ORIGINAL INVESTIGATION

Anxiolytic-like profiles of histamine H3 receptor agonistsin animal models of anxiety: a comparative studywith antidepressants and benzodiazepine anxiolytic

Fumikazu Yokoyama & Miki Yamauchi & Masayo Oyama & Kunihiro Okuma &Kaname Onozawa & Takako Nagayama & Rie Shinei & Makoto Ishikawa & Yasuo Sato &Nobukazu Kakui

Received: 15 June 2008 /Accepted: 22 March 2009 /Published online: 9 April 2009# Springer-Verlag 2009

AbstractRationale Histamine H3 receptor functions as a presynapticauto- and hetero-receptor on histaminergic and non-histaminergic neurons in the brain regulating the synapticrelease of numerous neurotransmitters. Therefore, theligands for this receptor have been proposed to be oftherapeutic interest for the treatment of various neuropsy-chiatric disorders. At present, however, the psychopharma-cological profiles of H3 ligands, particularly H3 agonists,have not been extensively studied.Objective The present study investigated the anxiolytic-likeprofiles of H3-selective agonists in a variety of classical(benzodiazepine-sensitive) and atypical (antidepressant-effective) animal models of anxiety. Comparator drugsused were diazepam and both fluvoxamine and desipraminein the former and latter models, respectively.Results H3 agonist R--methylhistamine and immepip wereinactive in rat elevated plus maze test and Vogel type conflicttest where diazepam (5 mg/kg) produced significantanxiolytic-like effects. Meanwhile, these H3 agonists (1030 mg/kg) significantly reduced isolation-induced vocal-izations in guinea pig pups and isolation-induced aggressivebehavior in mouse residentintruder test. Moreover, in ratconditioned fear stress test, R--methylhistamine (30 mg/kg)and immepip (10 mg/kg) significantly decreased freezingtime, which were completely reversed by concomitant

treatment with H3 antagonist, thioperamide (10 mg/kg). Incontrast to the limited efficacy obtained with desipramine(30 mg/kg), fluvoxamine (2060mg/kg) exhibited anxiolytic-like effects in all the latter three atypical models.Conclusions These data suggest that the H3 agonists mayhave anxiolytic-like effects similar to those of selective sero-tonin reuptake inhibitors but not benzodiazepine anxiolyticsand represent a novel strategy for the treatment of someanxiety disorders in which selective serotonin reuptakeinhibitors are prescribed.

Keywords Histamine H3 receptor .R--Methylhistamine .

Immepip . Selective serotonin reuptake inhibitor . Elevatedplus maze test . Vogel type conflict test . Isolation-inducedvocalization test . Residentintruder test . Conditioned fearstress test . Anxiolytic-like effects

Introduction

Since Schwartz (1975) found the existence of histaminergicneurons, increasing evidence supports a role for histamineas a neurotransmitter and neuromodulator in various brainfunctions, including cognition, emotion, stress, and feeding(Leurs et al. 2005). In four different subtypes of G protein-coupled histamine receptors (H1H4), H3 receptor wasdiscovered by Arrang et al. (1983) using cortical slice cul-tures, where auto-regulation of histamine release was demon-strated. This receptor subtype, as a presynaptic autoreceptor,suppresses the synthesis and release of histamine in the centralnervous system (CNS; Arrang et al. 1985, 1987). It is alsoshown to behave as a presynaptic heteroreceptor, modulatingthe release of several important neurotransmitters such asserotonin (5-HT; Schlicker et al. 1988), noradrenaline

Psychopharmacology (2009) 205:177187DOI 10.1007/s00213-009-1528-1

F. Yokoyama :M. Yamauchi :M. Oyama :K. Okuma :K. Onozawa : T. Nagayama : R. Shinei :M. Ishikawa :Y. Sato :N. Kakui (*)Pharmaceutical Research Center, Meiji Seika Kaisha, Ltd.,760 Morooka-cho,Kohoku-ku, Yokohama 222-8567, Japane-mail: nobukazu_kakui@meiji.co.jp

(Schlicker et al. 1994), dopamine (Schlicker et al. 1993),and acetylcholine (Clapham and Kilpatrick 1992). Inconsideration of the multiple roles of these neurotransmittersand the widespread existence of H3 receptor in the CNS,the specific ligands for this receptor could be expected tohave therapeutic potential in the treatment of various CNSdisorders.

With the latest advances in pharmacological investiga-tions, the putative therapeutic application of H3 antagonistshas been discussed in a recent review (Leurs et al. 2005).There are strong indications that H3 antagonists can beused to correct sleep disorder, cognitive disorder, or atten-tion deficit hyperactivity disorder (Witkin and Nelson2004). Oppositely, there is quite limited information avail-able regarding the psychopharmacological profiles ofH3 agonists, in particular their therapeutic values foranxiety disorders. A previous study reported that R--methylhistamine at the single dose did not affect time spentin open arms in elevated plus maze test (Perez-Garcia et al.1999). Their study, however, seems to require furtherpharmacological elucidation for dose response analysis(efficacy at the higher doses of R--methylhistamine) andevaluation of another H3 agonist.

Anxiety disorders are categorized by their diagnosticcriteria into obsessive-compulsive disorder, panic disorder,social phobia, generalized anxiety disorder, and posttrau-matic stress disorder. It is generally accepted that theclinical effectiveness of benzodiazepines is limited togeneralized anxiety disorder, social phobia, and panicdisorder (Borsini et al. 2002). Meanwhile, antidepressantssuch as selective serotonin reuptake inhibitors (SSRIs),tricyclic antidepressants (TCAs), and serotonin noradrena-line reuptake inhibitors (SNRIs) have been shown to beeffective in patients with a wider spectrum of anxietydisorders (Kent et al. 1998; McLeod et al. 1990). Thus, asis well known, the antidepressant drugs are clinicallybeneficial for not only depressive but also anxiety disor-ders. Here, the critical issue to be considered is that themost commonly used classical animal models of anxiety,such as the elevated plus maze test and the Vogel typeconflict test, which are sensitive to benzodiazepines, but notantidepressants (Griebel et al. 1997; Petersen and Lassen1981). This discrepancy clearly serves to indicate thenecessity for preclinical pharmacological studies, whichreflect the efficacy of clinically more useful drugs. Fromthese viewpoints, the following atypical anxiety modelsare a matter worthy of further preclinical examinations. Theefficacy of various types of antidepressants has beendemonstrated in the isolation-induced vocalization inguinea pig pups and the residentintruder test in mice(Borsini et al. 2002; Millan et al. 2001). The anxiolytic-likeaction of SSRIs but not TCAs has also been found in the ratconditioned fear stress test (Hashimoto et al. 1996).

Accordingly, these atypical models may have great poten-tial to predict clinical outcome of antidepressants or somerelated compounds in anxiety disorders.

The purpose of the present study was to characterize theanxiolytic-like profiles of H3 agonists with a variety ofanimal models, including classical and atypical ones. Thus,we first assessed the effects of R--methylhistamine andimmepip in comparison to the benzodiazepine anxiolytic,diazepam, in the classical anxiety models (elevated plusmaze test and Vogel type conflict test). And, second, weassessed the effects of the two H3 agonists in three atypicalanimal models (guinea pig isolation-induced vocalizationtest, mouse residentintruder test, and rat conditioned fearstress test) in comparison to the other reference drugs,fluvoxamine (serotonergic antidepressant) and desipramine(noradrenergic antidepressant). With our current observa-tions, the similarity or difference of the anxiolytic-likeprofiles among H3 agonists and the existing drugs and thepathophysiological and therapeutic insights of H3 receptoragonism will be discussed.

Materials and methods

Animals

Male Wistar rats (for Vogel type conflict test; 120200 g,Clea Japan, Tokyo, Japan: for conditioned fear stress test;200250 g, Japan SLC, Hamamatsu, Japan) and maleSpragueDawley rats (for elevated plus maze test, 150180 g, Charles River Japan, Atsugi, Japan) housed ingroups of five to six upon arrival were used. Male ICRmice (4048 g, Japan SLC) were housed ten to 12 per cageand used for residentintruder test and locomotion measure-ments. Pregnant guinea pigs were obtained from Japan SLCand housed in individual cages until parturition andthereafter with their pups throughout the study. All theseanimals were maintained under a 12-h light/dark cycle(light on 0700 hours, light off 1900 hours) in a temperature-and humidity-controlled room. Chow and water wereavailable ad libitum except during experiment. Animal carewas performed according to the protocols reviewed by theEthical Committee for Animal Experiment in Meiji SeikaPharmaceutical Research Center.

Drugs

R--methylhistamine dihydrochloride, thioperamide male-ate, desipramine hydrochloride, immepip dihydrobromide(TOCRIS, Bristol, UK), and fluvoxamine maleate (MeijiSeika, Tokyo, Japan) were dissolved in sterile saline andinjected intraperitoneally (i.p.) in a volume of 10 ml/kg(mice), 1 ml/kg (rats), or 2 ml/kg (guinea pig pups).

178 Psychopharmacology (2009) 205:177187

Diazepam (Wako Pure Chemical, Osaka, Japan) wassuspended in 0.5% (v/v) carboxymethylcellulose (Sigma-Aldrich) solution and injected i.p. to animals. In oraladministration tests, fluvoxamine and desipramine weredissolved in distilled water. All of these preparations weremade immediately before their injection on the day ofexperiment. The dosing regimen of H3 agonists (Cannonand Hough 2005; Malmberg-Aiello et al. 1994) and otherdrugs (Li et al. 2001; Molewijk et al. 1996; Sanchez andHyttel 1994; Wesolowska et al. 2006) used in this studywas based on previous reports and our pilot studies (datanot shown).

Apparatus and procedure

Elevated plus maze test

The test was performed according to the method previouslyreported (Pellow et al. 1985). Fifteen minutes after the i.p.administration of R--methylhistamine (10 and 30 mg/kg)and immepip (10 and 30 mg/kg) or 30 min after theadministration of diazepam (5 mg/kg), each rat was placedon the center of platform (1010 cm) facing one of theclosed arms (5010 cm) with walls (40 cm height) andallowed to explore within the maze for 5 min. The behaviorof rats was monitored by CCD camera located above themaze and analyzed by a computer system (Target System,Neuroscience, Tokyo, Japan). The assessment of anxiolytic-like activity was made with the measurement of the timespent in open arms.

Vogel type conflict test

A modified conflict test originally reported by Vogel et al.(1971) was employed to measure the anti-conflict activity.An operant chamber (312531 cm, Muromachi Kikai,Tokyo, Japan) equipped with a spout for water supply wasused. In the training session, the rats were deprived of waterfor 24 to 31 h and then individually placed in the chamberto be trained to lick the spout for water with no electricshock. On the following day, they were again deprived ofwater for 24 to 31 h, and the number of spout lickingresponse of each rat was counted for 10 min as a prevalueusing an automated licking counter. The rats were randomlyand evenly assigned to each of seven experimental groupsto avoid any significant difference among groups. The nextday, they were again deprived of water for 24 to 31 h andunderwent test session. During the measurement, an electricshock of 2 mA for 0.2 s from grids connected to a shockgenerator (Model SGS-002, Muromachi Kikai) was givento the animals for every three licking responses. At 15 minafter the administration of R--methylhistamine (10 and30 mg/kg) and immepip (10 and 30 mg/kg) or 30 min after

diazepam (5 mg/kg), the number of licking responses in theanimals was counted for 10 min.

Isolation-induced vocalization test

The test was conducted according to the previous report(Molewijk et al. 1996). The test plastic cage (2316.512 cm) equipped with illumination and microphone on theinside was placed in sound-proof boxes, and this box wasclosed during the test. From 7 days of age, guinea pig pupsentered a pretest in which their vocalizations were recordedby mini disc recorder for 5 min. Only animals that exhibitedmore than 600 counts of vocalization were used for drugevaluation. On the next day, R--methylhistamine (3, 10,and 30 mg/kg) and immepip (3, 10, and 30 mg/kg) wereinjected i.p., and the pups were immediately returned to thehome cage and remained there with their mother and littermates for 15 min before testing. Fluvoxamine (30 mg/kg)and desipramine (30 mg/kg) were administered i.p. 30 minprior to the test. Then, the pups were moved into the testplastic cage, and their vocalizations were recorded for 5 min.The counts of vocalization were measured by observersblinded to the experimental groups of the animals.

Residentintruder test

The test was performed as described by Sanchez and Hyttel(1994). After an acclimation period, the resident mice werekept isolated for 6 weeks in the metal cage (322218 cm). All of these mice underwent a pretest for thedetection of their levels of aggressiveness. In this pretest, anintruder mouse housed in groups was placed in a residenthome cage, and duration of attacks by the resident on theintruder mouse was measured for 10 min by observer blindto the treatment conditions. The attack was defined asbiting or as an attempt to bite the intruder mouse. Onlymice with total attack time of longer than 10 s wereincluded in drug administration test and randomly assignedto two or three groups based on their pretest values. In thetest session, isolated mice were administered i.p. with R--methylhistamine (10 and 30 mg/kg), immepip (10 and30 mg/kg), or vehicle 15 min prior to the test. Fluvoxamine(20 and 40 mg/kg) and desipramine (10 and 30 mg/kg)were given i.p 30 min before testing. Then, each of theintruder mice was confronted with each of the resident micein the resident home cage, and the total attack time wasmeasured for 10 min.

Conditioned fear stress test

In this procedure, rats were conditioned to anticipate a footshock. On the first day, they were individually subjected toinescapable electric foot shocks for a total of 5 min (1 mA

Psychopharmacology (2009) 205:177187 179

of scrambled shock, shock duration of 30 s5, and anintershock interval of 30 s) in an operant chamber (312531 cm, Muromachi Kikai) with a stainless grid floor. Ascrambled current shock was delivered via a shockgenerator (Model SGS-002). The test session was per-formed about 24 h after the exposure to electric foot shock.The rats were again placed in the same box and observedfor 5 min in the shock chamber with no current applied tothe floor of the chamber. Freezing was defined as theabsence of all observable movement of the skeleton and thevibrissae, except for those related to respiration, andmeasured by the observer who was blind to the experimen-tal groups of the animals. In acute treatment tests, R--methylhistamine (10 and 30 mg/kg) and immepip (3 and10 mg/kg) were injected i.p. 15 min before the observationperiod. Fluvoxamine (30 and 60 mg/kg) and desipramine(30 and 60 mg/kg) were administered orally 60 min beforetesting. As our preliminary studies strongly suggested thatSSRIs need to be repetitively injected for the expression ofsignificant anxiolytic actions, oral but not i.p. route wasadopted to minimize stress that the animals may receive atthe time of injections. In acute co-administration tests, H3agonists and thioperamide (10 mg/kg) were given i.p. 15 minbefore testing. In repeated treatment tests, fluvoxamine (30and 60 mg/kg) and desipramine (30 and 60 mg/kg) wereorally administered once a day for 5 days. On day 1, thesedrugs were given 60 min after the electric foot shock, andlater on days 24, the administration was conducted between1300 and 1700 hours. On the day of the experiment (day 5),the drugs were given 60 min before testing.

Locomotion measurements

Effects of H3 receptor ligands, fluvoxamine, and desipramineon spontaneous locomotion were evaluated at the same timeintervals as in the mouse residentintruder test. Immediatelyafter the injection of R--methylhistamine (15, 30, and60 mg/kg), immepip (10, 30, and 100 mg/kg), thioperamide(10 and 20 mg/kg), fluvoxamine (20 and 40 mg/kg), anddesipramine (10 and 30 mg/kg), mice were individuallyplaced in an open top plastic activity cage (172412 cm).Locomotion measurement system (DAS System MultidigitalCounter, Neuroscience) consisted of an infrared sensorpositioned just above the cage and the application software(Multidigital 32 Port Count System). After 15 min (for H3ligands) or 30 min (for reference drugs), the locomotoractivity was measured for 10 min and analyzed by thesoftware.

Statistics

All results were expressed as meanSEM. Statisticalanalysis was performed with the SAS statistical package

(SAS Institute Japan, Tokyo, Japan). Statistical significanceof the differences among multiple groups was testedby one-way analysis of variance followed by Dunnettsmultiple comparison test. A two-tailed Students t test wasused to evaluate the difference between two experimentalgroups. Differences were considered significant at p

Fig. 3a). A similar tendency was observed by the adminis-tration of immepip, but statistically significant changeswere detected only in the high dose group (345.884.5 in30 mg/kg, p

locomotion levels (96.136.9, p

mission in the classical animal models. An H1 receptorantagonist, chlorpheniramine, improved anxiety in the ratelevated plus maze test and the open field behavioral test(Hasenohrl et al. 1999). Yanai et al. (1998) reported thatanxiety was decreased in the elevated plus maze test ofthe mice lacking H1 receptors. Furthermore, the anxiolyticdrugs such as diazepam and serotonin 5-HT1A agonistbuspirone significantly inhibited the brain histamine turn-overs in rodents (Oishi et al. 1992; Oishi et al. 1986).Diazepam also inhibited an anxiogenic-like effect inducedby thioperamide in the mouse light/dark test (Imaizumiand Onodera 1995). Taken in concert, these findingssupport the notion that the histaminergic system enhancesan anxiogenic-like response mainly via activation of H1receptors and inhibition of H3 receptors. In the presentstudy, however, stimulation of H3 receptors by R--methylhistamine and immepip, at either of the tested doses(10 and 30 mg/kg), did not exhibit anxiolytic-like effects inboth the elevated plus maze test (Fig. 1) and the Vogel typeconflict test (Fig. 2), as opposed to diazepam. The result inthe former was consistent with the previous report that R--

methylhistamine (10 mg/kg) did not change the time spenton the open arm in the same test (Perez-Garcia et al. 1999).These findings strongly suggest the substantial contributionof postsynaptic (H1 receptor inhibition) but not presynaptic(H3 receptor stimulation) modulation of histamine-dependentanxiogenic responses in the benzodiazepine-sensitivemodels. One potential clue for interpretation is the existenceof complex postsynaptic histamine receptor systems con-sisting of anxiogenic-like H1 and anxiolytic-like H2 recep-tors, as demonstrated by the classical animal model (mouselight/dark test; Malmberg-Aiello et al. 2002). Changes in therelease of endogenous histamine by H3 agonists may there-fore affect postsynaptic signal transduction and subsequentexpression of anxiety responses via these receptors in ahighly complex manner. Future research should explore thisissue furthermore.

In clinical practice, antidepressants such as SSRIs andTCAs are now becoming the first-line therapy for treating a

Fig. 7 Effects of R--methylhistamine (RAMH; a), immepip (b), andreference drugs (c) on spontaneous locomotion in mice. R--methylhistamine (RAMH; 15, 30, and 60 mg/kg) and immepip (10,30, and 100 mg/kg) were i.p. administered 15 min before measure-ment of the locomotor activities for 10 min. Fluvoxamine (20 and40 mg/kg) and desipramine (10 and 30 mg/kg) were i.p. injected30 min before the test. Data are represented as mean SEM (n=8).*p

variety of anxiety disorders. In marked contrast to this,these antidepressants fundamentally exhibit no therapeuticefficacy in classical anxiety models like elevated plus mazetest and Vogel type conflict test, where pharmacologicalvalidation of benzodiazepine anxiolytics has often beenconducted (Griebel et al. 1997; Petersen and Lassen 1981).Current findings may challenge these issues and vocal-izations in mammals triggered by maternal separation havebeen regarded as an appropriate model that may detect theanxiolytic-like efficacy of antidepressants. In the ratmaternal separation studies, the serotonergic antidepres-sants are effective in reducing the number of vocalizations(Olivier et al. 1998), but the drugs enhancing noradrenergicneurotransmission are shown to rather worsen the anxietyresponses (Winslow and Insel 1990). Accordingly, theseappear to be not so suitable as to evaluate anxiolytic-likeactions of antidepressants in general. However, when it isthe case with guinea pig pups, the similar efficacy has beendemonstrated in common with various types of antidepres-sants, including TCAs, SSRIs, SNRIs, and monoamineoxidase inhibitors (Borsini et al. 2002). We indeed confirmedthe reduction of the number of vocalizations in isolatedguinea pig pups by fluvoxamine and desipramine (Fig. 3c),which are consistent with the previous reports (Molewijket al. 1996). Both R--methylhistamine and immepip dosedependently suppressed vocalizations (Fig. 3a), suggestingthat H3 agonists as well as antidepressants function as ananxiolytic. Although there has been no direct evidence thathistamine induces vocalizations in guinea pig pups via itspostsynaptic receptors, an H1 antagonist mepyramine sup-pressed foot shock-dependent vocalizations in rats (Sanchez2003). Combined with our data, there may be a causal rela-tionship between histaminergic neurotransmission andvocalization responses triggered by anxiety or stress. Thesepoints should also be addressed in the future pharma-cological studies.

To more precisely determine the neurobehavioral pheno-type of H3 agonist-induced anxiolytic-like effects and, inparticular, their similarities to the serotonergic and norad-renergic drugs, we next performed the mouse residentintruder test and the rat conditioned fear stress test.According to an earlier study, citalopram (SSRI) andclomipramine (TCA with relatively high affinity for 5-HTtransporter) are effective equally in both the residentintruder test and the forced swimming test at the testeddoses, whereas an effective dose of venlafaxine (SNRI) andreboxetine (NRI) in residentintruder test is higher than thatin forced swimming test (Millan et al. 2001). Thus, one canassume that the residentintruder test is relatively sensitiveto the serotonergic antidepressants. Consistent with theprevious studies (Sanchez and Hyttel 1994), fluvoxaminereduced isolation-induced aggression (Fig. 4b) accompa-nied by no significant changes in locomotor activities

(Fig. 7c). Likewise, R--methylhistamine and immepipsuppressed aggressive behavior induced by social isolation(Fig. 4a). In concert with the previous finding that H1receptor-deficient mice showed reduced aggressive behavior(Yanai et al. 1998), the efficacy of R--methylhistamine andimmepip is probably derived from decrease in bothhistamine release and its subsequent action to H1 receptors,at least in this model. Although the noradrenergic anti-depressant desipramine (30 mg/kg) significantly suppressedthe duration of attack behavior (Fig. 4c), it also inducedhypolocomotion at the apparently effective dose (Fig. 7c),highlighting relative importance of serotonergic neurotrans-mission for the specific expression of anxiolytic-like actionin the residentintruder test.

Differential responses between the two reference drugs(fluvoxamine and desipramine) in the conditioned fearstress test appear to partially coincide with a previous reportthat SSRIs but not classical noradrenergic antidepressantssuch as maprotiline exhibited the anxiolytic-like efficacy inthis test (Hashimoto et al. 1996). Comparable results sup-porting our data are also obtained where the effects offluvoxamine on conditioned freezing were enhanced by itschronic treatment (Li et al. 2001). Under these experimentalconditions, single treatment with R--methylhistamineand immepip significantly reduced freezing time (Fig. 5a),and these were completely antagonized by thioperamide(Fig. 5b). Contextual fear conditioning implies learning therelationship between aversive events and the environmentalstimuli that predict such events. Also, H3 receptor ligandsinfluence memory depending on the brain region, the natureof the cognitive task involved, and the pharmacologicalprofiles of tested compounds. Indeed, the cognitive perfor-mance of rats in object recognition and a passive avoidanceresponse, both of which are possibly related to the functionof the frontal cortex, was strongly impaired by R--methylhistamine (Blandina et al. 1996), while thioperamideimproved the response latency in the passive avoidance test(Meguro et al. 1995). Conversely, a beneficial effect of R--methylhistamine has been reported in rodent spatiallearning and memory in a water maze test that primarilyreflect the function of the hippocampus in rodents (Smithet al. 1994). Thus, such differences in the behavioraltasks and their relevant brain regions may explain thisdiscrepancy. As for the contextual fear conditioning,previous studies showed that local administration of H3agonists and H3 antagonists (or inverse agonists) immedi-ately after the fear conditioning into the basolateralamygdala enhanced (Cangioli et al. 2002) and impaired(Passani et al. 2001) memory consolidation, respectively, asdemonstrated by prolonged and shortened freezing time atretention test. These results are suggestive of a potentialmodulatory influence on cognition rather than anxietyof H3 receptor ligands that were administered in memory

184 Psychopharmacology (2009) 205:177187

acquisition phase. Compared with our studies, the mostdifferent is the timing of H3 agonist administration. Whenconsidering our dosing condition (just before the test), thecritical point is whether the H3 agonists administeredimpair the recall of memory. In addition to the foremen-tioned report (Smith et al. 1994), another (Rubio et al.2002) and our pilot study (unpublished data) also showedno amnesic effects of R--methylhistamine regarding therecovery of spatial memory in the rat water maze. Thesefindings now led us to conclude that the observed responsesby H3 agonists administered in the retention phase possiblyseem to reflect genuine anxiolytic-like properties of thecompounds in the conditioned fear stress test.

The exact mechanisms whereby H3 agonists exert theiraction in animal anxiety models have to await fullelucidation. H3 agonists behave as a presynaptic hetero-receptor to suppress the release of 5-HT (Schlicker et al.1988), noradrenaline (Schlicker et al. 1994), dopamine(Schlicker et al. 1993), -aminobutyric acid (Garcia et al.1997), and acetylcholine (Clapham and Kilpatrick 1992).Meanwhile, diazepam, one of the reference drugs used inthis study, is reported to directly or indirectly inhibitneurotransmission of 5-HT, noradrenaline, dopamine, andacetylcholine (Danneberg and Weber 1983). Our observa-tions that the H3 agonists failed to show anxiolytic-likeeffects in the elevated plus maze and the Vogel type conflicttests (Figs. 1, 2), therefore, suggest the importance of -aminobutyric acid receptor activation per se in diazepam-sensitive classical models. On the other hand, significantanxiolytic-like effects in three atypical models appeared tobe shared between the H3 agonists and fluvoxamine(Figs. 3, 4, 5, and 6). Whereas the basic mechanism ofaction of SSRIs depends on the increase of 5-HT in thesynaptic cleft (Bosker et al. 1995), H3 agonists ratherinhibit the release of 5-HT in the brain (Schlicker et al.1988). Thus, it is far less likely that the changes in mono-amine signals, in particular the reduction of serotonergicneurotransmission, underlie the anxiolytic-like effects of H3agonists. Other considerations to help explain our observa-tions are the following findings regarding an intimateassociation between histamine and corticotrophin-releasingfactor (CRF) in the regulation of anxiety and stress-relatedresponses. Originally described as a pivotal mediator ofacute neuroendocrine responses to stress, CRF is currentlyenvisioned as a peptide neurotransmitter involved in thepathogenesis of anxiety and depressive disorders (Haugeret al. 2006). Anxiety behavior can be triggered by centralinjection of CRF (Swerdlow et al. 1986), whereas CRF1receptor antagonists exhibit anxiolytic-like activities in theisolation-induced vocalization test (Hodgson et al. 2007),the residentintruder test (Farrokhi et al. 2004), and theconditioned fear stress test (Hikichi et al. 2000). In fact,selective CRF1 receptor antagonists have been developed as

a novel therapeutic approach for the treatment of anxietyand depression (Steckler and Dautzenberg 2006). On theother hand, earlier studies strongly suggested the patho-physiological importance of histamine by demonstratingthat a variety of stress exposures, including restraint, coldtemperature, and foot shock increase brain histamine turn-over (Taylor and Snyder 1971; Yoshitomi et al. 1986). Anintracerebroventricular infusion of histamine increased thelevel of CRF mRNA in the hypothalamic paraventricularnucleus (Kjaer et al. 1998) and the subsequent release ofACTH and corticosterone into plasma (Kjaer et al. 1994),suggesting positive regulation of CRF system by hista-mine. Moreover, stimulation of H3 receptor by R--methylhistamine, which is known to reduce the releaseof endogenous histamine (Ferretti et al. 1998), inhibitedthe ACTH and prolactin responses to restraint stress(Knigge et al. 1999). These observations enable us topropose that the putative mechanism of anxiolytic effectsof H3 agonists depends on their functional modulation ofCRF neurons via inhibition of histamine release.

The present study clearly demonstrated that R--methylhistamine and immepip are effective in all of theatypical anxiety models tested, and these appear to bequalitatively analogous to the profiles of SSRI but notbenzodiazepine. Although the underlying mechanismsremain to be elucidated, the observed competitive interac-tion between the two H3 agonists and thioperamide mayguarantee that the primary action of R--methylhistamineand immepip resides in the activation of H3 receptors.Additional mechanistic studies using H3 receptor knockoutmice will be supportive for a further substantiation of thisnotion. The present outcome may also provide an oppor-tunity to further investigate the pathophysiological roles ofH3 receptor in anxiety and stress-related responses andsimultaneously raise the possibility that the H3 agonistsoffer an effective therapeutic option for the treatment ofsome forms of anxiety disorders in which SSRIs arecommonly prescribed.

References

Arrang JM, Garbarg M, Schwartz JC (1983) Auto-inhibition of brainhistamine release mediated by a novel class (H3) of histaminereceptor. Nature 302:832

Arrang JM, Garbarg M, Schwartz JC (1985) Autoregulation ofhistamine release in brain by presynaptic H3-receptors. Neuro-science 15:553

Arrang JM, Garbarg M, Schwartz JC (1987) Autoinhibition ofhistamine synthesis mediated by presynaptic H3-receptors.Neuroscience 23:149

Blandina P, Giorgetti M, Bartolini L, Cecchi M, Timmerman H, LeursR, Pepeu G, Giovannini MG (1996) Inhibition of corticalacetylcholine release and cognitive performance by histamineH3 receptor activation in rats. Br J Pharmacol 119:1656

Psychopharmacology (2009) 205:177187 185

Borsini F, Podhorna J, Marazziti D (2002) Do animal models ofanxiety predict anxiolytic-like effects of antidepressants? Psycho-pharmacology (Berl) 163:121

Bosker FJ, Klompmakers AA, Westenberg HG (1995) Effects ofsingle and repeated oral administration of fluvoxamine onextracellular serotonin in the median raphe nucleus and dorsalhippocampus of the rat. Neuropharmacology 34:501

Cangioli I, Baldi E, Mannaioni PF, Bucherelli C, Blandina P, PassaniMB (2002) Activation of histaminergic H3 receptors in the ratbasolateral amygdala improves expression of fear memory andenhances acetylcholine release. Eur J Neurosci 16:521

Cannon KE, Hough LB (2005) Inhibition of chemical and low-intensity mechanical nociception by activation of histamine H3receptors. J Pain 6:193

Clapham J, Kilpatrick GJ (1992) Histamine H3 receptors modulate therelease of [3H]-acetylcholine from slices of rat entorhinal cortex:evidence for the possible existence of H3 receptor subtypes. Br JPharmacol 107:919

Danneberg P, Weber KH (1983) Chemical structure and biologicalactivity of the diazepines. Br J Clin Pharmacol 16(Suppl 2):231 S

Farrokhi C, Blanchard DC, Griebel G, Yang M, Gonzales C,Markham C, Blanchard RJ (2004) Effects of the CRF1 antagonistSSR125543A on aggressive behaviors in hamsters. PharmacolBiochem Behav 77:465

Ferretti C, Blengio M, Ghi P, Adage T, Portaleone P, Ricci Gamalero S(1998) Hypothalamic histamine release in normal and stressed ratsis affected by sex and aging. Pharmacol Biochem Behav 59:255

Garcia M, Floran B, Arias-Montano JA, Young JM, Aceves J (1997)Histamine H3 receptor activation selectively inhibits dopamine D1receptor-dependent [3H]GABA release from depolarization-stimulated slices of rat substantia nigra pars reticulata. Neuroscience80:241

Griebel G, Rodgers RJ, Perrault G, Sanger DJ (1997) Risk assessmentbehaviour: evaluation of utility in the study of 5-HT-related drugs inthe rat elevated plus-maze test. Pharmacol Biochem Behav 57:817

Haas H, Panula P (2003) The role of histamine and the tuberomamillarynucleus in the nervous system. Nat Rev Neurosci 4:121

Hasenohrl RU, Weth K, Huston JP (1999) Intraventricular infusion ofthe histamine H(1) receptor antagonist chlorpheniramineimproves maze performance and has anxiolytic-like effects inaged hybrid Fischer 344xBrown Norway rats. Exp Brain Res128:435

Hashimoto S, Inoue T, Koyama T (1996) Serotonin reuptake inhibitorsreduce conditioned fear stress-induced freezing behavior in rats.Psychopharmacology (Berl) 123:182

Hauger RL, Risbrough V, Brauns O, Dautzenberg FM (2006)Corticotropin releasing factor (CRF) receptor signaling in thecentral nervous system: new molecular targets. CNS NeurolDisord Drug Targets 5:453

Hikichi T, Akiyoshi J, Yamamoto Y, Tsutsumi T, Isogawa K, NagayamaH (2000) Suppression of conditioned fear by administration ofCRF receptor antagonist CP-154, 526. Pharmacopsychiatry 33:189

Hodgson RA, Higgins GA, Guthrie DH, Lu SX, Pond AJ, MullinsDE, Guzzi MF, Parker EM, Varty GB (2007) Comparison of theV1b antagonist, SSR149415, and the CRF1 antagonist, CP-154,526, in rodent models of anxiety and depression. PharmacolBiochem Behav 86:431

Imaizumi M, Onodera K (1995) Diazepam reduces anxiogenic effectsof thioperamide in a light/dark test in mice. Ann Psychiatry:Basic Clin Neurosci 5:239

Kent JM, Coplan JD, Gorman JM (1998) Clinical utility of theselective serotonin reuptake inhibitors in the spectrum of anxiety.Biol Psychiatry 44:812

Kjaer A, Larsen PJ, Knigge U, Warberg J (1994) Histaminergicactivation of the hypothalamic-pituitary-adrenal axis. Endocri-nology 135:1171

Kjaer A, Larsen PJ, Knigge U, Jorgensen H, Warberg J (1998)Neuronal histamine and expression of corticotropin-releasinghormone, vasopressin and oxytocin in the hypothalamus: relativeimportance of H1 and H2 receptors. Eur J Endocrinol 139:238

Knigge U, Soe-Jensen P, Jorgensen H, Kjaer A, Moller M, Warberg J(1999) Stress-induced release of anterior pituitary hormones:effect of H3 receptor-mediated inhibition of histaminergic activityor posterior hypothalamic lesion. Neuroendocrinology 69:44

Leurs R, Bakker RA, Timmerman H, de Esch IJ (2005) The histamineH3 receptor: from gene cloning to H3 receptor drugs. Nat RevDrug Discov 4:107

Li XB, Inoue T, Hashimoto S, Koyama T (2001) Effect of chronicadministration of flesinoxan and fluvoxamine on freezingbehavior induced by conditioned fear. Eur J Pharmacol 425:43

Malmberg-Aiello P, Ipponi A, Bartolini A, Schunack W (2002) Mouselight/dark box test reveals anxiogenic-like effects by activation ofhistamine H1 receptors. Pharmacol Biochem Behav 71:313

Malmberg-Aiello P, Lamberti C, Ghelardini C, Giotti A, Bartolini A(1994) Role of histamine in rodent antinociception. Br JPharmacol 111:1269

McLeod DR, Hoehn-Saric R, Zimmerli WD, De Souza EB, Oliver LK(1990) Treatment effects of alprazolam and imipramine: physio-logical versus subjective changes in patients with generalizedanxiety disorder. Biol Psychiatry 28:849

Meguro K, Yanai K, Sakai N, Sakurai E, Maeyama K, Sasaki H,Watanabe T (1995) Effects of thioperamide, a histamine H3antagonist, on the step-through passive avoidance response andhistidine decarboxylase activity in senescence-accelerated mice.Pharmacol Biochem Behav 50:321

Millan MJ, Dekeyne A, Papp M, La Rochelle CD, MacSweeny C,Peglion JL, Brocco M (2001) S33005, a novel ligand at bothserotonin and norepinephrine transporters: II. Behavioral profilein comparison with venlafaxine, reboxetine, citalopram, andclomipramine. J Pharmacol Exp Ther 298:581

Molewijk HE, Hartog K, van der Poel AM, Mos J, Olivier B (1996)Reduction of guinea pig pup isolation calls by anxiolytic andantidepressant drugs. Psychopharmacology (Berl) 128:31

Oishi R, Itoh Y, Saeki K (1992) Inhibition of histamine turnover by 8-OH-DPAT, buspirone and 5-hydroxytryptophan in the mouse andrat brain. Naunyn Schmiedebergs Arch Pharmacol 345:495

Oishi R, Nishibori M, Itoh Y, Saeki K (1986) Diazepam-induceddecrease in histamine turnover in mouse brain. Eur J Pharmacol124:337

Olivier B, Molewijk HE, van der Heyden JA, van Oorschot R,Ronken E, Mos J, Miczek KA (1998) Ultrasonic vocalizations inrat pups: effects of serotonergic ligands. Neurosci Biobehav Rev23:215

Passani MB, Cangioli I, Baldi E, Bucherelli C, Mannaioni PF,Blandina P (2001) Histamine H3 receptor-mediated impairmentof contextual fear conditioning and in-vivo inhibition ofcholinergic transmission in the rat basolateral amygdala. Eur JNeurosci 14:1522

Pellow S, Chopin P, File SE, Briley M (1985) Validation of open:closed arm entries in an elevated plus-maze as a measure ofanxiety in the rat. J Neurosci Methods 14:149

Perez-Garcia C, Morales L, Cano MV, Sancho I, Alguacil LF (1999)Effects of histamine H3 receptor ligands in experimental modelsof anxiety and depression. Psychopharmacology (Berl) 142:215

Petersen EN, Lassen JB (1981) A water lick conflict paradigm usingdrug experienced rats. Psychopharmacology (Berl) 75:236

Rubio S, Begega A, Santin LJ, Arias JL (2002) Improvement ofspatial memory by (R)-alpha-methylhistamine, a histamine H(3)-receptor agonist, on the Morris water-maze in rat. Behav BrainRes 129:77

Sanchez C (2003) Stress-induced vocalisation in adult animals. Avalid model of anxiety? Eur J Pharmacol 463:133

186 Psychopharmacology (2009) 205:177187

Sanchez C, Hyttel J (1994) Isolation-induced aggression in mice:effects of 5-hydroxytryptamine uptake inhibitors and involvementof postsynaptic 5-HT1A receptors. Eur J Pharmacol 264:241

Schlicker E, Betz R, Gothert M (1988) Histamine H3 receptor-mediated inhibition of serotonin release in the rat brain cortex.Naunyn Schmiedebergs Arch Pharmacol 337:588

Schlicker E, Fink K, Detzner M, Gothert M (1993) Histamine inhibitsdopamine release in the mouse striatum via presynaptic H3receptors. J Neural Transm Gen Sect 93:1

Schlicker E, Kathmann M, Detzner M, Exner HJ, Gothert M (1994)H3 receptor-mediated inhibition of noradrenaline release: aninvestigation into the involvement of Ca2+ and K+ ions, Gprotein and adenylate cyclase. Naunyn Schmiedebergs ArchPharmacol 350:34

Schwartz JC (1975) Histamine as a transmitter in brain. Life Sci17:503

Smith CP, Hunter AJ, Bennett GW (1994) Effects of (R)-alpha-methylhistamine and scopolamine on spatial learning in the ratassessed using a water maze. Psychopharmacology (Berl)114:651

Steckler T, Dautzenberg FM (2006) Corticotropin-releasing factorreceptor antagonists in affective disorders and drug dependencean update. CNS Neurol Disord Drug Targets 5:147

Swerdlow NR, Geyer MA, Vale WW, Koob GF (1986) Corticotropin-releasing factor potentiates acoustic startle in rats: blockade bychlordiazepoxide. Psychopharmacology (Berl) 88:147

Taylor KM, Snyder SH (1971) Brain histamine: rapid apparentturnover altered by restraint and cold stress. Science 172:1037

Vogel JR, Beer B, Clody DE (1971) A simple and reliable conflictprocedure for testing anti-anxiety agents. Psychopharmacologia 21:1

Wesolowska A, Nikiforuk A, Stachowicz K, Tatarczynska E (2006)Effect of the selective 5-HT7 receptor antagonist SB 269970 inanimal models of anxiety and depression. Neuropharmacology51:578

Winslow JT, Insel TR (1990) Serotonergic and catecholaminergicreuptake inhibitors have opposite effects on the ultrasonicisolation calls of rat pups. Neuropsychopharmacology 3:51

Witkin JM, Nelson DL (2004) Selective histamine H3 receptorantagonists for treatment of cognitive deficiencies and otherdisorders of the central nervous system. Pharmacol Ther 103:1

Yanai K, Son LZ, EndouM, Sakurai E, Nakagawasai O, Tadano T, KisaraK, Inoue I, Watanabe T, Watanabe T (1998) Behavioural charac-terization and amounts of brain monoamines and their metabolitesin mice lacking histamine H1 receptors. Neuroscience 87:479

Yoshitomi I, Itoh Y, Oishi R, Saeki K (1986) Brain histamine turnoverenhanced by footshock. Brain Res 362:195

Psychopharmacology (2009) 205:177187 187

Anxiolytic-like...AbstractAbstractAbstractAbstractAbstractIntroductionMaterials and methodsAnimalsDrugsApparatus and procedureElevated plus maze testVogel type conflict testIsolation-induced vocalization testResidentintruder testConditioned fear stress test

Locomotion measurementsStatistics

ResultsDiscussionReferences

/ColorImageDict > /JPEG2000ColorACSImageDict > /JPEG2000ColorImageDict > /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 150 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict > /GrayImageDict > /JPEG2000GrayACSImageDict > /JPEG2000GrayImageDict > /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 600 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict > /AllowPSXObjects false /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (None) /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org?) /PDFXTrapped /False

/SyntheticBoldness 1.000000 /Description >>> setdistillerparams> setpagedevice