NO20151209L - Methods of Preparation of Compounds - Google Patents

Abstract

Methods for preparing (S) - (+) - 3-aminomethyl-5-methyl-hexanoic acid and structurally related compounds are described.

Description

FIELD OF THE INVENTION

The present invention relates to methods for preparing compounds as stated in the preamble to claims 1 and 2. The methods are particularly useful for preparing γ-amino acids which exhibit binding affinity to the human α.25 calcium channel subunit, including pregabalin and related compounds.

DISCUSSION

Pregabalin, (S)-(+)-3-aminomethyl-5-methyl-hexasonic acid is related to the endogenous inhibitory neurotransmitter γ-aminobutyric acid (GABA), which is involved in the regulation of neuronal activity in the brain. Pregabalin exhibits anti-seizure activity, as described in US Patent No. 5,563,175 to R.B. Silverman et al., and is thought to be useful in treating, among other conditions, pain, physiological conditions associated with psychomotor stimulants, inflammation, gastrointestinal damage, alcoholism, insomnia and various psychiatric disorders, including mania and bipolar disorder. See, respectively, US Patent No. 6,242,488 to L. Bueno et al., US Patent 6,326,374 to L. Magnus & C. A. Segal, and US Patent 6,001,876 to L. Singh; US patent 6,194,459 to H.C. Akunne et al.; US Patent 6,329,429 to D. Schrier et al.; US Patent 6,127,418 to L. Bueno et al.; US Patent 6,426,368 to L. Bueno et al US Patent 6,306,910 to L. Magnus & C.A. Segal; and US Patent No. 6,359,005 to A. C. Pande, which is hereby incorporated by reference in its entirety for all purposes.

Pregabalin has been manufactured in different ways. Typically, a racemic mixture of 3-aminomethyl-5-methyl-hexanoic acid is synthesized and then resolved into its fl - and S -enantiomers. Such procedures may employ an azide intermediate, a malonate intermediate, or Hofmann synthesis. See respectively US patent 5,563,175 by R.B. Silverman et al.; US Patent 6,046,353, 5,840,956, and 5,637,767 to

T.M. Grote et al.; and US Patent 5,629,447 and 5,616,793 to B. K. Huckabee &

D. M. Sobieray, all of which are hereby incorporated by reference in their entirety for all purposes. In each of these procedures, the racemate is reacted with a chiral acid (a resolving agent) to form a pair of diastereoisomeric salts, which are separated by known techniques such as fractional crystallization and chromatography. These procedures thus involve considerable processing in addition to the production of the racemate, which, together with the resolving agent, results in increased production costs. However, the unwanted R-enantiomer is often discarded since it cannot be efficiently recycled, thus reducing the effective throughput for the process by 50%.

Pregabalin has also been synthesized directly using the chiral auxiliaries, (4f?,5S)-4-methyl-5-phenyl-2-oxazolidinone. See, for example, US Patents 6,359,169, 6,028,214, 5,847,151, 5,710,304, 5,684,189, 5,608,090, and 5,599,973, all to R.B. Silverman et al, which are hereby incorporated by reference in their entirety for all purposes. Although these procedures provide pregabalin in high enantiomeric purity, they are less suitable for large-scale synthesis due to the use of relatively expensive reagents (such as the chiral auxiliary) which are difficult to handle, as well as special cryogenic equipment to achieve the required operating temperature, which can be as low as -78°C.

A recently published US patent application describes a process for preparing pregabalin via asymmetric hydrogenation of a cyano-substituted olefin to produce a chiral cyano precursor of (S)-3-aminomethyl-5-methylhexanoic acid. See US Patent Application No. 2003/0212290 A1 by Burk et al., published November 13, 2003, which is hereby incorporated by reference in its entirety for all purposes. The cyano precursor is then reduced to give pregabalin. The asymmetric hydrogenation uses a chiral catalyst consisting of a transition metal bond to a biophosphine ligand such as (f?,f?)-Me-DUPHOS. The procedure results in significant enrichment of pregabalin over (f?)-3-(aminomethyl)-5-methylhexanoic acid.

The method described in US patent 2003/0212290 A1 represents a commercially feasible method for the production of pregabalin, but further improvements would be desirable for various reasons. For example, phosphine ligands, including the ligand (R,R)-Me-DUPHOS, are often difficult to prepare because they exhibit two chiral centers, which makes them more expensive. Furthermore, asymmetric hydrogenation requires the use of special equipment capable of handling H2, which also makes the reaction more expensive.

SUMMARY OF THE INVENTION

The present invention provides materials and methods for preparing enantiomerically enriched γ-amino acids (formula 1) such as pregabalin (formula 9).

In one aspect, the present invention relates to a method for producing a compound of formula 1,

or a pharmaceutically acceptable complex, salt, solvate or hydrate thereof,

R<1> and R<2> are different and are each independently selected from hydrogen atom, C1-12alkyl, C3-12cycloalkyl, and substituted C3-12cycloalkyl,

characterized by the procedure including:

(a) reducing a cyano unit of a compound of formula 8;

or a salt thereof to give the compound of formula 1, or a salt thereof, and (b) optionally converting the compound of formula 1 or a salt thereof into a pharmaceutically acceptable complex, salt, solvate or hydrate thereof, wherein

R<1> and R<2> in formula 8 are as defined in formula 1, and

R 5 in formula 8 is hydrogen atom, C 1-12 alkyl, C 3-12 cycloalkyl, or aryl-C 1-6 alkyl.

In a second aspect, the present invention relates to a method for preparing a compound of formula 9,

or a pharmaceutically acceptable complex, salt, solvate or hydrate thereof,

characterized by the procedure including:

(a) reducing a cyano unit of a compound of formula 16;

or a salt thereof, to give a compound of formula 9 or a salt thereof; (b) optionally converting the compound of formula 9 or a salt thereof into a pharmaceutically acceptable complex, salt, solvate or hydrate; wherein the compound of formula 16 is optionally prepared by decarboxylation of a compound of formula 11,

or a salt thereof; and

where R 3 in formula 11 is C 1-12 alkyl, C 3-12 cycloalkyl, or aryl-C 1-6 alkyl, and R 5 in formula 16 is hydrogen atom, C 1-12 alkyl, C 3-12 cycloalkyl, or aryl-C 1-6 alkyl.

The present invention includes all complexes and salts, whether pharmaceutically acceptable or not, solvates, hydrates and polymorphic forms of the described compounds. Certain compounds may contain an alkenyl or cyclic group so that cis/trans (or ZIE) stereoisomers are possible, or may contain a keto or oxime group so that tautomerism may occur. In such cases, the present invention generally includes all ZIE isomers and tautomeric forms, whether pure, substantially pure or mixtures.

BRIEF DESCRIPTION OF THE FIGURES

Fig. 1 depicts a scheme for the production of enantiomerically enriched γ-amino acids (formula 1). Fig. 2 depicts a scheme for the preparation of cyano-substituted diesters (formula 4).

DETAILED DESCRIPTION

DEFINITIONS AND ABBREVIATIONS

Unless otherwise stated, the description uses the definitions given below. Some of the definitions of the formulas may include a hyphen ("-") to indicate a bond between atoms or a point of attachment to a named or unnamed atom or group of atoms. Other definitions and formulas may include an equal sign ("=") or an identity symbol ("=') to indicate a double bond or a triple bond, respectively. Certain forms may also include one or more asterisks ("<*>") to indicate indicate stereogenic (asymmetric or chiral) centers, although the absence of an asterisk does not indicate that the compound lacks a stereocenter. Such formulas may refer to the racemate of individual enantiomers or to individual diastereomers, which may or may not be pure or substantially pure. ; "Substituted" groups are those in which one or more hydrogen atoms have been replaced by one or more non-hydrogen groups, provided that the valence requirements are met, and that a chemically stable compound results from the substitution. ;"Ca" or "about" as used in connection with a measurable numerical value, refers to the intended value of the variable and to all values of the variable that are within the experimental error of the indicated rte value, for example within a 95% confidence interval for the mean or within ±10 per cent of the indicated value, whichever is greater. ;"Alkyl" refers to a straight-chain and branched saturated hydrocarbon group, which generally has a specified number of carbon atoms (that is, C1-alkyl refers to an alkyl group that is 1, 2, 3, 4, 5, or 6 carbon atoms, and C1- 12alkyl refers to an alkyl group having 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, or 12 carbon atoms). Examples of alkyl groups include, without limitation, methyl, ethyl, n-propyl, /'-propyl, n-butyl, s-butyl, /-butyl, f-butyl, pent-1-yl, pent-2-yl, pent -3-yl, 3-methylbut-1-yl, 3-methylbut-2-yl, 2-methylbut-2-yl, 2,2,2-trimethyleth-1-yl, n-hexyl, and the like. "Alkenyl" refers to straight-chain or branched hydrocarbon groups having single or multiple unsaturated carbon-carbon bonds, and generally having a specific number of carbon atoms. Examples of alkenyl groups include, without limitation, ethenyl, 1-propen-1-yl, 1-propen-2-yl, 2-propen-1-yl, 1-buten-1-yl, 1-buten-2-yl, 3-buten-1-yl, 3-buten-2-yl, 2-buten-1-yl, 2-buten-2-yl, 2-methyl-1-propen-1-yl, 2-methyl-2- propen-1-yl, 1,3-butadiene-1-yl, 1,3-butadiene-2-yl, and the like. ;"Alkynyl" refers to straight chain or branched hydrocarbon groups having one or more carbon-carbon triple bonds, and generally having a specific number of carbon atoms. Examples of alkenyl groups include, without limitation, ethynyl, 1-propyn-1-yl, 2-propyn-1-yl, 1-butyn-1-yl, 3-butyn-1-yl, 3-butyn-2-yl, 2-butyn-1-yl, and the like. ;"Alkanoyl" and "alkanoylamino" refer, respectively, to alkyl-C(O)- and alkyl-C(O)-NH-, where alkyl is defined above, and generally includes a specified number of carbon atoms, including the carbonyl carbon. Examples of alkanoyl groups include, without limitation, formyl, acetyl, propionyl, butyryl, pentanoyl, hexanoyl, and the like. ;"Alkenoyl" and "alkynoyl" refer, respectively, to alkenyl-C(O)- and alkynyl-C(O)-, where alkenyl and alkynyl are as defined above. References to alkenoyl and alkynyl generally include a specified number of carbon atoms, excluding the carbonyl carbon. Examples of alkenoyl groups include, without limitation, propenoyl, 2-methylpropenoyl, 2-butenoyl, 3-butenoyl, 2-methyl-2-butenoyl, 2-methyl-3-butenoyl, 3-methyl-3-butenoyl, 2-pentenoyl, 3-pentenoyl, 4-pentenoyl, and the like. Examples of alkynoyl groups include, without limitation, propynoyl, 2-butynoyl, 3-butynoyl, 2-pentynoyl, 3-pentynoyl, 4-pentynoyl, and the like. ;"Alkoxy," "Alkoxycarbonyl," and "Alkoxycarbonylamino," refer, respectively, to alkyl-O-, alkenyl-O, and alkynyl-O; to alkyl-O-C(O)-, alkenyl-O-C(O)-, alkynyl-O-C(O)-; and to alkyl-O-C(0)-NH-, alkenyl-O-C(0)-NH-, and alkynyl-O-C(0)-NH-, where alkyl, alkenyl, and alkynyl are defined above. Examples of alkoxy groups include, without limitation, methoxy, ethoxy, n-propoxy, /-propoxy, n-butoxy, s-butoxy, t-butoxy, /7-pentoxy, s-pentoxy, and the like. Examples of alkoxycarbonyl groups include, without limitation, methoxy carbonyl, ethoxycarbonyl, n-propoxy carbonyl, /-propoxycarbonyl, n-butoxycarbonyl, s-butoxycarbonyl, f-butoxycarbonyl, n-pentoxycarbonyl, s-pentoxylcarbonyl, and the like. ;"Alkylamino", "alkylaminocarbonyl", "dialkylaminocarbonyl", "alkylsulfonyl", "sulfonylaminoalkyl" and "alkylsulfonylaminocarbonyl" refer, respectively, to alkyl-NH-, alkyl-NH-C(O)-, alkyl2-N-C(0)- , alkyl-S(02)-, HS(02)-NH-alkyl-, and alkyl-S(O)-NH-C(O)- where alkyl is as defined above. "Aminoalkyl" and "cyanoalkyl" refer, respectively, to NH 2 alkyl and N=C alkyl, where alkyl is defined above. ;"Halo," "halogen" and "halogeno" are used interchangeably and refer to fluorine, chlorine, bromine and iodine. ;"Haloalkyl", "haloalkenyl", "haloalkynyl", "haloalkanoyl", "haloalkenoyl", "halo-alkynoyl", "haloalkoxy" and "haloalkoxycarbonyl" refer, respectively, to alkyl, alkenyl, alkynyl, alkanoyl, alkenoyl, alkynyl . Examples of haloalkyl groups include, without limitation, trifluoromethyl, trichloromethyl, pentafluoroethyl, pentachloroethyl, and the like. "Hydroxyalkyl" and "oxoalkyl" refer, respectively, to HO-alkyl and O=alkyl, where alkyl is as defined above. Examples of hydroxyalkyl and oxoalkyl groups include, purification, hydroxymethyl, hydroxyethyl, 3-hydroxypropyl, oxomethyl, oxoethyl, 3-oxopropyl, and the like. ;"Cycloalkyl" refers to saturated monocyclic and bicyclic hydrocarbon rings, generally having a specified number of carbon atoms making up the ring (ie, C3-7cycloalkyl) refers to a cycloalkyl group having 3, 4, 5, 6 or 7 carbon atoms as ring members. The cycloalkyl may be attached to a parent group or to a substrate of any ring atom, unless such adaptation would destroy the valency requirements. Likewise, the cycloalkyl groups may include one or more non-hydrogen substituents at least such substitution would destroy the valency requirements. Useful substituents include, without limitation, alkyl, alkenyl, alkynyl, haloalkyl, haloalkenyl, haloalkynyl, alkoxy, alkoxycarbonyl, alkanoyl, and halo, as defined above, and hydroxy, mercapto, nitro, and amino. Examples of monocyclic cycloalkyl groups include, without limitation, cyclopropyl, cyclobutyl, cyclopentyl, cyclohexyl and the like. Examples of bicyclic cycloalkyl groups include, without limitation, bicyclo[1.1.0]butyl, bicyclo[1.1.1]pentyl, bicyclo[2.1.0]pentyl, bicyclo[2.1.1]hexyl, bicyclo[3.1.0]hexyl, bicyclo [2.2.1 ]heptyl, bicyclo[3.2.0]heptyl, bicyclo[3.1.1]heptyl, bicyclo[4.1.0]heptyl, bicyclo[2.2.2]octyl, bicyclo 3.2.1]octyl, bicyclo[4.1. 1]octyl, bicyclo[3.3.0]octyl, bicyclo [4.2.0]octyl, bicyclo[3.3.1]nonyl, bicyclo[4.2.1]nonyl, bicyclo[4.3.0]nonyl, bicyclo [3.3.2] decyl, bicyclo[4.2.2]decyl, bicyclo[4.3.1]decyl, bicyclo[4.4.0]decyl, bicyclo [3.3.3]undecyl, bicyclo[4.3.2]undecyl, bicyclo[4.3.3]dodecyl, and the like, which can be attached to a parent group or substrate at any of the ring atoms, unless such attachment would destroy the valence requirements. ;"Cycloalkenyl" refers to monocyclic and bicyclic hydrocarbon rings having one or more unsaturated carbon-carbon bonds and generally having a specified number of carbon atoms making up the ring (ie, C3.7cycloalkenyl refers to cycloalkenyl group having 3, 4, 5, 6 or 7 carbon atoms as Ting members). The cycloalkenyl may be attached to a parent group or to a substrate of any ring atom, unless such attachment would destroy the valence requirements. Likewise, the cycloalkenyl groups may include one or more non-hydrogen substituents unless such substitution would destroy the valency requirements. Useful substituents include, without limitation, alkyl, alkenyl, alkynyl, haloalkyl, haloalkenyl, haloalkynyl, alkoxy, alkoxycarbonyl, alkanoyl, and halo, as defined above, and hydroxy, mercapto, nitro, and amino. ;"Cycloalkanoyl" and "cycloalkenoyl" refer to cycloalkyl-C(O)- and cycloalkenyl-C(O)-, respectively, where cycloalkyl and cycloalkenyl are as defined above. References to cycloalkanoyl and cycloalkenoyl generally include a specified number of carbon atoms, excluding the carbonyl carbon. Examples of cycloalkanoyl groups include, without limitation, cyclopropanoyl, cyclobutanoyl, cyclopentanoyl, cyclohexanoyl, cycloheptanoyl, 1-cyclobutenoyl, 2-cyclobutenoyl, 1-cyclopentenoyl, 2-cyclopentenoyl, 3-cyclopentenoyl, 1-cyclohexenoyl, 2-cyclohexenoyl, 3-cyclohexenoyl, and such. ;"Cycloalkoxy" and "cycloalkoxycarbonyl" refer, respectively, to cycloalkyl-O- and cycloalkenyl-O and to cycloalkyl-O-C(O)- and cycloalkenyl-O-C(O)-, wherein cycloalkyl and cycloalkenyl are as defined above. References to cycloalkyloxy and cycloalkylcarbonyl generally include a specified number of carbon atoms, excluding the carbonyl carbon. Examples of cycloalkoxy groups include, without limitation, cyclopropoxy, cyclobutoxy, cyclopentoxy, cyclohexoxy, 1-cyclobutenoxy, 2-cyclobutenoxy, 1-cyclopentenoxy, 2-cyclopentenoxy, 3-cyclopentenoxy, 1-cyclohexenoxy, 2-cyclohexenoxy, 3-cyclohexenoxy, and the like . Examples of cycloalkoxycarbonyl groups include, without limitation, cyclopropoxycarbonyl, cyclobutoxycarbonyl, cyclopentoxycarbonyl, cyclohexoxycarbonyl, 1-cyclobutenoxycarbonyl, 2-cyclobutenoxycarbonyl, 1-cyclopentenoxycarbonyl, 2-cyclopentenoxycarbonyl, 3-cyclopentenoxycarbonyl, 1-cyclohexenoxycarbonyl, 2-cyclohexenoxycarbonyl, 3- cyclohexenoxycarbonyl, and the like. ;"Aryl" and "arylene" refer to monovalent and divalent aromatic groups, respectively, including 5- and 6-membered monocyclic aromatic groups containing 0 to 4 heteroatoms independently selected from nitrogen, oxygen and sulfur. Examples of monocyclic aryl groups include, without limitation, phenyl, pyrrolyl, furanyl, thiophenyl, thiazolyl, isothiazolyl, imidazolyl, triazolyl, tetrasolyl, pyrazolyl, oxazolyl, isoxasolyl, pyridinyl, pyrazinyl, pyridazinyl, pyrimidinyl, and the like. Aryl and arylene groups also include bicyclic groups, tricyclic groups, etc, including fused 5- and 6-membered rings described above. Examples of multicyclic aryl groups include, without limitation, naphthyl, biphenyl, anthracenyl, pyrenyl, carbazolyl, benzoxazolyl, benzodioazolyl, benzothiazolyl, benzoimidazolyl, benzothiophenyl, quinolinyl, isoquinolinyl, indolyl, benzofuranyl, purinyl, indolisinyl, and the like. The aryl and arylene groups may be attached to a target group or to a substrate in an arbitrary ring atom, unless such attachment would destroy the valence requirements. Likewise, aryl and arylene groups may include one or more non-hydrogen substituents unless such substitution would destroy the valency requirements. Useful substituents include, without limitation, alkyl, alkenyl, alkynyl, haloalkyl, haloalkenyl, haloalkynyl, cycloalkyl, cycloalkenyl, alkoxy, cycloalkoxy, alkanoyl, cycloalkanoyl, cycloalkenoyl, alkoxycarbonyl, cycloalkoxycarbonyl, and halo, as defined above, and hydroxy, mercapto, nitro , amino, and alkylamino. "Heteroring" "heterocyclyl" refers to saturated, partially unsaturated or unsaturated monocyclic or bicyclic rings having 5 to 7 or from 7 to 11 ring members, respectively. These groups are ring members consisting of carbon atoms and from 1 to 4 heteroatoms which are independently nitrogen, oxygen or sulfur, and may include any bicyclic group in which any of the above defined monocyclic heterorings is fused to a benzene ring. Nitrogen and sulfur heteroatoms can optionally be oxidized. The heterocyclic ring may be attached to a target group or to a substrate in an arbitrary heteroatom or carbon atom, unless such attachment would destroy the valence requirements. Likewise, any of the carbon or nitrogen ring members may include a non-hydrogen substituent unless such substitution would destroy the valency requirements. Useful substituents include, without limitation, alkyl, alkenyl, alkynyl, haloalkyl, haloalkenyl, haloalkynyl, cycloalkyl, cycloalkenyl, alkoxy, cycloalkoxy, alkanoyl, cycloalkanoyl, cycloalkenoyl, alkoxycarbonyl, cycloalkoxycarbonyl, and halo, as defined above, and hydroxy, mercapto, nitro , amino, and alkylamino. ;Examples of heterorings include, without limitation, acridinyl, azosinyl, benzimidazolyl, benzofuranyl, benzothiofuranyl, besothiophenyl, benzxazolyl, benzthiazolyl, benztriazolyl, benztetrazolyl, benzisoxazolyl, benzisothiazolyl, benzimidazolinyl, carbazolyl, 4a/-/-carbazolyl, carbolinyl, cromanyl, chromenyl, sinnolinyl, decahydroquinolinyl, 2H, 6H-1,5,2-dithiazinyl, dihydrofuro[2,3-bjtetrahydrofuran, furanyl, furazanyl, imidazolidinyl, imizolinyl, imidazolyl, 1A7-indazolyl, indolenyl, indolinyl, indolisinyl, indolyl, 3/-/ -indolyl, isobenzofuranyl, isocromanyl, isoindazolyl, isoindolinyl, isoindolyl, isoquinolinyl, isothiazolyl, isoxasolyl, morpholinyl, naphthyridinyl, octahydroisoquinolinyl, oxadiazolyl, 1,2,3-oxadiazolyl, 1,2,4-oxadiazolyl, 1,2,5-oxadiazolyl . pyrazinyl, pyrazolidinyl, pyrasolinyl, pyrazolyl, pyridazinyl, pyridooxazole, pyridoimidazole, pyridothiazole, pyridinyl, pyridyl, pyrimidinyl, pyrrolidinyl, pyrrolinyl, 2/-/-pyrrolyl, pyrrolyl, quinazolinyl, quinolinyl, 4/-/-quinolisinyl, quinoxalinyl, quinuclidinyl, tetrahydrofuranyl, tetrahydroisoquinolinyl, tetrahydroquinolinyl, 6H- 1,2,5-thiadiazinyl, 1,2,3-thiadiazolyl, 1,2,4-thiadiazolyl, 1,2,5-thiadiazolyl, 1,3,4-thiadiazolyl, tianthrenyl, thiazolyl, thienyl, thienothiazolyl, thienooxasolyl, thienoimidazolyl, thiophenyl, triazinyl, 1,2,3-triazolyl, 1,2,4-triazolyl, 1,2,5-triazolyl, 1,3,4-triazolyl, and xanthenyl. "Heteroaryl" and "heteroarylene" refer, respectively, to monovalent and divalent heterorings or heterocyclyl groups, as defined above, which are aromatic. The heteroaryl and heteroarylene groups represent a subset of aryl and arylene groups, respectively. ;"Arylalkyl" and "heteroarylalkyl" refer, respectively, to aryl-alkyl and heteroaryl-alkyl, where aryl, heteroaryl, and alkyl are as defined above, Examples include, without limitation, benzyl, fluorenylmethyl, imidazol-2-ylmethyl, and such. ;"Arylalkanoyl", "heteroarylalkanoyl", "arylalkenoyl", "heteroarylalkenoyl", "arylalkynoyl" and "heteroarylalkynoyl" refer, respectively, to arylalkanoyl, heteroarylalkanoyl, arylalkenoyl, heteroarylalkenoyl, arylalkynoyl, and heteroarylalkynoyl, where aryl, heteroaryl, alkanoyl, alkenoyl, and alkynoy are defined above. Examples include, without limitation, benzoyl, benzylcarbonyl, fluorenoyl, fluorenylmethylcarbonyl, imidazol-2-oyl, imidazol-2-ylmethylcarbonyl, phenylethenecarbonyl, 1-phenylethenecarbonyl, 1-phenyl-propenecarbonyl, 2-phenyl-propenecarbonyl, 3-phenyl- propenecarbonyl, imidazol-2-yl-ethenecarbonyl, 1-(imidazol-2-yl)-ethenecarbonyl, 1-(imidazol-2-yl)-propenecarbonyl, 2-(imidazol-2-yl)-propenecarbonyl, 3-(imidazole -2-yl)-propenecarbonyl, phenylethynecarbonyl, phenylpropynecarbonyl, (imidazol-2-yl)-ethynecarbonyl, (imidazol-2-yl)-propynecarbonyl, and the like. ;"Arylalkyloxy" and "heteroarylalkyloxy" refer, respectively, to arylalkyloxy and heteroarylalkyloxy, where aryl, heteroaryl, and alkyloxy are as defined above. Examples include, without limitation, benzyloxy, fluorenylmethyloxy, imidazol-2-ylmethyloxy, and the like. ;"Aryloxy" and "heteroaryloxy" refer, respectively, to aryl-O- and heteroaryl-O-, where aryl and heteroaryl are defined above. Examples include, without limitation, phenoxy, imidazol-2-yloxy, and the like. ;"Aryloxycarbonyl", "heteroaryloxycarbonyl", "aryloxycarbonyl" and "heteroaryl-alkoxycarbonyl" refer, respectively, to aryloxy-C(O)-, heteroaryloxy-C(O)-, arylalkoxy-C(O)-, and heteroarylalkoxy- C(O)-, where aryloxy, heteroaryloxy, arylalkyloxy, and heteroarylalkyloxy are defined above. Examples include, without limitation, phenoxycarbonyl, imidazol-2-yloxycarbonyl, benzyloxycarbonyl, fluorenylmethyloxycarbonyl, imidazol-2-ylmethyloxycarbonyl, and the like. "Leaving group" refers to any group that leaves a molecule during a fragmentation process, including substitution reactions, elimination reactions, and addition-elimination reactions. Leaving groups can be nucleofugal, where the group leaves with a pair of electrons that previously served as the bond between the leaving group and the molecule, or can be electrofugal, where the leaving group does not have this electron pair. The ability of a nucleofugal leaving group to leave depends on its base strength, with the strongest bases being the weakest leaving groups. Common nucleofugal leaving groups include nitrogen (for example from diazonium salts); sulfonates, including alkyl sulfonates (for example, mesylate), fluoroalkyl sulfonates (for example, triflate, hexaflat, nonaflate, and tresylate), and aryl sulfonates (for example, tosylate, brosylate, closylate, and nosylate). Others include carbonates, halide ions, carboxylations, phenolations, and alkoxides. Some strong bases such as NH2" and OH" can become better leaving groups with treatment with an acid. Common electrofugal leaving groups include the proton, CO2, and metals. ;"Enantiomeric excess" or "ee" is a measure, for a given sample, of the excess of one enantiomer relative to a racemic sample of a chiral compound and is expressed as a percentage Enantiomeric excess is defined as 100 x (er -1) / (er + 1), where "er" is the ratio of the more abundant enantiomer relative to a less abundant enantiomer. ;"Diastereomeric excess" or "de" is a measure, for a given sample of the excess of diastereomer over a sample having equal amounts of diastereomers and expression es as a percentage. Diastereomeric excess is defined as 100 x (dr -1) / (dr + 1), where "is" is the ratio of a more frequent diastereomer to a less frequent diastereomer. ;"Stereoselective," "enantioselective," "diastereoselective," and variants thereof, refer to a given process (eg, ester hydrolysis, hydrogenation, hydroformulation, %-allyl palladium coupling hydrosilylation, hydrocyanation, olefin metathesis, hydroacylation, allylamine isomerization, etc.) as gives more of one stereoisomer, enantiomer, or diastereoisomer than of the other, respectively. ;"High level of stereoselectivity", "high level of enantioselectivity", "high level of diastereoselectivity" and variants thereof refer to a given process that yields products having an excess of one stereoisomer, enantiomer, or diastereoisomer, comprising at least about 90% of the products. For a pair of enantiomers or diastereomers, a high level of enantioselectivity or diastereoselectivity will correspond to an ee or those of at least about 80%. ;"Stereoisomeric enriched," "enantiomerically enriched," "diastereomerically enriched," and variants thereof refer, respectively, to a sample of the compound having more of one stereoisomer, enantiomer, or diastereomer than the other.The degree of enrichment may be measured by % of total product, or for a pair of enantiomers or diastereomers, by ee or de. ;"Essentially pure stereoisomer," "essentially pure enantiomer," "essentially pure diastereomer," and variations thereof refer, respectively, to a sample containing one stereoisomer, enanti omer, or diastereomer, comprising at least 95% of the sample. For pairs of enantiomers and diastereomers, a substantially pure enantiomer or diastereomer will correspond to the sample having an ee or ones of about 90% or more. ;A "pure stereoisomer," "pure enantiomer," "pure diastereomer," and variants thereof, refer, respectively, to a sample containing one stereoisomer, enantiomer, or diastereomer, comprising at least about 99.5% of the sample. For pairs of enantiomers and diastereomers, a pure enantiomer or pure diastereomer" will correspond to samples that have an ee or those of 99% or more. ; "Opposite enantiomer" refers to a molecule that is in a non-superimposable mirror image of a reference molecule, which can be obtained by inverting all of the stereogenic centers in the reference molecule. For example, if the reference molecule has S absolute stereochemical configuration, then the opposite enantiomer has R absolute stereochemical configuration. Likewise, if the reference molecule has S,S absolute stereochemical configuration, then the opposite enantiomer has R, R stereochemical configuration, and so on. ;"Stereoisomers" of a specified compound refers to the opposite enantiomer of the compound and to any diastereoisomer or geometric isomers (ZIE) of the compound. For example, if the specified compound has S, R, Z stereochemical configuration, its stereoisomer will include its opposite enantiomers with R, S, Z configuration, its diastereomers have S,S,Z configuration and R, R, Z configuration, and the geometric isomers have S, R, E configuration, R, S, E configuration, S, S, E configuration, and R, R, E configuration. ;"Enantioselective value" or "E" refers to the ratio of specificity constants for each enantiomer of the compound undergoing a chemical reaction or transformation and can be calculated (for S-enantiomers) from the expression. ; ; where Ks and KR are first-order rate constants for the conversion of S- and R-enantiomers, respectively; Ksmog Krmer Michaelis constants for S- and R- ;enantiomers, respectively; x is the fractional conversion of the substrate; eep and eeser enantiomeric excess of the product and substrate (reactant), respectively. ;"Lipase unit" or "LU" refers to the amount of enzyme (ig) that releases 1 u.mol of titratable butyric acid/per min when it is combined with tributyrin and an emulsifier (gum arabic) at 30°C and pH 7. "Solvate" refers to a molecular complex comprising a described or claimed compound and a stoichiometric or non-stoichiometric amount of one or more solvent molecules (eg EtOH). "Hydrates" refers to a solvate comprising a described or claimed compound and a stoichiometric or non-stoichiometric amount of water. ;"Pharmaceutically acceptable complexes, salts, solvates or hydrates" refers to complexes, acids, base addition salts, solvates or hydrates of claimed and described compounds, which are within the scope of medical judgment, suitable for use in contact with tissues of patients without undue toxicity, irritation, allergic response and the like, consistent with a reasonable benefit/risk ratio, and effective for their intended use. ;"Pre-catalyst" or "catalyst precursor" refers to a compound or set of compounds that is converted to catalyst prior to use. ;"Treatment" refers to reversing, alleviating, inhibiting the process of, or preventing a disorder to which such terms apply, or preventing one or more symptoms of such disorders or conditions. ;"Processing" refers to the act of "processing" as defined immediately above. ;Table 1 indicates abbreviations used in the description. ; In some of the reaction schemes and examples below, certain compounds can be prepared by the use of protecting groups, which prevent unwanted chemical reaction in otherwise reactive sites. Protecting groups can also be used to improve solubility or otherwise modify physical properties of the compound. For a description of protecting group strategy, a description of materials and methods for installing and removing protecting groups, and a comparison of useful protecting groups for common functional groups, including amines, carboxylic acids, alcohols, ketones, aldehydes, and the like, see T. W. Greene and P.G. Wuts, Protecting Groups in Organic Chemistry (1999) and P. Kocienski, Protective Groups (2000), which are incorporated herein by reference in their entirety for all purposes. Additionally, some of the diagrams and examples below may omit details of common reactions, including oxidations, reductions, and the like, which are known to those skilled in the art of organic chemistry. The details of such reactions can be found in a number of treatises, including Richard Larock, Comprehensive Organic Transformations (1999), and the multi-volume series edited by Michael B. Smith and others, Compendium of Organic Synthetic Methods (1974-2003). In general, starting material and reagents can be obtained from conventional sources or prepared from literature sources. In general, the chemical transformations described in the specification can be carried out using substantially stoichiometric amounts of reactants, although certain reactions may benefit from using an excess of one or more of the reactants. Furthermore, many of the reactions described in the specification, including the enantioselective hydrolysis of the racemic diester (formula 4) described in detail below, can be carried out at about RT, but certain reactions may require the use of higher or lower temperatures, depending on reaction kinetics, yield and the like. Furthermore, many of the chemical transformations may use one or more compatible solvents, which may affect the reaction rate of the yield. Depending on the nature of the reactants, the one or more solvents may be polar protic solvents, polar aprotic solvents, non-polar solvents, or any combination. Any reference in the description to a concentration range, temperature range, pH range, catalyst loading range, and so on, whether or not the word "range" is expressly used, includes the endpoints indicated. Materials and procedures are described for producing optically active γ-amino acids (formula 1) including pharmaceutically acceptable salts, esters, amides or prodrugs thereof. The compounds of formula 1 include substituents R<1> and R<2>, which are defined above. Useful compounds of formula 1 involving those wherein R<1>is hydrogen atom and R2 is C1-12alkyl, C3-12cycloalkyl, or substituted C3-12cycloalkyl, or those wherein R<2>is a hydrogen atom and R<1>is C1- 12alkyl, C3-12cycloalkyl, or substituted C3-12cycloalkyl. Particularly useful compounds of formula 1 include those wherein R<1> is a hydrogen atom and R 2 is C 1-6 alkyl or C 3-7 cycloalkyl, or those wherein R 2 is a hydrogen atom and R< 1> is C 1-6 alkyl or C 3-7 cycloalkyl. Particularly useful compounds of formula 1 include those where R<1> is a hydrogen atom and R<2> is C 1-4 alkyl, such as pregabalin (formula 9). Fig. 1 shows a process for the production of optically active γ-amino acids (formula 1). The process includes the steps of contacting or combining a reaction mixture comprising a cyano-substituted diester (formula 4) and water with an enzyme to provide a product mixture which includes an optically active dicarboxylic acid monoester (formula 3) and an optically active diester (formula 5). The cyano-substituted diester (formula 4) has a stereogenic center, which is indicated by an asterisk ("<*>"), and as described below, can be prepared in accordance with a reaction scheme shown in Fig. 2. Before coupling with enzyme, the cyano-substituted diester (formula 4) typically comprises a racemic (equimolar) mixture of the diester of formula 5 and its opposite enantiomer. Substituents R<1>, R<2>, and R<3> in formula 3, formula 4, and formula 5, and substituent R<4> in formula 4 and formula 5 are as defined above in connection with formula 1. Generally , unless otherwise specified, with a specific identifier (R<1>,

R<2>, R<3>, etc.) is defined for the first time in connection with a formula, then the same substituent identifier used in subsequent formulas will have the same meaning as in the previous formula.

The enzyme (or biocatalyst) can be any protein which, having little or no effect on the compound of formula 5, will catalyze the hydrolysis of its opposite enantiomer to give dicarboxylic acid monoesters (formula 3). Thus, useful enzymes for enantioselective hydrolysis of the compound of formula 4 to formula 3 may include hydrolases, including lipases, certain proteases and other enantioselective esterases. Such enzymes can be obtained from a variety of natural sources, including animal organs and microorganisms. For example, see Table 2 for a non-limiting list of commercially available hydrolases.

As shown in the Examples section, useful enzymes for the enantioselective conversion of the cyano-substituted diester (Formula 4 and Formula 12) to the desired optically active dicarboxylic acid monoester (Formula 3 and Formula 11) include lipases. Particularly useful lipases include enzymes derived from the microorganism thermomyses lanuginosus, such as those available from Novo-Nordisk A/S under the trade name LIPOLASE® (CAS no. 9001-62-1). LIPOLASE® enzymes are obtained by submerged fermentation of an aspergillus orysae microorganism genetically modified with DNA from Thermomyses lanuginosus DSM 4109 which codes for the amino acid sequence of the lipase. LIPOLASE® 100L and LIPOLASE® 100T are available as a liquid solution and a granular solution, respectively. For every half nominal activity of 100 kLU/g. Other forms of LIPOLASE® include LIPOLASE® 50L, which has half the activity of LIPOLASE® 100L, and LIPOZYME® 100L, which has the same activity as LIPOLASE® 100L, but is graded for the nutrient.

Various screening techniques can be used to identify suitable enzymes. For example, a large number of commercially available enzymes can be screened using the high-throughput techniques described in the examples section below. Other enzymes (or microbial sources of enzymes) can be screened using enrichment isomerization techniques. Such techniques typically involve the use of carbon-limited or nitrogen-limited media supplemented with enrichment substrate, which may be the racemic substrate (formula 4) or a structurally similar compound. Potentially useful microorganisms are selected for further investigation based on their ability to grow in media containing the enrichment substrate. These microorganisms are then evaluated for their ability to enantioselectively catalyze ester hydrolysis by contacting the suspension of the microbial cells with racemic substrate and testing for the presence of the desired optically active dicarboxylic acid monoester (formula 3) using analytical methods such as chiral HPLC , gas liquid chromatography, LC/MS, and the like.

As soon as a microorganism that has the necessary hydrolytic activity has been isolated, enzyme engineering can be used to improve the properties of an enzyme it produces. For example, and without limitation, enzyme engineering can be used to increase the yield and enantioselectivity of the ester hydrolysis, to make the temperature and pH ranges wider for the enzyme, and to improve the enzyme's tolerance to organic solvents. Useful enzyme engineering techniques include rational design methods, such as site-directed hydrogenesis, and in vitro directed evolution techniques that employ successive rounds of random mutagenesis, gene expression, and high-throughput screening to optimize desired properties. See for example, K. M. Koeller & C.-H. Wong, "Enzymes for chemical synthesis," Nature 409:232-240 (Jan. 11, 2001), and references therein, the complete description of which is incorporated herein by reference.

The enzyme can be in the form of whole microbial cells, permeabilized microbial clays, the extract of microbial cells, partially purified enzymes, purified enzymes and the like. The enzyme may comprise a dispersion of particles having an average particle size, based on volume, of less than about 0.1 mm (finely distributed dispersion) or of about 0.1 mm or greater (coarse dispersion). Coarse enzyme dispersions offer possible processing advantages over finely divided dispersions. For example, coarse enzyme particles can be used repetitively in batch processes or in semi-continuous or continuous processes, and can usually (for example by filtration) from other components of the bioconversion easier than finely divided enzymes.

Useful coarse enzyme dispersions include cross-linked enzyme crystals (CLECs) and cross-linked enzyme aggregates (CLEAs) which are mainly composed of enzymes. Other coarse dispersions may include enzymes immobilized on or within an insoluble support medium. Useful solid support media include polymer matrices, calcium alginate, polyactrylamide, EUPERGIT®, and other polymeric materials, as well as inorganic matrices such as CELITE®. For general description of CLECs and other enzyme immobilization techniques, see U.S. Pat. patent 5,618,710 by M.A. Navia & N.L. St. Clair. For general discussion of CLEAs, including formulation and application, see U.S. patent application no. 2003/0149172 by L. Cao & J. Elzinga et al. See also A. M. Anderson, Biocat. Biotransform, 16:181

(1998) and P. Lopez-Serrano et al., Biotechnol. Easy. 24:1379-83 (2002) for a discussion of the application of CLEC and CLEA technology for a lipase. The complete descriptions of the above-mentioned reference are incorporated herein by reference for all purposes.

The reaction mixture may comprise a single phase or may comprise multiple phases (for example a two- or three-phase system). Thus, for example, the enantioselective hydrolysis shown in fig. 1 can take place in a single aqueous phase, which contains the enzyme, the original racemic substrate (formula 4), the undesired optically active diester (formula 5) and the desired optically active dicarboxylic acid monoester (formula 3). Alternatively, the reaction mixture may comprise a multiphase system that includes an aqueous phase in contact with a solid phase (eg enzyme or product), an aqueous phase in contact with an organic phase, or an aqueous phase in contact with an organic phase and a solid phase. For example, the enantioselective hydrolysis can be carried out in a two-phase system consisting of a solid phase, which contains the enzyme, and an aqueous phase, which contains the original racemic substrate, the undesired optically active diester, and the desired optically active dicarboxylic acid monoester.

Alternatively, the enantioselective hydrolysis can be carried out in a three-phase system composed of a solid phase, which contains the enzyme, an organic phase which initially contains the racemic substrate (formula 4), and an aqueous phase which initially contains a small fraction of the racemic substrate. Since the desired optically active dicarboxylic acid monoester (formula 3) has a lower pKa than the unreacted optically active diester (formula 5) and therefore exhibits greater aqueous solubility, the organic phase is enriched with the unreacted diester, while the aqueous phase was enriched with the desired dicarboxylic acid monoester as the reaction takes place.

The amounts of the racemic substrate (formula 4) and the biocatalyst used in the enantioselective hydrolysis will depend, among other things, on the ratio between particulate cyano-substituted diester and enzyme. In general, however, the reaction may utilize a substrate that has an initial concentration of about 0.1 M to about 3.0 M, and in many cases has an initial concentration of about 1.5 M to about 3.0 M. Furthermore, the reaction can generally use an enzyme loading of about 1% to about 10%, and in many cases can use an enzyme loading of about 35 to about 4% (vol/vol).

The enantioselective hydrolysis can be carried out over a wide range of temperatures and pH. For example, the reaction can be carried out at a temperature of from about 10°C to a temperature of about 50°C, but is typically carried out at about RT. Such temperatures generally allow complete conversion (eg, about 42% to about 50%) of the racemate (Formula 4) within a reasonable period of time (about 21 to about 241) without deactivation of the enzyme. Further, the enantioselective hydrolysis can be carried out at a pH of about 5 to a pH of about 10, more typically at a pH of about 6 to a pH of about 9, and often at a pH of about 6.5 to a pH of about 7.5.

In the absence of pH regulation, the pH of the reaction mixture will be reduced as the hydrolysis of the substrate (formula 4) proceeds due to formation of the dicarboxylic acid monoester (formula 3). To compensate for this change, the hydrolysis reaction can be run with internal pH control (ie in the presence of a suitable buffer) or can be run with external pH control through the addition of a base. Suitable buffers include potassium phosphate, sodium phosphate, sodium acetate, ammonium acetate, calcium acetate. BES, BICINE, HEPES, MES, MOPS, PIPES, TAPS, TES, TRICINE, Tris, TRIZMA®, or other buffers that have a pKa of about 6 to a pKa of about 9. The buffer concentration is generally in the range of about 5 mM to about 1 mM, and typically ranges from about 50 mM to about 200 mm. Suitable bases include aqueous solutions consisting of KOH, NaOH, NH4OH, etc., having concentrations in the range from about 0.5 M to about 15 M, and more typically in the range from about 5 M to about 10 M. Other inorganic additives such as calcium acetate can also be used.

After or during the enzymatic conversion of the racemate (formula 4), the desired optically active dicarboxylic acid monoester (formula 3) can be isolated from the product mixture using standard techniques. For example, in the case of a single (aqueous) phase batch reaction, the product mixture can be extracted one or more times with a non-polar organic solvent, such as hexane or heptane, which separates the desired dicarboxylic monoester (formula 2) and the non- reacted diesters (formula 5) in aqueous and organic phases, respectively. Alternatively, in the case of a multi-phase reaction using aqueous and organic phases enriched with the desired monoester (formula 3) and the unreacted diester (formula 5), respectively, the monoester and diester can be separated batchwise after reaction, or can are separated semi-continuously or continuously during the enantioselective hydrolysis.

As indicated in Figure 1, the unreacted diester (formula 5) can be isolated from the organic phase and racemized to give the racemic substrate (formula 4). The resulting racemic feed (formula 4) can be recycled or combined with unconverted racemic substrate, which then undergoes enzymatic conversion to formula 3 as described above. Recycling of unreacted diester (formula 5) increases the overall yield of the enantioselective hydrolysis above 50%, thereby increasing the atom economy of the process and lowering the costs associated with disposal of undesired enantiomers.

Treatment of the diester (formula 5) with a base sufficiently strong to abstract an acidic α-proton for the malonate unit generally results in inversion of the stereogenic center and generation of the racemic substrate. Useful bases include organic bases such as alkoxides (eg, sodium ethoxide), linear aliphatic amines, and cyclic amines, and inorganic bases such as KOH, NaOH, NH 4 OH, and the like. The reaction is carried out in a compatible solvent, including polar optical solvents such as EtOH or aprotic polar solvents such as MTBE. Reaction temperatures above RT will typically improve the yield of the racemization process.

As shown in fig. 1, the essentially enantio-pure dicarboxylic acid monoester (formula 3) can be converted to an optically active γ-amino acid (formula 1) using at least three different methods. In one method, the monoester (formula 3) is hydrolyzed in the presence of an acid catalyst or a base catalyst to give an optically cyano-substituted dicarboxylic acid (formula 6) or corresponding salt. The cyano unit of the resulting dicarboxylic acid (or its salt) is reduced to give an optically active γ-amino dicarboxylic acid (formula 7) or a corresponding salt, which is then decarboxylated by treatment with an acid, with heating, or both, to to give the desired optically active γ-amino acid (formula 1). The cyano unit can be reduced via reaction with H 2 i in the presence of a catalytic amount of Raney nickel, palladium, platinum, and the like, or by reaction with a reducing agent such as iAIH 4 , BH 3 -Me 2 S, and the like. Useful acids for hydrolysis and decarboxylation reactions include mineral acids, such as HCIO 4 , H 1 , H 2 SO 4 , HBr, HCI, and the like. Useful base catalysts for the hydrolysis reaction include various alkali and alkaline earth metal hydroxides, and oxides, including LiOH, NaOH, KOH, and the like.

In another approach, the dicarboxylic acid monoester (formula 3) undergoes reductive cyclization to form the optically active cyclic 3-carboxy-pyrrolidin-2-one

(formula 2) which is then treated with an acid to give the desired enantiomerically enriched γ-amino acid (formula 1). The reductive cyclization can be carried out by reacting the monoester (formula 3) with H 2 i in the presence of a catalytic amount of Raney nickel, palladium, platinum and the like. One or more acids may be used to hydrolyze and decarboxylate the resulting lactam acid (formula 2) including mineral acids such as HCI04, H1, H2SO4, HBr, and HCl, and organic acids such as HOAc, TFA, p-TSA, and the like. The concentration of the acids can be in the range from about 1N to about 12 N, and the amount of the acids can be in the range from about 1 ek to about 7 ek. The hydrolysis and decarboxylation reactions can be carried out at a temperature of about RT or higher, or at a temperature of about 60°C or higher, or at a temperature in the range from about 60°C to about 130°C.

In a third method, the ester unit of the dicarboxylic acid monoester (formula 3) is first hydrolyzed! to give the cyano-substituted dicarboxylic acid (formula 6 or its salt) as described above. The resulting dicarboxylic acid (or its salt) is then decarboxylated to give an optically active cyano-substituted carboxylic acid or its salt (formula 8 where R<5>is a hydrogen atom, although R5 may be C1-12alkyl, C3-12cycloalkyl, or aryl-C 1-6 alkyl as indicated below) The same conditions are used to decarboxylate the lactam acid (formula 2) or γ-amino dicarboxylic acid (formula 7) can be used. Instead of first hydrolyzing the ester unit, the dicarboxylic acid monoester (formula 3) can first be directly decarboxylated to a cyano-substituted monoester (formula 8) by heating the aqueous solution of the dicarboxylic acid monoester (as a salt) to a temperature of from about 50°C to reflux. Krapcho conditions (DMSO/NaCl/water) can also be used. In either case, the cyano unit of the compound of formula 8 can then be reduced to give the optically active γ-amino acid (formula 1). In addition to Raney nickel, a number of other catalysts can be used to reduce the cyano unit of the compound of formula 3, 6 and 8. These include, without limitation, heterogeneous catalysts containing from about 0.1% to about 20%, and more typically from about 1% to about 5%, based on weight, of transition metals such as Ni, Pd, Pt, Rh, Re, Ru, and Ir, including oxides and combinations thereof, typically supported on various materials, including Al 2 O 3 , C , CaCO 3 , SrCO 3 , BaSO 4 , MgO, SiO 2 , TiO 2 , ZrO 2 , and the like. Many of these metals, including Pd, can be doped with an amine, a sulfide, or another metal such as Pb, Cu, or Zn. Useful catalysts thus include palladium catalysts such as Pd/C, Pd/SrCO 3 , Pd/Al 2 O 3 , Pd/MgO, Pd/CaCO 3 , Pd/BaSO 4 , PdO, Pd black, PdCl 2 , and the like, containing from about 1% to about 5% Pd, based on weight. Other useful catalysts include Rh/C, Ru/C, Re/C, PtO 2 , Rh/C, RuO 2 , and the like.

The catalytic reduction of the cyano moiety is typically carried out in the presence of one or more polar solvents, including without limitation, water, alcohols, ethers, esters, and acids, such as MeOH, EtOH, IPA, THF, EtOAc, and HOAc. The reaction can be carried out at temperatures in the range from about 5°C to about 100°C, and reactions at RT are most common. In general, the substrate-to-catalyst ratio may range from about 1:1 to about 1000:1, based on weight, and the H 2 pressure may range from atmospheric pressure, 0 psig, to about 1500 psig. More typically, substrate-to-catalyst ratios range from about 4:1 to about 20:1, and H 2 pressures range from about 25 psig to about 150 psig.

All of the preceding methods can be used to convert the essentially enantiomeric monoester (formula 3) into the optically active γ-amino acid (formula 1), but each provides certain advantages over the others. For example, after acid work-up of the process using a reductive ring formation, the lactam acid (formula 2) can be isolated and purified by extracting it into an organic solvent, while the cyano-substituted dicarboxylic acid (formula 6) can be more difficult to isolate due to its comparatively higher aqueous solubility. Isolation of the lactam acid (formula 2) reduces the transfer of water-soluble impurities to the final product mixture, and enables higher reactant concentrations (for example, about 1 M to about 2 M) during hydrolysis and decarboxylation, thereby increasing process throughput. Furthermore, direct decarboxylation by heating the aqueous solution of the dicarboxylic acid monoester (formula 3) gives the cyano monoester (formula 8) in high enantiomeric purity. This compound can be separated from the reaction medium by extraction in an organic solvent or by direct phase separation, which ensures effective removal of inorganic impurities with the water phase. High reaction throughput and avoiding strong acid conditions are two advantages of this solution.

Fig. 2 illustrates a procedure for producing cyano-substituted diesters (formula 4), which can function as substrates for the enzymatic enantioselective hydrolysis shown in fig. 1. The process includes a crossed aldol condensation, which involves converting an unsymmetrical ketone or an aldehyde (formula 17) with a malonic acid diester (formula 18) in the presence of a catalytic amount of a base to give an a,p-non- saturated malonic acid diester/ (formula 19), where R<1>, R<2>, R<3>, and R<4> are as defined in connection with formula 1. This type of crossed aldol reaction is known as a Knoevenagel condensation, which is described in a number of literature reviews. See, e.g., B. K. Wilk, Tetrahedron 53:7097-7100 (1997) and the references cited therein, the complete descriptions of which are incorporated herein by reference for all purposes.

In general, any base capable of generating an enolate ion from the malonic acid diester (formula 18) can be used, including secondary amines, such as di-n-propylamine, di-/-propylamine, pyrrolidine, etc., and their salts. The reaction may include a carboxylic acid, such as HOAc, to neutralize the product to minimize enolization of the unsymmetrical ketone or aldehyde (Formula 17). Reactions involving unsymmetrical ketones may also utilize Lewis acids, such as titanium tetrachloride, zinc chloride, zinc acetate, and the like to promote reaction. The reaction is typically run in a hydrocarbon solvent under reflux conditions. Useful solvents include hexane, heptane, cyclohexane, toluene, methyl f-butyl ether, and the like, with azeotropic removal of water.

In a subsequent step, a cyano source such as HCN, acetone cyanohydrin, an alkali metal cyanide (NaCN, KCN, etc), or an alkaline earth metal cyanide (magnesium cyanide etc), undergoes conjugate addition to the p-carbons of the a,p-unsaturated malconic acid diesters (formula 19). The reaction is typically carried out in one or more polar protic solvents, including EtOH, MeOH, n-propanol, isopropanol, or polar aprotic solvents, such as DMSO, and the like. Subsequent acid work-up gives the cyano-substituted diester (formula 4). For an application of the method depicted in fig. 2 to prepare a pregabalin precursor (Formula 2) See US Patent 5,637,767 to Grote et al., which is hereby incorporated by reference in its entirety for all purposes.

The desired (S)- or (R)-enantiomers of any compound described herein can be further enriched through classical resolution, chiral chromatography, or recrystallization. For example, the optically active amino acids (formula 1 or formula 9) can be converted with an enantiomerically pure compound (eg acid or base) to give a pair of diastereoisomers, each composed of a single enantiomer, which are separated via fractional recrystallization or chromatography. The desired enantiomer is then regenerated from the appropriate diastereoisomer. Furthermore, the desired enantiomer can often be further enriched by recrystallization in a suitable solvent, if it is available in sufficient quantity (for example typically not much less than about 85% ee, and in some cases not much less than about 90% ee).

As described throughout the specification, many of the disclosed compounds have stereoisomers. Some of these compounds can exist as single enantiomers (enantiomers) or mixtures of pure enantiomers (enriched or racemic samples) which, depending on the relative excess of one enantiomer in relation to the other in a sample, can exhibit optical activity. Such stereoisomers, which are non-superimposable mirror images, exhibit a stereogenic axis or one or more stereogenic centers (ie, chirality). Other disclosed compounds may be stereoisomers that are not mirror images. Such stereoisomers, which are known to be diastereoisomers, may be chiral or achiral (containing no stereogenic center). They include molecules containing an alkynyl or cyclic group, such that cis/trans (or ZIE) stereoisomers are miscible, or molecules containing two or more stereogenic centers, where inversion of a single stereogenic center generates a corresponding diastereoisomer. Unless otherwise stated or otherwise apparent (for example, through the use of stereobonds, stereocenter descriptions, etc.), the scope of the present invention generally includes the reference compound and its stereoisomers, whether each pure (for example, the enantiomer) or mixtures (for e.g. enantiomerically enriched or racemic).

Many of the compounds may also contain a keto or oxime group, so that tautomerism may occur. In such cases, the present invention generally includes tautomeric forms, whether each pure or mixtures.

Many of the compounds described in this specification, including those represented by Formula 1 and Formula 9, are capable of forming pharmaceutically acceptable salts. These salts include, without limitation, acid addition salts (including diacidic) and base salts. Pharmaceutically acceptable acid addition salts include non-toxic salts derived from inorganic acids such as hydrochloric, nitric, phosphoric, sulfuric, hydrobromic, hydroiodic, hydrofluoric, phosphoric and the like, and likewise non-toxic salts derived from organic acids such as aliphatic mono- and dicarboxylic acids , phenyl-substituted alkanoic acids, hydroxyalkanoic acids, alkanedioic acids, aromatic acids, aliphatic and aromatic sulfonic acids, acids, etc. Such salts thus include sulfate, pyrosulfate, bisulfate, sulfite, bisulfite, nitrate, phosphate, monohydrogen phosphate, dihydrogen phosphate, metaphosphate, pyrophosphate, chloride, bromide, iodide, acetate, trifluoroacetate, propionate, caprylate, isobutyrate, oxalate, malonate, succinate, suberate, sebacate, fumarate, maleate, mandelate, benzoate, chlorobenzoate, methylbenzoate, dinitrobenzoate, phthalate, benzenesulfonate, toluenesulfonate, phenylacetate, citrate, lactate, malate, tartrate, methanesulfonate, and the like.

Pharmaceutically acceptable base salts include non-toxic salts derived from bases including metal cations, such as alkali and alkaline earth metal cations, as well as amines. Examples of suitable metal cations include, without limitation, sodium cations, (Na<+>), potassium cations (K<+>), magnesium cations (Mg<2+>), calcium cations, (Ca<2+>), and the like. Examples of suitable amines include, without limitation, N,N'-dibenzylethylenediamine, chloroprozain, choline, diethanolamine, dicyclohexylamine, ethylenediamine, N-methylglucamine, procaine, and t-butylamine. For a description of useful acid addition and base salts, see S. M. Berge et al., "Pharmaceutical Salts," 66 J. of Pharm. Sci., 1-19 (1977); see also Stahl and Wermuth, Handbook of Pharmaceutical Salts: Properties, Selection, and Use (2002).

A pharmaceutically acceptable acid addition salt (or base salt) can be prepared by contacting a compound-free base (or free acids) or zwitterion with a sufficient amount of a desired acid (or base) to produce a non-toxic salt. If the salt is precipitated from solution, it can be isolated by filtration; otherwise, the salt can be recovered by solvent evaporation. One can also generate the free base (or free acid) by combining the acid addition salt with a base (or with the base salt with an acid). Although certain physical properties of the free base (or free acid) and its respective acid addition salt (or base salt) may differ, (for example solubility, crystal structure, hygroscopicity, etc), a compound's free base and acid addition salt (or its free acid and base salt) otherwise the same for purposes of this description.

Described and claimed compounds can exist in both unsolvated and solvated forms, and as other types of complexes in addition to salts. Useful complexes include clathrates or compound-host inclusion complexes where the compound and host are present in stoichiometric or non-stoichiometric amounts. Useful complexes may also contain two or more organic, inorganic or organic and inorganic components in stoichiometric or non-stoichiometric amounts. The resulting complexes can be ionized, partially ionized, or non-ionized. For a review of such complexes, see J. K. Haleblian, J. Pharm. Pollock. 64(8): 1269-88

(1975). Pharmaceutically acceptable solvates also include hydrates and solvates where the crystallization solvent may be isotopically substituted, for example, D2O, d6-acetone, d6-DMSO, etc. In general, for the purposes of this specification, reference to an unsolvated form of a compound also includes the corresponding solvate or hydrated form of the compound.

The described compounds also include all pharmaceutically acceptable isotopic variations, where at least one atom is replaced by an atom having the same atomic number, but an atomic mass different from the atomic mass normally found in nature. Examples of isotopes suitable for inclusion in the disclosed compounds include, without limitation, isotopes of hydrogen, such as <2>H and <3>H; isotopes of carbon, such as 13C and<14>C; isotopes of nitrogen, such as 15N; isotopes of oxygen, such as<17>0 and<18>0; isotopes of phosphorus, such as31P and<32>P; isotopes of sulfur, such as<35>S; isotopes of fluorine, such as 18F; and isotopes of chlorine, such as<36>CI. The use of isotopic variations (eg, deuterium,<2>H) may provide certain therapeutic advantages resulting from greater metabolic stability, eg, increased in vivo half-life or reduced dosage requirements. Furthermore, certain isotopic variations of the disclosed compounds may incorporate a radioactive isotope, (eg, tritium, 3H, or <14>C), which may be useful in drug and/or substrate tissue distribution studies.

EXAMPLES

The following examples are intended to be illustrative and non-limiting, and represent specific embodiments of the present invention.

GENERAL MATERIALS AND METHODS

Enzyme screening was performed using a 96 well plate as described in D. Yazbeck et al., Synth. Catal. 345:524-32 (2003), and the full description thereof is incorporated herein by reference for all purposes. All enzymes used for the screening plate, (see Table 2) were available from commercial enzyme suppliers including Amano (Nagoya, Japan), Roche (Basel, Switzerland), Novo Nordisk (Bagsvaerd, Denmark), Altus Biologics Inc. (Cambridge, MA), Biocatalytics (Pasadena, CA), Toyobo (Osaka, Japan), Sigma-Aldrich (St. Louis, MO) and Fluka (Buchs, Switzerland). The screening reactions were performed in an Eppendorf Thermomixer-R (VWR). Subsequent enzymatic resolutions on a larger scale used LIPOLASE® 100L and LIPOLASE® 100T, which are available from Novo-Nordisk A/S (CAS no. 9001-62-1).

NUCLEAR MAGNETIC RESONANCE

Three hundred MHz<1>H NMR and 75 MHz<13>C NMR spectra were acquired on a BRUKER 300 UltraShield™ equipped with a 5 mm auto switchable PHQNP probe. Spectra were generally acquired close to RT, and standard autolock, autoshim and autogain routines were used. The samples were typically spun at 20 Hz for the 1D experiment.<1>H NMR spectra were acquired using 30-degree tip angle pulses, 1.0 s recycle delay and 16 scans at a resolution of 0.25 Hz/point. The recording window was typically 8000 Hz from +18 to -2 ppm (Reference TMS at 0 ppm) and processing was with 0.3 Hz line broadening. Typical acquisition times were 5-10 s. Regular<13>C NMR spectra were acquired using 30-degree tip angle pulses, 2.0 s recycle delay and 2048 scans with a resolution of 1 Hz/point. Spectral width was typically 25 KHz from +235 to -15 ppm (Reference TMS at 0 ppm). Proton decoupling was applied continuously and 1 Hz line broadening was applied during processing. Typical recording times were 102 min.

MASS SPECTROMETRY

Mass spectrometry was performed on a HEWLETT PACKARD 1100MSD using HP Chemstation Plus software. The LC was equipped with an Agilent 1100 quaternary LC system and an Agilent liquid processor as an autosampler. Data were acquired during electron spray ionization with ACN/water (containing 0.1% formic acid) as solvent (10% ACN to 90%, 7 min). Temperatures: probe was 350 °C, source was 150 °C. Coron depletion was 3000 V for positive ion and 3000 V for negative ion.

HIGH PERFORMANCE LIQUID CHROMATOGRAPHY

High performance liquid chromatography (HPLC) was performed on a Series 1100 AGILENT TECHNOLOGIES instrument equipped with an Agilent 220 HPLC autosampler, quaternary pumps and a UV detector. LC was PC-controlled using HP Chemstation Plus software. Normal phase chiral HPLC was performed using chiral HPLC columns available from Chiral Technologies (Exton, PA) and Phenomenex (Torrance, CA).

GAS CHROMATOGRAPHY

Gas chromatography (GC) was performed on an Agilent 6890N network GC system equipped with an FID detector with electrometer, a 7683 Series split/splitless capillary injector, a trackpad that monitors four external events, and an inboard printer/plotter. Enantiomeric excesses of diesters (formula 13, R<3>=R<4>=Et) and monoesters (formula 11, R<3>=Et) were carried out using CHIRALDEX G-TA column (30 m x 0.25 mm ), with helium carrier gas, and at 135°C. Under such conditions, the monoester decomposes to S-3-cyano-5-methyl-hexanoic acid ethyl ester, and ee was determined based on the decomposition product. The chiral GC columns used in the analysis were available from Astec, Inc (Whippany, NJ).

EXAMPLE 1: Enzyme screening via enzymatic hydrolysis of (R/S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexasonic acid ethyl ester (Formula 20) to give (3S)-3-cyano-2-ethoxycarbonyl-5- methyl hexanoic acid (formula 21).

Enzyme screening was performed using a screening kit composed of individual enzymes deposited in separate wells in a 96-well plate that was prepared in advance according to a procedure described in D. Yazbeck et al., Synth. Catal. 345:524-32 (2003). Each of the wells had an empty volume of 0.3 ml, (shallow well plate). One well of the 96-well plate contained only phosphate buffer (10 u.L, 0.1 M, pH 7.2), and a second well contained only ACN (10 u.L), and each of the remaining wells contained one of the 94 enzymes listed in Table 2 ( 10 µl, 100 mg/mil). Before use, the screening kit was removed from storage at -80°C and the enzymes were allowed to thaw at RT for 5 min. Potassium phosphate buffer (85 µl, 0.1 M, pH 7.2) was dispensed into each well using a multichannel pipette. Concentrated substrate (formula 20, 5 µl) was then added to each well via a multichannel pipette and the 96 reaction mixtures were incubated at 30°C and 750 rpm. The reactions were stopped and sampled after 241 by transferring each of the reaction mixtures to separate wells in a 96 well plate. Each of the wells had a void volume of 2 mil (deep well plate) and contained EtOAc (1 mil) and HCl (1N, 100 ul). The components in each well were mixed by aspiration of the well contents with a pipette. The second plate was centrifuged and 100 DL of organic supernatant was transferred from each well to separate wells in a third 96 well plate (shallow plate). Wells in the third plate were then sealed using a penetrable mat cover. As soon as the wells were sealed, the third plate was transferred to a GC system for determination of optical purity (ee).

Table 3 lists the enzyme, brand name, supplier and E-value for some of the enzymes that were screened. For a given enzyme, the E value was interpreted as the relative reactivity of a pair of enantiomers (substrates). The E values listed in Table 3 were calculated for HPLC data (fraction conversion, x, and ee) using a computer program named Ee2, which is available from the University of Graz. In general, enzymes exhibiting S selectivity and an E value greater than 35 are suitable for scale-up.

EXAMPLE 2. Enzymatic resolution of (R/S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (Formula 20) to give (3S)-3-cyano-2-ethoxycarbonyl-5-methyl- hexanoic acid potassium salt (formula 23) and (R)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 22)

A reactor (392 L) equipped with overhead stirring was charged with potassium phosphate buffer (292.2 L, 10 mM, pH 8.0) and LIPOLASE® 100L, type EX (3.9 L). The mixture was stirred at 800 RPM for 1 min and KOH (2 M) was added to adjust the pH to 8.0. (R/S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (Formula 20, 100 kg) was added and the resulting mixture was titrated with NaOH aq (50%) under hydrolysis to maintain a pH of 8.0. The degree of reaction was monitors! with HPLC

(Ci8 column, 4.6 mm x 150 mm, detection at 200 nm). When a conversion of about 40-45% had been reached (e.g. after about 24 h), the reaction mixture was transferred to a separation funnel. The aqueous mixture was extracted with heptane (205 L). EtOH (absolute) was added (up to about 5% vol/vol) to break a light emulsion that had formed, and the aqueous and organic layers were separated. The extraction steps were repeated twice and the aqueous layer containing (3S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid potassium salt (Formula 23) can be further concentrated under vacuum (eg 25 to 50% of its original volume). The organic layers containing (R)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 22) were combined, dried and concentrated. The resulting diethyl ester was then racemized in accordance with Example 6. MS m/z [M+H]<+>227.<1>H NMR (300 MHz, D 2 O): δ 2.35 (dd, 6H), 2.70 ( t, 3H), 2.85 (m, 1H), 2.99 (m, 1H), 3.25 (m, 1H), 4.75 (m, 1H), 5.60 (q, 2H). <13>C NMR (75 ppm, D 2 O) δ 0172.19, 171.48, 122.85, 62.70, 59.49, 40.59, 31.83, 27.91, 23.94, 21, 74, 14,77.

EXAMPLE 3. Enzymatic resolution of (R/S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (Formula 20) to give (3S)-3-cyano-2-ethoxycarbonyl-5-methyl- hexanoic acid potassium salt (formula 23) and (R) 3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 22).

A reactor (3.92 L) equipped with overhead stirring is charged with calcium acetate buffer (1.47 L, 100 mM, pH 7.0) and ( R / S )-3-Cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 20.1 kg). The mixture is stirred at 1100 RPM for 5 min and KOH (5 M) is added to adjust the pH to 7.0. LIPOLASE® 100L, type EX (75 ml) is added and the resulting mixture is titrated with KOH (5 M) under hydrolysis to maintain a pH of 7.0. The degree of reaction is monitored with HPLC (Cis column,

4.6 mm x 150 mm, detection at 200 nm). When a conversion of about 42% to 45% was reached (eg after about 20-251), the reaction mixture was transferred to a separatory funnel. The aqueous mixture is extracted with hexane (100% vol/vol). EtOH (absolute) is added (up to about 5% vol/vol) to destroy a small emulsion that had formed, and the aqueous and organic layers are separated. The extraction steps are repeated twice to obtain an aqueous layer containing (3S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester potassium salt (formula 23) which can

used in subsequent conversions without insulation. The organic layers containing (R)-3-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 11) are combined, dried and concentrated. The resulting dimethyl ester is then racemized in accordance with Example 6.

EXAMPLE 4. Preparation of (S)-4-isobutyl-2-oxo-pyrrolidine-3-carboxylic acid (formula 10) from (3S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid potassium salt (formula 23 ).

A vessel was filled with an aqueous solution containing (3S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid potassium salt (Formula 23, 411 I from Example 2). Raney nickel (50% aq solution, Sigma-Aldrich) was added to the mixture, and hydrogen gas was introduced into the vessel over a 20 h period to maintain a pressure of 50 psig in the liquid vessel space throughout the reaction. The hydrogenation reaction was monitored by uptake of H2 and HPLC analysis (Cis column, 4.6 mm x 150 mm, detection at 200 nm) of the vessel contents. After the reaction, the aqueous mixture was filtered to remove the Raney Ni catalyst. The pH of the concentrated solution was adjusted to

3.0 using 37% HCI (about 14 I). The resulting solution was extracted three times with EtOAc (50% v/v). The combined organic layers were concentrated in vacuo to give (S)-4-isobutyl-2-oxo-pyrrolidine-3-carboxylic acid (Formula 10). MS m/z [M+H]<+>186.1130.<13>C NMR (75 ppm, CDCl3) 5 □175.67, 172.23, 54.09, 47.62, 43.69, 37 .22, 26.31, 23.34, 22.54. Yield 40-42%; 97% ee.

EXAMPLE 5. Preparation of pregabalin (formula 9) from (S)-4-isobutyl-2-oxo-pyrrolidine-3-carboxylic acid (formula 10).

A reaction vessel (60 L) was charged with (S)-4-isobutyl-2-oxo-pyrrolidine-3-carboxylic acid (Formula 10), HCl (36-38%, 30 L), and water (29 L). HOAc (1 L) was added to the solution and the resulting solid was treated for 36-381 at 80°C and for a further 61 at 110°C. The degree of reaction was monitored by HPLC (Ci8 column 4.6 mm x 150 mm, detection at 200 nm). Water and excess HCl were evaporated to give an oil, which was washed with MTBE (2x15 L). Water was added to the oil and the mixture was stirred until the solution became clear. The pH of the solution was adjusted to 5.2-5.5 using KOH (about 6 kg), which resulted in precipitation of pregabalin. The mixture was heated to 80°C and then cooled to 4°C. After 10 h, crystalline pregabalin was filtered and washed with IPA (12 L). The filtrate was concentrated under vacuum to give a residual oil. Water (7.5 L) and EtOH (5.0 L) were added to the residual oil and the resulting mixture was heated to 80°C and then cooled to 4°C. After 10 h, a second amount of pregabalin crystals was filtered and washed with IPA (1 L). The combined pregabalin crystals were dried in a vacuum oven at 45°C for 241. MS m/z [M+H]<+>160.1340. 1H NMR (300 MHz, D 2 O): δ 2.97 (dd, J = 5, 4, 12.9 Hz, 1H), 2.89 (dd, J = 6, 6, 12.9 Hz, 1H), 2.05-2.34 (m, 2H), 1.50-1.70 (sept, J = 6.9Hz, 1H), 1.17 (t, J = 7.0 Hz, 2H), 0, 85 (dd, J = 2.2, 6.6 Hz, 6H).<13>C NMR (75 ppm, D2O) 5 □ 181.54, 44.32, 41.28, 32.20, 24.94 , 22.55, 22.09. Yield 80-85%; ee > 99.5%.

EXAMPLE 6. Preparation of (R/S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 20) via racemization of (R)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid acid ethyl ester (formula 22)

A reactor was charged with (R)-3-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 22, 49.5 kg) and EtOH (250 L). Sodium ethoxide (21% w/w in EtOH, 79.0 L, 1.1 eq) was added to the mixture, which was heated to 80 °C in 201. After completion of the reaction, the mixture was allowed to cool to RT and was neutralized with addition of HOAc (12.2 L). After evaporation of the EtOH, MTBE (150 L) was added to the mixture, and the resulting solution was filtered and evaporated to give ( R / S )-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (Formula 20 ) in quantitative yield.

EXAMPLE 7. Preparation of (S)-3-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 24) from (3S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 21)

A 50 mL round bottom flask was charged with (3S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (Formula 21, 3.138 g, 13.79 mmol), NaCl (927 mg, 1.15 eq) , deionized water (477 µL, 1.92 eq) and DMSO (9.5 mL). The resulting mixture was heated to 88°C and maintained at this temperature for 17 h. A sample was taken from LC and LC/MS analysis, which showed the presence of the starting material (Formula 21) and the products (Formula 245 and Formula 25). The temperature of the mixture was then raised to 135°C and allowed to react for an additional 3.51. A second sample was taken for LC and LC/MS analysis, which showed the absence of the starting material (formula 21) and showed, in addition to the desired products (formula 24 and formula 25), the presence of unidentified by-products. (S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (Formula 24): 97.4% ee after 88°C; 97.5% ee after 135°C.

EXAMPLE 8. Determination of optical purity (ee) of (S)-4-isobutyl-2-oxo-pyrrolidine-3-carboxylic acid (formula 10).

The optical purity of (S)-4-isobutyl-2-oxo-pyrrolidine-3-carboxylic acid (formula 10) was determined via a derivatization method. A sample of (S)-4-isobutyl-2-oxo-pyrrolidine-3-carboxylic acid was esterified with EtOH in the presence of a catalytic amount of dry HCl in dioxane at 70°C. The resulting lactam ester was analyzed by HPLC (CHIRALPAK AD-H, 4.6 mm x 250 mm) using a mobile phase of hexane and EtOH (95:5), with a flow rate of 1.0 ml/min, injection volume of 10u.L, column temperature of 35°C, and detection at 200 nm.

EXAMPLE 9. Determination of optical purity (ee) of pregabalin (formula 9).

The optical purity of pregabalin was analyzed via a derivatization method. A sample of pregabalin was derivatized with Marfey's reagent (1-fluoro-2-4-dinitrophenyl-5-L-alanine amide) and then analyzed by HPLC (LUNA Cis(2) column,

0.46mm x 150mm, 3u.m) using a mobile vase of aqueous NaPO4

(20 nM, pH 2.0) and ACN (90:10 for 10 min, 10:90 for 3 min, 90:10 for 5 min), a flow rate of 1.2 mL/min, an injection volume of 10 µL, column temp. at 35°C, and detection at 200 nm.

EXAMPLE 10. Enzymatic resolution of (R/S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (Formula 20) to give (3S)-3-cyano-2-ethoxycarbonyl-5-methyl- hexanoic acid ethyl ester sodium salt (formula 23) and (R)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 22)

A reactor (16000 L) equipped with overhanging agitation is charged with calcium acetate (254 kg), deionized water (1892.7 kg) and LIPOZYME® TL 100 I (food grade LPOLASE®, 983.7 kg). After complete mixing, (R/S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (Formula 20, 9000 kg, 85% purity analysis) was charged and the mixture was stirred for 24 h. NaOH (2068 kg of a 30% solution) is added during the reaction time to maintain the pH at 7.0. The degree of reaction is monitored by HPLC (Cis column 4.6 mm x 150 mm, detection at 200 nm) When a conversion of about 42% to 45% was reached (e.g. after about 20-251) the titrator and stirring were stopped. The organic phase is immediately separated, and the aqueous phase is washed twice with toluene (780 kg). The aqueous layer containing (3S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester sodium salt (formula 23) is used in subsequent transformations (Example 11) without isolation. The organic layers containing (R)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 22) are combined and concentrated. 3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester. The resulting diethyl ester is then racemized in accordance with Example 6.

EXAMPLE 11. Preparation of (S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 24) from (3S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid sodium salt (formula 23)

A reactor (16000 L) equipped with overhead stirring is charged with a final aqueous solution from Example 10 (9698.6 L, containing (3S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid sodium salt (Formula 23), NaCl (630 kg) and toluene (900 L). The mixture is stirred for 2 h under refluxing conditions (75-85°C). Stirring is stopped; the organic phase is immediately separated and the aqueous phase is washed twice with toluene (900 L) The organic layers containing (S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 24) are combined and concentrated. The ethyl ester (formula 24) is then hydrolyzed in accordance with Example 12.

EXAMPLE 12. Preparation of (S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid potassium salt (formula 26) from (S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (formula 24).

A reactor (12000 L) equipped with overhanging stirring is charged with (S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid ethyl ester (Formula 24, 2196 L from Example 11). KOH (1795.2 kg, 45% solution, w/w) and H 2 O (693.9 kg) are added to the reaction mixture with vigorous stirring. The temperature is maintained at 25°C. After 4 h, the reaction mixture is transferred to a hydrogenation vessel (Example 13) without further work-up.

EXAMPLE 13. Preparation of pregabalin (formula 9) from (S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid potassium salt (formula 26).

A hydrogenator (12000 L) is charged with water (942.1 L) and with the reaction mixture from Example 12, containing (S)-3-cyano-2-ethoxycarbonyl-5-methyl-hexanoic acid potassium salt (Formula 26, 4122.9 IN). A Raney nickel suspension (219.6 kg, 50% w/w in H 2 O) is added. The hydrogenation is carried out below 50 psig at 35°C. After 61, the Raney nickel is filtered off and the resulting filtrate is transferred to a reactor, (16000 L) for crystallization. After addition of H 2 O (1098 I), the pH of the solution is adjusted to 7.0-7.5 using HOAc (864.7 kg). The resulting precipitate is filtered and washed once with H 2 O (549 L) and twice with IPA (2586 L each time). The solid is recrystallized with IPA (12296 L) and H 2 O (6148 L). The mixture is heated to 70°C and then cooled to 4°C. After 5-101 the crystalline solid is filtered, washed with IPA (5724 L), and dried in a vacuum oven at 45°C for 241 to give pregabalin as a white crystalline solid (1431 kg, 30.0% overall yield, 99.5 % purity and 99.75% ee).

It shall be understood that, as used in this specification and the accompanying claims, the singular forms such as "a", "an", and "the", "that", may refer to a single object or to a plurality of objects unless the context is clear indicates otherwise. Thus, for example, reference to a material containing "a compound" may include a single compound or two or more compounds. It should be understood that the above description is intended to be illustrative and not limiting. Many embodiments will be obvious to those skilled in the art upon reading the description. Therefore, the scope of the invention shall be determined by reference to the accompanying patent claims and include equivalents of such embodiments. The descriptions of all articles and references, including patents, patent applications and publications, are incorporated herein by reference in their entirety for all purposes.

Claims (2)

1. Process for the preparation of a compound of formula 1, or a pharmaceutically acceptable complex, salt, solvate or hydrate thereof, R<1> and R2 are different and are each independently selected from hydrogen atom, C1-12 alkyl, C3-12 cycloalkyl, and substituted C3-12 cycloalkyl, characterized in that the method comprises: (a) reducing a cyano unit of a compound of formula 8, or a salt thereof to give the compound of formula 1, or a salt thereof, and (b) optionally converting the compound of formula 1 or a salt thereof into a pharmaceutically acceptable complex, salt, solvate or hydrate thereof, wherein R<1> and R2 in formula 8 are as defined in formula 1, and R<5> in formula 8 is hydrogen atom, C1-12 alkyl, C3-12 cycloalkyl, or aryl-C1-12 alkyl. 2. Process for the preparation of a compound of formula 9, or a pharmaceutically acceptable complex, salt, solvate or hydrate thereof, characterized in that the method comprises: (a) reducing a cyano unit of a compound of formula 16, or a salt thereof, to give a compound of formula 9 or a salt thereof; (b) optionally converting the compound of formula 9 or a salt thereof into a pharmaceutically acceptable complex, salt, solvate or hydrate; wherein the compound of formula 16 is optionally prepared by decarboxylation of a compound of formula 11, or a salt thereof; and where R <3> in formula 11 is C1-12 alkyl, C3-12 cycloalkyl, or aryl-Ci-6 alkyl, and R <5> in formula 16 is hydrogen atom, C1-12 alkyl, C3-12 cycloalkyl, or aryl -C 1-6 alkyl.