Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Although many factors have been attributed to the cause of AD, a consistent finding is a depressed central cholinergic system, characterized by decreased presynaptic cholinergic markers such as choline acetyltransferase (Bowen et al., 1979; Davies and Maloney, 1976), and by degeneration of cholinergic neurons in the nucleus basalis of Meynert (Bartus et al., 1982; Coyle et al., 1983; Whitehouse et al., 1981, 1982). This association, and the production of memory impairments from induced cholinergic hypofunction, have prompted considerable interest in cholinergic replacement therapy. Several cholinesterase inhibitors have been investigated clinically in AD patients, including THA (Cognex) (Summers et al., 1986; Davis et al., 1993), velnacrine (Siegfried, 1995), galanthamine (Kewitz et al., 1994; Thomsen et al., 1990) and HEP (Brufani et al., 1987). Of these, THA has shown clinical efficacy in 20 to 30% of AD patients and is the only agent to receive FDA approval for AD therapy (Knapp et al., 1994).

In general, clinical studies with THA show that memory can be moderately improved with an AChEI. However, the maximum efficacy of AChEI in AD therapy has not been satisfactorily determined because of dose-limiting cholinergic side effects and liver toxicity, especially with THA. In addition to the limitations associated with individual AChEI, the heterogeneous pathophysiology of AD is another factor to consider when assessing the efficacy of AChEI therapy. For example, a subpopulation of AD patients, e.g., those lacking the ApoE4 allele, show preferential improvement during THA treatment. In contrast, AD patients with at least one copy of the ApoE4 allele are insensitive to THA treatment (Poirier et al., 1995). Therefore, the development of chemically novel AChEI that are tolerable and relatively safe remains a therapeutic goal. Ongoing clinical investigations with donepezil (Aricept; E-2020) affirm this approach (Rogers and Friedhoff, 1996).

The studies in this report describe the pharmacological, behavioral and safety profiles of P10358 (fig. 1), a chemically novel AChEI for the treatment of AD. The discovery and characterization of P10358 started with in vitro and ex vivo potency determinations of AChE and BuChE inhibition. This was followed by preliminary in vivo observations of cholinergic stimulation (motor activity and hypothermia) and side effects. AChEI-induced body temperature reduction was assessed because hypothermia can be used as a physiological measure of central cholinergic stimulation and duration of drug action (Gordon, 1994). Behavioral correlates of central cholinergic function were performed in rats (reversal of scopolamine dementia in the water maze; enhancement of step-down passive avoidance) and mice (enhancement of social recognition). The ability of P10358 to alter brain dopamine neurotransmission was evaluated since muscarinic agonists and AChEI (e.g., THA, E2020) can increase extracellular levels of dopamine and its metabolite HVA (Yamanishi et al., 1992; Xu et al., 1989). Finally, cardiovascular studies were performed to assess autonomic liability and estimate a safety index, e.g., ratio between behaviorally active doses and those that affect arterial pressure and heart rate.

View larger version (8K): [in this window] [in a new window]   Fig. 1.   Chemical structure of P10358, a novel AChE inhibitor.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Animals. Male Wistar rats (150-250 g), Sprague-Dawley rats (275-325 g) and Swiss-Webster, CD-1 and CFW mice (18-30 g) were purchased from Charles River Laboratories (Wilmington, MA). ND4 mice were purchased from Harlan (Indianapolis, IN). All animals were housed in our colony before testing and were kept on a 12-hr light cycle from 0600 to 1800 hr with free access to food and water. Studies were carried out in accordance with the Declaration of Helsinki and with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the National Institutes of Health.

Chemicals. P10358 and HEP were synthesized by Chemical Research at Hoechst Marion Roussel, Inc. (Bridgewater, NJ) (Davis et al., 1992). These agents were dissolved in acidified distilled water and all doses refer to the free base. All other reagents were obtained from commercial sources.

In Vitro Studies

Cholinesterase activity determinations. AChE and BuChE activities were determined by a modification of the Ellman method (Ellman et al., 1961) as described previously (Bores et al., 1996). Wistar rat striatal homogenates (AChE) or human serum (BuChE) were assayed with 5 mM acetylthiocholine or butyrylthiocholine as the substrates, respectively.

Ex vivo AChE activity. Groups of three male Wistar rats or Swiss Webster mice were dosed s.c. or p.o.with several doses of P10358 or HEP and killed at various times thereafter. All animals were fasted overnight (17 hr) before oral dosing. Residual AChE activity was assayed as above using rat striatal or mouse forebrain homogenate preparations (1:19 w/v) and 5 mM acetylthiocholine (Bores et al., 1996). Statistical differences were determined by the Newman-Keuls test after a one-way analysis of variance and ED₅₀ values determined by log-probit analysis. It must be noted that ex vivo measures of AChE inhibition may underestimate the actual degree of inhibition in vivo due to dissociation that occurs during tissue dilution (Bores et al., 1996).

Radioligand binding profile. The ability of P10358 to nonselectively bind various neurotransmitter receptors, amine transporters and ion channels was evaluated with conventional radioligand binding methods. The following receptor (radiolabel) types were evaluated in-house: alpha-1 ([³H]prazosin) and alpha-2 adrenergic ([³H]yohimbine); D2 dopaminergic ([³H]N-methylspiroperidol); muscarinic ([³H]QNB); 5-HT uptake ([³H]5-HT); NE uptake ([³H]NE); DA uptake ([³H]DA). The following were assayed at Hoechst Central Screening (Frankfurt, Germany): A1 ([³H]DPCPX) and A2 adenosine ([³H]CGS-21680); angiotensin II ([³H]angiotensin II); bradykinin ([³H]bradykinin); endothelin-B ([³H]endothelin); GABA-A ([³H]muscimol); NMDA ([³H]MK-801); nicotinic ([³H]cystisine); µ-opioid ([³H]naloxone); quisqualate ([³H]AMPA); 5-HT_2A ([³H]ketanserin); Ca⁺⁺ channel ([³H]nitrendipine); K⁺_ATP channel ([³H]glibenclamide) and adenosine uptake ([³H]NBTI).

In Vivo Studies

Primary cholinergic effects. Overt cholinergic activity in whole animals was assessed as described earlier (Irwin, 1968; Fielding et al., 1974). Groups of four Wistar rats or four Swiss Webster mice were administered P10358 and observed for a number of overt behavioral effects (tremor; lethality) for up to 6 hr after dosing. Lethality was also assessed 24 hr after administration. Cholinergic activity was expressed as the number of animals exhibiting each parameter vs. the total number of animals treated. ED₅₀ values for tremor were determined using nonlinear regression analysis.

Hypothermia. Groups of four Swiss-Webster mice were randomly assigned to a vehicle control group or drug treatment groups, and two base-line pretreatment rectal temperatures were taken for each animal using a YSI series 500 temperature probe with readouts recorded on the YSI model 43 tel-thermometer (Yellow Springs Instrument Company, Yellow Springs, OH). Immediately on completion of base-line temperature readings, the animals were dosed with P10358 or vehicle control and rectal temperature measurements were made for each animal at 30-min intervals, starting 30 min postdose and ending after 5 hr. Group means were analyzed using a two-way repeated measures analysis of variance, followed by a LSD post hoc test.

In vivo effects on dopamine metabolism. Fasted male Wistar rats were killed 1 hr after treatment with HEP or P10358. Striata were removed and frozen at -80°C until assay. The striata were homogenized in 100 mM perchloric acid with 4.3 mM EDTA and centrifuged for 10 min at 10,000 rpm. The supernatants were transferred to microfilterfuge tubes (0.2 µm, Rainin Instrument Co., Woburn, MA) and recentrifuged as above. The high performance liquid chromatography system used a C18-ODS Hypersil, 3 µm, 100 × 4.6 mm column with electrochemical detection. The aqueous mobile phase consisted of 0.07 M sodium acetate, 0.04 M citric acid, 130 µM EDTA and 230 µM sodium octane sulfonate and contained methanol (buffer:methanol at 92.5:7.5, v/v) (Wagner et al., 1982). External standards were injected at 10-sample intervals. The flow rate was 1.0 ml/min. Retention times for DA, DOPAC and HVA were 5.2, 7.4 and 15.8 min, respectively.

Assessment of spatial memory. The ability of P10358 to reverse SCOP-induced spatial memory deficits was evaluated in the Morris water maze. In these studies, Sprague-Dawley rats were pretreated with SCOP (1 mg/kg, i.p.) and P10358 (1.25 and 2.5 mg/kg, p.o.) 30 min before testing each day. In a separate experiment, we determined that the oral administration of 1 or 3 mg/kg of P10358 caused significant and sustained inhibition of rat striatal AChE (1 mg/kg, p.o.: 1 hr, 25.8 ± 5.7% inhibition; two hr, 27.6 ± 4.3% inhibition; 3 mg/kg, p.o.: 1 hr, 25.6 ± 3.2% inhibition; 2 hr, 36.7 ± 1.4% inhibition. All values were significantly different from control (P < .01, grouped t test). Each rat's acquisition of spatial learning was assessed over 2 days. Briefly, a large black circular pool (140 cm diameter) was half filled with 22°C tap water. The tank was divided into four equal quadrants, and a 12 × 12 cm black plastic, platform with small holes to provide a gripping surface was submerged 2 cm below the water level in the center of one quadrant. The platform was not visible to the rats and remained in one location for the entire test. Preliminary tests with a different set of animals were conducted to verify the platform's hidden nature. Undrugged rats that had previously learned the platform's location were administered a probe trial in which the platform location had been changed to the opposite side of the water maze. During this trial animals swam directly to the original platform location and continued to search that quadrant of the maze. Only after this erroneous escape attempt did the rats explore the maze and randomly locate the platform in its new location. It was concluded that because the rats swam to the original platform location and not the new location, they were unable to see the platform itself. Had they been using the platform as a visual cue, they probably would have gone to it during the probe trial. A Panasonic CCTV camera was suspended over the center of the pool, its image monitored by a video tracking system [HVS VP 112 image analyzer; San Diego Instruments (San Diego, CA) software and interface; IBM PS/2 386 computer]. The water maze was surrounded by several distinct extra-maze visual cues (lights, posters, video equipment, etc.).

Enhancement of social recognition. Social recognition was evaluated in male CD-1 mice as described (Winslow and Camacho, 1995). Animals were singly housed for 1 to 2 wk prior to testing in polycarbonate cages with wood chip bedding and nesting material. An ovariectomized female (>40 days old) served as intruder stimulus animal and was housed in groups of five. All testing was conducted in the home cage of the male mouse in the vivarium. Tests were scheduled for the latter part of the light period.

Enhancement of step-down passive avoidance. The step-down assay described by Camacho et al. (1996) was used in our study. On day one, groups of 10 Sprague-Dawley rats were acclimated to the laboratory, experimenter and experimental chamber (31.5 × 24.5 × 27 cm), a plexiglas box equipped with white noise, soft white light, shock grid bars spaced 2-cm apart and a wooden platform (14 × 10.5 × 3.5 cm). On day 2, rats were again acclimated to the experimental area and then administered P10358 (0.3-2.5 mg/kg, p.o.). Thirty min after dosing, the rats were gently placed on the wooden platform, and their latencies (sec) to step-down were recorded. When all four paws touched the grid, a low level electric shock (0.20-22 mA) was delivered for 3 seconds by means of a Coulbourn Instruments (Lehigh Valley, PA) grid floor shocker. When the shock ceased, the rats were immediately removed from the experimental chamber and returned to their home cage. Fifteen min later this training procedure was repeated. On day 3, the same rats were again acclimated to the experimental area and then gently placed on the wooden platform. Their step-down latencies were measured (maximum 300 sec) and no shock was applied. Median latencies from the third day (24-hr retention trial) for each drug group were then compared to the median latency of the vehicle control group by the nonparametric Mann-Whitney U test.

Hemodynamic activity. Sprague-Dawley rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) then instrumented for arterial pressure recording as described (Vargas et al., 1993). Briefly, the surgical site was shaved and scrubbed with antiseptic. Polyvinyl catheters (ID/OD: 0.23/0.039 inch; Bolab, Lake Havasu City, AZ) prefilled with heparinized saline (100 U/ml) were placed in the left carotid artery and right jugular vein for blood pressure recording and intravenous drug injection, respectively. A 23-gauge wire stylet was inserted to plug catheter and the catheters were routed across the trachea and exteriorized at the nape of the neck. The incisions were closed with 3-O surgical silk and Nexaband Liquid adhesive (Veterinary Products Laboratories, Phoenix AZ). Animals were examined daily and allowed to recover for 2 days. On the experimental day, blood pressure was measured in the conscious freely moving rat by attaching the arterial catheter to a length of PE50 tubing attached to a swivel mounted 30 cm above the cage. The swivel attached to a Transpac Disposable transducer (Abbott Critical Care; Chicago, IL). Pulsatile arterial pressure and HR were measured simultaneously and recorded on a Beckman R611 Dynograph. MAP was collected for 1 hr before and 4 hr after oral administration of P10358.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

In Vitro Studies

Inhibition of AChE and BuChE activity in vitro. P10358 inhibited AChE and BuChE activity with potencies (IC₅₀) of 0.10 ± 0.02 µM (n = 3) and 0.08 ± 0.05 µM (n = 3), respectively. Regarding AChE inhibition, P10358 was 11 times less potent than HEP and 2.5 times more potent than THA (table 1). P10358 was an equipotent inhibitor of AChE and BuChE activity, whereas THA and HEP were more selective BuChE inhibitors (table 1).

Ex vivo inhibition of brain AChE by P10358. The dose-response relationships (fig. 2) of mouse forebrain AChE inhibition after p.o. and s.c. administration of P10358 shows that after 1 hr, the degrees of inhibition were not different (2-fold) when the two routes of administration were compared (ED₅₀ s.c.: 3.8 ± 1.2 mg/kg; p.o.: 6.7 ± 1.2 mg/kg). The inhibition of forebrain AChE activity observed at 7 mg/kg, p.o. recovered to normal levels within 4 hr (data not shown). Daily oral administration of 7 mg/kg for 5 days did not result in cumulative inhibition of mouse forebrain AChE activity (fig. 3). As with the mouse, dose-response analysis of rat striatal AChE inhibition induced by p.o. or s.c. P10358 administration also showed a 2-fold difference in the ED₅₀ values (s.c.: 3.6 ± 1.0 mg/kg; p.o.: 8.3 ± 1.2 mg/kg) (fig. 4A). Further characterization showed that the inhibition of striatal AChE activity produced by 10 mg/kg P10358 (p.o.) recovered to normal levels after 6 hr, but the dose of 20 mg/kg still showed significant enzyme inhibition after 24 hr (fig. 4B). Although HEP was 11 times more potent in vitro on AChE inhibition than was P10358 (table 1), HEP (ED₅₀ = 10.2 ± 1.1 mg/kg) and P10358 were equipotent when measured 1 hr after oral administration in the rat (fig. 4A).

View larger version (18K): [in this window] [in a new window]   Fig. 2.   Inhibition of mouse forebrain AChE activity after P10358 administration: dose-response. Groups (n = 3-6) of male Swiss-Webster mice (17-25 g) were dosed in equal volumes. Forebrain AChE activity was determined as described in "Materials and Methods." Each value (mean ± S.E.) represents the percent inhibition vs. the mean from control. All animals were fasted overnight (approximately 17 hr) prior to oral dosing. ED50 values (with 95% confidence limits) at the 1 hr time point were determined by log-probit analysis. The inhibition values for each point did not differ depending on route.

View larger version (75K): [in this window] [in a new window]   Fig. 3.   Inhibition of mouse forebrain AChE activity after P10358 administration: repetitive doses. Male Swiss-Webster mice were dosed once per day for 1, 2, 3, 4 or 5 days with P10358 (7 mg/kg, p.o.). One hr after the last dose, forebrains were removed and assayed for AChE activity as described in "Materials and Methods." Each value (mean ± S.E.M.) represents the result from a single group, expressed as the percent inhibition vs the mean from the control group. All animals were fasted overnight (approximately 17 hr) before oral dosing. All groups were significantly different from control (*P < .01, Newman Keuls). After five daily doses there was slightly less inhibition than after one dose (aP < .05). There were no significant differences between other groups.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Inhibition of rat striatal AChE activity after P10358 administration: dose (A) and time (B) responses. Groups (n = 3-14) of male Wistar rats were dosed and sacrificed at various time points. Striatal AChE activity was determined as described in "Materials and Methods." Each value (mean ± S.E.) represents the percent inhibition vs control. A, Rats were sacrificed one hr after dosing. HEP included as standard. All values were significantly different from control (*P < .01,grouped t test). B, Black bars represent values after 10 mg/kg, p.o., P10358. Hatched bars represent values after 20 mg/kg, p.o., P10358. *P < .05, grouped t test.

Effects of P10358 on brain dopaminergic function. The results in table 2 show the effect of P10358, HEP, MOCLO and TRANYLCYP on rat striatal DA, DOPAC and HVA. As with HEP, P10358 increased HVA levels, but had no significant effect on either DA or DOPAC. In contrast, the MAO inhibitors MOCLO and TRANYLCYP significantly decreased DOPAC and HVA levels. These findings indicate that P10358 is cholinomimetic through the specific inhibition of AChE, and does not affect MAO activity after oral administration.

Receptor profile. Based on conventional radioligand binding assays, P10358 had weak interactions with a variety of neurotransmitter receptors, ion channels and uptake carriers (table 3). The low affinity of P10358 for these recognition sites implies that this agent is a relatively specific inhibitor of cholinesterase activity.

In Vivo Studies

Primary cholinergic effects in rats and mice. Oral administration of P10358 produced tremor in both Swiss-Webster mice and Sprague-Dawley rats (fig. 5, A and B). Nonlinear regression analysis of the dose-response data showed that the oral ED₅₀ for tremor was very similar in mice (7.1 ± 1.1 mg/kg) and rats (4.8 ± 1.0 mg/kg). Lethality emerged at high doses in mice (40 mg/kg, p.o.) and rats (80 mg/kg, p.o.).

View larger version (15K): [in this window] [in a new window]   Fig. 5.   Potency curves for P10358-induced tremor (open squares) and lethality (closed squares) in mice (A) and rats (B). Various doses of P10358 were orally administered to groups of male Swiss-Webster mice and male Sprague Dawley rats, then evaluated for tremor. The number of animals (four/group) displaying tremor were normalized to 100%. Acute lethality was determined as the number of animals that died in the 24-hr period after dosing. Peak tremor was generally observed within the first hour.

Hypothermia in mice. P10358 produced significant hypothermic responses in Swiss-Webster mice (fig. 6). All animals had initial basal rectal temperatures of 37 to 37.5°C. Vehicle treated controls had temperatures which decreased slightly (1-2°C) over the course (6 hr) of the experiment. P10358 (5 to 20 mg/kg) dose-dependently induced pronounced and long lasting hypothermia after oral administration. The peak hypothermic effect and its duration varied as a function of dose.

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Hypothermic effect of oral P10358 in mice. Groups of Swiss-Webster mice (n = 8) were treated with 0, 5, 10 or 20 mg/kg and absolute rectal temperature (°C) was measured at regular intervals (min) after dosing. Basal temperature (B) was equivalent among the four treatment groups. *P < .05 vs. control time point (0 mg/kg). All values to the left of the asterisk were also statistically significant from control at same time point (two-way analysis of variance, time as a repeated measure, post hoc LSD test) .

Reversal of spatial memory deficits in Morris water maze. A repeated measures analysis of variance performed on mean percent of first trial latency indicated a significant treatment effect over the duration of water maze testing [F(3,38) = 9.5; P < .001]. Post hoc LSD tests revealed that the SCOP group took significantly longer to locate the submerged platform than the vehicle and the two SCOP + P10358 interaction groups (1.25 and 2.5 mg/kg; LSD: P < .001 for each comparison). In addition, both SCOP + P10358 groups did not significantly differ from the vehicle group (fig. 7).

View larger version (25K): [in this window] [in a new window]   Fig. 7.   Effect of P10358 on spatial memory in the Morris water maze. P10358 (1.25 or 2.5 mg/kg, p.o.) reversed scopolamine-induced water maze deficits. On each day of testing, male Spargue-Dawley rats were allowed four 90-sec trials with 15-min intertrial intervals. All drugs were administered 30 min before the initial trial on each day. Rats were injected with either vehicle, scop (1 mg/kg, i.p.) or a combination of scop + P10358. The mean percent of first trial swim distance is shown for each treatment group (n = 8) at each trial. *P < .001 vs. vehicle and the scop + P10358 groups (post hoc LSD analysis on combined trials for each day).

Enhancement of social recognition. Acute administration of P10358 (0.32-1.25 mg/kg, i.p.; [F(5,42) = 5.22, P < .01] significantly affected olfactory investigation (fig. 8). Moderate doses of P10358 had no measurable effect on initial levels of olfactory investigation but did significantly enhance the rate of decline resulting in overall decreases in time spent investigating. The two highest doses of P10358 (1.25 and 2.5 mg/kg) also significantly decreased walking (P = .04) and aggression (P = .17), possibily reflecting broader effects on arousal or motor ability. These disturbances were not detected at lower doses (0.312 and 0.625 mg/kg).

View larger version (30K): [in this window] [in a new window]   Fig. 8.   Effect of P10358 on social investigation and aggressive behavior in mice. Various doses of P10358 were evaluated on (A) olfactory investigation and (B) aggressive behavior in male CD-1 mice. Asterisks represent Newman-Keuls comparisons between the means averaged across trials for vehicle control and P10358 treatment groups (*P < .05).

Enhancement of step-down passive avoidance. In groups of rats, delivery of a low amperage shock during the first training trial produced a modest but significant increase in the latency to step off a platform on the second trial measured within subject (fig. 9, left). P10358 had no effect on latencies during the first training trial and increased step-down latencies at 0.625 mg/kg during the second training trial (fig. 9, middle). P10358 enhanced the 24-hr retention of the passive avoidance response (i.e., oral 0.625 and 1.25 mg/kg significantly enhanced the median step-down latencies compared to the vehicle control group) (fig. 9, right).

View larger version (16K): [in this window] [in a new window]   Fig. 9.   Effect of P10358 on step-down passive avoidance. Oral doses of P10358 were assessed on the median latency (horizontal bars) ± the 25th percentile (vertical bars) to step off a platform before training (left panel), 15 min after the first training trial (middle panel) and 24 hr after passive avoidance training and P10358 administration (right panel). Studies were performed with Sprague-Dawley rats. Asterisks represent significant differences between dose and vehicle control detected with the Mann-Whitney U one-tailed test (*P < .05).

Hemodynamic assessment. Oral administration of 5 mg/kg P10358 did not alter MAP or HR in freely moving rats (n = 8) from base-line levels of 113 ± 5 mm Hg and 351 ± 13 bpm, respectively. At this dose, overt cholinergic signs (e.g., tremor and salivation), were not observed. At 20 mg/kg P10358 (p.o.) significantly elevated MAP above pretreatment values. The pressor effect was evident 15 min (27 ± 11%) after dosing and was sustained up to 4 hr (19 ± 8%). In contrast, HR was unaffected throughout the observation period. This high dose also produced peripheral and central cholinergic signs (e.g., tremor, salivation, piloerection and fasiculations), in each animal during the 4-hr period.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

Despite the number of cholinergic therapies in preclinical and clinical development for AD therapy, several have already dropped out of development because of low efficacy or side effect liability (e.g., hepatotoxicity or blood dyscrasia). In addition, low oral activity and bioavailability may limit the clinical development of others. In clinical trials thus far, the presence or absence of the ApoE4 allele has not been used as a criteria for patient selection or screening, an important factor that may impact treatment outcome, especially with an AChEI (Poirier et al., 1995; Strittmatter and Roses, 1995; Pomara et al., 1995). Therefore, the maximum efficacy that can be achieved with an AChEI has probably not yet been realized in AD. Although the neurodegeneration found in AD is likely to be a multifactorial process involving neurochemical and genetic factors (Arai et al., 1992; Raskind et al., 1995; Schellenberg, 1995), cholinesterase inhibition remains the only therapeutic approach to demonstrate clinical efficacy in patients suffering from AD (Knapp et al., 1994). Therefore, a novel AChEI that is orally active, efficacious and tolerable remains a goal to further validate the cholinergic hypothesis of AD and to successfully treat AD patients. The pharmacological studies described indicate that P10358 has a preclinical profile that may be advantageous in the symptomatic treatment of AD. These studies demonstrate that the behavioral efficacy of P10358 in learning and memory paradigms is attributed specifically to brain AChE inhibition, because this compound has poor affinity for a wide variety of other neurotransmitter receptors.

Biochemical and physiological measures of central AChE inhibition. Biochemical studies showed that P10358 displayed high affinity for rat striatal AChE and was 2.5 times more potent that THA. P10358 was equipotent in its ability to antagonize AChE and BuChE activity in vitro, whereas THA and HEP are more potent toward BuChE. The significance of this finding in not immediately apparent but it does suggest that the two types of cholinesterase may bind P10358 in a common domain shared by both enzymes, whereas BuChE binds THA and HEP preferentially. The slightly higher potency of THA for BuChE identified in our study is consistent with a recent review of the biochemical activity of THA which reported that this agent is a more potent inhibitor of BuChE (Freeman and Dawson, 1991). Currently, it is uncertain whether P10358 behaves as a competitive or noncompetitive inhibitor of cholinesterase activity, but future studies are planned.

Behavioral efficacy of P10358. Research and drug discovery in AD is a formidable challenge because of the lack of definitive animal models that mimic the etiology of the disease process. Due to the high number of false positives that can be identified in learning and memory paradigms (Sarter et al., 1992a, 1992b), it is highly desirable to have biochemical and other mechanistic data to support behavioral observations. Given the clear cholinergic mechanism of P10358, this agent was tested in a number of behavioral assays of memory. As such, P10358 was evaluated for its ability to enhance a step-down passive avoidance response. Previous work indicates that drug-treated rats showing an increase in their median latency to step-down are considered to show an enhancement of learning (Cumin et al., 1982). After oral administration, relatively low doses of P10358 (0.63 and 1.25 mg/kg) enhanced learning in the step-down paradigm, a behavioral indication that this compound effectively stimulated central cholinergic function. In the same experimental paradigm, Camacho et al., (1996) demonstrated that other AChE inhibitors, such as galanthamine (1.25 and 2.5 mg/kg, i.p.), HEP (2.5 mg/kg, i.p.), velnacrine (2.5 mg/kg, i.p.) and THA (2.5 and 5 mg/kg, p.o.) also enhanced the 24-hr retention latency which is additional proof that brain acetylcholine is involved in forming the memory trace for this task. When the lowest active oral doses were compared, P10358 was 4- to eight-times more potent than THA and proved to be the most potent anticholinesterase evaluated in this procedure. The oral potency of P10358 indicates favorable bioavailability in comparison to the similar carbamate HEP, which was 10-times more potent toward inhibiting AChE in vitro than P10358, yet was four times less potent toward enhancing step-down passive avoidance.

Safety profile of P10358. Cardiovascular studies in the freely moving rat indicated that P10358 raised systemic arterial blood pressure at the high oral dose of 20 mg/kg, an effect that was not evident at a 4-fold lower dose. The hypertensive effect of this AChE inhibitor may be related to the activation of central cholinergic stimulation since systemically or centrally administered AChE inhibitors or muscarinic agonists can elevate arterial pressure in a number of species, including man (Brezenoff and Guiliano, 1982; Vargas and Ringdahl, 1990). The hypertensive effect of P10358 was only observed at 20 mg/kg, a dose that also produced profound (75%) and long-lasting inhibition of striatal AChE activity. Therefore, there is a clear separation (16-fold) between efficacious doses of P10358 that enhance central cholinergic function (i.e., step-down passive avoidance in rat) and doses that produce central motor and autonomic side effects (e.g., tremor and hypertension). Lethality was not observed until even higher doses were given. For example, acute mouse and rat toxicity, defined as the occurrence of lethality over a 24-hr observation period, was not observed until oral doses of 40 mg/kg in mice and 80 mg/kg in rats.