Reducciones asimétricas y oxidaciones selectivas catalizadas por deshidrogenasas inmovilizadas y estabilizadas

Resumen

In many industrial fields such as fine chemistry, pharmacy, cosmetology, agriculture, food, etc., the need for safer and purer products leads to the use of exquisite selective processes. This goal may be achieved by exploiting enzymes as catalysts since their excellent properties. Enzyme technology in the last decades has enabled the enzyme to catalyze industrial reaction due to the numerous advances in enzyme isolation, production, purification, stabilization, as well as the great efforts done on process design to optimize their use. Dehydrogenases (DHs), which depend on nicotinamide cofactors, are among the most interesting enzymes in biocatalysis because they perform selective reductions and specific oxidations that may be key steps in the synthesis routes for fine chemicals, such as pharmaceuticals, food additives, etc. The search of new DHs has been enormously encouraged in the last times resulting in an increased number of available ones. On the other hand, due to the high costs of the nicotinamide cofactors, their stoichiometric use is not acceptable from an economical point of view. For these reasons industrial application of these enzymes requires efficient in situ replenishment of the concomitant cofactor in its right oxidation state. Therefore, the use of simple and low cost purification systems, the use of enzymes from thermophilic microorganisms and the development of cost-effective systems for nicotinamide cofactors regeneration would overcome some of the limitation of DHs in order to apply them industrially. The main objective of this PhD Thesis project has been designing and developing bi-enzymatic systems using DHs with in situ enzymatic regeneration of the corresponding redox cofactors. These bi-enzimatic systems were formed by two DHs, one catalyzing the main reaction and the other acting as cofactor recycling partner. Thus, in this work has described the purification, characterization, stabilization by immobilization on solid supports and the reactivation of two new DHs from Thermus thermophilus HB27 recombinantly expressed in E.coli: an alcohol dehydrogenase (Tt27-ADH2) and NADH oxidase (Tt27-NOX). In one hand, the new short-chain alcohol Tt27-ADH2 has been described to catalyze both oxidative and reductive reactions at neutral pH with a broad range of substrates. Its highest activity was found towards the reduction of 2,2¿,2¿¿-trifluoroacetophenone (85 U/mg). Moreover, the enzyme was stabilized more than 200-fold by multipoint covalent immobilization on agarose matrixes via glyoxyl chemistry. Such heterogeneous catalyst was coupled to an immobilized cofactor recycling partner performed the quantitative reduction of 2,2¿,2¿¿-trifluoroacetophenone and (rac)-2-phenylpropanal to (S)-(+)-¿-(trifuoromethyl)benzyl alcohol and (R)-(+)-2-phenyl-1-propanol with enantiomeric excesses of 96 % and 90% respectively. Such activity at low temperature along its anti-prelog selectivity for pro-chiral aryl-ketones and its preference for the (R)-2-phenylpropanal enhances its potential use as selective catalyst under mild operation conditions often required in fine chemistry. On the other hand, the Tt27-NOX is a flavoenzyme which needed exogenous flavin cofactor (FAD or FMN) to reach its optimal performance. Interestingly, this enzyme presented a catalytic efficiency 6-fold higher than its counterpart in strain HB8. This enzyme also performs redox reaction at low temperatures showing high specific activity values (70 U/mg) despite being isolated from a thermophile source. Immobilization of Tt27-NOX on agarose matrixes via glyoxyl chemistry yielded the most stable enzyme preparation (until 300-fold more stable than soluble enzyme). The immobilized derivative was able to be reactivated under physiological conditions after inactivation by high solvent concentrations. The heterogeneous and robust biocatalyst was used as recycling partner in the kinetic resolution of (rac)-1-phenylethanol. The high stability along with its capability to be reactivated makes this biocatalyst suitable for cofactor recycling in redox biotransformations. In this work, we also purified and stabilized a recombinant glycerol dehydrogenase from Citrobacter braakii (Cb-GyDH). The enzyme oxidizes glycerol to dihydroxyacetone with a specific activity of 65 U/mg but undergoes a strong non-competitive inhibition by low concentrations of the corresponding oxidation product. The immobilized enzyme on agarose heterofunctionally activated with both glyoxyl and metal chelate groups showed an inhibition 1.7 times lower than its soluble counterpart. Moreover, this immobilized preparation was up to 90-fold more stable than the soluble form. Finally, we have optimized bi-enzymatic redox systems using different enzyme immobilization strategies in order to minimize the amount of soluble cofactor needed to perform different bio-redox reactions. The two main approaches used in this work were: the immobilization of each enzyme involved in the bi-enzymatic system on two different carriers, and the co-immobilization of the two enzymes on the same carrier. In this regard, we also optimize the enzyme distribution within the porous structure of the co-immobilized biocatalysts. In all cases, the co-immobilization of the two protein results in better cofactor recycling efficiencies than when the two enzymes were separately immobilized on two different carriers. In principle, a system where both enzymes are co-immobilized seems to be more favorable for efficiently regenerating the cofactor pool in its right oxidation state. Nevertheless, the co-immobilization itself does not guarantee the success of the bi-enzyme system. Hence, we have optimized the co-immobilization chemistry for each of the three bio-redox systems here studied. For one specific case (one thermophilic system), NADH is recycled 4600 times using 50 microeq of cofactor per equivalent of substrate at 55 ºC. Moreover, it has been found that uniform distribution of both DHs across the porous surface enhances the cofactor recycling by 1.5-fold factor, likely because of vicinal cooperation effects. Therefore, the rational design of the co-immobilization strategy of different biredox systems formed by two enzymes may be extended to other biocatalytic cascades, opening a window for the optimization of others multi-enzyme biotransformations where cofactor recycling is needed.