Yves L. Janin

Methyl Orthocarboxylates as Methylating Agents of Heterocycles

Methylation reactions occurring between trimethyl orthocarboxylates or N,N-dimethylcarboxamide dimethyl acetals and various hydroxylated heterocycles, involving a lactam-lactim tautomeric equilibrium, were investigated as an alternative to classic methylation methods. The corresponding O-methylated or N-methylated compounds were isolated in a number of instances and the reaction’s regioselectivity was shown to sometimes follow and sometimes differ from the corresponding outcome using standard methylation methods. In the course of this work, previously unreported effects were noticed. In one case the use of toluene as a reaction solvent led to much more N-methylated material. In other instances, the influence of the reagent’s steric bulk and/or stability (orthoformate vs. orthoacetate or N,N-dimethylformamide dimethyl acetal vs. its acetamide homologue) was also noticed. An unwanted formylation reaction could sometimes be avoided and, less often, an increase of O-methylated material was observed. The previously unreported 1-dimethoxymethylpyridin-2(1H)-one was characterized and its acid-catalyzed rearrangement into, mostly, 1-methylpyridin-2(1H)-one was studied. The new techniques described here (methanol trapping with 4 A molecular sieves and Lewis acid-catalyzed reaction) greatly increase the potential of trimethyl orthocarboxylates. These reagents can be considered as possible alternatives to the dimethyl formamide-producing N,N-dimethylformamide dimethyl acetal and may sometimes be attractive options compared to the usual carcinogenic and salt-producing methylating agents. (© Wiley-VCH Verlag GmbH, 69451 Weinheim, Germany, 2002)

2-Amino-3-(3-hydroxy-1,2,5-thiadiazol-4-yl)propionic acid: resolution, absolute stereochemistry and enantiopharmacology at glutamate receptors

In order to identify new subtype-selective (S)-glutamate (Glu) receptor ligands we have synthesized (RS)-2-amino-3-(3-hydroxy-1,2,5-thiadiazol-4-yl)propionic acid [(RS)-TDPA]. Resolution of (RS)-TDPA by chiral chromatography was performed using a Crownpac CR(+) column affording (R)- and (S)-TDPA of high enantiomeric purity (enantiomeric excess=99.9%). An X-ray crystallographic analysis revealed that the early eluting enantiomer has R-configuration. Both enantiomers showed high affinity as well as high agonist activity at (RS)-2-amino-3-(3-hydroxy-5-methylisoxazol-4-yl)propionic acid (AMPA) receptors, determined using a [(3)H]AMPA binding assay and an electrophysiological model, respectively. The affinities and agonist activities obtained for (R)-TDPA (IC(50)=0.265 microM and EC(50)=6.6 microM, respectively) and (S)-TDPA (IC(50)=0.065 microM and EC(50)=20 microM, respectively) revealed a remarkably low AMPA receptor stereoselectivity, (S)-TDPA showing the highest affinity and (R)-TDPA the most potent agonist activity. In addition, (S)-TDPA was shown to interact with synaptosomal Glu uptake sites displacing [(3)H](R)-aspartic acid (IC(50 ) approximately 390 microM). An enantiospecific and subtype-selective agonist activity was observed for (S)-TDPA at group I metabotropic Glu (mGlu) receptors (EC(50)=13 microM at mGlu(5) and EC(50)=95 microM at mGlu(1)).

3-Pyrazolone analogues of the 3-isoxazolol metabotropic excitatory amino acid receptor agonist homo-AMPA. Synthesis and pharmacological testing

We have previously shown that the higher homologue of (S)-glutamic acid [(S)-Glu], (S)-alpha-aminoadipic acid [(S)-alpha-AA] is selectively recognized by the mGlu(2) and mGlu(6) subtypes of the family of metabotropic glutamic acid (mGlu) receptors. Furthermore, a number of analogues of (S)-alpha-AA, in which the terminal carboxyl group has been replaced by various bioisosteric groups, such as phosphonic acid or 3-isoxazolol groups, have been shown to interact selectively with different subtypes of mGlu receptors. In this paper we report the synthesis of the 3-pyrazolone bioisosteres of alpha-AA, compounds (RS)-2-amino-4-(1,2-dihydro-5-methyl-3-oxo-3H-pyrazol-4-yl)butyric acid (1) and (RS)-2-amino-4-(1,2-dihydro-1,5-dimethyl-3-oxo-3H-pyrazol-4-yl)butyric acid (2). At a number of steps in the reaction sequences used, the reactions took unexpected courses and provided products which could not be transformed into the target compounds, and attempts to synthesize the 2,5-dimethyl isomer of 2, compound 3, failed. An X-ray crystallographic analysis of the intermediate 1,2-dihydro-4-(2-hydroxyethyl)-2,5-dimethyl-3H-pyrazol-3-one (5b) confirmed the expected regioselectivity of the reaction between methylhydrazine and alpha-acetylbutyrolactone (4). Neither 1 nor 2 showed significant effects at the different types of ionotropic glutamic acid receptors or at mGlu(1a) (group I), mGlu(2) (group II), and mGlu(4a) and mGlu(6) (group III) receptors, representing the three indicated groups of mGlu receptors.

A formal new access to the benzo[c]phenanthridine alkaloids, synthesis of nitidine and o-methyl fagaronine analogues.

Abstract Previously unreported 2-aryl-1-naphthylamines were obtained in good yields from 2-aryl-1-tetraloneoximes by using the Semmler-Wolf reaction but omitting the acetic anhydride usually present in the reaction mixture. From these amines, through the thermal cyclization of their corresponding ethyl carbamates, a new access to the benzo[c]phenanthridin-6(5H)-ones was found. Preparation of water-soluble Nitidine and O-Methyl Fagaronine analogues bearing an alkylamino side chain on the C-6 position was achieved from these compounds.

UNAMBIGUOUS SYNTHESIS OF 1-METHYL-3-HYDROXYPYRAZOLES

Abstract 2,3-Dihydropyrazolo[3,2-b]oxazoles were used as intermediates in a new method for preparation of N1-methyl-3-hydroxypyrazoles. Synthesis of this bicyclic system was achieved either by alkylation of 3-hydroxypyrazole with 1,2-dibromoethane or, with better yields, by cyclization of 1-tosyl-2-(2-hydroxyethyl)pyrazol-3-ones via a nitrogen to oxygen transfer of the tosyl group. Alkylation with methyl trifluoromethanesulfonate followed by dihydrooxazole ring-opening with sodium iodide, led to the 1-methyl-2-(2-iodoethyl)pyrazoles. Removal of the iodoethyl chain on N2 to give the target 3-hydroxypyrazoles was achieved either via a cyanation and then a decyanoethylation reaction or via an elimination of hydrogen iodide, followed by an iodine-based oxidation of the resulting vinylic derivative. Using the latter method, 1-methyl-3-hydroxypyrazoles were obtained in 58–73% yields from the corresponding 2,3-dihydropyrazolo[3,2-b]oxazoles.

On the methylation of 3-cyano-6-hydroxypyridine-2(1H)-ones

In order to prepare new heterocyclic derivatives as building block for compounds with potential biological activities, we were led to study the O-methylation of 3-cyano-6-hydroxypyridine-2(1H)-ones such as 1. This enabled us to develop a new two steps method to prepare the corresponding 2,6-dimethoxy derivative 5 in 80 % yield. It consisted first in heating 1 in trimethylorthoformate, with or without an acidic catalyst, which gave a mixture of the two mono-methoxy isomers, then a classical methylation of the second hydroxy moiety led almost exclusively to 5. In this paper we present this “methylation” method and various unexpected results recorded when we attempted to extend it to related derivatives or to other heterocycle containing lactim-lactam functions. An intramolecular transetherification mechanism requiring the simultaneous transfer of a hydrogen and a methyl is suggested.

SYNTHESIS OF CHLORAMPHENICOL AND MANDELONITRILE GALACTOSE-CONTAINING PRODRUGS

The synthesis of O 1-β-d-galactopyranosylchloramphenicol and O 1-β-d-galactopyranosylmandelonitrile as prodrugs potentially substrates of β-galactosidase, are reported. Preparation of O 1-(2,3,4,6-tetra-O-acetyl-β-d-galactopyranosyl) chloramphenicol from unprotected chloramphenicol was successful using β-d-galactopyranose pentaacetate and boron trifluoride diethyl etherate in acetonitrile. However, the β-galactosylated diastereoisomers of racemic mandelonitrile had to be made via O 1-(2,3,4,6-tetra-O-acetyl-β-d-galactopyranosyl)mandelamide in dichloromethane prior to dehydration to obtain the nitrile moiety. Indeed, galactosylation trials starting directly from mandelonitrile in acetonitrile led to the O 1-(2,3,4,6-tetra-O-acetyl-α-D-galactopyranosyl)mandelonitrile diastereoisomers. From a methodological point of view, this work extends the use of the galactosylation method to new hydroxyl bearing compounds. It also points out that the solvent used (acetonitrile or dichloromethane) and the purity of boron trifluoride diethyl etherate can be crucial factors in the use of this method as an eventual alternative to heavy metal-based Lewis acids usually employed in glycosylation reactions.

Synthetic approaches to 1-(2-chlorophenyl)isoquinoline-3-carboxylic acid

In connection with our research of new antitumor compounds, previously undescribed approaches to the 1-(2-chlorophenyl)isoquinoline-3-carboxylic acid 9 are reported here. Two related accesses from phenylethylamine or amphetamine were investigated and were found to be successful. A more robust synthesis, using Suzukis cross-coupling between 2-chlorophenylboronic acid 15 and the previously unreported methyl-1-bromoisoquinoline-3-carboxylate 14 was also developed. This synthetic route provides the ground for a combinatorial approach to the core structure of new potential peripheral benzodiazepine receptor ligands.