Influence of a selective histamine H3 receptor antagonist on hypothalamic neural activity, food intake and body weight

Abstract

OBJECTIVE:

This study was conducted to elucidate whether antagonistic targeting of the histamine H3 receptor increases hypothalamic histamine levels, in parallel with decreases in food intake and body weight.

METHODS:

The competitive antagonist potency of a recently synthesized histamine H3 receptor antagonist, NNC 38-1049, was studied in intact HEK293 cells expressing human or rat histamine H3 receptor, in which NNC 38-1049 was allowed to antagonize the effect of the H3 receptor agonist R-α-methylhistamine on isoprenaline-induced accumulation of cAMP. The affinity of NNC 38-1049 for a number of variants of the histamine receptor was also determined. Following single dosing of normal rats with NNC 38-1049, hypothalamic histamine levels were assessed by means of microdialysis. Plasma and brain levels of NNC 38-1049 and acute effects on food intake and energy expenditure were followed after oral doses of 3–60 mg/kg. Potential side effects were examined with rat models of behaviour satiety sequence (BSS), pica behaviour and conditioned taste aversion (CTA). Intakes of food and water together with body weight were recorded for 15 days during daily dosing of dietary obese rats.

RESULTS:

NNC 38-1049 was found to be a highly specific and competitive antagonist towards both human and rat histamine H3 receptors, and measurable amounts of NNC 38-1049 were found in the plasma of rats following single oral doses of 3–60 mg/kg and in the brain after 15–60 mg/kg. Following single intraperitoneal injections of NNC 38-1049 (20 mg/kg), significant increases in extracellular histamine concentrations were observed. The same dose did not change BSS or pica behaviour acutely, nor did it induce CTA following repeated administration for 7 days. Reductions in food intake were seen very soon after administration, and occurred in a dose-dependent fashion. Energy expenditure was unchanged, but the respiratory quotient (RQ) tended to decrease at higher doses, indicating an increase in lipid oxidation. Twice daily administration of 20 mg/kg of NNC 38-1049 in old and dietary obese rats resulted in sustained reduction of food intake throughout a 2-week study, and was associated with a highly significant (P<0.01) decrease in body weight compared with controls (−18.4±3.4 vs +0.4±2.7 g). The same dose of NNC 38-1049 produced an acute decrease of water intake, but 24 h intakes were not significantly changed.

CONCLUSIONS:

The results of this study strongly support the idea that an increase in the hypothalamic concentration of histamine produces a specific reduction of food intake and that this effect can be translated into a decrease in body weight.

Introduction

There is accumulating evidence that histaminergic neural circuits arising in the tuberomammillary nucleus and projecting into the satiety centres of the hypothalamus participate in regulation of energy homeostasis.1, 2, 3, 4 The very central function of neural histamine in regulation of food intake1, 5, 6, 7 is further underlined by the fact that leptin,8, 9 amylin10, 11 and bombesin12 have been suggested to exert their anorectic effects through histaminergic circuits. In accordance herewith, histamineric neurons project into hypothalamic centres known to participate in food intake regulation: the paraventricular nucleus (PVN) and ventromedial hypothalamus (VMH), where the anorectic effect is thought to be mediated by postsynaptic histamine H1 receptors.5, 13 The density of this receptor14 together with the H3-receptor-mediated control of the intrasynaptic concentration of histamine both seem to be crucial for the strength of the anorectic signal.

The intrasynaptic concentration of histamine is primarily controlled by feedback signals from presynaptic histamine H3 receptors15 that inhibit both the conversion of L-histidine to histamine7 and the release of histamine into the synaptic cleft.15, 16

Thus, by reducing this inhibition, using a selective histamine H3 receptor antagonist, the synaptic concentration of histamine should increase together with the signalling from the histamine H1 receptor, and food intake should consequently be inhibited.

The present work was conducted to investigate this hypothesis, and we report here the effects of the specific histamine H3 receptor antagonist, NNC 0038-000-1049 (NNC 38-1049), on hypothalamic histamine levels, energy homeostasis and body weight.

Methods

Test compound

By testing the series of monoacyldiamines, 1-alkyl-4-(4-aryl-4-oxobutanoyl) piperazines were identified as novel, imidazole-free histamine H3 receptor antagonists.17 Among these, 1-cyclopentyl-4-(4-(4-chlorophenyl)-4-oxobutanoyl) piperazine, denoted NNC 0038-0000-1049, abbreviated NNC 38-1049, (Figure 1) was selected for further in vivo profiling. This compound was prepared from 1-cyclopentylpiperazine (Emka-Chemie, D-85375 Neufahrn, Germany) and 4-(4-chlorophenyl)-4-oxobutanoic acid (Sigma-Aldrich, St Louis, MO, USA) using standard amide-bond-formation methodology (1-hydroxybenzotriazole, N-ethyl-N′-(3-dimethylaminopropyl) carbodiimide hydrochloride, dimethylformamide, 20°C, 4 h), and was subsequently converted to a hydrochloride salt by co-evaporation with hydrochloric acid and recrystallization of the resulting salt from ethanol.

Figure 1
figure1

Molecular structure of the imidazole-free histamine H3 receptor antagonist NNC 0038-000-1049 (NNC 38-1049).

In vitro characteristics

The in vitro characteristics of NNC 38-1049 were investigated using a Flash Plate® cAMP assay (NEN™ Life Science Products) as described previously.18 Briefly, intact HEK293 cells, stably expressing the human or rat histamine H3 receptor, were treated with isoprenaline in order to stimulate cAMP generation. The ability of the H3 receptor agonist R-α-methylhistamine (RAMHA) to inhibit the isoprenaline-induced accumulation of cAMP was determined, and EC50 values were calculated by nonlinear regression analysis of the dose–response curve using GraphPad Prism version 4.00 (GraphPad Software, San Diego, CA, USA). The competitive antagonist potency of NNC 38-1049, that is, its ability to inhibit the RAMHA-induced inhibition of cAMP accumulation, was determined by generating Schild plots and establishing Kb values as described previously.18

The affinity of NNC 38-1049 for the human H1, H2 and H4 receptors was also determined. For the H1 and H2 receptors, membranes from CHO cells stably expressing these receptors were used in a conventional filter-binding assay with 3H-pyrilamine (H1) and 125I-aminopotentidin (H2) as radiolabelled ligands. For the histamine H4 receptor, membranes from HEK293 cells stably expressing the receptor were used in a conventional filter-binding assay with 3H-histamine as radiolabelled ligand.

Animals, diets and ethics

In the present study, 8–9-week-old male Sprague–Dawley rats purchased from Møllegård Breeding and Research Centre (Lille Skensved, Denmark) and given normal rat chow were employed as the standard model for studies of various acute effects of NNC 38-1049. Longer-term effects (2 weeks) on food intake and body weight were studied in 17-month-old female Wistar rats (Møllegård Breeding and Research Centre, Lille Skensved, Denmark) in which dietary obesity had been induced as previously described (DIO rats).19 Effects of NNC 38-1049 on hypothalamic histamine release were studied in male albino Wistar rats from Harlan (Zeist, the Netherlands).

After purchase, rats were acclimatized to the experimental environment for 2–4 weeks before the start of experiments. On arrival, rats were placed in conventional rat cages harbouring 2–3 animals. About 7–10 days before recording of food intake was initiated, animals were moved to single cages. Unless otherwise stated, rats were subjected to a 12/12 h light/dark regime, and room temperature was held between 20 and 23°C. Food was either freely available at all times or during a certain time period of 5–7 h (schedule feeding, see below). Water remained freely available, and rats were weighed weekly. Diets were purchased from a local feed manufacturer (Brogaarden, Gentofte, Denmark). All experiments were performed in accordance with the Declaration of Helsinki and approved by the local ethical committies, viz. Animal Experiment Inspectorate (Dyreforsøgstilsynet, Copenhagen, Denmark) and Animal Care Committee of the Faculty of Mathematics and Natural Science (University of Groningen, the Netherlands).

Plasma and hypothalamic levels of NNC 38-1049 following single dose

Rats were given single oral doses of NNC 38-1049 in doses ranging from 0 to 60 mg/kg 1 h before being decapitated without prior anaesthesia. Each dose group consisted of four randomly assigned rats. Blood was collected in EDTA-coated tubes, and the hypothalami were quickly dissected out and frozen in liquid nitrogen. After decantation, plasma was stored at −80°C until analysed. The hypothalami were homogenized in methanol for 30 s at 13 500 rpm using an Ultra Turrax T25 (Janke & Kunkel, Staufen, Germany). Aliquots of 1000 μl of the homogenates diluted with 1500 μl methanol were homogenized for 20 s, followed by centrifugation for 10 min at 724 g and decantation. Further treatment of the supernatants is described in the ‘Analyses’ section.

Microdialysis of hypothalamic histamine following single dose

Animals were first anaesthetized using isoflurane (2.5%, 400 ml/min N2O, 600 ml/min oxygen (O2)), after which holes were drilled in the skull overlying the PVN. Home-made I-shaped probes (Hospal AN 69) with a dialysable surface of 2 mm were slowly inserted in the PVN. Co-ordinates of implantation were: posterior to bregma −1.9 mm, lateral to midline +1.5 mm and ventral to dura −9.0 mm, at an angle of 10°. Probes were secured with dental cement and surgical screws.

Experiments were performed 24–48 h after surgery. Probes were perfused with an artificial cerebrospinal fluid (aCSF), containing 147 mM NaCl, 3.0 mM KCl, 1.2 mM CaCl2 and 1.2 mM MgCl2 at a flow rate of 1.5 μl/min. Concomitantly, 20 min microdialysis samples were collected online in high-performance liquid chromatography (HPLC) loops and injected automatically into the column. After basal levels of histamine were established, NNC 38-1049 or vehicle (saline) was injected intraperitoneally (i.p.) in 1 ml/kg volumes.

Tests of unspecific effects

Behaviour satiety sequence (BSS)

Studies of the BSS were performed essentially as described by Halford and Blundell.20 Following a 24 h fast, rats received oral or i.p. administration of NNC 38-1049 in doses ranging from 0 to 20 mg/kg (n=8). They were thereafter allowed access to a food bowl (10 cm ; 3 cm high) filled with granulated standard lab chow mixed with tap water (50/50; 1 kg granulate/l tap water).

The behaviour of rats was recorded manually once every 15 s for 40 min, and their behaviour was classified as:

  • eating

  • active (locomotion, rearing or sniffing)

  • grooming (licking, scratching, stroking or biting of any part of the body)

  • resting

These recordings were clustered on an individual basis in eight bins of 5 min containing a total of 20 observations. Within these 5-min bins, the frequency of each type of behaviour was expressed as a proportion. Thus, for example, if an individual rat was observed eating five times out of 20, the corresponding proportion became 0.25. The same was done for all four classes of behaviour. Data of this type for animals belonging to the same group and 5-min bin were subsequently used for calculation of means and are represented as such.

Pica behaviour (kaolin consumption)

Standard food and kaolin pellets (99% kaolin and 1% gum arabic) were fed to rats in two identical containers for 7 days. The rats were housed in cages with an elevated grid floor and standard bedding underneath. Thereafter, standard bedding was substituted with filter paper, and baseline kaolin consumption was measured for 24 h. On day 9, about 30–45 min before the start of the dark period, the animals received vehicle or oral doses of NNC 38-1049 (10, 15 and 20 mg/kg) and were placed in clean cages with fresh filter paper. The anticancer drug Cisplatin (5 mg/kg, Sigma-Aldrich, St Louis, MO, USA) was used as positive control. All groups of rats received fresh water, food and kaolin pellets. Consumption of food and kaolin pellets was recorded over a 16-h period, including a 12-h night period.

Conditioned taste aversion (CTA)

A 7-day subchronic CTA paradigm was adopted in order to examine whether NNC 38-1049 induced CTA following repeated administration. Rats were allowed to adapt to single cages for 1 week before they were given water in two identical bottles instead of one, and were switched over to a reversed day–light cycle with lights off between 0900 and 2100 h. When daily water intake had become stable, animals were divided into three groups (n=8), which on average consumed the same amount of water. One of these groups received daily i.p. injections of vehicle, a second group received the reference compound lithium chloride (LiCl, 40 mg/kg) and a third group was given NNC 38-1049 (20 mg/kg). Immediately following the first dose, and throughout the 1-week injection period, the water in both bottles was replaced with fresh-made saccharin solution (1 g/l). This supply was withdrawn exactly 24 h after the last dosing, and both bottles were again filled with water alone for another 24 h period. Following this, rats were given access to one bottle filled with saccharin, and another filled with water. This two-bottle preference test was maintained for 24 h. During this time, saccharin intake, water intake, total fluid intake and the ratio saccharin/total fluid were registered. The volume of saccharine solution consumed in relation to total fluid intake was used as a measure of CTA, evoked by treatments, and is expressed as % saccharin intake.

Effects on the hypothalamic–pituitary–adrenal (HPA) axis

Rats were injected i.p. with vehicle (n=4) or NNC 38-1049 (n=4) immediately prior to the beginning of a 5-h schedule-feeding period. After 90 min, they were decapitated, and blood was collected in EDTA-coated tubes and centrifuged. The plasma was stored at 80°C until analysed for corticosterone.

Food intake, energy expenditure and substrate oxidation following single dose

Food-intake experiments were performed with animals that had been adapted to a 7-h feeding schedule within the light period. On the test day, groups (n=15) of animals received single oral administrations of NNC 38-1049 in doses ranging from 0 to 60 mg/kg just before presentation of food, after which food consumption was recorded for 3 h. Data are presented as accumulated food intake.

O2 consumption was used as a measure of energy expenditure and the respiratory exchange ratio (RER), that is, the ratio of carbon dioxide (CO2) production to O2 consumption, was employed as a measure of the type of substrate being oxidized. This exchange of gases was studied using an Oxymax equal flow system (Columbus Instruments, Columbus, OH, USA). The system was calibrated daily before the start of measurements.

After a 2.5-h period of adaptation to the Oxymax chambers, rats received a single oral administration of NNC 38-1049 in doses ranging from 0 to 60 mg/kg, and measurements were continued for an additional 2.5-h period. O2 and CO2 concentrations in the reference air and in the chambers were measured every 16.9 min. O2 consumption was calculated per metabolic weight (kg live weight0.75).

Intakes of food and water, body weight and plasma variables following repeated dosing

This 15-day experiment was performed in DIO rats that had been adapted to a 7-h feeding schedule within the light period. One administration was performed just before presentation of food and another was carried out 3.5 h later. Rats were randomly assigned to groups (n=14–15) receiving one of the following treatments: vehicle alone, vehicle in combination with 25% food restriction, or NNC 38-1049 in doses of either 2 × 10 or 2 × 20 mg/kg. NNC 38-1049 was given orally during the first 5 days, but this route of administration was subsequently changed to i.p. administration in order to avoid the day-to-day variation in the effect on food intake that was observed with oral administration, which might have been related to some discomfort to the animals resulting from the frequent oesophageal intubations. Animals were kept in single cages during food-intake recordings, with free access to water and with access to food for 7 h per day. Body weights were recorded at the start of treatment, after 9 days, and at the termination of treatment. The amount of food consumed during these time intervals was estimated by subtracting the residual food recovered from each cage from the total amount presented. The accumulated daily water intake was measured in a similar fashion, and, in addition, detailed 7-h recordings of water intake were performed in direct association with administrations on days 1, 8 and 11. These data were pooled for each single rat and are presented as the 7-h mean water intake. On the last day of treatment, the unanaesthetized animals were killed by decapitation. After bleeding, serum and plasma were prepared and frozen at −80°C. Body composition was determined by dual X-ray scanning (Dexa) on exsanguinated carcasses after decapitation.

Analyses

Quantitative analysis of NNC 38-1049 in plasma and brain

The concentration of NNC 38-1049 in supernatants from hypothalami extracts and plasma was measured by HPLC and tandem mass spectrometry (LC-MS-MS).

Shortly before analysis, 100 μl of plasma was mixed with 100 μl 0.1% formic acid in methanol, and 100 μl supernatant of brain extract was added to 100 μl 0.1% formic acid in water. The mixtures were centrifuged for 20 min at 16 452 g and the supernatants were analysed by use of Cohesive Technologies TurboFlow chromatography (Bucks, UK). An Oasis HLB column (1.0 × 50 mm) was employed with a 5–95% 0.1% formic acid–methanol gradient system and a total runtime of 1.67 min. A Sciex API 3000 mass spectrometer (Toronto, Canada) was run in positive ionization multiple reaction monitoring (MRM) mode for specific detection of the compound. Calibration curves were prepared with NNC 38-1049 dissolved either in plasma or in the supernatant of brain extract in the concentration range 10–3000 ng/ml.

Analysis of histamine in dialysates

Histamine from microdialysis samples was separated on a reversed-phase HPLC column (100 × 2.0 mm C18 Hypersil 3 μm column (Bester, Amstelveen, the Netherlands)). The mobile phase consisted of 0.16 M KH2PO4, 0.4 mM sodium octane sulphate, 0.1 mM EDTA and 1% of methanol (pH 4.5), and was delivered by a Pharmacia LKB 2150 pump (Pharmacia, Uppsala Sweden) at a flowrate of 0.4 ml/min.

Upon separation, histamine was derivatized postcolumn by mixing of the mobile phase with a 0.002% m/v solution of OPA in 0.15 M NaOH solution through a T-piece. The flowrate of the derivatization pump (Shimadzu LC-10AD) was 0.5 ml/min. The mobile phase and the derivatization agent were connected by a T-piece, followed by a metal mixing coil (ID 0.55 mm, OD 1.1 mm, length 1 m; dimension 3 × 10 cm, insulated with cotton wool). The derivatization reaction was performed at ambient temperature. Fluorescence intensity was measured with a Jasco FP-1520 fluorimeter (Jasco Co., Tokyo, Japan), with excitation set at 350 nm, emission at 450 nm and set at the highest sensitivity (1 × 1000). At these settings, the sensitivity of the system was 1 fmol/sample (3 × noise).

Metabolites and hormones

Plasma or serum concentrations of metabolites were analysed using a Synchron CX5 auto-analyser system (Beckman Instruments, Fullerton, CA, USA), employing standard methods. Corticosterone was analysed using a solid-phase radioimmunoassay (Diagnostic Products Corporation, CA, USA) with maximum intra- and interassay coefficients of variation of 12 and 15%, respectively.

Statistics

Statistical calculations were performed using Graph Pad Prism version 3.0. (GraphPad Software, San Diego CA, USA) or SAS version 8.2 (SAS Institute, Cary, NC, USA). Potential differences between treatment groups, exhibiting normally distributed data, were tested by one-way analyses of variance followed by Dunnett's post hoc test for multiple comparisons. In other cases, data were analysed by nonparametric one-way analysis of variance (Kruskal–Wallis test) using Dunn's post hoc test for multiple comparisons. Comparisons having a P-value of less than 0.05 were considered to be statistically significant. Data are presented as the means±s.e.m. unless otherwise stated. In the figures, significant statistical differences between means are marked as follows: *P<0.05, **P<0.01, ***P<0.001.

Results

In vitro characteristics

NNC 38-1049 was found to be a potent competitive antagonist at the human histamine H3 receptor. In the cAMP assay, Kb values of 2.3±0.63 nM (n=3) were determined. A representative Schild plot is shown in Figure 2, and in that particular experiment the Kb value was 2.7 nM. These Kb values agree well with the competitive antagonist potency values on the human H3 receptor determined in a membrane-based, functional [35S]GTPS assay (Ki=1.2±0.1 nM n=10)).17 NNC 38-1049 was also a potent, competitive antagonist at the rat H3 receptor (Ki=5.1±0.99 nM (n=10)) in a [35S]GTPS assay. Finally, NNC 38-1049 was found to be selective for the H3 receptor, in that the binding affinities for the human H1, H2 and H4 receptors were >10 000 nM (data not shown).

Figure 2
figure2

Schild plot determining the antagonist potency of NNC 38-1049 at the human histamine H3 receptor expressed in HEK293 cells. Dose–response curves for the RAMHA (H3 receptor agonist)-induced inhibition of isoprenaline-induced cAMP accumulation were generated with RAMHA alone, or in the presence of increasing concentrations of NNC 38-1049. Data points represent means of duplicate determinations from one representative experiment.

Plasma and hypothalamic levels of NNC 38-1049 following a single dose

Measurable amounts of NNC 38-1049 were found in plasma following oral doses of 3–60 mg/kg and in the brain after 15–60 mg/kg (Figure 3).

Figure 3
figure3

Plasma and hypothalamic concentrations of NNC 38-1049 following oral administration of doses ranging between 0 and 60 mg/kg of the same compound in rats. Administration was performed 1 h before the killing. Data represent means±s.e.m. of n=4. Due to the small number of observations, no statistical tests were performed on these data.

Hypothalamic histamine levels following single dose

NNC 38-1049 at an i.p. dose of 20 mg/kg, was found to produce a sharp rise in the concentration of histamine in dialysates recovered from the PVN. This rise was found to be statistically significant (P<0.05) in the period from 20 to 100 min following administration. A dose of 5 mg/kg also seemed to produce a slight, but yet statistically insignificant, increase (Figure 4).

Figure 4
figure4

Effect of i.p. administration of vehicle (saline, n=6), 5 mg/kg NNC 38-1049 (n=4) and 20 mg/kg NNC 38-1049 (n=4) on histamine levels in the PVN of rats. Basal histamine levels were 17.41±3.40 fmol per sample (n=14). All data are means±s.e.m. Following administration of 20 mg/kg, this rise was statistically significant (P<0.05) in the period from 20 to 100 min.

Tests of unspecific effects

Behaviour satiety sequence (BSS)

I.p. administration of NNC 38-1049 did not cause a disruption of the behavioural pattern of rats after presentation of food, as determined by observation of BSS (Figure 5). All the components of behaviour were represented in a fairly similar sequence regardless of whether rats were given vehicle or NNC 38-1049. A tendency to spend a little less time eating in the beginning of the 40-min observation period could, however, be noticed in rats given NNC 38-1049. When the areas under the curves were integrated for the whole 40-min period and statistically compared, no differences between groups were found for any of the behaviour classes. Following oral dosing, a slight activity increase was seen at doses of 10 and 15 mg/kg, whereas activity had returned to normal at 20 mg/kg (data not shown).

Figure 5
figure5

Representation of the BSS of rats after i.p. injection of either vehicle or NNC 38-1049 in a dose of 20 mg/kg. Data represent means±s.e.m. (n=8). When the area under the curves were integrated for the whole 40-min period and statistically compared, no differences between groups were found for any of the behaviour classes.

Pica behaviour (kaolin consumption)

There were no significant effects of NNC 38-1049 on pica behaviour (Figure 6), whereas the reference compound cisplatin produced a dramatic and highly significant (P<0.001) effect.

Figure 6
figure6

Pica behaviour as measured with kaolin consumption after oral administration of vehicle or increasing doses of NNC 38-1049 (a) and after either vehicle or cisplatin in a dose of 5 mg/kg (b) as positive control. Data represent means±s.e.m. (n=8) and statistical differences between means are marked as follows: *P<0.05, **P<0.01, ***P<0.001.

Conditioned taste aversion (CTA)

NNC 38-1049 did not induce any significant (P<0.05) CTA, as measured by saccharin preference (Figure 7) following a 7-day conditioning period in which rats were given a daily i.p. dose of 20 mg/kg. In contrast, the reference compound LiCl was found to produce a highly significant (P<0.01) reduction of saccharine consumption.

Figure 7
figure7

CTA measured as saccharine preference (saccharin intake/total fluid intake) after 7 days of i.p. administration of vehicle, NNC 38-1049 (20 mg/kg) or LiCl (40 mg/kg), which served as positive control. Data are means±s.e.m. (n=8) and statistical differences between means are marked as follows: *P<0.05, **P<0.01, ***P<0.001.

Effects on the HPA axis

No signs of activation of the HPA axis were found, as evidenced by the plasma levels of corticosterone, for which no significant difference was found between vehicle and NNC 38-1049-treated rats (175±35.1 vs 114.6±21.7 ng/ml).

Food intake, energy expenditure and substrate oxidation following single dose

Oral doses of NNC 38-1049 ranging from 15 to 60 mg/kg were found to produce significant (P<0.05) reductions in the accumulated amount of food consumed during a 3 h period following presentation of food in schedule-fed rats (Figure 8).

Figure 8
figure8

Accumulated food intakes during 3 h following doses ranging from 0 to 60 mg/kg of NNC 38-1049 in schedule-fed male rats. Data represent means±s.e.m. (n=15) and statistical differences between means are marked as follows: *P<0.05, **P<0.01, ***P<0.001.

None of the doses of NNC 38-1049 tested was associated with any significant changes in energy expenditure following single dosing, although the respiratory quotient (RQ) (RER) tended to decrease at higher doses (Figure 9). This tendency was particularly pronounced after administration of the highest dose, that is, 60 mg/kg.

Figure 9
figure9

Energy expenditure (a) and whole-body substrate oxidation (b) as determined by indirect calorimetry following doses of 0–60 mg/kg of NNC 38-1049. Data represent means±s.e.m. (n=6) and statistical differences between means are marked as follows: *P<0.05, **P<0.01, ***P<0.001.

Intakes of food and water, body weight and plasma variables following repeated dosing

In old DIO rats that had been trained to consume their daily food intake during a 7-h period, two daily administrations of NNC 38-1049 (2 × 20 mg/kg), were found to result in a significantly (P<0.01) sustained reduction in food intake for 15 days. This inhibition was of the order of 25%, and was associated with a highly significant (P<0.01) decrease in body weight. The decrease in body weight was of the same magnitude as seen in animals that were restricted to eat no more than 75% of their normal intake of food (Figure 10). Following the loss in body weight, there were no specific changes in the proportions of lean tissue and body fat (data not shown).

Figure 10
figure10

Accumulated food intake and changes in body weight in old DIO rats after dosing with either vehicle or NNC 38-1049 at 0 and 3.5 h. Vehicle was given alone or in combination with a 25% food restriction, and NNC 38-1049 was given in doses of 2 × 10 or 2 × 20 mg/kg for 15 days. Data are means±s.e.m. (n=14–15) and statistical differences between means are marked as follows: *P<0.05, **P<0.01, ***P<0.001. The food intake for the restricted group was not tested.

At a dose of 20 mg/kg, NNC 38-1049 significantly (P<0.01) decreased water intake in direct association with dosing (Figure 11). Later, this was compensated for so that 24-h measurements of water intake in rats receiving NNC 38-1049 were not different from those found in rats receiving vehicle.

Figure 11
figure11

Mean 7-h (a, n=5) and 24-h (b, n=14–15) water intake in old DIO rats after administration of either vehicle or NNC 38-1049 at 0 and 3.5 h. Vehicle was given alone or in combination with a 25% food restriction, and NNC 38-1049 was given in doses of 2 × 10 or 2 × 20 mg/kg for 15 days. Data are means±s.e.m. and statistical differences between means are marked as follows: *P<0.05, **P<0.01, ***P<0.001.

Plasma glucose, FFA and insulin remained generally unchanged after NNC 38-1049 dosing, whereas serum triglycerides were significantly (P<0.05) decreased in rats that had received the highest dose (Table 1). What, in a first instance, looked like a concomitant increase in β-hydroxybutyrate and decrease in leptin was not found to be statistically significant following post hoc testing.

Table 1 Plasma variables in dietary obese rats following 15 days of administration of vehicle or NNC 38-1049

Discussion

In the present study, we demonstrate that NNC 38-1049 is a very potent and specific competitive antagonist at both the human and rat histamine H3 receptor. This was evidenced by its ability to antagonize the effects of the well-known H3 receptor agonist RAMHA on the intracellular cAMP concentration. These observations provide evidence that the in vivo effects discussed below can be attributed to antagonism at the histamine H3 receptor.

NNC 38-1049 produced an increase in extracellular levels of histamine in the PVN. This original finding is interesting from the point of view that it has previously been demonstrated that the PVN plays a central role in the histaminergic regulation of food intake21 and that it harbours a population of H1 and H3 receptors.22 This finding adds further support to the view that the histamine H3 receptor is essential for regulating the histamine release and synaptic histamine concentration,15, 16, 23, 24, 25 and that this concentration can be increased by antagonistic targeting of the histamine H3-receptor.

It has previously been shown that intracerebroventricular (i.c.v.) injections of histamine H3 receptor antagonists such as thioperamide26, 27 are associated with an acute decrease of food intake. In one study performed by Attoube et al,28 these findings were reproduced also after peripheral administration of the same drug. The present work with NNC 38-1049 adds further support to these observations. In addition, we show that it is possible to sustain this effect on food intake over a prolonged period of time, and that a significant reduction in body weight is achieved. Our observations of BSS also imply that the effect on food intake can be separated from other unwanted behavioural changes, that is, that the reduction of food intake is specific. These data are in accordance with recently published observations with an H3 receptor antagonist belonging to another structurally different class of compounds,29 and increase the possibility for a future clinical use of an histamine H3 receptor antagonist in obese humans. Moreover, NNC 38-1049, obviously, did not evoke visceral illness, since no increase in pica behaviour was observed.

CTA is often used to examine potential side effects such as nausea or malaise. In most cases, this test is performed in acute settings. In an initial study, we found that NNC 38-1049 acutely induced CTA in a dose of 20 mg/kg, but not in doses of 10 or 30 mg/kg (data not shown). However, the most relevant question from a therapeutic and specificity point of view is, of course, whether the support of a potential CTA will be sustained during repeated administration. Previous experiences with other drugs have shown that the response to conditioning stimuli can change during repeated administration.30 For this reason, we adopted a subchronic CTA paradigm, in which the conditioning period was prolonged to 7 days. In this paradigm, NNC 38-1049 did not induce a significant CTA. This adds further support to the conclusion that effects of NNC 38-1049 on food intake and body weight are specific and not due to inflictions on well-being.

The fact that obese rats used in the present study lost a significant amount of weight may seem to contradict to the recent findings by Takahashi et al that histamine H3 receptor knockout mice actually gained in weight and became obese.31 However, we do not believe that this is necessarily the case. Intermittent increases in neural histamine by pharmacological means might have a quite different effect compared with a congenital form of histamine H3 receptor disruption. In fact, Takahashi et al31 found a certain decrease in H1 receptor abundance in their histamine H3 receptor knockout mice, which might explain the increase in food intake and concomitant increase in body weight. In the present study, no measurements of H1 receptor abundance was performed, but such measurements would certainly help to clarify this issue, which of course is of vital importance for a thorough understanding of an anticipated difference between the pharmacological antagonism of the histamine H3 receptor and complete genetic disruption. In this context, it is worth mentioning that NNC 38-1049 has a relatively short plasma half-life of about 20 min, which was the reason for the twice-daily dosing regimen. This also suggests that, in the present study, there was no carryover of drug exposure from one day to the other. This ‘recovery’ period could conceivably be of importance for the possibility of achieving a sustained reduction of food intake over a prolonged period of time.

In the present study, the reduction of food intake at the highest dose of NNC 38-1049 was approximately 25%, which was similar to that imposed on the food-restricted group. This similarity in energy intakes between the two groups was also reflected in a very similar loss of body weight. Moreover, body composition after weight loss did not differ between these groups. This observation fits well with our findings that NNC 38-1049 did not induce changes in daily water consumption and that there were no acute specific effects on energy expenditure. It has, however, to be emphasized that the measurements of energy expenditure were performed after administering a single dose in young growing animals and not in the dietary obese rats used in connection with repeated sampling. This fact may have decreased our possibilities of detecting such an effect on energy expenditure. Although neural histamine has been suggested to affect energy expenditure, it is actually difficult, not to say impossible, to find direct evidence for that in the literature. Instead, most of these suggestions are based on circumstantial evidence such as increases in the expression of mRNA of uncoupling protein-1,4 often derived from experiments employing genetically modified animals with defects in the leptin–melanocortin axis. Even a decrease in energy expenditure has recently been reported following i.p. injections of the relatively unspecific H3 receptor antagonist thioperamide.32 The fact that, in the present study, we could not confirm an upregulation of energy expenditure in the acute situation does not exclude the possibility of this occurring after repeated dosing, and new studies investigating the more long-term effects of a histamine H3 receptor antagonist on energy expenditure are in progress.

It has also been suggested that histamine promotes lipolysis via stimulation of the sympathetic β-adrenoreceptor.33 The reduction of the RQ that was observed after a single dose with the highest dose level of NNC 38-1049 indicates that lipid oxidation was increased in this situation. The depressed triglyceride levels observed after repeated administration of NNC 38-1049 are also in line with this. However, the reason why these changes in lipid metabolism did not translate into an accelerated loss of adiposity (data not shown) after prolonged administration compared with caloric restriction alone is presently not known, but could imply that the acute effects of NNC 38-1049 on lipid oxidation might change over time. In any case, these observations merit further investigation, and future studies will reveal the long-term effects of H3 receptor antagonism on whole-body lipid oxidation and adiposity.

No major changes were seen in plasma levels of glucose, insulin and corticosterone, implying that specific targeting of the histamine H3 receptor could be a sound and effective way of correcting obesity. It has previously been reported that i.c.v. injections of histamine elevate circulating levels of ACTH and corticosterone in rats.34 One of the reasons that this was not observed in the present study could be that such effects are relatively transient.34 Our animals were not bled in direct association with dosing of NNC 38-1049. Other differences in experimental design, for example, route of administration, could also have played a role.

In conclusion, the present study has addressed a number of critical issues in the evaluation of the concept of using antagonistic targeting of the histamine H3 receptor to decrease food intake and body weight. The results presented here with NNC 38-1049, and also by others (for a review, see Leurs et al35), strongly support the idea that such selective antagonism represents a promising avenue for further research. The most urgent issue seems to be the identification of a candidate with similar or superior efficacy profile, as shown here with NNC 38-1049, but with a somewhat longer plasma half-life.

References

  1. 1

    Sakata T, Ookuma K, Fujimoto K, Fukagawa K, Yoshimatsu H . Histaminergic control of energy balance in rats. Brain Res Bull 1991; 3–4: 371–375.

    Article  Google Scholar 

  2. 2

    Fujise T, Yoshimatsu H, Kurokawa M, Oohara A, Kang M, Nakata M, Sakata T . Satiation and masticatory function modulated by brain histamine in rats. Proc Soc Exp Biol Med 1998; 2: 228–234.

    Article  Google Scholar 

  3. 3

    Tsuda K, Yoshimatsu H, Niijima A, Chiba S, Okeda T, Sakata T . Hypothalamic histamine neurons activate lipolysis in rat adipose tissue. Exp Biol Med (Maywood) 2002; 3: 208–213.

    Article  Google Scholar 

  4. 4

    Masaki T, Chiba S, Yoshimichi G, Yasuda T, Noguchi H, Kakuma T, Sakata T, Yoshimatsu H . Neuronal histamine regulates food intake, adiposity, and uncoupling protein expression in agouti yellow (A(y)/a) obese mice. Endocrinology 2003; 6: 2741–2748.

    Article  Google Scholar 

  5. 5

    Mercer LP, Kelley DS, Humphries LL, Dunn JD . Manipulation of central nervous system histamine or histaminergic receptors (H1) affects food intake in rats. J Nutr 1994; 7: 1029–1036.

    Article  Google Scholar 

  6. 6

    Mercer LP . Histamine and the neuroregulation of food intake. Nutrition 1997; 6: 581–582.

    Article  Google Scholar 

  7. 7

    Vaziri P, Dang K, Anderson GH . Evidence for histamine involvement in the effect of histidine loads on food and water intake in rats. J Nutr 1997; 8: 1519–1526.

    Article  Google Scholar 

  8. 8

    Morimoto T, Yamamoto Y, Yamatodani A . Leptin facilitates histamine release from the hypothalamus in rats. Brain Res 2000; 2: 367–369.

    Article  Google Scholar 

  9. 9

    Toftegaard CL, Knigge U, Kjaer A, Warberg J . The role of hypothalamic histamine in leptin-induced suppression of short-term food intake in fasted rats. Regul Pept 2003; 1–3: 83–90.

    Article  Google Scholar 

  10. 10

    Rushing PA, Hagan MM, Seeley RJ, Lutz TA, D'Alessio DA, Air EL, Woods SC . Inhibition of central amylin signaling increases food intake and body adiposity in rats. Endocrinology 2001; 11: 5035.

    Article  Google Scholar 

  11. 11

    Mollet A, Lutz TA, Meier S, Riediger T, Rushing PA, Scharrer E . Histamine H1 receptors mediate the anorectic action of the pancreatic hormone amylin. Am J Physiol Regul Integr Comp Physiol 2001; 5: R1442–R1448.

    Article  Google Scholar 

  12. 12

    Kent P, Plamondon H, Merali Z . Pharmaco-ontogeny of bombesin's suppression of food intake and its attenuation by histamine H3 receptor agonists. Brain Res Dev Brain Res 1997; 1: 87–95.

    Article  Google Scholar 

  13. 13

    Lecklin A, Etu-Seppala P, Stark H, Tuomisto L . Effects of intracerebroventricularly infused histamine and selective H1, H2 and H3 agonists on food and water intake and urine flow in Wistar rats. Brain Res 1998; 1–2: 279–288.

    Article  Google Scholar 

  14. 14

    Haq AU, Bundrant HM, Mercer LP . Food intake is inversely correlated with central nervous system histamine receptor (H1) concentrations in male Sprague–Dawley rats fed normal, low protein, low energy or poor quality protein diets. J Nutr 1996; 12: 3083–3089.

    Article  Google Scholar 

  15. 15

    Arrang JM, Garbarg M, Schwartz JC . Auto-inhibition of brain histamine release mediated by a novel class (H3) of histamine receptor. Nature 1983; 5911: 832–837.

    Article  Google Scholar 

  16. 16

    Arrang JM, Garbarg M, Schwartz JC . Autoregulation of histamine release in brain by presynaptic H3-receptors. Neuroscience 1985; 2: 553–562.

    Article  Google Scholar 

  17. 17

    Zaragoza F, Stephensen H, Knudsen SM, Pridal L, Wulff BS, Rimvall K . 1-Alkyl-4-acylpiperazines as a new class of imidazole-free histamine H(3) receptor antagonists. J Med Chem 2004; 11: 2833–2838.

    Article  Google Scholar 

  18. 18

    Wulff BS, Hastrup S, Rimvall K . Characteristics of recombinantly expressed rat and human histamine H3 receptors. Eur J Pharmacol 2002; 1: 33–41.

    Article  Google Scholar 

  19. 19

    Malmlöf K, Johansen T . Growth hormone-mediated breakdown of body fat: insulin and leptin responses to GH are modulated by diet composition and caloric intake in old rats. Hormone Metab Res (Hormon Stoffwechselforsch Hormones Metab) 2003; 4: 236–242.

    Google Scholar 

  20. 20

    Halford JC, Blundell JE . Metergoline antagonizes fluoxetine-induced suppression of food intake but not changes in the behavioural satiety sequence. Pharmacol Biochem Behav 1996; 4: 745–751.

    Article  Google Scholar 

  21. 21

    Ookuma K, Yoshimatsu H, Sakata T, Fujimoto K, Fukagawa F . Hypothalamic sites of neuronal histamine action on food intake by rats. Brain Res 1989; 2: 268–275.

    Article  Google Scholar 

  22. 22

    Palacios JM, Wamsley JK, Kuhar MJ . The distribution of histamine H1-receptors in the rat brain: an autoradiographic study. Neuroscience 1981; 1: 15–37.

    Article  Google Scholar 

  23. 23

    Itoh Y, Oishi R, Nishibori M, Saeki K . Characterization of histamine release from the rat hypothalamus as measured by in vivo microdialysis. J Neurochem 1991; 3: 769–774.

    Article  Google Scholar 

  24. 24

    Jansen FP, Mochizuki T, Yamamoto Y, Timmerman H, Yamatodani A . In vivo modulation of rat hypothalamic histamine release by the histamine H3 receptor ligands, immepip and clobenpropit. Effects of intrahypothalamic and peripheral application. Eur J Pharmacol 1998; 2–3: 149–155.

    Article  Google Scholar 

  25. 25

    Westerink BH, Cremers TI, De Vries JB, Liefers H, Tran N, De Boer P . Evidence for activation of histamine H3 autoreceptors during handling stress in the prefrontal cortex of the rat. Synapse 2002; 4: 238–243.

    Article  Google Scholar 

  26. 26

    Doi T, Sakata T, Yoshimatsu H, Machidori H, Kurokawa M, Jayasekara LA, Niki N . Hypothalamic neuronal histamine regulates feeding circadian rhythm in rats. Brain Res 1994; 2: 311–318.

    Article  Google Scholar 

  27. 27

    Itoh E, Fujimiya M, Inui A . Thioperamide, a histamine H3 receptor antagonist, powerfully suppresses peptide YY-induced food intake in rats. Biol Psychiatry 1999; 4: 475–481.

    Article  Google Scholar 

  28. 28

    Attoub S, Moizo L, Sobhani I, Laigneau JP, Lewin MJ, Bado A . The H3 receptor is involved in cholecystokinin inhibition of food intake in rats. Life Sci 2001; 4: 469–478.

    Article  Google Scholar 

  29. 29

    Hancock AA, Bennani YL, Bush EN, Esbenshade TA, Faghih R, Fox GB, Jacobson P, Knourek-Segel V, Krueger KM, Nuss ME, Pan JB, Shapiro R, Witte DG, Yao BB . Antiobesity effects of A-331440, a novel non-imidazole histamine H3 receptor antagonist. Eur J Pharmacol 2004; 1–3: 183–197.

    Article  Google Scholar 

  30. 30

    De Beun R, Lohmann A, Schneider R, De Vry J . Ethanol intake-reducing effects of ipsapirone in rats are not due to simple stimulus substitution. Pharmacol Biochem Behav 1996; 4: 891–898.

    Article  Google Scholar 

  31. 31

    Takahashi K, Suwa H, Ishikawa T, Kotani H . Targeted disruption of H3 receptors results in changes in brain histamine tone leading to an obese phenotype. J Clin Invest 2002; 12: 1791–1799.

    Article  Google Scholar 

  32. 32

    Sindelar DK, Shepperd ML, Pickard RT, Alexander-Chacko J, Dill MJ, Cramer JW, Smith DP, Gadski R . Central H3R activation by thioperamide does not affect energy balance. Pharmacol Biochem Behav 2004; 2: 275–283.

    Article  Google Scholar 

  33. 33

    Tsuda K, Yoshimatsu H, Niijima A, Chiba S, Okeda T, Sakata T . Hypothalamic histamine neurons activate lipolysis in rat adipose tissue. Exp Biol Med (Maywood) 2002; 3: 208–213.

    Article  Google Scholar 

  34. 34

    Kjaer A, Larsen PJ, Knigge U, Warberg J . Histaminergic activation of the hypothalamic–pituitary–adrenal axis. Endocrinology 1994; 3: 1171–1177.

    Article  Google Scholar 

  35. 35

    Leurs R, Bakker RA, Timmerman H, de Esch IJ . The histamine H3 receptor: from gene cloning to H3 receptor drugs. Nat Rev Drug Discov 2005; 2: 107–120.

    Article  Google Scholar 

Download references

Acknowledgements

We acknowledge the late Dr René de Beun for initiating the BSS and pica studies. The excellent technical assistance provided by Mr Frank Strauss and Mrs Hanne Jepsen is particularly acknowledged.

Author information

Affiliations

Authors

Corresponding author

Correspondence to K Malmlöf.

Additional information

Parts of this work were presented in abstract form at the 12th European Congress of Obesity in Helsinki 2003, Finland, and at the European Histamine Research Society 2003 meeting in Noordwijkerhout, the Netherlands.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Malmlöf, K., Zaragoza, F., Golozoubova, V. et al. Influence of a selective histamine H3 receptor antagonist on hypothalamic neural activity, food intake and body weight. Int J Obes 29, 1402–1412 (2005). https://doi.org/10.1038/sj.ijo.0803036

Download citation

Keywords

  • histamine
  • H3 receptor antagonist
  • food intake
  • body weight

Further reading