Abstract
Linopirdine (3,3-bis(4-pyridinylmethyl)-1-phenylindolin-2-one, DUP996) is an extensively studied representative of a class of cognition enhancing compounds that increase the evoked release of neurotransmitters. Recent studies suggest that these agents act through the blockade of specific K+ channels. We have recently identified more potent anthracenone analogs of linopirdine: 10,10-bis(4-pyridinylmethyl)-9(10H)-anthracenone (XE991) and 10,10-bis(2-fluoro-4-pyridinylmethyl)-9(10H)-anthracenone (DMP 543). Although linopirdine possesses an EC50 of 4.2 μM for enhancement of [3H]ACh release from rat brain slices, XE991 and DMP 543 have EC50s of 490 and 700 nM, respectively. In addition to greater in vitro potency relative to linopirdine, both compounds show greater in vivopotency and duration of action. Although 5 mg/kg (p.o.) linopirdine does not lead to statistically significant increases in hippocampal extracellular acetylcholine levels, 5 mg/kg (p.o.) XE991 leads to increases (maximal effect > 90% over baseline) which are sustained for 60 min. Moreover, DMP 543 at 1 mg/kg causes more than a 100% increase in acetylcholine levels with the effect lasting more than 3 hr. At doses relevant to their release-enhancing properties, the only overt symptom consistently observed was tremor, possible via a cholinergic mechanism. These results suggest that XE991 and DMP 543 may prove to be superior to linopirdine as Alzheimer’s disease therapeutics. In addition, these agents should be useful pharmacological tools for probing the importance of particular ion channels in the control of neurotransmitter release.
The progressive neuronal degeneration and neurotransmitter deficits that occur in AD are likely the underlying cause of the cognitive decline observed in patients having the disease. The premise that reversal of the neurotransmitter deficits, particularly cholinergic, would lead to improvements in function in AD patients has, at least in part, been supported by cognitive assessment scores of patients receiving cholinesterase (Farlow et al., 1992, Rogers, 1996) or muscarinic agonist (Bodick and Offen, 1996) therapies.
In addition to cholinesterase inhibition and muscarinic agonism, neurotransmitter release enhancement is another means of boosting neurotransmission. Compounds, such as the aminopyridines, are known to increase the release of neurotransmitters. However, these agents increase both basal (release in the absence of a stimulus) and stimulus-evoked release (Maciag et al., 1994). Compounds with specificity for enhancing stimulus-evoked release should be superior to the aminopyridines, because they would be more effective in increasing the signal to noise in neuronal circuits. Unlike agonists or cholinesterase inhibitors, these agents should enhance neurotransmission under the control of and in synchrony with natural brain activation.
A variety of structurally dissimilar agents have been identified that enhance the evoked release of neurotransmitters (Wilkerson et al., 1993; Earl et al., 1992; Wilkerson et al., 1996; Chorvat et al., 1995). Linopirdine (3,3-bis(4-pyridinylmethyl)-1-phenylindolin-2-one, DUP 996) is an extensively-studied representative of this class of compounds (Nickolson et al., 1990; for reviews see Zaczek and Saydoff, 1993 and Aiken et al., 1996). The release-enhancing properties of this agent have been suggested to underlie its ability to increase the performance of laboratory animals in a number of learning and memory paradigms (Cook et al., 1990; Brioni et al., 1993). Because linopirdine positively affects animal behavioral performance and enhances the release of neurotransmitters known to be decreased in AD, its potential as a therapy for the disease was tested. However, the results of its clinical trials in AD patients were equivocal, and may have been a reflection of suboptimal pharmacokinetic and distribution properties of the drug. Thus, the therapeutic potential of this type of neurotransmitter release enhancer for AD might not have been adequately tested with this agent.
Apart from possible therapeutic uses, linopirdine and analogs have potential value in the study of neurotransmission. Neurotransmitter release is mediated and controlled through the opening of a number of ligand- and voltage-activated channels for various ionic species. Compounds that are found to interact with these channels have been important tools for furthering our understanding of events underlying neuronal signaling. Recent studies have suggested that linopirdine enhances ACh release through an interaction at tetraethylammonium sensitive K+ channels (Maciag et al., 1994). In addition, linopirdine has been shown to block a voltage-dependent K+ channel current identified as the M-current, IM (Aiken et al., 1995). Thus, compounds such as linopirdine should prove increasingly valuable in evaluating the contribution of IM or related channels in the control of neurotransmitter release.
We have recently identified additional functional analogs of linopirdine based upon an anthracenone backbone (see fig. 1). Two of these compounds, XE991 and DMP 543, are superior to linopirdine both in vitro and in vivo. Therefore these agents may prove to be superior to linopirdine as AD therapeutics and as tools for preclinical studies on the involvement of specific K+ channels in the control of neurotransmitter release. We report on the release enhancing properties of XE991 and DMP 543 in vitro and in vivo as well as overt symptomatology at efficacious doses and compare these activities to those of linopirdine.
Methods
Neurotransmitter Release Enhancement In Vitro
Our 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.
Tissue preparation and release assays were performed as described previously (Nickolson et al., 1990). Male Wistar rats (Charles River, Wilmington, DE) were decapitated and specific brain areas were immediately dissected. Brain areas dissected were chopped into 0.25 × 0.25 cm2 squares using a McIlwain tissue chopper. Approximately 100 mg of the slices were transferred to 10 ml of Krebs-Ringer solution, made up of 116 mM NaCl, 3 mM KCl, 1.3 mM CaCl2, 1.2 mM MgCl2, 1.2 mM KH2PO4, 1.2 mM Na2SO4, 25 mM NaHCO3 and 11 mM glucose. The solution also contained either 10 μCi of DA (30 Ci/mmol), 10 μCi [3H]d-aspartic acid (78 Ci/mmol) or 10 nmol choline chloride containing 10 μCi of [3H]choline chloride (80 Ci/mmol) for DA, d-aspartic acid or ACh release, respectively; all radiochemicals were purchased from Du Pont NEN, Boston, MA. [3H]d-Aspartic acid was used as a metabolically inert functional analog the neurotransmitterl-glutamic acid. The preparation was allowed to incubate for 30 min at 37°C under an atmosphere of 95% O2: 5% CO2. For DA preloading, the monoamine oxidase inhibitor nialamide (0.01 mM) was added to retard oxidation during the incubation. After the incubation period, the slices were washed three times with fresh Krebs-Ringer buffer and an aliquot of the slices (approximately 10 mg) was transferred to perfusion chambers of a Brandel SF-18 superfusion apparatus. The slices were superfused with oxygenated Krebs-Ringer solution at a rate of 0.25 ml/min for 10 min before fractions of the effluent were collected. For [3H]ACh release, 10 μM hemicholinium-3 was added to the superfusion medium to inhibit reuptake of [3H]choline during the release assay. After the 10-min washout period, fractions were collected in 4-min intervals (1.0-ml fractions) and were collected directly into scintillation vials; at the end of experiments, the chambers were emptied into scintillation vials and residual radioactivity was extracted from the slices in 1.0 ml of 0.1 N HCl. Scintillation cocktail was subsequently added to the vials that were then assessed for radioactivity in a Packard 1900 TR scintillation counter.
A total of 10 fractions were collected from each chamber during an experiment. Stimulated release was elicited by raising the KCl concentration to 25 mM, with a reduction in NaCl to balance osmolarity, for a period of 4 min (i.e., the period of one fraction) immediately before fraction 4 (S1) and fraction 8 (S2). Drugs, if present, were introduced immediately after the first stimulation period and remained in the superfusion medium through fraction 8. Fractional release was calculated by dividing the radioactivity (dpm) found in each fraction by the total radioactivity in the tissue at the start of the given collection period and is expressed in percent. S1 is defined as release found in fraction 4 plus fraction 5 minus release found in fraction 3 plus fraction 6; S2, the stimulation period during which compounds were tested, is defined as release found in fraction 8 plus fraction 9 minus release found in fraction 7 plus fraction 10.
EC50 values from dose response curves were calculated as follows. For each curve, percent peak response for each drug concentration was calculated. Peak response was defined as the highest value for S2/S1 for each individual curve. EC50 values were determined from a log-logit regression analysis of the percent peak response data.
ACh Release In Vivo
Surgery.
Male CD rats (Charles River) were housed individually on alpha-dry bedding, with food and water ad libitum on a 12-hr light/dark cycle. Rats weighing 200 to 250 g were anaesthetized with a mixture of Ketamine (110 mg/kg)/Rompin (10 mg/kg) administered i.m. and placed in a Kopf stereotaxic frame. Using aseptic surgical techniques, an incision was made in the skin to expose the skull. Two small pilot holes were drilled in the skull, anterior to bregma, to hold anchor screws. A cannula guide, to accommodate the microdialysis probe, was stereotaxically implanted over the hippocampus (A/P at the midpoint between lambda and bregma, M/L-2.00 mm, D/V-1.2 mm) and cemented in place with acrylic dental cement. A dummy cannula was inserted into the guide to insure its patency. The incision was sutured, the area bathed with Penicillin G and the animal was returned to its cage and allowed to recover for a minimum of 72 hr before testing.
Microdialysis.
Rats were fasted overnight before testing. On test day, rats were weighed and placed in individual clear Lucite boxes (12′w × 12′d × 15′h). An animal being tested was fitted with a harness attached to a dual channel swivel (for delivery of perfusate/collection of dialysate fractions) and supported by a counter-balanced lever arm. The dummy cannula was removed from the cannula guide and a microdialysis probe (CMA-12, BAS) was inserted in its place. The probe was perfused at a rate of 2 μl/min with artificial CSF (Harvard Bioscience, South Natick, MA) containing 100 μM physostigmine sulfate (a cholinesterase inhibitor). After a 2-hr acclimation period, four dialysate fractions were collected to determine baseline ACh levels. Fractions were collected over a 20-min intervals, at a rate of 2 μl/min for a total of 40 μl of dialysate per fraction. After baseline fractions were collected, drugs or vehicle control were administered orally (0.01 ml/g body weight) to the rats. Dialysate fractions were collected and analyzed (on-line high-performance liquid chromatography assay) to determine drug effect on extracellular ACh levels for 3 hr after drug administration. Animals were used once and after testing were euthanized and probe placement within the hippocampus was verified by visual inspection of the probe track.
CMA-12 (4 mm membrane) microdialysis probes (BAS) were used in our studies. Each probe was tested in a known concentration of ACh 1 μM to insure adequate recovery of probe immediately before implantation in an animal.
High-performance liquid chromatography analysis.
Resolution and enzymatic conversion of ACh to H2O2 was accomplished using Hamilton PRP-1 analytical column and a CMA immobilized enzyme reaction reactor column (CMA/BAS). Flow rate was 1 ml/min. The mobile phase consisted of 0.2 M Na2HPO4, 0.1 mM EDTA, 0.5 mM SOS, 0.9 mM tetramethylammonium chloride, with the pH adjusted to 8.0 with phosphoric acid. Fifty μl of Kathon CG (ESA) was added to each liter of mobile phase to retard bacterial growth. H2O2 reflective of extracellular ACh was measured by electrochemical detection using Waters 460 detector with a platinum electrode at a potential of +0.5 mV.
Data Analysis
H2O2 peak heights were used to determine changes in ACh levels. The ACh levels from three dialysis fractions before compound administration were averaged to determine the baseline ACh level for each rat. The percent change from baseline was then calculated for each fraction. Statistical significance was determined by repeated measures analysis of variance followed by means comparison between the individual fractions and baseline values as defined above.
Linopirdine Brain/Plasma Concentrations
Linopirdine was synthesized by the Du Pont Merck Pharmaceutical Company (Wilmington, DE) and radiolabeled (3H) by Du Pont New England Nuclear Research Products (Boston, MA) to a specific activity of 89.5 Ci/mmol.
Male Sprague-Dawley Rats (Charles River Breeding Laboratories) weighing 200 to 250 g received doses of linopirdine (1 and 10 mg/kg, p.o.) with trace amounts of [3H] linopirdine (30–40 mCi) and euthanized 15 and 30 min post-dose. Plasma samples were collected using an anticoagulant mixture of EDTA and aprotinin. Hippocampi were dissected, weighed and solubilized in Solvable (Du Pont, NEN) overnight at room temperature. Radioactivity was then determined by liquid scintillation counting on a Packard Tri-carb 2500, scintillation counter (Packard Corp., Meriden, CT). The concentrations of linopirdine in rat brain and plasma were determined based on the amount of radioactivity detected per sample, dose of linopirdine administered and the weight or volume of tissue samples.
Overt Behavioral and Physiological Signs
Male Sprague-Dawley rats (Cesarean-derived, 90–150 g), fasted overnight, were orally gavaged with either DMP 543 (0.625, 1.25 mg/kg), XE991 (6.25, 12.5 mg/kg) or linopirdine (12.5, 25 mg/kg) and observed for overt signs and the absence of basic reflexes. Readings were made at 0.25, 0.5, 1, 2, 4 and 6 hr after compound administration. XE991 was dissolved in a small volume of 1 N HCl and diluted in saline to achieve a concentration sufficient for a dose of 12.5 mg/kg when administered at 0.1 ml/10 g of body weight (10 ml/kg). DMP 543 and linopirdine were suspended in 0.25% methyl cellulose and placed on a beadmill either for 1 hr (linopirdine) or overnight (DMP 543). Signs and reflexes tested included: righting reflex, corneal reflex, placing reflex, myotatic reflex, mortality, exploratory loss, alterations in muscle tone, ptosis, ataxia, loss of lift and grip, catatonia, salivation, lacrimation, diarrhea, excitement, hypersensitivity to handling, tremor and convulsion.
Results
Neurotransmitter release enhancement from brain slices in vitro.
One of the characteristics of the effects of linopirdine on neurotransmitter release from slices in vitro is the dependence on a depolarizing stimulus (Nickolson et al., 1990). We examined if this property is also observed for XE991 and DMP 543. As shown in figure 2, at a concentration that led to maximal release enhancing response (3 μM, see fig. 3) neither XE991 nor DMP 543 increased the efflux of tritium before a depolarizing concentration of K+ was introduced into the superfusion medium. When K+ was increased, a robust, statistically significant enhancement of release was observed with both compounds. Thus, as has been previously shown for linopirdine, a depolarizing stimulus is required in order for XE991 and DMP 543 to influence release.
In figure 3, the dose response curves for linopirdine, XE991 and DMP 543 enhancement of [3H]ACh release from rat hippocampal slices in vitro are shown. Half maximal responses were observed at 0.49 ± 0.08 and 0.70 ± 0.06 μM for XE991 and DMP 543, respectively. These data indicate that the two compounds are between 5- and 10-fold more potent than linopirdine that had an EC50 value of 4.2 ± 0.8 μM.
The enhancement of the release of multiple neurotransmitters reported for linopirdine was also observed for both XE991 and DMP 543. Figure4 shows dose response curves for the enhancement of [3H]DA and [3H]d-aspartic acid release from striatal and hippocampal slices, respectively. The potencies of the compounds in the enhancement of the release of tritium from slices preloaded with [3H]DA and [3H]d-aspartic acid were similar. The EC50values for [3H]DA release enhancement were 0.20 ± 0.02 and 0.25 ± 0.08 μM, respectively, for XE991 and DMP 543. The EC50 values for [3H]d-aspartic acid release enhancement were 0.25 ± 0.08 and 0.22 ± 0.02 μM for XE991 and DMP 543, respectively.
Acetylcholine release in vivo.
To examine if linopirdine enhances ACh release in vivo, a microdialysis procedure was used in freely moving animals. Microdialysis probes were placed into the dorsal hippocampus for sampling of the extracellular fluid. Figure5 shows the linopirdine dose response for the elevation of extracellular ACh. The stress of oral administration of vehicle led to a small, but statistically significant, increase in extracellular ACh. This increase was observed only in the sample taken immediately after oral administration. Linopirdine, at 5 mg/kg p.o. did not produce a significant increase in ACh levels. Higher doses of linopirdine led to increasing peak levels of extracellular ACh as well as increasing duration of effect. At 10 mg/kg, linopirdine gave rise to statistically significant elevations in ACh levels that were maintained for 60 min post-dosing. Peak increases of approximately 60% above baseline were observed at this dose. A dose of 20 mg/kg linopirdine led to statistically significant increases that were maintained for 100 min and the maximum response exceeded 80% above baseline.
To determine if the linopirdine-induced increases in hippocampal extracellular ACh occur at a drug concentration range consistent with its in vitro potency in the slice ACh release assay, the concentration of the compound was estimated in plasma and brain 15 and 30 min after administration of [3H]linopirdine at doses of 1 mg/kg and 10 mg/kg p.o. It should be noted that total tritiated species were measured, therefore the data are reflective of both linopirdine and its metabolites. The data are shown in table 1. The hippocampus/plasma ratio of radioactivity ranged between 13 and 31%. At 1 mg/kg, a dose not expected to give rise to increases in extracellular ACh levels, 0.043 and 0.075 μM of tritium was observed in the hippocampus at 15 and 30 min, respectively; 10 mg/kg, a dose that gave rise to statistically significant increases in ACh levels (see fig. 4), led to 2.5 and 1.9 μM of total radiolabeled species, at 15 and 30 min, respectively. These data, although representing a rough estimate of linopirdine concentrations, indicate that the range of linopirdine concentrations in the hippocampus at a time when extracellular levels of ACh are elevated is consistent with that leading to ACh release enhancement from hippocampal slices in vitro.
In the ACh microdialysis paradigm, XE991 was more potent than linopirdine giving rise to greater maximal responses and longer durations of action at equivalent doses. As seen in figure 6, 5 mg/kg XE991 increased ACh levels to 100% over baseline; the overall duration of effect at this dose was 60 min. At 10 mg/kg XE991 ACh levels were increased to approximately 140% over baseline and statistically significant increases were observed over 100 min.
Of the compounds tested in our study, DMP 543 was the most potent and possessed the greatest duration of action in elevating extracellular ACh levels. At doses as low as 0.5 mg/kg p.o., statistically significant 40% increases were observed. Increases of 100% over baseline were observed after 1.0 mg/kg DMP 543 and these increases were maintained for the duration of the assay that lasted 180 min.
Overt signs.
To ascertain if observable side effects of linopirdine, XE991 and DMP 543 accompany their in vivo ACh release enhancing effects, overt symptomatology was examined at doses shown to be effective in the ACh microdialysis paradigm (see table 2). Tremors were observed in 4 of 20 rats and diarrhea was observed in 5 of 20 rats administered 0.625 mg/kg, p.o., DMP 543. No other signs were observed at this dose. When male rats were given 1.25 mg/kg, p.o., only tremors (15 of 20 rats) were seen. Tremors were present in 14 of 15 male rats at a dose of 6.25 mg/kg, p.o., XE991. Loss of lift occurred in 2 of 15 rats at the same dose. At the higher dose of XE991 all rats exhibited tremors, 6 of 10 rats were unable to lift their hindquarters when dangling from a wire and 3 of 10 had curled toes. Linopirdine-treated rats also showed tremors (8 of 10 rats at 12.5 and 25 mg/kg, p.o.) as well as diarrhea (3/10 at 25 mg/kg, p.o.) and loss of lift (2/10 at 25 mg/kg, p.o.). Thus, all of these compounds produced similar signs with tremors being the most consistently observed sign at the doses tested.
Discussion
Our study was aimed at comparing the neurotransmitter release enhancing agent linopirdine to two more recently discovered compounds, XE991 and DMP 543. We showed that, as with linopirdine, these agents enhance ACh release only in the presence of a depolarizing stimulus. XE991 and DMP 543 have EC50 values for in vitroACh release enhancement of 0.49 and 0.70 μM, respectively, indicating that they are 5 to 10 times more potent than linopirdine that has an EC50 value of 4.2 μM. This difference points to the superior potency of anthracenones, such as XE991 and DMP 543, over oxindoles, such as linopirdine.
We demonstrated that the release enhancing effects of XE991 and DMP 543 are not limited to ACh. The release of both DA and d-aspartic acid (used as a metabolically inert analog of glutamic acid) were enhanced by these compounds. Similar potencies (EC50 values ∼ 0.2 μM) were observed for both compounds. XE991 and DMP 543 maintain the 5- to 10-fold potency differential over linopirdine (for linopirdine potency see Nickolson et al., 1990) in enhancing the release of transmitters other than ACh.
Previous work using an in vivo cortical cup method indicated that linopirdine increases ACh release from rat cerebral cortical surface after s.c. doses under 1 mg/kg (Nickolson et al., 1990). However, subsequent studies demonstrated that 10 mg/kg linopirdine s.c. (Smith et al., 1993) or i.p. (Marynowskiet al., 1993) was required to reach brain concentrations of the compound (∼1 μM) consistent with those necessary to enhance release in vitro. Elevations in extracellular levels of ACh in the hippocampus as assessed by in vivo microdialysis were also observed in these studies after a dose of 10 mg/kg linopirdine. However, a dose of 1 mg/kg was not effective (Marynowski et al., 1993).
In our study, we demonstrated that linopirdine gives rise to a dose-dependent elevation in hippocampal ACh. At 5 mg/kg linopirdine p.o., no statistically significant increase in ACh levels were observed. However, progressive increases in both peak levels of ACh and duration of statistically significant elevations were observed at doses of 10 and 20 mg/kg p.o.
We estimated that the concentration of linopirdine reaching the brain was equivalent to that capable of enhancing release in vitrowhen the dose of the compound was 10 mg/kg. This, together with the observation that 10 mg/kg and not 5 mg/kg linopirdine gave rise to statistically significant increases in brain ACh levels, is consistent with linopirdine acting at the same site to enhance ACh release from slices in vitro and to increase extracellular ACh levelsin vivo.
In identifying second generation functional analogs of linopirdine, two issues were focused upon. The first was greater potency and, as indicated above, we found that compounds having anthracenone core structures, such as XE991, were 5- to 10-fold more potent than linopirdine at enhancing the release of ACh in vitro. We also showed that, although linopirdine at 5 mg/kg did not significantly increase extracellular levels of ACh in the hippocampus in vivo, XE991 at 5 mg/kg significantly elevated ACh levels approximately 100% over baseline.
Linopirdine has a relatively short half-life (Pieniaszek et al., 1995). The metabolic lability of the compound in humans was shown to be the result of rapid N-oxidation of the pyridinylmethyl pendant substituents (Rakestraw, 1990) producing molecules devoid of release-enhancing properties. Substitution of electron withdrawing groups, such as fluorine, alpha to the nitrogen in the pyridinylmethyl side-chain decreases basicity, thereby increasing the ionization potential of the nitrogen and making N-oxidation at that position less likely (Ramsey and Walker, 1974). This structural modification to XE991 produced a compound, DMP 543, with resistance to N-oxidation. Incubation of DMP 543 and XE991 with rat liver microsomes has demonstrated that the fluoro compound (i.e., DMP 543) shows a greatly decreased rate of NADPH-dependent metabolism (Richards LE, DuPont Merck Pharmaceutical Company, personal communication).
We showed that the superior potency of anthracenones combined with the fluorine substitution to the pendant groups, both of which are found in DMP 543, gives rise to a molecule with far greater in vivopotency over linopirdine. DMP 543 elevates extracellular ACh at dose of 0.5 mg/kg, and at 1.0 mg/kg ACh levels were elevated over baseline for the duration of our experimental procedure (180 min). These data indicate that DMP 543 is 20-times more potent than linopirdine and the duration of a 1.0 mg/kg dose of DMP 543 is more than 2-fold greater than the 10 mg/kg linopirdine dose.
When linopirdine, DMP 543 and XE991 were evaluated in rats, all elicited similar overt signs although the potencies of the compounds were different. Tremor was the only consistently observed overt sign with all the compounds eliciting this response at both of the doses tested. Given that the compounds enhance the release of ACh in the central nervous system and that tremors are evoked by many centrally acting cholinergic agonists it is likely that the tremors seen in this study are the result of central cholinergic stimulation. Thus, overall, the observable effects of the compounds are in keeping with their ACh release enhancing properties.
In summary, we have found that linopirdine enhances the release of AChin vivo at concentrations consistent with those necessary to elicit release enhancement in vitro. We have also identified two compounds XE991 and DMP 543 that are superior to linopirdine at enhancing release both in vitro and in vivo. These agents may prove to be superior to linopirdine as AD therapeutics. Current understanding of the mechanism of action of linopirdine and its analogs points to the blockade of specific K+ channels underlying the release enhancing effects of the compounds. Studies of superior pharmacological tools such as XE991 and DMP 543 will further our understanding of the importance of particular ion channels in the control of neurotransmitter release.
Footnotes
-
Send reprint requests to: Dr. Robert Zaczek, The DuPont Merck Research Laboratories, Experimental Station E400/4458, Wilmington, DE 19880-0400.
- Abbreviations:
- XE991
- 10,10-bis(4-pyridinylmethyl)-9(10H)-anthracenone
- DMP543
- 10,10-bis(2-fluoro-4-pyridinylmethyl)-9(10H)-anthracenone
- ACh
- acetylcholine
- DA
- dopamine
- IM
- M potassium current
- AD
- Alzheimer’s disease
- Received August 25, 1997.
- Accepted January 9, 1998.
- The American Society for Pharmacology and Experimental Therapeutics