SEARCH

SEARCH BY CITATION

Keywords:

  • BDNF ;
  • CRH ;
  • feeding behavior;
  • histamine;
  • hypothalamus

Abstract

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Brain-derived neurotrophic factor (BDNF), corticotropin-releasing factor (CRF), and hypothalamic neuronal histamine are anorexigenic substances within the hypothalamus. This study examined the interactions among BDNF, CRF, and histamine during the regulation of feeding behavior in rodents. Food intake was measured after treatment with BDNF, α-fluoromethyl histidine (FMH; a specific suicide inhibitor of histidine decarboxylase that depletes hypothalamic neuronal histamine), or CRF antagonist. We measured food intake in wild-type mice and mice with targeted disruption of the histamine H1 receptor (H1KO mice) after central BDNF infusion. Furthermore, we investigated CRF content and histamine turnover in the hypothalamus after BDNF treatment, and conversely, BDNF content in the hypothalamus after histamine treatment. We used immunohistochemical staining for histamine H1 receptors (H1-R) in BDNF neurons. BDNF-induced feeding suppression was partially attenuated in rats pre-treated with FMH or a CRF antagonist, and in H1KO mice. BDNF treatment increased CRF content and histamine turnover in the hypothalamus. Histamine increased BDNF content in the hypothalamus. Immunohistochemical analysis revealed that H1-Rs were expressed on BDNF neurons in the ventromedial nucleus of the hypothalamus. These results indicate that CRF and hypothalamic neuronal histamine mediate the suppressive effects of BDNF on feeding behavior and body weight.

Abbreviations used
BBB

blood-brain barrier

BDNF

brain-derived neurotrophic factor

BST

bed nucleus of stria terminalis

CeA

central nucleus of the amygdala

CRF

corticotropin-releasing factor

CRH

corticotropin-releasing hormone

FMH

α-fluoromethyl histidine

HDC

histidine decarboxylase

LH

lateral hypothalamus

MCH

melanin concentrating hormone

PBS

phosphate buffered saline

PVN

paraventricular nucleus

t-MH

tele-methylhistamine

TMN

tuberomammillary nucleus

TrkB

tropomyosin-related kinase receptor type B

VMH

ventromedial nucleus

Brain-derived neurotrophic factor (BDNF) plays an important role in the central regulation of energy metabolism. Intracerebroventricular (i.v.t.) BDNF administration results in decreased food intake and body weight (Pelleymounter et al. 1995). In animals with conditional BDNF deletion, mutation, or in BDNF (+/−) heterozygous mice, hyperphagia and obesity are accompanied by significantly reduced BDNF gene expression in the hypothalamus, including the ventromedial nuclei (VMH; Lyons et al. 1999; Rios et al. 2001). Further, exogenous BDNF reverses the phenotype of these animals, suggesting that endogenous BDNF reduces feeding and body weight gain (Kernie et al. 2000). Tropomyosin-related kinase receptor type B (TrkB) is a high-affinity BDNF receptor that is widely expressed in the adult central nervous system, notably in several nuclei involved in energy balance, including the VMH (Yan et al. 1997). Mutant mice that express TrkB in the brain at approximately one-quarter the normal level exhibit hyperphagia and excessive weight gain on high-fat diets (Xu et al. 2003).

Neuronal histamine is one of anorexigenic substances produced mainly in the tuberomammillary nucleus (TMN) of the posterior hypothalamus, which has diffuse projections throughout the brain, including projections to the VMH and paraventricular nucleus (PVN), where the histamine H1-receptors (H1-R) are localized (Fukagawa et al. 1989). In addition, corticotropin-releasing factor (CRF), which is mainly synthesized in the PVN, directly activates histamine neurons (Gotoh et al. 2005). Local BDNF infusion into the PVN and VMH markedly decreased food intake and body weight (Wang et al. 2007a, b). These effects of locally infused BDNF were attenuated by local infusion of α-helical CRF into the PVN, suggesting that the PVN is targeted by the BDNF-evoked CRF pathway (Toriya et al. 2010). From these behavioral and neuroanatomical studies, we hypothesized that BDNF, CRF, and hypothalamic neuronal histamine constitute a neuronal network within the hypothalamus that regulates food intake.

We examined this hypothesis from several different perspectives. First, we investigated whether the anorectic effect of BDNF was modulated in rats pre-treated with FMH, a suicide inhibitor of histidine decarboxylase (HDC), or α-helical CRF9-41 (α-helical CRF), a competitive CRF receptor antagonist that blocks both CRF type 1 receptors (CRF1-R) and type 2 receptors (CRF2-R). We also examined the effect of BDNF treatment on food intake in H1-R knockout (H1KO) mice, which lack the H1-R for neuronal histamine. Then, we determined whether the BDNF treatment affected CRF content and histamine turnover in the hypothalamus. Finally, we investigated whether the central infusion of histamine activated BDNF neurons and whether the blockade of endogenous BDNF biological activity modulated the suppressive effect of histamine on food intake.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Animals

Male Sprague–Dawley rats weighting 250–280 g (Seac Yoshitomi, Fukuoka, Japan), male C57BL/6N mice weighting 25–30 g (Seac Yoshitomi), and male H1KO mice weighting 25–30 g (Kyushu University, Fukuoka, Japan) were used. They were housed in a room with daily illumination from 07:00 to 19:00 (12/12 h light/dark cycle) and maintained at 21 ± 1°C with 55 ± 5% humidity. H1-R knockout (H1KO) mice that lack the H1-R neuronal histamine receptor for backcrossing were maintained at Oita University (Yufu, Japan). Backcrossing H1-R−/− homozygous mice with the C57BL/6N strain (Kyudo, Fukuoka, Japan) for six generations resulted in incipient congenital N5 mice with two genotypes (H1-R−/− and H1-R+/+). All genotypes were confirmed by Southern blotting as described (Inoue et al. 1996). Animals were fed standard chow (Clea chow, Clea, Japan), allowed access to and tap water ad libitum, and handled for 5 min each on four successive days to habituate arousal levels before the experiment. On test days, all animals had recovered to at least their pre-experiment body weight. All experiments proceeded in accordance with the Oita University Guidelines, which are based on the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health. We used total 156 male Sprague–Dawley rats, 30 male C57BL/6N (wild type) mice and 30 H1KO mice.

Surgery

Rats were placed in a stereotaxic apparatus under intraperitoneal (i.p.) anesthesia with 100 mg/kg of sodium pentobarbital (Narishige, Tokyo, Japan). The implantation of a chronic cannula (a 23-gauge stainless steel; length, 15 mm) into the third cerebroventricle (3vt) was performed at least 10 days before starting infusions. A 30-gauge stainless steel wire stylet was inserted into the guide cannula to prevent cerebrospinal fluid (CSF) leakage and cannula obstruction. A chronic cannula was inserted into the 3vt along the midline, 6.0 mm anterior to the zero ear bar coordinate, to a depth of 7.8 mm from the cortical surface, into the VMH, 0.8 mm lateral to the midline and 2.3 mm posterior to the bregma at a depth of 8.5 mm from the cortical surface, or into the PVN, 0.8 mm lateral to the midline and 0.8 mm posterior from the bregma at a depth of 7.5 mm from the cortical surface, according to the atlas of Paxinos and Watson (1997).

In mice, at least 1 week before starting infusions, a cannula (a 29-gauge stainless steel) was similarly implanted into the lateral cerebroventricle (l.v.t.), 1.0 mm lateral to the midline and 0.22 mm posterior to the bregma at a depth of 2.0 mm from the cortical surface. The coordinates for the PVN were 0.15 mm lateral to the midline and 0.8 mm posterior to the bregma at a depth of 4.0 mm from the cortical surface, according to the atlas of Franklin and Paxinos (1997).

Reagents

On the day of the experiment, l-histamine (11.1 μg/μL), BDNF (0.1 μg/μL; Sigma, St. Louis, MO, USA), α-helical CRF9-41 (α-helical CRF, 1 μg/μL; Sigma) which is a competitive CRF receptor antagonist that blocks both CRF types 1 (CRF1-R) and 2 (CRF2-R) receptors, anti-BDNF antibody (0.1 μg/μL; Millipore, Billerica, MA, USA), pargyline hydrochloride (1 mmol/mL; Sigma), and α-fluoromethylhistidine (FMH; Sigma), an inhibitor of HDC (100 μg/μL; Research Biochemical International, Natick, MA, USA) were freshly dissolved in phosphate buffered saline (PBS). The pH of each solution was adjusted to 6.5–7.5.

Experimental protocol

Overall, 66 rats were assigned to one of eleven different groups (n = 6 in each group) as follows: Group 1: PBS (10 μL/10 min, 3vt) + PBS (10 μL/10 min, 3vt), Group 2: FMH (1 mg/10 μL/10 min, 3vt) + PBS (10 μL/10 min, 3vt), Group 3, 4: PBS (10 μL/10 min, 3vt) + BDNF (1 μg/10 μL/10 min, 3vt or i.p.), Group 5, 6; FMH (1 mg/10 μL/10 min, 3vt) + BDNF (1 μg/10 μL/10 min, 3vt or i.p.), Group 7: α-helical CRF (10 μg/10 μL/10 min, 3vt) + PBS (10 μL/10 min, 3vt), Group 8: α-helical CRF (10 μg/10 μL/10 min, 3vt) + BDNF (1 μg/10 μL/10 min, 3vt), Group 9: anti-BDNF antibody (1 μg/10 μL/10 min, 3vt) + PBS (10 μL/10 min, 3vt), Group 10: PBS (10 μL/10 min, 3vt) + histamine (111 μg/10 μL/10 min, 3vt), and Group 11: anti-BDNF antibody (1 μg/10 μL/10 min, 3vt) + histamine (111 μg/10 μL/10 min, 3vt). Food was withheld for 24 h before the experiment. They were pre-treated with a central administration of FMH, α-helical CRF, anti-BDNF antibody, or PBS at two hour before injection of BDNF treatment (3vt or i.p.), histamine (3vt), or PBS (3vt). The dose of FMH was determined by referring the previous study and we confirmed this dose was enough to suppress the HDC activity without poisoning (Doi et al. 1994).

Measuring food intake in rats and mice

Food was withheld for 24 h before the experiment. Cumulative food intake for a 24 h period was measured after each treatment described above.

BDNF-regulated changes in CRF and tele-methylhistamine contents, and histamine-regulated changes in hypothalamic and plasma BDNF levels

In this study, 36 rats were divided into six groups (= 6 per group). At the start of the experiment, the drugs were administered as follows: Group 1, 2: PBS (10 μL/10 min, 3vt) + PBS (10 μL/10 min, 3vt or i.p.), Group 3: α helical CRF (10 μg/10 μL/10 min, 3vt) + PBS (10 μL/10 min, 3vt), Group 4, 5; PBS (10 μL/10 min, 3vt) + BDNF (1 μg/10 μL/10 min, 3vt or i.p.), and Group 6; α helical CRF (10 μg/10 μL/10 min, 3vt) + BDNF (1 μg/10 μL/10 min, 3vt). They were pre-treated with a central administration of α-helical CRF or PBS at two hour before injection of BDNF treatment (3vt or i.p.) or PBS (3vt or i.p.). Furthermore, 24 rats were divided into four groups as follows, intrahypothalamic injection of BDNF (0.1 μg/1 μL/1 min) or PBS (1 μL/1 min) in the PVN, intrahypothalamic injection of histamine (11.1 μg/1 μL/1 min), or PBS (1 μL/1 min) in the VMH (n = 6, in each).

All rats were fasted for 24 h before the experiment. On the day of the experiment, each group was pre-treated by i.p. administration of pargyline hydrochloride (0.33 mmol/kg), which inhibits monoamine oxidase B, and induces the extraneuronal accumulation of tele-methylhistamine (t-MH), a major metabolite of released neuronal histamine, 2 h prior to the start of the experiment. All rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) 2 h after above treatment and then exsanguinated following transcardiac perfusion with 100 mL saline containing 200 U heparin. After decapitation, brains were removed to measure t-MH, CRF and BDNF contents. Brain regions that contained hypothalamic sites [the lateral hypothalamus (LH), PVN, VMH, and TMN], and extrahypothalamic sites [central nucleus of the amygdala (CeA) and bed nucleus of stria terminalis (BST)] were dissected with a frozen razorblade, at the appropriate levels, based on the rat brain map by Paxinos and Watson (1997). The dissected tissue was homogenized in 400 μL of 0.5 M acetic acid, boiled for 10 min, and then protein was assayed (Bio-Rad, Hercules, CA, USA) in 50 μL aliquots. Finally, blood from all rats was obtained by cardiac puncture and centrifuged at 4000 g for 10 min at 4°C. Plasma was immediately frozen and stored at −80°C until analyzed. CRF content in each dissected section was measured using a CRF ELISA kit (Yanaihara Co., Shizuoka, Japan). BDNF levels in plasma and each specific section of the brain were measured with a BDNF ELISA kit (Insight Genomics, Falls Church, VA, USA). t-MH contents were assayed via HPLC, using the deproteinized supernatants containing the amine extracts. The details of amine assays have been described elsewhere (Sakata et al. 1981).

BDNF-regulated changes in food intake and BDNF content in H1-R knockout (H1KO) mouse hypothalamus

We used 30 H1KO and 30 wild-type mice. 24 H1KO mice and 24 wild-type mice were divided into four groups (n = 6 in each), respectively. In Group 1: l.v.t. administration of BDNF (0.1 μg/1 μL/1 min). In Group 2: l.v.t. administration of PBS (1 μL/1 min). In Group 3: Intrahypothalamic administration of BDNF in the PVN (0.1 μg/1 μL/1 min). In Group 4: Intrahypothalamic administration of PBS in the PVN (1 μL/1 min). Food was withheld for 24 h before the experiment. Then food intake was measured over a 24 h period following the treatment. Hypothalamic tissue blocks of H1KO (n = 6) and wild-type (n = 6) mice were dissected after 24-h deprivation and homogenized as mentioned above. Hypothalamic BDNF contents were measured using a BDNF ELISA kit (Insight Genomics).

Immunohistochemistry

In this study, 30 rats were anesthetized using 100 mg/kg sodium pentobarbital i.p., and killed by transcardiac perfusion with 50 mL PBS that contained 50 U heparin, followed by 50 mL 4% p-formaldehyde in PBS. The brains were removed and divided into forebrain, diencephalon, and brainstem segments. The specimens were rapidly frozen at −80°C and sectioned at a thickness of 40 μm using a cryostat at −20°C to stain for BDNF, H1-R, (n = 6) and BDNF or H1-R negative control (n = 6). The sections were incubated overnight at 4°C with polyclonal rabbit antiserum against rat HDC (1 : 2000; Chemicon Inc., Temecula, CA, USA), followed by biotin-conjugated goat anti-rabbit IgG and FITC-conjugated streptavidin (ABC reagent; Vector Laboratories, Burlingame, CA, USA). Next, they were incubated with a specific polyclonal rabbit antiserum against full-length (gp145TrkB, specific for rat, 1 : 100; Novus Biologicals, Litteton, CO, USA) and polyclonal rabbit antiserum against the truncated (gp95TrkB, specific for rat, 1 : 100; Abcam Biochemicals, Cambridge, MA, USA) isoforms of TrkB in 0.3% Triton X-100 containing 1% normal rabbit serum. Biotin-conjugated secondary IgG antibodies (ABC reagent; Vector Laboratories) were added, followed by rhodamine-conjugated streptavidin (ABC reagent; Vector Laboratories).

Other sections were incubated overnight at 4°C with a specific polyclonal sheep antiserum against rat BDNF (1 : 2000; Millipore) and a specific polyclonal rabbit antiserum against rat H1-R (1 : 1000; Millipore). The sections were then incubated with biotin-conjugated secondary IgG antibody (ABC reagent; Vector Laboratories), FITC-conjugated streptavidin for H1-R and rhodamine-conjugated streptavidin (ABC reagent; Vector Laboratories) for BDNF. The negative controls in each experiment were incubated with normal serum instead of the primary antibody (BDNF or H1-R), followed by both secondary antibodies. We determined the specificity of the BDNF and H1-R antisera, after incubating each with recombinant BDNF (Sigma) or H1-R (Gene Tex Inc., Irvine, CA, USA) protein. Increasing amounts of protein were added to a fixed concentration of BDNF and H1-R antiserum at a 1 : 1, 1 : 5, and 1 : 10 molar ratio to the concentration of each recombinant protein (n = 6 in each). Stained sections were analyzed using a confocal immunofluorescence microscope (Olympus, Tokyo, Japan) and imaging software (Lumina Vision; Mitsutani Corp., Tokyo, Japan). The resultant digital images of the same section were not adjusted or altered, except for an occasional change in brightness. BDNF or H1-R immunoreactivity was determined by drawing areas of VMH in hypothalami and applying an optical density scale using automated counting tools from the imaging software (Lumina Vision; Mitsutani Corp.). Every fifth section of serial hypothalamus samples from six rats (seven sections per rat) was analyzed.

Western blotting analysis

We examined the molecular weight of proteins recognized by the gp145TrkB, gp95TrkB, BDNF, and H1-R antibodies by Western blotting with solutions containing baculovirus system-derived TrkB lysate (including gp145TrkB and gp95TrkB protein; Novus Biologicals), human BDNF (Sigma), and H1-R (Gene Tex Inc.) protein. These proteins were diluted to final concentrations in sample buffer (50 mM Tris–HCl, pH 8.0, 10% glycerol, 1 mM EDTA, 2 mM dithiothreitol, and 1 mM o-vanadate), heated at 94°C for 4 min and then separated by centrifugation (12 000 g) for 5 min. The supernatants were resolved by 7.5% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and electrophoretically transferred onto polyvinylidene difluoride membranes. After blocking non-specific binding with 0.5% non-fat milk, the membranes were incubated overnight at 4°C with the same antiserum against gp145TrkB, gp95TrkB, BDNF, and H1-R used in the immunohistochemistry experiments. The membrane was repeatedly washed and then incubated with IgG secondary antibodies. Immunopositive bands were visualized on Hyperfilm (GE Healthcare Bioscience, Piscataway, NJ, USA) using enhanced chemiluminescence.

Statistics

The data are expressed as mean ± SEM. Statistical tests included the two-tailed Student's t-test and two-way analysis of variance (anova) followed by Scheffe's test for post hoc comparisons. Differences in t-MH and CRH contents between PBS and BDNF groups, plasma and hypothalamic BDNF levels between PBS and histamine group, and hypothalamic BDNF levels between wild-type and H1KO mice were analyzed using Student's t-test. For comparison of data among more than three groups, anova was performed. For all tests, the level of significance was set at < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Effect of FMH pre-treatment on BDNF-induced food intake suppression and effect of BDNF on t-MH content in rat hypothalamus

3vt administration of BDNF significantly decreased daily food intake compared to the PBS treatment, and FMH pre-treatment partially attenuated BDNF-induced food intake suppression (F(3, 20) = 18.1, < 0.05; Fig. 1a). FMH administration alone did not affect the cumulative food intake compared to the PBS administration (Fig. 1a). BDNF treatment also increased the pargyline-induced accumulation of t-MH as compared to the PBS treatment in the PVN, VMH, and TMN, but not LH, CeA, and BST (n = 6 per group, < 0.05; Fig. 1b). However, ip administration of BDNF had no effect on appetite (Fig. 1c) and the t-MH accumulation in all discrete regions (Fig. 1d). Figure 1e depicts change in body weight over 24 h following each administration. BDNF treatment decreased body weight compared to PBS treatment and FMH pre-treatment diminished BDNF-induced reduction of body weight although FMH administration alone did not influence body weight (F(3, 20) = 50.43, n = 6 per group, < 0.05; Fig. 1e). Moreover, there was no significant difference in plasma BDNF levels among PBS (3vt and i.p.) and BDNF (3vt and i.p.) (Fig. 1f). Moreover, the intrahypothalamic injection of BDNF in the PVN also increased t-MH contents in the PVN, VMH, and TMN, compared to injection of PBS (n = 6 per group, < 0.05; Fig. 1g).

image

Figure 1. (a and c) Daily food intake after administration [3vt (a), i.p. (c)] of phosphate buffered saline (PBS), α-fluoromethyl histidine (FMH), brain-derived neurotrophic factor (BDNF), or FMH followed by BDNF. (b and d) Tele-methylhistamine (t-MH) content in each region after infusion of PBS or BDNF [3vt (b), i.p. (d)]. (e) Mean of the percentage body weight gained or lost after 24 h of each infusion relative to the weight of each group before infusion. (f) Plasma BDNF levels after 3vt or i.p. administration of BDNF. (g) t-MH content in each region after intrahypothalamic infusion of PBS or BDNF in paraventricular nucleus (PVN). *< 0.05 versus PBS and FMH, #< 0.05 versus BDNF.

Download figure to PowerPoint

Immunohistochemical staining for gp145TrkB and gp95TrkB in histaminergic neurons in rats, and effect of α-helical CRF pre-treatment on BDNF-induced food intake suppression and effect of BDNF infusion on CRF and t-MH content in rat hypothalamus

Figure 2a and b show regions of TMN and VMH we selected to investigate whether TrkB, BDNF receptor, is expressed on histamine neurons in the TMN area and/or localized in the VMH area, using Paxinos and Watson atlas (Paxinos and Watson 1997). Immunohistochemical analysis of double labeling demonstrated that no histamine neurons were co-localized with gp145TrkB or gp95TrkB in the TMN (Fig. 2c). In contrast, several neurons in the VMH expressed gp145TrkB and gp95TrkB (Fig. 2d). To confirm the specificity of the gp145TrkB and gp95TrkB antibodies used in the immunohistochemistry experiments, Western blot analysis was performed. We detected an immunopositive band at the predicted molecular weight of gp145TrkB (∼ 140 kDa) and gp95TrkB (∼ 90 kDa) proteins, suggesting that the two primary antibodies were indeed capable of recognizing gp145TrkB and gp95TrkB (data not shown).

image

Figure 2. (a and b) Schema indicating the area of tuberomammillary nucleus (TMN) (a) and ventromedial nucleus (VMH) (b) in the hypothalamus. (c) Immunohistochemical staining for gp145TrkB and gp95TrkB (red) in histaminergic neurons (green; yellow arrows) of the TMN. (d) Immunohistochemical staining for gp145TrkB and gp95TrkB (red; yellow arrows) in the VMH. Scale bar, 10 μm. (e) Tele-methylhistamine (t-MH) content in each region after 3vt administration of phosphate buffered saline (PBS), α-helical corticotropin-releasing factor (CRF), brain-derived neurotrophic factor (BDNF), or α-helical CRF followed by BDNF. *< 0.05 versus PBS and α-helical CRF, #< 0.05 versus BDNF.

Download figure to PowerPoint

3vt administration of BDNF resulted in significant reduction of daily food intake by about 40% (32.8 ± 2.9 vs. 20.2 ± 2.1 g, F(3, 20) = 127.1, n = 6 per group, < 0.05), compared to PBS group. The BDNF-induced suppression of food intake was also attenuated by pre-treatment with α-helical CRF (26.3 ± 1.8 vs. 20.2 ± 2.1 g, n = 6 per group, < 0.05); α-helical CRF treatment alone did not alter feeding compared to the PBS group (32.4 ± 2.1 vs. 32.8 ± 2.9 g). In addition, BDNF reduced body weight significantly, compared to PBS group (−0.96 ± 0.05 vs. +1.22 ± 0.01%, F(3, 20) = 46.35, n = 6 per group, < 0.05) and the pre-treatment of α-helical CRF also attenuated BDNF-induced decrease of body weight (+0.03 ± 0.01 vs. −0.96 ± 0.05%, n = 6 per group, < 0.05) although α-helical CRF administration alone did not affect body weight (+1.31 ± 0.47 vs. +1.22 ± 0.01%). BDNF treatment significantly increased CRF contents only in the PVN area compared to the PBS treatment (4.35 ± 0.42 vs. 2.83 ± 0.31 pmol/μg protein, F(3, 20) = 17.39, n = 6 per group, < 0.05). BDNF treatment also increased pargyline-induced t-MH accumulation and α-helical CRF pre-treatment attenuated this effect in the PVN (F(3, 20) = 19.89, n = 6 per group, < 0.05), VMH (F(3, 20) = 13.46, < 0.05) and TMN (F(3, 20) = 12.33, < 0.05), but not LH, CeA, and BST (Fig. 2e).

Attenuation of the anorexic effects of BDNF in H1KO mice

Lvt treatment of BDNF led to a significant reduction in food intake in wild-type mice (approximately, 40% reduction in 24 h), whereas the same treatment had a significantly lesser effect in H1KO mice (approximately, 20% reduction in 24 h, F(3, 20) = 29.53, n = 6 per group, < 0.05; Fig. 3a). Figure 3b depicts changes in body weight over 24 h following administration of PBS and BDNF to wild-type and H1KO mice. BDNF-induced reduction of body weight was also attenuated in H1KO mice (F(3, 20) = 26.52, n = 6 per group, < 0.05, Fig. 3b). To confirm whether site-specific infusion of BDNF induces a more robust anorexigenic effect in the H1KO mice, BDNF was infused into specific PVN. Placement of the canula was compared to the atlas of Franklin and Paxinos (1997). Figure 3c shows a typical BDNF-infused site in mice. It was verified that the tip of guide cannula was just into the PVN. The insertion of the thin infusion cannula did not cause excessive tissue destruction. Microinfusion of BDNF into the PVN also decreased food intake (F(3, 20) = 23.86, n = 6 per group, < 0.05, Fig. 3d) and body weight (F(3, 20) = 25.16, n = 6 per group, < 0.05, Fig. 3e), compared with PBS infusion in wild-type mice. These BDNF-induced reductions of feeding and body weight were attenuated in H1KO mice (n = 6 per group, < 0.05, Fig. 3d and e).

image

Figure 3. (a) Daily food intake after l.v.t. administration of phosphate buffered saline (PBS) or brain-derived neurotrophic factor (BDNF) in wild-type and H1KO mice. (b) Mean of the percentage body weight gained or lost after 24 h of each infusion relative to the weight of each group before infusion. *< 0.05 versus PBS (wild-type and H1KO mice) and #< 0.05 versus BDNF (wild type). (c) Schema indicating the area of paraventricular nucleus (PVN) in the hypothalamus and histological identification of cannula tips in a representative infusion into PVN. Scale bar, 50 μm. 3V, third ventricle. (d) Daily food intake after intrahypothalamic injection of PBS or BDNF into the PVN in wild-type and H1KO mice. (e) Mean of the percentage body weight gained or lost after 24 h of each infusion relative to the weight of each group before infusion. *< 0.05 versus PBS (wild-type and H1KO mice) and #< 0.05 versus BDNF (wild type).

Download figure to PowerPoint

Effect of histamine infusion on food intake and BDNF content in rat hypothalamus

The histamine-induced inhibition of food intake was attenuated by pre-treatment with anti-BDNF antibody (F(3, 20) = 152.36, < 0.05, Fig. 4a). Histamine treatment reduced body weight compared to PBS treatment and pre-treatment of anti-BDNF antibody attenuated histamine-induced decrease of body weight although the administration of anti-BDNF antibody alone did not influence body weight (F(3, 20) = 50.43, n = 6 per group, < 0.05; Fig 4b). Histamine treatment significantly increased the BDNF content in the PVN and VMH, but not in the LH, TMN, CeA and BST, compared to PBS treatment (n = 6 per group, < 0.05; Fig. 4c). There was no difference in plasma BDNF levels between histamine and PBS treatments (Fig. 4d).

image

Figure 4. (a) Daily food intake after 3vt administration of phosphate buffered saline (PBS), anti-brain-derived neurotrophic factor (BDNF) antibody, histamine, or anti-BDNF antibody followed by histamine. *< 0.05 versus PBS and A-BDNF, #< 0.05 versus histamine. (b) Mean of the percentage body weight gained or lost after 24 h of each infusion relative to the weight of each group before infusion. (c) BDNF content in each region after 3vt infusion of PBS or histamine. *< 0.05 versus PBS. (d) Plasma BDNF levels after 3vt administration of PBS or histamine. (e) BDNF content in the hypothalamus in wild-type and H1KO mice. *< 0.05 versus wild-type mice. A-BDNF; anti-BDNF.

Download figure to PowerPoint

Differential hypothalamic BDNF content between wild-type and H1KO mice

BDNF content in the hypothalamus was significantly decreased in H1KO mice compared to wild-type mice (1.9 ± 0.21 vs. 1.45 ± 0.22 ng/mg protein; n = 6 per group, < 0.05; Fig. 4e).

Immunohistochemical staining for BDNF and H1-R in the VMH area in rats

Immunohistochemical analysis with double labeling demonstrated that the majority of BDNF neurons expressed H1-R in the VMH (Fig. 5a). To further confirm the specificity of BDNF and H1-R antibodies, we performed immunohistochemistry after pre-absorbing the antibodies with purified BDNF and H1-R protein. When BDNF and H1-R antibodies were pre-absorbed with BDNF and H1-R peptides, respectively, antibody labeling of both was greatly diminished, and labeling specificity was controlled by omitting the BDNF and H1-R antiserum (data not shown). The results from the semi-quantitative analysis of the BDNF immunolabeling following pre-absorption with BDNF protein are shown in Fig. 5h and j. With increasing concentrations of the recombinant BDNF protein relative to primary antibody (1 : 1, 1 : 5, and 1 : 10 molar ratio of BDNF antibody concentration to recombinant protein concentration), the BDNF immunolabeling significantly decreased compared to immunolabeling with the primary antibody alone. The diminution of the anti-BDNF antibody signal with BDNF protein showed a linear reduction in staining up to five times the molar ratio of antigen to antibody (data not shown). Pre-absorption with the H1-R peptide yielded similar results. Moreover, we detected an immunopositive band at the predicted molecular weight of BDNF (∼ 18 kDa) and H1-R (∼ 57 kDa) proteins, indicating that the two primary antibodies were indeed capable of recognizing BDNF and H1-R (data not shown). To verify that histamine activates BDNF neurons directly, we performed VMH-specific injection of histamine. Microinfusion of histamine into the VMH increased BDNF contents in the PVN and VMH, compared with PBS infusion (n = 6 per group, < 0.05, Fig. 5b).

image

Figure 5. (a) Immunohistochemical staining for brain-derived neurotrophic factor (BDNF) and H1-R in the ventromedial nucleus (VMH) of the hypothalamus. H1-Rs (green) are expressed in BDNF neurons (red). (b) BDNF contents in each region after intrahypothalamic infusion of phosphate buffered saline (PBS) or BDNF in paraventricular nucleus (PVN). *< 0.05 versus PBS.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We investigated that the relationship among BDNF, CRF, and neuronal histamine on feeding behavior. Our results showed that the FMH-induced depletion of neuronal histamine partially weakened the reduction of food intake by BDNF. We suggest that HDC activity is important to regulate energy metabolism because our previous studies revealed the decrease of HDC activity in Zucker fatty (fa/fa) rats which have a missense mutation in the leptin receptor gene (Yoshimatsu et al. 1992; Sakata and Yoshimatsu 1997). In this study, we used FMH to investigate whether BDNF-induced anorexia is related to the activity of HDC. In addition, this reduced BDNF-induced response was also mimicked in H1KO mice, indicating that neuronal histamine and H1-R partially mediate the BDNF-induced hypophagia. We demonstrated that BDNF treatment elevated t-MH levels in the PVN, VMH, and TMN of hypothalamus. However, TrkB was not localized in histamine neurons, indicating that BDNF might stimulate histamine neurons indirectly. In addition, BDNF also increased CRF content in the PVN, which is compatible with the observation that ectopic expression of hypothalamic BDNF increases CRF mRNA expression (Jeanneteau et al. 2012). Therefore, we speculate that BDNF activates histamine neurons through CRF neurons.

First, we demonstrated that central BDNF infusion increased the CRF content in the PVN of the hypothalamus. The central administration of α-helical CRF prevented the reduction in food intake induced by BDNF treatment, suggesting that CRF mediates the anorectic effect of BDNF in the hypothalamus. Moreover, it has been reported that CRF suppresses feeding when it is injected into the PVN, but not the LH, of rats (Krahn et al. 1988). There is the report that few VMH neurons project to the PVN, suggesting that it does not make much sense to propose that BDNF-expressing neurons in the VMH project to the PVN to affect CRF-expressing neurons (Canteras et al. 2004; Lin and York 2004). However, we identified that microinfusion of BDNF into the PVN also increased t-MH contents in the PVN, VMH, and TMN in the hypothalamus. Considering other studies that VMH neurons project to the PVN using retrograde tracer fluorogold and that VMH-selective knockdown of BDNF induces hyperphagia and obesity, BDNF neurons in the VMH may innervate and stimulate CRF in the PVN (Buijs et al. 2001; Unger et al. 2007).

Our previous study indicated that the CRF activated histamine neuron in the TMN through CRF1-R (Gotoh et al. 2005). Furthermore, we hypothesized that BDNF regulates neuronal histamine through CRF neurons. Our results demonstrated that BDNF induces the reduction of feeding by activating histamine neurons in the TMN through CRF neurons in the PVN, supporting our hypothesis. A previous study indicated that the suppressive effect of the CRF2-R antagonist on BDNF-induced hypophagia was less than that of α-helical CRF, the dual antagonist of both CRF1-R and CRF2-R, suggesting that the CRF1-R pathway can cooperate with the CRF2-R pathway to fully mediate the BDNF action to regulate feeding (Toriya et al. 2010).

Finally, we examined whether neuronal histamine activated BDNF neurons because neuronal histamine stimulates CRF neurons through H1-R (Kjaer et al. 1998). We showed that histamine also elevates BDNF expression in the VMH, suggesting that it is possible that some BDNF neurons may be stimulated with activated histamine neurons by CRF.

It seems reasonable to speculate that BDNF administered peripherally enters the hypothalamus by crossing the blood–brain barrier (BBB) and regulates energy metabolism and appetite because it has been reported that blood BDNF stored in platelets crosses the BBB and reaches the central nervous system (Pan et al. 1998). Furthermore, it is always possible that BDNF infused centrally might leak into the periphery via the venous system. This study showed that a single central administration of BDNF at a dose too low to have an effect when delivered peripherally was able to immediately suppress feeding behavior and there was no significant difference in plasma BDNF levels between ivt and ip administrations of BDNF, supporting the findings that BDNF has a short serum half-life of just minutes and the dose of peripherally injected BDNF that required to show any effect on feeding is 500 times greater than the dose that induced appetite suppression centrally (Kishino et al. 2001; Suwa et al. 2010).

A previous study demonstrated that selective microinfusion of FMH to the VMH and PVN induces feeding (Sakata et al. 1990). Here, essential questions can be raised as to why hyperphagia was not observed in histamine-depleted rats by FMH and food intake did not differ between wild-type and H1KO mice. There is no definite answer to the query to date. However, these results are compatible with previous findings that histamine depletion by the treatment of FMH did not affect daily food intake and basal food intake in H1KO mice tended to be higher than in wild-type mice, but this effect was not significant (Yoshimatsu et al. 1999; Mollet et al. 2001). There might be some possible and reasonable explanations. One is that FMH infusion to the third ventricle is weaker to deplete histamine than the infusion into specific nuclei. The other is that the differences in mean daily food intake between PBS treatment and FMH treatment as well as between wild-type and H1KO mice is relatively small compared with the deviation of daily food intake in both groups. In addition, the increase of food intake with the treatment of α-helical CRF alone was not observed. This finding is also compatible with previous reports that α-helical CRF attenuates leptin-induced reduction of food intake although a single administration of α-helical CRF does not significantly affect food consumption, and α-helical CRF locally infused into the PVN had no significant effect on food intake (Uehara et al. 1998; Toriya et al. 2010). There may be a possibility that a higher dose of α-helical CRF would have resulted in a significant reduction of food intake compared to control group as well as a more sustained attenuation of the anorexigenic effect of BDNF. Considering that BDNF stimulates histamine neurons through CRF neurons and neuronal histamine activates BDNF neurons directly, there might be a close relationship between histamine and BDNF in circadian rhythm regulation.

BDNF is synthesized in several areas of the hypothalamus, including the PVN, VMH, and LH (Conner et al. 1997). Hypothalamic BDNF expression is highest in the VMH (Unger et al. 2007). This region is important in the regulation of energy metabolism. The VMH is connected to several hypothalamic areas and contains receptors for nutritional status signals, including leptin (King 2006). However, far less is known regarding the role of BNDF and TrkB in the LH. The LH has often been referred to as a feeding center, as lesions to this region result in hypophagia and weight loss (Anand and Brobeck 1951). It contains cells that synthesize melanin concentrating hormone (MCH) and hypocretin, two orexigenic factors (Date et al. 1999). We found that central administration of histamine increased BDNF content in the PVN and VMH, but not in the LH, supporting a previous finding that BDNF injected in the LH did not reduce feeding and body weight, although BDNF and its receptors are expressed in the LH (Wang et al. 2007a, b).

This study supports the possibility that the anorectic effects of BDNF are because of the proposed VMH-PVN-TMN network (Fig. 6), providing novel insight into the action of BDNF in the hypothalamic regulation of energy metabolism.

image

Figure 6. Proposed feeding model for the interactions among brain-derived neurotrophic factor (BDNF), corticotropin-releasing factor (CRF), and histamine neurons. BDNF regulates histamine neurons in the tuberomammillary nucleus (TMN) through CRF neurons, whose receptors are expressed on the histamine neurons. Neuronal histamine suppresses food intake directly via H1-R, which is expressed on feeding-related neurons, including BDNF neurons, in the ventromedial nucleus (VMH) and paraventricular nucleus (PVN). ‘Feeding-related neurons’ indicates neurons that produce peptides such as BDNF and neuropeptide Y (NPY) or are co-localized with receptors such as H1-R and influence feeding behavior.

Download figure to PowerPoint

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

This study was supported by a grant for Research on Measures for Intractable Diseases from Japan's Ministry of Health, Labour, and Welfare. No other potential conflicts of interest relevant to this article were reported.

References

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • Anand B. K. and Brobeck J. R. (1951) Localization of a “feeding center” in the hypothalamus of the rat. Proc. Soc. Exp. Biol. Med. 77, 323324.
  • Buijs R. M., Chun S. J., Niijima A., Romijn H. J. and Nagai K. (2001) Parasympathetic and sympathetic control of the pancreas: a role for the suprachiasmatic nucleus and other hypothalamic centers that are involved in the regulation of food intake. J. Comp. Neurol. 431, 405423.
  • Canteras N. S., Simerly R. B. and Swanson L. W. (2004) Organization of projections from the ventromedial nucleus of the hypothalamus: a Phaseolus vulgaris-leucoagglutinin study in the rat. J. Comp. Neurol. 348, 4179.
  • Conner J. M., Lauterborn J. C., Yan Q., Gall C. M. and Varon S. (1997) Distribution of brain-derived neurotrophic factor (BDNF) protein and mRNA in the normal adult rat. J. Neurosci. 17, 22952313.
  • Date Y., Ueta Y., Yamashita H., Yamaguchi H., Matsukura S., Kangawa K., Sakurai T., Yanagisawa M. and Nakazato M. (1999) Orexins, orexigenic hypothalamic peptides, interact with autonomic, neuroendocrine and neuroregulatory systems. Proc. Natl Acad. Sci. USA 96, 748753.
  • Doi T., Sakata T., Yoshimatsu H., Machidori H., Kurokawa M., Jayasekara L. A. and Niki N. (1994) Hypothalamic neuronal histamine regulates feeding circulation in rats. Brain Res. 641, 311318.
  • Franklin K. B. J. and Paxinos G. (1997) The Mouse Brain in Stereotaxic Coordinates. Academic Press, San Diego.
  • Fukagawa K., Sakata T., Shiraishi T., Yoshimatsu H., Fujimoto K., Ookuma K. and Wada H. (1989) Neuronal histamine modulates feeding behavior through H1-receptor in rat hypothalamus. Am. J. Physiol. 256, R605R611.
  • Gotoh K., Fukagawa K., Fukagawa T., Noguchi H., Kakuma T., Sakata T. and Yoshimatsu H. (2005) Glucagon-like peptide-1, corticotrophin-releasing hormone, and hypothalamic neuronal histamine interact in the leptin-signaling pathway to regulate feeding behavior. FASEB J. 19, 11311133.
  • Inoue I., Yanai K., Kitamura D., Taniuchi I., Kobayashi T., Niimura K., Watanabe T. and Watanabe T. (1996) Impaired locomotor activity and exploratory behavior in mice lacking histamine H1 receptors. Proc. Natl Acad. Sci. USA 93, 1331613320.
  • Jeanneteau F. D., Lambert W. M., Ismaili N., Bath K. G., Lee F. S., Garabedian M. J. and Chao M. V. (2012) BDNF and glucocorticoids regulate corticotropin-releasing hormone (CRH) homeostasis in the hypothalamus. Proc. Natl Acad. Sci. USA 109, 13051310.
  • Kernie S. G., Liebl D. J. and Parada L. F. (2000) BDNF regulates eating behavior and locomotor activity in mice. EMBO J. 19, 12901300.
  • King B. M. (2006) The rise, fall, and resurrection of the ventromedial hypothalamus in the regulation of feeding behavior and body weight. Physiol. Behav. 87, 221244.
  • Kishino A., Katayama N., Ishige Y., Yamamoto Y., Ogo H., Tatsuno T., Mine T., Noguchi H. and Nakayama C. (2001) Analysis of effects and pharmacokinetics of subcutaneously administered BDNF. NeuroReport 12, 10671072.
  • Kjaer A., Larsen P. J., Knigge U., Johrgensen H. and Warberg J. (1998) Neuronal histamine and expression of corticotropin-releasing hormone, vasopressin and oxytocin in the hypothalamus: relative importance of H1 and H2 receptors. Eur. J. Endocrinol. 139, 238243.
  • Krahn D. D., Gosnell B. A., Levine A. S. and Morley J. E. (1988) Behavioral effects of corticotropin-releasing factor: localization and characterization of central effects. Brain Res. 443, 6369.
  • Lin L. and York D. A. (2004) Amygdala enterostatin induces c-Fos expression in regions of hypothalamus that innervate the PVN. Brain Res. 1020, 147153.
  • Lyons W. E., Mamounas L. A., Ricaurte G. A., Coppola V., Reid S. W., Bora S. H., Wihler C., Koliatsos V. E. and Tessarollo L. (1999) Brain-derived neurotrophic factor-deficient mice develop aggressiveness and hyperphagia in conjunction with brain serotonergic abnormalities. Proc. Natl Acad. Sci. USA 96, 1523915244.
  • Mollet A., Lutz T. A., Meier S., Riediger T., Rushing P. A. and Scharrer E. (2001) Histamine H1 receptors mediate the anorectic action of the pancreatic hormone amylin. Am. J. Physiol. Regul. Integr. Comp. Physiol. 281, R1442R1448.
  • Pan W., Bank W. A., Fasold M. B., Bluth J. and Kastin A. J. (1998) Transport of brain-derived neurotrophic factor across the blood-brain-barrier. Neuropharmacology 37, 15531561.
  • Paxinos G. and Watson C. (1997) The Rat Brain in Stereotaxic Coordinates. Academic Press, New York.
  • Pelleymounter M. A., Gullen M. J. and Wellman C. L. (1995) Characteristics of BDNF-induced weight loss. Exp. Neurol. 131, 229238.
  • Rios M., Fan C., Fekete C., Kelly J., Bates B., Kuehn R., Lechan R. M. and Jaenisch R. (2001) Conditional deletion of brain-derived neurotrophic factor in the postnatal brain leads to obesity and hyperactivity. Mol. Endocrinol. 15, 17481757.
  • Sakata T. and Yoshimatsu H. (1997) Hypothalamic neuronal histamine: implications of homeostatic maintenance in its control of energy metabolism. Nutrition 13, 403411.
  • Sakata T., Fujimoto K., Kurata K., Etou H., Fukagawa K., Okabe Y. and Ookuma K. (1981) Feeding and hyperglycemia induced by 1,5-anhydroglucitol in the rat. Physiol. Behav. 27, 401405.
  • Sakata T., Fukagawa K., Ookuma K., Fujimoto K., Yoshimatsu H., Yamatodani A. and Wada H. (1990) Hypothalamic neuronal histamine modulates ad libitum feeding by rats. Brain Res. 537, 303306.
  • Suwa M., Yamamoto K. I., Nakano H., Sasaki H., Radak Z. and Kumagai S. (2010) Brain-derived neurotrophic factor treatment increases the skeletal muscle glucose transporter 4 protein expression in mice. Physiol. Res. 59, 619623.
  • Toriya M., Maekawa F., Maejima Y., Onaka T., Fujiwara K., Nakagawa T., Makata M. and Yada T. (2010) Long-term infusion of brain-derived neurotrophic factor reduces food intake and body weight via a corticotrophin-releasing hormone pathway in the paraventricular nucleus of the hypothalamus. J. Neuroendocrinol. 22, 987995.
  • Uehara Y., Shimizu H., Ohtani K., Sato N. and Mori M. (1998) Hypothalamic corticotropin-releasing hormone is a mediator of the anorexigenic effect of leptin. Diabetes 47, 890893.
  • Unger T. J., Calderon G. A., Bradley L. C., Sena-Esteves M. and Rios M. (2007) Selective deletion of BDNF in the ventromedial and dorsomedial hypothalamus of adult mice results in hyperphagic behavior and obesity. J. Neurosci. 27, 1426514274.
  • Wang C., Bomberg E., Billington C., Levine A. and Kotz C. M. (2007a) Brain-derived neurotrophic factor in the hypothalamic paraventricular nucleus reduces energy intake. Am. J. Physiol. Regul. Integr. Comp. Physiol. 293, R1003R1012.
  • Wang C., Bomberg E., Billington C., Levine A. and Kotz C. M. (2007b) Brain-derived neurotrophic factor in the ventromedial nucleus of the hypothalamus reduces energy intake. Am. J. Physiol. Regul. Integr. Comp. Physiol. 293, R1037R1045.
  • Xu B., Goulding E. H., Zang K., Cepoi D., Cone R. D., Jones K. R., Tecott L. H. and Reichart L. F. (2003) Brain-derived neurotrophic factor regulates energy balance downstream of melanocortin-4 receptor. Nat. Neurosci. 6, 736742.
  • Yan Q., Radeke M. J., Matheson C. R., Talvenheimo J., Welcker A. A. and Feinstein S. C. (1997) Immunocytochemical localization of TrkB in the central nervous system of the adult rat. J. Comp. Neurol. 378, 135157.
  • Yoshimatsu H., Machidori H., Doi T., Kurokawa M., Ookuma K., Kang M. and Sakata T. (1992) Abnormalities in obese Zuckers: defective control of histaminergic functions. Physiol. Behav. 54, 487491.
  • Yoshimatsu H., Itateyama E., Kondou S., Tajima D., Himeno K., Hidaka S., Kurokawa M. and Sakata T. (1999) Hypothalamic neuronal histamine as a target of leptin in feeding behavior. Diabetes 48, 22862291.