Resolution of R,S

At first, a protocol similar to that developed by the discovery group was adopted in order to determine whether (R,S)-9 would undergo enantioselective hydrolysis. A mixture of (R,S)-9 and Amano PS-30 lipase was stirred overnight in dimethyl sulfoxide (DMSO) in phosphate buffer at 38—40°C. Analysis by chiral HPLC indicated that hydrolysis was selective for the (f )-enantiomer. 

While it is a powerful solvent, DMSO is undesirable for industrial use. Instead, the surfactant Triton X-100 can be an effective replacement for DMSO in enzymatic hydrolyses [14], and in our case, worked very well. Triton X-100 forms micelles when mixed with water and would solubilize low concentrations of ester for hydrolysis. However, increasing the surfactant charge did not accelerate hydrolysis. In an attempt to reduce the somewhat long reaction time, we screened the enzymes Amano PS Lipase XIII, Pseudomonas jluorescens and Candida cylindracea versus Amano PS-30. PS-30 was still considered the best choice of enzyme when 

The preparation was now increased to 50 g of (R,S)-9. The elimination of DMSO provided a convenient bonus in that unreacted (S)-9 was not appreciably soluble in the aqueous reaction mixture and was recovered with the other insoluble components by filtration. Subsequently, (R)-l could be isolated as a solid upon acidification of the filtrate. The chemical and optical purity of the crude (R)-1 could be efficiently raised by recrystallization from ethanol. These results solved our work-up concerns and allowed us to scale-up to the kilo-lab. 

The resolution of 1.0 kg of (R,S)-9 in the kilo-lab required 48 hours instead of the 25-35 hours observed on a smaller scale, but 94% of the available yield of (R)-9 was obtained in 96% enantiomeric excess (eep). With the exception that the filtration was slower than expected, all proceeded well and we moved into the plant using 50 and 200 gallon glass-lined vessels to resolve 7.90 kg of (R,S)-9. After 48 hours, the (R)-ester had decreased to a satisfactory level of 1.7% and the isola-tion/purification was conducted as described above to produce 2.42 kg of (R)-l in 75% available yield (98.5 wt.-%, 99.9% ee). We were delighted to note that our process development work had anticipated all potential problems while this had been our first plant-scale biotransformation of any sort, the chemistry and subsequent work-up had proceeded without incident. 

Over the years, we have scaled this resolution up to 40 kg of isobutyl ester 9 in 300 gallon vessels without problems. Although the larger scale runs are somewhat limited by the slow filtrations, yields have remained reproducible. We consider this preparation a mature and robust process and have devoted our further energy to other aspects of the roxifiban synthesis. 

---

Chen, C.Y, Cheng, Y.C. and Tsai, S.W., Lipase-catalyzed dynamic kinetic resolution of (R,S)-fenoprofen thioester in isooctane. J. Chem. Technol. Biotechnol., 2002, 77, 699-705. 

Yun, H., Cho, B.-K. and Kim, B.-G., Kinetic resolution of (R,S)-jec-butylamine using omega-transaminase from Vibrio fluvialis JS17 under reduced pressure. Biotechnol. Bioene., 2004, 87, 772-778. 

Screening of eight lipases and two esterases 10 different ionic liquids Kinetic resolution of (R,S)-l-phenylethanol 44... 

Hydrolase Resolution of (R,S)-ethyl mandelate and (R,S)-ethyl 2-chloromandelate in phosphate buffer / iso-octane [68, 69]... 

Recent studies in the pharmaceutical field using MBR technology are related to optical resolution of racemic mixtures or esters synthesis. The kinetic resolution of (R,S)-naproxen methyl esters to produce (S)-naproxen in emulsion enzyme membrane reactors (E-EMRs) where emulsion is produced by crossflow membrane emulsification [38, 39], and of racemic ibuprofen ester [40] were developed. The esters synthesis, like for example butyl laurate, by a covalent attachment of Candida antarctica lipase B (CALB) onto a ceramic support previously coated by polymers was recently described [41]. An enzymatic membrane reactor based on the immobilization of lipase on a ceramic support was used to perform interesterification between castor oil triglycerides and methyl oleate, reducing the viscosity of the substrate by injecting supercritical CO2 [42],... 

FIGURE 7.8 Dutch Resolution of (R,S)-jt>-hydroxyphenylglycine with (+)-camphor-10-sulfonic acid and (R,S) or (R)-phenylglycine as a family member. 

The chromatographic resolution of (R-S)-7-chloro-l,3-dihydro-3-methyl-5-phenyl-2H-l,4-benzodiazepin-2-one (III) is another example of the previous method that can be cited [21], Figure, 2. 

Figure 5. The chromatographic resolution of (R,S)-IV on an ionic Pirkle column. Flow rate 1 ml/min. Eluant hexane/2-propanol, 60/40 (a), 80/20 (b), and 90/10 (c). Absorption (A) and CD detection ( ) at 254 nm. (From Ref. 21)...

The resolution of the racemic mixture of 2-octanol and of other secondary alcohols (2-butanol, 2-pentanol, 2-hexanol, and 2-heptanol) by direct esterification in organic solvent was studied by using lyophiUzed mycelia of R. oryzae CBS112.07 as catalysts [18]. The profile of the resolution of (R,S)-2-octanol under optimized conditions (lg/1 of alcohol and equimolar butanoic acid as acylating agent in n-heptane, 30g/l of dry biocatalyst, and 30 °C) is shown in Figure 6.1. 

Resolution of Amino Acids. For the optical resolution of racemic threonine via replacing resolution, (S)-proline was utilized as an optically active cosolute although the structure of the imino acid is different from that of threonine. The same procedure was applied less sucessfully to the resolution of (R,S)-thiazolidine carboxylic acid. ... 

Lou WY, Zong MH (2006) Efficient kinetic resolution of (R, S)-l-trimethylsilylethanol via lipase-mediated enantioselective acylation in ionic liquids. Chirality 18 814-821... 

Xin J-Y, Zhao Y-J, Zhao G-L et al (2005) Enzymatic resolution of (R, S)-naproxen in water-saturated ionic liquid. Biocatal Biotransform 23 353-361... 

Fig- 5 The resolution of (R,S)-ethyl-3,3,3-trifluoro-2-hydroxy-2-methylpropionate with the esterase from Candida lipolytica. 

We then carried out the optical resolution of (R.S-)-BNOAc (10%) on a 500 mL scale to produce (S)-BNOAc (Fig. 12), and (S)-BONAc was obtained with >98% ee in a yield of 32% in 24 hours. This resolution reaction was reproduced up to a 1 kL scale, so that this technology should be suitable for industrial-scale production. Moreover, in order to make this method more efficient, if (R)-BN with 45% ee was reacted to (R)-BNOAc after extraction of (S)-BNOAc, and treated by the reverse type of strain, Pseudomonas sp. DS-K-717, the productivity would be substantially increased. 

Our proposed synthesis of 1 using a hetero Diels-Alder-biocatalysis approach is shown in Fig. 4. In this study, we report the details of the development of a commercially viable process to make 1 based on the enzyme resolution of (R,S) 4 [12]. 

The results presented above clearly indicate that the resolution of (R,S)-4 can be easily carried out using Bacillus lentus proteases. As has been mentioned before we chose Bacillus lentus protease-III for further development because of its efficacy and lower cost. Commercial formulations of this enzyme are stable and avail-... 

The enzymatic resolution of (R,S)-2-ethoxycarbonyl-3,6-dihydropyran has been carried out repeatedly in a pilot plant on a 100-125 kg scale. The conditions of pH, agitation, concentration, and enzyme/substrate ratio have been further optimized, but for the most part conditions developed during the laboratory optimization studies have been found to work well. This technology is currently being used to produce ton quantities of (S)-4. It has to be pointed out that no major difficulties were encountered during the scale-up. The work-up of the product in this process, unlike in many other enzyme-catalyzed processes, is quite simple and volume efficiencies are better than some chemical reactions. The recovery and recycling of the enzyme is not needed since it is commercially available and relatively inexpensive. 

Goswami A, Howell JM, Hua EY, et al. Chemical and enzymatic resolution of (R,S)-N-(tert-butoxycarbonyl)-3-hydroxymethylpiperidine. Organic Process Research and Development 5 4), 415, 2001. 

Palomo JM, Munoz G, Fernandez-Lorente G et al. (2003) Modification of Mucor miehei lipase properties via directed immobilization on different heterofunctional epoxy resins. Hydrolytic resolution of (R,S)-2-butyroyl-2-phenylacetic acid. J Mol Catal B Enzym 21 201-210 Palomo JM, Ortiz C, Fernandez-Lorente G et al. (2005) Lipase-lipase interaction as a new tool to immobilize and modulate the lipase properties. Enzyme Microb Technol 36 447-454 Park EY, Sato M, Kojima S (2006) Fatty acid methyl ester production using Upase-immobilizing silica particles with different particle sizes and different specific surface areas. Enzyme Microb Technol 39(4) 889-896... 

---