74483 729.734

British Journal of Nutrition (2003), 90, 729–734 The cannabinoid system: a role in both the homeostaticand hedonic control of eating? Neuroendocrine and Obesity Biology Unit, Department of Medicine, University of Liverpool,University Clinical Departments, Liverpool L69 3GA, UK (Received 16 May 2003 – Revised 19 May 2003 – Accepted 20 May 2003) Knowledge of the cannabinoid system and its components has expanded greatly over the past decade. There is increasing evidence for itsrole in the regulation of food intake and appetite. Cannabinoid system activity in the hypothalamus is thought to contribute to the homeo-static regulation of energy balance, under the control of the hormone leptin. A second component of cannabinoid-mediated food intakeappears to involve reward pathways and the hedonic aspect of eating. With the cannabinoid system contributing to both regulatory path-ways, it presents an attractive therapeutic target for the treatment of both obesity and eating disorders.
Anandamide: Endocannabinoids: 2-Arachidonoyl glycerol: SR 141716: Food intake: Obesity: Reward The regulation of energy homeostasis and feeding beha- viour is highly complex. It depends on the brain beingable to read, interpret and integrate a wide range of signals The cannabinoid system consists of two receptors (termed and to make appropriate changes in food intake and energy CB1 and CB2), their endogenous ligands (the endocannabi- expenditure as a result of the information. Responsibility noids) and the uptake mechanisms and hydrolysing for this control is shared between several brain regions, enzymes that regulate ligand levels.
spanning both higher and lower centres (cortex to brain- The cannabinoid receptors belong to the 7-transmembrane stem), within which are located numerous neurochemical G-protein coupled receptor family. CB1 is known as the transmitters. Regulatory activities of this complexity are central receptor subtype and is expressed at particularly likely to be controlled by a number of transmitters operat- high levels in brain regions including the cortex, basal ing at a variety of levels. Novel information regarding the ganglia, cerebellum and hippocampus (Glass et al. 1997; neuronal circuits that control food intake continues to Harrold et al. 2002). However, the distribution of CB1 is extend our understanding of energy homeostasis. The not limited to brain circuitry, with receptors recently ident- present review will focus on one neuronal system, the ified on nerve terminals innervating the gastrointestinal tract (Croci et al. 1998; Hohmann & Herkenham, 1999).
In the past decade, cannabinoid receptors and their puta- By contrast, expression of CB2, the peripheral receptor, is tive ligands have been discovered within the central nervous restricted to sites at the periphery, mostly within immune system and linked to a number of aspects of feeding cells. There is evidence for the existence of a further centrally behaviour, including a potential role in the regulation of located cannabinoid receptor, as certain effects of centrally food intake. Recently, interest has revived in the effects on administered cannabinoid ligands are not inhibited by the appetite of the plant-derived cannabinoids and analogous CB1-specific antagonist SR 141716 (Welch et al. 1998). It molecules. The present article will discuss current advances is also possible that the endocannabinoids may exert some in this area and will also consider the potential of the of their pharmacological actions by non-receptor-mediated cannabinoid system as a therapeutic target in the control of mechanisms, e.g. membrane perturbations and gap junction Abbreviations: AG, arachidonoyl glycerol; CB, cannabinoid receptor; FAAH, fatty acid amide hydoxylase; THC, tetrahydrocannabinol.
* Corresponding author: Dr Joanne A. Harrold, fax þ 44 151 706 5797, email harrold@liverpool.ac.uk The psychoactive ingredient of marijuana, D9-tetrahy- exogenous cannabinoids, e.g. D9-THC and the endogenous drocannabinol (THC) is known to interact with CB1 recep- cannabinoids, anandamide and 2-AG, are reliably reported tors (Ledent et al. 1999). It mimics the effects of the to stimulate feeding (Williams et al. 1998; Williams & endogenous cannabinoids, the first of which was identified Kirkham, 1999; Hao et al. 2000). The hyperphagia in porcine brain in 1992 and termed anandamide from ‘ananda’ meaning ‘bliss’ (Devane et al. 1992; Di Marzo D9-THC stimulates feeding as potently as central injection et al. 1998a). To date, three endocannabinoids have been of neuropeptide Y (Corp et al. 1990). As the hyperphagia identified, with the inclusion of 2-arachidonoyl glycerol is selectively blocked by the CB1 receptor antagonist SR (AG) and very recently nolodin ether (Hanus et al.
141716, but not by an antagonist of the peripheral CB2 2001). Anandamide is widely distributed within the brain.
receptors (SR 144258), this suggests that the actions are However, its basal levels are low compared with most neu- mediated by the central receptors. This is further sup- rotransmitters, with the lipophilic compound being syn- ported by the observation that mice with genetically thesised on demand and immediately released from nerve impaired CB1 receptors eat less than their wild type litter- terminals by a Ca2þ dependent mechanism. Anandamide mates in response to food deprivation (Di Marzo et al.
is inactivated by reuptake via the anandamide membrane transporter (Day et al. 2001) and rapid degradation by These observations suggest that tonic cannabinoid fatty acid amide hydroxylase (FAAH)-mediated hydrolysis release may be crucial to the normal regulation of feed- (Giuffrida et al. 2001). 2-AG is thought to be similarly ing. Direct measurements of brain endocannabinoid regulated by the anandamide membrane transporter and levels in response to fasting, feeding and satiation further FAAH, both of which are distributed in brain areas in a support this observation. Fasting increases levels of ana- pattern corresponding to that of CB1 receptors. It is too damide and 2-AG in the nucleus accumbens, and to a early to apply these principles to nolodin ether.
lesser extent the hypothalamus, where 2-AG levels also Endocannabinoids are implicated in a variety of physio- declined with feeding (Kirkham et al. 2002). No changes logical functions including pain reduction, motor regulation, were detected in satiated rats and levels in the cerebellum, learning and memory, appetite stimulation and reward. In a control region not directly involved in the control of some of these functions the cannabinoids play a modulatory feeding, were unaffected regardless of nutritional state role, whilst in others they are essential system components.
The mechanisms of cannabinoid-induced hyperphagia remain to be elucidated. However, there is a body of evi-dence that points towards an involvement of both reward processes and established homeostatic pathways, many of There is increasing evidence for a role of the cannabinoid which are regulated by the hormone leptin and operate system in the regulation of food intake and appetite. Both Table 1. Summary of the cannabinoid system-mediated regulation of energy homeostasis, indicating the known influence of perturbations of energy balance and drug administration on receptor density and endocannabinoid levels CNS, central nervous system; VMH, ventromedial hypothalamic nucleus.
1 Subpopulations of cannabinoid receptor containing neurones appear to exist, playing roles in both the hedonic and homeostatic control of food intake. Whether these represent functional independent populations is unclear. It has been shown that cannabinoid receptor-containing neurones in the hypothalamus are allintrinsic to this brain region (Romero et al. 1998). However, modulation of brain reward circuitry by leptin has also been reported (Fulton et al. 2000).
2 Peripheral anandamide may promote feeding by acting on specific hypothalamic areas important in the control of food intake. However, endocannabinoids are rapidly hydrolysed in the intestine and may not reach the brain in sufficient quantities to interact with central CB1 receptors (Di Marzo et al. 1998b). Alternatively,signals from the viscera indicating cannabinoid-mediated alterations of gastric activity may converge on the nucleus of the solitary tract in the medulla, fromwhere inputs are relayed to the hypothalamus.
Hedonic mechanisms for regulating food intake satiated rats induces significant hyperphagia (Jamshidi &Taylor, 2001). Finally, defective leptin signalling in ob/ob Brain reward systems allow the reinforcement of responses and db/db mice and fa/fa Zucker rats is associated with elev- that have no homeostatic value. Motivation and reward ated hypothalamic endocannabinoid levels, with these levels have been studied most extensively in the context of being reduced in ob/ob mice following leptin treatment drug addiction. However, a number of studies suggest that food reward and drug reward pathways may share Genetically obese rodents exhibit a continuous motivation some common components, including evidence that the to eat and thus demonstrate extreme hyperphagia. The cannabinoid system plays roles in both feeding and evidence presented earlier implicates the endocannabinoid system in this mechanism, possibly acting as a component Association of the cannabinoid system with reward pro- of a leptin-sensitive regulatory pathway. Interestingly, a cesses is indicated by a number of lines of evidence.
deficient leptin function has been reported in dietary-obese SR 141716 antagonises the hunger induced by anandamide animals, with the development of leptin resistance and 2-AG. However, the compound also produces changes (Widdowson et al. 1997a), which could modulate the in ingestive behaviour when administered alone. SR 141716 relationship between plasma leptin levels and cannabinoid selectively inhibits consumption of palatable food and drink, system activity. It is tempting to speculate that the with decreased intake of sucrose, alcohol and a sweet diet hyperphagia demonstrated by dietary obese animals may observed in rats, mice and marmosets respectively (Arnone also arise from an increased hypothalamic endocannabinoid et al. 1997; Simiand et al. 1998). However, it has little system activity occurring as a consequence of reduced leptin effect on bland food consumption. These results suggest regulation. However, evidence has recently come to light that that the central cannabinoid system may act to amplify refutes this argument. Hypothalamic 2-AG levels have been found to increase with food deprivation and decline with This is further supported by the observation that CB1 feeding (Kirkham et al. 2002), suggesting that once initiated, receptors are expressed particularly in areas of the brain eating no longer depends on hypothalamic endocannabinoids such as the nucleus accumbens, the hippocampus and the for maintenance. Furthermore, no relationship has been entopeduncular nucleus; these areas are either directly identified between CB1 receptor binding density and leptin involved in hedonic aspects of eating or are connected to in dietary obese animals (as discussed later).
reward-related brain areas (Finkelstein et al. 1996; Gorba-chevskaia, 1999; Pecina & Berridge, 2000). In addition, thecannabinoids appear to interact with known opioidergic A role for the cannabinoid system in common human reward pathways, indicated by synergistic actions of SR 141716 and the opioid receptor antagonist, naloxone, on Unselected Wistar rats given a palatable diet overeat to a food intake (Welch & Eads, 1999; Kirkham & Williams, variable degree, with approximately half the animals becoming significantly obese (Harrold et al. 2000). This Evidence in human subjects also supports specific dietary-induced obesity, attributable to voluntary hyperpha- cannabinoid involvement in food (orosensory) reward.
gia, is the closest approximation to common lifestyle-related For example, hyperphagic effects of marijuana in human obesity in man, in which overconsumption of palatable food volunteers were principally attributed to an increase in is an important contributing factor. Recent evidence points the consumption of highly palatable sweet foods such as to the conclusion that the endogenous cannabinoids, acting chocolate and biscuits (Iverson, 2000).
on discrete extrahypothalamic populations of CB1 receptors,may drive appetite for palatable food and thus lead to thedevelopment of dietary-induced obesity. Rats fed a palatable Cannabinoids, leptin and the hypothalamus diet for 10 weeks demonstrated reduced CB1 receptor den- Several lines of evidence suggest that the cannabinoids are sity in the forebrain and hippocampus, consistent with modulated by leptin and this may be involved in the control increased activation of the receptors by endogenous cannabi- of feeding First, leptin administration decreases noids. By contrast, CB1 receptor binding in the hypothala- hypothalamic levels of anandamide and 2-AG; endogenous mus was low and unaltered. Furthermore a lack of cannabinoid levels in the only extrahypothalamic site exam- correlation between receptor density and plasma leptin ined, the cerebellum, were reportedly unaffected (Di Marzo suggests that this receptor activity is not regulated by the et al. 2001). Curiously, CB1 receptor density, as determined circulating hormone (Harrold et al. 2002).
by autoradiography, is relatively sparse within the hypo- The anatomical localisation of these changes is notable, thalamus (Harrold et al. 2002), although studies using drawing attention away from the hypothalamus. The unal- [35S]guanylyl 50-[g-[35S]thio]-triphosphate binding indicate tered hypothalamic receptor density argues against a role that receptor coupling to G proteins is more efficient in the for hypothalamic cannabinoids in driving appetite in dietary hypothalamus than in areas with a higher receptor density obesity. It is possible that hypothalamic cannabinoids act to (Breivogel et al. 1997). In addition, studies have shown stimulate feeding under particular circumstances, e.g. star- that anandamide increases Fos expression in the paraventric- vation, when falling leptin and insulin levels are known to ular nucleus of the rat hypothalamus, which plays an import- activate other orexigenic systems such as neuropeptide ant role in the regulation of energy balance (Wenger et al.
Y. Unlike the cannabinoid system, these pathways are 1997; Patel et al. 1998). Furthermore, anandamide adminis- reported to be switched off under conditions of excess tration into the ventromedial hypothalamic nucleus of intake of palatable food (Widdowson et al. 1997b).
Therefore, pharmacological targeting of the cannabinoid oleoylethanolamide (the oleic acid analogue of ananda- system may prove particularly useful in the treatment of human obesity. This is supported by the recent observation increases after a meal in conjunction with reductions in ana- that SR 141716-treated dietary obese mice demonstrate tran- ndamide (Rodrı´guez de Fonseca et al. 2001; Go´mez et al.
sient reductions in food intake. Sustained falls in body 2002). It is possible that both act in a coordinated way to weight and adiposity were also reported. These were attrib- control food intake and gastric motility via opposing actions uted to the hypophagia, potentially in conjunction with a thermogenic or metabolic effect, as treated animals demon- Recently, a peripheral role for CB1 receptors in meta- strated significantly greater weight loss following a 24 h fast bolic regulation has been indicated by the observation than vehicle-treated controls (Ravinet Trillou et al. 2003).
that SR 141716 increases Acrp30 (more commonlyknown as adiponectin) mRNA expression in adiposetissue of obese fa/fa rats and in cultures of adipocytes (Bensaid et al. 2003). Adiponectin induces non-esterified Despite the existence of central mechanisms for the regu- fatty acid oxidation, decreases hyperglycaemia and hyper- lation of food intake by the endocannabinoids, evidence insulinaemia and reduces body weight. This regulation suggests that they may also promote feeding by acting at per- may play a role in the body weight reduction induced by ipheral sites Indeed, CB1 receptors are located on SR 141716, with metabolic regulation contributing to its nerve terminals innervating the gastrointestinal tract, which are involved in mediating gut-derived satiety signals (Crociet al. 1998; Hohmann & Herkenham, 1999). In addition, cap- saicin-induced deafferentation prevents changes in feedingelicited by the administration of cannabinoid drugs (Go´mez The ability of marijuana to increase hunger has been et al. 2002). Moreover, the peripheral administration of noticed for centuries. Despite the public concern related CB1 agonists and antagonists and the acute administration to the abuse of marijuana and its derivatives, scientific of peripherally acting satiety factors or feeding inhibitors, studies have highlighted their ability to stimulate appetite, such as gastrointestinal hormones and the non-cannabinoid especially for sweet and palatable food, and point to the anandamide analogue oleamide, induce similar patterns of future therapeutic potentials of cannabinoid compounds c-fos expression in hypothalamic and brainstem areas regu- in the treatment of obesity and eating disorders.
lating food intake (Rodrı´guez de Fonseca et al. 1997).
Finally, central administration of SR 141716 has no effect on food intake in food-deprived animals. SR 141716 isactive only after intraperitoneal or oral administration, but Application of cannabinoid effects include the treatment of not after subcutaneous injection, further supporting the wasting diseases in which patients are unable or unwilling hypothesis of peripheral actions of cannabinoids on food to eat. Indeed, D9-THC is used clinically for this purpose, particularly in AIDS and cancer patients (Mechoulam & There is some controversy as to whether peripheral Fride, 2001). Anorexia also develops with old age in anandamide also promotes feeding by acting on specific man. This is analogous to the decline in food intake hypothalamic areas involved in energy homeostasis. For observed in old mice. An age-dependent decline in alcohol example, diets containing polyunsaturated non-esterified preference has also been observed (Wang et al. 2003). This fatty acids are known to enhance anandamide levels in is absent in CB1 receptor knockout mice, independent of some brain structures of newborn pigs and mice (Berger their age, suggesting that the decline is related to loss of et al. 2001). Furthermore, food deprivation for 24 h increases cannabinoid signalling in relevant brain areas (Wang et al.
intestinal anandamide concentrations 7-fold, reaching levels 2003). No age-dependent change in anandamide, 2-AG or that are 3-fold greater than those needed to half-maximally CB1 receptor density have been detected in wild type activate CB1 receptors (Devane et al. 1992). However, as mice, suggesting that a decrease in ligand or receptor only 1·6 – 5·0 % of orally administered cannabinoids survive number is unlikely to account for the decline. In fact, a their passage through the digestive system and enter the reduction in agonist stimulated guanylyl 50-[g-[35S]thio]- bloodstream (Di Marzo et al. 1998b), probably due to the triphosphate labelling in old mice suggests that a localised high levels of the enzyme that degrades the compounds in decline in the coupling of CB1 receptors to G-proteins may the gastrointestinal tract (FAAH), this suggests that levels account for reductions in food intake and alcohol prefer- are too low to cause considerable central effects. This is sup- ence. Accordingly, treatment with low doses of ananda- ported by the observation that no increases in brain levels of mide is able to cause a small but significant increase in anandamide occur after 24 h of food deprivation (Go´mez voluntary alcohol intake in old mice (Wang et al. 2003).
Although anandamide binds and activates the CB1 It is hypothesised that the raised gut anandamide levels receptor in vitro, the compound produces only weak and following food deprivation may serve as a short-range transient cannabinoid effects in vivo, thus limiting its effec- hunger signal to promote feeding. Elevated anandamide tiveness as a means of treatment. This probably arises as a may also play a role in regulating gastric emptying and result of anandamide’s rapid catabolism (Adams et al.
intestinal peristalsis, both processes being inhibited by the 1998). Indeed, the half-life of anandamide appears to endocannabinoids (Calignano et al. 1997; Izzo et al.
be in the order of minutes (Willoughby et al. 1997). One 1999). Interestingly, intestinal levels of anandamide and candidate enzyme for regulating anandamide activity is FAAH. Mice lacking FAAH are severely impaired in their The location of the anorectic actions of SR 141716 is ability to degrade anandamide, and when treated with the not clear. As the receptor antagonist is able to cross the CB1 receptor ligand exhibit intense CB1 mediated effects blood – brain barrier, it has been assumed that the effects that are inhibited by SR 141716 (Cravatt et al. 2001).
have a central origin. However, CB1 receptors are not Thus, FAAH may represent an attractive pharmacological exclusive to the brain. High effectiveness of intraperitoneal target, with inhibitors of the enzyme (whose actions and oral administration suggests that further studies of gut would only be evident as sites where endocannabinoid pro- mechanisms of action are warranted. Furthermore, it may duction and release is taking place) serving as therapeutic prove useful to combine cannabinoid antagonists with agents for the treatment of cachexia. To this end, several agents acting at other neurotransmitter systems implicated exceptionally potent inhibitors of FAAH are being investi- in the control of food intake, e.g. opioid systems.
Another potential compound is oleamide, a lipid found in cerebrospinal fluid which causes similar pharmacologi-cal effects to anandamide in mice. Only anandamide binds to the CB1 receptor, but by inhibiting FAAH andthus increasing the concentration of anandamide, oleamide Adams IB, Compton DR & Martin BR (1998) Assessment of anandamide interaction with the cannabinoid brain receptor: potentiates the endocannabinoid binding and enhances its SR 141716A antagonism studies in mice and autoradiographic analysis of receptor binding in ratbrain. J Pharmacol Exp Ther A similar entourage effect has been observed for 2-AG with actions of the endocannabinoid being enhanced Arnone M, Maruani J, Chaperon F, et al. (1997) Selective inhi- in vitro and in vivo by co-administration of other fatty bition of sucrose and ethanol intake by SR 141716, an antagonist acid glycerol esters that coexist with 2-AG in the brain of central cannabinoid (CB1) receptors. Psychopharmacology (Ben-Shabat et al. 1998). These compounds (e.g. 2-lino- leoylglycerol and 2-palmitoylglycerol) have no intrinsic Ben-Shabat S, Fride E, Sheskin T, et al. (1998) An entourage activity at the CB1 receptor, but amplify both the binding effect: inactive endogenous fatty acid glycerol esters enhance of 2-AG and its post-receptor signalling, probably by inhi- 2-arachidonoyl-glycerol cannabinoid activity. Eur J Pharmacol353, 23 – 31.
biting the degradation of 2-AG by FAAH. These com- Bensaid M, Gary-Bobo M, Esclangon A, et al. (2003) The canna- pounds have been shown to enhance certain central binoid CB1 receptor agonist SR 141716 increases Acrp30 actions of 2-AG (motor activity and analgesia) when mRNA expression in adipose tissue of obese fa/fa rats and in given intraperitoneally with the endocannabinoid, in the cultured adipocyte cells. Mol Pharmacol 63, 908 – 914.
ratios at which they are found in the brain. As yet, the Berger A, Crozier G, Bisogno T, Cavaliere P, Innis S & effects of these endogenous enhancers on feeding have Di Marzo V (2001) Anandamide and diet: Inclusion of dietary not been investigated, but it is predicted that they would arachidonate and docosahexaenoate leads to increased brains enhance the ability of endogenous 2-AG to increase levels of the corresponding N-acylethanolamines in piglets.
Proc Natl Acad Sci USA 98, 6402 – 6406.
Boger DL, Sato H, Lerner AE, Guan X & Gilula NB (1999) Arachidonic acid amide inhibitors of gap junction cell – cell communication. Bioorg Med Chem Lett 9, 1151 – 1154.
Boger DL, Sato H, Lerner AE, et al. (2000) Exceptionally potent Unlike the cannabinoid system, other appetite stimulating inhibitors of fatty acid amide hydrolase: the enzyme responsible systems such as neuropeptide Y/agouti gene-related pep- for degradation of endogenous oleamide and anandamide. Proc tide are reportedly switched off under conditions of excess Natl Acad Sci USA 97, 5044 – 5049.
intake (Widdowson et al. 1997b). Therefore, pharmacologi- Breivogel CS, Sims LJ & Childers SR (1997) Regional cal targeting of the cannabinoid system may prove useful in differences in cannabinoid receptor/G-protein coupling in rat the treatment of lifestyle-related obesity in human subjects.
brain. J Pharmacol Exp Ther 282, 1632 – 1642.
It is not yet clear to what extent pharmacological agents Calignano A, La Rana G, Makriyannis A, Lin SY, Beltramo M & Piomelli D (1997) Inhibition of intestinal motility by acting on this system may have sustained actions and appli- anandamide, an endogenous cannabinoid. Eur J Pharmacol cability to different feeding regimens. For CB1 antagonists to be useful anorectic drugs, their effects would have to be Colombo G, Agabio R, Diaz G, Lobina C, Reali R & Gessa GL sustained over days and apply to more than just sweet (1998) Appetite suppression and weight loss after the food. It has been reported that the anorectic actions of SR cannabinoid antagonist SR 141716. Life Sci 63, 113 – 117.
141716 disappear within 3 – 6 d of treatment in rats, Corp ES, Melville LD, Greenberg D, Gibbs J & Smith GP suggesting that tolerance develops, but weight loss is sus- (1990) Effect of fourth ventricular neuropeptide Y and peptide tained (Colombo et al. 1998). However, the dosing protocol YY on ingestive and other behaviors. Am J Physiol 259, used in this study was not optimal for sustained effects.
Early results from clinical trials on the experimental drug Supersensitivity to anandamide and enhanced cannabinoid Rimonabant (SR 141716) in the USA and Europe have signaling in mice lacking fatty acid amide hydrolase. Proc Natl been promising, with patients losing up to 4 kg over a 16-week period. However, more research is needed to deter- Croci T, Manara L, Aureggi G, et al. (1998) In vitro functional mine the long-term effects of the drug. Phase III clinical evidence of neuronal cannabinoid CB1 receptors in human trials are currently underway, being due for completion in ileum. Br J Pharmacol 125, 1393 – 1395.
Day TA, Rakhshan F, Deutsch DG & Barker EL (2001) Role of fatty acid amidehydrolase in the transport of the endogenous Endocannabinoid levels in rat limbic forebrain and hypothala- cannabinoid anandamide. Mol Pharmacol 59, 1369 – 1375.
mus in relation to fasting, feeding and satiation: stimulation Devane WA, Hanus L, Breuer A, et al. (1992) Isolation and struc- of eating by 2-arachidonoyl glycerol. Br J Pharmacol 136, ture of a brain constituent that binds to the cannabinoid recep- Ledent C, Valverde O, Cossu G, et al. (1999) Unresponsiveness Di Marzo V, Goparaju SK, Wang L, et al. (2001) Leptin-regulated to cannabinoids and reduced addictive effects of opiates in endocannabinoids are involved in maintaining food intake.
CB1 receptor knockout mice. Science 283, 401 – 404.
Mechoulam R & Fride E (2001) A hunger for cannabinoids.
Di Marzo V, Melck D, Bisogno T & De Petrocellis L (1998a) Endocannabinoids: endogenous cannabinoid receptor ligands Mechoulam R, Fride E, Hanus L, et al. (1997) Anandamide may with neuromodulatory action. Trends Neurosci 21, 521 – 528.
mediate sleep induction. Nature 389, 25 – 26.
Di Marzo V, Sepe N, De Petrocellis L, et al. (1998b) Trick or Patel NA, Moldow RL, Patel JA, Wu G & Chang SL (1998) Arachi- treat from endocannabinoids. Nature 396, 636 – 637.
donylethanolamide (AEA) activation of FOS proto-oncogene Finkelstein DI, Reeves AK & Horne MK (1996) An electron micro- protein immunoreactivity in the rat brain. Brain Res 797, scopic tracer study of the projections from the entopeduncular nucleus to the ventrolateral nucleus of the rat. Neurosci Lett Pecina S & Berridge KC (2000) Opioid sites in nucleus accumbens shell mediate eating and hedonic ‘liking’ for food: map based on Fulton S, Woodside B & Shizgal P (2000) Modulation of brain microinjection Fos plumes. Brain Res 863, 71 – 86.
reward circuitry by leptin. Science 287, 125 – 128.
Rodrı´guez de Fonseca F, Carrera MR, Navarro M, Koob GF & Giuffrida A, Beltramo M & Piomelli D (2001) Mechanisms of Weiss F (1997) Activation of corticotropin-releasing factor in endocannabinoid inactivation: biochemistry and pharmacology.
the limbic system during cannabinoid withdrawal. Science J Pharamacol Exp Ther 298, 7 – 14.
Glass M, Dragunow M & Faull RLM (1997) Cannabinoid receptors Rodrı´guez de Fonseca F, Navarro M, Go´mez R, et al. (2001) An in the human brain: a detailed anatomical and quantitative auto- anorexic lipid mediator regulated by feeding. Nature 414, radiographic study in the fetal, neonatal and adult human brain.
Romero J, Wenger T, de Miguel R, Ramos JA & Fernandez-Ruiz Go´mez R, Navarro M, Ferrer B, et al. (2002) A peripheral JJ (1998) Cannabinoid receptor binding did not vary in several mechanism for CB1 cannabinoid receptor-dependent modulation hypothalamic nuclei after hypothalamic deafferentation. Life of feeding. J Neurosci 22, 9612 – 9617.
Gorbachevskaia AL (1999) Projections from the substantia nigra, Simiand J, Keane M, Keane PE & Soubrie´ P (1998) SR 141716, a ventral tegmental area and amygdala to the pallidium in dog CB1 cannabinoid receptor antagonist, selectively reduces sweet food intake in marmosets. Behav Pharmacol 9, 179 – 181.
Hanus L, Abu-Lafi S, Fride E, et al. (2001) 2-Arachidonyl Ravinet Trillou CR, Arnone M, Delgorge C, et al. (2003) Anti- glycerol ether, an endogenous agonist of the cannabinoid obesity effect of SR 141716, a CB1 receptor antagonist, in CB1 receptor. Proc Natl Acad Sci USA 98, 3662 – 3665.
diet-induced obese mice. Am J Physiol 284, R345 – R353.
Hao S, Avraham Y, Mechoulam R & Berry EM (2000) Low dose Wang L, Liu J, Harvey-White J, Zimmer A & Kunos G (2003) anandamide affects food intake, cognitive function, neurotrans- Endocannabinoid signaling via cannabinoid receptor 1 is mitter and corticosterone levels in diet-restricted mice. Eur J involved in ethanol preference and its age-dependent decline in mice. Proc Natl Acad Sci USA 100, 1393 – 1398.
Harrold JA, Elliott JC, King PJ, Widdowson PS & Williams G Welch SP & Eads M (1999) Synergistic interactions of endogenous (2002) Down-regulation of cannabinoid-1 (CB-1) receptors in opioids and cannabinoid systems. Brain Res 848, 183 – 190.
specific extrahypothalamic regions of rats with dietary obesity: Welch SP, Huffman JW & Lowe J (1998) Differential blockade of a role for endogenous cannabinoids in driving appetite for the antinociceptive effects of centrally administered cannabi- palatable food? Brain Res 952, 232 – 238.
noids by SR 141716A. J Pharmacol Exp Ther 286, 1301 – 1308.
Harrold JA, Williams G & Widdowson PS (2000) Early leptin Wenger T, Jamali KA, Juaneda C, Leonardelli J & Tramu G response to a palatable diet predicts dietary obesity in rats: key (1997) Arachidonylethanolamide (anandamide) activates the role of melanocortin-4 receptors in the ventromedial hypothala- parvocellular part of the hypothalamic paraventricular nucleus.
mic nucleus. J Neurochem 74, 1224 – 1228.
Biochem Biophys Res Commun 237, 724 – 728.
Hohmann AG & Herkenham M (1999) Localization of central can- Widdowson PS, Upton R, Buckingham R, Arch J & Wiliams G nabinoid receptor messenger RNA in neuronal subpopulations of (1997a) Inhibition of food response to intracerebroventricular rat dorsal root ganglia: a double-label in situ hybridization study.
injection of leptin is attenuated in rats with diet-induced obesity.
Iverson LL (2000) The Science of Marijuana. Oxford: Oxford Widdowson PS, Upron R, Henderson L, Buckingham R, Wilson S & Williams G (1997b) Reciprocal regional changes in brain NPY Izzo AA, Mascolo N, Capasso R, Germano MP, De Pasquale R receptor density during dietary restriction and dietary-induced & Capasso F (1999) Inhibitory effect of cannabinoid agonists obesity in the rat. Brain Res 774, 1 – 10.
on gastric emptying in the rat. Naunyn Schmiedebergs Arch Williams CM & Kirkham TC (1999) Anandamide induces over- eating: mediation by central cannabinoid (CB1) receptors.
Jamshidi N & Taylor DA (2001) Anandamide administration into Psychopharmacology 143, 315 – 317.
the ventromedial hypothalamus stimulates appetite in rats. Br J Williams CM, Rogers PJ & Kirkham TC (1998) Hyperphagia in pre-fed rats following oral delta 9-THC. Physiol Behav 15, Kirkham TC & Williams CM (2001) Synergistic effects of opioid and cannabinoid antagonists on food intake. Psychopharma- Willoughby KA, Moore SF, Martin BR & Ellis EF (1997) The bio- disposition and metabolism of anandamide in mice. J Pharmacol Kirkham TC, Williams CM, Fezza F & Di Marzo V (2002)

Source: http://www.nutritionsociety.org.uk/bjn/090/0729/0900729.pdf