INTRODUCTION

Increasing evidence supports a role for histamine as a neurotransmitter and neuromodulator in various brain functions, including cognition, emotion, and feeding (Timmerman, 1990; Onodera et al, 1994; Leurs et al, 1998). Histamine is also released by activation of heteroreceptors (Hill and Straw, 1988; Gulat-Marnay et al, 1989a, 1990; Prast et al, 1991; Ono et al, 1992; Chikai et al, 1994; Laitinen et al, 1995). Histaminergic neurons are concentrated in the tuberomammillary nucleus of the posterior basal hypothalamus (Panula et al, 1984; Watanabe et al, 1984) and project to various brain regions (Inagaki et al, 1988; Panula et al, 1989).

H3 receptors were first characterized as autoreceptors that modulate histamine synthesis and release through negative feedback (Arrang et al, 1983, 1987; Van der Werf et al, 1987; Jansen et al, 1998). They are highly expressed throughout the brain (Arrang et al, 1998; Lovenberg et al, 2000; Drutel et al, 2001) on presynaptic terminals of histaminergic and nonhistaminergic neurons and also modulate the release of other neurotransmitters (Schlicker et al, 1988, 1989, 1993; Fink et al, 1990; Clapham and Kilpatrick, 1992). As H3 receptors display high constitutive activity, previously known antagonists were reclassified as inverse agonists (Morisset et al, 2000; Wieland et al, 2001).

Apolipoprotein E (apoE), which is important in lipoprotein and cholesterol metabolism (Mahley, 1988), has been implicated in nerve development and regeneration, neurite outgrowth, and neuroprotection (Weisgraber and Mahley, 1996). Mice deficient in apoE (Apoe–/–) (Piedrahita et al, 1992; Plump et al, 1992) have been used to define the role of apoE in brain function. Cardiac and serum histamine levels are higher in Apoe–/– than wild-type mice and Apoe–/– mice have 37% more cardiac mast cells (Huang et al, 2002). Mast cells migrate from blood to brain (Silverman et al, 2000) and brains of Apoe–/– mice might also be exposed to higher histamine levels than wild-type mice. In the present study, we analyzed Apoe–/– mice to investigate the possible role of apoE in the regulation of the H3-receptor-mediated signaling.

In the present study, we demonstrate that H3 antagonists increase anxiety in wild-type, but not Apoe–/–, mice. The differential effects of H3 antagonists on measures of anxiety were not seen on nonspatial and emotional learning and memory. As Apoe–/– mice showed higher sensitivity to the anxiety-reducing effects of the H1 receptor antagonist mepyramine than wild-type mice, reduced negative feedback via H3 receptors could contribute to increased signaling of H1 receptors in Apoe–/– mice.

Experimental Procedures

Animals

Male Apoe–/– (C57BL/6J-Apoetm1Unc) and wild-type C57BL/6J mice were obtained from The Jackson Laboratory (Bar Harbor, ME). Young (3–5 months old) Apoe–/– and wild-type mice were used in all the experiments, as indicated (n=5–19 mice per group). Young Apoe–/– mice were studied to evaluate a potential role of apoE on H3-receptor-mediated signaling, as adult Apoe–/– mice show age-dependent structural and functional alterations in the cortex and hippocampus (Masliah et al, 1995; Raber et al, 1998; Buttini et al, 1999). Such alterations could cause secondary changes in H3-receptor-mediated signaling. Mice were housed under conditions of constant temperature (18°C), light from 06:00 to 13:00, and free access to food and water. To minimize the effects of social influences on behavior, mice were housed individually 24 h before behavioral testing. Otherwise, they were group-housed. To avoid circadian variation, the mice were tested and killed between 10:00 and 14:00. All animal experiments were carried out in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals.

Drugs

The H3 ligands thioperamide and clobenpropit, the H1 ligand mepyramine maleate (pyrilamine maleate), and the H2 ligand zolantidine were gifts from Dr H Timmerman and Dr R Leurs (Department of Pharmacochemistry, Free University of Amsterdam, The Netherlands). Drugs were dissolved in saline and administered by intraperitoneal injection 1 h before behavioral testing at a dose of 5 mg/kg (thioperamide), 5.6 mg/kg (mepyramine), or 10 mg/kg (clobenpropit and zolantidine). The person testing the mice was blinded to genotype and treatment. The doses were selected based on reported studies.

Elevated plus maze

Anxiety levels were assessed with an elevated, plus-shaped maze consisting of two open arms and two closed arms equipped with rows of infrared photocells interfaced with a computer (Hamilton-Kinder, Poway, CA), as described (Raber et al, 2000). Mice were placed individually in the center of the maze and allowed free access for 10 min. They could spend their time either in a closed safe area (closed arms) or in an open area (open arms). Recorded beam breaks were used to calculate the time spent in the open arms, the distance moved in the open arms, entries into the open arms, and the number of times the mice extended over the edges of the open arms. Reductions in these measures indicate increased anxiety. After behavioral testing, the equipment was cleaned with 5% acetic acid to remove odors.

Novel object recognition

Novel object recognition was used to evaluate nonspatial learning and memory, as described (Raber et al, 2002) according to Rampon et al (2000). On each of 3 consecutive days, the mice were habituated to a Plexiglas enclosure (16 × 16 inches, open from the top) for 5 min. On day 4, they were allowed to explore the enclosure containing two objects, placed in equivalent positions (1 cm from the edge), for 15 min. On day 5, the mice were put in the enclosure for 15 min. One of the objects was replaced by a replica and the other one with a novel object. The amount of time the animals spent exploring the objects on days 4 and 5 was recorded with two stopwatches. Exploring was defined as close investigation within 1 cm of the objects, including sniffing of the objects. Between trials, the enclosure was cleaned with 5% acetic acid.

Dissection of brain regions

The mice were killed by cervical dislocation, and their brains were rapidly removed. A sagittal cut was made along the midline, and the whole hippocampus and cortex were dissected. Hypothalamic and amygdaloid regions were dissected out as described (Raber et al, 1994). The hypothalamus was defined as the region between two vertical cuts starting from the two lateral hypothalamic sulci and a horizontal cut through the mammillothalamic tracts and the rhomboid thalamic nucleus. The dissection of the amygdala was defined as the region between a vertical cut tangential to the external capsule and a diagonal cut along the medial border of the ipsilateral optic tract. The dissected brain regions were frozen on dry ice and stored at −80°C until use.

Histamine receptor binding

Membranes from the brain regions described above were prepared by homogenization in 10 volumes (wt/vol) of 50 mM sodium-PBS, pH 7.5, at 4°C containing EDTA (10 mM), 0.1 mMphenylmethylsulfonyl fluoride, 0.004 mg/ml chymostatin, 0.004 mg/ml leupeptin, and centrifugation at 40 000g for 30 min at 4°C. The pellet was resuspended in 5 ml of water and lysed for 30 min on ice. The centrifugation and lysing steps were repeated, followed by centrifugation at 40 000g for 30 min at 4°C. The final membrane pellets were resuspended in water and stored frozen at −80°C until use. H3 receptor binding was determined with [3H]Nα-methylhistamine ([3H]NAMH) as described (Tedford et al, 1995). Before use, the pellets were dissolved in distilled water and homogenized for 2 s by sonication. The homogenates (100 μg of protein) were incubated for 40 min at 25°C with increasing concentrations of [3H]NAMH (82.0 Ci/mmol) in 50 mM sodium phosphate buffer, pH 7.4, in the presence or absence of 10 μM thioperamide. The reaction was terminated by rapid dilution with 3 ml of ice cold 50 mM Tris, pH 7.4, and filtration through Whatmann GF/B filters that had been pretreated with polyethylenimine (0.3%), followed by two subsequent washes with 3 ml of 50 mM Tris, pH 7.4. Retained radioactivity was determined by liquid scintillation counting. Nonspecific binding was defined with 10 μM thioperamide as competing ligand. H1 receptor binding was determined with [3H]mepyramine (20 Ci/mmol), binding was performed similar to H3 using conditions as described (Yanai et al, 1998). Briefly, the homogenates were incubated for 30 min at 25°C in a 50 mM Na+/K+-phosphate buffer, pH 7.4, in the presence or absence of 1 μM mianserin. The reaction was terminated by rapid dilution with 3 ml of ice cold 50 mM Na+/K+-phosphate buffer, pH 7.4. Protein concentrations were determined spectrophotometrically with the Bradford reagent (Bradford, 1976), with bovine serum albumin as a standard.

Plasma corticosterone and ACTH

Plasma corticosterone and ACTH were determined using commercial radioimmunoassays from ICN. The intra- and interassay variation were both 7%. Plasma ACTH was determined using commercial radioimmunoassays from Phoenix Pharmaceuticals. The intra- and interassay variation were 9 and 10%, respectively.

Statistical analysis

Data are expressed as mean±SEM. Statistical analyses were carried out with Prism 3.0. The statistical differences between groups were determined by ANOVA, followed by a Tukey–Kramer post hoc test when appropriate; p<0.05 was considered significant.

RESULTS

Effects of the H3 Antagonist Thioperamide on Measures of Anxiety in Wild-Type and Apoe–/– Mice

Anxiety levels in young wild-type and Apoe–/– mice (3–5 months old) were assessed in the elevated plus maze 1 h after intraperitoneal administration of thioperamide or saline. Wild-type mice treated with thioperamide showed increased measures of anxiety as compared to wild-type mice treated with saline (Figure 1a–d). The total activity in the closed arms was comparable and not significantly different between the saline- and thioperamide-treated wild-type mice, indicating that the differences in measures in the open arms were not caused by differences in activity levels. In contrast, thioperamide had no effect on measures of anxiety in Apoe–/– mice (Figure 1a–d).

Figure 1
figure 1

Measures of anxiety levels in thioperamide-treated male wild-type and Apoe–/– mice in the elevated plus maze. Compared with saline controls, wild-type mice that received thioperamide showed significant reductions in time spent in the open arms (a), in distance moved in the open arms (b), in the number of extensions over the edges of the open arms (c), and in the number of entries into the open arms (d), indicating increased measures of anxiety after treatment with thioperamide. Apoe–/– mice showed no significant change in any of these measures of anxiety. None of the differences between wild-type and Apoe–/– mice were significant. ***p<0.001, **p<0.01, *p<0.05 vs saline controls; n=5–15 mice per group. Wild-type mice saline, n=15; thioperamide, n=15; Apoe–/– mice: saline, n=7; thioperamide, n=8.

Full size image

Effects of H3 Ligands on Novel Object Recognition in Wild-Type and Apoe–/– Mice

Next we determined whether in wild-type and Apoe–/– mice H3 antagonists also have differential effects on novel object recognition (Rampon et al, 2000), as described previously (Raber et al, 2002). During the training session, mice were allowed to explore for 15 min an open field containing two objects. For the retention session (24 h later), they were placed back into the same open field for 15 min, after one of the familiar objects had been replaced with a novel object and the other familiar object with an exact replica. The percentage of time the mice spent exploring the novel vs the familiar object relative to the total amount of time they explored either object in the retention session was used as a measure of object recognition memory. Wild-type and Apoe–/– mice (3–5 months old) received saline, thioperamide, or clobenpropit during the training (day 4) and retention (day 5) sessions. The recently cloned H4 receptor (Oda et al, 2000; Liu et al, 2001; Morse et al, 2001; Nguyen et al, 2001; Zhu et al, 2001) was found to have an affinity for H3-specific ligands. To rule out the possible contribution of the H4 receptor to the effects of thioperamide, we also treated wild-type and Apoe–/– mice with clobenpropit, a H3-specific antagonist that was reported to be an H4 receptor agonist as well (Oda et al, 2000). In the training session, all groups of mice spent a comparable amount of time exploring each object. In the retention session, the saline-treated wild-type and Apoe–/– mice spent significantly more time exploring the novel object (wild type: 8.37±0.93 s; Apoe–/–: 9.34±2.76 s) than the familiar object (wild type: 5.43±0.69 s; Apoe–/–: 5.15±1.25 s) (Figure 2a, b), whereas the thioperamide- and clobenpropit-treated wild-type and Apoe–/– mice spent equal amounts of time exploring both objects. The similar effects of thioperamide and clobenpropit on novel object recognition suggest that these effects are mediated by the H3 receptor and not the H4 receptor. Thus, the H3 antagonists similarly impaired object recognition in wild-type and Apoe–/– mice (Figure 2a, b).

Figure 2
figure 2

Novel object recognition in thioperamide- and clobenpropit-treated wild-type and Apoe–/– male mice. The percentages of time spent exploring the novel and familiar objects on day 5 of testing are shown. Only saline-treated mice spent significantly more time exploring the novel object. ***p<0.001 novel vs familiar object; n=5–19 mice per group. Wild-type mice saline, n=19; thioperamide, n=15; clobenpropit, n=6; Apoe–/– mice: saline, n=10; thioperamide, n=5; clobenpropit, n=6.

Full size image

H3 Expression Levels in Wild-Type and Apoe–/– Mice

To determine whether there are differences in H3 receptor expression in young Apoe–/– and wild-type mice (3–5 months old), which could have contributed to their differential response to H3 antagonists on measures of anxiety, we performed saturation analysis with [3H]NAMH in brain regions that have been implicated in cognition or emotion. The total number of receptors (Bmax in fmol/mg protein) in the amygdala (wild type: 87.3±2.5; Apoe–/–: 81.8±2.3), cortex (wild type: 119.9±3.0; Apoe–/–: 56.8±5.8), and hippocampus (wild type: 108.4±10.5; Apoe–/–: 29.1±1.7) was significantly lower in Apoe–/– than in wild-type mice (Figure 3). In the hypothalamus, Bmax was not significantly different between the groups. No significant difference was found in the binding affinities (Kd) of [3H]NAMH in any of the brain regions. Thus, compared to wild-type controls, Apoe–/– mice have a region-specific downregulation of the H3 receptor, without a change in affinity for H3-specific ligands in brain regions implicated in cognition and emotion.

Figure 3
figure 3

Saturation curve of [3H]NAMH in amygdala (a), hypothalamus (b), cortex (c), and hippocampus (d), of male wild-type (solid line, •) and Apoe–/– (dashed line, □) mice. Nonspecific binding was determined in the presence of 10 μM thioperamide. Significant differences in total number of receptors were seen in amygdala (p<0.05), cortex (p<0.001), and hippocampus (p<0.001). n=7 pooled mice per brain region.

Full size image

Increased Sensitivity of Apoe–/– Mice to the Effects of the H1 Antagonist Mepyramine on Measures of Anxiety

In Apoe–/– mice, reduced negative feedback via H3 receptors could increase signaling of H1 and H2 receptors. To determine whether in Apoe–/– mice there is increased H1- or H2-receptor-mediated signaling, wild-type and Apoe–/– mice were assessed in the elevated plus maze 1 h after intraperitoneal administration of the H1 antagonist mepyramine, the H2 antagonist zolantidine, or saline. Apoe–/– mice treated with mepyramine, but not with zolantidine, showed reduced measures of anxiety as compared to Apoe–/– mice treated with saline (Figure 4a–d). The total activity in the closed arms was comparable and not significantly different between the saline-, mepyramine-, and zolantidine-treated Apoe–/– mice, indicating that the differences in measures in the open arms were not caused by differences in activity levels. In contrast, mepyramine and zolantidine had no effect on measures of anxiety in wild-type mice (Figure 4a–d). Thus, compared to wild-type controls, Apoe–/– mice show an increased sensitivity to the effects of H1 receptor blockade on measures of anxiety. To determine whether in Apoe–/– the effects of mepyramine on measures of anxiety in the plus maze were associated with an altered HPA axis response (Knigge et al, 1999), we also measured plasma ACTH and corticosterone levels directly after plus maze testing, as described previously (Raber et al, 2000). Compared to saline controls, mepyramine reduced the plasma corticosterone levels in wild-type (saline: 179±38 pg/ml, n=6; mepyramine: 89±26 pg/ml, n=6; p<0.05 Tukey–Kramer), but not in Apoe–/– (saline: 206±30 pg/ml, n=8; mepyramine: 224±10 pg/ml, n=9), mice. Mepyramine also reduced plasma ACTH levels in wild-type (saline: 121±20 pg/ml, n=6; mepyramine: 77±9 pg/ml, n=6; p<0.05 Tukey–Kramer), but not in Apoe–/–, mice (saline: 57±5 pg/ml, n=8; mepyramine: 62±11 pg/ml, n=9). Thus, in the plus maze, the effects of mepyramine on measures of anxiety are dissociated from those on the HPA axis.

Figure 4
figure 4

Measures of anxiety in mepyramine- and zolantidine-treated male wild-type and Apoe–/– mice in the elevated plus maze. Compared with saline controls, Apoe–/– mice that received mepyramine showed significant increases in time spent in the open arms (a), in distance moved in the open arms (b), in the number of extensions over the edges of the open arms (c), and in the rest time in the open arms (d), indicating decreased measures of anxiety after treatment with mepyramine. Wild-type mice showed no significant change in any of these measures of anxiety. None of the differences between wild-type and Apoe–/– mice were significant. **p<0.01, *p<0.05 vs saline controls. Wild-type mice: saline, n=9; mepyramine, n=8; zolantidine, n=8; Apoe–/– mice: saline, n=7; mepyramine, n=7; zolantidine, n =5.

Full size image

H1 Expression Levels in Wild-Type and Apoe–/– Mice

To determine whether there are differences in H1 receptor expression in young Apoe–/– and wild-type mice (3–5 months old), which could have contributed to their differential response to mepyramine on measures of anxiety, we performed saturation analysis with [3H]mepyramine in brain regions that have been implicated in cognition or emotion. There was no difference in the total number of H1 receptors or H1 receptor binding affinity in the amygdala, cortex, hippocampus, or hypothalamus (Figure 5 and Table 1).

Figure 5
figure 5

Saturation curve of [3H]mepyramine in the amygdala (a), hypothalamus (b), cortex (c), and hippocampus (d), of male wild-type (solid line, •) and Apoe–/– (dashed line, □) mice. Nonspecific binding was determined in the presence of 10 μM thioperamide. No Significant differences in total number of receptors or receptor affinity were seen in any brain region. n=7 pooled mice per brain region.

Full size image
Table 1 H1 Expression Levels in Wild-Type and Apoe–/– Micea
Full size table

DISCUSSION

This study shows that H3 antagonists increased measures of anxiety in wild-type, but not in Apoe–/–, mice. In contrast, Apoe–/– mice showed a higher sensitivity to the anxiety-reducing effects of H1 receptor blockade than wild-type mice. The effects of H1 receptor blockade on measures of anxiety were dissociated from those on the HPA axis response. These differential effects of H3 antagonists on measures of anxiety were not seen on object recognition.

H1 receptor blockade did not reduce measures of anxiety in wild-type C57Bl/6J mice. This is consistent with the lack of effect of H1 receptor blockade on measures of anxiety in wild-type ddY mice. The reduced measures of anxiety in Apoe–/– mice after H1 receptor blockade might be caused by increased levels of histamine release. H1 receptors become activated at levels of histamine release higher than normal (Yuzurihara et al, 2000), mepyramine has antagonizing effects of on experimental anxiety induced by histamine releasers (Imaizumi and Onodera, 1993; Yuzurihara et al, 2000), and the H1 receptor agonist and H3 receptor antagonist betahistine has anxiogenic effects (Imaizumi et al, 1996).

The effects of H1 receptor blockade on measures of anxiety were dissociated from those on the HPA axis response. While mepyramine reduced measures of anxiety in Apoe–/–, but not wild-type, mice, it reduced plasma ACTH and corticosterone levels in wild-type, but not Apoe–/–, mice. These data suggest that in Apoe–/– mice mepyramine does not reduce measures of anxiety by inhibiting the HPA axis response. The dissociation of the effects of H1 receptor blockade on anxiety from those on the HPA axis in Apoe–/– and wild-type mice and the differential effects of H3 receptor blockade on novel object recognition and anxiety in Apoe–/–, but not wild-type, mice suggest that differential pharmacokinetic profiles of histaminergic drugs in the two genotypes do not underlie the behavioral results.

Thioperamide, at 5 mg/kg, increased measures of anxiety in wild-type mice. In a previous study in which the elevated plus maze was used to assess the anxiety of rats after the administration of 2 mg/kg thioperamide, the time spent in the open arms was reduced, but not significantly (Perez-Garcia et al, 1999). These data indicate that thioperamide doses higher than 2 mg/kg are required to obtain significant increased measures of anxiety.

In rodents, H3 antagonists improved performance in the modified elevated plus maze test (Miyazaki et al, 1995a; Onodera et al, 1998; Perez-Garcia et al, 1999). However, in the elevated plus maze H3 antagonists increase measures of anxiety, which could have contributed to the altered performance in the modified elevated plus maze tests.

While not statistically significant, several measures of anxiety (Figures 1 and 4) showed subtle genotype differences. It is unlikely though that the differential sensitivity of wild-type and mice to thioperamide or mepyramine are due to subtle differences in baseline behaviors. In Figure 1d, in which there were no baseline differences in wild-type and Apoe–/– mice, thioperamide decreased extending over the edges in wild-type but not Apoe–/– mice. Similarly, in Figure 4b, in which there were no baseline differences in wild-type and Apoe–/– mice, mepyramine increased the distance moved in the open arms in Apoe–/– but not in wild-type mice.

Apoe–/– mice have lower levels of H3 expression, without a change in affinity for H3-specific ligands, in the amygdala, hippocampus, and cortex than age-matched wild-type controls. In contrast, Apoe–/– and wild-type mice have similar levels of H1 expression and a similar affinity for H1-specific ligands in the amygdala, hypothalamus, hippocampus, and cortex. These data indicate that there is no simple association between levels of H1 and H3 receptor expression in structures associated with anxiety vs cognition which could explain why H3 antagonists impaired hippocampus- and cortex-dependent novel object recognition (Rampon et al, 2000) but did not increase more amygdala-dependent measures of anxiety in the plus maze or why Apoe–/– mice were more sensitive to the anxiety-reducing effects of mepyramine. Interestingly, in young Apoe–/– mice the enhancement of the α-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) receptors by phosphatidylserine (PS) was abolished without genotype changes in AMPA receptor binding (Valastro et al, 2001). These effects on the function of specific receptors may be related to the role of apoE in cholesterol transport. In Apoe–/– mice, the distribution of cholesterol in the synaptic plasma membrane is altered (Igbavboa et al, 1997), which can change the function of specific membrane receptors (Sooksawate and Simmonds, 2001). For example, altered membrane cholesterol in Apoe–/– mice reduces the potency of GABA at the GABAA receptor (Sooksawate and Simmonds, 2001). Similarly, by altering the lipid environment, apoE deficiency may reduce the total number of H3 receptor binding sites available for ligand binding in the amygdala, hippocampus, and cortex.

Recently, the H4 receptor was cloned and characterized (Nakamura et al, 2000; Oda et al, 2000; Liu et al, 2001; Morse et al, 2001; Nguyen et al, 2001; Zhu et al, 2001). To determine whether H4 signaling was involved in the observed effects on object recognition, we tested mice treated with the H4 receptor agonist and H3 receptor antagonist clobenpropit (Oda et al, 2000). As clobenpropit, like thioperamide, impaired object recognition test, it is likely that the effects of histamine signaling do not involve a direct histamine H4 receptor activation.

Our data demonstrate that H3 receptor blockade impairs novel object recognition in wild-type and Apoe–/– mice. Consistent with our data, H3 receptor stimulation by Rα-methylhistamine improved memory retention of naive rats in the water maze (Rubio et al, 2002) and H3 receptor blockade impaired social memory (Prast et al, 1996). Administration of thioperamide, histamine, or histidine, decreased investigation time of a juvenile rat by an adult rat while immepip or inhibition of histamine synthesis by α-fluoromethylhistidine prolonged recognition time (Prast et al, 1996). However, H3 receptor blockade might be beneficial under condition of cognitive impairments caused by cholinergic dysfunction (Miyazaki et al, 1995a, 1995b, 1997; Blandina et al, 1996, 1998; Gulat-Marnay et al, 1989a, 1989b; Giovannini et al, 1999; Molinengo et al, 1999). In a two-arm maze using a metal box and a glass bottle as objects in the retention trial and two identical objects in the first trial, H3 receptor blockade antagonized scopolamine-induced object recognition impairments and object recognition impairments at 90- and 120-min intertrial intervals at which male and female rats, respectively, did not exhibit object recognition anymore in this version of the test (Ghi et al, 1999).

Several studies have implicated the histaminergic system in cognitive tasks (Passani et al, 2000). Based on the effects of H3 receptor blockade on scopolamine-induced amnesia and on retention of animals trained to avoid a footshock in repeated acquisition avoidance models of attention deficit hyperactivity disorder (ADHD) and other disorders in which vigilance and impulsivity are impaired (Fox et al, 2002), H3 receptor antagonists were suggested as treatment for cognitive disorders, including Alzheimer's disease (AD) (Airaksinen et al, 1991; Nakamura et al, 1993). However, reduced brain histamine levels in AD (Mazurkiewicz-Kwilecki and Nsonwah, 1989; Panula et al, 1998) have not been consistently demonstrated (Cacabelos et al, 1989). The data reviewed above suggest that H3 receptor blockade might only be beneficial under condition of cognitive impairments caused by cholinergic dysfunction, which could relate to the ability of thioperamide to increase the release of acetylcholine (Clapham and Kilpatrick, 1992). However, H3 receptor agonists improved cognition and reduced muscarinic antagonist-induced cognitive deficits in the water maze (Smith et al, 1994). As AD patients already show cognitive deficits and increased anxiety (Ferretti et al, 2001), the use of H3 antagonists may be counterproductive and should be carefully evaluated.