Histamine has a well-established role as a neurotransmitter in the central nervous system (1). Among variety of functions, the histaminergic system plays a regulatory role in feeding behaviour, and histamine acts as an inhibitory modulator (1, 2). After portocaval anastomosis (PCA) sustained high plasma level of aromatic amino acids (L-tryptophan, L-phenylalanine, L-tyrosine, L-histidine) results in excessive amine transmitter synthesis in the brain, leading to alterations in the cerebral aminergic systems, including the histaminergic one. In the brain of PCA rats, both the content (3) and release (4-5) of histamine from neurons are increased. However, despite that, shunted rats eat and drink more than their sham-operated counterparts and have the body mass between 20 to 30% below their control pairs (3-5). Underlying regulatory mechanisms were, therefore, addressed. Considering a close anatomical localization of histaminergic and orexinergic systems (6) and a stimulatory effect of neuropeptide orexin A on food intake (7), we investigated the importance of histamine H1 and H2 and orexin type 1 (OX1) receptors. Moreover, we tested an involvement of cannabinoid type 1 (CB1) receptor (8), as endocannabinoids play an important role in the control of food intake and energy homeostasis. Finally, hypothalamic levels of orexin A and histamine as well as blood concentrations of some circulating satiety peptides were examined.


MATERIAL AND METHODS
Animals and treatments

All procedures strictly followed the EU legislation, concerning animal experiments and were approved by the local ethical committee. Adult male Lewis rats with an end-to-side portocaval anastomosis and sham-operated animals according to Lee and Fisher (9), 6-8 months after operations were used. To enable intracerebroventricular (icv) drug administration, the rats were implanted with a polyethylene cannulae inserted into the right brain lateral ventricle under ketamine/xylazine (100mg/10mg per kg, intraperitoneally [(ip]) anaesthesia, 5-7 days prior to the study.

In a single treatment, rats were rendered motivated for food intake by 16 hours fasting with water available ad libitum. The involvement of histamine and orexins in food intake regulation was evaluated by using H1 and H2 histamine receptor antagonists pyrilamine (5 mg/kg; ip) and ranitidine (70 µg; icv) and OX1 receptor blocker SB 334867 (10 mg/kg; ip). All ip injections were performed 30 min, while icv administrations - 5 min before food presentation.

At zero time the rats were placed individually in metabolic cages with free access to a pre-measured amount and the volume of regular lab diet for rats and a tap water, respectively. The consumption and excretion records were done hourly for 6 h and thereafter at 24 and 48h.

There was no food restriction in the chronic treatment with CB1 receptor blocker rimonabant. The drug was given by a gastric tube in a dose of 1.5, 3.0 or 10 mg daily for 10 days. Consumption records were done before, during and post-therapy.

Analytical procedures.

One week after the completion of studies the rats were sacrificed by decapitation. Blood was collected from truncal vessels to measure the concentrations of circulating leptin, insulin and cholecystokinin. Brain was quickly removed from the skull and the hypothalamus dissected to analyse orexin A and histamine concentrations.

Estimation was accomplished for peptides by radioimmunoassays with RIA kits provided either by Phoenix Pharmaceuticals, Inc. USA (rat orexin A and cholecystokinin octapeptide CCK 26-33) or by LINCO Research, USA (rat insulin and leptin). The concentration of histamine was analyzed by radioenzymatic method (10).

Statistics

Means ± SEM were calculated and differences between groups were assessed with paired or unpaired t-tests as appropriate.


RESULTS
The PCA rats, smaller by weight up to 30%, ate about one and half and drank twice as much daily as sham-operated animals (Fig. 1, left panel). They also excreted 2.5 fold more urine. They had smaller livers but somewhat stomachs (Fig. 1, right panel) with thicker gastric mucosa.

Fig. 1. (Left) PCA rats (n=36) expressed higher feed and water intakes. (Right) They have liver atrophy and stomach enlargement. *** p<0.001 vs Sham (n=14)

Centrally applied ranitidine did not influence feed consumption, either in PCA or in control rats (data not shown), while pyrilamine stimulated food intake in both groups, the response being lower and its duration shorter in PCA rats (Fig. 2). As can be further seen in Fig. 2. SB 334867 given before food presentation decreased consumption in all rats. Noteworthy, the sham rats were more susceptible to OX1 receptor antagonist; a reduced food intake was present over 24 h.

Fig. 2. The effect of pretreatment with Pyrilamine H1 receptor antagonist (5 mg/kg, i.p.) or SB 334867 an orexin OX1R antagonist (10 mg/kg, i.p). Cumulative intake for 6h, 24h (D1) and 48h (D2). Values are means with SEM for 4-6 rats per treatment.

Likewise, in a different manner feeding behavior was modified in the two groups by chronically administered rimonabant (Fig. 3). Two panels on the left side of Fig. 3. show the avarage daily food intake per sham (upper) and PCA (lower) rat during 10 days of treatment, with the used drug doses: 1.5 mg/kg - curve with empty squares, 3 mg/kg - curve with gray squares, 10 mg/kg - curve with black squares. The corresponding: empty-, gray- and black- filled columns on the right side panel show means with SEM for appropriate groups, before, during and post therapy.

Fig. 3. (Upper left) Avarage daily food intake per sham (upper) and (Lower left) per PCA rat during 10 days of treatment with the used Rimonabant doses (i.g): 1.5mg/kg - curve with empty squares, 3mg/kg- curve with gray squares, 10 mg/kg-curve with black squares. The corresponding: empty-, gray- and black- filled columns on Right panel show means with SEM for appropriate groups, before, during and post therapy. All groups consist of n=6-7 rats with exception of n=17 PCA treated with 10 mg/kg Rimonabant. a statistically significant with p<0.05 (at least) therapy vs before therapy; b p<0.05 post therapy vs therapy.

Paradoxically, in PCA rats, the lowest dose of rimonabant (1.5 mg/kg) stimulated food intake, so, during and post therapy it was significantly higher than before the treatment. In contrast, PCA rats treated with 10 mg/kg, ate significantly more after the therapy completion (ap<00.5 post therapy vs. before therapy, bp<0.05 post therapy vs. therapy). Only sham rats treated with the highest rimonabant dose ate significantly less throughout the therapy than in preceding period but then they increased the consumption. Similarly, sham rats treated with 1.5 and 3.0 mg/kg of rimonabant consumed more after the therapy (p<0.05) than before and during the treatment.

Concerning the effect of rimonabant on body mass modification, again sham rats were better responders and after 10 days of treatment with 3.0 mg/kg and 10 mg/kg they lost on average 12% of initial body mass (p<0.05), as compared with only up to 5% in PCA rats - Fig. 4.

Fig. 4. The effect of chronic Rimonabant (i.g) on body mass in sham operated and portocavally shunted (PCA) rats. Values are means with SEM. a p<0.05 vs before therapy

Table 1. presents concentrations of circulating satiety signalling peptides as measured in blood serum. In PCA rats, CCK was almost doubled whilst leptin concentration was about half of that of sham-operated animals. There were no differences in insulin concentrations. In the hypothalamus of PCA rats both, orexin A and histamine concentrations were significantly increased, i.e. orexin: 38.7 ± 2.54 vs. 26.64 ± 0.73 ng/g, p<0.002; histamine: 789 ± 59 vs. 171 ± 35 ng/g, p<0.0001.

Table 1. Plasma cholecystokinin (CCK), leptin and insulin concentrations in PCA and Sham-operated rats

The values represent the means ± SEM. *p<0.05, One way analysis of variance and Student-Keuls-Newman test

DISCUSSION
The present data confirm an involvement of histamine, orexins and endocannabinoids in the regulation of food intake in PCA rats. Moreover, we demonstrate for the first time a lower susceptibility of the central energy homeostatic mechanisms to pyrilamine, SB 334867 and rimonabant in these animals.

Neuro-hormonal control of food intake includes central neuropeptidergic, monoaminergic and endocannabinoid systems activated by peripheral peptides (CCK, orexins, ghrelin, polypeptyde YY, glucagon-like peptide-1, leptin, oxyntomodulin) (11,12). In the brain areas involved in the food intake control, i.e. in the arcuate and paraventricular nuclei, lateral hypothalamus and perifornical area, many orexigenic (neuropeptide Y, agouti-related peptide, orexin A, melanin) and anorexigenic substances (proopiomelanocortin-derived peptides, histamine, thyrotropin releasing hormone, corticotropin releasing hormone, oxytocin) are secreted to establish an individual response to a meal (13-15). The central nervous system (CNS) modifies food intake mainly by vagal nerves that provide the major neuroanatomical link between gastrointestinal system stimulated during food intake and CNS sites that control feeding behavior (16). Because of morphological and functional connections, in the present study, we have concentrated on the central histaminergic, orexinergic and endocannabinoid systems. Our results demonstrate a rise in the brain histamine concentration in PCA rats, and show that, despite of the potential inhibitory effect of the histaminergic system (2), the food intake in PCA rats is not reduced, just the opposite (Fig.1), which is in agreement with the previous studies (3-5). We suggest that the effect can be explained by a compensatory activation of the central orexigenic systems to maintain energy homeostasis. Indeed, our present results demonstrate a rise in the hypothalamic orexin A concentration in PCA rats. However, we cannot exclude the activation of other orexigenic neuronal systems in these animals. The hypothalamus is a primary site of integration of several factors of central and peripheral origin. Studies have shown that critical to normal regulation of food intake and proper body mass is leptin. Leptin signals nutritional status and may alter interactions between orexigenic and anorectic signals to fulfill its role in energy homeostasis. Rats with long term portocaval shunt have greatly reduced leptin (Table 1). Leptin is produced in adipose tissue and also in stomach. It has recently been reported that not only the lean body mass but also the whole body fat mass are significantly lower in PCA rats within 4 postoperative weeks (17). Noteworthy, among hormones that have been measured weekly in this study was also leptin. The decreased hormone level was noted already on the first week following shunt (17), and as we could show here, the decrement in leptin concentration appears permanent feature of PCA.

As already pointed out despite higher food and water consumption, PCA rats have significantly lower body weight. We suggest that, in addition to possible intestinal digestive and absorption limitations associated with the PCA implantation, other centrally mediated effects influencing energy expenditure are activated in PCA rats, causing the lower food efficiency in these animals. Indeed, the interactions between the orexinergic and histaminergic systems concern not only food intake control, but also the sleep/arousal (18) and central cardiovascular regulations (19). Under life-threatening situations, for example in haemorrhagic shock in rats, both histamine and orexin A are able to activate the sympathetic nervous system and produce the resuscitating effect (20,21). Our recent data demonstrate also an inhibitory effect of SB 334867 on central histamine-induced resuscitating action in haemorrhage-shocked WKY rats (22). Moreover, we have shown the activation of central histamine-connected compensatory mechanisms in PCA rats subjected to haemorrhagic shock (23)

Histamine acting centrally influences both food and water intake in rats (24). Pharmacological blockade of H1 and H2 receptors increases eating and reduces drinking behaviors (24 ). In addition, dietary supplementation with the histamine precursor L-histidine suppresses food consumption and lowers adipose tissue accumulation in rats (25). Our present results suggest that pyrilamine, but not ranitidine, increases food intake in both groups of rats. Interestingly, in rats with PCA, H1 antagonist demonstrates less pronounced effects in comparison to the control group. We suggest that the difference can be explained by a compensatory increase in orexinergic system activity.

Central orexins, acting via OX2 receptor, are able to directly excite histaminergic neurons of the tuberomammillary nucleus (26). However, there are some discrepancies concerning mechanisms of orexinergic-histaminergic interactions. Studies by Ishizuka et al. demonstrate that orexin A-mediated increase in histamine release results in pronounced locomotor activity, but not in food intake (27). It appears, the histaminergic system participates rather in arousal than feeding evoked by orexin A (27). On the other hand, Jorgensen et al. have shown that the orexigenic effect of orexin A is dependent on an intact histaminergic system (28).

Our results demonstrate that SB 334867 significantly reduces food intake in both groups, however, the antagonist has been less effective in PCA rats. We postulate that the difference can be related to a stronger activation of the orexinergic system and only partial inhibition of the effect by SB 334867 in PCA rats. However, we cannot exclude other mechanisms, for example an increase in plasma CCK concentration, which can stimulate different central orexigenic, not only orexinergic, systems.

Our present results demonstrate high plasma concentrations of CCK in PCA rats. In contrast, the study by Nylander et al., performed on rats 4 weeks after PCA implantation does not reveal any differences in CCK concentrations in these animals (29), the discrepancy possibly connected with the short term shunting. On the other hand, these authors convincingly demonstrated that associated with PCA the activation of endocrine cells containing histamine, the enterochromaffin-like (ECL) cells in the oxyntic mucosa of rat is not related under these circumstances to gastrin but to CCK (30). CCK exerted its enhanced trophic effect via CCK2 receptor activation located on ECL cells in rat stomach (29).

Interestingly, high blood levels of CCK can be implicated in reflex-mediated increase in pancreatic exocrine function (31). Additionally, recent data demonstrate that CCK enhances the permeability of the blood-brain barrier to leptin and, in this way, CCK increases energy expenditure by activating hypothalamic the janus kinase/signal transducer and activator of transcription (JAK/STAT) signalling pathway, which is coupled to leptin receptors (32).

Under physiological conditions, endocannabinoids (anandamide and 2-arachidonoylglycerol), acting via CB1 receptor, promote food intake and energy expenditure both at a central and peripheral level, and the blockade of this receptor with rimonabant leads to the reduction of feeding (33). Interestingly, our study demonstrates a dose-related response to rimonabant. In PCA rats, paradoxically, the lowest dose of the antagonist evoked an increase in food intake, whereas the animals treated with the highest dose (10 mg/kg) have eaten significantly more in the post-therapy period. Only in sham rats, a dose of 10 mg/kg was effective in suppression of food intake. We suggest that a lack of the anorexigenic effect of rimonabant in PCA rats can be associated with the increased activity of the orexinergic system, since the study by Crespo et al. indicates that endocannabinoid and orexinergic systems share a common mechanism in food intake regulation (34). Moreover, the hypothalamic orexinergic neurons are involved in cannabinoid CB1 receptor antagonism-mediated reduction of appetite (34). The effect could be associated with an inhibition of orexinergic neurones by cannabinoids (35).

In conclusion, there are adaptive changes in the controlling mechanisms responsible for the food intake in PCA rats. We suggest a compensatory increase in the orexigenic systems activity leading to diminution of the anorexigenic effects of the histaminergic system.

Acknowledgements: Work supported by 502-17-520 and 502-17-689 UM grants. Rimonabant was obtained from the USA NIMH's Chemical Synthesis and Drug Supply Program.

Conflict of interest statement: None declared.


REFERENCES

1. Schwartz J.C, Arrang J.M, Garbarg M, Pollard H, Ruat M. Histaminergic transmission in the mammalian brain. Physiol Rev 1991; 71: 1-51.
2. Gotoh K, Fukagawa K, Fukagawa T, et al. Hypothalamic neuronal histamine mediates the thyrotropin-releasing hormone-induced suppression of food intake. J Neurochem 2007; 103: 1102-1110.
3. Fogel WA, Tuomisto L, Sasiak K, et al. Effect of pargyline on brain N-telemetylhistamine in portocaval shunted rats. Relation to amine neurotransmit-ters. J Neurochem 1994; 62: 615-620.
4. Fogel WA, Michelsen KA, Panula P, Sasiak K, Andrzejewski W. Cerebral and gastric histamine system in atered after portocaval shunt. J Physiol Pharmacol 2001; 52: 657-670.
5. Fogel W.A., Michelsen K.A., Granerus G et al. Neuronal storage of histamine in the brain and tele-methylimidazoleacetic acid excretion in portocaval shunted rats. J Neurochem 2002; 80, 375-382.
6. Eriksson KS, Sergeeva O, Brown RE, Haas HL. Orexin/hypocretin excites the histaminergic neurons of the tuberomammillary nucleus. J Neurosci 2001;21: 9273-9279.
7. Sakurai T. Orexins and orexin receptors: implication in feeding behavior. Regul Pept 1999; 85: 25-30.
8. Di Marzo V, Matias I. Endocannabinoid control of food intake and energy balance. Nat Neurosci. 2005; 8: 585-589.
9. Lee SH, Fisher B. Portocaval shunt in the rat. Surgery 1961; 50: 668-672.
10. Taylor K.M, Snyder S.H. Isotopic microassay of histamine, histidine, histidine decarboxylase and histamine methyltransferase in brain tissue. J Neurochem 1972; 19: 1343-1358.
11. Konturek PC, Konturek JW, Czesnikiewicz-Guzik M, Brzozowski T, Sito E, Konturek SJ. Neuro-hormonal control of food intake: basic mechanisms and clinical implications. J Physiol Pharmacol 2005; 56 (Suppl. 6): 5-25.
12. Schioth HB. G protein-coupled receptors in regulation of body weight. CNS Neurol Disord Drug Targets 2006; 5: 241-249.
13. Valassi E, Scacchi M, Cavagnini F. Neuroendocrine control of food intake. Nutr Metab Cardiovasc Dis 2008; 18: 158-168.
14. Trojniar W, Plucinska K, Ignatowska-Jankowska B, Jankowski M. Damage to the nucleus accumbens shell but not core impairs ventral tegmental area stimulation-induced feeding. J Physiol Pharmacol 2007: 58(Suppl. 3): 63-71.
15. Bojanowska E, Nowak A. Interactions between leptin and exendin-4, a glucagon-like peptide-1 agonist, in the regulation of food intake in the rat. J Physiol Pharmacol 2007; 58: 349-360.
16. Bugajski AJ, Gil K, Ziomber A, Zurowski D, Zaraska W, Thor PJ. Effect of long-term vagal stimulation on food intake and body weight during diet induced obesity in rats. J Physiol Pharmacol 2007; 58(Suppl. 1): 5-12.
17. Dasarathy S, Muc S, Hisamuddin K. et al. Altered expression of genes regulating skeletal muscle mass in the portacaval anastomosis rat. Am J Physiol Gastrointest Liver Physiol 2007; 292:G1105-G1113.
18. McGinty D, Szymusiak R. Hypothalamic regulation of sleep and arousal. Front Biosci 2003; 8: 1074-1083.
19. Tanida M, Niijima A, Shen J, et al. Dose-different effects of orexin-A on the renal sympathetic nerve and blood pressure in urethane-anesthetized rats. Exp Biol Med (Maywood) 2006; 231: 1616-1625.
20. Jochem J, Irman-Florjanc T, Zwirska-Korczala K. Influence of amitriptyline on central histamine-induced reversal of haemorrhagic shock in rats. Inflamm Res 2007, 56(Suppl. 1): S27-S28.
21. Jochem J, Zwirska-Korczala K, Zabielski R, Kato I, Kuwahara A. Cardiovascular effects of centrally acting orexin A in haemorrhage-shocked rats. J Physiol Pharmacol 2006; 57(Suppl. 11): 115-124.
22. Jochem J. Orexin type 1 receptor antagonist SB 334867 inhibits central histamine-induced resuscitating effect in rats subjected to haemorrhagic shock. Inflamm Res, in press.
23. Jochem J, Fogel WA, Lewinski A, Zwirska-Korczala K, Maksymowicz M. Involvement of the histaminergic system in the central cardiovascular regulation in haemorrhage-shocked rats with portocaval anastomosis. Inflamm Res 2007; 56(Suppl. 1): S29-S30.
24. Magrani J, de Castro e Silva E, Varjao B, et al. Histaminergic H1 and H2 receptors located within the ventromedial hypothalamus regulate food and water intake in rats. Pharmacol Biochem Behav 2004; 79: 189-198.
25. Kasaoka S, Tsuboyama-Kasaoka N, Kawahara Y, et al. Histidine supplementation supresses food intake and fat accumulation in rats. Nutrition 2004; 20: 991-996.
26. Bayer L, Eggermann E, Serafin M, et al. Orexins (hypocretins) directly excite tuberomammillary neurons. Eur J Neurosci 2001; 14: 1571-1575.
27. Ishizuka T, Nomura S, Hosoda H, Kangawa K, Watanabe T, Yamatodani A. A role of the histaminergic system for the control of feeding by orexigenic peptides. Physiol Behav 2006; 89: 295-300.
28. Jorgensen EA, Knigge U, Watanabe T, Warberg J, Kjaer A. Histaminergic neurons are involved in the orexigenic effects of orexin-A. Neuroendocrinology 2005; 82: 70-77.
29. Nylander AG, von Friesen CP, Monstein HJ, et al. Increased expression of cholecystokinin-A receptor mRNA in pancreas and cholecystokinin-B receptor mRNA in oxyntic mucosa after porta-caval shunting in rats. Pharmacol Toxicol. 1997; 80:147-51
30. Chen D, Nylander AG, Rehfeld JF, Sundler F, Hakanson R. Hypercholecystokininemia produced by pancreaticobiliary diversion causes gastrin-like effects on enterochromaffin-like cells in the stomach of rats subjected to portacaval shunting or antrectomy. Scand J Gastroenterol. 1993;28:988-92
31. Nawrot-Porabka K, Jaworek J, Leja-Szpak A, et al. Involvement of vagal nerves in the pancreatostimulatory effects of luminal melatonin, or its precursor tryptophan. Study in the rat. J Physiol Pharmacol 2007; 58(Suppl. 6): 81-95.
32. Merino B, Somoza B, Ruiz-Gayo M, Cano V. Circadian rhythm drives the responsiveness of leptin-mediated hypothalamic pathway of cholecystokinin-8. Neurosci Lett 2008; 442: 156-168.
33. Lambert DM, Muccioli GG. Endocannabinoids and related N-acylethanolamines in the control of appetite and energy metabolism: emergence of new molecular players. Curr Opin Nutr Metab Care 2007; 10: 735-744.
34. Crespo I, Gomez de Heras R, Rodriguez de Fonseca F, Navarro M. Pretreatment with subeffective doses of rimonabant attenuates orexigenic actions of orexin A-hypocretin 1. Neuropharmacology 2008; 54: 219-225.
35. Huang H, Acuna-Goycolea C, Li Y, Cheng HM, Obrietan K, van den Pol AN. Cannabinoids excite hypothalamic melanin-concentrating hormone but inhibit hypocretin/orexin neurons: omplications for cannabinoid actions on food intake and cognitive arousal. J Neurosci 2007; 27: 4870-4881.