| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
St. Vincents Institute (G.R.S., B.E.K.) and Department of Medicine (G.R.S., B.E.K.), St. Vincents Hospital, Fitzroy 3065, Australia; Department of Medicine (B.C.F., J.P., S.A.), Heidelberg Repatriation Hospital, 3081 Heidelberg, Germany; Department of Physiology (A.M.A.), University of Melbourne, and Commonwealth Scientific and Industrial Research Organization Molecular and Health Technologies (B.E.K.), Parkville 3052, Australia; and Cellular and Molecular Metabolism Laboratory (M.J.W., M.A.F.), School of Medical Sciences, Royal Melbourne Institute of Technology, Bundoora, Victoria 3083, Australia
Address all correspondence and requests for reprints to: Gregory R. Steinberg, St. Vincents Institute, 9 Princes Street, Fitzroy, Victoria 3065, Australia. E-mail: gsteinberg{at}svi.edu.au.
| Abstract |
|---|
|
|
|---|
2 activity, AMPK
2Thr172 phosphorylation, and acetyl-coenzyme A carboxylase phosphorylation, effects not observed 2 or 6 h after injection. Intracerebroventricular CNTFAx15 reduced food intake, increased arcuate nucleus (ARC) signal transducer and activator of transcription 3 phosphorylation, and reduced AMPK signaling but not in the paraventricular nucleus (PVN), posterior hypothalamus, or cortex. To compare the effects of leptin and CNTFAx15 in a diet-induced model of obesity, mice were fed a control carbohydrate or high-fat diet (HFD) for 12 wk. Leptin treatment ip reduced food intake in control mice but not in mice fed a HFD. In contrast, ip CNTF markedly reduced food intake in both control and HFD animals. Both leptin and CNTF reduced AMPK activity and acetyl-coenzyme A carboxylase phosphorylation in the ARC and PVN of control-fed mice. A HFD blunted leptin but not CNTF effects on AMPK signaling in the ARC and PVN. In summary, these data demonstrate that CNTFAx15 bypasses diet-induced leptin resistance to reduce hypothalamic AMPK activity. | Introduction |
|---|
|
|
|---|
-subunit activation loop by the upstream kinases, LKB1 (3, 4) and calmodulin-dependent kinase kinase (5, 6, 7). Whereas most research has focused on mechanisms mediating AMPK activity in peripheral tissues such as skeletal muscle and liver, more recently a critical role of AMPK as a regulator of food intake has been revealed. The inhibition of AMPK activity mediates the hypothalamic suppression of appetite by leptin (8, 9),
-lipoic acid (10), and the fatty synthase inhibitor, C75 (11, 12), whereas activation of AMPK by glucose deprivation (9) and ghrelin (8) stimulates appetite. Ciliary neurotrophic factor (CNTF) is a 22-kDa cytokine that was found to induce severe anorexia and weight loss during clinical trials for the treatment of amyotrophic lateral sclerosis (13). Whereas CNTF-induced weight loss was initially attributed to a cachectic response, subsequent studies in diet-induced (14) and genetic (ob/ob, MC4R/) obesity (15, 16) demonstrated that low doses of CNTF and the CNTF homolog, axokine (CNTFAx15) induced weight loss without causing the typical deleterious effects of other related cytokines, such as IL-1. Indeed, the safety and efficacy of CNTFAX15 as a weight loss agent have been confirmed in a recent clinical trial, although the development of neutralizing antibodies may restrict its general application (17). Obese rodents (14, 18) and humans (17) chronically administered CNTF exhibit reduced rebound weight gain after the cessation of treatment due to a sustained reduction in caloric intake, which results in the prolonged maintenance of weight loss. This effect has recently been attributed to hypothalamic neurogenesis, which appears to be capable of resetting the energy-balance set point (18).
The IL-6 family of cytokines consisting of IL-6, IL-11, leukemia inhibitory factor, oncostatin M, cardiotrophin-1, and CNTF are highly redundant in their ability to transduce biological activities, but specificity of responses is ensured by tissue-specific expression of their ligand-specific
-subunits (19). Both systemic leptin (20) and CNTF (15) reduce food intake and body mass in leptin-deficient ob/ob mice. CNTF receptor expression is localized within the hypothalamic neurons of the arcuate nucleus (ARC), an area that overlaps substantially with neurons activated by leptin (18). The effects of both leptin and CNTF are attributed to the similar expression patterns and signaling homology of both the leptin and CNTF receptor within hypothalamic regions involved in food intake and similar activation of the signal transducer and activator of transcription (STAT) 3 signaling (14, 21). The activation of STAT3 within the ARC is associated with the suppression of the synthesis of orexigenic peptides such as neuropeptide Y (NPY) and agouti-related peptide (AgRP) that in turn leads to suppression of food intake through a diverse and multifaceted signaling cascade (for review see Ref. 22).
In addition to regulation by STAT3, orexigenic peptides are also regulated by AMPK. In elegant experiments using an in vivo adenovirus approach in which dominant-negative (DN) and constitutively active (CA) mutations of AMPK were injected into the ventral medial hypothalamus. Minokoshi et al. (9) showed that a CA-AMPK increased body weight and food intake, whereas a DN-AMPK had the opposite effect. Increased food intake in animals treated with a CA-AMPK was related to increased expression of arcuate NPY and AgRP when in the fasted state, whereas the reciprocal relationship was observed in the fed state with the DN-AMPK. Lee et al. (23) extended these findings in a neuroblastoma cell line by showing that the modulation of cellular ATP and therefore AMPK by glucose, 2-deoxyglucose, pyruvate, or ATP synthesis inhibitors also altered the expression of AgRP. Similarly, the fatty acid synthase inhibitor C75 was also found to suppress ARC AMPK and NPY expression, an effect that was reversed in the presence of the AMPK activator 5-amino-imidazole-4-carboxamide-1-ß-D-ribofuranoside (12). In agreement with these observations, elevated hypothalamic AMPK
2 activity in diabetic rats was associated with elevated NPY and suppressed proopiomelanocortin (POMC) mRNA, effects that were reversed in the presence of the AMPK inhibitor, compound C (24).
Despite the pronounced effects of leptin on reducing food intake in a model of leptin deficiency (ob/ob mice), these effects are abrogated in diet-induced mice fed a high-fat diet (25). In contrast, the effect of CNTF on food intake persists in diet-induced obesity and in db/db mice, which lack a functional leptin receptor (15). In this study, we tested the hypothesis that like leptin, CNTFAx15 might also suppress hypothalamic AMPK signaling. Moreover, in light of the observations of the persistent effects of CNTFAx15 but not leptin on food intake in diet-induced obesity, we examined the effects of both cytokines on AMPK activity after 12 wk of high-fat feeding.
| Materials and Methods |
|---|
|
|
|---|
In preliminary experiments to test our ability to isolate specific sections of the hypothalamus, experiments in six untreated C57BL6/J mice were conducted. The following four hypothalamic dissections were made: 1) an area containing predominantly the paraventricular nucleus (PVN), 2) an area containing predominantly the anterior hypothalamus, 3) an area containing predominantly the ARC and 4) an area containing predominantly posterior hypothalamus (PH). These hypothalamic sections were dissected on dry ice as previously described for rats with minor modifications (26). Briefly, mice were decapitated and coronal slices were made with the first between the optic chiasm and approximately 0.5 mm caudal hemisected into a dorsal region that contained the PVN and a ventral region containing the anterior hypothalamus. The remainder of the hypothalamus was dissected into a ventral region containing the ARC and a dorsal region containing the PH. A sample of the cortex was also collected. All samples were freeze clamped and stored at 80 C until analysis by real-time quantitative PCR analysis for POMC and AgRP mRNA, which are markers for the ARC. POMC and AgRP mRNA were highly enriched in the ARC samples but were not detected in samples from the PVN, PH, or cortex (Table 1
). These data suggest that the hypothalamic regions isolated were correct and representative of their respective nuclei, although it is also likely that our dissections contained other regions of the brain not described such as ventromedial nuclei in the case of the ARC and the lateral hypothalamus and perifornical area in the case of the PVN.
|
186200), Cys17Ala, Arg63Gln)]; Regeneron Pharmaceuticals, Tarrytown, NY) was injected ip into conscious mice, and hypothalami were excised on dry ice after decapitation after 45 min, 2h, and 6 h, freeze clamped, and stored at 80 C until further analysis. The ip administration of CNTFAx15 at a dose of 300 µg/kg·d reduces food intake and body weight to a similar degree to leptin (1.0 mg kg·d) in ob/ob mice (15). This leptin concentration has been demonstrated to reduce hypothalamic AMPK activity (8). Individually housed mice were injected ip at the start of the light cycle on two occasions separated over a 24-h period with PBS (vehicle), 300 µg CNTFAx15/kg body mass, or 1 mg leptin/ kg body mass (R&D Systems, Minneapolis, MN). Food intake and body mass were monitored daily over the 48-h period after treatments. Mice were then allowed to recover for 6 d on their respective diets. On the day before terminal experiments, mice were fasted overnight and the following morning injected ip with PBS, leptin, or CNTFAx15 as described above. Forty-five minutes after injection, mice were decapitated and the brain rapidly removed as described above.
Intracerebroventricular experiments
Mice were anesthetized with xylazine (10 mg/kg) and ketamine (75 mg/kg) via ip injection and secured in a stereotaxic frame. An incision was made in the skin and the skull was then leveled between lambda and bregma. Using a dental drill, a small hole was made in the skull to enable a 30-gauge guide cannula (Plastics One, Roanoke, VA) to be lowered toward the lateral ventricle (0.3 mm posterior to bregma, 1.0 mm lateral to the midline, and 3.0 mm below the surface of the skull). The cannula was then secured using dental acrylic and fixed in place with a skull screw. The cannula was blocked by a capped insert until the time of injection to maintain cannula patency and sterility and individually housed to avoid removal of cannula. Correct placement of the cannula into the lateral ventricle was confirmed the following day by injection of angiotensin II (50 ng). Animals not displaying a prompt and sustained drinking response were excluded from further study. This was approximately 8% of animals.
One week after these initial procedures PBS (vehicle) or CNTFAx15 (90 ng) were microinjected, at the start of the light cycle, into the lateral ventricle in a volume of 1 µl over 30 sec as previously described (27), and food intake was monitored over a 24-h period. Mice were allowed to recover for 6 d. On the day before terminal experiments, mice were fasted overnight and the following morning injected intracerebroventricularly (icv) with PBS or CNTFAx15. Forty-five minutes after injection, animals were killed via decapitation for harvesting of the brain as described above.
Analytical methods
Real-time quantitative PCR.
RNA was isolated from hypothalamic sections using the RNeasy minikit (QIAGEN, Doncaster, Australia) for total RNA isolation. An on-column DNase treatment was performed using an RNase-free DNase set (QIAGEN). Reverse transcription of mRNA (1.5 µg) was performed using the thermoscript RT-PCR system (Invitrogen, Mount Waverly, Australia) with random hexamer priming as recommended by the manufacturer. Quantitative real-time PCR was performed on the Rotorgene 3000 (Corbett Research, Sydney, Australia) using Assay-on-Demand gene expression kits for mouse POMC and AgRp (Applied Biosystems, Foster City, CA). cDNA was amplified using 20 µl TaqMan PCR containing 3 mM MgCl2, 200 mM deoxynucleotide triphosphates, 100 nm primers, 50 nM TaqMan probe, 1x gold reaction buffer, and 0.5 U Amplitaq gold (Applied Biosystems). Assays were performed in triplicate and normalized using 18s ribosomal RNA (Applied Biosystems) as an internal control. The relative quantities of each transcript were then calculated using the comparative critical threshold method.
AMPK activity assay.
Hypothalamic sections were homogenized in ice-cold buffer [20 mM HEPES (pH 7.5), 2 mm EDTA, 50 mM NaF, 5 mm Na4P2O7, 1% Nonidet P40 + 1% protease inhibitor cocktail (Complete, Roche, Stockholm, Sweden)] centrifuged at 14,000 x g for 25 min, and the supernatant was incubated with 10 µl AMPK
1 and AMPK
2 antibody-bound protein A agarose beads for 16 h. Immunocomplexes were washed and suspended in 50 mM Tris (pH 7.5) buffer for the AMPK assay in the presence of 200 µM AMP (28). Activities were calculated as picomoles of phosphate incorporated into the SAMS synthetic peptide per minute per milligram protein subjected to immunoprecipitation as previously described (28). Five microliters of the remaining beads were solubolized in sample buffer exposed to SDS-PAGE and transferred to polyvinyl difluoride membranes, blocked, and immunoblotted with antibodies detecting AMPK Thr172 phosphorylation and AMPK
1 and -
2 as previously described (29). After incubation with horseradish peroxidase-conjugated secondary antibody (1:2000; Amersham Biosciences, Castle Hill, New South Wales, Australia), the immunoreactive proteins were detected with enhanced chemiluminescence (PerkinElmer, Rowville, Victoria, Australia) and quantified by densitometry. For the determination of ACC samples, lysates were immunoprecipitated with streptavidin and immunoblotted as described (29).
Statistical analysis
Results are presented as the mean ± SEM. Data were analyzed for differences by one-way ANOVA with specific differences located with a Tukeys post hoc test. P < 0.05 was considered to be statistically significant.
| Results |
|---|
|
|
|---|
1 activity (Fig. 1A
1 Thr172 phosphorylation (Fig. 1B
2 activity (Fig. 1C
|
2 activity and phosphorylation (Fig. 2
1 phosphorylation was suppressed (Fig. 2D
1 activity did not achieve statistical significance after icv CNTFAx15 (Fig. 2C
|
|
To assess the effects of leptin and CNTFAx15 on food intake and body mass, mice were injected ip with saline, leptin, or CNTF Ax15 for 2 d at 0900 h. Both CNTFAx15 and leptin treatment reduced food intake to a similar degree in animals fed a CON diet (Fig. 3A
). Animals fed a HFD had a greater daily caloric intake than CON-fed animals (Fig. 3A
). Importantly despite the pronounced effects of leptin in animals fed a CON diet, leptin was ineffective in reducing food intake in HFD animals, whereas CNTFAx15 reduced food intake to a similar degree as that observed in animals fed the CON diet (Fig. 3A
). In line with reductions in food intake, both CNTFAx15 and leptin reduced body mass over the 2-d treatment period in animals fed a CON diet, whereas only CNTFAx15 reduced body mass in animals fed a HFD (Fig. 3B
). It should be noted that reductions in body mass were greater than expected from calorie restriction alone and were likely attributed to increased energy expenditure as previously reported (30).
|
2 activity (Fig. 4
2 activity in ARC and tended (P = 0.11) to do so in the PVN (Fig. 4
2 activity or ACC phosphorylation in the PH or cortex (Table 3
1 activity was unaltered by leptin or CNTFAx15 in the ARC, PVN, PH, and cortex (data not shown).
|
|
| Discussion |
|---|
|
|
|---|
We have shown that like leptin, CNTFAx15 injected both ip and icv selectively reduces hypothalamic AMPK
2 activity acutely, effects that are not maintained 2 or 6 h after ip injection. This transient reduction of AMPK
2 activity is consistent with the short half-life (
45 min) of CNTFAx15 in circulation (30), and our icv experiments support a direct role for CNTFAx15 in acutely reducing hypothalamic AMPK activity, an effect associated with chronic reductions in food intake. It should be noted that the icv injection of CNTFAx15 was not specific to hypothalamic neurons and may have had effects on other regions of the brain accessible by the ventricular system. Indeed, supporting this possibility, Bjorbaek et al. (32) demonstrated the rapid stimulation of the suppressor of cytokine signaling 3 (SOCS3) in the cerebellum after CNTF administration. Therefore, in the present study, we cannot discount the possibility that extrahypothalamic regions may have contributed to reductions in food intake by CNTFAx15.
Acute reductions in hypothalamic AMPK
2 activity after CNTFAx15 were associated with reduced phosphorylation of ACC, a downstream target of AMPK. Reduced phosphorylation of ACC is associated with elevated levels of malonyl-CoA, thereby reducing carnitine palmitoyltransferase (CPT)-1 activity (33). Chemical and genetic inhibition of hypothalamic CPT-1 have been shown to reduce food intake (34); thus, our results showing an inhibition of AMPK by CNTFAx15 are consistent with the role of CPT-1 as a regulator of appetite and suggests a mechanism by which reduced AMPK activity may alter food intake. In addition to direct modulation of CPT-1 activity, AMPK has been demonstrated to have direct effects on the expression of orexigenic NPY and AgRP neurons (9). Previous studies (14, 35) reported reduced NPY after CNTFAx15 treatment, and our data support the idea that AMPK is a transcriptional regulator of NPY. Taken together, our data suggest that the acute modulation of AMPK activity by CNTFAx15 may chronically alter food intake by both inhibition of CPT-1 and transcriptional regulation of AgRP and NPY. To assess the quantitative importance of hypothalamic AMPK in the regulation of food intake by CNTFAx15, future studies in hypothalamic-specific AMPK null mice are warranted.
Leptin signaling in the hypothalamus is essential for the regulation of body mass and neuroendocrine homeostasis (36). These effects are attributed, in part, to the binding of leptin to the long form of the leptin receptor (LRb) within hypothalamic nuclei, which regulates food intake and energy expenditure. Specifically, tyrosine phosphorylation of 1138 of LRb by leptin is central to these regulatory effects (37). Despite the pronounced effects of leptin in models of leptin deficiency (20) or lipodystrophy (38), rodents fed a HFD are resistant to the effects of leptin (25). Human obesity is similarly characterized by elevated levels of leptin (39), and recombinant leptin infusion has minimal effects on body mass and food intake under these conditions (40). In the present study, we demonstrate that diet-induced obesity blunts the acute effects of leptin on AMPK signaling, suggesting that the inability of leptin to reduce AMPK signaling may be a contributing factor to the elevated caloric intake observed in high-fat-fed animals, but hypothalamic-specific deletion of AMPK will be necessary to directly test this hypothesis.
A potential mediator of leptin resistance is SOCS3. Elegant studies by Bjorbaek et al. (41, 42, 43) illustrated that SOCS3 inhibits leptin signaling through Src homology 2 binding to Tyr 985 of the leptin receptor, resulting in suppressed hypothalamic STAT3 activation. More recently elevated SOCS3 has been observed within the ARC after high-fat feeding, an effect associated with suppressed leptin-dependent STAT3 phosphorylation (44). In support of a critical role of SOCS3 in the regulation of hypothalamic leptin resistance, studies in SOCS3 hypothalamic-specific null mice (45) or SOCS3 mice with haplo-insufficiency (46) demonstrated protection from diet-induced leptin and insulin resistance. Whereas the effects of SOCS3 on food intake is considered to be due to the inhibition of STAT3, SOCS3 also inhibits leptin activation of ERK signaling (47). These data indicate that SOCS3 inhibits alternative leptin signaling pathways and suggest that elevated SOCS3 after diet-induced obesity may also inhibit leptins effects on AMPK signaling.
Previous studies (14, 18, 21, 48) have shown that CNTF increases phosphorylation of STAT3 in hypothalamic neurons that overlap with those activated by leptin. Surprisingly, despite the similarities in signaling between leptin and CNTF in chow-fed animals, HFDs suppressed leptin but not CNTFAx15 effects on AMPK signaling. Whereas SOCS3 has been demonstrated to bind to the SHP-2-binding site of gp130 (49), we demonstrate that CNTFAx15 suppression of hypothalamic AMPK is not blunted by high-fat feeding and that CNTFAx15 is capable of reducing food intake in leptin-resistant animals. Whereas the LRb shares significant sequence homology with both gp130 and the leukemia inhibitory factor receptor (50), signaling is independent of these IL-6 class of receptors (51). The mechanism(s) mediating the sustained sensitivity to CNTF but not leptin are presently not understood.
In summary, the present study has demonstrated that acute reductions in hypothalamic AMPK signaling by CNTFAx15 represent a potentially important mechanism mediating CNTFAx15 effects on food intake. Importantly, the effects of CNTFAx15 are maintained in diet-induced obesity, whereas the suppressive effects of leptin on AMPK signaling are blunted, highlighting a potential mechanism contributing to leptin resistance and highlighting the potential therapeutic potential of CNTFAx15 as a treatment for human obesity. Further investigations examining the role of hypothalamic AMPK in mediating the chronic effects of CNTFAx15 for the treatment of obesity are warranted.
| Acknowledgments |
|---|
| Footnotes |
|---|
The authors declare no potential conflicts of interest.
First Published Online May 4, 2006
Abbreviations: ACC, Acetyl-coenzyme A carboxylase; AgRP, agouti-related peptide; ARC, arcuate nucleus; AMPK, AMP-activated protein kinase; CA, constitutively active; CNTF, ciliary neurotrophic factor; CNTFAx15, CNTF homolog, axokine; CON, control; CPT, carnitine palmitoyltransferase; DN, dominant negative; HFD, high-fat diet; icv, intracerebroventricularly; LRb, long form of the leptin receptor; NPY, neuropeptide Y; PH, posterior hypothalamus; POMC, proopiomelanocortin; PVN, paraventricular nucleus; SOCS3, suppressor of cytokine signaling 3; STAT, signal transducer and activator of transcription.
Received December 15, 2005.
Accepted for publication April 24, 2006.
| References |
|---|
|
|
|---|
/ß and MO25
/ß are upstream kinases in the AMP-activated protein kinase cascade. J Biol 2:28[CrossRef][Medline]
-lipoic acid mediated by suppression of hypothalamic AMP-activated protein kinase. Nat Med 10:727733[CrossRef][Medline]
-melanocyte stimulating hormone in the diet-induced obese rat. Brain Res 892:130137[CrossRef][Medline]This article has been cited by other articles:
![]() |
L. F Rezende, L. F Stoppiglia, K. L A Souza, A. Negro, F. Langone, and A. C Boschero Ciliary neurotrophic factor promotes survival of neonatal rat islets via the BCL-2 anti-apoptotic pathway J. Endocrinol., October 1, 2007; 195(1): 157 - 165. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Simler, A. Malgoyre, N. Koulmann, A. Alonso, A. Peinnequin, and A. X. Bigard Hypoxic stimulus alters hypothalamic AMP-activated protein kinase phosphorylation concomitant to hypophagia J Appl Physiol, June 1, 2007; 102(6): 2135 - 2141. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Cai, A. V Gyulkhandanyan, M. B Wheeler, and D. D Belsham Glucose regulates AMP-activated protein kinase activity and gene expression in clonal, hypothalamic neurons expressing proopiomelanocortin: additive effects of leptin or insulin J. Endocrinol., March 1, 2007; 192(3): 605 - 614. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. L. Carey, G. R. Steinberg, S. L. Macaulay, W. G. Thomas, A. G. Holmes, G. Ramm, O. Prelovsek, C. Hohnen-Behrens, M. J. Watt, D. E. James, et al. Interleukin-6 Increases Insulin-Stimulated Glucose Disposal in Humans and Glucose Uptake and Fatty Acid Oxidation In Vitro via AMP-Activated Protein Kinase. Diabetes, October 1, 2006; 55(10): 2688 - 2697. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |