Cerulenin, a natural fatty acid synthase (FAS) inhibitor, and its synthetic analog C75 are hypothesized to alter the metabolism of neurons in the hypothalamus that regulate ingestive behavior to cause a profound decrease of food intake and an increase in metabolic rate, leading to body weight loss. The bulk of data exclusively originates from mammals (rodents); however, such effects are currently lacking in nonmammalian species. We have, therefore, addressed this issue in broiler chickens because this species is selected for high growth rate and high food intake and is prone to obesity. First, we demonstrate that FAS messenger and protein are expressed in the hypothalamus of chickens. FAS immunoreactivity was detected in a number of brain regions, including the nucleus paraventricularis magnocellularis and the nucleus infundibuli hypothalami, the avian equivalent of the mammalian arcuate nucleus, suggesting that FAS may be involved in the regulation of food intake. Second, we show that hypothalamic FAS gene expression was significantly (P < 0.05) decreased by overnight fasting similar to that in liver, indicating that hypothalamic FAS gene is regulated by energy status in chickens. Finally, to investigate the physiological consequences of in vivo inhibition of fatty acid synthesis on food intake, we administered cerulenin by intravenous injections (15 mg/kg) to 2-wk-old broiler chickens. Cerulenin administration significantly reduced food intake by 23 to 34% (P < 0.05 to P < 0.0001) and downregulated FAS and melanocortin receptors 1, 4, and 5 gene expression (P < 0.05). However, the known orexigenic (neuropeptide Y, agouti gene-related peptide, orexin, and orexin receptor) and anorexigenic (pro-opiomelanocortin and corticotropin-releasing hormone) neuropeptide mRNA levels remained unchanged after cerulenin treatment. These results suggest that the catabolic effect of cerulenin in chickens may be mediated through the melanocortin system rather than the other neuropeptides known to be involved in food intake regulation.
energy homeostasis and feeding are regulated by the central nervous system. It has been reported that the naturally occurring fatty acid synthase (FAS) inhibitor cerulenin (39) and its more potent synthetic analog C75 (28) profoundly reduce food intake and body weight and increase metabolic rate when injected into mice (6, 8, 21, 24, 29, 33, 35, 43, 53, 58). The metabolic effects of these compounds were hypothesized to involve central nervous system pathways. Loftus et al. (33) have suggested in their original paper that these compounds might, like leptin, act in part by modulating the hypothalamic neuropeptide gene expression. However, studies investigating this hypothesis revealed conflicting results. Indeed, Shimokawa et al. (43) reported that a single intraperitoneal injection of C75 modulates the expression of hypothalamic orexigenic and anorexigenic neuropeptides in a manner similar to that of feeding and leptin. Cha et al. (6) also showed that long-term treatment with C75 mimics the effect of leptin on hypothalamic neuropeptide gene expression in mice, whereas Kumar et al. (29) reported that repeated injections of C75 either have no effect on or regulate hypothalamic neuropeptide mRNA levels in a manner opposite to that of feeding and leptin. Makimura et al. (35) also reported that repeated injections with cerulenin decreases body weight and increases metabolic rate in mice without modulating hypothalamic neuropeptide gene expression. These discrepancies might be related to the physiological context and models (36, 44). Such studies are presently lacking in chickens (nonmammalian species), which manifest some particularities: 1) broiler chickens were selected for high growth and high food intake and are prone to obesity, 2) the mechanism of leptin action on food intake regulation in chickens is different from that described in mammals (13), 3) chickens are characteristically hyperglycemic, insulin resistant (14, 45), and lacking glucose transporter GLUT4 (42), and 4) as in humans, lipogenesis occurs essentially in the liver of chickens (18, 31, 32); however, in rodents, lipogenesis occurs in both adipose tissue and liver (54). These peculiarities make chicken an interesting model for understanding appetite and satiety at molecular levels, which may help to shed light on mechanisms of body weight and energy homeostasis control. Therefore, the present study aimed 1) to characterize FAS in chicken brain, 2) to investigate the effect of nutritional state and genotype on hypothalamic FAS gene expression, and 3) to assess whether intravenous cerulenin administration affects food intake and hypothalamic neuropeptide gene expression in chickens.
MATERIALS AND METHODS
Animal studies were conducted with research protocols approved by the Ethical Commission for Experimental Use of Animals of the Catholic University of Leuven (Belgium).
Experiment 1: characterization of FAS in chicken hypothalamus.
Day-old broiler chickens (Ross strain) were purchased from a commercial hatchery (Avibel, Halle-Zoersel, Belgium) and housed in a controlled-light (a 14:10-h light-dark photoperiod; lights on at 7:00 AM) environment and allowed ad libitum access to standard diet (Hendrix, Merksem, Belgium) and water. At 3 wk of age, tissues (brain, hypothalamus, liver, and muscle) (n = 3) were dissected, quickly frozen in liquid nitrogen, ground in powder, and stored at −80°C.
Experiment 2: genotype and nutritional status.
To assess whether hypothalamic FAS gene expression is regulated by nutritional state and genotype, two broiler chicken lines were used. These two lines were established by long-term divergent selection for ratio of abdominal fatness to live weight, in which the fat line (FL) had ∼1.5- to 2-fold higher proportional abdominal fat weight than the lean line (LL) at 9 wk of age (30). Animals (male) of each line were kept on a conventional floor pen, fed ad libitum with a balanced diet (3,100 kcal/kg, 22% protein), and exposed to a daily 14-h light period (lights on at 7:00 AM). At 9 wk of age, animals were submitted to two different nutritional states: fasting for 16 h (from 3:00 PM to 7:00 AM) and fed ad libitum (n = 3). After blood sampling and cervical dislocation were completed, tissues (hypothalamus and liver) were quickly removed, frozen in liquid nitrogen, and stored at −80°C until use.
Experiment 3: cerulenin administration.
Day-old broiler chickens (Ross) were purchased from a commercial hatchery (Avibel) and reared on floor pens until 1 wk of age, at which time the birds were transferred to individual cages and provided with individual feeders and drinking nipples. Food and water were available for ad libitum consumption, and lighting schedule provided 14 h of light per day (lights on at 7:00 AM). After 1 wk of adaptation, birds were divided into two homogenous weight- and food intake-matched groups (n = 4) and fasted for 2 h (from 7:00 to 9:00 AM) to increase their appetite. Each bird received an intravenous injection of 15 mg/kg cerulenin (Sigma) or equal volume of vehicle (10% DMSO in RPMI 1640 medium; Invitrogen). Food was returned to the cages, and intake was measured continuously (every hour). Cerulenin (same dose) was administered again after 4 and 24 h. At the end of the experiment, blood was taken from a wing vein for hormone and metabolite analysis. The chickens were then killed by decapitation, and tissues (hypothalamus) were removed, frozen in liquid nitrogen, and stored at −80°C until use.
Total RNA was extracted from 100 mg of tissue (brain, hypothalamus, liver, and muscle) using Trizol reagent (Invitrogen) according to the manufacturer’s recommendations. Pellets were suspended in 20–30 μl of DEPC-treated water. The quantity and integrity of isolated RNA were determined for each sample by using both ultraviolet absorbance (260/280 nm) and 1% agarose gel electrophoresis.
Reverse Transcription and Polymerase Chain Reaction
Reverse transcription was carried out with 1 μg of total RNA. First-strand cDNA was PCR amplified with specific primers for FAS (GenBank accession No. J04485), neuropeptide Y (NPY; GenBank accession No. M87294), agouti gene-related peptide (AgRP; GenBank accession No. AB029443), orexin (ORX; GenBank accession No. AB056748), orexin receptor (ORXR; GenBank accession No. AB110634), pro-opiomelanocortin (POMC; GenBank accession No. AB019555), corticotropin-releasing hormone (CRH; GenBank accession No. AJ621492), melanocortin receptors (MCR-1, -4, and -5; GenBank accession Nos. AY220305, XM426042, and AB012868, respectively), leptin receptor (extracellular domain named total Ob-R; GenBank accession No. AB033383), and ribosomal 18S (GenBank accession No. AF173612) as a control (for primers and conditions of PCR, see Ref. 13). For the intracellular domain of leptin receptor (named long isoform Ob-R), the regions selected for primer design correspond to sense 5′-GCTTGCTCAGGTAGCTCCTG-3′ and antisense 5′-TGCGGCACGTATGGCACGAT-3′. PCR was performed in 50 μl containing 2 μl of RT reaction, 1 unit of Taq DNA polymerase (Roche Diagnostic, Belgium), 0.1 mM dNTP mixture, and 10 pmol of each forward and reverse primer. After denaturation (94°C, 6 min), 27 cycles, including denaturation (94°C, 30 s), annealing (60°C, 30 s), and extension (72°C, 1 min) were performed, followed by a final elongation step (72°C, 10 min). The number of cycles used for each gene was in the linear amplification range.
The amplified fragments were separated on a low melting point agarose gel (1%), and the appropriate bands were cut out, purified by using Qiaquick gel extraction kit protocol (Qiagen), and stored at −20°C. The cDNA fragments were cloned in the pPCR Script Amp SK(+) cloning vector using the pPCR Script Amp cloning kit (Stratagene, La Jolla, CA) or TOPO cloning vector using TOPO PCR cloning kit (Invitrogen) and automatically sequenced by an ABI automated sequencer. The cloned fragments (25–30 ng) were labeled by random priming with [α-32P]dCTP (16).
Southern Blot Analysis
The amplified PCR products were transferred to nylon membrane by using a vacuum-blotting apparatus (Amersham Biosciences), cross linked by ultraviolet irradiation, and baked at 80°C for 20–30 min. Membranes were hybridized with heat-denatured 32P-labeled DNA probes, prepared as described above, at 42°C overnight.
During the following day, the membranes were rinsed twice with 1× SSC, 0.1% SDS at 55°C. Each wash was for 20 min, and then membranes were subjected to storage phosphor autoradiography cassette. Hybridization signals were quantified by phosphorimagery (Bio-Imaging analyzer BAS 1000 Mac BAS, Fujix, TINA software, version 2.09).
Western Blot Analysis
Tissues (hypothalamus, brain, liver, muscle) were homogenized in lysis buffer (10 mM Tris base, pH 7.4, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 0.5% NP-40, protease inhibitor cocktail; Sigma). Homogenates were centrifuged at 600 g for 20 min at 4°C, and supernatants were then ultracentrifuged for 45 min at 45,000 rpm. Protein concentrations were determined with a Bradford assay kit (Bio-Rad) with BSA as a standard. Proteins were run on 4–10% Novex Tris-glycine gels (Invitrogen Life Technologies). The transferred membranes were blocked in enhanced chemiluminescence blocking agent for 1 h at room temperature and incubated with rabbit polyclonal anti-FAS antibody (1:1,000) (Interchim) at room temperature overnight. The peroxidase-conjugated goat anti-rabbit secondary antibody was used (1:2,000) for 1 h at room temperature. The signal was visualized by enhanced chemiluminescence (ECL plus; Amersham Biosciences).
Male chickens (Ross; Avibel) were anesthetized and perfused transcardially first with 4°C PBS, pH 7.4, and then with 4°C filtered 4% paraformaldehyde in PBS (pH 7.4). On perfusion, brains were removed and postfixed in bouin-hollande solution (Sigma) for 24 h. Brains were washed successively in sterile water (24 h), ethanol (50%, 70%; 2 h), ethanol 95% (4 h), xylol (4 h and then overnight), xylol/paraffin (1/1; 4 h), and paraffin (4 h and then overnight). Brains were sectioned into 7-μm coronal sections with a historange CBK microtome.
Deparaffinized and rehydrated sections were initially washed in Tris-buffered saline (TBS; 0.01 M, 0.9% NaCl, 0.3% Triton X-100, pH 7.6), subsequently incubated in 0.3% H2O2 for 20 min at room temperature to inhibit endogenous peroxidase activity, rinsed for 10 min with TBS, and then preincubated in normal goat serum (1:5, 45 min, room temperature). The sections were incubated with the primary rabbit polyclonal anti-FAS antibody (1:500, overnight, room temperature; Interchim). After several rinses in TBS, the sections were incubated with biotinylated goat anti-rabbit IgGs (1:500, 30 min, room temperature) (Dako, Glostrup, Denmark) and with peroxidase-conjugated streptavidin (1:600, 30 min, room temperature; Dako). After a thorough rinsing in TBS, the peroxidase reaction was revealed with a fresh solution containing glucose oxidase, 3,3′-diaminobenzidine, and nickel ammonium sulfate (Sigma-Aldrich). The sections were dehydrated, coverslipped, and observed with a DM RBE microscope (Leica, Heidelberg, Germany). All dilutions were made in TBS, and all incubations were performed under gentle agitation.
Plasma Parameters Determination
Commercial colorimetric diagnostic kits were used to measure plasma glucose (IL test kit, no. 182508-00), triglycerides (IL test kit 181610-60), uric acid (IL test kit 181685-00), nonesterified fatty acid (Wako Chemicals, Neuss, Germany), creatine kinase (IL test kit 181605-90), albumin (IL test kit 18250040), lactate (Sigma kit 826), and total protein levels with an automated spectrophotometer (Monarch Chemistry Systems, Instrumentation Laboratories, Zaventem, Belgium). Plasma leptin concentrations were measured by RIA (Linco Research). Samples were assayed in a single assay, and the intra-assay coefficient of variation was 6.3%. We measured plasma thyroxine (T4) and tri-iodothyronine (T3) levels by RIA using commercial rabbit anti-T4 and anti-T3 antibodies (Cambridge Medical Technology/Ventrex, Billerica, MA) and unlabeled T4 and 3,3′,5-triiodo-l-thyronine standards (Sigma) (10). Intra-assay coefficients of variation were 5.9 and 4.9% for T4 and T3, respectively. The plasma corticosterone concentrations were measured by RIA kit (IDS). The intra-assay variability was 4.6%.
We performed statistical analyses using Statistical Analysis System (SAS) software (SAS Institute, 2000, version 8.1). Repeated-measures ANOVAs, followed by post hoc multiple comparisons and Student-Newman-Keuls tests, were performed to detect changes in food intake over time during cerulenin treatment. A P < 0.05 was considered significant.
Data from Southern blot analysis were expressed in arbitrary densitometry units normalized to the 18S rRNA levels, and values are expressed as means ± SE. Data from experiment 2 were analyzed by two-factor ANOVA with genotype (FL and LL) and nutritional state (fed or fasted) as classification variables. When a significant model was discerned, treatment means were separated by the post hoc Tukey’s test. All other data were compared by using the Student’s unpaired t-test. For RIA curve analysis, Graph Pad software (version 2, 94–95) was used.
Characterization of FAS mRNA and Protein Expression in Chicken Hypothalamus
RT-PCR, with total RNA derived from the brain and the hypothalamus, demonstrated the presence of FAS mRNA (Fig. 1A). By using specific primers for chicken FAS, a band of the predicted size (423 bp) was obtained in the brain and in the hypothalamus (Fig. 1A), as well as in the liver and in the muscle, which were chosen as the positive controls. The fragments were cloned and sequenced. The sequences obtained were 100% identical to those previously described by Chirala et al. (7). The presence of FAS protein in whole brain and hypothalamus lysates was demonstrated by Western blotting with polyclonal rabbit anti-FAS. As depicted in Fig. 1B, an immunoreactive band located at ∼250 kDa, was revealed in both brain and hypothalamus, as well as in the liver and muscle, which were chosen as the positive controls. The same data were observed in the other animals (n = 2). In the hypothalamus, FAS-immunoreactive cells were observed along the walls of the third ventricle inside the nucleus paraventricularis magnocellularis (PVN) (Fig. 2D) and the nucleus periventricularis hypothalami (Fig. 2E). In the posterior hypothalamic region, FAS-positive cells were localized in the nucleus infundibuli hypothalami (IN), the avian equivalent of the mammalian arcuate nuclei (Fig. 2I). These results indicated that FAS (mRNA and protein) is expressed in chicken hypothalamus and are widely distributed in the appropriate nuclei involved in feeding behavior, whereas the physiological relevance is yet speculative.
The next study aimed to establish whether hypothalamic FAS gene expression responds to a change in nutritional state and genotype in broiler chickens.
Effect of Fasting and Genotype on Hypothalamic FAS and NPY Gene Expression in Broiler Chickens
One group of each line (FL and LL) (n = 3) was maintained on a normal diet and fed ad libitum, whereas the second group was denied access to food for 16 h before tissue harvesting (hypothalamus and liver). Liver tissue was used as a positive control because FAS mRNA levels are downregulated in chicken liver during starvation (3). Two-factor ANOVA revealed a significant effect of nutritional state on hypothalamic FAS gene expression (P < 0.05, Fig. 3A). Compared with the fed state, overnight fasting significantly reduced the hypothalamic FAS gene expression, although this effect was only apparent in the FL chickens and not in the LL chickens as reflected in the significant genotype by nutritional state interaction (P < 0.01). FL chickens were characterized by significantly higher FAS mRNA levels in the liver than their LL counterparts (genotype effect P < 0.0001; Fig. 3C). As expected, food deprivation caused a significant (effect of nutritional state, P < 0.0001) decrease in hepatic FAS gene expression. However, this depressive effect was more pronounced in the LL chickens (genotype by nutritional state interaction: P < 0.01; Fig. 3C).
In contrast to FAS gene expression, hypothalamic NPY mRNA levels were significantly upregulated in both the fasted FL and LL chickens compared with that of their ad libitum-fed counterparts (effect of nutritional state, P < 0.0001; Fig. 3B). This upregulation of NPY was more pronounced in the FL than in the LL chickens (genotype by nutritional state interaction, P < 0.05). Furthermore, in the fasted state, hypothalamic FAS and NPY gene expression did not differ between the two lines (Fig 3, A and B). However, the hepatic FAS gene expression was significantly higher (P < 0.05; Fig. 3C) in FL compared with LL chickens.
Effect of Cerulenin Administrations on Food Intake and Body Weight in Broiler Chickens
Intravenous injection of 2-wk-old broiler chickens with cerulenin (15 mg/kg body wt) significantly reduced food intake relative to RPMI 1640 control (P < 0.05 to P < 0.0001; Fig. 4). This reduction was already observed 2 h after administration. After the second injection (4 h after the first one), the reduction in food intake was still significant (P < 0.001), reached an average of 34% relative to the placebo, and was maintained during the next 18 h. The third injection (24 h after the first injection) led to a 23% mean reduction in food intake compared with the RPMI 1640-treated group (P < 0.0001). Cerulenin did not significantly affect body weight of chickens compared with the placebo. In both groups, body weight was increased, although the increment was significantly lower (P < 0.05) in the cerulenin-treated group than in the control group (Table 1).
Effect of Cerulenin Administrations on Plasma Hormone and Metabolite Concentrations
Cerulenin administrations in 2-wk-old broiler chickens did not elicit a significant change in plasma hormone and metabolite levels monitored in this study except for plasma lactate levels, which were significantly increased after cerulenin administrations (P < 0.05; Table 2).
Effect of Cerulenin Administrations on Hypothalamic FAS and Neuropeptide Gene Expression
Cerulenin administrations significantly downregulated the hypothalamic FAS gene expression in 2-wk-old broiler chickens compared with the control (0.88 ± 0.06 vs. 1.11 ± 0.04; means ± SE, n = 4, P < 0.05; Fig. 5A). However, cerulenin did not significantly affect the expression of either orexigenic (NPY, AgRP, ORX, and ORXR) (Fig. 5B) or anorexigenic neuropeptides (POMC, CRH) (Fig. 5C). Despite the absence of an effect on expression of genes for POMC [the precursor of α-melanocyte-stimulating hormone (α-MSH), the agonist of the MCR-3 and-4] and AgRP [the antagonist of MCR-3 and-4], cerulenin injections significantly downregulated the expression of MCR-1, -4, and -5 in the hypothalamus of broiler chickens compared with the RPMI 1640-treated group (1.10 ± 0.05 vs. 0.611 ± 0.12 for MCR-1; 1.03 ± 0.04 vs. 0.52 ± 0.03 for MCR-4, and 1.30 ± 0.03 vs. 0.81 ± 0.02 for MCR-5; means ± SE, n = 4, P < 0.05; Fig. 5C).
Both FAS mRNA and protein are expressed in the hypothalamus of chickens. The observation of FAS immunoreactivity in feeding-related hypothalamic nuclei (PVN and IN) corroborates previous observations in rodents and humans (25) and suggests that FAS may interact with the expression and/or release of peptides that control food intake.
To determine whether hypothalamic FAS is regulated by feeding status or fat pad size, we compared FAS mRNA levels in hypothalamic and hepatic tissues in fed and fasted FL and LL chickens (30). Fatty acid synthesis occurs during periods of energy surplus, and concomitantly FAS gene expression is downregulated during starvation in liver (3), which is the major site of lipogenesis in avian species (18, 31, 32). The regulation of hypothalamic FAS gene expression in response to starvation was similar to that of liver FAS. Indeed, both hypothalamic and hepatic FAS mRNA levels were significantly reduced by fasting compared with the fed state in the FL chickens. However, this reduction was not significant in LL chickens for hypothalamic FAS gene expression but was even more pronounced for hepatic FAS mRNA levels. These differences may be related to genotype-specific, as well as tissue-specific, regulation of FAS gene expression, as previously reported for lipogenic genes (20, 26). In addition, the effect of starvation on chicken hypothalamic FAS mRNA levels seems to be different from that described in mammals in which fasting did not significantly affect the hypothalamic FAS gene expression (25). The mechanism behind this discrepancy is not clear and may be related to duration of fasting and/or subtle species-dependent differences. Interestingly, fasting induces a pattern of two bands of FAS, and the highest one was only apparent after labeling; therefore, we could not sequence it. We speculate that this fasting-induced pattern might represent other chicken FAS homolog or splice variants, especially because avian liver contains two discrete FAS transcripts (3).
We have also shown that the expression of NPY, the best-described anabolic effector, significantly increased with overnight fasting in both lines. This result is in good agreement with previous findings in mammals (41) and in birds (4, 23). More interestingly, independent of the nutritional state, hypothalamic NPY and FAS gene expression did not differ between the two genotypes. These data explain, at least partly, the same ingestive behavior in FL and LL chickens. Indeed, in contrast to hyperphagic obese rodent models such as ob/ob mice (59), food intake, body weight, basal energy requirement, digestive use of dietary energy, and body temperature are similar between FL and LL chickens (30). Hepatic FAS gene expression was higher in FL than in LL chickens, corroborating previous findings that the FL chickens are characterized by a higher hepatic de novo fatty acid synthesis compared with their LL counterparts (39). These results showed that hypothalamic FAS gene expression was decreased by feeding status and suggested that FAS may be involved in the brain control of appetite and satiety in chickens.
Recent data have shown that administration of FAS inhibitors such as cerulenin and C75 potently reduce food intake and cause profound weight loss in mice and rats (35, 43, 44). To investigate the physiological consequences of in vivo inhibition of fatty acid synthesis on food intake in chickens, we administered cerulenin to 2-wk-old broiler chickens by intravenous injections. As in mammals, cerulenin administration caused a significant reduction (23–34%) of food intake in chickens. The treatment was well tolerated by the chickens with no obvious toxicity (55). Necropsy analysis of all major organs in the treated group revealed no pathology. Furthermore, the dose of cerulenin used in this study has been shown to be not toxic in rodents. In addition, we did not observe any differences in the behavior, including physical activity between vehicle- and cerulenin-treated chickens, suggesting that the effect of cerulenin is not related to known effects of toxins, malaise, or illness. Nevertheless, no conditioned taste aversion tests have yet been performed; therefore, such effects cannot yet be ruled out completely.
We next assessed the mode of cerulenin action on food intake by determining its effect on the expression of hypothalamic feeding-related neuropeptide genes. Cerulenin administrations significantly downregulated the expression of FAS gene in chicken hypothalamus; however, the expression of orexigenic (NPY, ORX, ORXR, and AgRP) and anorexigenic (POMC and CRH) neuropeptide mRNA remained unchanged. This result suggests that cerulenin inhibits food intake in chickens independent of the known hypothalamic anabolic and catabolic neuropeptide gene expression. However, this result should be interpreted cautiously because gene expression is not the only measure to consider, and a possible effect of cerulenin on the release of these neuropeptides cannot be excluded. In addition, we sampled the whole hypothalamus and whether the effect of cerulenin on the neuropeptide gene expression may be different according to the hypothalamic nuclei.
Previous studies investigating the interaction between FAS inhibitors (cerulenin and C75) and hypothalamic neuropeptides in the regulation of food intake in mammals have produced conflicting results. These discrepancies may be related to the physiological context and models and whether FAS inhibition occurs acutely or chronically; the subsequent catabolic responses could be mediated via different pathways, as previously proposed by Mobbs and Makimura (36). Among these pathways, leptin and neuropeptides have been the subject of intense research and discussion in the past few years.
In birds, the leptin system seems to be quite different from that described in mammals. Chicken leptin is expressed not only in adipose tissue but also in the liver (2, 52), whereas in mammals, it is mainly expressed in adipose tissue (59). We have recently shown that leptin induces hepatic FAS gene expression in chickens (12). However, it inhibits lipogenesis and stimulates β-oxidation in mammals (5, 17, 56, 57). Despite these differences, leptin inhibits food intake in chickens, and this inhibition was selectively mediated via a downregulation of orexigenic hypothalamic neuropeptides (NPY, ORX, and ORXR) and MCR-4 and MCR-5 rather than AgRP and anorexigenic neuropeptides (POMC and CRH) (13). The downregulation of MCR-4/5 gene expression by leptin treatment without modification of AgRP [the antagonist of MCR-3 and -4 (34)] and POMC mRNA levels [the precursor of α-MSH: the potent agonist of MCR-3 and -4 (1)] indicates that MCRs may play an important role in mediating leptin effects on feeding behavior in chickens. These observations, in combination with the present results (no effect of cerulenin on anabolic and catabolic hypothalamic neuropeptide gene expression), suggest that MCRs may be a potential pathway to mediate the observed effect of cerulenin on food intake in chickens. Our data support this concept and showed for the first time that cerulenin administration decreases MCR-1, -4, and -5 mRNA levels in chicken hypothalamus without modulating AgRP and POMC gene expression. However, whether the effects of cerulenin on MCRs are direct or indirect are unclear and yet to be determined. Furthermore, these observations do not exclude a possible compensatory response for the MCRs after changes of the release of one or some feeding-related neuropeptide(s), such as α-MSH by cerulenin treatment.
It is known that the melanocortin system coordinates the maintenance of energy balance in mammals through the regulation of both food intake and energy expenditure (22). In Agouti yellow obese (Ay) mice, which develop obesity because of blockade of hypothalamic MCR-4 secondary to ectopic expression of AgRP, cerulenin treatment reduced food intake and body weight (34), suggesting that cerulenin may act through mechanism(s) that is (are) independent of MCRs. This hypothesis should not be readily accepted as definitive because nature is rich in backup mechanisms and a compensated mechanism has not been excluded.
In birds, five receptor genes belonging to the MCR family have been recently cloned (47–51). MCR-1 is implicated in melanogenesis within melanocytes because it has been mapped at the genetic locus, which acts to control feather color pigmentation in chickens (49). MCR-2 has been suggested to regulate steroidogenesis in chickens (47); however, no clear function has yet been ascribed to the other MCRs. MCR-3 is expressed exclusively in the adrenal gland (51), whereas MCR-4 and -5 showed a ubiquitous expression in peripheral tissues and in the brain (50). Our data suggest that the effect of cerulenin on food intake in chickens may involve changes in signaling of MCRs.
Interestingly, cerulenin injections significantly increased the plasma lactate levels in chickens, whereas the plasma glucose levels remained unchanged, indicating that cerulenin may affect the Cori cycle and could increase the malonyl-CoA levels in the chicken hypothalamic neurons by increasing the acetyl-CoA substrate. This observation raises the hypothesis that cerulenin may reduce food intake in chickens by also acting on hypothalamic glucose-sensing neurons through malonyl-CoA accumulation as previously suggested for mammals (21, 33, 58). We were not able to measure the malonyl-CoA activity in this study because of a limited quantity of hypothalamic tissues.
In conclusion, the present study is the first to report the effect and the mode of cerulenin’s action on food intake in chickens (nonmammalian species). The catabolic effect of cerulenin seems to be independent of gene expression for the known orexigenic (NPY, ORX, and AgRP) and anorexigenic (POMC and CRH) neuropeptides but dependent on the melanocortin system. Furthermore, FAS-positive cells are widely distributed in feeding-related hypothalamic nuclei, especially in the PVN and IN nuclei, and its gene expression varied depending on energy status (higher in fed state and lower in fasted state), suggesting that FAS may play a crucial role in the brain-food intake and body weight control. Further studies of FAS function would shed light on its role in food intake regulation in chickens.
This work was supported by a research grant (G.0402.05) from the Fonds voor Wettenschappelijk Onderzoek-Vlaaderen (Research Foundation)-Flanders.
The authors thank Gerda Nackaerts and Inge Vaesen for skilled technical assistance and Anne Collin (INRA Tours) for providing the FL and LL chickens.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society