gp130 signaling in proopiomelanocortin neurons mediates the acute anorectic response to centrally applied ciliary neurotrophic factor

POMC-Restricted Inactivation of gp130.

Mice carrying the loxP-flanked (flox) gp130 allele (13) were crossed with mice expressing the Cre recombinase under control of the Pomc promoter (14). Cre-mediated recombination was visualized through the use of a reporter mouse strain (Z/EG) in which transcription of a green fluorescent protein (GFP) gene under control of the ubiquitously expressed Rosa26 promoter is prevented by a floxed neomycin-resistance gene (15). These mice showed a pattern of GFP immunoreactivity in the ARC of the hypothalamus similar to POMC (Fig. 1a). Consistent with this, Western blot analysis revealed no alterations in overall brain and hypothalamic gp130 protein expression (Fig. 1b). Accordingly, gp130 expression in peripheral tissues remained unchanged in gp130^ΔPOMC mice (Fig. 1b).

Abolished CNTF-Stimulated STAT3 Tyrosine Phosphorylation in POMC Cells of gp130^ΔPOMC Mice.

To determine the effect of POMC cell-restricted deficiency of gp130 on CNTF’s induction of STAT3 phosphorylation, control and gp130^ΔPOMC mice expressing β-gal in POMC neurons were injected with recombinant CNTF i.v. In control mice, CNTF treatment resulted in profound activation of STAT3 phosphorylation in both POMC- and non-POMC-expressing neurons of the ARC (Fig. 1c). In contrast, in gp130^ΔPOMC mice, CNTF stimulation resulted in tyrosine phosphorylation of STAT3 in non-POMC cells, but it failed to activate STAT3 phosphorylation in POMC neurons (Fig. 1 c and d). To determine the specificity of gp130 ablation in POMC neurons for gp130-dependent cytokine signaling, both control and gp130^ΔPOMC mice were treated by i.v. injection of leptin. Leptin treatment evoked a clear stimulation of STAT3 tyrosine phosphorylation in both POMC and non-POMC cells of control and gp130^ΔPOMC mice (Fig. 1c). These data indicate, that POMC-restricted inactivation of gp130 selectively inhibits the signaling via gp130-coupled cytokine receptors such as the CNTF receptor, without abolishing gp130-independent cytokine receptor signaling such as that mediated by the leptin receptor.

Unaltered POMC Neuron Number and Stress Response in gp130^ΔPOMC Mice.

gp130 has been demonstrated to play a critical role in neurogenesis (16). Thus, we next determined whether lack of functional gp130 signaling alters the number or distribution of hypothalamic POMC neurons. This analysis revealed equal numbers and distribution of POMC neurons in both control and gp130^ΔPOMC mice (Fig. 1e). These data indicate that disrupted gp130 signaling in our animals did not affect neurogenesis and cell survival of POMC neurons.

Consistent with the expression pattern of endogenously expressed POMC, the only non-hypothalamic site of recombination detectable in PomcCre-transgenic mice was the pituitary (Fig. 2a). Because cytokines that signal through gp130-dependent receptors such as leukemia inhibiting factor (LIF) and IL-6 have been demonstrated to regulate adrenocorticotropin expression and release in cultured pituitary cells (17), we next characterized the functional stress response of gp130^ΔPOMC mice. Pomc mRNA expression in the pituitary of control and gp130^ΔPOMC mice did not show a significant difference (Fig. 2b). To directly determine the stress response in control and gp130^ΔPOMC mice, we used two paradigms activating the corticotropic response via distinct mechanisms, i.e., restraint stress and LPS injection. Both stimuli dramatically increased plasma corticosterone concentrations to a comparable extent in control and gp130^ΔPOMC mice (Fig. 2c). As expected, LPS injection resulted in a profound increase of circulating IL-6 concentrations (18) (Fig. 2d). This increase was indistinguishable between control and gp130^ΔPOMC mice (Fig. 2d). Taken together, these data indicate that inactivation of gp130 signaling in POMC cells of the hypothalamus and pituitary does not impair responses to restraint and LPS-induced stress in these animals.

Normal Energy and Glucose Homeostasis in gp130^ΔPOMC Mice.

To determine the importance of endogenous gp130-dependent signaling in POMC cells on the regulation of energy homeostasis, we first monitored the body weight of control and gp130^ΔPOMC mice from weaning until 6 months of age. This analysis revealed no change in body weight in the presence of POMC-restricted gp130 deficiency both under standard and high-fat diets (Fig. 3a). White adipose tissue mass and circulating plasma leptin concentrations were indistinguishable between control and gp130^ΔPOMC mice (Fig. 3 b and c). In control and gp130^ΔPOMC mice, there was an increase in adipose tissue mass and plasma leptin concentrations upon exposure to high-fat diet (Fig. 3 b and c). Food intake and basal metabolic rate were also indistinguishable between control and gp130^ΔPOMC mice (Fig. 3 d and e). The expression of anorexigenic neuropeptides such as POMC and cocaine- and amphetamine-related transcript (CART) was unaltered in control and gp130^ΔPOMC mice (Fig. 3f). Body length was indistinguishable between control and gp130^ΔPOMC mice, a further indication of intact function of the melanocortin pathway in the absence of gp130 signaling in POMC neurons (Fig. 3g).

Consistent with unaltered energy homeostasis in gp130^ΔPOMC mice, these animals exhibited normal glucose metabolism as assessed by glucose tolerance tests, insulin tolerance tests, and blood glucose and plasma insulin concentrations (Fig. 5 a, c, e, and f, which is published as supporting information on the PNAS web site). Exposure to high-fat diet resulted in impaired glucose tolerance, increased blood glucose concentrations, and elevated plasma insulin concentrations as an indirect measure of insulin resistance in a similar manner in control and gp130^ΔPOMC mice (Fig. 5 b and d–f). Taken together, these data indicate that gp130 signaling in POMC neurons is dispensable for normal control of body weight, food intake, energy expenditure, and glucose metabolism.

Blunted Anorectic Response to Intracerebroventricularly (i.c.v.) Injected CNTF in gp130^ΔPOMC Mice.

To analyze the effect of exogenous CNTF on food intake, we implanted cannulae into the lateral ventricle of female and male control and gp130^ΔPOMC mice. In females, analysis of food intake 4 h after CNTF injection revealed a reduction in food intake in control mice by ≈80% compared with 0.9% NaCl injection (Fig. 4a). Strikingly, however, this effect was severely blunted in gp130^ΔPOMC mice, which exhibited only a 35% reduction in food intake (Fig. 4a). Similarly, in females, food intake over 24 h after a single injection of CNTF was significantly reduced by 40% in control animals but only resulted in a decrease of 18% in gp130^ΔPOMC mice (Fig. 4a). In addition, CNTF reduced body weight significantly 24 h after injection in control females but exhibited only a minor effect in gp130^ΔPOMC mice (Fig. 4b). Similar results were obtained when male control and gp130^ΔPOMC mice were acutely treated with CNTF: 6-h food intake after single injection was significantly reduced by 60% in control animals, whereas gp130^ΔPOMC mice failed to exhibit a significant reduction of food intake (Fig. 4c). Body weight significantly decreased by 4% in CNTF-treated controls, whereas it remained unchanged in gp130^ΔPOMC mice (Fig. 4d). Taken together, CNTF very rapidly and severely impairs food intake over a 24-h period in control animals, an effect that is blunted in gp130^ΔPOMC mice.

CNTF Fails to Induce c-Fos Immunoreactivity in the Paraventricular Nucleus (PVN) of gp130^ΔPOMC Mice.

The PVN of the hypothalamus is a major site of action of POMC efferents in the regulation of feeding (19). Therefore, we analyzed the activation of PVN neurons in response to acute CNTF treatment in control and gp130^ΔPOMC mice. In control animals, CNTF evoked a strong induction of c-Fos immunoreactivity in the PVN (Fig. 4 e and f). Strikingly, however, CNTF treatment completely failed to enhance c-Fos immunoreactivity in the PVN of male gp130^ΔPOMC mice (Fig. 4 e and f). Thus, CNTF application acutely and strongly activates neurons in the PVN that are characterized as important mediators of energy homeostasis regulation (19). The absence of PVN neuronal activation in response to CNTF application in gp130^ΔPOMC mice clearly indicates that PVN activation critically depends on functional CNTF signaling in POMC neurons.

Animals.

All animal procedures were conducted in compliance with protocols approved by local government authorities (Bezirksregierung Köln) and were in accordance with National Institutes of Health guidelines. Mice were housed in groups of three to five at 22–24°C in a 12-h light/12-h dark cycle with lights on at 6 a.m. Animals were fed either regular chow food (Global Rodent T.2018.R12 from Harlan Teklad, containing 12% of calories from fat) or a high-fat diet (C1057 from Altromin, containing 55.2% of calories from fat). Water was available ad libitum, and food was only withdrawn if required for an experiment. Body weight was measured once per week. At the end of the study period, the animals were killed under isoflurane anesthesia.

Generation of PomcCre-gp130^lox/lox Mice.

PomcCre mice (14) were mated with gp130^lox/lox mice (13), and a breeding colony was maintained by mating gp130^lox/lox with PomcCre-gp130^lox/lox mice. gp130^lox mice had been backcrossed for at least five generations on a C57BL/6 background, and PomcCre mice (initially established on an FVB background) had been backcrossed for two generations on a C57BL/6 background before intercrossing them with gp130^lox mice. Only animals from the same mixed background strain generation were compared with each other. PomcCre mice were genotyped by PCR as described in ref. 14. gp130 mice were genotyped by PCR with the following primers: 5gp130, 5′-GGT GGC TGA TTC ACC TGC A-3′; and 3gp130, 5′-TAC GCT GGG CAG CGT CCT-3′. For visualization of Cre-mediated recombination, mice were crossed with either the indicator strain Rosa(lacZ reporter) (Fig. 6, which is published as supporting information on the PNAS web site) (29) or the Z/EG reporter strain (15).

Restraint Stress and LPS Treatment.

Mice were adapted to gentle handling for ≈8 weeks before the experiment. To determine basal serum corticosterone and IL-6 levels, blood was drawn from the tail vein during the first 3 h of the light phase. The next day, mice were subjected to 1 h of restraint stress at the same time of the light phase. Restraint stress was achieved by enclosing the animals in a plastic tube with a diameter of 3 cm and openings at both ends for tail and nose. At the end of the experiment, blood samples were again drawn from the tail vein.

Seven days later, the animals were injected i.p. with 0.9% NaCl during the first 3 h of the light phase, and blood was collected from the tail vein 1 h after injection. After 5 days of recovery, each mouse received an i.p. injection of 300 μg of LPS (L2630 from Sigma) 1 h after injection blood samples were taken. Two days after LPS injection, the animals were killed under isoflurane anesthesia.

Glucose and Insulin Tolerance Tests.

Glucose and insulin tolerance tests were performed as described in ref. 30.

Indirect Calorimetry and Food Intake.

All measurements were performed with the Comprehensive Laboratory Animal Monitoring System (CLAMS) (oxymax; Columbus Instruments, Columbus, OH). Mice were placed at room temperature (22–24°C) in 3.0-liter chambers of the CLAMS open circuit calorimetry. Settling time was set at 150 s, and measuring time was set at 60 s with room air as reference. Food and water were provided ad libitum in the appropriate devices. Mice were allowed to acclimatize in the chambers for 24 h. Parameters of indirect calorimetry were measured for at least the following 72 h. Food intake was measured continuously in the CLAMS during the experiment. The data presented are average values obtained in these recordings.

Western Blot Analysis.

Indicated tissues were dissected and homogenized in homogenization buffer with a Polytron homogenizer (IKA Werke, Staufen, Germany), and Western blot analyses were performed as described in ref. 31 with antibodies raised against gp130 (sc-656 from Santa Cruz Biotechnology) and Pten (MS-1250-P from NeoMarkers) as loading control.

Hypothalamic Neuropeptide Expression.

Measurements of mRNA levels were carried out by quantitative RT-PCR on RNA extracted from dissected hypothalamic or pituitary tissue. Total RNA for each hypothalamus was quantified by spectrophotometry after purification with the Qiagen RNeasy kit. Two hundred nanograms of each total RNA sample was reverse-transcribed and PCR-amplified by using the TaqMan-principles ABI PRISM 7700 sequence detection system (Applied Biosystems). Efficiency for the primers was estimated from standard curves made with serial cDNA dilutions.

Analytical Procedures.

Blood glucose levels and insulin and leptin serum levels were determined as described in ref. 30. IL-6 serum levels were measured by ELISA with mouse standards according to the manufacturer’s guidelines (BD OptEIA mouse IL-6 ELISA kit; BD Biosciences Pharmingen). Corticosterone serum levels were determined by RIA using mouse standards according to the manufacturer’s guidelines (ImmuChem Double Antibody Corticosterone ¹²⁵I RIA kit; ICN).

Statistical Methods.

Data were analyzed for statistical significance by using a two-tailed unpaired Student t test.

i.c.v. Cannula Implantation.

Twelve-week-old mice were anesthetized by i.p. injection of Avertin (240 mg/kg) (2,2,2-tribromoethanol; Sigma) and placed in a stereotactic device. A sterile osmotic pump connector cannula (Bilaney Consultants, Düsseldorf, Germany) was implanted into the lateral brain ventricle (0.2 mm posterior and 1.0 mm lateral relative to bregma and 2.3 mm below the surface of the skull). The support plate of the catheter was attached to the skull with superglue. The catheter was prefilled with 0.9% NaCl and connected to a sealed microrenathane catheter (MRE-025 from Braintree Scientific, Braintree, MA).

i.c.v. Injection.

After 7 days of recovery, 2 μl of recombinant rat CNTF (0.5 mg/ml; 557-NT/CF from R & D Systems) or 0.9% NaCl was injected, and 1 μl of 0.9% NaCl was postinjected to ensure a complete drug administration to the brain. To avoid a backflow, the catheter was sealed. Injections were performed under isoflurane anesthesia 1 h before the onset of the dark cycle.

Food intake and body weight were measured 4 and 24 h after the injection for females and 6 h after the injection for males. The correct position of the i.c.v. cannula was verified by injection of methyl blue after the mice were killed.

Tissue Preparation for Immunocytochemistry and Immunocytochemical Procedures.

Control and gp130^ΔPOMC mice were mated with RosaArte26 reporter mice (29). At the age of 10 weeks, mice fed ad libitum received one i.v. injection of 5 μg of recombinant rat CNTF, 20 μg of leptin (L3772 from Sigma), or 0.9% NaCl into the tail vein. Thirty minutes after injection, the mice were perfused with 0.9% NaCl followed by 4% paraformaldehyde, and the brains were extracted, kept in 4% paraformaldehyde overnight at 4°C, and stored in 20% sucrose at 4°C.

For one series, we performed double-labeling of p-STAT3 and β-galactosidase (indicating POMC-expressing neurons) using free-floating immunohistochemistry as described in ref. 32. In brief, sections were washed, pretreated with 100% methanol/0.3% glycine/0.03% SDS, and blocked in 3% normal donkey serum followed by incubation of the primary antibodies [rabbit anti-p-STAT3 (1:3,000 dilution; Cell Signaling Technology, Beverly, MA) and goat anti-β-galactosidase (1:3,000 dilution; Biogenesis, Poole, U.K.)] for 48 h at 4°C. Secondary labeling was done with anti-rabbit Alexa 488 (Invitrogen/Molecular Probes) and biotinylated anti-goat antibody (1:200; Jackson ImmunoResearch) followed by streptavidin-Alexa 568 (1:200) for detection of p-STAT3 and β-galactosidase antibodies. Sections were mounted onto gelatin-coated slides, dried, and coverslipped by using mounting media (ProLong Gold; Invitrogen/Molecular Probes). Fluorescence signals were detected under a fluorescent microscope (BX51; Olympus), and representative pictures of the ARC were taken with a digital color camera (DP70; Olympus).

Fot better visualization of GFP expression after Cre-recombination, brain sections (25 μm) of double-heterozygous PomcCre Z/EG mice were washed, pretreated with 0.3% H₂O₂, blocked with PBT-Azide containing 3% donkey serum, and incubated overnight with primary antibody (anti-GFP rabbit serum, 1:10,000; A6455 from Invitrogen/Molecular Probes). Incubation with secondary antibody (anti-rabbit IgG, 1:500; 711-065-125 from Jackson ImmunoResearch) was followed by incubation with the VECTASTAIN Elite ABC kit (Vector Laboratories) for 1 h and 0.4% DAB/0.01% H₂O₂. Afterward the sections were mounted onto gelatin-coated slides and covered with glycerin.

Pituitary was stained with the same antibodies, using the Tyramide Signal Amplification kit (TSA Plus Fluorescence Systems; PerkinElmer).

POMC Cell Counting.

Quantification of β-galactosidase-positive POMC cells was done as described in ref. 32. In brief, every fourth section throughout the ARC was taken and allocated rostral to caudal to visualize the distribution of POMC neurons throughout the ARC. By using Adobe photoshop software, β-galactosidase-positive neurons were counted and marked digitally to prevent multiple counts. Cell counts were performed on three animals per group. The percentage of double-positive cells compared with the total amount of β-galactosidase-positive POMC cells after CNTF stimulation was also determined by cell counting.

Immunostaining for c-Fos and Quantification.

As described above, catheter-implanted animals received an injection of either 0.9% NaCl or CNTF 1 h before onset of the dark phase. Six hours later, the mice were anesthetized by using Avertin and perfused with 0.9% NaCl followed by Somogyi–Takagi fixative. The brains were dissected, kept in glutaraldehyde-free fixative at 4°C for at least 2 h, and stored in 0.1 M phosphate buffer at 4°C until further preparation.

Free-floating brain sections (50 μm) containing the PVN were immunostained for c-Fos [rabbit α-c-Fos, Oncogene (EMD Bioscience) used at 1:20,000] by our standard protocol using diaminobenzidine as the chromagen as described in ref. 33. Sections were examined under a Zeiss Axioplan 2 microscope outfitted with an AxioCam HRc camera and axiovision 4.2 imaging software. Every section containing the PVN (n = 4) was examined for each animal (with the operator blinded to the experimental conditions). For each hemisphere of the PVN, at a magnification of ×40, a single plane of focus was chosen and a black and white image was captured; a 100-μm² box was placed in the center of the PVN. All of the c-Fos-labeled cell nuclei were counted within the box for each section. Only those labeled nuclei that were clearly in the plane of focus were selected. The results were expressed as the number of c-Fos-labeled cells per 100-μm² area of the PVN.