Tissue preparation

Male Pomc tau-lacZ^+/−, ob/ob, and wild-type mice were terminally anesthetized with pentobarbitone (50 mg/kg of body weight, ip) and perfused transcardially with heparinized saline, followed by 4% paraformaldehyde or 10% neutral buffered formalin (Sigma, St. Louis, MO). Brains were removed, placed in 15% sucrose/fixative solution, and immersed in 30% sucrose/PBS at 4 C. Twenty-five-micrometer coronal sections were cut using a freezing sliding microtome.

Dual-label ³⁵S-POMC and X-gal

Pomc tau-lacZ^+/− mice are expected to express the β-galactosidase gene under the Pomc promoter in sites of endogenous POMC expression (15). Histochemical validation of exclusive expression of β-galactosidase activity in endogenous POMC-containing neurons was performed in Pomc tau-lacZ^+/− mice and control^+/+ littermates. First, free-floating sections were incubated with X-gal staining buffer (15) in darkness at 37 C for 12 h. Sections were then mounted onto slides and air dried before processing for ³⁵S-POMC in situ hybridization histochemistry (ISHH).

The protocol for ISHH used was a modification of that previously reported (16,17). Briefly, an antisense POMC ³⁵S-labeled riboprobe was generated from cDNA template. The linearized plasmid was subjected to in vitro transcription with SP6 polymerase according to the manufacturer’s protocol (Promega, Madison, WI). The ³⁵S-POMC probe was diluted to 2 × 10⁷ cpm/ml. Sections were incubated in the hybridization solution for 12–16 h at 56 C. Sections were next immersed in 0.002% RNase A (Roche Molecular Biochemicals, Indianapolis, IN) for 30 min. After stringency washes, slides were dipped in 3% parlodion (Fisher Scientific, Fair Lawn, NJ), air dried, dipped in photographic emulsion (NTB2; Kodak, Rochester, NY), and stored in light-tight boxes at 4 C for 1 wk. Sections were developed (D-19, Kodak) and fixed (Fixer, Kodak) and analyzed with an Axioskop 2 mot plus microscope (Zeiss, Thornwood, NY).

5-HT_2CR and POMC colocalization

To assess colocalization of 5-HT_2CR and POMC, dual-label immunohistochemistry (IHC) for β-galactosidase immunoreactivity (IR) and 5-HT_2CR-IR was performed in Pomc tau-lacZ^+/− and wild-type mice. Modified standard IHC procedures, as previously reported (18,19), were used. Briefly, tissue was washed with PBS and then blocked in 3% normal donkey serum (NDS) in 0.25% Triton X-100 in PBS (PBT) for 1 h. Tissue was then incubated with goat anti-5-HT_2CR (1:200; Santa Cruz Biotechnology, Santa Cruz, CA) and mouse anti-β-galactosidase (1:5000; Promega) antibodies in 3% NDS and PBT-azide (0.02% sodium azide in PBT) overnight at room temperature. The anti-5-HT_2CR antibody has previously been shown to be specific, with no background staining in 5-HT_2CR-deficient mice (20). After PBS washes, sections were incubated for 1 h with biotinylated donkey antimouse serum (1:1000; Jackson Laboratories) in 3% NDS and PBT. Sections were washed and then incubated for 1 h with appropriate fluorophores, Alexa Fluor 594 conjugated to streptavidin (1:1000; Molecular Probes, Eugene, OR) and Alexa Fluor 488 conjugated to donkey-antigoat IgG (1:500; Molecular Probes) in 3% NDS and PBT. Finally, after PBS washes, sections were mounted onto slides, air dried, and coverslipped with mounting medium (Vectashield; Vector Laboratories, Burlingame, CA).

Effect of BVT.X on acute food intake, locomotor activity, and oxygen consumption (VO₂)

The effect of saline and BVT.X on 6 h dark-cycle HFD paste (58.4 kcal percent fat) intake in murine models of obesity, ob/ob and DIO mice, was assessed. HFD paste was provided in glass jars in the home cage. DIO mice were maintained on HFD after weaning, whereas ob/ob mice were maintained on a chow diet until 3 d before the initiation of the study, at which time they were switched to the HFD. Forty-five minutes before the onset of the dark cycle, the HFD was removed from the home cage and saline or BVT.X (20 or 60 mg/kg) was administered by ip injection. At the onset of the dark cycle, fresh preweighed HFD was returned to the home cage, and intake was recorded for the next 6 h. These studies were performed using a within-subject experimental design (i.e. each animal was assessed with all treatments), such that animals received one of the three treatments on alternating days, with a minimum of 3 d elapsing between experimental treatments.

The effect of saline and BVT.X on 6-h, dark-cycle, powdered laboratory rodent chow (Purina) intake in young Mc4r null and wild-type littermates was also examined. Mice were recently weaned and maintained on the chow diet provided in the Comprehensive Lab Animal Monitoring System (CLAMS; Columbus Instruments, Columbus, OH). Mice were housed in the CLAMS chambers 3 d before the initiation of the study. CLAMS chambers are rectangular 1-liter containers with a two-level photobeam array for measuring locomotor activity. Powdered chow diet is provided under a spring-loaded grate that is fixed to a balance. VO₂ is measured using indirect calorimetry. All data are collected and stored electronically in a time-stamped data file. While mice were acclimating to the chambers, food intake and body weight were measured and compared with home cage measurements to ensure stable feeding patterns. To assess the effect of BVT.X on ingestive behavior, food was removed 45 min before the onset of the dark cycle, and mice were injected with saline or BVT.X (60 mg/kg, ip). At the onset of the dark cycle, food was returned, and 6-h food intake, locomotor activity, and VO₂ were measured. This study was performed using a within-subject experimental design, such that each animal received saline or BVT.X on different days, with 2 d elapsing between the experimental treatments.

Effect of prolonged BVT.X infusion on POMC mRNA, body weight, percent body fat, and food intake

To assess the prolonged effect of a selective 5-HT_2CR agonist on POMC mRNA, body weight, percent body fat, and food intake, ob/ob mice were treated with saline or BVT.X (60 mg/kg·d) via sc osmotic minipump for 7 d. Body weight and chow pellet intake were recorded daily. Immediately before pump implantation (d 0) and after 7 d of treatment (d 7), percent body fat was assessed with dual-energy x-ray absorptiometry (DEXA; Lunar PIXImus, MEC Lunar Corp., Minster, OH). After the DEXA scan on d 7, mice were perfused with fixative, and brain tissue was collected and prepared as described above. Arcuate nucleus of the hypothalamus (ARC) POMC mRNA was determined using densitometric quantification of ³⁵S-labeled POMC with ISHH. ISHH was performed as outlined above, and then slides were placed in x-ray film cassettes with BMR-2 film (Kodak). The film was developed 24 h later. Comparisons of POMC mRNA after saline or BVT.X treatment were made by assessing the autoradiographic ARC (coordinates from Bregma, −1.34 to −2.70 mm) ³⁵S-labled POMC signal as measured with a light box, a digital camera interface, and Scion Image software (Frederick, MD). Determination of the level of bregma for each section of brain tissue on the film was made by examining the sections on the slides counterstained with thionin on a Zeiss Axioskop 2 mot plus microscope. At comparable levels of the rostral to caudal ARC for each mouse, the ³⁵S-POMC signal within each section was analyzed by computing the mean density minus background using Scion Image software.

Drug

The selective 5-HT_2CR agonist used was BVT.X [K_i (nm): 5-HT_2CR, 9; 5-HT_2AR, > 1000; 5-HT_2BR, > 1000; 5-HT_1AR, > 800; 5-HT_1BR, > 1000; other 5-HT receptors, not active] which was kindly provided by Biovitrum (Stockholm, Sweden). BVT.X was dissolved in 0.9% pyrogen-free saline.

Coexpression of 5-HT_2CR and POMC

Whereas the general expression of 5-HT_2CR mRNA in the rodent has been reported (21), the chemical phenotype of 5-HT_2CR expressing neurons in satiety centers within the mouse brain has not been described. We hypothesized that 5-HT_2CRs are coexpressed with POMC neurons synthesizing the anorectic endogenous melanocortin agonist α-MSH in the ARC, as is observed in the rat (17). To investigate this hypothesis, we used Pomc tau-lacZ^+/− mice to facilitate the identification of POMC-expressing cells. We first confirmed the exclusive expression of β-galactosidase activity in endogenous POMC-containing neurons in these mice but not in their wild-type littermates using dual labeling with X-gal and ³⁵S-POMC mRNA (n = 10). As expected, no β-galactosidase activity was found in wild-type mice, but β-galactosidase activity was evident in Pomc tau-lacZ^+/− mice (Fig. 2​2,, A–D). Dual-label analyses revealed that approximately 100% of X-gal-stained neurons contained ³⁵S-POMC mRNA. These data illustrate that Pomc tau-lacZ^+/− mice may be used as a tool to visualize cells expressing endogenous POMC mRNA.

Open in a separate window

Figure 2

5-HT_2CRs are coexpressed with POMC in the ARC. A, Wild-type (+/+) mice exhibit endogenous POMC mRNA using ISHH with ³⁵S-POMC (identified by clusters of black grains) but no β-galactosidase activity (X-gal stain of blue cytoplasm) in the ARC. B, Pomc tau-lacZ^+/− mice display ³⁵S-POMC in every neuron positive for X-gal. C, Higher level magnification of wild-type^+/+ mouse ARC, showing ³⁵S-POMC grains but no X-gal stain. D, Higher-level magnification of Pomc tau-lacZ^+/− mouse ARC, showing a typical cell containing both ³⁵S-POMC grains and X-gal stain (arrow). E–F, Fluorescent IHC for 5-HT_2CR (E) and β-galactosidase (F) in Pomc tau-lacZ^+/− mouse ARC. G, Merge of E and F, showing cells coexpressing 5-HT_2CR and β-galactosidase (arrows). Scale bar (B), 200 μm and applies to A; (D), 20 μm and applies to C; (G), 20 μm and applies to E–G.

Consistent with earlier reports in rats using ISHH (17,21), we observed that 5-HT_2CR protein is widely expressed in the wild-type mouse hypothalamus using IHC (n = 4). We next investigated the hypothesis that 5-HT_2CRs are positioned to affect melanocortin action in the mouse by performing dual-label IHC using antibodies directed against the 5-HT_2CR and β-galactosidase in Pomc tau-lacZ^+/− mice. 5-HT_2CR-IR neurons were abundantly coexpressed with β-galactosidase-IR cells in the ARC (Fig. 2​2,, E–G). Specifically, approximately 75% of β-galactosidase-IR neurons were 5-HT_2CR-IR positive, and 28% of 5-HT_2CR-IR neurons were β-galactosidase-IR positive. These findings indicate that 5-HT_2CRs are positioned to affect the activity of the majority of ARC POMC neurons in the mouse. Furthermore, they suggest that 5-HT_2CRs are also expressed in non-POMC cells in the ARC.

Effect of prolonged BVT.X infusion on POMC mRNA, body weight, percent body fat, and food intake

Obese ob/ob mice (n = 8, mean body weight 46 g) received saline or BVT.X (60 mg/kg·d) via sc osmotic minipump for 7 d. Using densitometric quantification of ARC POMC mRNA, we observed that BVT.X treatment significantly elevated POMC expression throughout the extent of the ARC (Fig. 3A​3A).). This BVT.X-induced increase in POMC mRNA was associated with a significant reduction in percent body fat (Fig. 3B​3B)) and body weight (Fig. 3C​3C)) on d 7, compared with pretreatment levels. Food intake was significantly reduced during the first 2 d of treatment but then normalized after that (Fig. 3D​3D).

Open in a separate window

Figure 3

Prolonged 5-HT_2CR agonist infusion increases POMC mRNA expression and reduces percent body fat, body weight, and initial food intake. Obese ob/ob mice (n = 8, mean body weight 46 g) were treated with saline or BVT.X (60 mg/kg·d for 7 d, sc minipump). A, Prolonged BVT.X treatment increased POMC mRNA in the ARC, compared with saline treatment, as determined by densitometry analysis after ISHH with a ³⁵S-labeled antisense POMC riboprobe. Prolonged BVT.X treatment significantly decreased percent body fat as determined by DEXA (B) and body weight (grams) (C), compared with pretreatment levels. D, BVT.X treatment decreased chow pellet intake on d 1 and 2, but no further significant differences were observed. Data are presented as mean ± sem. Significant differences are indicated as follows: *, P < 0.05; **, P < 0.01.

The authors are grateful to Dr. Loraine Tung for technical assistance and Dr. Anthony Coll for the generous gift of Pomc tau-lacZ ^+/− mice.