Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Histamine (HA) exerts a neurotransmitter function in the peripheral and central nervous systems through its action at three HA receptor subtypes: H₁, H₂, and H₃ (reviewed by Arrang, 1994; Ruat et al., 1994). The HA H₃ receptor is highly localized in the central nervous system (CNS), and low levels are seen in peripheral tissues (Korte et al., 1990). Autoradiographic studies have described the detailed heterogeneous distribution of CNS HA H₃ receptors within brain regions (Pollard et al., 1993). HA H₃ receptors are present on histaminergic nerve terminals in the brains of rats and humans (Arrang et al., 1983, 1988). The HA H₃ autoreceptor modulates the amount of HA synthesized and released from histaminergic neurons (Arrang et al., 1983, 1987a). Furthermore, HA H₃ receptors are present as heteroreceptors in other neurotransmitter systems. The release of serotonin (Fink et al., 1990), dopamine (Schlicker et al., 1993), acetylcholine (Clapham and Kilpatrick, 1992), norepinephrine (Molderings et al., 1992), and Substance P (Taylor and Kilpatrick, 1992) is attenuated by HA H₃ receptor agonists. Conversely, HA H₃ receptor antagonists block these receptors, resulting in increased neurotransmitter release. Theoretically, the use of HA H₃ antagonists may provide for a clinically relevant enhancement of neurotransmitter release and thus provide effective therapies in the treatment of a number of CNS disorders (Timmerman, 1990; Leurs et al., 1998).

Selective pharmacological tools have been developed for each of the HA receptor subtypes, and selective HA H₃ antagonists such as GT-2016, thioperamide, and clobenpropit (VUF-9153) (Fig. 1) have been described (Arrang et al., 1987b; Van der Goot et al., 1992; Tedford et al., 1995; reviewed by Leurs et al., 1995; Stark et al., 1996). However, several structural features of HA H₃ receptor antagonists have yet to be fully exploited. For example, although a stereoselective bias in the binding of HA H₃ agonists has been demonstrated for several potent compounds (i.e., (R)--methylhistamine and (R)-, (S)--dimethyhistamine), such a relationship has yet to be demonstrated for HA H₃ antagonists. Furthermore, variations in heteroatom substitution have been previously investigated through the incorporation of a number of linker moieties, including amide-, guanidine- thiourea-, isothiourea-, carbamate- and ether-containing substitutions (Arrang et al., 1987b; Van der Goot et al., 1992; Schunack and Stark, 1994; Tedford et al., 1995; Brown et al., 1996; Schlicker et al., 1996). The absence of such heteroatom substitution through the incorporation of an unsaturated carbon spacer has not previously been considered in the development of HA H₃ receptor antagonists.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Structures of verongamine, thioperamide, clobenpropit, and GT-2016.

In addition to the compounds discussed above, the natural product verongamine (Fig. 1), isolated from the marine sponge Verongula gigantea, was determined to have moderate HA H₃ receptor affinity with an IC₅₀ of 0.5 µM for guinea pig CNS HA H₃ receptors (Mierzwa et al., 1994). Analysis of the structural features of verongamine from a conformational and stereochemical view point provided us with conceptual insight that led to the development of several new series of HA H₃ receptor ligands with high affinity and receptor selectivity. Structure-activity relationship (SAR) studies focused on several structural features illustrated by the derivative shown in Fig. 2. These features include the imidazole ring (A), ethylene bridge (B), amide bond or replacements (C), chirality and substitution of the amino constituent (D), and the terminal hydrophobic region (E). Evaluation of these specific elements has led to the development of new, potent, and selective HA H₃ receptor ligands. Herein, the in vitro and in vivo binding characteristics of a variety of newly developed compounds are described (see Tedford et al., 1999). Moreover, the structural requirements for ligand interaction with the HA H₃ receptor and initial pharmacological profiles of selective HA H₃ receptor ligands are discussed.

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Model template for HA H3 receptor ligand development.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Chemicals. GT compounds and thioperamide were synthesized by Gliatech chemists (Ali et al., 1998, 1999). Other reagents were purchased as indicated: triprolidine (Research Biochemicals International, Natick, MA), aminopotentidine (Tocris Cookson, Bristol, UK), [³H]N-methylhistamine ([³H]NAMHA; 81.5 Ci/mmol), [³H]pyrilamine (>20 Ci/mmol; DuPont NEN Research Products, Boston, MA), and [¹²⁵I]iodoaminopotentidine (2000 Ci/mmol; Amersham, Arlington Heights, IL or provided by Drs. Leurs and Timmerman at the Leiden/Amsterdam Center for Drug Research, the Netherlands). Clobenpropit was kindly provided by Dr. Timmerman.

Animals. Male Sprague-Dawley rats were purchased from Harlan Laboratories (Indianapolis, IN) and housed two per cage on a 12-h light/dark schedule with ad libitum access to Teklad Mouse/Rat Diet 7012 (Harlan Laboratories) and water in accordance with the Animal Welfare Act of 1994 and amendments. Animals were acclimated to laboratory conditions for a minimum of 1 week before tissue harvesting.

In Vitro HA H₃ Receptor-Binding Analysis. HA H₃ receptor affinity was determined in rat cerebral cortical membranes with [³H]NAMHA as described previously (Tedford et al., 1995). Animals were euthanized by rapid decapitation and cerebral cortical tissues were harvested and frozen on dry ice. Cerebral cortical membranes were prepared in 50 mM sodium-PBS (pH 7.5 at 4°C) containing EDTA (10 mM), phenylmethylsulfonyl fluoride (0.1 mM), chymostatin, and leupeptin (each 0.2 mg/50 ml). The final membrane pellets were resuspended in water and stored frozen at 80°C before use. Protein concentrations were determined using the Coomassie Plus Protein Assay (Pierce, Rockford, IL).

Ex Vivo HA H₃ Receptor-Binding Analysis. Ex vivo HA H₃ receptor occupancy was determined in rat cerebral cortical membranes with [³H]NAMHA as described previously (Taylor et al., 1992; Tedford et al., 1995). Rats were acutely treated with a single i.p. injection of compound (n = 4/group). Drugs were administered in a final volume of 1 ml/kg. The animals were euthanized by a lethal injection of sodium pentobarbital (150 mg/kg i.p., Nembutal; Abbott Laboratories, North Chicago, IL) at the indicated times postadministration. Following euthanasia, the upper torso was transcardially perfused through the aortic arch with 60 ml of 0.9% saline to remove potential vascular drug contamination. Brains were removed, dissected, and frozen on dry ice. The tissue was stored at 80°C before conducting the binding studies. On the day of binding experiments, the tissue was homogenized using a motor-driven tissue grinder (Omni 1000) in 9 volumes (w/v) of 50 mM sodium-phosphate buffer (pH 7.4). Ex vivo binding was carried out in a total volume of 0.4 ml of 50 mM sodium-phosphate buffer containing ~1 nM [³H]NAMHA and 0.15 to 1 mg of protein. Nonspecific binding was determined using 10 µM thioperamide. Samples were treated as described above for the in vitro binding. ED₅₀ values (doses that produced 50% inhibition of [³H]NAMHA binding) in milligrams per kilograms were determined by linear regression analysis of the data on a log-linear plot.

In Vitro Human HA H₁ and H₂ Receptor-Binding Analysis. Before binding, human HA H₁ or H₂ receptor-expressing cells were harvested, washed, and stored as a pellet at 80°C. Stable human HA H₁ receptor-expressing Chinese hamster ovary cells were provided by Drs. Leurs and Timmerman (Leiden/Amsterdam Center for Drug Research). The HA H₁ receptor was expressed at ~1 pmol/mg protein in these cells. In preparation for binding, cell cultures were expanded and homogenates were prepared by sonication in water. HA H₁ receptor affinity was determined using membranes prepared from HA H₁ receptor-transfected cells and [³H]pyrilamine (Tedford et al., 1995). Binding was performed in a total volume of 0.4 ml of 50 mM sodium/potassium-phosphate buffer containing 1 mM MgCl₂ (pH 7.4, 25°C). Nonspecific binding was determined in the presence of 10 µM triprolidine. For competition studies, 4 nM [³H]pyrilamine was incubated with ~8 µg of membrane protein for 30 min at 25°C in polypropylene tubes with increasing concentrations of test compounds. Binding was terminated and samples were quantitated as described for HA H₃ receptor binding.

Spontaneous Locomotor Activity. Spontaneous locomotor activity (SLA) was measured in adult Sprague-Dawley rats (200-300 g) using automated Omnitech digiscan animal activity monitors (42.5 × 42.5 × 40 cm; Omnitech, Columbus OH) between 9:00 AM and 2:00 PM (lights out). Data were collected for the following parameters: total, horizontal, central and peripheral activities, stereotypes, and vertical movements. Total counts, number of occurrences, and total time involved in each behavior were determined. Thirty-minute baseline readings (5-min collection intervals) were collected, after which the animals were administered drug or vehicle. Following treatment, an additional 90 min of activity monitoring was conducted.

Passive Avoidance Training and Drug Treatment Studies. Postnatal day 15 to 16 (P15-16) rat pups were used for the 10-trial passive avoidance response (PAR) testing. Animals were singly placed in a Coulbourn Instruments (Allentown, PA) small animal shuttle box. The shuttle box has a lighted compartment divided by a door leading to a dark compartment. The animals were placed facing away from the dark compartment and the time to turn and enter the dark compartment was measured (latency). Upon entering the dark compartment, the animal trips an automated door closure and, after a 5-s delay, a mild foot shock (0.5 mA, 60 Hz, 2 s) is delivered through a metal grid floor. Immediately after the shock, the pup is removed and returned to its home cage for 3 min. Animals were returned to the lighted compartment after 3 min and the step-through latency (maximum of 180 s) was recorded for a total of 10 trials.

Statistical Analysis. PAR statistical analysis was restricted to comparisons between the drug-treated animals and their vehicle-treated littermate controls for a particular dose of drug. Both the median and mean latencies (seconds) were determined for each trial. Statistically significant differences (p < .05) for a given trial were determined using a nonparametric Kruskal-Wallis test. Mean latencies ± S.E. are depicted for graphical representation.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

In Vitro HA H₃ Receptor-Binding and SAR Studies. The initial synthesis of several HA amides provided several key SAR for the development of a series of potent HA H₃ receptor ligands (Table 1). The amide resulting from the coupling of HA with (L)-phenylalanine afforded a compound, GT-2130, with good HA H₃ receptor affinity (K_i = 104 nM). In contrast, the (D)-phenylalanine analog exhibited a marked reduction in HA H₃ receptor affinity (K_i = 2,814 nM). This gave direct evidence for a stereoselective bias for the HA H₃ receptor and illustrated the importance of the chiral amino center for enhancing HA H₃ receptor affinity. Elongation of the carbon tail (GT-2148 versus GT-2130) and/or replacement of the terminal phenyl group with a cyclohexyl moiety (GT-2130 versus GT- 2140) provided additional optimization of HA H₃ affinity within the amide series. Substitution on the NH₂ group (GT-2142) resulted in significant loss in HA H₃ receptor affinity, suggesting the importance of the primary chiral amino group. In addition, analogs of compounds without the primary amino group (e.g., GT-2174 versus GT-2140) also demonstrated potent HA H₃ receptor activity, provided that optimization of the carbon chain length and hydrophobic tail were made. Moreover, incorporation of the chiral NH₂ group into various conformationally restricted heterocyclic ring systems (data not shown) indicated that an octahydroindole moiety was an acceptable substitution.

Receptor Selectivity Profile. The HA receptor selectivity profile was evaluated for one of the more potent olefin-containing analogs, GT-2227. GT-2227 had minimal affinity for the human HA H₁ and H₂ receptor subtypes (Table 3). The binding of GT-2227 to human HA H₃ receptors was evaluated using membranes prepared from human forebrain autopsy tissue. These affinities provided receptor selectivity ratios of 4190 and 1397 for the human HA H₃ receptor with respect to the human HA H₁ and H₂ receptors, respectively. A general receptor screening profile for GT-2227 was also performed for a variety of receptors and ion channels. Overall, 1 µM GT-2227 produced minimal inhibition of ligand binding in these assays (Table 4). This concentration of GT-2227 did, however, produce >50% inhibition of ligand binding to ₂ and I₂ receptors in preliminary binding studies (Table 4). More detailed analysis of binding to these receptors provided IC₅₀ values of 1274 and 1342 nM for ₂ and I₂ receptors, respectively.

Ex Vivo HA H₃ Receptor-Binding Studies. The CNS penetration for several of the more potent HA H₃ ligands was evaluated using the ex vivo binding technique. Rats were dosed with 0.3 to 30 mg/kg (i.p.) compound and, after 1 h, the rats were euthanized and the penetration of compound into the CNS was evaluated. Following 0.3, 1, and 3 mg/kg (i.p.), three of the most potent GT compounds, GT-2227, GT-2260, and GT-2286, produced dose-dependent inhibition of [³H]NAMHA binding in cerebral cortical homogenates prepared from drug- versus vehicle-treated animals (Fig. 3A). Each of the three compounds produced very similar inhibition profiles, with 25 to 35% inhibition of [³H]NAMHA binding at 0.3 mg/kg and near maximal receptor occupancy at the 3-mg/kg doses 1 h postadministration. Log-linear regression analysis of these data provided similar ED₅₀ values of 0.7 ± 0.05, 0.43 ± 0.12, and 0.48 ± 0.14 mg/kg (mean ± S.E.) for GT-2227, GT-2260, and GT-2286, respectively. These compounds represent some of the most potent, CNS available HA H₃ ligands developed to date. The ex vivo binding profile for GT-2227 in postnatal day 16 rat pups was also evaluated. Pups were dosed with 0.3, 1, and 3 mg/kg GT-2227 (i.p.) for 1 h. The ED₅₀ was 0.49 ± 0.04 mg/kg (mean ± S.E.) and is similar to what was observed in adult rats. Two known HA H₃ antagonists, thioperamide and clobenpropit, were also compared for CNS penetration. Thioperamide produced a dose-dependent inhibition of [³H]NAMHA binding at 1, 3, and 10 mg/kg (i.p.), as well as clobenpropit at 3, 10, and 30 mg/kg (i.p.) (Fig. 3). The ED₅₀ values for these compounds were 1.49 ± 0.64 and 3.88 ± 4.41 mg/kg (mean ± S.E.), respectively.

View larger version (37K): [in this window] [in a new window]   Fig. 3.   Ex vivo HA H3 receptor-binding profile for selected ligands in rat brain cortex. A, GT-2227, GT-2260, and GT-2286 (0.3, 1, and 3 mg/kg) and the respective vehicles were administered i.p. 1 h before the animals were euthanized. Ex vivo HA H3 receptor binding was determined for each animal (n = 3-4/group) in femtomoles per milligram of protein and expressed as a percentage of values from vehicle-treated animals. Absolute values for vehicle-dosed animals were 38.00 ± 2.07, 27.76 ± 0.88, and 31.10 ± 0.92 fmol/mg protein for GT-2227, GT-2260, and GT-2286, respectively (mean ± S. E., n = 4). B, thioperamide (1, 3, and 10 mg/kg), clobenpropit (3, 10, and 30 mg/kg), and GT-2231 (1, 3, 10, and 30 mg/kg) and the respective vehicles were administered i.p. 1 h before the animals were euthanized. Ex vivo HA H3 receptor binding was determined for each animal (n = 3-4/group) in fmol/mg protein and expressed as a percentage of values from vehicle-treated animals. Absolute values for vehicle-dosed animals were 30.75 ± 1.9, 29.47 ± 6.99, and 31.87 ± 2.34 fmol/mg protein for thioperamide, clobenpropit, and GT-2231, respectively (mean ± S. E., n = 4). ND, not determined.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Twenety-four-hour time course ex vivo binding profiles for GT-2227 and thioperamide in rat brain cortex. GT-2227 or thioperamide (10 mg/kg) and vehicle were administered i.p. Animals were euthanized at 1, 2, 4, 8, 16, or 24 h after administration of drug. Ex vivo HA H3 receptor-binding was determined for each animal (n = 4/group) in femtomoles per milligram of protein and expressed as percentage of vehicle-treated values. Absolute values for vehicle-dosed animals were 33.12 ± 3.36 and 37.65 ± 3.74 fmol/mg protein for GT-2227 and thioperamide, respectively (mean ± S. E., n = 4).

Effects of GT-2227 and Methylphenidate on Cognitive Deficits in Juvenile Rat Pups. Cognitive impairments have been described in developing rat pups that display rapid development of normal cognitive capabilities at approximately 3 weeks after birth (Duméry and Blozovski, 1987). This immature rat model can be used to assess whether compounds may enhance learning and memory, without the conflicting use of neurotoxins, lesions, or amnesic agents to induce a cognitive deficit. We tested juvenile rat pups at various postnatal times for cognitive function in a single-trial PAR. Initially, 48-h retention latencies (seconds) were determined in juvenile rat pups at P15 to 35 following a single-trial PAR. Deficits in the 48-h recall of a single-trial PAR test were seen in the P15 to 16 rat pups (data not shown).

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Effect of GT-2227 and methylphenidate on acquisition of a 10-trial PAR in rat pups. P15 to 16 rat pups were tested for acquisition of a multitrial PAR with a 3-min intertrial period. Littermates were equally divided into vehicle- or drug-treatment groups. GT-2227 (A) or methylphenidate (B) was administered i.p. at a dose of 1 or 3 mg/kg, respectively, 30 min before training. Animals were returned to their home cage with their littermates for the intertrial time period. Data are expressed as mean latency ± S. E. (n = 12-18/group). *Indicates statistically significant differences (p < .05) between drug- and vehicle-treatment groups at the specific trial number. Nonparametric statistical analysis (Kruskal-Wallis test) was conducted on median latencies (seconds).

Effects of GT-2227 and Methylphenidate on SLA in Adult Rats. HA H₃ receptor blockade may result in increased CNS HA release and has been shown to induce electroencephalogram activation or arousal (Lin et al., 1990; Monti et al., 1991). Previous studies have not demonstrated profound psychomotor stimulant properties for HA H₃ antagonists, compared to known psychomotor stimulants such as methylphenidate. GT-2227 (1, 3, and 10 mg/kg i.p.) was tested in adult rats for potential psychomotor stimulant effects on SLA. The doses tested were chosen to provide for 50 to 100% HA H₃ cerebral cortical receptor blockade and sustained duration. Normal habituation to the novel environment was evident before either vehicle or drug treatment in the adult rats (>30 days old). Thirty minutes after exposure to the activity chambers, animals (n = 12-18/group) were administered GT-2227 or methylphenidate and SLA was recorded for the next 90 min. GT-2227 had no effect on SLA at any of the tested doses in adult rats as measured by horizontal activity (Fig. 6A) as well as total, central, and peripheral cage activities, stereotypes, or vertical movements (data not shown). Methylphenidate (1, 3, and 10 mg/kg i.p.), however, produced a dose-dependent increase in SLA as measured by horizontal activity. This psychomotor stimulation induced by methylphenidate was dose-dependent (data not shown), and the effect of the 3-mg/kg dose was maintained for approximately 1 h (Fig. 6B).

View larger version (31K): [in this window] [in a new window]   Fig. 6.   Effect of GT-2227 and methylphenidate on SLA in adult rats. Adult rat (>30 days) SLA patterns were measured every 5 min for a total of 2 h. GT-2227 or methylphenidate (1, 3, or 10 mg/kg, i.p.) was administered after a 30-min baseline SLA was established. Group mean horizontal activity measurements (beam breaks) are presented for the 10-mg/kg dose of GT-2227 in A and for the 3-mg/kg dose of methylphenidate in B, as well as the respective vehicle-treated groups (mean ± S. E., n = 12-18/group).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present studies, a new series of 1H-4-substituted-imidazoyl HA H₃ receptor ligands were synthesized and tested for HA H₃ receptor affinity and selectivity. The development of this series of HA H₃ ligands was conceptually derived using the natural product verongamine (Mierzwa et al., 1994) as a synthetic template. Several structural activity relationships were demonstrated that provide new insights into the binding characteristics of the HA H₃ receptor. Importantly, the chiral amino functionality provided evidence that the HA H₃ receptor exhibits a stereoselective bias in antagonist ligand-binding interactions. This has been demonstrated previously with several HA H₃ agonists [i.e., (R)--methylhistamine and (R)-, (S)--dimethyhistamine], but had not been evaluated for HA H₃ antagonists. Additionally, although amide-, guanidine- thiourea-, isothiourea-, carbamate-, and ether-containing series of HA H₃ receptor antagonists have been described (Arrang et al., 1987b; Van der Goot et al., 1992; Schunack and Stark, 1994; Tedford et al., 1995; Brown et al., 1996; Schlicker et al., 1996), the incorporation of a three to four unsaturated carbon spacer (Fig. 2C) had not previously been disclosed in the development of potent HA H₃ receptor antagonists. Clearly, the present studies provide support for the use of acetylene and olefin moieties as spacer groups. Most surprising is that potent HA H₃ receptor affinity was maintained despite the removal of all heteroatoms which would transcend into the binding domain associated with the terminal amino group of HA.

Molecular modeling efforts further support our data and suggest that a nonpolar, planar spacer (olefin and acetylene functionalities) can effectively be used as equivalents to the polar and planar amide or amide-oxime spacer in verongamine (Ali et al., 1998). However, it is clear from these studies that the imidazole head, as well as the position, orientation, and chemical makeup of the middle spacer and terminal hydrophobic pieces must act harmoniously to provide a preferred conformation for interaction with the HA H₃ receptor. The orientation of these structural features is illustrated by the enhanced affinity of the cis-olefin versus the trans-olefin isomer, GT-2227 versus GT-2228.

The use of a template which systematically explored the various structural features required for HA H₃ receptor activity has led to the development of new compounds which maintain high HA H₃ receptor affinity and selectivity. Further optimization has produced a series of compounds that are orally active, can penetrate the blood-brain barrier very effectively, and provide a sustained duration of action. This was illustrated with GT-2227, GT-2260, and GT-2286, which show similar CNS penetration profiles in ex vivo binding studies. These penetration profiles are better than those of clobenpropit, GT-2231, and thioperamide. Moreover, GT-2227 had an excellent time course profile with near maximal receptor occupancy maintained out to 8 h. This profile was notably better than that of thioperamide. Additional studies by Tedford et al. (1998b) have also characterized the HA H₃ functional antagonist activity of GT-2227 using the isolated guinea pig jejunum and cardiac synaptosomes. There, the pA₂ for GT-2227 was determined to be 7.9 for the inhibition of neurogenic contractions of the guinea pig jejunum, which is in agreement with the K_i for GT-2227.

Several studies have indicated a role for HA in cognitive processes (de Almeida and Izquierdo, 1988; Kamei and Tasaka, 1991; Kamei et al., 1993; Prast et al., 1996). These observations, combined with the neuroanatomical localization of HA H₃ receptors in cerebral cortical regions (Cumming et al., 1991; Pollard et al., 1993), are supportive of a functional role for HA in higher learning. Recently, the HA H₃ antagonists thioperamide and GT-2016 were shown to enhance learning and recall in several cognitive models (Meguro et al., 1995; Miyazaki et al., 1995; Tedford, 1998a). In addition, the in vivo and in vitro release of acetylcholine was shown to be modulated by presynaptic HA H₃ heteroreceptors (Clapham and Kilpatrick, 1992; Blandina et al., 1996). Blandina et al. (1996) further showed that in addition to reduced in vivo cerebral cortical acetylcholine release, the HA H₃ agonists imetit and (R)--methylhistamine impaired cognitive performance in both object recognition and PAR learning models. Together, these biochemical and behavioral studies are supportive of a role for HA and HA H₃ receptor blockade in cognition enhancement. The present studies show that GT-2227 produced significant improvements in the performance (latency) of developmentally immature rat pups in the PAR at doses which provide for 50 to 100% occupancy of cerebral cortical HA H₃ receptors in vivo. These doses of GT-2227 had no effect on the overall locomotor activities of adult rats as measured by total, horizontal, central and peripheral activities, stereotypes, and vertical movements.

Although the use of selective HA H₁ and H₂ ligands has been effectively exploited in the treatment of allergy and excessive gastric acid secretion, respectively, the pharmacological benefits of selective HA H₃ compounds have yet to be fully realized. Potential CNS clinical uses have been identified for both HA H₃ receptor antagonists or agonists based on the reports of the high density of HA H₃ receptor levels and their distribution within the CNS combined with the role of the HA H₃ receptor at both auto- and heteroreceptor sites. A therapeutic role for the use of HA H₃ receptor antagonists in cognitive dysfunction, sleep disorders, obesity, and hormonal dysfunction has been suggested (Timmerman, 1990; Leurs et al., 1998, Tedford, 1998a). The limited availability of clinical agents, however, has hindered further exploration of the clinical usefulness of such agents. Furthermore, the biochemical and pharmacological characterization of the HA H₃ receptor has been limited as well due to the poor CNS availability of a variety of early HA H₃ agents. The present studies describe the development of a new series of potent and selective HA H₃ receptor ligands with excellent CNS penetration. Initial behavioral assessment of a prototype compound (GT-2227) indicates that improvements in learning are seen at doses that provide significant cerebral cortical HA H₃ receptor occupancy without locomotor stimulant properties. These studies lend support for the use of HA H₃ receptor antagonists in diseases associated with cognitive or attentional deficits. Importantly, these new CNS-penetrating ligands can be used to conduct behavioral assessments to further explore the full potential and clinical utility of blockade of the HA H₃ receptor.