Brain-Derived Neurotrophic Factor Plays a Role as an Anorexigenic Fact
http://www.100md.com
《内分泌学杂志》
Laboratoire de Physiologie Neurovegetative, Unite Mixte de Recherche Universite Paul Cezanne Aix-Marseille III, Centre National de la Recherche Scientifique (Unite Mixte de Recherche 6153), Institut National de la Recherche Agronomique (Unite Mixte de Recherche 1147), Faculte des Sciences et Techniques, 13397 Marseille cedex 20, France
Abstract
Brain-derived neurotrophic factor (BDNF) has recently been implicated as an anorexigenic factor in the central control of food intake. Previous studies focused on the hypothalamus as a probable site of action for this neurotrophin. It was demonstrated that BDNF is an important downstream effector of melanocortin signaling in the ventromedial hypothalamus. In this study, we addressed whether BDNF can modulate food intake in the hindbrain autonomic integrator of food intake regulation, i.e. the dorsal vagal complex (DVC). To this end, we used two complementary methodological approaches in adult rats. First, we measured the effects of intraparenchymal infusions of exogenous BDNF within the DVC on food intake and body weight. Second, we measured the endogenous BDNF protein content in the DVC and hypothalamus after food deprivation, refeeding, or peripheral treatments by the anorexigenic hormones leptin and cholecystokinin (CCK). BDNF infusion within the DVC induced anorexia and weight loss. In the DVC, BDNF protein content decreased after 48 h food deprivation and increased after refeeding. Acute and repetitive peripheral leptin injections induced an increase of the BDNF protein content within the DVC. Moreover, peripheral CCK treatment induced a transient increase of BDNF protein content first in the DVC (30 min after CCK) and later on in the hypothalamus (2 h after CCK). Taken together, these results strongly support the view that BDNF plays a role as an anorexigenic factor in the DVC. Our data also suggest that BDNF may constitute a common downstream effector of leptin and CCK, possibly involved in their synergistic action.
Introduction
RECENT EVIDENCE INDICATES that brain derived neurotrophic factor (BDNF), a member of the neurotrophin family, contributes to food intake and body weight control, acting as an anorexigenic factor in adult rodents. Indeed, infusion of BDNF in the lateral ventricles induces a reduction in food intake associated with weight loss in rats (1). Conversely, mice heterozygous for targeted disruption of BDNF, as well as conditional BDNF mutants, show hyperphagia and obesity (2, 3). Moreover, the same phenotype was observed in mice with a reduced expression of BDNF high-affinity tyrosine kinase receptor type B (TrkB) at a quarter of the normal amount (4). Mechanistic interpretations of these data all focused on the hypothalamus, which is indeed the best documented among autonomic centers involved in energy homeostasis integration. Xu et al. (4) demonstrated that BDNF is expressed at high levels in the ventromedial hypothalamus (VMH), in which its expression is regulated by nutritional state and melanocortin-4 receptor (MC4R) signaling. In addition, the same authors showed that intracerebroventricular infusion of BDNF suppresses the hyperphagia and excessive weight gain observed on high-fat diets in mice with deficient MC4R signaling, suggesting that BDNF functions as a downstream effector through which MC4R signaling regulates energy balance.
However, the dorsal vagal complex (DVC) of hindbrain constitutes another autonomic integrator of food intake regulation. The DVC is located in the caudal brainstem and comprises three structures, namely the nucleus tractus solitarii (NTS), area postrema, and dorsal motor nucleus of the vagus nerve. The DVC contains leptin and insulin receptors, glucose-sensing mechanisms, and neuropeptide mediators relevant to energy balance (5). The food intake-inhibiting hormone leptin has been shown to act directly within the DVC (6, 7). It has also been demonstrated that intraparenchymal injections of MC4R agonists or antagonists in the DVC, respectively, reduce and enhance food intake (8), consistent with the facts that the MC4R has its highest density in the DVC, and the NTS is one of the two structures in the brain containing proopiomelanocortin (POMC) neurons. More recently it was shown that brainstem NTS neurons expressing POMC are activated by the hormone cholecystokinin (CCK) and that activation of the brainstem neuronal MC4R is required for CCK-induced suppression of feeding (9). The DVC was also shown to contain high densities of BDNF and TrkB (10, 11). In this study, we addressed the question of whether BDNF can modulate food intake directly inside the DVC. We demonstrated that intraparenchymal infusions of exogenous BDNF in the DVC induced anorexia and weight loss and that endogenous BDNF protein content in the same structure was regulated by nutritional state as well as by peripheral leptin. Moreover, we showed that peripheral CCK-8S treatment induced a transient increase of BDNF protein content first in the DVC and later in the hypothalamus. Preliminary accounts of these results have been published in abstract form [Digestive Disease Week 2004 (12) and International Society of Autonomic Neuroscience 2005 (13)].
Materials and Methods
Animals and housing conditions
Adult male Wistar Han rats from Charles River Laboratories (L’Arbresle, France), weighting 290 ± 10 g at the time of experiments, were housed singly in metal hanging cages at constant temperature (25 C) under a 12-h light, 12-h dark cycle with lights off at 1800 h. Water and standard pellet food (pellet AO4, UAR, Villemoisson-sur-Orge, France) were available ad libitum unless otherwise noted. All rats were habituated to the hanging cage and to handling for at least 5 d before the experiments. An additional 5-d habituation period with a daily ip injection of saline was performed when necessary. All procedures were performed in accordance with Principles of Laboratory Animal Care (National Institutes of Health publication 86-23, revised 1985) as well as with French national law.
Experiment 1: intracerebral infusion of exogenous BDNF
The human recombinant BDNF used in this study was a kind gift from Regeneron (Tarrytown, NY). Osmotic pumps (model 1002, 0.25 μl/h for 14 d; Alzet, Cupertino, CA) were filled immediately before implantation under sterile atmosphere, with either human recombinant BDNF (167 ng/μl BDNF for 1 μg/d treatment, n = 7 and 16.7 ng/μl BDNF for 0.1 μg/d treatment, n = 3) diluted in artificial cerebrospinal fluid (aCSF) [mixture 1:1 of the A and B solutions; solution A: 296.4 mM NaCl, 6 mM KCl, 2.8 mM CaCl22H2O, 1.6 mM MgCl26H2O; solution B: 1.6 mM Na2HPO47H2O, 0.4 mM NaH2PO4H2O (pH 7.4)] or aCSF alone (n = 4). Osmotic pumps were then connected to a cannula (model 330OP/DW/Spc, 30G; Plastics One, Roanoke, VA,) with 6 cm of vinyl tubing (model C312VT; Plastics One) filled with the same solution. Rats weighting 290 ± 10 g received a sc injection of atropine (0.18 mg/kg) and were anesthetized 30 min later with an ip injection of ketamine and xylazine (100 and 15 mg/kg ip, respectively). The tip of the infusion cannula was implanted stereotaxically into the DVC in reference to the stereotaxic atlas of Paxinos and Watson (area postrema: –5 mm from interaural line; lateral: –1.3 mm from midline; dorsoventral: 1.8 mm above interaural line) (14), and the cannula was cemented to three jeweler’s screws attached to the skull. The osmotic pump was then placed into a sc pocket in the subscapular area. Rats recovered for 1 d in individual thermostated cage with food and water ad libitum and were then placed back in their individual hanging cages.
Weight, food, and water intake measurements.
Body weight and food and water intakes were measured every day. Rats were removed from their cages 1 h before lights off (1700 h) and weighed. Food hoppers, food debris collected under each cage, and water bottles were weighed, and daily food and water intake values were determined. Rats were returned to their cages and supplied with fresh preweighed food and water at the onset of the dark period.
Postmortem localization of infusion sites
On the 15th day after implantation, rats were anesthetized, osmotic pumps were removed, and the volume remaining in each pump was determined. Infusion sites were labeled by injecting 0.8 μl pontamine blue 2% through the cannula (the estimated dead volume of the cannula and connecting tube was 0.6 μl). Rats were perfused transcardially with 150 ml 0.1 M PBS and then fixed with 250 ml of a solution of paraformaldehyde 4% (in phosphate buffer 0.1 M at 4 C). Brains were removed, postfixed 1 h in paraformaldehyde 4% in PBS 0.1 M at 4 C, and then cryoprotected 24–48 h in a solution of sucrose 30% (in PBS at 4 C). After sedimentation in sucrose, brains were frozen by immersion in isopentane at –40 C for 20 sec and then stored at –80 C. Thirty-micrometer sections were cut in the coronal plane, and the cannula tip location was determined.
Experiment 2: modulation of the endogenous BDNF protein content
Food deprivation.
Rats were maintained on food pellets and water ad libitum (control) or food deprived for 48 h, with water ad libitum. DVCs and hypothalami were collected as described below. The BDNF protein content of the collected samples was analyzed by either Western blotting only (n = 6 per group) or both Western blotting and ELISA (n = 6 per group), as described below.
Refeeding.
Rats (n = 12) were given access to chow at 0900–1000 and 1400–1800 h daily for 5 d before the experiment, according to a previously published method (9). On the day of the experiment, one group (refeed, n = 6) received chow at 0900–1000 h, and the other group (control, n = 6) did not. Rats were killed at 1000 h, and DVCs and hypothalami were collected and their BDNF protein content analyzed by ELISA, as described below.
Leptin injections
The murine leptin used in this study was a kind gift from Amgen (Thousand Oaks, CA). Leptin dose was chosen to produce a significant but moderate change in body weight gain, within a repetitive treatment protocol. We used a protocol previously described to reduce body weight gain in Long-Evans rats (15) and verified its effects on body weight gain in Wistar Han rats.
For acute treatment, rats fed ad libitum (290 ± 10 g) were injected with leptin (0.2 mg/rat; ip, n = 6) or saline (n = 6) and were killed 6 h after injection, given the slow onset of leptin-induced decrease in food intake (16 ; for review see also Ref.17). DVCs and hypothalami were collected and their BDNF protein content analyzed by ELISA, as described below.
For repetitive treatment, rats fed ad libitum (290 ± 10 g) were injected daily (30 min before the end of the light period) with saline (control) or leptin (0.2 mg/rat; ip), during 3 d. Rats were killed 20 h after the last injection. In a first series of experiments, only the DVCs were collected and submitted to Western blot analysis (n = 6 per group). In a second series of experiments, DVCs and hypothalami were collected and their BDNF protein content analyzed by both Western blotting and ELISA, as described below (n = 6 per group). These rats were also weighed daily at 1400 h to assess the effects of leptin treatment on body weight gain.
CCK-8S injections
Rats fed ad libitum (290 ± 10 g) were injected with saline or CCK-8S (6 μg/kg; ip; Sigma, Saint Quentin Fallavier, France) and killed 30 min, 2 h, or 20 h after injection (six groups; n = 6 per group). DVCs and hypothalami were collected and their BDNF protein content analyzed by both Western blotting and ELISA, as described below.
DVC and hypothalamus dissection
Rats were anesthetized with halothane and decapitated, the brain was removed from the skull, and the hypothalamus was dissected out under binocular control. The brainstem and upper cervical spinal cord were removed rapidly and glued to the cutting stage of a vibratome. Sectioning of 500-μm slices and microdissection of DVC from the slices of interest, under binocular control, were performed in chilled aCFS. Tissue samples (DVC and hypothalamus) were placed in microcentrifuge tubes, immediately frozen in liquid nitrogen, and stored at –80 C until use. For each rat, the total amount of time spent from killing to freezing of sample was 7 min, 15 sec ± 10 sec for the hypothalamus and 16 min, 02 sec ± 13 sec for the DVC. To minimize circadian variations of BDNF in our experiments, test and control rats were killed at equal daytime on alternate days, and whole groups were processed within less than 2 h.
Protein extraction and dosage
Tissue samples were homogenized in lysis buffer (100 μl for DVC, 500 μl for hypothalamus) [NaCl 137 mM; Tris-HCl 20 mM; Triton X-100 1%, glycerol 10%, sodium orthovanadate 1 mM, phenylmethylsulfonyl fluoride 1 mM, protease inhibitors cocktail (Sigma)], incubated 30 min at 4 C, and centrifuged at 15,000 x g (30 min at 4 C). The supernatants were collected and protein concentration determined by the Bradford’s procedure (18).
Protein gel electrophoresis and immunoblots
Protein extracts were mixed with sodium dodecyl sulfate sample buffer (sample buffer Laemmli; Sigma), denatured for 3 min at 100 C, and transferred on ice. Equal amounts of protein extracts were separated by electrophoresis in 12% polyacrylamide gels containing sodium dodecyl sulfate and transferred on a nitrocellulose membrane (Immobilon-P transfer membranes; Millipore, Molsheim, France). The BenchMark prestained protein ladder (Invitrogen, Cergy Pontoise, France) was used for determination of apparent molecular mass. Because BDNF immunorevelation gave no signal at an apparent molecular mass superior to 27 kDa, the nitrocellulose membranes were subsequently cut through a line corresponding to an apparent molecular mass of 37.1 kDa for a separate BDNF and -actin immunorevelation. BDNF and -actin were revealed separately on their corresponding membrane parts by incubation with anti-BDNF rabbit polyclonal IgG (1:750; N-20; Santa Cruz Biotechnology, Santa Cruz, CA) or anti--actin monoclonal mouse IgG2 (1:2000; Sigma), followed by horseradish peroxidase-conjugated secondary antibody (goat antirabbit IgG for BDNF and goat antimouse IgG for -actin; 1:2000; Dako, Trappes, France). After rinsing with buffer, the immunocomplexes were detected by chemiluminescence using the enhanced chemiluminescence kit (Amersham Biosciences, Freiburg, Germany) according to the manufacturer’s instructions. The film signals were digitally scanned and then quantified using Quantity One software (Bio-Rad Laboratories, Marnes la Coquette, France). The -actin was used as internal standard for normalization.
BDNF immunoassay
BDNF protein was also measured with a conventional two-site ELISA. BDNF Emax immunoassay (Promega, Charbonniere, France) was performed according to the manufacturer’s protocol. For each sample, 30 μg of total proteins were used to determine BDNF content. The sensitivity of the assay was 15 pg/ml, and the cross-reactivity with other related neurotrophic factors was less than 3%. The within-assay variability was less than 3%, and all relevant comparisons were made within the same assay. BDNF concentration was determined as picograms per milligram of total protein, and relative change in BDNF content between experimental groups expressed as percentage of control values.
Statistical analysis
Statistical analyses were performed using STATView (version 5.0.1; Statview Software, Cary, NC). Body weight, food and water intake were analyzed using a two-way, repeated-measures ANOVA in which difference from presurgical weight or food or water intake was the within-groups variable, and dose was the between-groups variable. A simple-way ANOVA was used to assess the effects of food deprivation, refeeding, and leptin treatments on BDNF protein content. A two-way ANOVA was used to assess the effects of CCK treatment on BDNF protein content and the effects of repetitive leptin treatment on cumulative body weight gain. In all cases, when a significant main effect was present, Fischer test was used for post hoc analyses. All data are expressed as mean ± SEM.
Results
Exogenous BDNF infusion in the DVC
Data for 14 rats with verified infusion site in the DVC (see Fig. 1) and pump functioning are included in these analyses. Data from three rats were excluded based on cannula localization outside the DVC (n = 2) or a disconnected pump (n = 1).
Body weight
Control rats infused with aCSF showed a linear growth (4.87 g/d, R2 = 0.998) throughout the infusion period. Intraparenchymal infusion of BDNF in the DVC reduced the cumulative weight gain in a dose- (F2 = 12.89; P = 0.0013) and time- (F13 = 30.81; P < 0.0001) dependent manner, as shown in Fig. 2A. There was a significant interaction between dose and time (F26 = 3.99; P < 0.0001). At the highest BDNF dose tested, body weight first decreased gradually, reaching its minimum 5 d after the beginning of the infusion period and then increased linearly at a rate remaining lower than that of control rats (2.49 g/d, R2 = 0.9598). At a dose of 0.1 μg/d, BDNF induced a reduction of growth rate until the fourth postimplant day, followed by a linear growth comparable with that of control rats (4.63 g/d, R2 = 0.9977). Simple-effects analysis of the interaction between time and dose revealed that infusion of BDNF significantly decreased cumulative body weight gain, compared with controls, from the second postimplant day to the end of the infusion period at a dose of 1 μg/d and from the fifth postimplant day to the end of the infusion period at a dose of 0.1 μg/d.
Food and water intakes
Control rats infused with aCSF showed a relatively constant food intake over the infusion period (24.41 ± 0.71 g/d). Intraparenchymal infusion of BDNF in the DVC reduced the daily food intake in a dose- (F2 = 19.51; P = 0.0002) and time- (F13 = 2.71; P < 0.0001) dependent manner, as can be seen in Fig. 2B. There was also a significant interaction between time and dose (F26 = 2.52; P = 0.0005). Simple-effects analysis of this interaction revealed a significant anorexigenic effect of BDNF, compared with control, from the first postimplant measurement of food intake, to the eighth day of infusion for the 0.1 μg/d dose and until the end of the infusion period for the 1 μg/d dose. The anorexigenic effect of BDNF was highest on the first measurement day (second postimplant day) (daily food intake reduced by –54.7% for BDNF 0.1 μg/d and –67.4% for BDNF 1 μg/d) and decreased gradually over time.
Control rats infused with aCSF showed a relatively constant water intake over the infusion period (29.22 ± 0.48 g/d). Although the mean daily water intake of BDNF-treated rats appeared lower than that of controls (23.26 ± 3.24 g/d for the 1 μg/d BDNF dose; 21.07 ± 3.12 g/d for the 0.1 μg/d BDNF dose), the repeated-measures ANOVA analysis did not show any significant effect of treatment on water intake.
The above results show that intraparenchymal infusion of exogenous BDNF in the DVC induced anorexia and weight loss in rats, suggesting that this neurotrophin acts as an anorexigenic signaling factor within the DVC.
Regulation of endogenous BDNF protein content
To further evaluate the implication of BDNF in the regulation of food intake within the DVC, we next asked whether its endogenous protein content could be modulated by nutritional status and peripheral injections of leptin or CCK-8S.
Endogenous BDNF protein content in DVC and hypothalamus were first measured by Western blotting. Immunorevelation of BDNF produced a single band corresponding to an apparent molecular mass of 27 kDa, i.e. twice the size of monomeric mature BDNF (13.5 kDa), and well below the apparent molecular mass of pro-BDNF (35 kDa). We next used a commercially available sandwich ELISA kit for quantifying BDNF unambiguously (BDNF Emax Immunoassay System, Promega). These two quantification methods gave similar results in terms of relative changes in BDNF protein content both after a 48-h fasting, and after leptin or CCK-8S treatment. Results presented in Figs. 3–5 were obtained by ELISA. Results obtained by Western blotting are shown in Fig. 6 and Table 1 for comparison.
Effect of 48 h fasting and refeeding on BDNF content in DVC and hypothalamus
Food deprivation for 48 h led to a statistically significant reduction in BDNF protein content in the DVC when compared with rats fed ad libitum: 100 ± 14.93% for control (n = 6) vs. 63 ± 3.16% for fasted rats (n = 6) (F1 = 9.26; P = 0.03) (see Fig. 3A). By contrast, there was no significant effect of a 48 h fasting on BDNF protein content in the hypothalamus taken as a whole: 100 ± 6.88% for control (n = 6) vs. 99.87 ± 6.76% for fasted rats (n = 6) (data not shown). Moreover, refeeding during 1 h after a 15-h fast induced a statistically significant increase in BDNF protein content in the DVC when compared with rats fasted during 16 h (control group): 100 ± 3.87% for control (n = 6) vs. 131.06 ± 11.47% for refed rats (n = 6) (F1 = 6.586; P = 0.0281) (see Fig. 3B). By contrast, BDNF protein content in the hypothalamus was not affected by refeeding: 100 ± 5.26% for control (n = 5; one sample was lost during freezing) vs. 108.40 ± 7.04% for refed rats (n = 6) (data not shown).
Effect of acute and repeated leptin injections on BDNF content in DVC and hypothalamus
In our hands, repetitive daily leptin injections (0.2 mg/rat per day ip) induced a reduction of cumulative body weight gain that became statistically significant only after the third treatment day, compared with control: 8.6 ± 1.26 g for leptin-treated rats (n = 6) vs. 13.2 ± 0.54 g for saline-treated rats (n = 6) (F1 = 8.138; P = 0.0139) (data not shown).
Single peripheral leptin injection (0.2 mg/rat ip) induced a statistically significant increase in BDNF protein content in the DVC when measured 6 h after injection, compared with control: 100 ± 4.52% for control (n = 6) vs. 137.56 ± 15.5% for leptin-treated rats (n = 6) (F1 = 5.408; P = 0.0424) (see Fig. 4A). By contrast, BDNF content measured in the whole hypothalamus from the same rats was not modified by leptin treatment: 100 ± 5.40% for control (n = 6) vs. 90.08 ± 5.89% for leptin-treated rats (n = 6) (data not shown).
Cumulative leptin treatment (0.2 mg/rat per day ip) for 3 consecutive days led to a highly significant increase in BDNF protein content in the DVC, as measured 20 h after the last injection, compared with control: 175.25 ± 5.04% for leptin-treated rats (n = 6) vs.100 ± 2.84% for saline-treated controls (n = 6) (F1 = 143.40; P < 0.0001) (see Fig. 4B). Conversely, BDNF content measured in the whole hypothalamus from the same rats was not modified by leptin treatment: 100 ± 4.88% for control (n = 6) vs. 103.08 ± 9.71% for leptin-treated rats (n = 6) (data not shown).
Effect of a single CCK-8S ip injection on BDNF content in DVC and hypothalamus
An ip injection of CCK-8S (6 μg/kg) in rats fed ad libitum led to a rapid and transient increase in BDNF protein content in the DVC. Thirty minutes after injection, BDNF levels were significantly higher in CCK-8S-treated rats (125.20 ± 4.59%; n = 6), compared with saline-treated controls (100 ± 2.95%; n = 6) (F1 = 21.40; P < 0.001) (Fig. 5A), whereas there were no significant differences at later time points (2 h and 20 h).
In the hypothalamus, CCK-8S treatment did not modulate BDNF protein content 30 min after injection but led to a significant increase 2 h after injection [190.51 ± 8.72% for CCK-8S-treated rats (n = 6) vs. 100 ± 6.94% for saline-treated rats (n = 6) (F1 = 65.89; P < 0.0001)]. This increase in BDNF levels was also transient because there were no statistically significant differences between CCK-8S- and saline-treated rats 20 h after injection.
Discussion
In this study we provide for the first time evidence for a role of BDNF as an anorexigenic factor in the DVC of the adult rat. First, we showed that intraparenchymal infusion of exogenous BDNF within the DVC induced a dose-dependent anorexia. Second, we showed that endogenous BDNF protein content in the DVC decreased after fasting, increased after refeeding, and increased after peripheral leptin or CCK treatment.
Exogenous BDNF infusion in the DVC-induced anorexia
Intraparenchymal infusion of BDNF at a dose of 0.1 μg/d resulted in a transient anorexia, with an initial 54% decrease in daily food intake, compared with aCSF-treated rats, and an overall 30% reduction in the cumulative food intake measured over the first 8 d of treatment. It should be noted that, whereas BDNF-treated rats did not eat significantly less than control rats during the remaining infusion period, their weight curve tended to be parallel to, but below, that of control rats. We interpret the return of food intake after 8 d of treatment to, but not above, control levels, as implying that BDNF had reduced but not totally lost its effects on food intake at this point. At a higher dose (1 μg/d), the anorexigenic effect of BDNF was more severe and remained significant all along the infusion period, with an initial 67% decrease in daily food intake, followed by a gradual decrease in severity, resulting in an overall 43% reduction of cumulative food intake over the 14 d of infusion and a transient weight loss for the first 5 d of infusion. Therefore, our results show that BDNF causes anorexia and weight loss when infused in the DVC, albeit in a less severe manner than that observed by others after infusion in the lateral ventricle (1, 19). Thus, the TrkB receptors within the DVC appear as one of the targets at which BDNF exert its anorexigenic signaling.
Other target sites for BDNF/TrkB anorexigenic signaling have been pointed out by others. In the hypothalamus, several lines of evidence have identified the VMH as a key site in which BDNF expression is modulated by nutritional state and MC4R signaling (2, 4). Kernie et al. (2) noticed that heterozygous BDNF+/– mice segregate in two phenotypes, one obese and one nonobese, which display a differential retention of BDNF mRNA expression in the VMH. Xu et al. (4) showed that in wild-type mice, BDNF mRNA is highly expressed in the VMH, in which its expression is lowered by fasting. Moreover, the same authors showed that the agouti yellow mice (Ay), in which ectopic expression of the agouti protein antagonizes MC4R, as well as the MC4R –/– mice, display a reduced BDNF mRNA expression level in the VMH, compared with wild-type mice. Neurons of the VMH project to many targets, within and outside the hypothalamus, in which TrkB receptors are expressed. Two extrahypothalamic projection sites of the VMH have been implicated in the regulation of food intake: the NTS within the DVC (6, 20, 21, 22) and the periaqueductal gray (19, 23). It has been demonstrated previously that continuous infusion of BDNF in the midbrain near the periaqueductal gray of adult rats also results in a significant but transient weight loss at high doses (60 μg/d), whereas it appears inefficient at a dose of 12 μg/d (19). The results obtained in our study suggest that the DVC is more sensitive to the anorexigenic action of BDNF than the periaqueductal gray.
The mechanism of the progressive reduction of BDNF efficiency in the periaqueductal gray (19) and the DVC (our study) is unknown. A tempting hypothesis could be that continuous treatment with BDNF may result in TrkB receptor down-regulation; however, it does not seem to be the case because previous studies suggest that in the adult brain, the expression of TrkB mRNA is remarkably resistant to modulation by its ligand BDNF. For instance, infusion of BDNF in the periaqueductal gray decreases TrkB receptor protein at the site of infusion, but this is associated with neither altered expression of TrkB mRNA nor attenuation of the analgesic effects of BDNF, suggesting that the observed decrease in TrkB represents receptor turnover (24). Moreover, it has been shown that in BDNF+/– mice, the expression of TrkB mRNA is not altered in the hypothalamus, compared with wild-type mice (2).
Endogenous BDNF protein content in the DVC was decreased by fasting, increased by refeeding, and increased by anorexigenic hormones
We also addressed whether endogenous BDNF within the DVC could be modulated by anorexigenic hormones and nutritional state.
Endogenous BDNF protein content was first measured by Western blotting. In our hands, immunorevelation of BDNF produced a single band corresponding to an apparent molecular mass of 27 kDa, i.e. twice more than that of monomeric mature BDNF (13.5 kDa), and well below the apparent molecular mass of pro-BDNF (35 kDa). We observed that this BDNF-like immunoreactive band was modulated by peripheral leptin or CCK treatment and also nutritional state. We next used a commercially available sandwich ELISA kit for quantifying BDNF unambiguously. These two quantification methods gave similar results in terms of relative changes in BDNF protein content after 48 h fasting and leptin or CCK treatment. Therefore, we interpret the 27-kDa immunoreactive band observed in Western blot as mature BDNF dimers, despite the use of denaturing electrophoresis conditions.
We showed here that within the DVC, BDNF protein content decreased by 40% after 48 h fasting and increased by 31% by refeeding after a 15-h fast. Because infusion of exogenous BDNF in the DVC induces anorexia, the decrease in endogenous BDNF in the same structure when nutritional status is deficient and the increase observed after refeeding strongly suggests that BDNF takes part in the signaling pathways whereby the DVC controls food intake. We also tested the effects of peripheral anorexigenic hormone treatments on BDNF protein content in the DVC. We showed that: 1) a single leptin ip injection induced an increase in BDNF content within the DVC; 2) a peripheral leptin treatment during 3 consecutive days led to a major and long-lasting increase in BDNF protein levels within the DVC; and 3) a single ip CCK injection led to a rapid and transient increase of BDNF protein content in the DVC. These results are consistent with a role of BDNF as an anorexigenic factor within the DVC, downstream of anorexigenic hormones signaling.
BDNF mRNA-expressing cell bodies have been detected in the NTS (10). Therefore, modulating the activity of neural networks in the DVC could result in a local regulation of BDNF synthesis. Several inputs to the DVC could be involved in this process: descending hypothalamic projections, vagal afferents, or a direct humoral action of anorexigenic hormones on DVC neurons. The DVC is indeed a direct target site for leptin (6, 7). Vagal afferents are also responsive to leptin (25, 26) and are a necessary target of CCK-mediated reduction of food intake (27). So leptin, for instance, could modulate neural networks in the DVC by acting directly on this structure or indirectly through the hypothalamus and its descending projections on NTS neurons or through vagal afferents.
Because BDNF is known to be transported either anterogradely (10, 28) or retrogradely (24, 29, 30), several possible sources could contribute to the BDNF protein levels measured in the DVC in the present study. Because vagal efferents have been shown to retrogradely transport BDNF in adult rat (30) and BDNF to be expressed in viscera (31), we cannot exclude that part of BDNF protein observed in the DVC is of peripheral origin. Nevertheless, we are not aware of a modulation of BDNF content in the viscera after leptin treatment or fasting. The VMH is known to project to the NTS (32, 33) and express BDNF in a nutritional state-dependent manner (4). NTS also receives projections from the lateral parabrachial nucleus (32), a structure of dense BDNF mRNA expression (10). In the case of CCK treatment, however, the increase of BDNF protein content in the DVC is far too rapid to consider a possible involvement of anterograde or retrograde transport from a distant source. Conversely, rapid neosynthesis of BDNF is not unprecedented. Indeed, BDNF neosynthesis has been demonstrated to be necessary and sufficient for long-term phrenic facilitation to occur in response to intermittent hypoxia (34), a synaptic plasticity that develops within 15 min (35). Moreover, up-regulation of BDNF mRNA was also observed in the hypothalamus within 15 min during immobilization stress in rat (36). Our data therefore suggest that the increase of BDNF protein content observed in the DVC in response to CCK injection could be due to a local neosynthesis.
Interestingly, it has been shown recently that CCK-mediated suppression of feeding involves the brainstem melanocortin system (9). It is tempting to speculate that activation of the brainstem melanocortin system could be involved in the increase in BDNF protein content observed in this study after CCK treatment. In this hypothesis, the melanocortin signaling would regulate BDNF expression in the DVC, as it does in the VMH (4).
BDNF as a possible common downstream effector of anorexigenic hormones signaling
Leptin and CCK have been shown to reduce food intake in a synergistic manner (for review see Ref.17). This synergistic interaction can occur at the periphery on vagal afferents (26, 37, 38). It can also involve central nervous system sites because intracerebroventricular leptin injections also act in synergy with peripheral CCK (39, 40, 41).
In this study, we have analyzed the BDNF protein content in parallel in the DVC and the whole hypothalamus, in the same rats, after fasting and peripheral treatment by leptin or CCK. Kinetically, we showed that CCK induced an increase of BDNF protein content first in the DVC and later in the hypothalamus. To our knowledge, this is the first time that an increase of BDNF protein content is shown to occur in the hypothalamus after ip CCK injection. However, nerve growth factor was previously shown to be up-regulated in the hypothalamus after CCK treatment through a mechanism involving CCK-A receptors (42). Surprisingly, we could not detect any effect of leptin treatment or fasting on the BDNF protein content in the hypothalamus. A 48-h fasting was previously shown to reduce BDNF mRNA solely in the VMH but having no effect in other hypothalamic nuclei (4). However, when measuring BDNF mRNA in the hypothalamus taken as a whole, Rios et al. (3) did not detect any effect of a 68-h fasting. Thus, the contribution of VMH to the BDNF protein content of the total hypothalamus might have been too small to allow the detection of fasting effect in our experiments. Similarly, genetic models of altered melanocortin signaling display a reduced hypothalamic expression of BDNF mRNA limited to the VMH (4). Therefore, the activation of arcuate POMC neurons by leptin might have been unable to significantly modulate the BDNF protein content in the total hypothalamus for the same reasons as discussed above concerning fasting. Nevertheless, our data point out the quantitative importance of the CCK-induced increase of BDNF protein content in the hypothalamus.
Also noticeable is the sequential manner in which CCK induces an increase of BDNF protein content in the DVC and then in the hypothalamus. Whereas the increase of BDNF protein content observed in the DVC correlates well with the rapid and short-lived CCK-mediated reduction in food intake, one may wonder what role could be played by a delayed increase of BDNF protein content in the hypothalamus. A tempting hypothesis is that this signaling could be involved in the synergistic action of CCK and leptin to decrease body weight. It was shown that CCK treatment enhances not only the magnitude but also the duration of the body weight loss-induced by an intracerebroventricular leptin injection (40, 41). Taken together, our results suggest that BDNF could be a common downstream effector of leptin and CCK in the DVC and probably also in the hypothalamus. We suggest that a convergent signaling of these two hormones could be a mechanism by which they exert a synergistic action on food intake and body weight.
In summary, our results showed that BDNF plays a role as an anorexigenic factor in the DVC in which its protein level is regulated by leptin, CCK, and nutritional state. The increase of BDNF protein content observed after leptin or CCK treatment also points out that BDNF may constitute a common downstream effector of these two hormones and thereby play a role in their synergistic action.
Acknowledgments
We are indebted to Regeneron for the supply of BDNF and Amgen for the supply of leptin. We gratefully thank C. Tardivel for critical reading of the manuscript.
Footnotes
This work was supported by grants from Centre National de la Recherche Scientifique, Institut National de la Recherche Agronomique, University Paul Cezanne-Aix-Marseille III, Conseil General 13 (32/2003), and Conseil Regional Provence-Cte d’Azur (10732/2003). B.B. was supported by a doctoral fellowship from the Ministere de l’Education Nationale, de la Recherche, et de la Technologie.
First Published Online September 15, 2005
Abbreviations: aCSF, Artificial cerebrospinal fluid; BDNF, brain-derived neurotrophic factor; CCK, cholecystokinin; DVC, dorsal vagal complex; MC4R, melanocortin-4 receptor; NTS, nucleus tractus solitarii; POMC, proopiomelanocortin; TrkB, tyrosine kinase receptor type B; VMH, ventromedial hypothalamus.
Accepted for publication September 8, 2005.
References
Pelleymounter MA, Cullen MJ, Wellman CL 1995 Characteristics of BDNF-induced weight loss. Exp Neurol 131:229–238
Kernie SG, Liebl DJ, Parada LF 2000 BDNF regulates eating behavior and locomotor activity in mice. EMBO J 19:1290–1300
Rios M, Fan G, Fekete C, Kelly J, Bates B, Kuehn R, Lechan RM, Jaenisch R 2001 Conditional deletion of brain-derived neurotrophic factor in the postnatal brain leads to obesity and hyperactivity. Mol Endocrinol 15:1748–1757
Xu B, Goulding EH, Zang K, Cepoi D, Cone RD, Jones KR, Tecott LH, Reichardt LF 2003 Brain-derived neurotrophic factor regulates energy balance downstream of melanocortin-4 receptor. Nat Neurosci 6:736–742
Grill HJ, Kaplan JM 2002 The neuroanatomical axis for control of energy balance. Front Neuroendocrinol 23:2–40
Grill HJ, Schwartz MW, Kaplan JM, Foxhall JS, Breininger J, Baskin DG 2002 Evidence that the caudal brainstem is a target for the inhibitory effect of leptin on food intake. Endocrinology 143:239–246
Hosoi T, Kawagishi T, Okuma Y, Tanaka J, Nomura Y 2002 Brain stem is a direct target for leptin’s action in the central nervous system. Endocrinology 143:3498–3504
Williams DL, Kaplan JM, Grill HJ 2000 The role of the dorsal vagal complex and the vagus nerve in feeding effects of melanocortin-3/4 receptor stimulation. Endocrinology 141:1332–1337
Fan W, Ellacott KL, Halatchev IG, Takahashi K, Yu P, Cone RD 2004 Cholecystokinin-mediated suppression of feeding involves the brainstem melanocortin system. Nat Neurosci 7:335–336
Conner JM, Lauterborn JC, Yan Q, Gall CM, Varon S 1997 Distribution of brain-derived neurotrophic factor (BDNF) protein and mRNA in the normal adult rat CNS: evidence for anterograde axonal transport. J Neurosci 17:2295–2313
Yan Q, Radeke MJ, Matheson CR, Talvenheimo J, Welcher AA, Feinstein SC 1997 Immunocytochemical localization of TrkB in the central nervous system of the adult rat. J Comp Neurol 378:135–157
Bariohay B, Lebrun B, Hay M, Moyse E, Jean A 2004 Intraparenchymal infusion of brain-derived neurotrophic factor in the dorsal vagal complex induces anorexia, hypodipsia and weight loss in rat. Gastroenterol 126(Suppl 2):A258
Bariohay B, Lebrun B, Moyse E, Jean A 2005 BDNF plays a role as an anorexigenic factor in the dorsal vagal complex. Auton Neurosci 119:107
Paxinos G, Watson C 1998 The rat brain in stereotaxic coordinates. 4th ed. San Diego: Academic Press
Richy S, Burlet A, Max JP, Burlet C, Beck B 2000 Effect of chronic intraperitoneal injections of leptin on hypothalamic neurotensin content and food intake. Brain Res 862:276–279
Nishiyama M, Makino S, Asaba K, Hashimoto K 1999 Leptin effects on the expression of type-2 CRH receptor mRNA in the ventromedial hypothalamus in the rat. J Neuroendocrinol 11:307–314
Wang L, Barrachina MD, Martinez V, Wei JY, Tache Y 2000 Synergistic interaction between CCK and leptin to regulate food intake. Regul Pept 92:79–85
Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254
Siuciak JA, Boylan C, Fritsche M, Altar CA, Lindsay RM 1996 BDNF increases monoaminergic activity in rat brain following intracerebroventricular or intraparenchymal administration. Brain Res 710:11–20
Swanson LW, Kuypers HG 1980 A direct projection from the ventromedial nucleus and retrochiasmatic area of the hypothalamus to the medulla and spinal cord of the rat. Neurosci Lett 17:307–312
Smith GP 2000 The controls of eating: a shift from nutritional homeostasis to behavioral neuroscience. Nutrition 16:814–820
Faulconbridge LF, Cummings DE, Kaplan JM, Grill HJ 2003 Hyperphagic effects of brainstem ghrelin administration. Diabetes 52:2260–2265
Skultety FM 1966 Changes in caloric intake following brain stem lesions in cats. 3. Effects of lesions of the periaqueductal gray matter and rostral hypothalamus. Arch Neurol 14:670–680
Frank L, Wiegand SJ, Siuciak JA, Lindsay RM, Rudge JS 1997 Effects of BDNF infusion on the regulation of TrkB protein and message in adult rat brain. Exp Neurol 145:62–70
Buyse M, Ovesjo ML, Goiot H, Guilmeau S, Peranzi G, Moizo L, Walker F, Lewin MJ, Meister B, Bado A 2001 Expression and regulation of leptin receptor proteins in afferent and efferent neurons of the vagus nerve. Eur J Neurosci 14:64–72
Gaige S, Abysique A, Bouvier M 2002 Effects of leptin on cat intestinal vagal mechanoreceptors. J Physiol 543(Pt 2):679–689
Smith GP, Gibbs J, Jerome C, Pi-Sunyer FX, Kissileff HR, Thornton J 1981 The satiety effect of cholecystokinin: a progress report. Peptides 2(Suppl 2):57–59
Altar CA, Cai N, Bliven T, Juhasz M, Conner JM, Acheson AL, Lindsay RM, Wiegand SJ 1997 Anterograde transport of brain-derived neurotrophic factor and its role in the brain. Nature 389:856–860
DiStefano PS, Friedman B, Radziejewski C, Alexander C, Boland P, Schick CM, Lindsay RM, Wiegand SJ 1992 The neurotrophins BDNF, NT-3, and NGF display distinct patterns of retrograde axonal transport in peripheral and central neurons. Neuron 8:983–993
Helke CJ, Adryan KM, Fedorowicz J, Zhuo H, Park JS, Curtis R, Radley HE, Distefano PS 1998 Axonal transport of neurotrophins by visceral afferent and efferent neurons of the vagus nerve of the rat. J Comp Neurol 393:102–117
Lommatzsch M, Braun A, Mannsfeldt A, Botchkarev VA, Botchkareva NV, Paus R, Fischer A, Lewin GR, Renz H 1999 Abundant production of brain-derived neurotrophic factor by adult visceral epithelia. Implications for paracrine and target-derived Neurotrophic functions. Am J Pathol 155:1183–1193
Saper CB, Loewy AD 1980 Efferent connections of the parabrachial nucleus in the rat. Brain Res 197:291–317
Canteras NS, Simerly RB, Swanson LW 1994 Organization of projections from the ventromedial nucleus of the hypothalamus: a Phaseolus vulgaris-leucoagglutinin study in the rat. J Comp Neurol 348:41–79
Baker-Herman TL, Fuller DD, Bavis RW, Zabka AG, Golder FJ, Doperalski NJ, Johnson RA, Watters JJ, Mitchell GS 2004 BDNF is necessary and sufficient for spinal respiratory plasticity following intermittent hypoxia. Nat Neurosci 7:48–55
Baker-Herman TL, Mitchell GS 2002 Phrenic long-term facilitation requires spinal serotonin receptor activation and protein synthesis. J Neurosci 22:6239–6246
Rage F, Givalois L, Marmigere F, Tapia-Arancibia L, Arancibia S 2002 Immobilization stress rapidly modulates BDNF mRNA expression in the hypothalamus of adult male rats. Neuroscience 112:309–318
Wang YH, Tache Y, Sheibel AB, Go VL, Wei JY 1997 Two types of leptin-responsive gastric vagal afferent terminals: an in vitro single-unit study in rats. Am J Physiol 273(2 Pt 2):R833–R837
Yuan CS, Attele AS, Dey L, Xie JT 2000 Gastric effects of cholecystokinin and its interaction with leptin on brainstem neuronal activity in neonatal rats. J Pharmacol Exp Ther 295:177–182
Emond M, Schwartz GJ, Ladenheim EE, Moran TH 1999 Central leptin modulates behavioral and neural responsivity to CCK. Am J Physiol 276(5 Pt 2):R1545–R1549
Matson CA, Ritter RC 1999 Long-term CCK-leptin synergy suggests a role for CCK in the regulation of body weight. Am J Physiol 276(4 Pt 2):R1038–R1045
Matson CA, Reid DF, Cannon TA, Ritter RC 2000 Cholecystokinin and leptin act synergistically to reduce body weight. Am J Physiol Regul Integr Comp Physiol 278:R882–R890
Tirassa P, Stenfors C, Lundeberg T, Aloe L 1998 Cholecystokinin-8 regulation of NGF concentrations in adult mouse brain through a mechanism involving CCK(A) and CCK(B) receptors. Br J Pharmacol 123:1230–1236(Bruno Bariohay, Bruno Lebrun, Emmanuel M)
Abstract
Brain-derived neurotrophic factor (BDNF) has recently been implicated as an anorexigenic factor in the central control of food intake. Previous studies focused on the hypothalamus as a probable site of action for this neurotrophin. It was demonstrated that BDNF is an important downstream effector of melanocortin signaling in the ventromedial hypothalamus. In this study, we addressed whether BDNF can modulate food intake in the hindbrain autonomic integrator of food intake regulation, i.e. the dorsal vagal complex (DVC). To this end, we used two complementary methodological approaches in adult rats. First, we measured the effects of intraparenchymal infusions of exogenous BDNF within the DVC on food intake and body weight. Second, we measured the endogenous BDNF protein content in the DVC and hypothalamus after food deprivation, refeeding, or peripheral treatments by the anorexigenic hormones leptin and cholecystokinin (CCK). BDNF infusion within the DVC induced anorexia and weight loss. In the DVC, BDNF protein content decreased after 48 h food deprivation and increased after refeeding. Acute and repetitive peripheral leptin injections induced an increase of the BDNF protein content within the DVC. Moreover, peripheral CCK treatment induced a transient increase of BDNF protein content first in the DVC (30 min after CCK) and later on in the hypothalamus (2 h after CCK). Taken together, these results strongly support the view that BDNF plays a role as an anorexigenic factor in the DVC. Our data also suggest that BDNF may constitute a common downstream effector of leptin and CCK, possibly involved in their synergistic action.
Introduction
RECENT EVIDENCE INDICATES that brain derived neurotrophic factor (BDNF), a member of the neurotrophin family, contributes to food intake and body weight control, acting as an anorexigenic factor in adult rodents. Indeed, infusion of BDNF in the lateral ventricles induces a reduction in food intake associated with weight loss in rats (1). Conversely, mice heterozygous for targeted disruption of BDNF, as well as conditional BDNF mutants, show hyperphagia and obesity (2, 3). Moreover, the same phenotype was observed in mice with a reduced expression of BDNF high-affinity tyrosine kinase receptor type B (TrkB) at a quarter of the normal amount (4). Mechanistic interpretations of these data all focused on the hypothalamus, which is indeed the best documented among autonomic centers involved in energy homeostasis integration. Xu et al. (4) demonstrated that BDNF is expressed at high levels in the ventromedial hypothalamus (VMH), in which its expression is regulated by nutritional state and melanocortin-4 receptor (MC4R) signaling. In addition, the same authors showed that intracerebroventricular infusion of BDNF suppresses the hyperphagia and excessive weight gain observed on high-fat diets in mice with deficient MC4R signaling, suggesting that BDNF functions as a downstream effector through which MC4R signaling regulates energy balance.
However, the dorsal vagal complex (DVC) of hindbrain constitutes another autonomic integrator of food intake regulation. The DVC is located in the caudal brainstem and comprises three structures, namely the nucleus tractus solitarii (NTS), area postrema, and dorsal motor nucleus of the vagus nerve. The DVC contains leptin and insulin receptors, glucose-sensing mechanisms, and neuropeptide mediators relevant to energy balance (5). The food intake-inhibiting hormone leptin has been shown to act directly within the DVC (6, 7). It has also been demonstrated that intraparenchymal injections of MC4R agonists or antagonists in the DVC, respectively, reduce and enhance food intake (8), consistent with the facts that the MC4R has its highest density in the DVC, and the NTS is one of the two structures in the brain containing proopiomelanocortin (POMC) neurons. More recently it was shown that brainstem NTS neurons expressing POMC are activated by the hormone cholecystokinin (CCK) and that activation of the brainstem neuronal MC4R is required for CCK-induced suppression of feeding (9). The DVC was also shown to contain high densities of BDNF and TrkB (10, 11). In this study, we addressed the question of whether BDNF can modulate food intake directly inside the DVC. We demonstrated that intraparenchymal infusions of exogenous BDNF in the DVC induced anorexia and weight loss and that endogenous BDNF protein content in the same structure was regulated by nutritional state as well as by peripheral leptin. Moreover, we showed that peripheral CCK-8S treatment induced a transient increase of BDNF protein content first in the DVC and later in the hypothalamus. Preliminary accounts of these results have been published in abstract form [Digestive Disease Week 2004 (12) and International Society of Autonomic Neuroscience 2005 (13)].
Materials and Methods
Animals and housing conditions
Adult male Wistar Han rats from Charles River Laboratories (L’Arbresle, France), weighting 290 ± 10 g at the time of experiments, were housed singly in metal hanging cages at constant temperature (25 C) under a 12-h light, 12-h dark cycle with lights off at 1800 h. Water and standard pellet food (pellet AO4, UAR, Villemoisson-sur-Orge, France) were available ad libitum unless otherwise noted. All rats were habituated to the hanging cage and to handling for at least 5 d before the experiments. An additional 5-d habituation period with a daily ip injection of saline was performed when necessary. All procedures were performed in accordance with Principles of Laboratory Animal Care (National Institutes of Health publication 86-23, revised 1985) as well as with French national law.
Experiment 1: intracerebral infusion of exogenous BDNF
The human recombinant BDNF used in this study was a kind gift from Regeneron (Tarrytown, NY). Osmotic pumps (model 1002, 0.25 μl/h for 14 d; Alzet, Cupertino, CA) were filled immediately before implantation under sterile atmosphere, with either human recombinant BDNF (167 ng/μl BDNF for 1 μg/d treatment, n = 7 and 16.7 ng/μl BDNF for 0.1 μg/d treatment, n = 3) diluted in artificial cerebrospinal fluid (aCSF) [mixture 1:1 of the A and B solutions; solution A: 296.4 mM NaCl, 6 mM KCl, 2.8 mM CaCl22H2O, 1.6 mM MgCl26H2O; solution B: 1.6 mM Na2HPO47H2O, 0.4 mM NaH2PO4H2O (pH 7.4)] or aCSF alone (n = 4). Osmotic pumps were then connected to a cannula (model 330OP/DW/Spc, 30G; Plastics One, Roanoke, VA,) with 6 cm of vinyl tubing (model C312VT; Plastics One) filled with the same solution. Rats weighting 290 ± 10 g received a sc injection of atropine (0.18 mg/kg) and were anesthetized 30 min later with an ip injection of ketamine and xylazine (100 and 15 mg/kg ip, respectively). The tip of the infusion cannula was implanted stereotaxically into the DVC in reference to the stereotaxic atlas of Paxinos and Watson (area postrema: –5 mm from interaural line; lateral: –1.3 mm from midline; dorsoventral: 1.8 mm above interaural line) (14), and the cannula was cemented to three jeweler’s screws attached to the skull. The osmotic pump was then placed into a sc pocket in the subscapular area. Rats recovered for 1 d in individual thermostated cage with food and water ad libitum and were then placed back in their individual hanging cages.
Weight, food, and water intake measurements.
Body weight and food and water intakes were measured every day. Rats were removed from their cages 1 h before lights off (1700 h) and weighed. Food hoppers, food debris collected under each cage, and water bottles were weighed, and daily food and water intake values were determined. Rats were returned to their cages and supplied with fresh preweighed food and water at the onset of the dark period.
Postmortem localization of infusion sites
On the 15th day after implantation, rats were anesthetized, osmotic pumps were removed, and the volume remaining in each pump was determined. Infusion sites were labeled by injecting 0.8 μl pontamine blue 2% through the cannula (the estimated dead volume of the cannula and connecting tube was 0.6 μl). Rats were perfused transcardially with 150 ml 0.1 M PBS and then fixed with 250 ml of a solution of paraformaldehyde 4% (in phosphate buffer 0.1 M at 4 C). Brains were removed, postfixed 1 h in paraformaldehyde 4% in PBS 0.1 M at 4 C, and then cryoprotected 24–48 h in a solution of sucrose 30% (in PBS at 4 C). After sedimentation in sucrose, brains were frozen by immersion in isopentane at –40 C for 20 sec and then stored at –80 C. Thirty-micrometer sections were cut in the coronal plane, and the cannula tip location was determined.
Experiment 2: modulation of the endogenous BDNF protein content
Food deprivation.
Rats were maintained on food pellets and water ad libitum (control) or food deprived for 48 h, with water ad libitum. DVCs and hypothalami were collected as described below. The BDNF protein content of the collected samples was analyzed by either Western blotting only (n = 6 per group) or both Western blotting and ELISA (n = 6 per group), as described below.
Refeeding.
Rats (n = 12) were given access to chow at 0900–1000 and 1400–1800 h daily for 5 d before the experiment, according to a previously published method (9). On the day of the experiment, one group (refeed, n = 6) received chow at 0900–1000 h, and the other group (control, n = 6) did not. Rats were killed at 1000 h, and DVCs and hypothalami were collected and their BDNF protein content analyzed by ELISA, as described below.
Leptin injections
The murine leptin used in this study was a kind gift from Amgen (Thousand Oaks, CA). Leptin dose was chosen to produce a significant but moderate change in body weight gain, within a repetitive treatment protocol. We used a protocol previously described to reduce body weight gain in Long-Evans rats (15) and verified its effects on body weight gain in Wistar Han rats.
For acute treatment, rats fed ad libitum (290 ± 10 g) were injected with leptin (0.2 mg/rat; ip, n = 6) or saline (n = 6) and were killed 6 h after injection, given the slow onset of leptin-induced decrease in food intake (16 ; for review see also Ref.17). DVCs and hypothalami were collected and their BDNF protein content analyzed by ELISA, as described below.
For repetitive treatment, rats fed ad libitum (290 ± 10 g) were injected daily (30 min before the end of the light period) with saline (control) or leptin (0.2 mg/rat; ip), during 3 d. Rats were killed 20 h after the last injection. In a first series of experiments, only the DVCs were collected and submitted to Western blot analysis (n = 6 per group). In a second series of experiments, DVCs and hypothalami were collected and their BDNF protein content analyzed by both Western blotting and ELISA, as described below (n = 6 per group). These rats were also weighed daily at 1400 h to assess the effects of leptin treatment on body weight gain.
CCK-8S injections
Rats fed ad libitum (290 ± 10 g) were injected with saline or CCK-8S (6 μg/kg; ip; Sigma, Saint Quentin Fallavier, France) and killed 30 min, 2 h, or 20 h after injection (six groups; n = 6 per group). DVCs and hypothalami were collected and their BDNF protein content analyzed by both Western blotting and ELISA, as described below.
DVC and hypothalamus dissection
Rats were anesthetized with halothane and decapitated, the brain was removed from the skull, and the hypothalamus was dissected out under binocular control. The brainstem and upper cervical spinal cord were removed rapidly and glued to the cutting stage of a vibratome. Sectioning of 500-μm slices and microdissection of DVC from the slices of interest, under binocular control, were performed in chilled aCFS. Tissue samples (DVC and hypothalamus) were placed in microcentrifuge tubes, immediately frozen in liquid nitrogen, and stored at –80 C until use. For each rat, the total amount of time spent from killing to freezing of sample was 7 min, 15 sec ± 10 sec for the hypothalamus and 16 min, 02 sec ± 13 sec for the DVC. To minimize circadian variations of BDNF in our experiments, test and control rats were killed at equal daytime on alternate days, and whole groups were processed within less than 2 h.
Protein extraction and dosage
Tissue samples were homogenized in lysis buffer (100 μl for DVC, 500 μl for hypothalamus) [NaCl 137 mM; Tris-HCl 20 mM; Triton X-100 1%, glycerol 10%, sodium orthovanadate 1 mM, phenylmethylsulfonyl fluoride 1 mM, protease inhibitors cocktail (Sigma)], incubated 30 min at 4 C, and centrifuged at 15,000 x g (30 min at 4 C). The supernatants were collected and protein concentration determined by the Bradford’s procedure (18).
Protein gel electrophoresis and immunoblots
Protein extracts were mixed with sodium dodecyl sulfate sample buffer (sample buffer Laemmli; Sigma), denatured for 3 min at 100 C, and transferred on ice. Equal amounts of protein extracts were separated by electrophoresis in 12% polyacrylamide gels containing sodium dodecyl sulfate and transferred on a nitrocellulose membrane (Immobilon-P transfer membranes; Millipore, Molsheim, France). The BenchMark prestained protein ladder (Invitrogen, Cergy Pontoise, France) was used for determination of apparent molecular mass. Because BDNF immunorevelation gave no signal at an apparent molecular mass superior to 27 kDa, the nitrocellulose membranes were subsequently cut through a line corresponding to an apparent molecular mass of 37.1 kDa for a separate BDNF and -actin immunorevelation. BDNF and -actin were revealed separately on their corresponding membrane parts by incubation with anti-BDNF rabbit polyclonal IgG (1:750; N-20; Santa Cruz Biotechnology, Santa Cruz, CA) or anti--actin monoclonal mouse IgG2 (1:2000; Sigma), followed by horseradish peroxidase-conjugated secondary antibody (goat antirabbit IgG for BDNF and goat antimouse IgG for -actin; 1:2000; Dako, Trappes, France). After rinsing with buffer, the immunocomplexes were detected by chemiluminescence using the enhanced chemiluminescence kit (Amersham Biosciences, Freiburg, Germany) according to the manufacturer’s instructions. The film signals were digitally scanned and then quantified using Quantity One software (Bio-Rad Laboratories, Marnes la Coquette, France). The -actin was used as internal standard for normalization.
BDNF immunoassay
BDNF protein was also measured with a conventional two-site ELISA. BDNF Emax immunoassay (Promega, Charbonniere, France) was performed according to the manufacturer’s protocol. For each sample, 30 μg of total proteins were used to determine BDNF content. The sensitivity of the assay was 15 pg/ml, and the cross-reactivity with other related neurotrophic factors was less than 3%. The within-assay variability was less than 3%, and all relevant comparisons were made within the same assay. BDNF concentration was determined as picograms per milligram of total protein, and relative change in BDNF content between experimental groups expressed as percentage of control values.
Statistical analysis
Statistical analyses were performed using STATView (version 5.0.1; Statview Software, Cary, NC). Body weight, food and water intake were analyzed using a two-way, repeated-measures ANOVA in which difference from presurgical weight or food or water intake was the within-groups variable, and dose was the between-groups variable. A simple-way ANOVA was used to assess the effects of food deprivation, refeeding, and leptin treatments on BDNF protein content. A two-way ANOVA was used to assess the effects of CCK treatment on BDNF protein content and the effects of repetitive leptin treatment on cumulative body weight gain. In all cases, when a significant main effect was present, Fischer test was used for post hoc analyses. All data are expressed as mean ± SEM.
Results
Exogenous BDNF infusion in the DVC
Data for 14 rats with verified infusion site in the DVC (see Fig. 1) and pump functioning are included in these analyses. Data from three rats were excluded based on cannula localization outside the DVC (n = 2) or a disconnected pump (n = 1).
Body weight
Control rats infused with aCSF showed a linear growth (4.87 g/d, R2 = 0.998) throughout the infusion period. Intraparenchymal infusion of BDNF in the DVC reduced the cumulative weight gain in a dose- (F2 = 12.89; P = 0.0013) and time- (F13 = 30.81; P < 0.0001) dependent manner, as shown in Fig. 2A. There was a significant interaction between dose and time (F26 = 3.99; P < 0.0001). At the highest BDNF dose tested, body weight first decreased gradually, reaching its minimum 5 d after the beginning of the infusion period and then increased linearly at a rate remaining lower than that of control rats (2.49 g/d, R2 = 0.9598). At a dose of 0.1 μg/d, BDNF induced a reduction of growth rate until the fourth postimplant day, followed by a linear growth comparable with that of control rats (4.63 g/d, R2 = 0.9977). Simple-effects analysis of the interaction between time and dose revealed that infusion of BDNF significantly decreased cumulative body weight gain, compared with controls, from the second postimplant day to the end of the infusion period at a dose of 1 μg/d and from the fifth postimplant day to the end of the infusion period at a dose of 0.1 μg/d.
Food and water intakes
Control rats infused with aCSF showed a relatively constant food intake over the infusion period (24.41 ± 0.71 g/d). Intraparenchymal infusion of BDNF in the DVC reduced the daily food intake in a dose- (F2 = 19.51; P = 0.0002) and time- (F13 = 2.71; P < 0.0001) dependent manner, as can be seen in Fig. 2B. There was also a significant interaction between time and dose (F26 = 2.52; P = 0.0005). Simple-effects analysis of this interaction revealed a significant anorexigenic effect of BDNF, compared with control, from the first postimplant measurement of food intake, to the eighth day of infusion for the 0.1 μg/d dose and until the end of the infusion period for the 1 μg/d dose. The anorexigenic effect of BDNF was highest on the first measurement day (second postimplant day) (daily food intake reduced by –54.7% for BDNF 0.1 μg/d and –67.4% for BDNF 1 μg/d) and decreased gradually over time.
Control rats infused with aCSF showed a relatively constant water intake over the infusion period (29.22 ± 0.48 g/d). Although the mean daily water intake of BDNF-treated rats appeared lower than that of controls (23.26 ± 3.24 g/d for the 1 μg/d BDNF dose; 21.07 ± 3.12 g/d for the 0.1 μg/d BDNF dose), the repeated-measures ANOVA analysis did not show any significant effect of treatment on water intake.
The above results show that intraparenchymal infusion of exogenous BDNF in the DVC induced anorexia and weight loss in rats, suggesting that this neurotrophin acts as an anorexigenic signaling factor within the DVC.
Regulation of endogenous BDNF protein content
To further evaluate the implication of BDNF in the regulation of food intake within the DVC, we next asked whether its endogenous protein content could be modulated by nutritional status and peripheral injections of leptin or CCK-8S.
Endogenous BDNF protein content in DVC and hypothalamus were first measured by Western blotting. Immunorevelation of BDNF produced a single band corresponding to an apparent molecular mass of 27 kDa, i.e. twice the size of monomeric mature BDNF (13.5 kDa), and well below the apparent molecular mass of pro-BDNF (35 kDa). We next used a commercially available sandwich ELISA kit for quantifying BDNF unambiguously (BDNF Emax Immunoassay System, Promega). These two quantification methods gave similar results in terms of relative changes in BDNF protein content both after a 48-h fasting, and after leptin or CCK-8S treatment. Results presented in Figs. 3–5 were obtained by ELISA. Results obtained by Western blotting are shown in Fig. 6 and Table 1 for comparison.
Effect of 48 h fasting and refeeding on BDNF content in DVC and hypothalamus
Food deprivation for 48 h led to a statistically significant reduction in BDNF protein content in the DVC when compared with rats fed ad libitum: 100 ± 14.93% for control (n = 6) vs. 63 ± 3.16% for fasted rats (n = 6) (F1 = 9.26; P = 0.03) (see Fig. 3A). By contrast, there was no significant effect of a 48 h fasting on BDNF protein content in the hypothalamus taken as a whole: 100 ± 6.88% for control (n = 6) vs. 99.87 ± 6.76% for fasted rats (n = 6) (data not shown). Moreover, refeeding during 1 h after a 15-h fast induced a statistically significant increase in BDNF protein content in the DVC when compared with rats fasted during 16 h (control group): 100 ± 3.87% for control (n = 6) vs. 131.06 ± 11.47% for refed rats (n = 6) (F1 = 6.586; P = 0.0281) (see Fig. 3B). By contrast, BDNF protein content in the hypothalamus was not affected by refeeding: 100 ± 5.26% for control (n = 5; one sample was lost during freezing) vs. 108.40 ± 7.04% for refed rats (n = 6) (data not shown).
Effect of acute and repeated leptin injections on BDNF content in DVC and hypothalamus
In our hands, repetitive daily leptin injections (0.2 mg/rat per day ip) induced a reduction of cumulative body weight gain that became statistically significant only after the third treatment day, compared with control: 8.6 ± 1.26 g for leptin-treated rats (n = 6) vs. 13.2 ± 0.54 g for saline-treated rats (n = 6) (F1 = 8.138; P = 0.0139) (data not shown).
Single peripheral leptin injection (0.2 mg/rat ip) induced a statistically significant increase in BDNF protein content in the DVC when measured 6 h after injection, compared with control: 100 ± 4.52% for control (n = 6) vs. 137.56 ± 15.5% for leptin-treated rats (n = 6) (F1 = 5.408; P = 0.0424) (see Fig. 4A). By contrast, BDNF content measured in the whole hypothalamus from the same rats was not modified by leptin treatment: 100 ± 5.40% for control (n = 6) vs. 90.08 ± 5.89% for leptin-treated rats (n = 6) (data not shown).
Cumulative leptin treatment (0.2 mg/rat per day ip) for 3 consecutive days led to a highly significant increase in BDNF protein content in the DVC, as measured 20 h after the last injection, compared with control: 175.25 ± 5.04% for leptin-treated rats (n = 6) vs.100 ± 2.84% for saline-treated controls (n = 6) (F1 = 143.40; P < 0.0001) (see Fig. 4B). Conversely, BDNF content measured in the whole hypothalamus from the same rats was not modified by leptin treatment: 100 ± 4.88% for control (n = 6) vs. 103.08 ± 9.71% for leptin-treated rats (n = 6) (data not shown).
Effect of a single CCK-8S ip injection on BDNF content in DVC and hypothalamus
An ip injection of CCK-8S (6 μg/kg) in rats fed ad libitum led to a rapid and transient increase in BDNF protein content in the DVC. Thirty minutes after injection, BDNF levels were significantly higher in CCK-8S-treated rats (125.20 ± 4.59%; n = 6), compared with saline-treated controls (100 ± 2.95%; n = 6) (F1 = 21.40; P < 0.001) (Fig. 5A), whereas there were no significant differences at later time points (2 h and 20 h).
In the hypothalamus, CCK-8S treatment did not modulate BDNF protein content 30 min after injection but led to a significant increase 2 h after injection [190.51 ± 8.72% for CCK-8S-treated rats (n = 6) vs. 100 ± 6.94% for saline-treated rats (n = 6) (F1 = 65.89; P < 0.0001)]. This increase in BDNF levels was also transient because there were no statistically significant differences between CCK-8S- and saline-treated rats 20 h after injection.
Discussion
In this study we provide for the first time evidence for a role of BDNF as an anorexigenic factor in the DVC of the adult rat. First, we showed that intraparenchymal infusion of exogenous BDNF within the DVC induced a dose-dependent anorexia. Second, we showed that endogenous BDNF protein content in the DVC decreased after fasting, increased after refeeding, and increased after peripheral leptin or CCK treatment.
Exogenous BDNF infusion in the DVC-induced anorexia
Intraparenchymal infusion of BDNF at a dose of 0.1 μg/d resulted in a transient anorexia, with an initial 54% decrease in daily food intake, compared with aCSF-treated rats, and an overall 30% reduction in the cumulative food intake measured over the first 8 d of treatment. It should be noted that, whereas BDNF-treated rats did not eat significantly less than control rats during the remaining infusion period, their weight curve tended to be parallel to, but below, that of control rats. We interpret the return of food intake after 8 d of treatment to, but not above, control levels, as implying that BDNF had reduced but not totally lost its effects on food intake at this point. At a higher dose (1 μg/d), the anorexigenic effect of BDNF was more severe and remained significant all along the infusion period, with an initial 67% decrease in daily food intake, followed by a gradual decrease in severity, resulting in an overall 43% reduction of cumulative food intake over the 14 d of infusion and a transient weight loss for the first 5 d of infusion. Therefore, our results show that BDNF causes anorexia and weight loss when infused in the DVC, albeit in a less severe manner than that observed by others after infusion in the lateral ventricle (1, 19). Thus, the TrkB receptors within the DVC appear as one of the targets at which BDNF exert its anorexigenic signaling.
Other target sites for BDNF/TrkB anorexigenic signaling have been pointed out by others. In the hypothalamus, several lines of evidence have identified the VMH as a key site in which BDNF expression is modulated by nutritional state and MC4R signaling (2, 4). Kernie et al. (2) noticed that heterozygous BDNF+/– mice segregate in two phenotypes, one obese and one nonobese, which display a differential retention of BDNF mRNA expression in the VMH. Xu et al. (4) showed that in wild-type mice, BDNF mRNA is highly expressed in the VMH, in which its expression is lowered by fasting. Moreover, the same authors showed that the agouti yellow mice (Ay), in which ectopic expression of the agouti protein antagonizes MC4R, as well as the MC4R –/– mice, display a reduced BDNF mRNA expression level in the VMH, compared with wild-type mice. Neurons of the VMH project to many targets, within and outside the hypothalamus, in which TrkB receptors are expressed. Two extrahypothalamic projection sites of the VMH have been implicated in the regulation of food intake: the NTS within the DVC (6, 20, 21, 22) and the periaqueductal gray (19, 23). It has been demonstrated previously that continuous infusion of BDNF in the midbrain near the periaqueductal gray of adult rats also results in a significant but transient weight loss at high doses (60 μg/d), whereas it appears inefficient at a dose of 12 μg/d (19). The results obtained in our study suggest that the DVC is more sensitive to the anorexigenic action of BDNF than the periaqueductal gray.
The mechanism of the progressive reduction of BDNF efficiency in the periaqueductal gray (19) and the DVC (our study) is unknown. A tempting hypothesis could be that continuous treatment with BDNF may result in TrkB receptor down-regulation; however, it does not seem to be the case because previous studies suggest that in the adult brain, the expression of TrkB mRNA is remarkably resistant to modulation by its ligand BDNF. For instance, infusion of BDNF in the periaqueductal gray decreases TrkB receptor protein at the site of infusion, but this is associated with neither altered expression of TrkB mRNA nor attenuation of the analgesic effects of BDNF, suggesting that the observed decrease in TrkB represents receptor turnover (24). Moreover, it has been shown that in BDNF+/– mice, the expression of TrkB mRNA is not altered in the hypothalamus, compared with wild-type mice (2).
Endogenous BDNF protein content in the DVC was decreased by fasting, increased by refeeding, and increased by anorexigenic hormones
We also addressed whether endogenous BDNF within the DVC could be modulated by anorexigenic hormones and nutritional state.
Endogenous BDNF protein content was first measured by Western blotting. In our hands, immunorevelation of BDNF produced a single band corresponding to an apparent molecular mass of 27 kDa, i.e. twice more than that of monomeric mature BDNF (13.5 kDa), and well below the apparent molecular mass of pro-BDNF (35 kDa). We observed that this BDNF-like immunoreactive band was modulated by peripheral leptin or CCK treatment and also nutritional state. We next used a commercially available sandwich ELISA kit for quantifying BDNF unambiguously. These two quantification methods gave similar results in terms of relative changes in BDNF protein content after 48 h fasting and leptin or CCK treatment. Therefore, we interpret the 27-kDa immunoreactive band observed in Western blot as mature BDNF dimers, despite the use of denaturing electrophoresis conditions.
We showed here that within the DVC, BDNF protein content decreased by 40% after 48 h fasting and increased by 31% by refeeding after a 15-h fast. Because infusion of exogenous BDNF in the DVC induces anorexia, the decrease in endogenous BDNF in the same structure when nutritional status is deficient and the increase observed after refeeding strongly suggests that BDNF takes part in the signaling pathways whereby the DVC controls food intake. We also tested the effects of peripheral anorexigenic hormone treatments on BDNF protein content in the DVC. We showed that: 1) a single leptin ip injection induced an increase in BDNF content within the DVC; 2) a peripheral leptin treatment during 3 consecutive days led to a major and long-lasting increase in BDNF protein levels within the DVC; and 3) a single ip CCK injection led to a rapid and transient increase of BDNF protein content in the DVC. These results are consistent with a role of BDNF as an anorexigenic factor within the DVC, downstream of anorexigenic hormones signaling.
BDNF mRNA-expressing cell bodies have been detected in the NTS (10). Therefore, modulating the activity of neural networks in the DVC could result in a local regulation of BDNF synthesis. Several inputs to the DVC could be involved in this process: descending hypothalamic projections, vagal afferents, or a direct humoral action of anorexigenic hormones on DVC neurons. The DVC is indeed a direct target site for leptin (6, 7). Vagal afferents are also responsive to leptin (25, 26) and are a necessary target of CCK-mediated reduction of food intake (27). So leptin, for instance, could modulate neural networks in the DVC by acting directly on this structure or indirectly through the hypothalamus and its descending projections on NTS neurons or through vagal afferents.
Because BDNF is known to be transported either anterogradely (10, 28) or retrogradely (24, 29, 30), several possible sources could contribute to the BDNF protein levels measured in the DVC in the present study. Because vagal efferents have been shown to retrogradely transport BDNF in adult rat (30) and BDNF to be expressed in viscera (31), we cannot exclude that part of BDNF protein observed in the DVC is of peripheral origin. Nevertheless, we are not aware of a modulation of BDNF content in the viscera after leptin treatment or fasting. The VMH is known to project to the NTS (32, 33) and express BDNF in a nutritional state-dependent manner (4). NTS also receives projections from the lateral parabrachial nucleus (32), a structure of dense BDNF mRNA expression (10). In the case of CCK treatment, however, the increase of BDNF protein content in the DVC is far too rapid to consider a possible involvement of anterograde or retrograde transport from a distant source. Conversely, rapid neosynthesis of BDNF is not unprecedented. Indeed, BDNF neosynthesis has been demonstrated to be necessary and sufficient for long-term phrenic facilitation to occur in response to intermittent hypoxia (34), a synaptic plasticity that develops within 15 min (35). Moreover, up-regulation of BDNF mRNA was also observed in the hypothalamus within 15 min during immobilization stress in rat (36). Our data therefore suggest that the increase of BDNF protein content observed in the DVC in response to CCK injection could be due to a local neosynthesis.
Interestingly, it has been shown recently that CCK-mediated suppression of feeding involves the brainstem melanocortin system (9). It is tempting to speculate that activation of the brainstem melanocortin system could be involved in the increase in BDNF protein content observed in this study after CCK treatment. In this hypothesis, the melanocortin signaling would regulate BDNF expression in the DVC, as it does in the VMH (4).
BDNF as a possible common downstream effector of anorexigenic hormones signaling
Leptin and CCK have been shown to reduce food intake in a synergistic manner (for review see Ref.17). This synergistic interaction can occur at the periphery on vagal afferents (26, 37, 38). It can also involve central nervous system sites because intracerebroventricular leptin injections also act in synergy with peripheral CCK (39, 40, 41).
In this study, we have analyzed the BDNF protein content in parallel in the DVC and the whole hypothalamus, in the same rats, after fasting and peripheral treatment by leptin or CCK. Kinetically, we showed that CCK induced an increase of BDNF protein content first in the DVC and later in the hypothalamus. To our knowledge, this is the first time that an increase of BDNF protein content is shown to occur in the hypothalamus after ip CCK injection. However, nerve growth factor was previously shown to be up-regulated in the hypothalamus after CCK treatment through a mechanism involving CCK-A receptors (42). Surprisingly, we could not detect any effect of leptin treatment or fasting on the BDNF protein content in the hypothalamus. A 48-h fasting was previously shown to reduce BDNF mRNA solely in the VMH but having no effect in other hypothalamic nuclei (4). However, when measuring BDNF mRNA in the hypothalamus taken as a whole, Rios et al. (3) did not detect any effect of a 68-h fasting. Thus, the contribution of VMH to the BDNF protein content of the total hypothalamus might have been too small to allow the detection of fasting effect in our experiments. Similarly, genetic models of altered melanocortin signaling display a reduced hypothalamic expression of BDNF mRNA limited to the VMH (4). Therefore, the activation of arcuate POMC neurons by leptin might have been unable to significantly modulate the BDNF protein content in the total hypothalamus for the same reasons as discussed above concerning fasting. Nevertheless, our data point out the quantitative importance of the CCK-induced increase of BDNF protein content in the hypothalamus.
Also noticeable is the sequential manner in which CCK induces an increase of BDNF protein content in the DVC and then in the hypothalamus. Whereas the increase of BDNF protein content observed in the DVC correlates well with the rapid and short-lived CCK-mediated reduction in food intake, one may wonder what role could be played by a delayed increase of BDNF protein content in the hypothalamus. A tempting hypothesis is that this signaling could be involved in the synergistic action of CCK and leptin to decrease body weight. It was shown that CCK treatment enhances not only the magnitude but also the duration of the body weight loss-induced by an intracerebroventricular leptin injection (40, 41). Taken together, our results suggest that BDNF could be a common downstream effector of leptin and CCK in the DVC and probably also in the hypothalamus. We suggest that a convergent signaling of these two hormones could be a mechanism by which they exert a synergistic action on food intake and body weight.
In summary, our results showed that BDNF plays a role as an anorexigenic factor in the DVC in which its protein level is regulated by leptin, CCK, and nutritional state. The increase of BDNF protein content observed after leptin or CCK treatment also points out that BDNF may constitute a common downstream effector of these two hormones and thereby play a role in their synergistic action.
Acknowledgments
We are indebted to Regeneron for the supply of BDNF and Amgen for the supply of leptin. We gratefully thank C. Tardivel for critical reading of the manuscript.
Footnotes
This work was supported by grants from Centre National de la Recherche Scientifique, Institut National de la Recherche Agronomique, University Paul Cezanne-Aix-Marseille III, Conseil General 13 (32/2003), and Conseil Regional Provence-Cte d’Azur (10732/2003). B.B. was supported by a doctoral fellowship from the Ministere de l’Education Nationale, de la Recherche, et de la Technologie.
First Published Online September 15, 2005
Abbreviations: aCSF, Artificial cerebrospinal fluid; BDNF, brain-derived neurotrophic factor; CCK, cholecystokinin; DVC, dorsal vagal complex; MC4R, melanocortin-4 receptor; NTS, nucleus tractus solitarii; POMC, proopiomelanocortin; TrkB, tyrosine kinase receptor type B; VMH, ventromedial hypothalamus.
Accepted for publication September 8, 2005.
References
Pelleymounter MA, Cullen MJ, Wellman CL 1995 Characteristics of BDNF-induced weight loss. Exp Neurol 131:229–238
Kernie SG, Liebl DJ, Parada LF 2000 BDNF regulates eating behavior and locomotor activity in mice. EMBO J 19:1290–1300
Rios M, Fan G, Fekete C, Kelly J, Bates B, Kuehn R, Lechan RM, Jaenisch R 2001 Conditional deletion of brain-derived neurotrophic factor in the postnatal brain leads to obesity and hyperactivity. Mol Endocrinol 15:1748–1757
Xu B, Goulding EH, Zang K, Cepoi D, Cone RD, Jones KR, Tecott LH, Reichardt LF 2003 Brain-derived neurotrophic factor regulates energy balance downstream of melanocortin-4 receptor. Nat Neurosci 6:736–742
Grill HJ, Kaplan JM 2002 The neuroanatomical axis for control of energy balance. Front Neuroendocrinol 23:2–40
Grill HJ, Schwartz MW, Kaplan JM, Foxhall JS, Breininger J, Baskin DG 2002 Evidence that the caudal brainstem is a target for the inhibitory effect of leptin on food intake. Endocrinology 143:239–246
Hosoi T, Kawagishi T, Okuma Y, Tanaka J, Nomura Y 2002 Brain stem is a direct target for leptin’s action in the central nervous system. Endocrinology 143:3498–3504
Williams DL, Kaplan JM, Grill HJ 2000 The role of the dorsal vagal complex and the vagus nerve in feeding effects of melanocortin-3/4 receptor stimulation. Endocrinology 141:1332–1337
Fan W, Ellacott KL, Halatchev IG, Takahashi K, Yu P, Cone RD 2004 Cholecystokinin-mediated suppression of feeding involves the brainstem melanocortin system. Nat Neurosci 7:335–336
Conner JM, Lauterborn JC, Yan Q, Gall CM, Varon S 1997 Distribution of brain-derived neurotrophic factor (BDNF) protein and mRNA in the normal adult rat CNS: evidence for anterograde axonal transport. J Neurosci 17:2295–2313
Yan Q, Radeke MJ, Matheson CR, Talvenheimo J, Welcher AA, Feinstein SC 1997 Immunocytochemical localization of TrkB in the central nervous system of the adult rat. J Comp Neurol 378:135–157
Bariohay B, Lebrun B, Hay M, Moyse E, Jean A 2004 Intraparenchymal infusion of brain-derived neurotrophic factor in the dorsal vagal complex induces anorexia, hypodipsia and weight loss in rat. Gastroenterol 126(Suppl 2):A258
Bariohay B, Lebrun B, Moyse E, Jean A 2005 BDNF plays a role as an anorexigenic factor in the dorsal vagal complex. Auton Neurosci 119:107
Paxinos G, Watson C 1998 The rat brain in stereotaxic coordinates. 4th ed. San Diego: Academic Press
Richy S, Burlet A, Max JP, Burlet C, Beck B 2000 Effect of chronic intraperitoneal injections of leptin on hypothalamic neurotensin content and food intake. Brain Res 862:276–279
Nishiyama M, Makino S, Asaba K, Hashimoto K 1999 Leptin effects on the expression of type-2 CRH receptor mRNA in the ventromedial hypothalamus in the rat. J Neuroendocrinol 11:307–314
Wang L, Barrachina MD, Martinez V, Wei JY, Tache Y 2000 Synergistic interaction between CCK and leptin to regulate food intake. Regul Pept 92:79–85
Bradford MM 1976 A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 72:248–254
Siuciak JA, Boylan C, Fritsche M, Altar CA, Lindsay RM 1996 BDNF increases monoaminergic activity in rat brain following intracerebroventricular or intraparenchymal administration. Brain Res 710:11–20
Swanson LW, Kuypers HG 1980 A direct projection from the ventromedial nucleus and retrochiasmatic area of the hypothalamus to the medulla and spinal cord of the rat. Neurosci Lett 17:307–312
Smith GP 2000 The controls of eating: a shift from nutritional homeostasis to behavioral neuroscience. Nutrition 16:814–820
Faulconbridge LF, Cummings DE, Kaplan JM, Grill HJ 2003 Hyperphagic effects of brainstem ghrelin administration. Diabetes 52:2260–2265
Skultety FM 1966 Changes in caloric intake following brain stem lesions in cats. 3. Effects of lesions of the periaqueductal gray matter and rostral hypothalamus. Arch Neurol 14:670–680
Frank L, Wiegand SJ, Siuciak JA, Lindsay RM, Rudge JS 1997 Effects of BDNF infusion on the regulation of TrkB protein and message in adult rat brain. Exp Neurol 145:62–70
Buyse M, Ovesjo ML, Goiot H, Guilmeau S, Peranzi G, Moizo L, Walker F, Lewin MJ, Meister B, Bado A 2001 Expression and regulation of leptin receptor proteins in afferent and efferent neurons of the vagus nerve. Eur J Neurosci 14:64–72
Gaige S, Abysique A, Bouvier M 2002 Effects of leptin on cat intestinal vagal mechanoreceptors. J Physiol 543(Pt 2):679–689
Smith GP, Gibbs J, Jerome C, Pi-Sunyer FX, Kissileff HR, Thornton J 1981 The satiety effect of cholecystokinin: a progress report. Peptides 2(Suppl 2):57–59
Altar CA, Cai N, Bliven T, Juhasz M, Conner JM, Acheson AL, Lindsay RM, Wiegand SJ 1997 Anterograde transport of brain-derived neurotrophic factor and its role in the brain. Nature 389:856–860
DiStefano PS, Friedman B, Radziejewski C, Alexander C, Boland P, Schick CM, Lindsay RM, Wiegand SJ 1992 The neurotrophins BDNF, NT-3, and NGF display distinct patterns of retrograde axonal transport in peripheral and central neurons. Neuron 8:983–993
Helke CJ, Adryan KM, Fedorowicz J, Zhuo H, Park JS, Curtis R, Radley HE, Distefano PS 1998 Axonal transport of neurotrophins by visceral afferent and efferent neurons of the vagus nerve of the rat. J Comp Neurol 393:102–117
Lommatzsch M, Braun A, Mannsfeldt A, Botchkarev VA, Botchkareva NV, Paus R, Fischer A, Lewin GR, Renz H 1999 Abundant production of brain-derived neurotrophic factor by adult visceral epithelia. Implications for paracrine and target-derived Neurotrophic functions. Am J Pathol 155:1183–1193
Saper CB, Loewy AD 1980 Efferent connections of the parabrachial nucleus in the rat. Brain Res 197:291–317
Canteras NS, Simerly RB, Swanson LW 1994 Organization of projections from the ventromedial nucleus of the hypothalamus: a Phaseolus vulgaris-leucoagglutinin study in the rat. J Comp Neurol 348:41–79
Baker-Herman TL, Fuller DD, Bavis RW, Zabka AG, Golder FJ, Doperalski NJ, Johnson RA, Watters JJ, Mitchell GS 2004 BDNF is necessary and sufficient for spinal respiratory plasticity following intermittent hypoxia. Nat Neurosci 7:48–55
Baker-Herman TL, Mitchell GS 2002 Phrenic long-term facilitation requires spinal serotonin receptor activation and protein synthesis. J Neurosci 22:6239–6246
Rage F, Givalois L, Marmigere F, Tapia-Arancibia L, Arancibia S 2002 Immobilization stress rapidly modulates BDNF mRNA expression in the hypothalamus of adult male rats. Neuroscience 112:309–318
Wang YH, Tache Y, Sheibel AB, Go VL, Wei JY 1997 Two types of leptin-responsive gastric vagal afferent terminals: an in vitro single-unit study in rats. Am J Physiol 273(2 Pt 2):R833–R837
Yuan CS, Attele AS, Dey L, Xie JT 2000 Gastric effects of cholecystokinin and its interaction with leptin on brainstem neuronal activity in neonatal rats. J Pharmacol Exp Ther 295:177–182
Emond M, Schwartz GJ, Ladenheim EE, Moran TH 1999 Central leptin modulates behavioral and neural responsivity to CCK. Am J Physiol 276(5 Pt 2):R1545–R1549
Matson CA, Ritter RC 1999 Long-term CCK-leptin synergy suggests a role for CCK in the regulation of body weight. Am J Physiol 276(4 Pt 2):R1038–R1045
Matson CA, Reid DF, Cannon TA, Ritter RC 2000 Cholecystokinin and leptin act synergistically to reduce body weight. Am J Physiol Regul Integr Comp Physiol 278:R882–R890
Tirassa P, Stenfors C, Lundeberg T, Aloe L 1998 Cholecystokinin-8 regulation of NGF concentrations in adult mouse brain through a mechanism involving CCK(A) and CCK(B) receptors. Br J Pharmacol 123:1230–1236(Bruno Bariohay, Bruno Lebrun, Emmanuel M)