PKA activator inhibits both schedule feeding and NPY-induced feeding in rats

PKA activator inhibits both schedule feeding and NPY-induced feeding in rats

Peptides 24 (2003) 245–254 Hypothalamic administration of cAMP agonist/PKA activator inhibits both schedule feeding and NPY-induced feeding in rats S...

230KB Sizes 0 Downloads 87 Views

Peptides 24 (2003) 245–254

Hypothalamic administration of cAMP agonist/PKA activator inhibits both schedule feeding and NPY-induced feeding in rats Sulaiman Sheriff a,∗ , William T. Chance a,b,c , Sabahat Iqbal a , Tilat A. Rizvi d , Chun Xiao a , John W. Kasckow b,c,e , A. Balasubramaniam a,c a

Department of Surgery, College of Medicine, University of Cincinnati, 231 Albert Sabin Way, Cincinnati, OH 45267, USA b Veterans Affairs Medical Center, Cincinnati, OH 45221, USA c University of Cincinnati Neurosciences Program, University of Cincinnati, Cincinnati, OH 45267, USA d Department of Anatomy, Cell Biology and Neurobiology, University of Cincinnati, Cincinnati, OH 45267, USA e Department of Psychiatry, University of Cincinnati, Cincinnati, OH 45267, USA Received 26 June 2002; accepted 30 September 2002

Abstract Following central administration, neuropeptides that decrease the level of cAMP induce feeding. Conversely, cAMP activating neuropeptides tend to elicit satiety. When the inhibitory effect of neuropeptide Y (NPY) on the hypothalamic cAMP production was blocked by pertussis toxin, the potent orexigenic effect of NPY was lost. These findings suggest that there may be a link between hypothalamic cAMP and the central regulation of food intake. In this report, we show that the injection of the membrane-permeable cAMP agonist, adenosine-3 ,5 -cyclic monophosphorothioate Sp-isomer (Sp-cAMP), into perifornical hypothalamus (PFH) significantly inhibited schedule-induced and NPY-induced food intake for up to 4 h. This inhibitory effect was normalized within 24 h. A taste aversion could not be conditioned to Sp-cAMP treatment, suggesting that the anorectic response was not due to malaise. Sp-cAMP administration significantly increased the active protein kinase A (PKA) activity in dorsomedial (DMH) and ventromedial (VMH), but not in lateral (LH) hypothalamus. Consistently, food deprivation lowered, while refeeding normalized endogenous cAMP content in DMH and VMH, but not in LH areas. No significant effect of adenosine-3 ,5 -cyclic monophosphorothioate Rp-isomer (Rp-cAMP, cAMP antagonist) was observed on hypothalamic PKA activity, schedule-induced, or NPY-induced food intake. These findings suggest that the increase in cAMP level and PKA activity in DMH and VMH areas may trigger a satiety signal. © 2003 Elsevier Science Inc. All rights reserved. Keywords: Satiety; cAMP; PKA; Hunger; Hypothalamus

1. Introduction Abbreviations: ad lib, ad libitum; AGRP, agouti-related peptide; CART, cocaine- and amphetamine-regulated transcript; CREB, cAMP response element binding protein; CREM, cAMP response element modulator; CRH, corticotropin-releasing hormone; CSF, cerebrospinal fluid; DMH, dorsomedial hypothalamus; GLP-1, glucagon-like peptide 1; ICER, inducible cAMP early repressor; i.c.v., intracerebroventricular; LC, locus coeruleus; LH, lateral hypothalamus; iht, intrahypothalamic; MCH, melanin-concentrating hormone; ␣-MSH, melanocyte-stimulating hormone; NPY, neuropeptide Y; PACAP, pituitary adenylate cyclase activating polypeptide; PKA, protein kinase A; PFH, perifornical hypothalamus; PMSF, phenylmethylsulfonyl fluoride; PVN, paraventricular nucleus; Rp-cAMP, adenosine-3 ,5 -cyclic monophosphorothioate Rp-isomer; Sp-cAMP, adenosine-3 ,5 -cyclic monophosphorothioate Sp-isomer; TCA, trichloroacetic acid; VMH, ventromedial hypothalamus ∗ Corresponding author. Tel.: +1-513-558-2720; fax: +1-513-558-0750. E-mail address: [email protected] (S. Sheriff).

Multiple chemical signals, including neurotransmitters and neuropeptides, act through specific receptors in several regions of the brain to regulate food intake. These signals can be grouped into hunger-inducing orexigenic stimuli and satiety-producing anorexigenic stimuli. Among the neuropeptides that activate hunger signals are neuropeptide Y (NPY), agouti-related peptide (AGRP), melaninconcentrating hormone (MCH), orexin A and B, and galanin. Conversely, several neuropeptides, such as corticotropinreleasing hormone (CRH), pituitary adenylate cyclase activating polypeptide (PACAP), amylin, glucagon-like peptide 1 (GLP-1), melanocyte-stimulating hormone (␣-MSH), and cocaine- and amphetamine-regulated transcript (CART), are known to induce satiety [5,22]. Most of these neuropeptides

0196-9781/03/$ – see front matter © 2003 Elsevier Science Inc. All rights reserved. doi:10.1016/S0196-9781(03)00037-8


S. Sheriff et al. / Peptides 24 (2003) 245–254

act through rhodopsin-like G protein-coupled receptors that are functionally coupled to adenylate cyclase by a stimulatory (Gs) or inhibitory (Gi) G protein. Binding of anorectic neuropeptides to their respective receptor subtypes is known to induce cAMP production from cellular ATP via Gs protein and adenylate cyclase activation. In contrast, binding of orexigenic neuropeptides appears to activate a Gi protein complex, leading to the inhibition of cAMP synthesis. Thus, coupling of a neuropeptide receptor to either a Gs or Gi protein determines whether intracellular concentration of cAMP will increase or decrease, respectively. The negative correlation between the cAMP synthesis and food intake suggests that cAMP may be an important biochemical signal in the regulation of feeding behavior. Valases et al. [28] have reported in rats that hypothalamic cAMP level was high in the morning and decreased at night. Similar changes in cAMP content were observed in ventromedial hypothalamus (VMH) areas [17]. Since the predominant feeding in rats occurs at night, decreased hypothalamic cAMP may act as central signal for appetite, while increased hypothalamic cAMP may exert satiety. Furthermore, intrahypothalamic (iht) administration of NPY inhibited hypothalamic adenylate cyclase activity and stimulated food intake in rats. Uncoupling NPY receptor activation from the Gi protein by pertussis toxin treatment attenuated NPY-induced food intake [4]. A recent study by Gillard et al. [10] suggests that an increase in cellular cAMP in the perifornical hypothalamus (PFH) may stimulate eating behavior in satiated rats. These findings prompted us to investigate the direct effect of altering hypothalamic cAMP on feeding behavior in rats. Despite the correlations of cAMP changes with feeding behavior, a direct link between the alterations in hypothalamic cAMP levels and feeding is lacking. In the present study, we examined possible involvement of a cAMP agonist and antagonist on ad libitum (ad lib) feeding, schedule-induced feeding, and NPY-induced food intake in Sprague–Dawley rats bearing unilateral PFH cannulae. The two cAMP analogs used in this study were adenosine-3 ,5 -cyclic monophosphorothioate Sp-isomer (Sp-cAMP) and adenosine-3 ,5 -cyclic monophosphorothioate Rp-isomer (Rp-cAMP). Sp-cAMP is a specific and more potent protein kinase A (PKA) activator than endogenous cAMP due to its higher affinity to the regulatory subunit of PKA. Although Rp-cAMP is a stereoisomer of Sp-cAMP, it acts as a cAMP antagonist by occupying the cAMP-binding site in PKA, preventing it from becoming activated [8]. The two cAMP analogs used in this study are also membrane permeable, more lipophilic, and resistant to phosphodiesterase than cAMP. We show in this study that hypothalamic administration of Sp-cAMP inhibited schedule-induced feeding and NPY-induced food intake that paralleled the increase in active PKA activity in hypothalamic areas. Furthermore, food deprivation decreased cAMP levels in dorsomedial hypothalamus (DMH) and VMH areas, while refeeding normalized it. These results are consis-

tent with our hypothesis that hypothalamic cAMP may play a significant role in the central regulation of food intake.

2. Materials and methods 2.1. Materials Adult male Sprague–Dawley rats were purchased from Zivic Miller (Zelienopole, PA). Other chemicals and reagents used were as follows: 32 P-␥ATP, 125 I-cAMP kit (NEN Company, Boston, MA); PKA assay system (Life Technologies Inc., Rockville, MD); Sp-8-bromo-cAMP and Rp-8-bromo-cAMP (Biolog Life Science Institute, c/o Ruth Langhorst International Marketing, La Jolla, CA); LiCl, saccharin, (Sigma, St. Louis, MO). Porcine NPY was obtained from American Peptide Company Inc. (Sunnyvale, CA). Rat diet Teklad LM-485, 7012 was purchased from Harlan Teklad (Madison, WI). Insulin, corticosterone, aldosterone radioimmunoassay (RIA) kits (ICN Pharmaceuticals, Inc., Orangeburg, NY) and [Arg8 ]-vasopressin RIA kit (Peninsula Laboratories, Inc., San Carlos, CA) were obtained and used as recommended by the suppliers. All chemicals and reagents were the highest grade available and obtained commercially. 2.2. Animals Male Sprague–Dawley rats (350–480 g) were housed and maintained individually in a temperature- and humiditycontrolled environment under a 12-h light/dark cycle and acclimated to laboratory conditions for 1 week. Rats had ad lib access to water and food (Teklad rat diet 7012 containing 19% crude protein, 5% crude fat, and 5% crude fiber), except during the schedule feeding or the food-deprivation period. The rats were anesthetized with ketamine/xylazine (80/15 mg/kg) and 24-ga stainless steel cannulae (Plastics One, Roanoke, VA) were implanted unilaterally into PFH as described in our previous publications [4]. Experiments were conducted following a 2-week recovery period. At the conclusion of the studies, the rats were euthanized by decapitation, brains were removed, and 2 mm thick coronal sections of mid hypothalamus were taken using a Zivic Miller brain matrix. This isolated section, extends rostrally to just behind the optic chiasma and caudally just anterior to the mammilary bodies, occupying the central region of the hypothalamus. The tissues were transferred to cold cerebrospinal fluid (CSF), and sectioned into DMH, VMH, and lateral hypothalamus (LH) regions with the aid of a dissecting microscope and prominent landmarks (fornix, third ventricle, and optic tract). In these sections, the VMH contains the arcuate nucleus of the hypothalamus, while the DMH section includes the paraventricular nucleus (PVN). The LH includes the hypothalamic area lateral to the fornix and extending to the optic tract. These hypothalamic areas were frozen individually in liquid nitrogen and used for cAMP or PKA assays.

S. Sheriff et al. / Peptides 24 (2003) 245–254

2.3. Food deprivation and refeeding Food-deprived rats were deprived of food but not of water for 48 h, while the refed group was food deprived for 48 h but allowed to eat for 8 h before sacrifice. 2.4. Effects of cAMP analogs on ad lib feeding To test the effects of altering the hypothalamic cAMP levels, perifornical injections of 1 ␮l artificial CSF or an equal volume of Sp-cAMP (40 nmol) or Rp-cAMP (40 nmol) in artificial CSF were administered. As a comparison, NPY (0.24 nmol) was also tested. Five to 11 rats were included in each group. Access to rat chow was permitted immediately (NPY or CSF) or 15 min (Sp-cAMP or Rp-cAMP) following the injections. Food intake was monitored at 1, 2, 4, and 24 h after the treatment. 2.5. Effects of cAMP analogs on schedule-induced feeding In another set of experiments, rats with PFH cannula were placed on a feeding schedule under which access to rat chow was permitted only from 09:00 until 17:00 h. Water was always available to these rats. A 2-week time period allowed the rats to adapt to this feeding schedule. These rats were injected by treating with 1 ␮l CSF or 40 nmol of either Sp-cAMP or Rp-cAMP in 1 ␮l artificial CSF. Fifteen minutes after the iht injections the rats were allowed access to food. Intake of rat chow was monitored at 1, 2, 4, and 24 h. Five to eight rats were included in each group. 2.6. Effects of cAMP analogs on NPY-induced feeding The effects of cAMP anologs on NPY-induced feeding were assessed in another experiment with additional rats. Two weeks after recovering from cannula implantation into the PFH, different groups of rats were treated with 1 ␮l artificial CSF, 40 nmol Sp-cAMP, or 40 nmol Rp-cAMP contained in 1 ␮l artificial CSF. Fifteen minutes later, either 0.24 nmol NPY in 1 ␮l artificial CSF or 1 ␮l artificial CSF treatments were given to the rats in order to obtain the following groups: CSF-CSF, CSF-NPY, Sp-cAMP-NPY, and Rp-cAMP-NPY. Eight to 14 rats were included in each group. Access to rat chow was permitted immediately following the second injections. Food intake was monitored at 1, 2, 4, and 24 h after the second treatment. 2.7. Conditioned taste aversion test Following the termination of the schedule feeding experiment, a group of 31 of these rats were placed on ad lib feeding for 1 week prior to assessing conditioned taste aversion. The rats were next placed on a drinking schedule for 1 week, under which access to water was permitted for 1 h each day.


On the conditioning day, 24 of these rats were allowed access to a 0.1% saccharin solution instead of normal drinking water, the control animals were given water without saccharin. Fifteen minutes after the initiation of this paradigm, nine of these rats were treated with 1 ␮l CSF while seven rats received 40 nmol Sp-cAMP in 1 ␮l CSF. The seven rats that consumed water instead of saccharin were also treated with 1 ␮l CSF. The remaining eight saccharin-exposed rats were treated with 0.6 M LiCl (i.p., 5 ml/kg) as a positive control. The next day all rats were offered a choice between 0.1% saccharin and water in two identical water bottles, randomized for position. Intake of water and saccharin was monitored for 15 and 60 min during this two-bottle test and a saccharin preference ratio was calculated by dividing the intake of saccharin by the total fluid intake for each time point. 2.8. RIA and glucose measurement Plasma insulin, corticosterone, aldosterone (ICN Pharmaceuticals, Inc., Orangeburg, NY) and [Arg8 ]-vasopressin (Peninsula Laboratories, Inc., San Carlos, CA) were measured by RIA method. The percentage coefficient of variations (CVs) of RIAs are as follows: (a) insulin, <8.25% (intra-assay), <8.81% (interassay); (b) corticosterone, <10.3% (intra-assay), <7.1% (interassay); (c) aldosterone, <6.3% (intra-assay), <14.2% (interassay); and (d) cAMP, <3.5% (intra-assay), <11.0% (interassay). The sensitivity of the assay for vasopressin is 4 pg/tube. Glucose levels were monitored by the glucose oxidation method. 2.9. cAMP levels Isolated hypothalamic areas were individually homogenized in 100 ␮l of homogenization buffer (50 mM triethanolamine, pH 7.4, supplemented with leupeptin (in ␮g/␮l), aprotinin (in ␮g/␮l), and phenylmethylsulfonyl fluoride (PMSF, 1 mM)). Ten microliters of homogenate was saved for protein estimation, while the remaining 90 ␮l of homogenate was extracted with 6% trichloroacetic acid (TCA). The TCA extract was treated with 5 volumes of diethyl ether (four times) to remove TCA. Next, the cAMP-containing extract was evaporated to dryness under slow stream of air, reconstituted in 200 ␮l of 50 mM acetate buffer (pH 6.2), and assayed for cAMP by RIA [25]. The cAMP values were normalized to milligrams of protein. 2.10. PKA assay Hypothalamic areas were homogenized in 70 ␮l of Tris–HCl buffer, pH 7.5, containing 5 mM EDTA and 1 mM PMSF in ice. The homogenate was centrifuged at 10,000 × g for 3 min. PKA activity was assessed in the supernatant, as suggested by the supplier’s protocol. In brief, 10 ␮l protein extract was incubated with 10 ␮l PKA activator (40 ␮M cAMP in 50 mM Tris–HCl,


S. Sheriff et al. / Peptides 24 (2003) 245–254

pH 7.5) for 15 min at room temperature. Then, 10 ␮l of 25 ␮Ci/ml 32 P-␥ATP-containing substrate (200 ␮M kemptide: Leu-Arg-Arg-Ala-Ser-Leu-Gly, 400 ␮M ATP, 40 mM MgCl2 , BSA (in mg/ml) in Tris–HCl buffer, pH 7.5) and 10 ␮l of 50 mM Tris buffer were added and incubated for 5 min at 30 ◦ C. The final reaction volume was 40 ␮l. The reaction was terminated by transferring 20 ␮l aliquots of reaction mixture onto phosphocellulose discs and washing twice with 10 ml of 1% (v/v) phosphoric acid. The amount of radioactivity incorporated into the substrate was monitored by counting in an LKB-Wallac 1217 Rackbeta scintillation counter (Turku, Finland). Substrate phosphorylation in the presence of cAMP reflected the total amount of PKA activity in each sample. Non-specific reaction was measured by using 6–22 amino acid fragment of protein kinase inhibitor (PKI; 1 ␮M in

50 mM Tris–HCl, pH 7.5) protein in parallel treatments. PKA substrate phosphorylation in the presence of PKI was defined as background and subtracted from each sample. Total PKA activity was expressed as nanomoles of kemptide phosphorylated per milligram of protein. Active PKA levels were monitored by measuring the substrate phosphorylation in the absence of cAMP and expressed as percent to the total PKA activity. 2.11. Statistical analysis Data were analyzed by one-way analysis of variance (ANOVA), followed by the Tukey–Kramer’s multiple comparisons post-test (Instat Program, GraphPad Software Inc., San Diego, CA). Active PKA activity values were directly compared by Student’s t-test.

Fig. 1. Effects of cAMP analogs on ad lib, or NPY-treated or schedule-fed rats. (A) Intake of rat chow by satiated rats following the iht injection of 1 ␮l artificial CSF, Sp-cAMP (40 nmol), or Rp-cAMP (40 nmol). (B) Intake of rat chow by satiated rats following the iht injection of NPY (1 ␮g) preceded 15 min earlier by Sp-cAMP (40 nmol), Rp-cAMP (40 nmol), or artificial CSF (1 ␮l). (C) Intake of rat chow by schedule-fed rats following iht injection of 1 ␮l artificial CSF, Sp-cAMP (40 nmol), or Rp-cAMP (40 nmol). Data shown are the mean ± S.E.M.

S. Sheriff et al. / Peptides 24 (2003) 245–254

3. Results


feeding during the 1-, 2-, 4-, or 24-h measurement periods (Fig. 1B).

3.1. Effects of cAMP analogs on ad lib food intake The effects of iht injections with the PKA activator, Sp-cAMP, or PKA inhibitor, Rp-cAMP, on ad lib food intake in satiated rats are shown in Fig. 1A. There was no significant change in ad lib food intake by Rp-cAMP or Sp-cAMP up to 4 h in comparison to control (CSF-treated) rats. In addition, there was no significant difference in food intake between Sp-cAMP- or Rp-cAMP-treated groups across 1, 2-, and 4-h measurement periods. However, as a parallel, NPY treatment increased food intake significantly (P < 0.01) at each of the time point as compared to all groups. 3.2. Effects of cAMP analogs on NPY-induced food intake To determine the role of cAMP in mediating the orexigenic effect of NPY, Sp-cAMP or Rp-cAMP was administered 15 min before the peptide injection. As illustrated in Fig. 1B, NPY treatment increased food intake significantly compared to the CSF-treated group across the 4-h test period (P < 0.001). Pretreatment with Sp-cAMP significantly inhibited NPY-induced feeding by 74 ± 9% at 1-h time period (P < 0.01). Similar inhibitory effects were observed at 2 h (69 ± 7%, P < 0.01) and 4 h (58 ± 9%, P < 0.01). This inhibitory effect of Sp-cAMP on NPY-induced feeding was lost by 24 h (not shown), suggesting the reversible nature of this antagonism. In contrast, Rp-cAMP administration had no significant effect on NPY-induced

3.3. Effects of cAMP analogs on schedule-induced food intake The effect of Sp-cAMP and Rp-cAMP on schedule-induced feeding was examined in another set of rats. As shown in Fig. 1C, hypothalamic administration of Sp-cAMP reduced food intake significantly for 4 h in schedule-fed rats. Food intake was inhibited significantly by ∼50% for the first hour (P < 0.01) in comparison to CSF-treated group. This inhibition decreased to 39 and 34% by 2 and 4 h, respectively (P < 0.05). Twenty-four hours after injection, Sp-cAMP no longer reduced feeding significantly (not shown). The Rp-cAMP treatment had no significant effect on food intake during any of the measurement periods (Fig. 1C). 3.4. Effect of Sp-cAMP on conditioned taste aversion To determine whether Sp-cAMP reduced food intake due to causing malaise, its ability to establish a conditioned taste aversion was investigated. There was no significant difference in the mean saccharin preference ratios in the Sp-cAMP-treated (0.718 ± 0.15) and CSF-treated (0.70 ± 0.1) groups. In contrast, rats that had LiCl/paired with saccharin exhibited a significantly lower saccharin preference (0.08 ± 0.02, P < 0.01) in comparison to other groups. These findings demonstrate that the LiCl treatment induced a conditioned taste aversion, while Sp-cAMP was devoid of such effect (Fig. 2).

Fig. 2. Effect of Sp-cAMP treatment on saccharin preference of rats 24 h after pairing of 0.1% saccharin with the iht injection of 1 ␮l artificial CSF, Sp-cAMP (40 nmol), or i.p. injection of LiCl (5 ml/kg). Intake of saccharin by satiated rats 15 min after the iht injection of test compounds was monitored for 30 min. A lower saccharin preference to LiCl (not to Sp-cAMP) indicates that greater taste aversion and more malaise became classically conditioned to treatment with the compound. Data shown are the mean ± S.E.M.


S. Sheriff et al. / Peptides 24 (2003) 245–254

Fig. 3. Schematic diagram illustrating the areas (VMH, DMH, and LH) of rat hypothalamus taken in the experiments measuring PKA or cAMP levels. Each coronal section was 2 mm thick, and extended approximately 1.0 mm anterior and posterior to this representative section. Modified from Pellegrino and Cushman [19]. Scale bar, 200 ␮m.

3.5. Effects of cAMP analogs on hypothalamic PKA activity PKA activity was monitored in DMH, VMH, and LH areas from three groups of rats 30 min after treatment with CSF (1 ␮l), Sp-cAMP (40 nmol/␮l), or Rp-cAMP (40 nmol/␮l). Activity was measured in the presence or

absence of exogenous cAMP in order to determine the total and activated PKA, respectively. The hypothalamic areas isolated for PKA assay are indicated in Fig. 3. Total PKA activity in DMH (753 ± 492 pmol/mg protein) appeared more elevated than VMH (238 ± 63 pmol/mg protein) or LH (362 ± 134 pmol/mg protein). No significant differences were observed in total PKA activity in

Fig. 4. Sp-cAMP-induced PKA activity in the distinct areas of hypothalamus. Active PKA activity was monitored in three areas of hypothalamus following the iht injection of 1 ␮l artificial CSF, Sp-cAMP (40 nmol), or Rp-cAMP (40 nmol) in schedule-fed rats. Hypothalamic areas were isolated 30 min after the iht injection of test compounds. Data shown are the mean ± S.E.M.

S. Sheriff et al. / Peptides 24 (2003) 245–254


Fig. 5. Effect of nutritional status on cAMP level in rat hypothalamic areas. Hypothalamic areas were extracted from three different groups: ad lib, rats had free access to food and water; food deprived, deprived of food but not of water for 48 h; and refed/food deprived as described earlier but allowed to eat for 8 h prior to sacrifice. All animals from the three different groups were killed at the same time. Data shown are the mean ± S.E.M.

Sp-cAMP-treated (VMH, 472±132 pmol/mg protein; DMH, 403 ± 140 pmol/mg protein; LH, 147 ± 24 pmol/mg protein) or Rp-cAMP-treated (VMH, 716 ± 192 pmol/mg protein; DMH, 264 ± 130 pmol/mg protein; LH, 159 ± 92 pmol/mg protein) groups when compared with CSF-treated rats. However, the Sp-cAMP-treated group exhibited significant increases (P < 0.05) in the percentage of active PKA in VMH (56±11.4%) and DMH (50.3±16.3%) in comparison to CSF-treated (VMH 27.3±5.2%; DMH 12.7±8.9%) controls (Fig. 4). Rp-cAMP-treated group showed no significant changes in VMH (24 ± 7.1%) or DMH (12.2 ± 4.8%) areas when compared with CSF- or Sp-cAMP-treated groups. No significant change in active PKA level in LH was observed in Sp-cAMP-treated (29.4 ± 12.6%) or Rp-cAMP-treated (30.5 ± 18.3%) group in comparison to CSF-treated group (9.8 ± 4.2%) (Fig. 4). 3.6. Effect of food deprivation and refeeding on hypothalamic cAMP level Hypothalamic cAMP content was detected using a standard RIA method from three different areas of the hypothalamus. The CV of this assay kit was <13.5%. In comparison to ad lib rats (11.9 ± 1.3 pmol/mg protein), the cAMP content in DMH area was significantly reduced in the food-deprived group (8.2 ± 1.2 pmol/mg protein, P < 0.01). Refeeding increased the cAMP content in DMH (10.3 ± 0.3 pmol/mg protein); however, this cAMP content increase was still lower than that observed in the ad lib-fed group (P < 0.05). In the VMH, the level of cAMP was comparable to that in the DMH (12.5 ± 2.2 pmol/mg protein) of the ad lib-fed group. A similar, though smaller

reduction, in cAMP levels (10.7 ± 0.5 pmol/mg protein, P < 0.05) was noticed in food-deprived rats in comparison to ad lib controls (P < 0.05). In contrast to the DMH, however, 8 h of refeeding reversed the cAMP level to control levels in the VMH (15.4 ± 4.3 pmol). The LH area had higher cAMP content than the VMH and DMH areas. Neither the food-deprived (18.8 ± 3 pmol/mg protein) nor refed groups (20.2 ± 1.2 pmol/mg protein) exhibited significant differences in cAMP levels when compared with the ad lib-fed (18.4 ± 1.1 pmol/mg protein) rats (Fig. 5).

4. Discussion The findings of the present study support the hypothesis that hypothalamic cAMP is involved in the regulation of food intake. Administration of Sp-cAMP into the PFH inhibited schedule-induced and NPY-induced feeding for 4 h. It has been demonstrated that the PFH is very sensitive to both the hunger-stimulating effects of NPY and satiety-eliciting beta-adrenergic catecholamines [14,27], and thus represents an appropriate area in which to assess these effects. Absence of an inhibitory effect by Sp-cAMP on ad lib food intake may be due to the fact that most feeding by rats occurs during the dark phase of the light/dark cycle. These tests were conducted in the early hours of the light phase of the light/dark cycle, when little ad lib feeding occurs. In addition, hypothalamic cAMP levels have been reported to be higher in the morning than at night [28]. Thus, additional stimulation may not reduce feeding due to a ceiling effect. Therefore, an increase in endogenous cAMP may have already activated cAMP-mediated satiety.


S. Sheriff et al. / Peptides 24 (2003) 245–254

Targeted disruption of the regulatory subunit (RII beta) of PKA strongly suggests a potential role of this enzyme in the peripheral regulation of energy homeostasis [7]. However, the central role of PKA in appetite regulation remains hypothetical. The inhibitory effect of Sp-cAMP on schedule-induced and NPY-induced feeding in the present investigation is suggestive evidence for a role of cAMP/PKA pathways in the central control of hunger and satiety mechanisms. The inability to establish a conditioned taste aversion to Sp-cAMP treatment suggests that the anorectic effect produced by the cAMP agonist is not due to malaise. Unlike cAMP, Sp-cAMP is resistant to phosphodiesterasemediated degradation, resulting in increased biological activity [8]. We demonstrated more than a two- to threefold increase in the catalytic subunit activity (active PKA) by Sp-cAMP in the VMH and the DMH, but not in the LH. Thus, in the present investigation, cAMP-induced anorexia appeared to be mediated by activating a PKA signaling cascade in DMH and VMH, but not in the LH area. However, we are uncertain whether the site-specific activation of PKA by Sp-cAMP in the present investigation is due to diffusion, axonal transport, or the actual amount of endogenous PKA in these areas. The second messenger cAMP has been demonstrated to diffuse more extensively than other second messengers, such as IP3 or Ca2+ , along axons due to its longer half-life and higher diffusion length [12]. Lack of a cAMP-induced effect in total PKA activity supports the hypothesis that the Sp-cAMP had no significant effect on PKA expression, but activated the catalytic subunits by binding to regulatory subunits of PKA and subsequently dissociating them. It remains to be determined whether the activation of PKA in the PVN neurons is a sign of physiological feedback with the purpose of quelling a hunger signal, or whether it acts as an anorectic signal induced by cAMP. The return of normal feeding following the administration of the cAMP agonist the next day indicates that this effect was reversible. Furthermore, we did not observe any significant change by Sp-cAMP on circulating insulin, corticosterone, aldosterone, vasopressin, or glucose (Table 1), suggesting that Sp-cAMP-mediated anorexia is independent of these factors. A trend towards a moderate increase in serum corticosterone and aldosterone following injections of both Sp-cAMP and Rp-cAMP analogs suggests a non-specific action of these compounds.

In contrast to the anorectic effect of Sp-cAMP, the administration of the cAMP antagonist Rp-cAMP failed to exhibit any stimulatory effect on ad lib food intake. Similarly, no significant effect of Rp-cAMP was observed on schedule-induced or NPY-induced feeding. The absence of any significant effect by Rp-cAMP suggests that the induction of a hunger stimulus may require the integration of other events in addition to the inhibition of cAMP accumulation in hypothalamus. These events may include the activation of receptor subtypes that regulate food intake, the mobilization of intracellular Ca2+ , or both. Alternatively, absence of Rp-cAMP action on PKA activity in hypothalamic areas may be due to the lack of specificity or the dose used in this study. Earlier studies have shown that pertussis toxin treatment of the NPY-Y1 receptor subtype-bearing cell line abolished peptide-induced mobilization of intracellular Ca2+ release and the inhibitory effect on cAMP production [1]. Also, our in vivo studies have demonstrated that the injection of NPY into PFH increased the CaM kinase II activity in rat hypothalamus within 5 min of treatment [24]. NPY-Y1 receptors coupled to Gq-phospholipase C signaling pathway may also be involved in the regulation of feeding behavior [18,23]. In our previous investigation, we have observed an increase in cAMP response element binding protein (CREB) phosphorylation in the hypothalamus following PFH injection of NPY or food deprivation [6,24]. Pair-fed and food-deprived mice also showed significant parallel increases in NPY expression and CREB phosphorylation in the PVN and arcuate nucleus [13,26]. Based on these findings, we hypothesize that the NPY-mediated inhibitory effect on cAMP/PKA pathways may enhance the feeding behavior by preventing the synthesis of the negative regulators of CREB, such as inducible cAMP early repressor (ICER) or cAMP response element modulator (CREM) [3,9,15,16]. This hypothesized inhibitory effect of NPY on the above transcriptional antagonists may augment the transcriptional activity of phosphoCREB. Additional factors regulating the hunger stimulus may include the activation of c-fos expression in the PVN, which has been suggested as a possible mediator for NPY-induced feeding in rats [29]. Higher doses of Rp-cAMP have been shown to inhibit cAMP production in cell lines [8] as well as to inhibit tyrosine hydroxylase phosphorylation in the locus coeruleus (LC) of rats when administered intracerebroventricular

Table 1 Measurement of plasma hormones and glucose levels 30 min after the iht administration of artificial CSF (1 ␮l), Sp-cAMP (40 nmol), or Rp-cAMP (40 nmol) CSF (n = 6) Insulin (mU/ml) Corticosterone (ng/ml) Aldosterone (pg/ml) [Arg8 ]-vasopressin (pg/ml) Glucose (mg/100 ml)

21 306 289 7.02 105

± ± ± ± ±

0.6 45 48 0.3 7.4

Sp-cAMP (n = 8) 29 449 479 7.5 117

± ± ± ± ±

2.8 25 110 1.3 2.6

Rp-cAMP (n = 8) 28 452 492 6.1 113

± ± ± ± ±

2.4 26 99 0.3 2.3

The CVs of the RIAs were as follows: (a) insulin, <13.9%, (b) corticosterone, <7.43%, (c) aldosterone, <13.0%, and (d) vasopressin, <16.9%. Data shown are the mean ± S.E.M.

S. Sheriff et al. / Peptides 24 (2003) 245–254

(i.c.v.) [20]. In the present investigation, although higher doses of Rp-cAMP at 80 nmol increased ad lib food intake, higher doses of Rp-cAMP also produced seizures and convulsions within 10–15 min after administration in approximately 20% of the rats tested. At the 40 nmol dose, no significant effect on motor activity was observed in the rats throughout the entire test period. Therefore, we utilized the 40 nmol dose throughout the study. An earlier study using dibutyryl cAMP reported that the injection of this cAMP analog has significantly increased NPY levels in arcuate and medial PVN, but not in other areas of hypothalamus. Although a 25 ␮g of dibutyryl cAMP produced no behavioral effects, a 100 ␮g dose produced seizures and convulsions within 30 min [2]. Using 8-bromo-cAMP, a slightly more potent compound than dibutyryl cAMP, an increase in food intake in satiated rats has been reported [10]. We do not know why our results with Sp-cAMP are discrepant to the previous report. It is not clear whether a different analog of cAMP, i.e. Sp-cAMP, should result in different findings than observed with 8-bromo-cAMP. One possibility accounting for these differences may be the rat diets. In the present study, the rats were maintained on regular protein-rich (19%) rat chow. While Gillard et al. [10] used sweetened milk-mash diet containing 46% rat chow powder, 37% sucrose, and 17% milk to monitor food intake. Neuronal stimuli, such as sight, smell, and taste, have been suggested to activate neurons regulating feeding behavior in the LH and PFH [21]. Furthermore, central injection of NPY has been shown to increase intake of carbohydrate diets [11]. It is not clear, however, whether injection of 8-bromo-cAMP or its metabolite in LH and PFH areas can activate the neurons which regulate smell- or taste-dependent food intake [21], or even hypothalamic NPY expression [2], which could account for the discrepancies between our findings and those of previous report [10]. In summary, the data from the current study support the hypothesis that elevated hypothalamic cAMP may act as a mediator for the central regulation of satiety. As suggested in our previous report [5], compounds that stimulate adenylate cyclase activity inhibit feeding behavior while inhibition of cAMP synthesis tends to facilitate food intake. If this hypothesis is valid, modulations of cAMP activity may represent an important second messenger in the signal transduction cascade that ultimately results in hunger or satiety.

Acknowledgments This work was supported in part by United States Public Health Service Grants DK-53548 (S.S.), GM-47122 (A.B.), and MH-001545 (J.W.K.), and the Department of Veterans Affairs (W.T.C). J.W.K. was supported by NIMH KO1 MH 001545-01. Portions of this work were presented at the 6th International NPY Conference, April 2001, Sydney, Australia.


References [1] Aakerlund L, Gether U, Fuhlendorff J, Schwartz TW, Thastrup O. Y1 receptors for neuropeptide Y are coupled to mobilization of intracellular calcium and inhibition of adenylate cyclase. FEBS Lett 1990;260:73–8. [2] Akabayashi A, Zaia CT, Gabriel SM, Silva I, Cheung WK, Leibowitz SF. Intracerbroventricular injection of dibutyryl cyclic adenosine 3 ,5 -monophosphate increases hypothalamic levels of neuropeptide Y. Brain Res 1994;660:323–8. [3] Borsook D, Smirnova O, Behar O, Lewis S, Kobierski LA. PhosphoCREB and CREM/ICER: positive and negative regulation of proenkephalin gene expression in paraventricular nucleus of the hypothalamus. J Mol Neurosci 1999;12:35–51. [4] Chance WT, Sheriff S, Foley-Nelson T, Fischer JE, Balasubramaniam A. Pertusis toxin inhibits neuropeptide Y-induced feeding in rats. Peptides 1989;10:1283–6. [5] Chance WT, Thompson H, Thomas I, Fischer JE. Anorectic and neurochemical effects of pituitary adenylate cyclase activating polypeptide in rats. Peptides 1995;16:1511–6. [6] Chance WT, Sheriff S, Peng F, Balasubramaniam A. Antagonism of NPY-induced feeding by pretreatment with cyclic AMP response element binding protein antisense oligonucleotide. Neuropeptides 2000;34:167–72. [7] Cummings DE, Brandon EP, Planas JV, Motamed K, Idzerda RL, McKnight GS. Genetically lean mice result from targeted disruption of the RII beta subunit of protein kinase A. Nature 1996;382: 622–6. [8] Dostmann WRG, Taylor SS, Genieser H-G, Jastorff B, Doskeland SO, Ogreid D. Probing the cyclic nucleotide binding sites of cAMP-dependent protein kinases I and II with analogs of adenosine 3 ,5 -cyclic phosphorothioates. J Biol Chem 1990;265:10484–91. [9] Foulkes NS, Borrelli E, Sassone-Corsi P. CREM gene: use of alternative DNA-binding domains generates multiple antagonists of cAMP-induced transcription. Cell 1991;64:739–49. [10] Gillard RG, Khan AM, Grewal RS, Mouradi B, Wolfsohn SD, Stanley GB. The second messenger cAMP elicits eating by an anatomically specific action in the perifornical hypothalamus. J Neurosci 1998;18:2646–52. [11] Jhanwar-Uniyal M, Beck B, Jhanwar YS, Burlet C, Leibowitz SF. Neuropeptide Y projection from arcuate nucleus to parvocellular division of paraventricular nucleus: specific relation to the ingestion of carbohydrate. Brain Res 1993;631:97–106. [12] Kasai H, Paterson O. Spatial dynamics of second messengers: IP3 and cAMP as long-range and associative messengers. Trends Neurosci 1994;17:95–101. [13] Lambert PD, Anderson KD, Sleeman MW, Wong V, Tan J, Hijarunguru A, et al. Ciliary neurotrophic factor activates leptin-like pathways and reduces body fat, without cachexia or rebound weight gain, even in leptin-resistant obesity. Proc Natl Acad Sci USA 2001;98:4652–7. [14] Leibowitz SF, Stanley BG. Neurochemical controls of appetite. In: Ritter RC, Ritter S, Barnes CD, editors. Feeding behavior: neural and humoral controls. Orlando, FL: Academic Press; 1986. p. 191–234. [15] Luckman SM, Cox HJ. Expression of inducible cAMP early repressor (ICER) in hypothalamic magnocellular neurons. Mol Brain Res 1995;34:231–8. [16] Molina CA, Foulkes NS, Lalli E, Sassone-Corsi P. Inducibility and negative autoregulation of CREM: an alternative promotor directs the expression of ICER, an early response repressor. Cell 1993;75:875– 86. [17] Murakami N, Takashi K. Circadian rhythm of adenosine-3 ,5 monophosphate content in suprachiasmatic nucleus (SCN) and ventromedial hypothalamus (VMH) in the rat. Brain Res 1983;276: 297–304. [18] Parker SL, Parker MS, Sweatman T, Crowley WR. Characterization of G protein and phospholipase C-coupled agonist binding to the Y1







S. Sheriff et al. / Peptides 24 (2003) 245–254 neuropeptide Y receptor in rat brain: sensitivity to G protein activators and inhibitors and to inhibitors of phospholiase C. J Pharmacol Exp Ther 1998;286:382–91. Pellegrino LJ, Cushman AJ. A stereotaxic atlas of the rat brain. In: Elliot RM, Lindzey G, MacCorquodale K, editors. The century psychology series. New York: Meredith; 1967. Punch LJ, Self DW, Nestler EJ, Taylor JR. Opposite modulation of opiate withdrawal behaviors on microinfusion of a protein kinase A inhibitor versus activator into the locus coeruleus or periaqueductal gray. J Neurosci 1997;17:8520–7. Rolls ET, Murzi E, Yaxley S, Thorpe SJ, Simpson SJ. Sensoryspecific satiety: food-specific reduction in responsiveness of ventral forebrain neurons after feeding in the monkey. Brain Res 1986; 368:79–86. Schwartz MW, Woods SC, Porte Jr D, Seeley RJ, Baskin DG. Central nervous system control of food intake. Nature 2000;404:661– 71. Selbie LA, Darby K, Schmitz-Peiffer C, Browne CL, Herzog H, Shine J, et al. Synergistic interaction of Y1-neuropeptide Y and aib-adrenergic receptors in the regulation of phospholipase C, protein kinase C, and arachidonic acid production. J Biol Chem 1995;270:11789–96.

[24] Sheriff S, Chance WT, Fischer JE, Balasubramaniam A. Neuropeptide Y treatment and food deprivation increase cyclic AMP response element-binding in rat hypothalamus. Mol Pharmacol 1997;51:597– 604. [25] Sheriff S, Balasubramaniam A. Inhibitory and stimulatory effects of neuropeptide Y (17–36) on rat cardiac adenylate cyclase activity. Structure–function studies. J Biol Chem 1992;267:4685–90. [26] Shimizu-Albergine M, Ippolito DL, Beavo JA. Downregulation of fasting-induced cAMP response element-mediated gene induction by leptin in neuropeptide Y neurons of the arcuate nucleus. J Neurosci 2001;21:1238–46. [27] Stanley BG, Magdalin W, Seirafi A, Thomas W, Leibowitz SF. The perifornical area: the major focus of (a) patchily distributed hypothalamic neuropeptide Y sensitive feeding system(s). Brain Res 1993;604:304–17. [28] Valases C, Wright SJ, Gatravas GN. Diurnal changes in cyclic nucleotide levels in the hypothalamus of the rat. Exp Brain Res 1980;40: 261–4. [29] Yokosuka M, Kalra PS, Kalra SP. Inhibition of neuropeptide Y (NPY)-induced feeding and c-fos response in magnocellular paraventricular nucleus by a NPY receptor antagonist: a site of NPY action. Endocrinology 1999;140:4494–500.