No abstract

A group of 3‐(hydroxyalkyl) 5‐alkyl 1,4‐dihydro‐2,6‐dimethyl‐4‐aryl‐3,5‐pyridinedicarboxylates (11–15) were prepared using a modified Hantzsch reaction, which were then elaborated to valproate (16–18), valerate (19, 20), and 1‐methyl‐1,4‐dihydropyridyl‐3‐carbonyloxy (25, 26) derivatives. Alternatively, the valproate derivative 3‐(2‐n‐propylpentanoyloxymethyl) 5‐isopropyl 1,4‐dihydro‐2,6‐dimethyl‐4‐(2,3‐dichlorophenyl)‐3,5‐pyridinedicarboxylate (34) was prepared by the reaction of isopropyl 1,4‐dihydro‐2,6‐dimethyl‐4‐(2,3‐dichlorophenyl)‐3,5‐pyridinedicarboxylate (30) with chloromethyl valproate (33). This class of lipophilic compounds possess partition coefficients (Kp) in the 149–452 range, relative to the reference drug nimodipine (Kp = 187). All compounds exhibited potent calcium channel antagonist (CCA) activity (IC50 = 10–7 to 10–10 M range), relative to the reference drug nimodipine (IC50 = 1.49 × 10–8 M). CCA structure–activity relationships showed the parent C‐3 2‐hydroxyethyl compounds were more potent than their valproate derivatives, but less active than their valerate derivatives. Compounds having a 1‐methyl‐1,4‐dihydropyridyl‐3‐carbonyloxy chemical delivery system (CDS) were approximately equiactive to the parent C‐3 2‐hydroxyethyl compounds. Anticonvulsant activity was determined in the maximal electroshock (MES) and subcutaneous metrazol (scMet) screens. 3‐(2‐Hydroxyethyl) 5‐isopropyl 1,4‐dihydro‐2,6‐dimethyl‐4‐(2,3‐dichlorophenyl)‐3,5‐pyridinedicarboxylate (12) provided modest protection in the MES and scMet screens. In the C‐3 valproate [CO2(CH2)nO2CCH(n‐Pr)2] group of compounds, those possessing an ethylene spacer (n = 2) provided protection in the MES screen, whereas those having a propylene spacer (n = 3) or methylene spacer (n = 1) were inactive. Related C‐3 valerate esters [CO2(CH2)2O2C‐n‐Bu] also provided protection in the MES screen, whereas those having a 1‐methyl‐1,4‐dihydropyridyl‐3‐carbonyloxy CDS moiety were inactive. Drug Dev. Res. 48:26–37, 1999. © 1999 Wiley‐Liss, Inc.

3-(2-Hydroxyethyl) 5-methyl 1,4-dihydro-2,6-dimethyl-4-(2,3-dichlorophenyl)-3,5-pyridinedi-carboxyla te (7) was prepared using a modified Hantzsch reaction, which was then elaborated to 3-[2-[[(1-methyl-1,4-dihydropyrid-3-yl)carbonyl]oxy]ethyl]5-methyl 1,4-dihydro-2,6-dimethyl-4-(2,3-dichlorophenyl)-3,5-pyridinedicarboxylat e [10, felodipine-chemical delivery system (CDS)]. The equipotent 3-(2-hydroxyethyl) 7 (IC50 = 3.04 x 10(-8) M) and felodipine-CDS (10, IC50 = 3.10 x 10(-8) M) were, respectively, 2- and 21-fold less potent calcium channel antagonists than the reference drugs nimodipine (IC50 = 1.49 x 10(-8) M) and felodipine (IC50 = 1.45 x 10(-9) M). Compounds 7, 10, nimodipine, and felodipine are highly lipophilic (Kp = 236, 366, 187, and 442, respectively). 3-(2-Hydroxyethyl) 7, felodipine-CDS (10), and felodipine provided protection against maximal electroshock-induced seizures in mice but were inactive in the subcutaneous metrazol anticonvulsant screen. In vitro incubation studies of felodipine with rat plasma and 20% brain homogenates showed felodipine was very stable in both biological media. Similar incubations of felodipine-CDS showed its rate of biotransformation followed psuedo-first-order kinetics with half-lives of 15.5 h in rat plasma and 1.3 h in 20% rat brain homgenates. In vivo biodistribution of felodipine and felodipine-CDS was studied. Uptake of felodipine in brain produced a peak brain concentration of 5 micrograms/g of brain tissue at 5 min, after which it rapidly egressed from brain resulting in undetectable levels at 60 min. Peak blood concentrations of 10 occurred at about 7 min followed by a rapid decline to a near undetectable concentration by 17 min. The pyridinium salt species 9, resulting from oxidation of 10, also reached peak concentrations at about 7 min but it slowly decreased to undetectable concentrations at 2 h. 3-(2-Hydroxyethyl) 7 remained at near undetectable concentrations throughout a 2 h time period. Localization of 10 in brain provided a peak concentration of 4.2 micrograms/g of brain tissue at 5 min and then decreased to negligible concentrations at 15 min. The concentration of oxidized pyridinium species 9 in brain remained high providing detectable concentrations up to 4 days. In contrast, the concentration of the 3-(2-hydroxyethyl) hydrolysis product 7 in brain remained at very low levels throughout the study. The slow hydrolysis rate of the pyridinium ester 9 to the 3-(2-hydroxyethyl) 7 and the rapid egression of felodipine-CDS from brain are believed to contribute to the moderate anticonvulsant activity exhibited hy the felodipine-CDS (10).

The cuprous iodide catalyzed reaction of 2-methyl-2-[1-(3-benzoyl-4-phenylpyridinium)]acetic acid methyl ester bromide (5), prepared by reaction of 3-benzoylpyridine (4) with racemic methyl 2-bromopropionate, with phenylmagnesium chloride at −23 °C afforded the 2-methyl-2-[1-(3-benzoyl-4-phenyl-1,4-dihydropyridyl)acetic acid methyl ester (6, 74%), which was a mixture of four diastereomers. Recrystallization of this diastereomeric mixture from diethyl ether afforded a solid product (6a-solid, 30%, 4R*,2R* and 4S*,2S*) and an oil product (6b-oil, 39%, 4R*,2S* and 4S*,2R*), each consisting of a mixture of two diastereomers that differ in relative configuration (R* or S*) at the 1,4-dihydropyridine C-4 position and the-CH(Me)CO2Me moiety. Addition of the Grignard reagent from either of the two diastereotopically different faces of the planar pyridinium salt (5) gives rise to two diastereomeric products 6a-solid and 6b-oil in which the C-4 phenyl substituent is pseudo-axial to the boat-shaped 1,4-dihydropyridine ring. Alkaline hydrolysis, or ammonolysis, of the acetic acid methyl ester (6) afforded the respective acetic acid (7), or acetamide (8), derivative. Keywords: 1,4-dihydropyridines, diastereomers.

A group of 3-alkyl 5-isopropyl 4-aryl-1,4-dihydro-2,6-dimethyl- 3,5-pyridinedicarboxylates 10-20 containing an amine, quaternary ammonium, aryl (heteroaryl)alkenyl, 4-(4-fluorophenyl)- piperazin-1-yl or methoxy moiety in the C-3 alkyl ester R-substituent in combination with a C-4 phenyl ring bearing a 2,3-Cl2, 3-NO2, 3-NMe2, 4-NMe2 or 3,4,5-(OMe)3 X-substituent were prepared using the Hantzsch 1,4-dihydropyridine reaction. In vitro calcium channel antagonist activity (CCA) was determined using a guinea pig ileum longitudinal smooth muscle assay. In the C-4 3-nitrophenyl series of compounds, the C-3 ester R-substituent was a determinant of CCA activity where the relative potency order was -CH2CH2CH=C-(2-methylphenyl)2 > or = -CH2CH2NMe2.HCI> -CH2CH2CH=C-(3-methyl-2-thienyl)2 > -CH2CH2+NMe3I -. The position and nature of the C-4 phenyl X-substituent, were also determinants of CCA activity where the relative activity order was 3-NMe2 > 4-NMe2 > 3,4.5-(OMe)3. Anticonvulsant activities were determined in mice using the subcutaneous metrazol (scMet) and maximal electroshock (MES) screens. The compounds investigated were generally not effective for protecting against scMet induced seizures, except for 10 (X = 2,3-Cl2, R = 2-[4-(4- fluorophenyl)piperazin-l-yl]ethyl] and 14a (X = 3-NMe2.HCl, R = CH2CH2OMe), which exhibited modest activity. Compound 11a (X = 3-NO2, R = -CH2CH2NMe2.HCl) was the most effective agent in the MES screen. All of the compounds investigated, except for 11b (X = 3-NO2, R = -CH2CH2+NMe3 I-, Kp = 0.15), are lipophilic with n-octanol/aqueous phosphate buffer (pH = 7.4) partition coefficients (Kp) in the 121-424 range relative to the reference drug nimodipine (Kp = 187). The structure-activity relationships acquired reinforce the concept that calcium is only one of several factors that are involved in seizure generation.