WO1991019002A1

WO1991019002A1 - A process for chiral enrichment of asymmetric primary amines

Info

Publication number: WO1991019002A1
Application number: PCT/DK1990/000134
Authority: WO
Inventors: Stig Aasmul-Olsen; Søren Hansen
Original assignee: Carlbiotech Ltd. A/S; Chemical Dynamics Development; Chiron Laboratories A.S
Priority date: 1990-06-01
Filing date: 1990-06-01
Publication date: 1991-12-12

Abstract

Chiral enrichment of a mixture of enantiomers of an asymmetric primary amine by stereoselective transacylation catalyzed by a subtilisin-like enzyme in a substantially anhydrous organic medium using a simple alkyl ester of an aliphatic alkanoic acid as the acyl donor. It is possible to achieve a high degree of conversion with a high stereoselectivity so that the formed amid and/or the remaining amine exhibits a very high enantiomeric excess. An excess of the acyl donor may be used as the organic reaction medium or part thereof; this makes possible the use of high concentrations of amine and acyl donor to obtain a high concentration of the chirally enriched product which makes this process well suited for industrial scale application.

Description

A process for chiral enrichment of asymmetric primary amines

This invention relates to a process for chiral enrichment of a mixture of enantiomers of an asymmetric primary amine.

Chemical compounds containing an asymmetric center occur in one or the other of the two possible mirror image forms. These mirror images, enantiomers, have different reactivity in chiral surroundings. Consequently, living organisms exhibit chiral discrimination in their inter¬ actions with chiral compounds, i.e. if a chiral molecule is biologically active, then the desired activity will predominantly reside in one of the enantiomers.

For the preparation of pure enantiomers several distinct and basic methods are currently in use. These methods can be conveniently divided into three groups on the basis of the type of starting materials used: direct synthesis from the chiral pool; catalytic asymmetric synthesis from pro- chiral compounds; and racemate resolution via crystalli¬ zation or kinetic resolution. The crystallization method can again be divided into: (1) Diastereomer crystalliza¬ tion; and (2) Preferential crystallization, including crystallization induced asymmetric transformation. See, for example, Bruggink, A., Hulshof, L.A., and Sheldon, R.A. : "Industrial scale resolutions of racemates" in Pharmaceutical Manufacturing International 1990, publ. by Sterling Publications International Limited.

A group of chiral compounds which are of special interest for use as intermediates in the preparation of drugs and agrochemicals is the one consisting of primary aralkyl- amines branched in the 1-position of the alkyl group. For example, 1-phenylethylamine and 1-methyl-3-phenylpropyl- amine are intermediates in the preparation of N-phenyl- alkyl-anthranilic acid compounds with activity as chloride channel blockers; see Richardson, S. et al.: "The stereo- selective synthesis and analysis of two potential chloride channel blockers". Poster 17 in Abstracts of The Smith Kline & French Research Symposium on Chirality in Drug Design and Synthesis, Robinson College, Cambridge, U.K., 27th-28th March 1990. l-methyl-3-phenylpropylamine is also useful as an intermediate in the preparation of other drugs and agrochemicals; see JP patent applications Kokai Nos. 63-185943-A, 63-188655-A and 63-237796-A. The most interesting enantiomer is the R-isomer, which is i.a. a valuable intermediate in the preparation of (-)-5-[(R)-l- hydroxy-2-((R)-l-methyl-3-phenylpropylamino)ethyl]salicyl- amide, generic name (R,R)-labetalol, purchased by Schering Corp. under the trade name "Dilevalol©", which is a 0- adrenergic blocker and is useful as an anti-hypertensive drug.

The racemic amines, l-methyl-3-phenylpropylamine and 1- phenylethylamine, are commercial products which may be purchased i.a. from BASF and Fluka.

The interesting (R)-isomers are prepared either by chiral synthesis or by diastereomer crystallization from the ra- cemate. (R)-l-methyl-3-phenylpropylamine may also be pro¬ duced by contacting the racemate with microorganisms which decompose the (S)-isomer, see JP Kokai 63-237796-A. How- ever, chiral synthesis is a cumbersome multi-step process for which also the chiral starting material must be pro¬ vided; resolution by crystallization is often accompanied by heavy losses of material; and predominant fermentation of one enantiomer requires the use of specially selected strains of microorganisms, which must be kept genetically pure, and entails a complete loss of the fermented enan- tiomer, and also fermentations are performed with rather low concentrations of substrate thus requiring the hand¬ ling of very large volumes.

In the last few years it has been discovered that many en¬ zymes exert activity in nearly anhydrous organic solvents. In a poster by A. M. Klibanov: "Enzymatic Processes in Organic Media", International Conference on Enzyme Engi¬ neering, 25th/26th September 1986, The Cambridge Series on Biotechnology, the author reviews a number of articles on this topic and i.a. says: "Upon replacement of water as the reaction medium with organic solvents, enzymes acquire remarkable novel properties, e.g. they become extremely thermostable and show different substrate specificity. Many enzymatic reactions, while not feasible in water, readily take place in organic solvents".

In an article by V. Gotor, R. Brieva and F. Rebolledo: "Enantioselective Acylation of Amino Alcohols by Porcine Pancreatic Lipase", J. Chem. Soc, Chem. Commun. , 1988, it is said that upases are the most common enzymes acting as catalysts in anhydrous organic solvents and have been widely used for the resolution of racemic alcohols, carbo- xylic acids and esters via enzymatic transesterification. The authors then describe the enantioselective amide for¬ mation and esterification of 2-aminobutan-2-ol and amide formation from l-aminopropan-2-ol with >95% enantiomeric excess using ethylacetate as the acyl donor and solvent and porcine pancreatic lipase as the catalyst. However, attempts at using a corresponding method for the enantio¬ selective transacylation of chiral hindered primary amines have shown that under conditions where acylation is ob¬ tained the stereoselectivity is nil (see Reference Example 1). It is known i.a. from Martin Ottesen and lb Svendsen: "The Subtilisins", Methods in Enzymology, Vol. XIX, page 212, that the subtilisins are capable of hydrolyzing both peptide bonds and ester bonds and are effective in trans- esterification reactions and transpeptidation reactions, and further that subtilisin Carlsberg hydrolyzes even the methyl esters of simple aliphatic carboxylic acids. But these references apply to the conduct of the enzyme in aqueous solutions and from that no conclusions can be drawn as to its function in non-aqueous organic media and even less about its possible stereoselectivity.

N. Chinsky, A.L. Margolin and A.M. Klibanov, "Chemoselec- tive Enzymatic Monoacylation of Bifunctional Compounds", J. Am. Chem. Soc., 1989, 111, 386-388, report experiments with transacylating bifunctional compounds using specifi¬ cally activated acylating agents in the form of chloro- ethyl esters. They transacylated the bifunctional compound 6-amino-l-hexanol using various chloroethyl esters as acyl donors and lipase, subtilisin Carlsberg or subtilisin BNP' as catalysts and tert-amyl alcohol as the solvent, and they found that the chemoselectivity of acylation., i.e. the preference for acylating either the hydroxy group or the amino group, varies with the kind of enzyme used as well as with the acyl moiety. This article suggests the need for halogenactivated acyl donors when using sub¬ tilisin for transacylation in organic media. There is, however, no mention of stereoselectivity in this article.

A.L. Margolin and A.M. Klibanov in a review: "Enzymatic synthesis of peptides in anhydrous organic solvents", Peptides - Chemistry, Structure and Biology, Proc. of the eleventh American Peptide Symposium, July 9/14, 1989, La Jolla, California, USA, Ed. by Jean E. Rivier and Garland . Marshall, ESCOM, Leiden 1990, pages 355-359, i.a. refer to some experiments using anhydrous organic solvents as a reaction medium and the microbial protease subtilisin as a catalyst. They observed that, upon a transition from water to organic solvents, the enantioselectivity of several proteases dramatically relaxes. They measured the ratio of specificity constants ^(k _oat ^K _ta ⁾ _L/^(k _ca._t ^K _ta ⁾ _D/ ^{which re}~ flects enantioselectivity of an enzyme, in water and in several organic solvents. The enzymatic reaction studied in water was the hydrolysis of 2-chloroethyl esters of N- acetyl-L- and -D-amino acids. In organic solvents they examined the enzymatic transesterification reaction be¬ tween the same esters and propanol- For all non-aqueous solvents tested, this enantioselectivity factor was found to be 10- to 100-fold lower. Later, they investigated enantioselectivity of five other serine proteases in water and in butyl ether and found that elastase, -lytic pro¬ tease, subtilisin BPN', o-chymotrypsin and trypsin all ex¬ hibit striking enantioselectivity in water, but not in organic solvents.

H. Kitaguchi, P.A. Fitzpatrick, J.E. Huber and A.M. Kli¬ banov: "Enzymatic Resolution of Racemic Amines: Crucial Role of the Solvent", J. Am. Chem. Soc. 1989, 111, 3094- 3095, set out to find a stereoselective enzyme which would catalyze the reaction between carboxylic acid esters and chiral amines in anhydrous organic solvents. With this ob¬ ject they tested the protease subtilisin Carlsberg and various upases as catalysts of the reaction between tri- fluoroethyl butyrate and the enantiomers of α-methylben- zylamine in octane, but for no enzyme was any substantial enantioselectivity observed. They then proceeded to exa¬ mine the dependence of the enantiomer conversion rates v_ς, v_, and their ratio (the enantioselectivity factor) on the reaction medium for subtilisin Carlsberg and found that subtilisin's enantioselectivity factor is a strong func- tion of the solvent. In particular they found a marked enantioselectivity of subtilisin in 3-methyl-3-pentanol. These authors worked with low concentrations of racemic amine (0.2 to 0.5 M) and of acyl donor (0.2 to 0.56 M) obtaining a very low concentration of product, and they found it necessary to use a strongly activated acyl donor namely trifluoroethyl butyrate. However, the manufacture of this activated ester is difficult (and expensive) and, moreover, the compound is unstable and per se react with the amine (without stereoselectivity). Further, it has not been possible to duplicate the results reported in the article (see Reference example 3). Attempts to acylate R,S-1-methyl-3-phenylpropylamine (0.2 M) with trifluoro¬ ethyl butyrate (0.4 M) in 3-methyl-3-pentanol as a solvent catalysed by crystalline subtilisin A (subtilisin Carls¬ berg), using 10 mg/ml of enzyme at 30^CC, gave a conversion of only 27% after 11 days and of 48% after 21 days. When using a subtilisin A preactivated with tryptophanamide in a known manner, the rate of reaction was increased, but the enantiomeric excess was only 53%. In all experiments employing the halogen-activated acyl donor a substantial amount of spontaneous (non-selective) conversion was observed. This amount would be even higher, if higher concentrations of this type of donor were to be employed, thereby further reducing the enantiomeric excess.

Now, surprisingly^ it has been found that it is possible to achieve a chiral enrichment of a mixture of enantiomers of an asymmetric primary amine by stereoselective trans- acylation catalyzed by a subtilisin-like enzyme in a sub¬ stantially anhydrous organic medium using a simple alkyl ester of an aliphatic alkanoic acid as the acyl donor.

This is contrary to what would have been expected from the prior art, and hereby i.a. the following advantages are obtained: 1) The simple esters are cheap, easy to handle and non- toxic to humans. They are chemically stable and do not tend to react spontaneously with the amine or the enzyme.

2) The enzyme retains most of its activity; it may simply be filtered off from the reaction mixture and reused in the process.

3) It is possible to achieve a high degree of conversion with a high stereoselectivity so that the formed amide and/or the remaining amine exhibits-a very high enantio¬ meric excess.

4) An excess of the acyl donor may be used as the organic reaction medium or part thereof; this makes possible the use of high concentrations of amine and acyl donor to ob¬ tain a high concentration of the chirally enriched product which makes this process well suited for industrial scale application.

5) The loss of valuable amine is much less in this process than in the conventional diastereomer crystallization.

The process according to the invention is characterized by subjecting the amine to a stereoselective transacylation with a simple (C₁-C₁₀)alkyl ester of an aliphatic (C.- C₂₂)alkanoic acid as the acyl donor catalyzed by a sub- tilisin-like enzyme in an organic medium containing not more than 8% by weight, preferably not more than 3% by weight, and most preferably not more than 1% by weight, of water at a temperature of from 10 to 80°C, preferably from 25 to 60°C, and most preferably from 30 to 40°C, until a substantial amount of one of the enantiomers has been con¬ verted to the amide. Hereby it is possible to obtain a high conversion of sub¬ stantially only one of the amine enantiomers to the amide, obtaining a high enantiomeric excess of the amide of that enanτ-iomer in the produced amide and a high enantiomeric excess of the other enantiomer in the remaining amine.

Due to the fact that all or part of the organic medium may be made up of an excess of the acyl donor, it is possible to use high concentrations of the amine and the acyl donor in the reaction mixture thus securing a high throughput in the process and a high yield of the chirally enriched amine.

If a co-solvent is used, it is preferably one which en- hances the velocity and/or stereoselectivity of the trans- acylation reaction. The best results have been obtained by using pyridine as the co-solvent.

Generally, satisfactory results are obtained when the con- centration of the amine in the reaction mixture is from 0.1 to 3.0 M, preferably from 0.5 to 2.0 M, and the con¬ centration of the acyl donor is from 0.5 to 8.0 M, prefer¬ ably from 2.0 to 4.5 M. The best results are obtained when the acyl donor is used in an excess concentration of at least 1 M in relation to the amine.

Conveniently, the concentration of the subtilisin-like enzyme in the reaction mixture is from 0.01 to 5.0 mM based on enzyme protein content.

Generally, the organic medium should be anhydrous, i.e. it should contain as little water as possible barring the water molecules chemically bound to the enzyme and neces¬ sary for its function. In practice, no conversion is seen, when the organic medium contains more than 8% by weight of water, and preferably the water content of the medium does not exceed 3% by weight, and most preferably 1% by weight.

The reaction temperature may be varied in the range from 10 to 80 °C depending on the activity of the particular enzyme used, the reactivities of the particular amine and ester, and the composition of the organic medium, i.e. the nature and amount of the co-solvent, if any. Preferably the reaction temperature will be in the range from 25 to 60 °C and most preferably in the range from 30 to 40 °C.

The remaining chirally enriched amine- may be isolated from the reaction mixture, for example by extraction, by frac¬ tional distillation, by ion exchange or liquid chormato- graphy or by precipitation of an amine salt.

After filtering off the enzyme and any precipitated ma¬ terial the amine may be extracted with water after acidi¬ fication of the reaction mixture and then after basifica- tion of the aqueous phase reextracted with an organic sol- vent such as dichloromethane. The amine is isolated by evaporation of the organic solvent.

If further chiral enrichment is desired, the isolated amine may be subjected to another stereoselective trans- acylation by the process of the invention, or it may be subjected to optical resolution by crystallization tech¬ nique. Such optical resolution may also be done by cry¬ stallization directly from the reaction mixture after filtering off the enzyme and precipitants. As the resolv- ing agent, for example, a chiral form of tartaric acid or a derivative thereof or a chiral amino acid or a deriva¬ tive thereof may be used.

The other enantiomer of the amine may be obtained by iso- lating the formed amide from the reaction mixture and hydrolyzing it to obtain the amine. In another embodiment the isolated amide is subjected to racemization and hydro¬ lysis to obtain the racemic amine for reuse in the pro¬ cess.

The asymmetric primary amine to be chirally enriched by the process of the invention may be any primary amine having at least one asymmetric carbon atom. Particularly interesting are the substituted or unsubstituted alkyl- amines which are branched in the 1-position and the sub- stituted or unsubstituted aralkylamines. The most inter¬ esting amines are the primary aryl-(C₂-C₁₄)alkylamines branched at the 1-position of the alkyl moiety. Examples of specific amines which may advantageously be chirally enriched by the process of the invention are l-methyl-3- phenylpropylamine, l-phenyl-(C.-C_fi)alkylamines, asymmetric amino acids and the amides and branched alkyl esters of said amino acids.

A unique feature of this invention is that the acyl donor is a simple (C.-C-_Q)alkyl ester of an aliphatic (^cι~^c 2^~ alkanoic acid, and preferably a simple straight chain (C.,-

Cg)alkyl ester of a (C,-^)alkanoic acid. Examples of specific acyl donors which may be used in the process of the invention are ethylacetate, propylacetate, isopropyl- acetate, butylacetate, ethylpropionate, propylpropionate, isopropylpropionate, butylpropionate, ethylbutyrate, pro- pylbutyrate, isopropylbutyrate, butylbutyrate, ethylisobu- tyrate, isopropylisobutyrate, ethylcaproate, propylcap- roate and ethylcaprylate. Especially good results have been obtained by using ethylbutyrate or butylbutyrate as the acyl donor.

The subtilisin-like enzyme to be used as a catalyst in the process of the invention is generally a serine endopro- tease of microbial origin having a molecular weight be¬ tween 15000 and 35000 and a neutral to alkaline pH- optimum. Examples of specific enzymes which may be used with good results are subtilisin A ( "Subtilisin Carls¬ berg"), subtilisin B ("Subtilisin Novo"), subtilisin BPN' and some related enzymes sold in crude form under the trade names of the manufacturer, such as "Alcalase"®, "Esperase"®, "Savinase"® and "Nagarse"®.

In some cases it may be advantageous that the enzyme has been covalently modified, for example by derivatization with glutaric dialdehyde and subsequent reaction with an amine. It may also be advantageous to immobilize the en¬ zyme.

Some enzymes may need to be preactivated by drying from an aqueous solution in the presence of a suitable ligand or inhibitor, if necessary followed by washing out ligand or inhibitor with an organic solvent, in order to give opti¬ mum results. For example, crystalline subtilisin A usually gives better results when preactivated, while "Esperase"® do not need to be preactivated.

"Esperase"® is the enzyme of choice for use in the process of the invention giving both a good conversion rate and a high enantioselectivity (% enantiomeric excess).

A great advantage in the process of the invention is that the enzyme may be isolated from the reaction mixture by simple filtration and reused in the process with good re¬ sults.

The alcohol formed from the acyl donor by the transacyla¬ tion tends to inhibit the activity of the enzyme. There¬ fore, it may be advantageous to perform the transacylation reaction in the presence of a molecular sieve which binds "the alcohol formed from the acyl donor. As both the amide and the alcohol product tend to inhibit the activity of the enzyme it may in some cases be advan¬ tageous to interrupt the reaction before the desired enan¬ tiomeric excess has been reached, isolate the remaining amine from the reaction mixture and subject it to another stereoselective transacylation by the process of the in¬ vention.

The process of the invention and the results obtained thereby will be further explained and illustrated in the following detailed description with Examples and Reference examples.

General Method for Examples

The reactions performed on an analytical scale with a reaction volume of 1, 2 or 5 ml were carried out with magnetic stirring or automatic shaking in a closed reaction glass or polypropylene vessels at different fixed temperatures. As reference standard was used 35°C unless otherwise indicated.

The amine was dissolved in the dry organic solvent mixed with dry acyl donor and the reaction initiated by addition of a dry preparation of the enzyme to give a suspension.

The tables also include concentrations, content of organic solvent, product yield in terms of combined R and S amine conversion to amide and the fixed temperature, controlled by thermostated stirred water or oil bath or by using a Thermo ixer 5437 (Eppendorf) with automatic shaking.

Reaction times are typically between 1 - 3 weeks and enzyme concentrations are 1.0 - 1.5 mM based on crude added weight unless otherwise stated.

Product identification and determination of product yield including enantiomeric excess (ee) were performed by means of reverse phase or chiral HPLC (Waters 6000 A pumps, 660 gradient blender, UK6 injector or Wisp model 712) on a 4 μm C₁O₀ NovaPak (8 x 100 mm, RCM, Waters), Lichro Cart

(250 x 4 mm, Merck) or Chiral 5 μm, α.-AGP (Chromtech AB, Sweden, 100 x 4 mm) column, for reverse phase or chiral HPLC respectively.

Elution systems used in reverse phase HPLC were gradients containing 50 mM triethyl ammonium phosphate (TEAP) pH 3 and from 0% to 80% acetonitrile with a flow of 1 or 2 ml/min. Elution was monitored normally at 254 nm or at 336 nm in the case of diastereomer seperation. Elution systems used in chiral HPLC were 10 mM NaH₂P0₄ buffer pH 7.0 containing 9 to 11% isopropanol according to Her ansson, J. , in Proceedings of The Thirteenth Symposium on Column Liquid Chro atography, (1989), p. 33, CLC 89, Stockholm. Elution was monitored at 254 nm.

The product (R-amine) was identified by HPLC comparison with a reference compound, chemically synthesized and resolved, following diastereomeric derivatization, as were the R and S amides in underivatized forms.

Enantiomeric excess was determined in two ways: By derivatization of the enantiomeric pair of amines with Marfey's reagent l-fluoro-2,4-dinitrophenyl-5-L- alanineamide, according to Marfey, P; Carlsberg Res. Commun. , 4£, (1984), 591-96, and separation of the formed diastereomeric pair by RP-HPLC on an analytical column or by direct seperation of the formed amides on the analytical chiral AGP column.

It was then calculated on basis of the peak areas for the two enantiomers or diastereomers according to the formulae

l^ArβaS amide ^{" Area}R amideI ^eeamide " ^{x 100%}

<^AreaS amide ^{+ AreΛ}R amide>

and

l^AreaS amine ^{" Area}R amineI ^eeamine = ^{x 100%}

^^AreΛS amine ^{+ Area}R amine>

Both equations were used during process monitoring depending on the analytical method applied.

Conversion was determined on the basis of the ratio between the integrated areas below the peaks in the HPLC elution chromatogram corresponding to the amide respectively the amine phenyl group which absorbs at 254 nm.

The reaction conditions in the preparative examples are described in the individual examples. The reaction was followed on analytical HPLC as described. No attempt has been made to optimize the work-up procedures used with regard to yield.

Solvents and acyl donors used were standard reagent grades obtained from Merck and Fluka, molecular sieves were obtained from Schweizerhall, USA, and enzyme preparations were obtained as crystalline or crude industrial preparations from Sigma, USA, A ano Pharmaceutical Corporation, Japan, Miles Corporation, USA, Tanabe, Japan, Novo-Nordic Industries, Denmark, Enzymatix, Great Britain, and the Nagase Company, Japan. The enzymes were then typically pretreated in the following manner: Following partial dissolution to 0.5 - 1.0% (w/w) in 3% DMF, undissolved material was filtered off and the filtrate taken to dryness either by lyophilisation or evaporation under reduced pressure.

Alcalase R , SavinaseR and EsperaseR are all registered trademarks belonging to the manufacturer for the industrial preparations used, being of grades 3.0T, 6.0T and 6.0T, respectively. Consequently, these contain less than 20% (w/w) active enzyme, the balance being salts and coating substances. In accordance with the product sheets of the manufacturer, all three enzymes are classified by the enzyme classification number EC 3.4.21.14, the first as Subtilisin and the other two as alkaline bacillus proteinases (CAS number 9073-77-2) . The Subtilisin enzymes present as active ingredients in these three industrial preparations and the baccillus strains used for producing them are thoroughly described in US Patent 3,723,250 (1973), which is hereby incorporated by reference. Particularly useful strains for the production of said enzymes of the genus Bacillus have been deposited in The National Collection of Industrial Bacteria, Torry Research Station, Aberdeen, Scotland by the enzyme manufacturer. These strains are designated by deposition numbers with the prefix NCIB. The Subtilisin enzymes were assayed for activity in aqueous solution using BzArgOEt as substrate according to Ottesen, M. and Svendsen, I., in Methods of Enzymology (Eds. Pearlman, G.E. and Lorand, L.), 19_, p. 199-215, Academic Press (1970).

Raσemic (R,S)-l-methyl-3-phenylpropylamine was obtained from BASF, W. Germany. Other amines in racemic form were obtained from Fluka or Sigma.

Reference example 1

Attempts to acylate R,S-l-methyl-3-phenylpropylamine (0.5 M) stereoselectively using butylbutyrate (3.0 M) as acyl donor and solvent and 3-methyl-3-pentanol or Pyridine as co-solvent (50% v/v) at 60°C in a process catalysed by a variety of lipases of animal, fungi and bacterial origin.

Lip ^case from Co-solvent Conversion(%) '^a^ eea_mi.ne(%) '

a) Determined at 28 days in 3M3P and 12 days in Pyridine.

b) N-butyric acid-l-methyl-3-phenylpropylamide, R and S.

Reference example 2

Attempts to acylate R,S-l-methyl-3-phenylpropylamine (1.0 M) using butylbutyrate (3.0 M) as acyl donor and solvent with 50% Pyridine (v/v) as a co-solvent at 40°C in a conversion catalysed by various dry plant thiolproteases (3-4 mM) .

Enzy ^Jme Time (days) Conversion (%) ee a_mi. ne (%)

Ficin 1 JD 0

14 0 0

Bromelain 1 0 0 14 0 0

Crude papain 1 0 0

14 0 0

Reference example 3

Attempts to acylate R,S-l-methyl-3-phenylpropylamine (0.2 M) and butylbutyrate or trifluoroethylbutyrate (0.4 M) as acyl donors and 3-methyl-3-pentanol as a solvent in a process catalysed by crystalline Subtilisin A or Subtilisin A preactivated with tryptophan amide according to Russell, A.J. and Klibanov, A.M., J.Biol.Chem. , vol. 263, p. 11624, (1988), using 10 mg/ml of enzyme at 30°C.

Time

Enzyme Acyldonor (days) C nv. (%)^a' ^ee _amide^

Subtilisin A Butylbutyrate 40 0 0^d)

Subtilisin A Trifluoro- 11 27 57 ethylbutyrate

Subtilisin A Trifluoro- 21 48 58 ethylbutyrate

. . . c Subtilisin A ' Trifluoro- 13 39 44 ethylbutyrate '

Preactivated Trifluoro- 12 49 53 Subtilisin A ethylbutyrate

a) Including spontaneous conversion of 5-25%.

b) N-butyric acid-l-methyl-3-phenylpropylamine, R and S.

c) 10 mg of 3A molecular sieve added to remove water.

d) eeamine Example 1: Preparative Two-Step Enzymatical Resolution of R,S-l-methyl-3-phenylpropylamine by Butylbuty¬ rate Conversion Catalysed by Esperase.

Preparative Synthesis of R-l-methyl-3-phenylpropylamine (1st step)

Procedure

83 ml (506 mmol) R,S-l-methyl-3-phenylpropylamine was dissolved in 125 ml Pyridine and 292 ml (1757 mmol) butylbutyrate. The reaction was initiated by addition of 20.7 g Esperase (1.5 mM) and kept for 19 days in an oil bath at 35°C under magnetic stirring in a closed vessel.

The reaction mixture was then filtrated for enzyme and precipitates. 300 ml H_0 was added and pH in the aqueous phase adjusted to 3.4 by adding of 180 ml of 32% hydrochloric acid. Extraction of the aqueous phase 2 times with 500 ml of ethylacetate. 300 ml of 6 N sodiumhydroxide was then added to the aqueous phase (pH 13.3) , which was extracted with 500 ml dichloromethane, followed by drying of the organic phase with MgSO. , filtration and removal of the organic solvents by evaporation under reduced pressure at 45°C to give 40 ml (52% RS corresponding to 83% R, 264 mmol) yellowish oil of the R-l-methyl-3-phenylpropylamine free base.

Purity

HPLC purity (254 nm) 80% R-l-methyl-3-phenylpropylamine, 20% S-l-methyl-3-phenylpropylamine (^eeami_de ^{= ηo}%) (determined by diastereomeric separation on a Merck Lichro Cart 250 x 4 mm column using a CH_CN gradient in aqueous TEAP buffer, pH 3, following derivatization with Marfey's reagent) . Preparative Synthesis of R-l-methyl-3-phenylpropylamine (2. step)

40 ml (244 mmol) R-l-methyl-3-phenylpropylamine (including 20% S-amine) was dissolved in 73 ml pyridine and 186 ml butylbutyrate. The reaction was initiated by addition of 7.63 g Esperase and kept for 8 days in an oil bath thermostated to 35°C under magnetic stirring in a closed vessel.

The reaction mixture was filtrated for enzyme and precipitates. 200 ml H_0 was added, followed by acidification of the reaction mixture to pH 3.4 with 100 ml of 32% hydrochloric acid. After the separation, the aqueous phase was extracted 2 times with 200 ml of ethylacetate.

To the aqueous phase was added 250 ml of 6 N sodiumhydroxide to pH 13.3, followed by extraction with 250 ml dichloromethane, drying of the organic phase with MgSO., filtration and removal of the organic solvent by evaporation under reduced pressure at 45°C to give 30.7 ml (77% RS corresponding to 71% R) of a yellowish oil of R-l- methyl-3-phenylpropylmine free base after redissolution in ethanol and repeated evaporation under reduced pressure.

Purity

HPLC purity (254 nm) : 93% R-l-methyl-3-phenylpropylaraine and 7% S-l-methyl-3-phenylpropylamine (determined by diastereomeric separation on a Merck Lichro Cart 250 x 4 mm column using a CH_CN gradient in aqueous TEAP buffer, pH 3, following derivatization with Marfey's reagent).

No amide could be detected by reverse phase HPLC. GC assays for solvents used or formed: Pyridine, butylbutyrate, butyric acid, methylene chloride and ethanol, showed a content of 3% (w/w) of the last and only trace amounts of others so the preparation was essentially free of other solvents.

Example 2

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (0.5 M) catalysed by Esperase ' and Subtilisin A_*^*">)^y using butylbutyrate as acyl donor and solvent and 34% (v/v) Pyridine as co-solvent at different temperatures.

Enzyme Temperature(°C) e)

Esperasec) 25 Esperasec) 30 Esperasec) 35

Esperase ,c) 40 Subtilisin A^d^ 50 Subtilisin A^d) 60 Subtilisin ^d) 70

a) Reaction parameters: 3.0 M butylbutyrate, 0.36 mM Subtilisin A.

b) Reaction parameters: 3.5 M butylbutyrate, 1 mM Esperase.

c) Determined after 16 days.

d) Determined after 9 days.

e) N-butyric acid-l-methyl-3-phenylpropylamide, R and S. Example 3

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (0.9 M) catalysed by Esperase (1.44 mM) using ethylbutyrate and butylbutyrate (2.6 M) as acyl donors and solvents and 43% (v/v) of different organic co- solvents at 35°C.

Solvent Substrate Conversion(%) ee . , (%)

_aλ Hexane ' Ethylbutyrate 23 83

Tetrahydrofurane 3.)' Butylbutyrate 12 86 Pyridinec) ' Ethylbutyrate 31 92 Pyridine ' Butylbutyrate 36 89

a) Determined after 10 days of reaction.

b) N-butyric acid-l-methyl-3-phenylpropylamide, R and S.

c) Determined after 13 days of reaction.

Example 4

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (0.9 M) catalysed by different enzymes (3.1 mM) using butylbutyrate (2.6 M) as acyl donor and solvent and 43% (v/v) Pyridine as a co-solvent at 50°C.

Enzyme Conversion(%) ' ^eeami_de^

62 76 67 84

70

a) Determined after 10 days of reaction.

b) These reactions were performed under other circumstan¬ ces:

3.5 M butylbutyrate, 1.25 M R,S-l-methyl-3- phenylpropylamine, 21% (v/v) Pyridine and 0.4 mM enzyme at 35°C, measured after 24 days.

c) N-butyric acid-l-methyl-3-phenylpropylamide, R and S.

Example 5

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (0.5 M) catalysed by preactivated Subtilisin Ac) ' (0.36 mM) with different ligands, using butylbutyrate (2.8 M) as acyl donor and solvent and 46%

(v/v) Pyridine as a co-solvent at 50°C.

Ligand

N-acetyl-L-Tryptophane amide .49 67

Butyric acid amide 42 76 N-butyric acid- R,S-l-methyl-3-phenylpropylamide 42 63

None^b) 17 63

a) The reactions were tested after 13 days.

b) This reaction was performed without preactivation of Subtilisin A.

c) Preactivation of Subtilisin A was performed by freeze- drying of an aqueous solution (0.01 M phosphate buffer, pH 7.8, containing 3% (v/v) dimethylforma ide to enhance ligand solubility) of Subtilisin A (5 mg/ml) in the presence of ligand (10 mM) . After freeze-drying, the enzyme ligand complex was washed twice with acetonitrile (0.05 ml/mg complex) to removed the added ligand.

d) N-butyric acid-l-methyl-3-phenylpropylamide, R and S. Example 6: Preparative One-Step Enzymatical Resolution of R,S-l-methyl-3-phenylpropylamine by Butylbuty¬ rate Conversion Catalysed by Esperase in Pre¬ sence of Molecular Sieve.

Procedure

83 ml (506.1 mmol) R,S-l-methyl-3-phenylpropylamine was dissolved in 125 ml Pyridine and 292 ml (1757.4 mmol) butylbutyrate. The reaction was initiated by addition of 20.7 g Esperase (1.5 mM) and 5.0 g 5A molecular sieve (1/16' pellets) and kept for 28 days in an oil bath at 35°C under magnetic stirring in a closed vessel.

The reaction mixture was then filtrated for enzyme, precipitates and molecular sieve. 300 ml H_0 was added and pH in the reaction mixture adjusted to 3.4 with 180 ml of 32% hydrochloric acid. The aqueous phase was then extracted two times with 500 ml of ethyl acetate. 300 ml of 6 N sodium hydroxide was then added to the aqueous phase (pH 13.3), which was extracted with 500 ml dichloromethane, followed by drying of the organic phase with MgSO., filtration and removal of the organic solvents by evaporation under reduced pressure at 45°C to give 44 ml (56% RS corresponding to 79% R, 268.3 mmol) yellowish oil of the R,S-l-methyl-3-phenylpropylamine free base.

Purity

HPLC purity (254 n ) 85% R-l-methyl-3-phenylpropylamine, 15% S-l-methyl-3-phenylpropylamine (^ee _aιnidp ⁼ 88%) (determined by diastereomeric seperation on a Merck Lichro Cart 250 x 4 mm column using a CH CN gradient in aqueous TEAP buffer, pH 3, following derivatization with Marfey's reagent) . Example 7

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (1 M) catalysed by Esperase (1 mM) using butylbutyrate (3.0 M) as acyl donor and solvent and different organic mixtures employing alkanols as co- solvents at 40°C.

Co-solvent mixture(% v/v/% v/v) Conversion(%) ^ee _aniia_e(*)

a) Determined after 11 days.

b) Determined after 15 days.

c) Determined after 29 days.

d) N-butyric acid-l-methyl-3-phenylpropylamide, R and S.

e) 0.5 M amine was used. Determined after 20 days.

f) 0.5 M amine and 4.6 M acyl donor was used. Determined after 21 days. Example 8

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine catalysed by Esperase or Subtilisin using different alkanoic acid alkyl esters as acyl donors and solvents and Pyridine (20-60% v/v) as co-solvent.

Acyldonor Conversion(%) ^eeami_de^

Butylbutyrate³ 35 67 _a\ Isopropylisobutyrate ' 12 84

Methylbutyrate²^ 41 37 _a\ Meth lacetate ' 30 6 Ethylbutyrate⁰* 36 89

Ethylcaproate ' 30 20

a) 3.5 M acyl donor, 1.0 M amine, 1 mM Esperase, 35°C. Determined at 29 days.

b) N-alkanoyl-l-methyl-3-propylamide, R and S.

c) 2.6 M ethylbutyrate, 0.9 M amine, 1.44 mM Esperase, 35°C. Determined at 13 days.

d) 3.0 M ethylcaproate, 0.5 M amine, 0.36 mM subtilisin, 60°C. Determined at 14 days. Example 9

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (1.5 M) catalysed by Esperase (1 mM) using butylbutyrate (3.5 M) as acyl donor and solvent and 17% (v/v) Pyridine as co-solvent at 30°C with different amounts of 5k molecular sieve of activated granulate added.

Amount of 5A molecular sieve (mg/ml)

None 5

10

15

25

a) Determined at 35 days.

b) N-butyric acid-l-methyl-3-propylamide, R and S.

Example 10

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (1.0 M) catalysed by glutaric dialdehyde modified Esperase using butylbutyrate (4.1 M) as acyl donor and solvent and 15% Pyridine as co-solvent at 30°C.

a.) Enzyme Conversion(%) ' ^eeamide^

Modified Esperase ' 41 89

Esperase 28 . 76

a) Determined after 13 days of reaction.

b) Purification of Esperase:

Crude Esperase was dissolved in water and filtrated. The aqueous phase was concentrated in an A icon Cell (pressure limit « 4 bars, filter (PM-10) with a cut off at 10,000). Water was added to the concentrated solution and the concentration procedure repeated, followed by lyophilisation of the concentrate to give the purified Esperase.

Glutaric dialdehyde derivatization of Esperase: Derivatization of purified Esperase was performed by dropwise adding of a solution of 2.5% glutaric dialdehyde (10 times excess) under stirring to a solution of purified Esperase (0.05 mg/ml) . Stirring for half an hour, followed by quenching by addition of a weight equivalent of glycine and freeze-drying to give the modified Esperase. Example 11

Enzymatic stereoselective acylation of R,S-1-phenylethyl- amine (0.5 M) catalysed by Esperase (1.5 M) using butylbutyrate (4.0 M) as acyl donor and solvent and 25% (v/v) Pyridine as co-solvent at 35°C.

^Conversion⁽%^{)a) ee}amide^(%)b)

32 28

a) Determined after 5 days of reaction.

b) N-butyric acid-l-methyl-3-propylamide, R and S.

Example 12

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine at various concentrations catalysed by Subtilisin in crystalline or preactivated form (0.36 mM) or Esperase (1.8 mM) using butylbutyrate (3.0 M) as acyl donor and solvent and 18-50% (v/v) Pyridine as a co- solvent at various temperatures.

Enzyme Amine cone. (M) Conversion(%) ^ee _am-_jr_te(%)

Subtilisin A^a) 0.1 36 27

Subtilisin A^a 0.3 36 38 Subtilisin ³ 0.5 33 39

Subtilisin A^a* 1.0 26 46

Preact. Subtilisin ' 1.0 24 50

Preact. Subtilisin ' 1.5 20 60

Preact. Subtilisin*³* 2.0 15 73 c_ Esperase ' 0.5 25 85 c_

Esperase ' 0.75 25 83 c_

Esperase ' 1.0 19 94 e)

Esperase ' 1.5 26 _.78

a) Determined at 14 days, 60°C.

b) Determined at 13 days, 60°C.

c) Determined at 6 days, 50°C.

d) N-butyric acid-l-methyl-3-phenylpropylamide, R and S.

e) 1 mM Esperase, 3.5 M butylbutyrate. Determined at 7 days, 30°C. Example 13

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (1.0 M) catalysed by Esperase at various concentrations using butylbutyrate (3.0 M) as acyl donor and solvent and 34% (v/v) Pyridine as a co-solvent at 40°C.

Enzyme cone. (mM) Conversion(%)^a* ^eeamid ^

0.36 27 77 0.72 25 71 1.08 28 80 1.44 30 77 1.8 36 83

a) Determined at 21 days.

b) N-butyric acid-l-methyl-3-phenylpropylamide, R and S.

Example 14

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine catalysed by Esperase or preactivated Subtilisin using butylbutyrate at different concentrations as acyl donor and solvent and Pyridine (0-43% v/v) as a co-solvent to comprise the rest of the mixture at various temperatures.

Butylbutyrate(M) Enzyme Conversion(%) ee .. (%) '

a) 1.0 M amine, 0.36 mM preactivated Subtilisin A. Determined after 13 days at 60°C.

b) 1. 0 M amine, 1.4 mM Esperase. Determined after 13 days at 35°C.

c) 1.0 M amine, 1.8 mM Esperase. Determined after 13 days at 50°C.

d) N-butyric acid-l-methyl-3-phenylpropylamide, R and S.

e) 1.0 mM Esperase, 0.5 M amine. Determined after 21 days at 40°C. Example 15

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (0.5 M) catalysed by Subtilisin A at various concentrations using butylbutyrate (3.0 M) as acyl donor and solvent and 43% (v/v) Pyridine as a co-solvent at 60°C.

Enzyme cone. (mM) Conversion(%)^{a ee}amid ^

0.1 15 47 0.2 33 35 0.72 46 29 1.44 53 21

a) Determined after 7 days of reaction.

b) N-butyric acid-l-methyl-3-phenylpropylamide, R and S.

Example 16

Enzymatic stereoselective acylation of different amines (0.5 M) catalysed by Esperase (1.5 mM) using butylbutyrate (4.0 M) as acyl donor and solvent and 25% Pyridine as a co-solvent at 35°C.

Amine Conversion(%)^{a) ee}amide^(%)1^

R,S-l-phenylethylamine 32 47

a) Determined after 5 days of reaction.

b) N-butyric acid-1-phenylethylamide, R and S.

Example 17

Reuse of Recovered Enzyme

Activity test of Esperase used for 23 days in enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (1.0 M) catalysed by Esperase (1.08 mM) using butylbutyrate (3.0 M) as acyl donor and solvent and 25% Pyridine as a co-solvent at 40°C and then recovered from the mixture by filtration.

Test of enzyme U/μla)'

Before reaction 0.058

After reaction (23 days) 0.040

a) Assay according to Ottesen, M. and Svendsen, I. in Methods of Enzymology (Eds. Pearlman, G.E. and Lorand, L.), 19.' p. 199-215, Academic Press (1970) using BzArgOEt as substrate.

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (1.25 M) catalysed by fresh and recovered Esperase (1 mM) using butylbutyrate (3.5 M) as acyl donor and solvent and Pyridine (21% v/v) as a co- solvent at 40°C for 13 days.

Enzyme preparation Conversion(%) ee . , (%) '

Fresh 34 86 Recovered 21 84

b) N-butyric acid-l-methyl-3-phenylpropylamide, R and S. Example 18

Enzymatic stereoselective acylation of R,S-l-methyl-3- phenylpropylamine (0.5 M) catalysed by Esperase in crude industrial form and pretreated by filtration after dissolution in 3% DMF followed by freeze-drying using butylbutyrate (4.0 M) as acyl donor and solvent and Pyridine (25% v/v) as a co-solvent at 35°C for 39 days.

Enzyme preparation Conversion(%) ^eea_ammi_.^d_e_^)

Crude 1.4 mM 30 54 Pretreated 1 mM 39 59

a) N-butyric acid-l-methyl-3-phenylpropylamide, R and S.

Example 19

Enzymatic stereoselective acylation of different chirally R-enriched batches of R,S-l-methyl-3-phenylpropylamine (1.5 M) catalysed by Esperase (1.5 mM) using butylbutyrate

(3.0 M) as acyl donor and solvent and Pyridine (25% v/v) as a co-solvent at 35°C for three days.

Starting content of R-amine (% of R,S)^b* Conversion(%) ^eeamide ^%^

73 8 74 77 7 69 82 7 56 87 5 51 93 3 16

a) N-butyric acid-l-methyl-3-phenylpropylamide, R and S formed in the conversion.

b) These batches were produced by mixtures of purified amine chirally enriched as described in examples 1 and

6 and racemic amine.

Claims

P a t e n t C l a i m s :

1. A process for chiral enrichment of a mixure of enan- tiomers of an asymmetric primary amine, c h a r a c ¬ t e r i z e d by subjecting the amine to a stereoselec¬ tive transacylation with a simple (^C _I~^C _IQ) lkyl ester of an aliphatic (C₁-C₂_)alkanoic acid as the acyl donor cata¬ lyzed by a subtilisin-like enzyme in an organic medium containing not more than 8% by weight, preferably not more than 3% by weight, and most preferably not more than 1% by weight of water at a temperature of from 10 to 80°C, pre¬ ferably from 25 to 60°C, and most preferably from 30 to 40°C, until a substantial amount of one of the enantiomers has been converted to the amide.

2. A process according to claim 1, c h a r a c t e r ¬ i z e d in that the remaining chirally enriched amine is isolated from the reaction mixture.

3. A process according to claim 2, c h a r a c t e r ¬ i z e in that the isolation is performed by extraction, fractional distillation, ion exchange or liquid chromato- graphy or precipitation of an amine salt.

4. A process according to claims 2 or 3, c h a r a c ¬ t e r i z e d in that the isolated amine is subjected to another stereoselective transacylation according to claim 1.

5. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the chirally enriched amine is subjected to optical resolution by crystallisa¬ tion technique.

6. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the formed amide is isolated from the reaction mixture and hydrolyzed to ob¬ tain the amine.

7. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the formed amide is isolated from the reaction mixture and is subjected to racemization and hydrolysis to obtain the racemic amine for reuse in the process.

8. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the amine is a sub¬ stituted or unsubstituted alkylamine being branched in the 1-position.

9. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the amine is a substi¬ tuted or unsubstituted aralkylamine.

10. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the amine is a primary aryl-(C₂-C.. )alkylamine branched at the 1-position of the alkyl moiety.

11. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the amine is selected from the group consisting of l-methyl-3-phenylpropylamine, l-phenyl-(C₁-C_β)alkylamines, asymmetric amino acids and the amides and branched alkylesters of said amino acids.

12. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the acyl donor is a simple straight chain (C-_j-Cg)alkyl ester of a (C_I-C_IQ)- alkanoic acid.

13. A process according to claim 12, c h a r a c t e r ¬ i z e d in that the acyl donor is selected from the group consisting of ethylacetate, propylacetate, isopropylace- tate, butylacetate, ethylpropionate, propylpropionate, isopropylpropionate, butylpropionate, ethylbutyrate, pro- pylbutyrate, isopropylbutyrate, butylbutyrate, ethylisobu- tyrate, isopropylisobutyrate, ethylcaproate, propylcapro- ate and ethylcaprylate.

14. A process according to claim 13, c h a r a c t e r ¬ i z e d in that the acyl donor is ethyl, propyl, isopro- pyl or butyl butyrate.

15. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the subtilisin-like enzyme is a serine 'endoprotease of microbial origin having a molecular weight between 15000 and 35000 and a neutral to alkaline pH optimum.

16. A process according to claim 15, c h a r a c t e r¬ i z e d in that the serine endoprotease has been produced by species of Bacillus, preferably selected among the strains NCIB Nos. 10144-10148, 10281, 10284, 10286, 10288, 10301, 10304, 10306-10313 and 10315-10327, related strains, and microorganisms into which genetic material from the mentioned strains has been introduced.

17. A process according to claim 15, c h a r a c t e r ¬ i z e d in that the subtilisin-like enzyme is selected from the group consisting of subtilisin A ("Subtilisin Carlsberg"), "Alcalase"®, subtilisin B ("Subtilisin Novo"), "Esperase"®, "Savinase"® and subtilisin BPN' ("Nagarse"®).

18. A process according to claim 17, c h a r a c t e r ¬ i z e d in that the subtilisin-like enzyme is "Esperase"®.

19. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the subtilisin-like enzyme has been covalently modified.

20. A process according to claim 19, c h a r a c t e r ¬ i z e d in that the enzyme has been modified by deriva¬ tion with glutaric dialdehyde and subsequent reaction with an amine.

21. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the subtilisin-like enzyme has been immobilized.

22. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the subtilisin-like enzyme has been preactivated by drying from an aqueous so¬ lution in the presence of a suitable ligand or inhibitor, if necessary followed by washing out ligand or inhibitor with an organic solvent.

23. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the subtilisin-like enzyme is isolated from the reaction mixture and reused in the process.

24. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the organic medium is made up of an excess of the acyl donor.

25. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the organic medium comprises an excess of the acyl donor and a co-solvent.

26. A process according to claim 25, c h a r a c t e r ¬ i z e d in that the co-solvent is one which enhances the velocity and/or stereoselectivity of the transacylation reaction.

27. A process according to claim 26, c h a r a c t e r ¬ i z e d in that the co-solvent is pyridine, cyclohexane or hexane.

28. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the concentration of the amine in the reaction mixture is from 0.1 to 3.0 M, preferably from 0.5 to 2.0 M.

29. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the concentration of the acyl donor in the reaction mixture is from 0.5 to 8.0 M, preferably from 2.0 to 4.5 M.

30. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the acyl donor is used in an excess concentration of at least 1 M in relation to the amine.

31. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the concentration of the subtilisin-like enzyme in the reaction mixture is from 0.01 to 5.0 mM based on enzyme protein content.

32. A process according to any of the preceding claims, c h a r a c t e r i z e d in that the transacylation re¬ action is performed in the presence of a molecular sieve which binds the alcohol formed from the acyl donor.