《有机化学》课程教学资源（文献资料）Chemoenzymatic and microbial dynamic kinetic resolutions.pdf

Review Article Chemoenzymatic and Microbial Dynamic Kinetic Resolutions AZLINA HARUN KAMARUDDIN,1* MOHAMAD HEKARL UZIR,1 HASSAN Y. ABOUL-ENEIN,2* AND HAIRUL NAZIRAH ABDUL HALIM3 1 School of Chemical Engineering, Engineering Campus, Universiti Sains Malaysia, Nibong Tebal, Pulau Pinang, Malaysia 2 Pharmaceutical and Medicinal Chemistry Department, National Research Centre, Dokki, Cairo, Egypt 3 School of Bioprocess Engineering, Universiti Malaysia Perlis (UniMAP), Arau, Perlis, Malaysia ABSTRACT This review tracks a decade of dynamic kinetic resolution developments with a biocatalytic inclination using enzymatic/microbial means for the resolution part followed by the racemization reactions either by means of enzymatic or chemocatalyst. These fast developments are due to the ability of the biocatalysts to significantly reduce the number of synthetic steps which are common for conventional synthesis. Future developments in novel reactions and products of dynamic kinetic resolutions should consider factors that are needed to be extracted at the early synthetic stage to avoid inhibition at scale-up stage have been highlighted. Chirality 21:449–467, 2009. VC 2008 Wiley-Liss, Inc. KEY WORDS: dynamic kinetic resolution; biocatalysts; chemocatalysts; chiral resoluion INTRODUCTION Dynamic kinetic resolution (DKR) has gained importance for more than 2 decades since the term was first coined in 1983 for its ability to allow for a maximal theoretical yield of 100%. It is an efficient technique that produces enantiomerically pure form of the intended products. It combines enzyme, chemocatalyst or microbial kinetic resolution with the in situ enzyme, metal or base-catalysed racemisation. The classical schematic route that represents DKR is given in Figure 1. With the current method, the process depends profoundly on the ability of the enzyme/cells to select between the two enantiomers as expressed by the enantiomeric ratio, E; a relative term of the rate of reaction of the two enantiomers.1,2 The process is a well-travelled route to optically active compounds that has overcome the limited yield of the required enantiomer if kinetic resolution alone is being used. With such a combination, one can in principle obtain a quantitative yield of one of the enantiomers where the resolution step is combined with an in situ racemisation of the unreacted enantiomer. The combined route of enzymatic resolution with metal and base catalyzed racemisation in DKR is not exactly straightforward. Martı´n-Matute and Ba¨ckvall3 highlighted possible problems of incompatibility of the processes combined together in one pot. For an efficient DKR, one of the important requirements is that the compatibility of the two catalysts must be achieved but most often this is the major problem. The interference of metal with enzyme in the resolution reaction may give poor resolution, or the enzyme could also act as an inhibitor in the racemisation process. However, some groups have shown otherwise.3–7 A cursory glance at the literatures on the aspects of DKR shows that a large number of new processes have been developed. These include reactions which produced pharmaceutical and agrochemical products.8,9 However, not many of these processes can be scaled up. From the chemical engineering point of view, Graviilidis’s and coworkers10 have given an excellent comprehensive review of factors that inhibit scalability of fine chemicals and pharmaceuticals. The researchers in their early-phase of synthetic work must already start bearing in mind that critical data for scaling up should be able to be extracted from their synthetic work to support the later-phase of the developmental work. Graviilidis and coworkers identified the most common factors that interact with chemistry causing a fall in the performance and they suggested ways to analyze these issues in order to generate suitable solutions. Wells11 has also mentioned a similar problem for scale up purposes, where a number of factors need to be considered. He concluded that the key factor to succeed in identifying and scaling up reactions with biocatalysts is an understanding of the physical nature of the biocatalyst itself, the number and action of the enzymes present, and how best to present the enzyme to the reaction. *Correspondence to: Azlina H. Kamaruddin, School of Chemical Engineering, Engineering Campus, Universiti Sains Malaysia, Seri Ampangan, 14300 Nibong Tebal, Pulau Pinang, Malaysia. E-mail: chazlina@eng. usm.my or Hassan Y. Aboul-Enein, Pharmaceutical and Medicinal Department, National Research Centre, Dokki, Cairo 12311, Egypt. E-mail: enein@gawab.com. Received for publication 27 January 2008; Accepted 20 May 2008 DOI: 10.1002/chir.20619 Published online 24 July 2008 in Wiley InterScience (www.interscience.wiley.com). CHIRALITY 21:449–467 (2009) VC 2008 Wiley-Liss, Inc

Several configurations of enzymatic reactors have been reported involving reaction of chiral centers, for instance, a packed bed column with immobilized lipase on resolution of naproxen and the use of enzymatic membrane reactors (EMR) in which high enantioselectivity for the (S)- enantiomer of the racemate was achievable.12–15 This shows that EMR has a great potential towards the development of chirotechnology. As the result of the role played by membrane in the separation of two immiscible fluids, EMRs are normally employed in biphasic reactions in which separation and reaction occur simultaneously. Although most of the reactions considered in the EMR are isolated to resolution reactions, but the possibility of combining both resolution and racemisation under a single operating EMR system has been described by Kamaruddin and coworkers16 in the production of (S)-ketoprofen. As mentioned earlier, microbial kinetic resolution as well as DKR has also been the focused of academia in developing novel reactions. There is a wide range of microorganisms which has been screened for microbial transformations especially for use in reactions involving dynamic kinetic resolution. However, baker’s yeast (Saccharomyces cerevisiae) has been one of the most popular organisms to perform DKR. Much have been written especially on its ability to catalyse reduction reaction on a and b-ketoesters.17–19 Other types of organisms with similar ability as that of baker’s yeast are tabulated in Table 1. The use of microbial cells as parts of the catalyst provides a stable environment as well as a complete cycle for a redox reaction in particular. As described by Uzir, in 2005 (unpublished data) in a reduction reaction, for a reaction to complete a cycle, oxidised cofactor NADP1 or NAD1 is required to release an electron to become NADPH or NADH, respectively. Such a compound is only available in a whole-cell where the cofactor is continuously produced during the growth31 and subsequently consumed during the reaction. This means that in a situation where a reaction requires a cofactor to be generated in situ, the use of whole cell is one of the best alternatives. In the reduction of cyclic b-oxoesters with 5- or 6-membered ring with Saccharomyces cerevisiae, considerably good results were obtained. In addition, for a reduction of 2-carbethoxycycloheptanone (1) with higher cyclic ring, a microorganism such as Kloekera magna gives a better yield and selectivities of the product (2). This is shown in the reaction given by Scheme 1. This review focuses on the most recent development in DKR reactions with a biocatalytic inclination using enzymatic means for the resolution part and followed by the racemisation reactions either by enzymatic or chemical means. The discussion is arranged in such a way that different types of reactions such as hydrolysis, esterification, alcoholysis, and transesterification, aminolysis, and ammonolysis and acylation for the preparation of cyanohydrin esters were discussed in separate headings. In addition to this scope, the use of microorganisms in DKR is also considered. ENZYMATIC HYDROLYSIS Enzyme-catalysed hydrolysis is an attractive method for the kinetic resolution of racemic esters into carboxylic acid and alcohol.32–38 The hydrolysis process under in situ racemisation (DKR) from various racemic esters as substrate is well known and can be carried out by lipase for the production of non-steroidal anti-inflammatory drugs (NSAIDs) such as (S)-suprofen, (S)-naproxen, (S)-ibuprofen, (S)-fenoprofen, etc. and pharmaceutical active intermediates.15,32–37 Tsai and coworkers32,33 reported the DKR of (rac)- suprofen 2,2,2- trifluoroethyl thioester (3) for the production of (S)-suprofen (4) in different organic solvents; isooctane32 and cyclohexane.33 The DKR of racemic suprofen is shown in Scheme 2. Candida rugosa lipase was employed as the biocatalyst for enantioselective hydrolysis of (rac)- suprofen 2,2,2-trifluorothioester, in which trioctylamine was added as the catalyst to perform in situ racemisation of the remaining (R)-thioester (5). The conversion of the racemic suprofen for the desired (S)-suprofen completed with 95 e.e.32 In their work, they described a detailed investigation of the trioctylamine catalyst on the kinetic behaviours of the thioester in the DKR process. Their results indicated that the racemisation catalyst not only activates the lipase, but also enhances the enzyme stability. Tsai and coworkers then successfully integrated a hydrophobic hollow-fiber membrane into the DKR process, where the desired (S)-suprofen is continuously removed from cyclohexane to the aqueous phase circulating in the shell-side of the membrane, giving high yields of high optical purity (S)-suprofen. They developed a kinetic model for the whole process (operating in batch and fed-batch modes). The model was an extensive one, incorporating enzymatic hydrolysis and deactivation, lipase activation, racemisation and non-selective hydrolysis of the substrate by trioctylamine, and reactive extraction of (rac)- suprofen into the aqueous phase in the membrane. The production of (S)-naproxen (7) from racemic naproxen methyl ester by DKR has been reported by Xin et al.15 Candida rugosa lipase was selected as biocatalyst for enantioselective continuous hydrolysis process under in situ racemisation of racemic naproxen methyl ester (6) using sodium hydroxide as racemisation catalyst to catalyse Fig. 1. Schematic representation of a dynamic kinetic resolution. SS and SR represent enantiomers and PS and PR represent product enantiomers. 450 KAMARUDDIN ET AL. Chirality DOI 10.1002/chir

the remaining (R)-methyl ester (8). The conversion of racemic naproxen methyl ester was greater than 60% with an enantiomeric excess (e.e.) of (S)-naproxen greater than 96%. The DKR of racemic naproxen ester is shown in Scheme 3. This reaction took place in an aqueous-organic biphasic system where a tubular silicone rubber membrane was used in a stirred-tank reactor. This whole set-up allows for the separation of the chemical catalytic racemisation and enzyme resolution processes which served to avoid the key problem associated with Naproxen conventional dynamic resolution. Besides working on (S)-suprofen as a valuable product, Tsai and coworkers also extended their research on the DKR to the production of (S)-naproxen. Two different substrates were used in their study which were 2,2,2-trifluoroethyl ester34 and 2,2,2-trifluoroethyl thioester.35 Candida rugosa lipases immobilized on polypropylene powders were employed as biocatalysts for the enantioselective hydrolysis of (rac)-naproxen ester in isooctane at 458C. The organic base of 1,5,7-triazabicyclo[4,4,0] dec-5-ene bound to polystyrene cross-linked with 2% divinylbenzene (DVB) (p-TBD)34 and trioctylamine35 was added as an in situ racemisation catalysts, respectively. The enantiomeric excess (e.e.) for the (S)-naproxen was 58.1% at the racemate conversion of 95.5%.34 Using the trioctylamine as racemization catalyst, they integrated the DKR process with a hollow- fiber membrane to reactively extract the desired (S)-naproxen out of the reaction medium. The DKR of (rac)-fenoprofen thioester (9) by using Lipase MY as biocatalyst was reported by Tsai and coTABLE 1. (Continued) Product Microorganism Yield d.e. (%) e.e. (%) Reference Kloekera magna 80 100 94 25 Yarrowia lipolytica 95 – 95 26 Saccharomyces cerevisiae 74 – 98 27, 28 Saccharomyces cerevisiae 71 98 85 29, 30 Scheme 1. Reduction of 2-carbethoxycycloheptanone via DKR technique. 452 KAMARUDDIN ET AL. Chirality DOI 10.1002/chir

workers.36 The DKR of (rac)-fenoprofen thioester is shown in Scheme 4. The enantioselective hydrolysis process of (rac)-fenoprofen 2,2,2- trifluoroethyl thioester in isooctane was catalyzed by Lipase MY from Candida rugosa under the in situ racemisation of the remaining (R)-thioester substrate (11) with trioctylamine as racemisation catalyst. The conversion of racemic (rac)-fenoprofen thioester was 91% with an enantiomeric excess of (S)-fenoprofen (10) of 91% when 75 mM of trioctylamine was added as racemisation catalyst. They showed that the racemisation process follows a first order reversible kinetics, in which a linear relationship between the inter-conversion constant with trioctylamine concentration was found. Their work also indicated that trioctylamine also acted as a lipase activator. Kamaruddin and coworkers,37 developed the enzymecatalysed enantioselective hydrolysis process for (S)-ibuprofen (13) production which is one of the most important members of NSAIDs that belongs to the family of propionic acid. It is widely used to treat rheumatoid arthritis, headache, muscular strain, cephalalgia, etc.39 The process involved the hydrolysis of (rac)-ibuprofen ester, 2-ethoxy- 2-4-(isobutylphenyl) propionate (12). The kinetic resolution of (rac)-ibuprofen ester with lipase from Candida rugosa has produced (S)-ibuprofen acid and unreacted (R)- ibuprofen ester (14). Under the in situ racemisation, the unreacted (R)-ibuprofen ester was racemised with sodium hydroxide as the base catalyst. Through this method, the enantiopure (S)-ibuprofen (13) can be obtained at 99.4% and the conversion higher than 85% can be achieved. The DKR of racemic ibuprofen ester is shown in Scheme 5. A lipase-catalysed dynamic hydrolytic resolution of (rac)-2,2,2-trifluoroethyl a-chlorophenyl acetate (15) and (16) in water-saturated isooctane for the production of tri- fluoroethyl (R)-a-chlorophenyl acetic acid (17) has been reported by Tsai and coworkers.40 The (R)-a-chlorophenyl acetic acid (17) is a type of a-haloarylacetic acids which are known as important intermediates for synthesizing many drugs such as prostaglandin, prostacyclin, semi-synthetic penicillin, and thiazolium salts.41–43 The DKR of (rac)-2,2,2-trifluoroethyl a-chlorophenyl acetate is shown in Scheme 6. The authors reported that the best hydrolysis reaction was catalysed by Lipase MY(I) at 358C and the racemisation was catalyzed by trioctylamine as the racemization agent. The reaction gave 93.0% yield and 89.5% e.e. of the desired (R)-a-chlorophenyl acetic acid (17). Recently, Fulop and coworkers44 reported lipase-catalysed kinetic and DKR of racemic 1,2,3,4-tetrahydroisoquinoline-1-carboxylic acid (ethyl ester (rac)-(18) to produce enantiopure (R)-1,2,3,4-tetrahydroisoquinoline-1-carboxylic acid (19). The drugs containing the isoquinoline skeleton can be applied to a wide range of therapies.44 Candida antarctica lipase B catalyzed the enantioselective hydrolysis of the corresponding ethyl ester (rac)-1 in toluene/acetonitrile (4:1) containing 1 equiv of added water and 0.25 equiv of dipropylamine as racemisation catalyst to catalyse the unreacted S-ethyl ester (20). However, they found that during the DKR process using base-catalyst, the rate of racemisation was much slower than the rate of enzymatic hydrolysis. They then solved the problem by starting the hydrolysis reaction in the presence of low enzyme content Scheme 2. DKR of racemic suprofen. Scheme 3. DKR of racemic naproxen ester. CHEMOENZYMATIC AND MICROBIAL DKR 453 Chirality DOI 10.1002/chir