Original Paper Neuroendocrinology 2005;81:273–282 DOI: 10.1159/000087925
Received: April 25, 2005 Accepted after revision: July 3, 2005 Published online: August 29, 2005
Cocaine- and Amphetamine-Related Transcript Is Involved in the Orexigenic Effect of Endogenous Anandamide Douglas Osei-Hyiamana Michael Depetrilloa Judith Harvey-Whitea Anthony W. Bannonb Benjamin F. Cravattc Michael J. Kuhard Ken Mackiee Miklós Palkovitsf George Kunosa a
Section on Neuroendocrinology, National Institute on Alcohol Abuse and Alcoholism, National Institutes of Health, Bethesda, Md., b Amgen Inc., Neuroscience, Thousand Oaks, Calif., c Scripps Institute, La Jolla, Calif., d Yerkes National Primate Research Center, Division of Neuroscience, Emory University, Atlanta, Ga., e Departments of Physiology and Anesthesiology, University of Washington, Seattle, Wash., and f Laboratory of Genetics, National Institute on Mental Health, National Institutes of Health, Bethesda, Md., USA
Key Words Food intake behavior Endocannabinoids Cannabinoid receptors Fatty acid amide hydrolase Cocaine- and amphetamine-related transcript peptide Transgenic mice Arcuate nucleus Paraventricular nucleus Periventricular hypothalamic area Amygdala Bed nucleus of the Stria terminalis Immunocytochemistry
Abstract Endocannabinoids acting at CB1 cannabinoid receptors (CB1) increase appetite. In view of the predominant presynaptic localization of CB1 in the brain, we tested the hypothesis that the orexigenic effect of endocannabinoids involves inhibition of the release of a tonically active anorexigenic mediator, such as the peptide product of the cocaine- and amphetamine-related transcript (CART). The CB1 antagonist rimonabant inhibited food intake in food-restricted wild-type mice, but not in their CART-deficient littermates. Mice deficient in fatty acid amide hydrolase (FAAH), the enzyme responsible for the in vivo metabolism of the endocannabinoid anandamide, have reduced levels of CART-immunoreactive nerve fibers and terminals in several brain regions implicated in appetite control, including the arcuate, dorsomedial and periventricular nuclei of the hypothalamus, the amyg-
© 2005 S. Karger AG, Basel Fax +41 61 306 12 34 E-Mail
[email protected] www.karger.com
Accessible online at: www.karger.com/nen
dala, the bed nucleus of the stria terminalis and the nucleus accumbens, and treatment of FAAH–/– mice with rimonabant, 3 mg/kg/day for 7 days, increased CART levels toward those seen in FAAH+/+ wild-type controls. In contrast, no difference in the density of CART-immunoreactive fibers was observed in the median eminence and the paraventricular nucleus of FAAH+/+ and FAAH–/– mice. Acute treatment of wild-type mice with the cannabinoid agonist HU-210 resulted in elevated CART levels in the dorsomedial nucleus and the shell portion of the nucleus accumbens. These observations are compatible with CART being a downstream mediator of the CB1-mediated orexigenic effect of endogenous anandamide. Copyright © 2005 S. Karger AG, Basel
Introduction
Appetite and food intake are regulated through a complex neural circuitry that involves structures in the hypothalamus, limbic forebrain, the amygdala and structures in the brainstem, as well as a growing number of neuropeptides and biogenic amines that are either stimulatory (orexigenic) or inhibitory (anorexigenic) [1–3]. Evidence has accumulated over the last few years to indicate that endocannabinoids act as orexigenic signals via cannabi-
George Kunos NIAAA/NIH, 5625 Fishers Lane Rm. 2S-24, MSC 9413 Bethesda, MD 20892-9413 (USA) Tel. +1 301 4432069, Fax +1 301 4800257, E-Mail
[email protected]
noid CB1 receptors in the central nervous system [4–7], and that blocking CB1 receptors inhibits food intake and results in weight loss in rodents [8–11]. The weight-reducing effect of CB1 antagonists is accentuated in obese animals [12–14] and is being explored for the treatment of human obesity. We have reported that the absence of CB1 receptors in mice attenuates the hunger-induced increase in food intake, and endocannabinoid levels in the hypothalamus are decreased by leptin and increased in leptindeficient states, which suggests that endocannabinoids are part of the leptin-regulated neural circuitry [7]. Fasting and overfeeding-induced changes in endocannabinoid levels detected in the hypothalamus and limbic forebrain implicated both of these brain regions in the appetite controlling effect of endocannabinoids [15]. Other findings suggest that endocannabinoids in the gut may also activate appetitive circuitries in the brain by stimulating CB1 receptors on peripheral sensory nerve terminals [16], and activation of CB1 receptors in the hindbrain can also increase food intake [17]. The synaptic organization of the endocannabinoid/ CB1 receptor system in the brain has unique features. Unlike classical neurotransmitters, endocannabinoids are not stored but rather generated ‘on demand’ in postsynaptic neurons, in response to depolarization and a rise in intracellular calcium or activation of metabotropic receptors [17]. They are released from neurons through a mechanism not yet clarified and are presumed to diffuse retrogradely to presynaptic axon terminals where they stimulate CB1 receptors to inhibit the release of transmitters such as GABA or glutamate [18]. CB1 receptors are the most abundant G-protein-coupled receptors in the mammalian brain and are located almost exclusively presynaptically [17]. This raises the question of how does their activation lead to an increase in appetite and food intake. A plausible explanation may involve a tonically active anorexigenic mediator, the release of which would be presynaptically inhibited by endocannabinoids. We reasoned that if this is the case, then in animals deficient in this particular anorexigenic factor the ability of a CB1 receptor antagonist to reduce food intake would be lost. The three most important anorexigenic factors in the brain are -melanocyte-stimulating hormone (-MSH) [19], the peptide products of cocaine- and amphetaminerelated transcript (CART) [20–22], and corticotropin-releasing hormone (CRH) [23, 24]. In the present study we have explored the role CART. The results indicate that CART is a downstream mediator of the orexigenic effect of endocannabinoids in mice. A preliminary account of some of these findings has been presented [25].
274
Neuroendocrinology 2005;81:273–282
Materials and Methods Materials An antibody raised in rabbit against the C-terminal tail of the rat CB1 receptor [26], and an antibody against the rat CART peptide raised in rabbit against the C-terminal C4 peptide [27] were used. Rimonabant (also known as SR141716, N-[piperidin-1-yl]-5[4-chlorophenyl]-1-[2,4-dichlorophenyl]-4-methyl-1H-pyrazole-3carboxamide HCl) was provided by the National Institute on Drug Abuse Drug Supply Program (Research Triangle Park, N.C., USA). HU-210 ((–)-11-OH-9-tetrahydrocannabinol dimethylheptyl) was from Tocris Cookson (Baldwin, Mo., USA). For intraperitoneal injections, drugs were dissolved in Tocrisolve (Tocris Cookson). Animals CART+/+ and CART–/– mice were generated at Amgen Inc. (Thousand Oaks, Calif., USA) as recently described [28]. Separate production colonies of CART+/+ and CART–/– mice originally derived from littermates were established at Charles River Laboratories (Wilmington, Mass., USA) for use in experimental procedures. The mice were maintained on the outbred (Black Swiss ! 129SvJ) background. All feeding experiments were conducted at Amgen Inc. under protocols approved by Amgen’s IACUC and were performed in Amgen’s AAALAC-approved animal facility. The age range of female mice utilized in these studies was 8–14 weeks. FAAH–/– and FAAH+/+ mice had been developed as described elsewhere [29] and had been backcrossed to a C57Bl/6 genetic background. All animals were housed under a 12-hour light/dark cycle and had access to standard laboratory chow and water ad libitum, except when stated otherwise. Experimental procedures were run under protocols approved by the Institutional Animal Care and Use Committees of NIAAA or Amgen, Inc., as appropriate. Food Intake Studies Food was removed from single-housed mice for approximately 20 h (i.e., overnight). On the day of the experiment, mice were treated with vehicle or 4 mg/kg rimonabant i.p. 15 min prior to placing the animals into individual cages with a pre-weighed amount of food. At 1, 2, 4 and 24 h following treatment, food was re-weighed and the difference between the original food weight and the weight at each time was used to compute cumulative food intake, correcting for spillage, which was minimal. Immunohistochemistry Animals were anesthetized with pentobarbital sodium (65 mg/ kg, i.p.) and the brains were fixed by intracardial perfusion with 4% paraformaldehyde in 0.1 M phosphate buffer (PBS), pH 7.4. The brains were removed, post-fixed in the same fixative solution for 4 h at 4 ° C, then transferred to 30% sucrose in 0.1 M PBS at 4 ° C overnight. Brains were cut in coronal sections of 10 m thickness by using a microtome. The avidin-biotin-peroxidase (ABC, Vectastain) method was used for immunostaining. Sections were kept in a 20% sucrose solution, rinsed in 0.1 M PBS, treated with 0.6% Triton X-100 for 1 h, then the endogenous peroxidase activity was blocked by a 10-min incubation in 3% H2O2. After washing in 0.1 M PBS, sections were treated with 10% normal goat serum for 1 h, rinsed in 0.1 M PBS and incubated with affinity-purified CART or CB1 receptor primary polyclonal antibodies overnight at 4 ° C. Next, sections were incubated for 1 h at room temperature in bio-
Osei-Hyiaman et al.
tinylated goat-anti-rabbit-IgG secondary antibody (1:1,000; Vector Laboratories Inc., Burlingame, Calif., USA), and the sections processed using the ABC technique with a rabbit Vectastain Elite ABCperoxidase kit (Vector Laboratories). The tissue-bond peroxidase was visualized by in 3,3-diaminobenzidine (DAB; 0.5 mg DAB/ml and 0.01% H2O2 in Tris-HCl buffer). Sections were washed in deionized distilled water 3 times for 4 min, dried and coverslipped. For double immunostaining, sections were incubated in the first antibody (CART or CB1 antibodies were used alternatively in different brains) and processed the same way as for single immunostaining. This first antibody was detected by using nickel-enhanced DAB (NiDAB, Sigma Chemical Co., St. Louis, Mo., USA) as the chromogen. The sections were then incubated in the second primary antibody overnight followed by biotinylated anti-rabbit IgG. This second antibody was detected by DAB without intensification. Following the immunohistochemical procedures, sections were dried overnight, dehydrated in ethanol followed by xylene, and coverslipped. The dark blue stained structure represents the first immunoreaction that is well distinguished from the brown secondary immunostaining.
Endocannabinoid Levels in Brain Mice were decapitated and the brain quickly removed. Dissection of the brain into four regions, extraction of the samples and measurement of endocannabinoid levels by liquid chromatography/in line mass spectrometry was done as described [30]. Extraction efficiency was determined using a deuterated internal standard (d4-anandamide), and the amounts of anandamide and 2-arachidonoylglycerol (2-AG) were determined using inverse linear regression of standard curves. Values are expressed as fmol or pmol per mg wet tissue. Statistical Methods For analyses of food intake, independent two-way ANOVA was performed on data from each genotype with treatment (between) and time (within) as factors. Follow-up analysis (where warranted) utilized an unpaired t-test at each time-point (comparing vehicle to rimonabant treatment). Significance level was set at p ! 0.05. All analyses were performed with software from GraphPad Prism (GraphPad, San Diego, Calif., USA). Densitometric images were analyzed using one-way ANOVA for between-group comparisons of FAAH+/+, FAAH–/–, and rimonabant-treated FAAH–/– groups, while the two-tailed Student’s t test was used for comparisons of FAAH+/+ vs. FAAH–/– or vehicletreated vs. HU-210-treated groups. Significant main and interac-
CART Is Involved in the Appetitive Effect of Endocannabinoids
Food intake (g)
✽
1 ✽ ✽✽✽
0 1
2
3
4
3
4
CART –/–
2
Food intake (g)
Quantification of Immunohistochemical Images After mounting, brain sections were analyzed by light microscopy. An Olympus microscope fitted with a Sony CCD-IRIS/RGB video camera linked to a personal computer was used to collect digital images of the sections for quantitative image analysis. The NIH Image 1.63 software was subsequently used to analyze the digitized video images. An 8-bit gray scale image of each stained section was adjusted for equal background so that areas on each section with no immunostaining would have no positive signals. Using the software, grains that were 110 pixels were selected and counted. Densitometric quantification was performed for each bilateral area on 3–8 sections per treated animal and averaged for each animal to yield a single data point for statistical analysis, as described below.
CART +/+
2
1
0 1
2 Time (h)
Fig. 1. Rimonabant inhibits food intake in wild-type but not in
CART-knockout mice. Food intake in 8 CART-knockout mice and 8 wild-type littermates pretreated with vehicle ($) or 4 mg/kg rimonabant i.p. 15 min before testing ()) was measured as described in Methods. Significant difference between corresponding vehicle- and rimonabant-treated groups is indicated by * (p ! 0.05) or *** (p ! 0.001).
tive effects were further analyzed with the appropriate post-hoc tests (Tukey-Kramer multiple comparisons test or Fisher’s exact test). Variations in brain area size were controlled for by careful brain slice selection paying particular attention to location as illustrated in the Rat Brain Atlas of Paxinos and Watson (1998, 4th ed.). Between-group brain region variations with p ! 0.05 were considered statistically significant.
Results
Rimonabant Fails to Influence Food Intake in CART Knockout Mice The effect of the CB1 antagonist rimonabant (4 mg/kg, i.p.) was tested on food intake in mice deficient in CART (CART–/–) and in their wild-type littermates (CART+/+).
Neuroendocrinology 2005;81:273–282
275
Anandamide (fmol/mg tissue)
150
✽
✽
✽
100
✽
50
0 Cb
Hyp
Lf
Amg
Cb
Hyp
Lf
Amg
2-AG (pmol/mg tissue)
15
10
5
0
2
Fig. 2. Anandamide and 2-arachidonoylglycerol (2-AG) levels in
the cerebellum (Cb), limbic forebrain (Lf), hypothalamus (Hyp) and amygdala (Amg) of FAAH+/+ (open columns) and FAAH–/– mice (filled columns). Columns and vertical bars represent means 8 SE from samples from 6–8 separate animals. Note that anandamide, but not 2-AG, levels were dramatically increased in FAAH–/– compared to FAAH+/+ mice (* p ! 0.001). Fig. 3. Network of CART peptide-immunoreactive fibers in the dorsomedial and arcuate nuclei of the hypothalamus and CART peptide-immunoreactive cell bodies in the perifornical area. Coronal sections. A FAAH+/+ mouse; B FAAH–/– mouse; C FAAH–/– mouse treated for 7 days with 4 mg/kg/day rimonabant. AN = Arcuate nucleus; F = fornix; VM = ventromedial nucleus; 3V = third ventricle. Scale bar: 200 m. Densitometric quantitation of immunopositive regions in four replicates in each group yielded values of 34.3 8 2.4 (A), 13.1 8 0.7* (B), and 20.0 8 3.1*# (C) for the dorsomedial nucleus, and 19.9 8 2.2 (A), 10.4 8 1.6* (B), and 29.8 8 3.2*# (C) for the arcuate nucleus. * and # indicate significant difference (p ! 0.05) from value in FAAH+/+ or untreated FAAH–/– mice, respectively.
In CART+/+ mice, rimonabant caused a reduction in cumulative food intake that was statistically significant at 1, 2 and 4 h following treatment, compared to corresponding values in vehicle-treated CART+/+ mice (fig. 1).
276
Neuroendocrinology 2005;81:273–282
3
By 24 h post-treatment, no significant differences were observed with food intake values of 5.8 8 0.3 and 5.4 8 0.2 g, in vehicle and rimonabant-treated mice, respectively. In contrast, in CART–/– mice, treatment with
Osei-Hyiaman et al.
rimonabant did not significantly alter fast-induced food intake at any time-point measured (fig. 1), including 24 h where food intake was 5.3 8 0.2 and 5.8 8 0.5 g in vehicle and rimonabant-treated mice, respectively. As these findings suggested that CART may be a mediator of the effect of endocannabinoids on food intake, we next examined the effects of CB1 stimulation by endogenous and exogenous cannabinoids on the levels of immunoreactive CART in brain regions implicated in appetite control. Rimonabant Treatment Reverses the Reduction of CART Levels Observed in Selected Brain Regions of FAAH–/– Compared to FAAH+/+ Mice The enzyme fatty acid amidohydrolase (FAAH) has been implicated in the in vivo metabolism of anandamide [31], and the genetic ablation or pharmacological inhibition of FAAH unmasks a CB1 receptor-mediated endocannabinergic tone in some behaviors [29, 32]. We tested whether CART levels in the brain of mice lacking FAAH are altered in a way that is compatible with tonic activation of CB1. Brain levels of anandamide were 10- to 15fold higher in FAAH–/– compared to FAAH+/+ mice, whereas 2-AG levels were not different (fig. 2), indicating that FAAH selectively degrades anandamide but not 2-AG. CART peptide was visualized by immunohistochemical staining in brain sections and quantified by densitometry. As illustrated in figure 3, the density of CARTimmunoreactive fibers was visibly lower in both the dorsomedial and arcuate nuclei of the hypothalamus of a FAAH–/– compared to a FAAH+/+ mouse, and treatment of a FAAH–/– mouse with 4 mg/kg/day rimonabant for 7 days resulted in a substantial increase in CART immunoreactivity in both regions. These changes were statistically significant as quantified in replicates obtained from 4 separate animals in each group (see legend of fig. 3). Similar differences were evident in the amygdala (fig. 4),
Fig. 4. Network of CART peptide-immunoreactive fibers in the medial nucleus of the amygdala. A FAAH+/+ mouse; B FAAH–/– mouse; C FAAH–/– mouse treated with 4 mg/kg/day rimonabant for
7 days. Coronal sections. BA = Basal amygdaloid nucleus; IC = internal capsule; OT = optic tract. Scale bar: 200 m. Densitometric quantitation of immunopositive fibers from four replicate animals in each group yielded values of 116.6 8 2.6 (A), 37.7 8 1.3* (B), and 701 8 2.8*# (C), where * and # indicate significant difference (p ! 0.05) from value in A or B, respectively.
CART Is Involved in the Appetitive Effect of Endocannabinoids
Neuroendocrinology 2005;81:273–282
277
Fig. 5. Fine network of CART peptide-im-
munoreactive fibers in the bed nucleus of the stria terminalis. A FAAH+/+ mouse; B FAAH–/– mouse; C FAAH–/– mouse treated with 4 mg/kg/day rimonabant for 7 days. Depletion of immunoreactivity in the FAAH–/– mouse and its restoration by rimonabant treatment is especially clear in the dorsolateral subdivision (arrow). AC = Anterior commissure; LV = lateral ventricle; MP = medial preoptic nucleus. Scale bar: 200 m. Densitometry in the dorsolateral + ventral subdivisions in four replicate animals in each group yielded values of 42.5 8 1.5 (A), 16.4 8 0.6* (B), and 40.0 8 2.1*# (C). Significance indicated as in figures 3 and 4.
Fig. 6. Network of CART peptide-immunoreactive fibers in the hypothalamic paraventricular (PVN) and periventricular nuclei (periVN) in FAAH+/+ (A) and FAAH–/– mice (B). Coronal sections. 3V = Third ventricle. No significant difference in CART immunoreactivity in the paraventricular nucleus (A 39.5 8 2.0 vs. B 36.5 8 3.2, n = 4), but marked reduction in the periventricular nucleus of FAAH–/– mice (A 38.0 8 1.0 vs. B 18.7 8 0.8, p ! 0.01, n = 4).
the bed nucleus of the stria terminalis (fig. 5), and in the shell portion of the nucleus accumbens (not shown). In contrast, no significant difference in CART immunoreactivity was observed in the median eminence (not shown) and in the paraventricular nucleus (PVN) of FAAH+/+ and FAAH–/– mice (fig. 6), although a reduction was seen in the periventricular nucleus ventral to the PVN (fig. 6). Treatment with the CB Agonist HU-210 Increases CART Immunoreactivity CART immunoreactivity was also quantified following acute treatment of wild-type C57Bl/6J mice with a low dose of the potent synthetic cannabinoid agonist HU210. Animals were treated with HU-210, 2 g/kg, i.p., and were sacrificed 1 h later. CART immunoreactivity
278
Neuroendocrinology 2005;81:273–282
was significantly increased in the dorsomedial hypothalamic nucleus (fig. 7) and nucleus accumbens (not shown) after acute HU-210 treatment. Co-Localization of CB1 Receptors and CART in Various Brain Regions Co-localization of CART- and CB1-immunoreactive fibers in the same brain areas was analyzed by double immunostaining. In brain regions where both CART and CB1 are expressed at high levels in nerve fibers and terminals, such as the amygdala or the prefrontal cortex, the co-localization of these two fiber types was evident, as illustrated in figure 8 for the amygdala. In other regions, such as the arcuate nucleus and dorsolateral hypothalamic nuclei, the expression of either CB1- or CART-immu-
Osei-Hyiaman et al.
Discussion
Fig. 7. Acute treatment of control mice with the CB agonist HU210 increases CART peptide immunoreactivity in the hypothalamic dorsomedial nucleus. A Vehicle-treated mouse. B Mouse treated with 2 g/kg HU-210 1 h before sacrifice. Densitometry in a total of 4 mice in each group: 25.3 8 0.9 (A) vs. 32.6 8 1.5 (B), p ! 0.05.
noreactive structures was too low to allow demonstration of co-localization by light microscopy. The interconnection between CART and CB1 in the amygdala is bidirectional: fine, varicose CART-immunoreactive fibers originating in the posteromedial amygdaloid nucleus run into the CB1-rich basolateral amygdaloid nucleus (fig. 8C), while in the basomedial amygdaloid nucleus, CB1-positive nerve terminals of unknown origin (probably from the basolateral amygdala) are located in the CART-immunopositive network of fibers (fig. 8B).
CART Is Involved in the Appetitive Effect of Endocannabinoids
We have observed a striking loss of the ability of the CB1 antagonist rimonabant to inhibit food intake in food-restrained CART knockout mice as compared to their wild-type littermates, which strongly suggests that CART is a downstream mediator of the orexigenic effect of endocannabinoids. In view of the predominantly presynaptic localization of CB1 receptors and their well-documented ability to inhibit transmitter release via calcium channel inhibition [33], it is plausible that CB1-mediated inhibition of the release of a tonically active anorexigenic mediator, such as CART, is responsible for the appetite-stimulating effect of cannabinoids. Indeed, endogenous CART may tonically inhibit food intake, as indicated by the ability of centrally administered CART antiserum to increase feeding in normal rats [20, 34]. The reduced hypothalamic levels of CART in leptin-deficient obese animals and the ability of leptin to increase CART levels [20, 35] is also compatible with CART being an anorexigenic mediator. This is further supported by the fasting-induced decrease in CART and its normalization on re-feeding in brain regions involved in appetite control, including the hypothalamus, nucleus accumbens and amygdala [36, 37]. In a recent study, adult male CART knockout mice of the same strain used in the present study were found to have increased body weight compared to wild-type controls [28]. Although CART knockout mice on a different background strain did not display the expected obese phenotype, they were found to be more prone to high-fat diet-induced obesity than their wild-type littermates [38]. Earlier studies have implicated the hypothalamus [6, 7, 39] as well as the limbic forebrain [15, 39] as sites of action of the appetitive effect of endocannabinoids, although activation of hypothalamic circuitries may also occur indirectly via activation of CB1 receptors on peripheral sensory nerve terminals [16]. In contrast to CART, endocannabinoids are negatively regulated by leptin and their brain levels are elevated in regions implicated in appetite control in leptin-deficient obese animals [7] or by fasting [15]. CART and CB1 receptor mRNA have been found to be co-expressed in the hypothalamic PVN [40], and here we documented co-localization of CART-immunoreactive fibers and CB1 receptors in the amygdala. Bidirectional connections between the two systems can be demonstrated immunohistochemically, although the highest levels of CART and CB1 immunoreactivity are in different amygdaloid nuclei (fig. 8). CB1 receptor-immunore-
Neuroendocrinology 2005;81:273–282
279
Fig. 8. Bilateral neuronal connections be-
tween the CB1-immunopositive basolateral and CART-immunoreactive basomedial amygdaloid nuclei. A Amygdala, coronal section, CART/CB1 double immunostaining. Boxed areas in A are shown enlarged in B and C. B Fine CB1 terminals in the basomedial amygdaloid nucleus are stained brown. C Fine CART-immunopositive varicose fibers (stained bluish-gray) in the CB1positive basolateral amygdaloid nucleus. bma = Basomedial amygdaloid nucleus; co = cortical amygdaloid nucleus; ep = endopeduncular nucleus; la = basolateral amygdaloid nucleus; lh = lateral hypothalamic area; ot = optic tract; pi = piriform cortex. Scale bars: 400 m (A), 25 m (B) and 50 m (C). The areas boxed in A had been rotated 90° in B and C, corresponding corners are marked by X.
active networks of fibers are particularly abundant in the basolateral amygdaloid region (fig. 8C) where they have been shown to be located on cholecystokinin-containing GABAergic interneurons [41, 42], whereas CART immunoreactivity is more abundant in the basomedial amygdaloid nucleus (fig. 8B). This suggests that rather than directly inhibiting CART release via CB1 located on CART-containing terminals, endocannabinoids may inhibit CART release indirectly, via an action on intermediary neurons projecting from the basolateral to the basomedial amygdaloid nucleus. Additional evidence for a
280
Neuroendocrinology 2005;81:273–282
functional interaction between CART and CB1 is provided by the observation that treatment of mice with the potent cannabinoid agonist HU-210 resulted in an increase in CART immunoreactivity in the arcuate and accumbens nuclei, reflecting accumulation of the CART peptide in axon terminals, likely as a result of its reduced release. The possibility that endogenous cannabinoids may have a similar effect was explored by using genetically altered mice deficient in FAAH. FAAH has been shown to be responsible for the in vivo enzymatic degradation
Osei-Hyiaman et al.
of the endocannabinoid anandamide, and FAAH-deficient mice have 10- to 15-fold elevated brain levels of anandamide as compared to their wild-type littermates [29]. Reduced FAAH activity due to genetic removal or pharmacological blockade of the enzyme has been shown to result in a CB1 receptor-mediated reduction in anxiety [43] and pain sensation [29, 44]. Anandamide rather than 2-AG is the likely mediator of such effects, as indicated by the selective increase in anandamide, but not 2-AG, levels in the brains of FAAH–/– vs. FAAH+/+ mice (fig. 2). Therefore, it is likely that the observed reduction of immunoreactive CART in the arcuate, dorsomedial and periventricular nuclei of the hypothalamus, the bed nucleus of the stria terminalis and the amygdala was due to a lifelong increase in endocannabinoid tone maintained by elevated levels of anandamide, which would lead to inhibition of CART synthesis as a secondary response to decreased release. Although all of the above structures have been implicated in the control of food intake [3, 45], it is possible that the observed functional interaction between CB1 and CART peptide may be also relevant for other behavioral processes regulated by both endocannabinoids and CART. In further confirmation of such a functional interaction, disruption of endocannabinoid ‘tone’ by treatment of FAAH–/– mice with the CB1 receptor antagonist rimonabant resulted in a restoration of CART content close to levels seen in wild-type mice. The absence of such changes in the median eminence and the PVN suggests
either that CB1 receptors are not co-localized with CART terminals in those regions or, in view of the documented co-localization of CB1 and CART mRNAs in the PVN [40], that the cellular distribution of FAAH in this latter region is such that it plays a lesser role in terminating the action of endogenous anandamide. Interestingly, the effect of intra-PVN microinjection of CART peptide on appetite is not quite clear, as both decreases [46] and increases in food intake have been reported [47]. Other anorexigenic mediators, such as CRH and MSH, may also be involved in the effects of endocannabinoids [40]. In this regard, it is noteworthy that co-localization of CB1 and CRH type-1 receptors have been recently documented in hypothalamic and limbic areas involved in appetite control, using double-label in situ hybridization histochemistry [48]. As for -MSH, results of a recent study indicate that CB1 receptors are located downstream from MC-4 receptors involved in the control of food intake [49], which is opposite in direction to the interaction of CART and CB1 indicated by the present results. Further experiments are in progress to clarify the neural circuitry involved in the orexigenic effects of endocannabinoids. Acknowledgment This work was supported in part by NIH grants DA00286 and DA11322 to K.M., RR00165, DA00418 and DA10732 to M.J.K. and DA017259 and DA015197 to B.F.C.
References 1 Kalra SP, Dube MG, Pu S, Xu B, Horvath TL, Kalra PS: Interacting appetite-regulating pathways in the hypothalamic regulation of body weight. Endocr Rev 1999;20:68–100. 2 Saper CB, Thou TC, Elmquist JK: The need to feed: homeostatic and hedonic control of eating. Neuron 2002;36:199–211. 3 Palkovits M: Hypothalamic regulation of the food intake. Clin Neurosci 2003;56:288–302. 4 Williams CM, Kirkham TC: Anandamide induces overeating: mediation by central cannabinoids (CB1) receptors. Psychopharmacology 1999;143:315–317. 5 Hao S, Avraham Y, Mechoulam R, Berry EM: Low-dose anandamide affects food intake, cognitive function, neurotransmitter and corticosterone levels in diet-restricted mice. Eur J Pharmacol 2000;392:147–156. 6 Jamshidi N, Taylor DA: Anandamide administration into the ventromedial hypothalamus stimulates appetite in rats. Br J Pharmcol 2001;134:1151–1154.
CART Is Involved in the Appetitive Effect of Endocannabinoids
7 Di Marzo V, Goparaju SK, Wang L, Liu J, Batkai S, Jarai Z, Fezza F, Miura GI, Palmiter RD, Sugiura T, Kunos G: Leptin-regulated endocannabinoids are involved in maintaining food intake. Nature 2001;410:822–825. 8 Colombo G, Agabio R, Diaz G, Lobina C, Reali R, Gessa GL: Appetite suppression and weight loss after the cannabinoids antagonist SR141716. Life Sci 1998;63:113–117. 9 Freedland CS, Poston JS, Porrino LJ: Effects of SR141716A, a central cannabinoids receptor antagonist, on food-maintained responding. Pharmacol Biochem Behav 2000;67:265– 270. 10 Kirkham TC, Williams CM: Synergistic effects of opioid and cannabinoid antagonists on food intake. Psychopharmacology 2001; 153: 267– 270.
11 Rowland NE, Mukherjee M, Robertson K: Effects of the cannabinoids receptor antagonist SR141716, alone and in combination with dexfenfluramine or naloxone, on food intake in rats. Psychopharmacology 2001; 159: 111– 116. 12 Ravinet Trillou C, Arnone M, Delgorge C, Gonalons N, Keane P, Maffrand JP, Soubrie P: Anti-obesity effect of SR141716, a CB1 receptor antagonist, in diet-induced obese mice. Am J Physiol 2003;284:R345–R353. 13 Hildebrandt AL, Kelly-Sullivan DM, Black SC: Antiobesity effects of chronic cannabinoids CB1 receptor antagonist treatment in diet-induced obese mice. Eur J Pharmacol 2003;462:125–132. 14 Vickers SP, Webster LJ, Wyatt A, Dourish CT, Kennett GA: Preferential effects of the cannabinoid CB1 receptor antagonist, SR141716, on food intake and body weight gain of obese (fa/ fa) compared to lean Zucker rats. Psychopharmacology 2003;167:103–111.
Neuroendocrinology 2005;81:273–282
281
15 Kirkham TC, Williams CM, Fezza F, Di Marzo V: Endocannabinoid levels in rat limbic forebrain and hypothalamus in relation to fasting, feeding and satiation: stimulation of eating by 2-arachidonoyl glycerol. Br J Pharmacol 2002;136:550–557. 16 Gómez R, Navarro M, Ferrer B, Trigo JM, Bilbao A, Arco ID, Cippitelli A, Nava F, Piomelli D, de Fonseca FR: A peripheral mechanism for CB1 cannabinoid receptor-dependent modulation of feeding. J Neurosci 2002; 22: 9612–9617. 17 Miller CC, Murray TF, Freeman KG, Edwards GL: Cannabinoid agonist, CP 55,940, facilitates intake of palatable foods when injected into the hindbrain. Physiol Behav 2004; 80: 611–616. 18 Freund TF, Katona I, Piomelli D: Role of endogenous cannabinoids in synaptic signaling. Physiol Rev 2003;83:1017–1066. 19 Fan W, Boston BA, Kesterson RA, Hruby VJ, Cone RD: Role of melanocortinergic neurons in feeding and the agouti obesity syndrome. Nature 1997;385:165–168. 20 Kristensen P, Judge ME, Thim L, Ribel U, Christjansen KN, Wulff BS, Clausen JT, Jensen PB, Madsen OD, Vrang N, Larsen PJ, Hastrup S: Hypothalamic CART is a new anorectic peptide regulated by leptin. Nature 1998;393: 72–76. 21 Bannon AW, Seda J, Carmouche M, Francis JM, Jarosinski MA, Douglass J: Multiple behavioral effects of cocaine- and amphetamineregulated transcript (CART) peptides in mice: CART 42-89 and CART 49-89 differ in potency and activity. J Pharmacol Exp Ther 2001;299:1021–1026. 22 Hunter RG, Philpot K, Vicentic A, Dominguez G, Hubert GW, Kuhar MJ: CART in feeding and obesity. Trends Endocrinol Metab 2004; 15:454–459. 23 Britton DR, Koob GF, Rivier J, Vale W: Intraventricular corticotrophin-releasing factor enhances behavioral effects of novelty. Life Sci 1982;31:363–367. 24 Krahn DD, Gosnell BA, Grace M, Levine AS: CRF antagonist partially reverses CRF- and stress-induced effects on feeding. Brain Res Bull 1986;17:285–289. 25 Kunos G, Osei-Hyiaman D, Palkovits M, Kuhar MJ, Mackie K, Bannon AW: Endocannabinoids and appetite: possible interactions with other neurotransmitter systems. International Cannabinoid Research Society – Symposium on the Cannabinoids, Burlington, Vt 2002. 26 Hájos N, Katona I, Naiem SS, Mackie K, Ledent C, Mody I, Freund TF: Cannabinoids inhibit hippocampal GABAergic transmission and network oscillations. Eur J Neurosci 2000; 12:3239–3249.
282
27 Koylu EO, Couceyro PR, Lambert PD, Ling NC, DeSouza EB, Kuhar MJ: Immunohistochemical localization of novel CART peptides in rat hypothalamus, pituitary and adrenal gland. J Neuroendocrinol 1997;9:823–833. 28 Wierup N, Richards WG, Bannon AW, Kuhar MJ, Ahren B, Sundler F: CART knockout mice have impaired insulin secretion and glucose intolerance, altered -cell morphology and increased body weight. Regul Pept 2005; 129: 203–211. 29 Cravatt BF, Demarest K, Patricelli MP, Bracey MH, Giang DK, Martin BR, Lichtman AH: Supersensitivity to anandamide and enhanced endogenous cannabinoid signaling in mice lacking fatty acid amide hydrolase. Proc Natl Acad Sci USA 2001;98:9371–9376. 30 Wang L, Liu J, Harvey-White J, Zimmer A, Kunos G: Endocannabinoid signaling via cannabinoid receptor 1 is involved in ethanol preference and its age-dependent decline in mice. Proc Natl Acad Sci USA 2003; 100: 1393– 1398. 31 Cravatt BF, Giang DK, Mayfield SP, Boger DL, Lerner RA, Gilula NB: Molecular characterization of an enzyme that degrades neuromodulatory fatty-acid amides. Nature 1996; 384:83–87. 32 Piomelli D: The molecular logic of endocannabinoid signalling. Nat Rev Neurosci 2003;4: 873–884. 33 Wilson RI, Kunos G, Nicoll RA: Presynaptic specificity of endocannabinoid signaling in the hippocampus. Neuron 2001;31:453–462. 34 Lambert PD, Couceyro PR, McGirr KM, Dall Vechia SE, Smith Y, Kuhar MJ: CART peptides in the central control of feeding and interactions with neuropeptide Y. Synapse 1998; 29:293–298. 35 Elias CF, Lee C, Kelly J, Aschkenasi C, Ahima RS, Couceyro PR, Kuhar MJ, Saper CB Elmquist JK: Leptin activates hypothalamic CART neurons projecting to the spinal cord. Neuron 1998;21:1375–1385. 36 McAlister ED, Van Vogt DA: Effect of leptin administration versus re-feeding on hypothalamic neuropeptide gene expression in fasted male rats. Can J Physiol Pharmacol 2004; 82: 1128–1134. 37 Vicentic A, Lakatos A, Hunter R, Philpot K, Dominguez G, Kuhar MJ: CART peptide diurnal rhythm in brain and effect of fasting. Brain Res 2005;1032:111–115. 38 Asnicar MA, Smith DP, Yang DD, Heiman ML, Fox N, Chen YF, Hsiung HM, Köster A: Absence of cocaine- and amphetamine-regulated transcript results in obesity in mice fed a high caloric diet. Endocrinology 2001; 142: 4394–4400. 39 Freedland CS, Whitlow CT, Smith HR, Porrino LJ: Functional consequences of the acute administration of the cannabinoid receptor antagonist, SR141716A, in cannabinoid-naïve and -tolerant animals: a quantitative 2[14C]deoxyglucose study. Brain Res 2003;962: 169–179.
Neuroendocrinology 2005;81:273–282
40 Cota D, Giovanni M, Tschöp M, Grübler Y, Flachskamm C, Schubert M, Auer D, Yossouridis A, Thöne-Reineke C, Ortmann S, Tomassoni F, Cervino C, Nisoli E, Linthorst ACE, Pasquali R, Lutz B, Stalla GK, Pagotto U: The endogenous cannabinoids system affects energy balance via central orexigenic drive and peripheral lipogenesis. J Clin Invest 2003;112: 423–431. 41 Katona I, Rancz EA, Acsady L, Ledent C, Mackie K, Hajos N, Freund TF: Distribution of CB1 cannabinoid receptors in the amygdala and their role in the control of GABAergic transmission. J Neurosci 2001; 21: 9506– 9518. 42 McDonald AJ, Mascagni F: Localization of the CB1-type cannabinoid receptor in the rat basolateral amygdala: high concentrations in a subpopulation of cholecystokinin-containing interneurons. Neuroscience 2001; 107: 641– 652. 43 Kathuria S, Gaetani S, Fegley D, Valino F, Duranti A, Tontini A, Mor M, Tarzia G, La Rana G, Calignano A, Giustino A, Tattoli M, Palmery M, Cuomo V, Piomelli D: Modulation of anxiety through blockade of anandamide hydrolysis. Nat Med 2003;9:76–81. 44 Lichtman AH, Shelton CC, Advani T, Cravatt BF: Mice lacking fatty acid amide hydrolase exhibit a cannabinoid receptor-mediated phenotypic hypoalgesia. Pain 2004;109:319–327. 45 Ciccocioppo R, Fedeli A, Economidou D, Policani F, Weiss F, Massi M: The bed nucleus is a neuroanatomical substrate for the anorectic effect of corticotropin-releasing factor and for its reversal by nociceptin/orphanin FQ. J Neurosci 2003;23:9445–9451. 46 Wang C, Billington CJ, Levine AS, Kotz CM: Effect of CART in the hypothalamic paraventricular nucleus on feeding and uncoupling protein gene expression. Neuroreport 2000;11: 3251–3255. 47 Abbott CR, Rossi M, Wren AM, Murphy KG, Kennedy AR, Stanley SA, Zollner AN, Morgan DGA, Morgan I, Ghatei MA, Small CJ, Bloom SR: Evidence of an orexigenic role for cocaineand amphetamine-regulated transcript after administration into discrete hypothalamic nuclei. Endocrinology 2001;142:3457–3463. 48 Hermann H, Lutz B: Coexpression of cannabinoid receptor type 1 with the corticotrophinreleasing hormone receptor type 1 in distinct regions of the adult mouse forebrain. Neurosci Lett 2005;375:13–18. 49 Verty ANA, McFarlane JR, McGregor IS, Mallet PE: Evidence for an interaction between CB1 cannabinoid and melanocortin MCR-4 receptors in regulating food intake. Endocrinology 2004;145:3224–3231.
Osei-Hyiaman et al.
