. et al. (2006) Oxyntomodulin increases energy expenditure in addition to decreasing energy intake in overweight and obese humans: a randomised controlled trial. Int. J. Obes. (Lond.) 30, 1729–1736 88. Wynne, K. et al. (2005) Subcutaneous oxyntomodulin reduces body weight in overweight and obese subjects: a double-blind, randomized, controlled trial. Diabetes 54, 2390–2395 89. Bhat, V.K. et al. (2013) A novel GIP–oxyntomodulin hybrid peptide acting through GIP, glucagon and GLP-1 receptors exhibits weight reducing and anti-diabetic properties. Biochem. Pharmacol. 85, 1655–1662