Enzyme-catalysed racemization

Enzymatic racemisation is an attractive option in DKR because the reactions catalysed by enzymes are performed under mild conditions. The Degussa group have recently described their successful commercialization of two DKR-based processes that employ racemases, namely (i) the DKR of 5-substituted hydantoins using whole cells coexpressing a L-carbamoylase, a hydantoin racemase and a hydantoinase and (ii) the DKR of N-acetyl amino acids using an acylase in combination with an N-acetyl amino acid racemase from Amycolatopsis orientalis. 

---

Another approach to the synthesis of chiral non-racemic hydroxyalkyl sulfones used enzyme-catalysed kinetic resolution of racemic substrates. In the first attempt. Porcine pancreas lipase was applied to acylate racemic (3, y and 8-hydroxyalkyl sulfones using trichloroethyl butyrate. Although both enantiomers of the products could be obtained, their enantiomeric excesses were only low to moderate. Recently, we have found that a stereoselective acetylation of racemic p-hydroxyalkyl sulfones can be successfully carried out using several lipases, among which CAL-B and lipase PS (AMANO) proved most efficient. Moreover, application of a dynamic kinetic resolution procedure, in which lipase-promoted kinetic resolution was combined with a concomitant ruthenium-catalysed racem-ization of the substrates, gave the corresponding p-acetoxyalkyl sulfones 8 in yields... 

Two types of racemic 3-hydroxy phosphonates, in which the phosphono and hydroxy moiehes are separated hy double bond, were successfully resolved using a common enzyme-catalysed acetylation. Both acyclic 52 (Equation 28) and cyclic 54 (Equation 29) derivatives underwent easy acetylation under the kinetic resolution conditions to give the products in high yield and with almost full stereoselechvity. 

In contrast to Mori s synthesis, Pawar and Chattapadhyay used enzymatically controlled enantiomeric separation as the final step [300]. Butanone H was converted into 3-methylpent-l-en-3-ol I. Reaction with trimethyl orthoacetate and subsequent Claisen-orthoester rearrangement yielded ethyl (E)-5-methyl-hept-4-enoate K. Transformation of K into the aldehyde L, followed by reaction with ethylmagnesium bromide furnished racemic ( )-7-methylnon-6-ene-3-ol M. Its enzyme-catalysed enantioselective transesterification using vinylacetate and lipase from Penicillium or Pseudomonas directly afforded 157, while its enantiomer was obtained from the separated alcohol by standard acetylation. 

Pellissier, H., Recent developments in dynamic kinetic resolution. Tetrahedron, 2008, 64, 1563-1601 Turner, N.J., Enzyme catalysed deracemisation and dynamic kinetic resolution reactions. Curr. Opin. Chem. Biol., 2004, 8, 114-119 Gmber, C.C., Lavandera, I., Faber, K. and Kroutil, W., From a racemate to a single enantiomer deracemisation by stereoinversion. Adv. Synth. Catal., 2006, 348, 1789-1805 Pellissier, H., Dynamic kinetic resolution. Tetrahedron, 2003, 59, 8291-8327 Pmnies, O. and Backvall, J.-E., Combination of enzymes and metal catalysts. A powerful approach in asymmetric catalysis. Chem. Rev., 2003, 103, 3247-3261. 

We have used a series of biocatalysts produced by site-directed mutations at the active site of L-phenylalanine dehydrogenase (PheDH) of Bacillus sphaericus, which expand the substrate specificity range beyond that of the wild-type enzyme, to catalyse oxidoreduc-tions involving various non-natural L-amino acids. These may be produced by enantiose-lective enzyme-catalysed reductive amination of the corresponding 2-oxoacid. Since the reaction is reversible, these biocatalysts may also be used to effect a kinetic resolution of a D,L racemic mixture. ... 

In connection with our own work on the enzyme-catalysed hydrolysis of cyclohexene epoxide with various fungi we made the unexpected observation that the microorganism Corynesporia casssiicola DSM 62475 was able to interconvert the (1R,2R) and (1S,2S) enantiomers of the product, trans cyclohexan-1,2-dioI 25. As the reaction proceeded the (1R,2R) enantiomer was converted to the (1S,2S) enantiomer [20]. If the racemic trans diol 25 was incubated with the growing fungus over 5 days, optically pure (> 99 % e. e.) (1 S,2S) diol 25 could be isolated in 85% yield. Similarly biotransformation of cis (meso) cycIohexan-1,2-diol 26 yielded the (1S,2S) diol 25 in 41 % (unoptimized) yield (Scheme 11). 

Enzymes as chiral catalysts play a role in all three methods. In nature enzymes catalyse all production of chiral compounds. In the laboratory enzymes can catalyse asymmetric synthesis, as well as resolve racemates. Which of the three methods is chosen in different cases depends on several factors, like price of starting materials, number of synthetic steps, available production technology and know-how etc. There is at present a constant ongoing development of synthetic methods and biotransformation is one field. Utilization of method i) requires knowledge of classical organic synthesis, enzymes have already played their role. Enzymes may play a part both in asymmetric synthesis and resolution. 

Enzyme catalysed hydrolysis of racemic epoxides is interesting from a practical point of view. This reaction is catalysed by epoxide hydrolases (EHs, EC 3.3.2.3) (Archelas and Furstoss, 1998). Mammalian EHs are the most widely studied and they are divided into five groups among which the soluble (cytosolic) epoxide hydrolases (sEH) and microsomal epoxide hydrolases (mEH) are best charactelised. The mechanism of sEH from rat starts with a nucleophilic attack by Asp333 on a carbon of the epoxide (usually the least hindered one) to form a glycol monoester intermediate which is stabilised by an oxyanion hole. A water molecule activated by His523 releases the 1,2-diol product. An... 

Nanda S, Rao AB et al (1999) Enzyme catalysed kinetic resolution of racemic 2,2-dimethyl-3-(2,2-disubstituted vinyl) cyclopropane carboxylic acids anchored on polymer supports. Tetrahedron Lett 40 5905-5908... 

Enzymic resolutions involve acceptance by the enzyme, which is a very finely honed chiral system, of one enantiomer of a racemic compound, but not the other. The selective acceptance arises because interactions between the enzyme and the enantiomers are diastereomeric. In its natural environment, the ability of an enzyme to discriminate between enantiomers is virtually absolute. In addition to their stereoselectivity, some enzymes can react at very high rates. Each round of catalysis by the enzyme carbonic anhydrase with its physiological substrate occurs in about 1.7 jus at room temperature, although for a small number of other enzymes, best exemplified by the more lethargic lysozyme, the corresponding figure is about a million times slower. Accordingly, the enzyme-catalysed hydrolysis of, say, one enantiomer of an ester proceeds at a finite rate and hydrolysis of the other not at all. Resolutions such as those of 39, 42 and 45 therefore have a kinetic basis and are also known as kinetic resolutions. 

It should be evident that the maximum yield of a particular enantiomer normally available from a racemic mixture is 50%. However, in some enzymic catalysed kinetic resolutions it is possible to obtain >50% yield of one enantiomer from a racemate. For this to occur, it is necessary to have the desired chemical reaction, e.g. enzyme-catalysed stereoselective esterification, occurring at the same time as the enantiomers of the racemic starting compound are interconverting under equilibrium conditions. A successful example of this technique is provided by ben-zaldehyde cyanhydrin (2-hydroxy-2-phenylacetonitrile), whose R and S enantiomers, 49 and 50, respectively, equilibrate in the presence of a basic anion-exchange resin (Scheme 3.5). In the presence of lipase, (S)-ben-zaldehyde cyanhydrin acetate 51 was formed in 95% yield and in 84% enantiomeric excess (see Inagaki et al.u and Ward15). 

Exploiting the ability of an enzyme to catalyse a reaction, usually either alkaline hydrolysis of an ester, or an esterification, of one enantiomer exclusively. If the enantiomers of the racemic starting material can be made to equilibrate while the enzyme-catalysed enantioselective reaction is taking place, yields of one enantiomer in excess of 50% can be obtained. 

Other possibilities to prepare chiral cyanohydrins are the enzyme catalysed kinetic resolution of racemic cyanohydrins or cyanohydrin esters [107 and references therein], the stereospecific enzymatic esterification with vinyl acetate [108-111] (Scheme 2) and transesterification reactions with long chain alcohols [107,112]. Many reports describe the use of fipases in this area. Although the action of whole microorganisms in cyanohydrin resolution has been described [110-116],better results can be obtained by the use of isolated enzymes. Lipases from Pseudomonas sp. [107,117-119], Bacillus coagulans [110, 111], Candida cylindracea [112,119,120] as well as lipase AY [120], Lipase PS [120] and the mammalian porcine pancreatic lipase [112, 120] are known to catalyse such resolution reactions. 

The same enzyme catalyses the esterification of racemic 129 in non-aqueous solution with vinyl acetate. The released alcohol is CH2=CHOH, the enol of acetaldehyde it immediately forms acetaldehyde which self condenses and is removed from the equilibrium. The enzyme is filtered off, the enantiomerically pure alcohol (S) -129 and acetate (R)-130 separated by flash chromatography, and the ester hydrolysed to the alcohol without racemisation. Either method (esterification or hydrolysis) gives both enantiomers of a range of secondary alcohols.31... 

The racemic amino alcohol 2 was easy to prepare so one obvious choice is an enzyme-catalysed acylation or deacylation process (chapter 26). These worked reasonably well. Deacylation of the ester-amide 6 gave the amide 5 which was difficult to hydrolyse. Acylation gave the monoester 4 but this equilibrated rapidly with the same amide 5. Both gave the same enantiomer of the amino alcohol and of course the maximum yield is only 50%. 

Finally, we should mention a series of synthetic approaches to L-pyrophosphino-tricine 144, a cyclic analogue of L-phosphinotricine, and its Ala-Ala peptide 145, a cyclic analogue of bialaphos (tripeptide of L-phosphinotricine produced in nature by Streptomyces hydroscopicus) [186], As the final step, the procedures comprised enzyme-catalysed hydrolysis and resolution of the racemic mixtures formed into optical antipodes by a-chymotripsin. Both 1,2-azaphospholane 144 and its tripeptide 145 exhibit antitumour activity. Moreover, tripeptide 145 displayed a greater bactericidal activity than the antibiotic bialaphos. 

Enantioselective cleavage of non-peptide amide bonds is also important in the production of optically active amino acids (Scheme 3.12). Carboxy-peptidases often are the enzymes of choice in this area of work these enzymes catalyse the hydrolysis of an amide function which is close to a carboxylic acid group. The rate of hydrolysis is usually increased if R (Scheme 3.12) is an aromatic unit or a large aliphatic moiety. For example, thrco-jS-phenylserine R = PhCH(OH) has been resolved by incubation of the racemic JV-trifluoroacetate with carboxypeptidase-A, with the optically pure (L)-enantiomer being obtained in a good yield. 

Lipases have also been widely applied for the resolution of racemic chiral amines. In principle, these reactions can be carried out in both the hydrolytic mode as well as under conditions favouring acylation. As amines are more nucleophilic than alcohols, it is necessary to use less reactive acyl donors in order to minimize the background reaction of non-enzyme catalysed acylation, and in this respect it appears that simple esters such as ethyl acetate are optimal. 

The enzyme-catalysed hydrolysis of SchifiF bases derived from racemic amino acid esters and aromatic aldehydes was studied by Parmar et al. The L-amino acid precipitated out from the solution as the reaction progressed and the liberated aldehyde and unhydrolysed o-ester remained in solution. 

---