Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The central noradrenergic system is considered to play important roles in attention, learning, and memory processes (Aston-Jones et al., 1991; Berridge et al., 1993; Sara et al., 1994; Ferry et al., 1999a,b). Noradrenaline has also been found to have an important permissive role in long-term potentiation, a form of synaptic plasticity associated with memory processes (Bliss et al., 1983), and also facilitates in a synergistic manner a similar role for acetylcholine in long-term potentiation (Brocher et al., 1992). Decline in memory is one of the major symptoms of Alzheimer's disease (AD), and this impairment has been ascribed to the central cholinergic (Bartus et al., 1986; Perry et al., 1992) and noradrenergic (Reinikainen et al., 1990; Haapalinna et al., 1998) pathologies that occur in this disorder. Neuronal cell loss in the locus coeruleus (LC) and a decrease in presynaptic noradrenergic markers are commonly observed in postmortem brains of patients with AD (Cowburn et al., 1988), Parkinson's disease, and related neuropathologies (Colpaert, 1994). This deficiency in the LC-noradrenergic (LC-NA) system has been hypothesized to be the critical factor in determining the progression of a family of neurodegenerative diseases that includes AD, because the LC-noradrenergic system is considered to be essential in controlling compensatory mechanisms that slow, prevent, or reverse the neurodegenerative process (Colpaert, 1994). Thus, the targeting of noradrenergic mechanisms represents a potential therapeutic strategy against progressive neurodegenerative diseases such as AD.

The activity of LC-noradrenergic neurons is regulated by presynaptic inhibitory ₂-adrenergic autoreceptors. By blocking these receptors, ₂-adrenoceptor antagonists disinhibit the LC system, leading to an increase in LC neuronal activity and a consequent increase in noradrenaline synthesis and release in target areas (Dennis et al., 1987). Concomitantly, ₂-antagonists facilitate by disinhibition the activity of other neurotransmitter systems downstream from the LC-NA system that are negatively influenced by ₂-adrenergic heteroreceptors, such as acetylcholine (Tellez et al., 1997), dopamine (Matsumoto et al., 1998), and serotonin (Raiteri et al., 1990), deficits in these transmitter systems being apparent from postmortem studies in AD (Francis et al., 1994). Thus, ₂-adrenoceptor antagonists facilitate in vivo the release of neurotransmitters that are involved in learning and memory and that are compromised in AD, including cortical acetylcholine and noradrenaline.

Dexefaroxan, the (+)-enantiomer of efaroxan, is a potent and selective antagonist of rodent and human ₂-adrenoceptors. In contrast to earlier ₂-adrenoceptor antagonists such as yohimbine or idazoxan, dexefaroxan acts as a specific and competitive antagonist with relatively little intrinsic agonist activity (Tellez et al., 1997; Chopin et al., 1999), and has no appreciable affinity for the imidazoline (I₁ and I₂) receptor binding sites in human brain (Vauquelin et al., 1999). Furthermore, dexefaroxan, which is devoid of cholinesterase inhibitory actions, produces a robust, dose-dependent, and sustained increase in endogenous acetylcholine outflow in the medial prefrontal cortex of the rat in vivo (Tellez et al., 1997), and has been found to have cholinergic neuroprotective effects in vivo (Martel et al., 1998; Debeir et al., 2001). These results support the potential therapeutic utility of dexefaroxan to provide both symptomatic and trophic benefits in neurodegenerative disorders that involve deficits in central cholinergic function (e.g., AD). To further assess this potential, the present studies were undertaken to investigate the capacity of dexefaroxan to modify memory processes in rodents. Effects in young and aged rats, and in rats with bilateral ibotenic acid induced-lesions of the nucleus basalis magnocellularis, were examined using the passive avoidance task and the Morris water test in rats. Effects in mice were investigated using an object recognition test. Tacrine, a potent, reversible, noncompetitive, and centrally acting inhibitor of acetylcholinesterase, which antagonizes the memory deficits produced by scopolamine in normal human volunteers and which has been used clinically for the symptomatic treatment of cognitive deficits in AD (Bartus, 2000), was evaluated as a comparative reference compound.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animals. Animals were housed, handled, and cared for in accordance with the Guide for the Care and Use of Laboratory Animals (National Research Council, 1996) and the European Directive No. 86/609. The experimental protocol was carried out in compliance with French regulations and with local ethical committee guidelines for animal research. Animals were housed with free access to food and water in a room maintained at 21 ± 1°C and 60 ± 5% humidity, under a 12-h light/dark cycle with lights on from 7:00 AM. Male NMRI mice (Ico: NMRI [IOPS] Han; IFFA Credo, Domaine des Oncins, France) weighing 30 to 32 g were housed in groups of 15. Male Sprague-Dawley rats (Ico: OFA SD [IOPS Caw]; IFFA Credo), weighing 200 to 250 g, were housed in groups of six. For experiments comparing drug effects in young and aged rodents, 2-month-old male Sprague-Dawley rats, weighing 200 to 250 g, and 24-month-old male Sprague-Dawley rats, weighing 500 to 700 g, were housed singly for 2 weeks before being tested. For the nucleus basalis magnocellularis (NBM)-lesioned animal experiments, male Sprague-Dawley rats, weighing 160 to 180 g at the start of the experiment, were housed singly for 2 weeks before being lesioned.

Bilateral Ibotenic Acid Lesions of NBM. This procedure has been described previously (Winkler and Thal, 1995). Rats were anesthetized with a halothane/oxygen mixture and placed in a David Kopf small animal stereotaxic apparatus (model 900). Body temperature was maintained at 37.0 ± 0.2°C with a thermostated heating pad. Bilateral infusions of 25 nmol/0.5 µl of ibotenic acid (lesioned animals) or saline (sham-operated animals) were made at a rate of 0.1 µl/min via a syringe pump (model CMA-100; Carnegie Medicine AB, Stockholm, Sweden) and a 30-gauge stainless steel cannula. After each infusion the cannula was left in place for 2 min to prevent reflux and to allow for toxin diffusion. Immediately after lesioning of the NBM on one side, the contralateral NBM was lesioned. Stereotaxic coordinates were as follows: 1.0 mm posterior from bregma, ±2.5 mm lateral from midline, and 8.5 mm ventral from the bone surface. Animals were allowed to recuperate from anesthesia in individual cages under an infrared heating lamp, and then housed individually with free access to food and water for the rest of the study.

Passive Avoidance. This procedure has been described previously (Chopin and Briley, 1992). The apparatus consisted of a larger white compartment (30 × 30 × 30 cm) connected by an opening (7 × 7 cm) to a smaller black compartment (20 × 20 × 12 cm) with a grid floor. A 60-W lamp was positioned centrally 60 cm above the base of the large compartment (750 lux). The room was dark during the experimental sessions, which were conducted between 9:00 AM and 3:00 PM. Initially, each rat was placed in the larger illuminated compartment and allowed 30 s to freely explore that side of the apparatus. The entrance to the dark compartment was then opened and, as soon as the rat had entered with all four paws on the grid floor, the door was closed and an inescapable scrambled footshock was delivered through the grid by a Coulbourn shock generator. Immediately afterward, the rat was returned to its home cage. Forty-eight hours, 1 week, 5 weeks, or 15 weeks later, the rat was again placed in the larger illuminated compartment and, after 30 s, the door was opened. The delay in entering the dark compartment was recorded to a maximum of 180 s.

Morris Water Maze. The water maze test was adapted from Morris (1981). The test apparatus consisted of a circular fiberglass tank (130 cm in diameter, 50 cm in depth). The pool was filled to a height of 30 cm with water at room temperature (21-22°C). The pool was divided into four virtual quadrants (Q₁, Q₂, Q₃, and Q₄) of equal surface area. A transparent escape platform made of Plexiglas (10 cm in diameter, 29 cm in height) was placed in a fixed location in the tank, 1 cm below the water surface. The platform was not visible from just above water level, and transfer trials have indicated that escape onto the platform was not achieved by visual or other proximal cues (Morris, 1981). Many extra-maze cues surrounded the maze and were available for the rats to use in locating the escape platform. On the training trials, the platform remained in a constant location in the center of one quadrant (Q₄) equidistant from the center and the edge of the pool. Each rat received three trials per day for 1, 2, 3, or 4 days. Each training trial involved placing the rat into the pool facing the wall at one of the three quadrants Q₁, Q₂, and Q₃. A different starting point was randomly used on each trial. The rats were allowed to swim freely until they found the escape platform. The latency to find the hidden platform was recorded and used as a measure of acquisition of the task. If a rat failed to locate the platform within 100 s it was then manually guided to the escape platform by the experimenter. The intertrial interval was 20 s during which the rat remained on the platform. Twenty hours after the last training trial, the platform was removed from the pool, the rats were allowed to swim for 60 s in the pool and the time spent in the target quadrant Q₄ (the quadrant in which the platform was during training) was recorded. The percentage of time spent in the previous training quadrant Q₄ was used as an index of memory. The higher the percentage, the better the memory was considered to be.

Object Recognition. The object recognition test was adapted from Ennaceur and Delacour (1988) and Bartolini et al. (1996). Each mouse was first placed (acquisition trial) in an open box (45 × 45 × 30 cm) made of wood with the inside painted black, and exposed to two identical objects (O_1a and O_1b) for 3 min. The mouse was then returned to its home cage. After a delay of 1, 4, 8, or 24 h the mouse was placed (recall trial) in another box (34 × 34 × 30 cm) made of Plexiglas and painted white, and then presented with one of the familiar objects (O₁) and a novel object (O₂) for a further 3 min. The objects were placed approximately 10 cm distant from two adjacent corners of the box. Two kinds of wood objects were used: a parallelepiped (2 × 2 × 5 cm) and a pyramid (4 × 4 × 3 cm). The role (familiar and novel), as well as the location of the two choice objects, was counterbalanced between mice. As far as could be ascertained, the objects had no natural significance for the mice and they had never been associated with a reinforcer. Care was taken to avoid olfactory stimuli by cleaning the objects carefully. The time (t) spent (in seconds) exploring the objects (O₁ and O₂) was recorded. Exploration was operationally defined as directing the nose to the object at a distance of less than 2 cm and/or touching it with the nose (Ennaceur and Delacour, 1988). A discrimination ratio was calculated using the formula [tO₂/(tO₁ + tO₂)] × 100.

Drugs. Dexefaroxan hydrochloride (2-[2-(2-ethyl-2,3-dihydrobenzofuranyl)]-2 imidazoline) and UK 14304 tartrate were synthesized at Pierre Fabre Medicament (Castres, France). Tacrine hydrochloride was obtained from Sigma-Aldrich (Saint Quentin, France), diazepam base from Interchim (Paris, France), and scopolamine hydrobromide from Fluka (Saint Quentin, France). All drugs were dissolved in distilled water. An injection volume of 10 ml/kg was used throughout. Doses refer to the free base and were selected from the geometrical series 0.0025, 0.01, ... 2.5, and 10 mg/kg. For the continuous infusion experiments, dexefaroxan was delivered subcutaneously (0.04, 0.16, 0.63, or 2.5 mg/rat/day) by Alzet osmotic minipumps (model 2 ML4; Alza, Palo Alto, CA). The mean pumping rate was 2.60 µl/h (62.4 µl/day), and the duration of pumping was 33.8 days. For the control rats, the minipumps were filled with the vehicle solution (0.9% sterile saline).

Statistics. In the passive avoidance test, all results were compared using a Kruskal-Wallis nonparametric one-way analysis of variance corrected for ties, followed by a two-tailed Mann-Whitney U test. Results are, however, expressed as the mean ± S.E.M. in spite of the probable nonnormality of the distribution of scores, because it was felt that these parameters provide a clearer indication for most investigators.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Passive Avoidance Behavior in Rats: UK 14304-Induced Deficits. The ₂-adrenoceptor agonist UK 14304, given 30 min before the acquisition (training) trial, produced dose-dependent deficits in the passive avoidance task (H = 18.39, p = 0.001) (Table 1). At the doses of 0.04 to 2.5 mg/kg i.p., dexefaroxan significantly reversed the effects of 1.25 mg/kg UK 14304 (H = 13.94, p = 0.016). The dose-response curve of dexefaroxan was bell-shaped with a maximal effect at 0.63 mg/kg (Fig. 1A). Under the same conditions, tacrine (0.16-10 mg/kg i.p.) was inactive (H = 0.99, p = 0.91) (Fig. 1A).

View larger version (31K): [in this window] [in a new window]   Fig. 1.   Effects of tacrine and dexefaroxan on memory deficits induced by amnesic compounds (AC) in a passive avoidance task in rats. A, AC = 1.25 mg/kg i.p. UK 14304. B, AC = 2.5 mg/kg i.p. scopolamine. C, AC = 20 mg/kg i.p. diazepam. Tacrine and dexefaroxan were administered i.p. 35 min before the initial trial (training). AC were administered 30 min before training. Animals were trained with a single electric footshock of 0.8 mA/2 s and were tested 48 h later. Results are mean ± S.E.M. of 11 to 24 rats. *, p < 0.05; **, p < 0.01; ***, p < 0.001 compared with respective [AC + vehicle] group (Mann-Whitney U test).

Passive Avoidance Behavior in Rats: Scopolamine-Induced Deficits. Scopolamine administered 30 min before training produced a robust and significant dose-dependent reduction in the step-through latencies, compared with the vehicle control group (H = 13.49, p = 0.004) (Table 1). At the doses of 0.04 to 2.5 mg/kg i.p., dexefaroxan significantly attenuated the effects of 2.5 mg/kg scopolamine (H = 18.06, p = 0.003). The dose-response curve of dexefaroxan was bell-shaped with a maximal effect at 0.63 mg/kg i.p. (Fig. 1B). Under the same conditions, tacrine was also active from 0.04 to 2.5 mg/kg i.p. with a bell-shaped dose-response curve and a maximal effect at 0.16 mg/kg i.p. (H = 33.70, p < 0.001) (Fig. 1B). No significant difference was found between the maximal effects obtained with dexefaroxan and tacrine (U = 63, p = 0.56).

Passive Avoidance Behavior in Rats: Diazepam-Induced Deficits. Diazepam given 30 min before the training trial significantly impaired the acquisition/retention of the passive avoidance task in rats (H = 34.56, p < 0.001) (Table 1). At the dose of 0.63 mg/kg i.p., dexefaroxan significantly attenuated the effects of 20 mg/kg diazepam (H = 21.46, p = 0.002). The dose-response curve of dexefaroxan was bell-shaped with a maximal effect at 0.63 mg/kg (Fig. 1C). Under the same conditions, tacrine was also active at the doses of 0.16 and 0.63 mg/kg (H = 19.76, p = 0.002) (Fig. 1C). No significant difference was found between the maximal effects obtained with dexefaroxan and tacrine (U = 68, p = 0.77).

Passive Avoidance Behavior in Rats: Training-Testing Interval of 15 Weeks. An interval of 15 weeks between the training trial and the testing trial significantly impaired the memory performance in a passive avoidance task, compared with an interval of 48 h or 5 weeks (H = 9.17, p = 0.011) (Table 2). At the doses of 0.16 and 0.63 mg/kg i.p., dexefaroxan significantly enhanced memory performance, and thus attenuated the spontaneous forgetting induced by a 15-week forgetting period (H = 11.27, p = 0.046). The dose-response curve of dexefaroxan was bell-shaped with a maximal effect at 0.16 mg/kg (Fig. 2). Under the same conditions, tacrine, tested from 0.04 to 10 mg/kg, was inactive (H = 6.11, p = 0.29) (Fig. 2).

View larger version (30K): [in this window] [in a new window]   Fig. 2.   Effects of tacrine and dexefaroxan on memory in a passive avoidance task with a training-testing interval (T-T) of 15 weeks in rats. Tacrine and dexefaroxan were administered i.p. 30 min before the initial trial (training). Animals were trained with a single electric footshock of 0.8 mA/2 s. Results are mean ± S.E.M. of 12 rats. *, p < 0.05; **, p < 0.01 compared with vehicle group with a T-T of 15 weeks (Mann-Whitney U test).

Passive Avoidance Behavior in Rats Trained with a Weak Electric Footshock. The step-through latency of rats trained with electric footshock from 0.2 to 0.8 mA/2 s was significantly higher than that of rats trained without electric footshock (H = 43.38, p < 0.001). A high intensity of 0.8 mA/2 s induced a maximal latency (180 s), and this performance gradually diminished with the progressive decrease of the intensity of the electric footshock (Table 2). At the doses of 0.16 and 0.63 mg/kg i.p., dexefaroxan given immediately after the training trial, significantly increased the memory performance of rats trained with a weak electric footshock of 0.1 mA/2 s (H = 10.46, p = 0.033). The dose-response curve of dexefaroxan was bell-shaped with a maximal effect at 0.63 mg/kg (Fig. 3). Under the same conditions, tacrine was also active but only at the dose of 0.63 mg/kg (H = 9.68, p = 0.046) (Fig. 3). No significant difference was found between the maximal effects obtained with dexefaroxan and tacrine (U = 71.5, p = 0.93).

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Effects of tacrine and dexefaroxan on passive avoidance behavior in rats trained with a weak electric footshock (E.F.) of 0.1 mA/2 s. Tacrine and dexefaroxan were administered i.p. immediately after the initial trial (training). Animals were tested 48 h later. Results are mean ± S.E.M. of 12 rats. *, p < 0.05; ***, p < 0.001 compared with vehicle group trained with an E.F. = 0.1 mA/2 s (Mann-Whitney U test).

Morris Water Maze Test of Spatial Memory. In the Morris water maze test, performances are dependent on the number of training days (Table 3). In a probe trial without platform, 24 h after three or four training days, rats spend significantly more than 25% of their time (chance performance) in the quadrant that contained the platform during the training sessions (Q₄), indicating that they have "learned" the location of the platform in this quadrant. However, after only one or two training days, equal time is spent in all the four quadrants of the pool, suggesting that rats have not yet learned the platform location (Table 3). Because, after two training days, the spatial memory performance of control animals is not significantly different from chance, a potential promnesic effect of a compound can be detected. This test paradigm was used to evaluate the potential promnesic effects of dexefaroxan, in comparison with tacrine, on spatial memory in rats. Tacrine and dexefaroxan were given immediately after the training sessions of days 1 and 2. On day 3 in a probe trial, without platform, 24 h after the second administration of drug, rats treated with dexefaroxan at 0.16, 0.63, and 2.5 mg/kg i.p. spent significantly more than 25% of their time in the target quadrant Q₄ (Fig. 4). The performance of these rats were also significantly higher than those of control animals [F_(5,74) = 2.38, p = 0.046]. The dose-response curve of dexefaroxan was bell-shaped with a maximal effect at 0.16 mg/kg i.p. Under the same conditions, tacrine was also active at 0.16, 0.63, and 2.5 mg/kg i.p. with a maximal effect at 0.63 mg/kg i.p. [F_(3,45) = 2.82, p = 0.048] (Fig. 4). No significant difference was found between the maximal effects obtained with dexefaroxan and tacrine (t = 0.16, p = 0.87).

View larger version (45K): [in this window] [in a new window]   Fig. 4.   Effects of tacrine and dexefaroxan in the Morris water maze task: percentage of time during the 60-s swim spent by rats in the target quadrant (without platform) during the probe trial on day 3. Tacrine and dexefaroxan were administered i.p. immediately after the training sessions of days 1 and 2. Results are mean ± S.E.M. of 8 to 16 rats. +, p < 0.05; ++, p < 0.01; +++, p < 0.001 compared with chance performance (25%); o, p < 0.10; *, p < 0.05; **, p < 0.01 compared with vehicle (Student's t test).

Object Recognition Test in Mice. During the acquisition trial, there was no significant difference in the time spent in exploring the two identical objects (O_1a and O_1b), thus indicating that individual animals had no preference for a specific object or place (Table 4). During the recall session, after a delay interval of 1 h, control mice spent significantly more time exploring the novel object (O₂), indicating that they recognized the familiar one (O₁). The discrimination ratio (71.3%) was significantly different from chance performance (50%) (t = 2.42, p = 0.038). After a 4, 8, or 24 h interval, control mice did not discriminate between the familiar and the novel object, and the discrimination ratios were not significantly different from chance performance (Table 4). To detect a potential promnesic effect of a compound, the acquisition-recall interval used was 4 h. Tacrine and dexefaroxan were given immediately after the acquisition session. In the recall session, mice treated with dexefaroxan at 0.0025 to 10 mg/kg i.p. spent significantly more than 50% of their object exploration time with the novel object (Fig. 5). The performance of these mice was also significantly higher than that of control animals [F_(8,91) = 4.26, p = 0.0002]. The dose-response curve of dexefaroxan was somewhat irregular with a maximal effect at 10 mg/kg i.p. Under the same conditions, tacrine was also active at 0.16 and 2.5 mg/kg i.p. with a maximal effect at 2.5 mg/kg i.p. [F_(6,103) = 3.27, p = 0.006] (Fig. 5). Note that the discrimination ratio of mice treated by 0.63 mg/kg i.p. tacrine was not significantly different from the ratio of control mice; only a tendency (t = 1.74, p = 0.098) to be different from the chance performance (50%) was found. No significant difference was found between the maximal effects obtained with dexefaroxan and tacrine (t = 1.54, p = 0.14).

View larger version (45K): [in this window] [in a new window]   Fig. 5.   Effects of tacrine and dexefaroxan on the discrimination ratio in a two-trial object recognition task in mice. In a first trial, mice were exposed to two identical objects (O1a and O1b) for 3 min and in a second trial (4 h after the first trial) to two dissimilar objects, a familiar (O1) and a novel one (O2), for 3 min. Tacrine and dexefaroxan were administered i.p. immediately after the first trial. Results are mean ± S.E.M. of 10 to 20 mice/group. Discrimination ratio = [tO2/(tO1 + tO2)] × 100. , p < 0.10; +, p < 0.05; ++, p < 0.01; +++, p < 0.001 compared with chance performance (50%); *, p < 0.05; **, p < 0.01; ***, p < 0.001 compared with vehicle (Student's t test).

Chronic Drug Treatments by Using Subcutaneous Osmotic Minipumps in Passive Avoidance Test. Twenty-one days after subcutaneous implantation of the osmotic minipumps, the continuous infusion of dexefaroxan at the doses of 0.16 and 0.63 mg/rat/day significantly antagonized the memory deficit induced by acute scopolamine (2.5 mg/kg i.p.) in a passive avoidance task in rats (H = 11.14, p = 0.025). The dose-response curve of dexefaroxan was bell-shaped (Fig. 6A). Note that a tendency (U = 26.5, p = 0.064) to increase the step-through latency was found with 0.04 mg/rat/day of dexefaroxan, but without attaining the criteria for statistical significance (p < 0.05) (Fig. 6A).

View larger version (37K): [in this window] [in a new window]   Fig. 6.   Effects of a continuous subcutaneous infusion of dexefaroxan via Alzet osmotic minipumps. Results are mean ± S.E.M. of 10 rats. A, effects on scopolamine-induced memory deficit in a passive avoidance task in rats. On day 21 of infusion, scopolamine was administered i.p. at a dose of 2.5 mg/kg, 30 min before training. Animals were trained with a single electric footshock of 0.8 mA/2 s and were tested 48 h later. , p < 0.10; **, p < 0.01 compared with [scopolamine + vehicle] group (Mann-Whitney U test). B, effects of dexefaroxan infusion for 25 days on the percentage of time during the 60-s swim spent by rats in the target quadrant (without platform) during the probe trial on day 3 in the Morris water maze task. ++, p < 0.01; +++, p < 0.001 compared with chance performance (25%); *, p < 0.05; **, p < 0.01 compared with vehicle (Student's t test).

Chronic Drug Treatments by Using Subcutaneous Osmotic Minipumps in Morris Water Maze Test. On day 3 in a probe trial, without platform, rats chronically infused for 25 days with dexefaroxan at 0.16 and 0.63 mg/rat/day spent significantly more than 25% of their time in the target quadrant Q₄. The performances of these rats were also significantly higher than those of control animals [F_(4,45) = 4.61, p = 0.003]. The dose-response curve of dexefaroxan was bell-shaped (Fig. 6B).

Cognitive Deficits in Aged Rats: Passive Avoidance Test. Aged (24-month-old) rats exhibited a significant decrease in retention of the inhibitory avoidance task compared with 2-month-old animals (U = 34, p = 0.024) (Fig. 7A). In the 24-month-old rats, 0.63 mg/kg i.p. dexefaroxan significantly increased the time taken to enter the dark compartment (U = 43, p = 0.031). Under the same conditions, a tendency (U = 48, p = 0.057) to increase the step-through latency was found with 0.63 mg/kg i.p. tacrine, but without attaining the criteria for statistical significance (p < 0.05) (Fig. 7A). The performances of old rats treated with dexefaroxan or tacrine were not significantly different from those of young control animals (U = 71, p = 0.47 and U = 70, p = 0.44, respectively). Note that the number of old rats available for study was limited, and thus did not permit a complete dose-response evaluation.

View larger version (34K): [in this window] [in a new window]   Fig. 7.   Effects of tacrine and dexefaroxan in aged rats (24 months). A, effects on passive avoidance behavior. Tacrine and dexefaroxan were administered i.p. 30 min before the initial trial (training). Animals were trained with a single electric footshock of 0.8 mA/2 s and were tested 1 week later. B, effects on the percentage of time during the 60-s swim spent by rats in the target quadrant (without platform) during the probe trial on day 4 in the Morris water maze task. Tacrine and dexefaroxan were administered i.p. immediately after the training sessions of days 1, 2, and 3. Results are mean ± S.E.M. of 12 to 14 rats. ++, p < 0.01 compared with chance performance (25%); , p < 0.10; *, p < 0.05; **, p < 0.01 compared with aged control rats (24-month-old vehicle group).

Cognitive Deficits in Aged Rats: Morris Water Maze Test. Rats received three training days and were tested in a probe trial, without platform, on the 4th day. Tacrine and dexefaroxan were given immediately after the training sessions of days 1, 2, and 3, at the dose of 0.63 mg/kg i.p. On day 4 in a probe trial, without platform, young rats (2 months) spent significantly more than 25% of their time in the target quadrant Q₄ (44.0%), indicating that they had learned the location of the platform in this quadrant. In contrast, old rats (24 months) spent no more time in the target quadrant Q₄ (23.8%) than in the other three quadrants, suggesting that they had not learned the platform location. Thus, old rats exhibited a significant decrease of performance in the water maze task compared with 2-month-old animals (t = 4.64, p < 0.001) (Fig. 7B). However, 24-month-old rats treated with 0.63 mg/kg i.p. dexefaroxan spent significantly more than 25% of their time in the target quadrant Q₄. The performances of these rats were also significantly higher than those of aged control animals (t = 3.32, p = 0.007). Under the same conditions, tacrine was also active at 0.63 mg/kg i.p. (t = 4.59, p = 0.001) (Fig. 7B). No significant difference was found between the effects obtained with dexefaroxan and tacrine (t = 1.52, p = 0.15). The performances of old rats treated with dexefaroxan or tacrine were not significantly different from those of young control animals (t = 1.77, p = 0.11 and t = 1.34, p = 0.21, respectively).

Cognitive Deficits in Nucleus Basalis-Lesioned Rats: Passive Avoidance Test. Two weeks postsurgery, rats with bilateral ibotenic acid induced-lesions of the NBM exhibited a significant decrease in retention of the inhibitory avoidance task compared with sham control animals (U = 24, p = 0.004) (Fig. 8A). In the NBM-lesioned rats, dexefaroxan significantly increased the time taken to enter the dark compartment with a minimal significant dose of 0.63 mg/kg i.p. (H = 8.40, p = 0.038). The dose-response curve was bell-shaped with a maximal effect at 0.63 mg/kg i.p. (Fig. 8A). Under the same conditions, tacrine was also active, but only at the dose of 0.63 mg/kg i.p. (H = 8.02, p = 0.046). The performances of NBM-lesioned rats treated with 0.16, 0.63, and 2.5 mg/kg i.p. dexefaroxan or 0.16 and 0.63 mg/kg i.p. tacrine were not significantly different from those of sham control animals (Fig. 8A).

View larger version (38K): [in this window] [in a new window]   Fig. 8.   Effects of tacrine and dexefaroxan in rats with bilateral ibotenic acid-induced lesions of the nucleus basalis magnocellularis. A, effects on passive avoidance behavior. Two weeks postsurgery, tacrine and dexefaroxan were administered i.p. 30 min before the initial trial (training). Animals were trained with a single electric footshock of 0.8 mA/2 s and were tested 1 week later. B, effects on the percentage of time during the 60-s swim spent by rats in the target quadrant (without platform) during the probe trial on day 4 in the Morris water maze task. Three weeks postsurgery, tacrine and dexefaroxan were administered i.p. immediately after the training sessions of days 1, 2, and 3. Results are mean ± S.E.M. of 8 to 20 rats. ++, p < 0.01 compared with chance performance (25%); *, p < 0.05; **, p < 0.01 compared with lesioned control rats (lesioned vehicle group).

Cognitive Deficits in Nucleus Basalis-Lesioned Rats: Morris Water Maze Test. Three weeks postsurgery, rats received three training days and were tested in a probe trial, without platform, on the 4th day. Tacrine and dexefaroxan (or vehicle) were given immediately after the third training trials on days 1, 2, and 3. On day 4 in a probe trial, without platform, sham control rats spent significantly more than 25% of their time in the target quadrant Q₄ (38.5%), indicating that they had learned the location of the platform in this quadrant. In contrast, rats with bilateral NBM lesions spent no more time in the target quadrant Q₄ (23.5%) than in the other three quadrants, suggesting that they had not learned the platform location. Thus, NBM-lesioned rats exhibited a significant decrease of performance in the water maze task compared with sham control animals (t = 5.22, p < 0.001) (Fig. 8B). However, NBM-lesioned rats treated with 0.63 and 2.5 mg/kg i.p. dexefaroxan spent significantly more than 25% of their time in the target quadrant Q₄. The dose-response curve of dexefaroxan was bell-shaped with a minimal significant dose and a maximal effect at the same dose of 0.63 mg/kg i.p. (Fig. 8B). The performance of these rats was also significantly higher than that of lesioned control animals [F_(3,52) = 10.08, p < 0.001]. Under the same conditions, tacrine was also active at 0.63 and 2.5 mg/kg i.p. [F_(3,46) = 7.78, p < 0.001]. The performances of NBM-lesioned rats treated with 0.63 and 2.5 mg/kg i.p. dexefaroxan or tacrine were not significantly different from those of sham control animals (Fig. 8B).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In the present studies, the selective ₂-adrenoceptor antagonist dexefaroxan was found to ameliorate the performances of rodents in tests evaluating different aspects of cognitive function. Overall, theses effects were comparable to or in some cases superior to those observed with the cholinesterase inhibitor tacrine. Dexefaroxan blocked the deficits induced by UK 14304, scopolamine, or diazepam in passive avoidance behavior in rats. A rationale for using these pharmacological probes to induce cognitive deficits in animals is provided by the amnesic properties that these classes of agents exert in humans. For example: 1) the ₂-adrenoceptor agonist clonidine disrupts memory performance in healthy volunteers (Frith et al., 1985) and in patients with AD (Riekkinen et al., 1999); 2) consistent with the cholinergic hypothesis of learning and memory, centrally active muscarinic cholinergic receptor antagonists such as scopolamine have amnesic properties in rodents, subhuman primates, and in humans (Chopin and Briley, 1992; Bartus, 2000). Similarities in the memory impairments between Alzheimer patients and scopolamine-treated animals have been reported, and it has been proposed that scopolamine could serve as a useful pharmacological tool to produce a partial model of the disorder (Bartus, 2000); 3) the deficit induced by diazepam is consistent with the anterograde amnesia induced by benzodiazepines in animals and nonanxious volunteer subjects (Curran, 1991). The acetylcholinesterase inhibitor tacrine was also active against the deficits induced by scopolamine or diazepam; however, unlike the positive effects observed with dexefaroxan, tacrine was without significant effects on the memory deficit induced by UK 14304, and was also inactive in young adult rats in the 15-week spontaneous forgetting paradigm.

A compound administered before the training trial can be thought to act on the memory processes of acquisition and/or retention, whereas administered after training it would be considered active only on retention. Dexefaroxan or tacrine, given immediately after training, increased the memory performance of rats trained with the weak electric footshock procedure. Thus, it would appear that dexefaroxan has a promnesic effect in this test by facilitating the memory retention processes. In two other tests (Morris water maze and object recognition), dexefaroxan and tacrine were also given after training, and thus could influence only the memory retention phase and not the processes of acquisition or the other noncognitive influences such as arousal and motivation. The results showed promnesic effects of dexefaroxan and tacrine by facilitating spatial memory processes in the Morris water maze task in rats, and by increasing the memory performance of mice in the object recognition test. Because this latter test essentially involves episodic memory (Bartolini et al., 1996), the results indicate that dexefaroxan has a promnesic effect by facilitating episodic memory processes. Interestingly, episodic memory has been considered to be a form of recent memory in humans, and one which is impaired in aging and in early stages of AD (M'Harzi et al., 1997; Bartus, 2000).

To investigate whether the effects of dexefaroxan to ameliorate performance in passive avoidance and spatial memory paradigms could be maintained during a chronic treatment period, adult rats were implanted with Alzet osmotic minipumps to deliver a continuous subcutaneous infusion of the drug for up to 33.8 days. When tested on day 21 of infusion, rats receiving dexefaroxan showed a significantly higher memory performance, compared with vehicle-infused controls, when challenged with scopolamine in the passive avoidance test. Similarly, rats continuously infused with dexefaroxan for 25 days showed a significant facilitation of spatial memory in the Morris water maze task. These results indicate that after a chronic subcutaneous drug infusion of up to 25 days, tolerance (tachyphylaxis) does not occur to the effect of dexefaroxan in facilitating cognitive performance in these two different tests in the rat.

Aged rodents exhibit impaired memory, especially in performance of single-trial inhibitory avoidance tasks (Bartus, 2000) and in the Morris water maze test in rats (Lindner, 1997). In the present study, 24-month-old rats exhibited a significant decrease of performance in both tests compared with 2-month-old animals. Notably, the age-related deficits observed in both tests were reversed by dexefaroxan and tacrine, and to an extent that the performances of old rats treated with dexefaroxan or tacrine were not significantly different from those of young controls.

Lesioning of the NBM in the rat by local injection of the excitotoxin ibotenic acid is known to partially mimic some of the complex behavioral and neurochemical deficits characteristic of AD, and has for this reason been widely used as a model for studying the role of central cholinergic pathways in some cognitive processes and for identifying potentially useful pharmacological treatments (Bartus et al., 1986; Pepeu et al., 1986). As expected, and as a validation of the lesioning procedure, bilateral NBM-lesioned rats showed significant performance deficits in both the passive avoidance and Morris water maze tests, in agreement with previous studies (O'Connell et al., 1994). Dexefaroxan or tacrine significantly attenuated these deficits and to an extent that the performances of lesioned rats were not significantly different from those of sham control animals.

In the present study, the effects of dexefaroxan were consistently dose-related, describing bell-shaped dose-response curves with optimal effects occurring at or around 0.63 mg/kg i.p. The bell shape of the dose-response curves and the same peak effective dose of dexefaroxan reported in this study was observed also on cortical acetylcholine outflow, as measured in vivo by microdialysis in the rat (Tellez et al., 1997), and on circling behavior in rats with unilateral 6-hydroxydopamine lesions of the nigrostriatal pathway (Chopin et al., 1999). Bell-shaped dose-response curves have also been reported for the ₂-adrenoceptor antagonists idazoxan and yohimbine in reversing the loss of the righting reflex in behavioral studies (Colpaert, 1986), and on cortical high-voltage spindles in electroencephalogram studies (Yavich et al., 1994). Although the inverted U-shape of the dose-response curve has been attributed to partial agonist actions at ₂-adrenoceptors (Colpaert, 1986) and to actions at non₂-receptors (Yavich et al., 1994), the precise mechanisms underlying this phenomenon are not known for certain.

The present results with dexefaroxan are in general agreement with previous studies showing that other ₂-adrenoceptor antagonists, such as atipamezole or RU-52583 (3-11, methyl-20, 21-dinoreburnamenine), augment cognitive performances in young adult rats (Haapalinna et al., 1998), and in animals with deficits associated with aging (Haapalinna et al., 2000) or induced by lesioning of the basal forebrain nuclei (M'Harzi et al., 1997).

In memory tests in which tacrine was active, no significant difference was apparent between the maximal effects obtained with dexefaroxan and tacrine. Dexefaroxan, by enhancing acetylcholine release (Tellez et al., 1997), may be effective in memory tests in part by increasing extracellular acetylcholine levels; tacrine presumably accomplishes the same but through a different mechanism of action, the inhibition of acetylcholinesterase. Remarkably, however, dexefaroxan was also active, where tacrine was not, in a passive avoidance task with a training-testing interval of 15 weeks and against UK 14304-induced deficits. It is not likely that a lack of effect of tacrine in those tests was simply due to using nonoptimal doses. Evidence for its efficacy in most other tasks argues against this. Furthermore, most experiments incorporated a wide enough (up to 250-fold) dose range to cover the pharmacologically relevant doses of tacrine that have been reported for the rat [i.e., 0.04 to 10 mg/kg (Bartus, 2000)]. These findings suggest that dexefaroxan has other mnesic properties that tacrine does not. Blockade of autoreceptors by ₂-adrenoceptor antagonists increases the release of noradrenaline in the brain (Dennis et al., 1987). The central noradrenergic system plays an important role in memory processes (Aston-Jones et al., 1991; Berridge et al., 1993; Sara et al., 1994), and the stimulation of this neurotransmitter system can improve cognitive functions (Haapalinna et al., 1998). For example, post-training infusions of noradrenaline or the -adrenoceptor agonist clenbuterol into the basolateral nucleus of amygdala enhance memory retention of rats tested in passive avoidance and water maze tasks (Ferry et al., 1999a,b). Furthermore, the memory-enhancing effects of dexefaroxan could also in part be the result of ₂-adrenoceptor antagonist-induced facilitation in the release of other neurotransmitters, such as dopamine or serotonin (Raiteri et al., 1990; Matsumoto et al., 1998), which also modulate learning and memory processes.

In conclusion, dexefaroxan clearly facilitates various aspects of memory in young adult rats, in a manner comparable to or superior to that found with the cholinesterase inhibitor tacrine. Furthermore, the facilitatory effects of dexefaroxan were found to persist even after a 21- to 25-day constant subcutaneous infusion by using osmotic minipumps, indicating that tolerance to the promnesic effect of the drug does not occur during this chronic treatment interval. Importantly, dexefaroxan ameliorates the age-related memory deficits of old rats, and to a level that is comparable to that of young adult animals, and reverses the memory deficits induced by excitotoxin lesions of the nucleus basalis magnocellularis region in the rat. Together, this body of evidence supports the potential utility of dexefaroxan in the treatment of cognitive deficits occurring in neurodegenerative disorders such as Alzheimer's disease.