Bioreduction biocatalyst

Itoh, N., Nakamura, M., Inoue, K. and Makino, Y. (2007) Continuous production of chiral 1,3-butanediol using immobilized biocatalysts in a packed bed reactor promising biocatalysis method with an asymmetric hydrogen-transfer bioreduction. Applied Microbiology and Biotechnology, 75 (6), 1249-1256. 

Oxidoreductases are, after lipases, the second most-used kinds of biocatalysts in organic synthesis. Two main processes have been reported using this type of enzymes-bioreduction of carbonyl groups [39] and biohydroxylation of non-activated substrates [40]. However, in recent few years other processes such as deracemization of amines or alcohols [41] and enzymatic Baeyer-Villiger reactions of ketones and aldehydes [42] are being used with great utility in asymmetric synthesis. 

The bioreduction of carbonyl compounds with reductases has been exploited for many years, especially in the case of ketones, with baker s yeast Saccharomyces cerevisiae) being the most popular biocatalyst [45]. For instance, yeast treatment of 3-chloropropiophenone affords the expected (lS)-3-chloro-l-phenylpropan-l-ol, which was treated with trifluorocresol in tertrahydrofuran in the presence of tri-phenylphosphine and diethyl azodicarboxylate at room temperature to give (3R)-l-chloro-3-phenyl-3-[4-(trifluoromethyl)phenoxy]propane and the later reaction with methylamine leads to (R)-fluoxetine that is an important serotonin uptake inhibitor (Scheme 10.19) [46]. 

Although reductases play an important role in the in vivo synthesis of many chemicals (see flavour example in Fig. 7.11), little attention has been paid to this type of biocatalyst. In most cases whole microbial or plant cells are used to perform a bioreductive reaction due to the requirement for (expensive) cofactors. Typical examples include the reduction of certain double bonds in terpenes by plant cells [27,41], the reduction of Massoi lactone to R(+)-8-decalactone by Basidiomycetes and S. cerevisiae [28], and the baker s yeast-catalyzed reduction of ketones to (chiral) alcohols [42]. 

A surprisingly broad substrate specificity has been found for the enoate reductases from Clostridium tyrohutyritMm and Clostridium kluyveri [126], however, up to now, no preparative-scale bioreduction of a,p-unsaturated carboxylic acids using these enzymes in their purified form has been reported, most likely due to the technical difficulties in handling these sensitive biocatalysts. 

Figure 6.1 Schematic illustration of asymmetric hydrogen-transfer bioreduction with E. coli biocatalysts in IPA-aqueous solution system.

In this chapter, we describe development of novel enzymes including engineered PAR, LSADH, and P-keto ester reductase (KER) from P. citrinum key advances in tailoring biocatalysts by protein engineering and future aspect of them for improved bioreduction of ketones to synthesize various chiral alcohols. 

Inoue, K, Makino, Y., and Itoh, N. (2005) Purification and characteri2ation of a novel alcohol dehydrogenase from Ldfionia sp. strain S749 a promising biocatalyst for an asymmetric hydrogen transfer bioreduction. Appl. Environ. Microbiol., 71, 3633-3641. 

The use of organic solvents as reaction media for biocatalytic reactions can not only overcome the substrate solubility issue, but also facilitate the recovery of products and biocatalysts as well. This technique has been widely employed in the case of lipases, but scarcely applied for biocatalytic reduction processes, due to the rapid inactivation and poor stability of redox enzymes in organic solvents. Furthermore, all the advantages for nonaqueous biocatalysis can take effect only if the problem of cofactor dependence is also solved. Thus, bioreductions in micro- or nonaqueous organic media are generally restricted to those with substrate-coupled cofactor regeneration. 

Toxic substrates and products to whole-cell biocatalysts. Finally, in whole-cell format, the substrate and/or product of the bioreduction can be toxic to the cells, preventing cofactor regeneration. Such irreversible loss of regeneration capacity is, of course, catastrophic for the process. In principle, this can be overcome by maintaining a low substrate concentration, but this will ultimately prevent a sufficiently high product concentration for an effective process. In some cases, dependent upon the water-solubility (and if the substrate is a liquid), it may be possible to feed the substrate, such that a low concentration is provided to the cells in the reactor, but at the end of the reaction a high product concentration is achieved. However, in nearly all cases at the required concentration for an... 

For the aforementioned reasons the use of isolated enzymes is often preferred due to reduction in side reactions and higher productivities (see Ref. [12] for a review of this topic). However this brings other challenges such as the need for effective cofactor regeneration. The choice between enzyme and whole-cell biocatalysts is complex and requires more work in the future to establish a clearer strategy to help the process design and implementation of bioreductions. 

Numerous bioreduction examples use an organic solvent to form a second phase from extractive ISPR. For example, extractive ISPR in a two-phase system was used to remove 2-phenylethanol, giving an increase of one order of magnitude in product concentration and also productivity [41]. The two-phase system was ultimately limited by mass transfer on account of high viscosity of the chosen organic phase (oleic acid). In common with many such processes, it is clear that far more work is required on solvent selection. Interestingly the dissolved level of solvent in water is always exposed to the biocatalyst (whether the ISPR is internal or external) and in many cases this causes a detrimental effect to the biocatalyst over time. 

Prior to the widespread awdlabdity of recombiant carbonyl reductases enzymes, the use of microbial reductions using either actively growing or dormant cells was commonplace Bakers yeast in particular, was a readily available source of stereoselective carbonyl reductases enzymes. Even with the widespread knowledge of the power of recombinant CRED biocatalysts, the literature is still rife with wild-type whole-cell microbial reductions. The reductions presented have advanced well beyond the early Bakers yeast reduction and have an apphcation even today. When the whole-cell fermentation is developed and finely tuned, high titers of product alcohol are possible and Scheme 6.4 shows m example of a keto-amide 12 bioreduction performing at 100 g/L with more than 98% ee with multi-kg isolation [12]. The bioprocess was performed over 8 days at pH 7 using the yeast Candida sorbophila. 

Using traditional organic cosolvents, the bioreduction of para-bromo-2,2,2-trifluoroacetophenone 83 (Scheme 6.34) was limited to a maximum product concentration of 10 g/L due to substrate-induced deactivation of the R. eryth-ropolis (ADH RE) biocatalyst. However, by employing 10% (v/v) [BMP][NTf2], a water immiscible ionic liquid, more than 5 times the ketone was converted to the chiral alcohol in less than 24 h. The ionic liquid improved the initial reaction rate by more than four times in the presence of ionic liquid compared to an aqueous-only reaction. Moreover, the ionic liquid improved the stability of both the ADH RE and the GDH coenzyme compared to reactions with either organic cosolvents or aqueous buffer systems [59]. 

Enzymes with different stereochemical preferences for 3-oxo esters and 2-alkyl 3-0X0 esters have been isolated [57,62,63,68]. They are NADPH-dependent enzymes and are able to catalyze the reduction of oxo esters of different type. However, they are not available for enzyme-catalyzed reaction in substitution of the whole-cell catalyst. Synthetic applications make use of whole-cell biocatalysts. Valuable intermediates in synthesis are keto esters possessing additional functionality. Thus 5 -4-chloro-3-hydroxybutanoic acid 33 (Scheme 12) has been obtained by reduction with suspended cells from cultures of G. candidum. The compound is the intermediate in the synthesis of the cholesterol antagonist 34. In the biotransformation process a reaction yield of 95% and optical purity of 96% were obtained at 10 g/L. The optical purity was increased to 99% by heat treatment of cell suspensions prior to conducting the bioreduction [69]. 

---