engineered whole cell systems. Nonheme oxygenases and oxidases may help to
circumvent these limitations. The regio- and stereoselective oxidation of (poly)
hydroxy compounds, which is notoriously difficult with nicotinamide-dependent
dehydrogenases, is more efficiently performed with Cu- or flavin-dependent alcohol oxidases. The deracemization of pharmacologically important (cyclic) secamines can be achieved by combining enantioselective oxidation using amine
oxidases combined with non-stereoselective (chemical) reduction in a cyclic fashion on industrial scale.
The stereoselective formation of carbon–carbon bonds by means of aldolases
and transketolases is a well-researched standard technique for the synthesis of
carbohydrate-like α- and β-hydroxy ketones. It is noteworthy that the full set of
stereo-complementary aldolases are now available, which accept cheap
(non-phosphorylated) donors, such as dihydroxy acetone, pyruvate, acetaldehyde
and glycin. Along these lines, PLP-dependent lyases for acyloin and benzoin
condensations are gaining importance due to their lack of sensitive phosphorylated
cosubstrates.
Glycosyl transfer reactions using glycosidases and glycosyl transferases are
gaining ground as more of these enzymes are made available by genetic engineering. Non-phosphorylated cheap high-energy sugar donors, such as sucrose, currently supersede the expensive glycosyl phosphates previously employed. Within
only a few years, synthesis of enantiopure amines became feasible on industrial
scale by the advent of (R)- and (S)-selective transaminases.
The synthesis of optically active (R)- and (S)-cyanohydrins by hydroxynitrile
lyases is now well established. Several of these enzymes are commercially available and were successfully implemented in industry. The lyase-catalyzed addition
of water and ammonia onto activated C¼C bonds has little counterpart in traditional chemical catalysis and provides a highly desirable atom efficiency of 100%.
Unfortunately, the presently used hydratases and ammonia lyases are impeded by a
narrow substrate tolerance, which possibly might be circumvented by the design of
suitable mutants.
Although halogenation and dehalogenation reactions can be catalyzed by
enzymes, it is doubtful whether these reactions will be used widely, mainly because
the corresponding conventional chemical methods are highly competitive.
Emerging fields exploit the applicability of lesser-used enzymes, such as
cyclases and racemases [25], sulfatases, diol dehydratases, aldoxime dehydratases
and carboxylate reductases. Notoriously ‘dirty’ and inefficient Friedel-Crafts-alkylation and -acylations are feasible by employing alkyl- and acyl-transferases and
enzymatic carboxylation reactions are currently being investigated for the selective
formation of carbon–carbon bonds using CO 2 as a raw material for organic
synthesis [26].
The methodology concerning the employment of enzymes in nonaqueous
organic solvents with respect to enzyme activity is well understood and highly
predictable. Thus, the synthesis of esters, lactones, amides, peptides, and peracids
by using enzymes is standard methodology in industry. On the other hand, the
influence of the nature of organic solvents on an enzyme’s selectivity is still
insufficiently understood.
4 State of the Art and Outlook
409
circumvent these limitations. The regio- and stereoselective oxidation of (poly)
hydroxy compounds, which is notoriously difficult with nicotinamide-dependent
dehydrogenases, is more efficiently performed with Cu- or flavin-dependent alcohol oxidases. The deracemization of pharmacologically important (cyclic) secamines can be achieved by combining enantioselective oxidation using amine
oxidases combined with non-stereoselective (chemical) reduction in a cyclic fashion on industrial scale.
The stereoselective formation of carbon–carbon bonds by means of aldolases
and transketolases is a well-researched standard technique for the synthesis of
carbohydrate-like α- and β-hydroxy ketones. It is noteworthy that the full set of
stereo-complementary aldolases are now available, which accept cheap
(non-phosphorylated) donors, such as dihydroxy acetone, pyruvate, acetaldehyde
and glycin. Along these lines, PLP-dependent lyases for acyloin and benzoin
condensations are gaining importance due to their lack of sensitive phosphorylated
cosubstrates.
Glycosyl transfer reactions using glycosidases and glycosyl transferases are
gaining ground as more of these enzymes are made available by genetic engineering. Non-phosphorylated cheap high-energy sugar donors, such as sucrose, currently supersede the expensive glycosyl phosphates previously employed. Within
only a few years, synthesis of enantiopure amines became feasible on industrial
scale by the advent of (R)- and (S)-selective transaminases.
The synthesis of optically active (R)- and (S)-cyanohydrins by hydroxynitrile
lyases is now well established. Several of these enzymes are commercially available and were successfully implemented in industry. The lyase-catalyzed addition
of water and ammonia onto activated C¼C bonds has little counterpart in traditional chemical catalysis and provides a highly desirable atom efficiency of 100%.
Unfortunately, the presently used hydratases and ammonia lyases are impeded by a
narrow substrate tolerance, which possibly might be circumvented by the design of
suitable mutants.
Although halogenation and dehalogenation reactions can be catalyzed by
enzymes, it is doubtful whether these reactions will be used widely, mainly because
the corresponding conventional chemical methods are highly competitive.
Emerging fields exploit the applicability of lesser-used enzymes, such as
cyclases and racemases [25], sulfatases, diol dehydratases, aldoxime dehydratases
and carboxylate reductases. Notoriously ‘dirty’ and inefficient Friedel-Crafts-alkylation and -acylations are feasible by employing alkyl- and acyl-transferases and
enzymatic carboxylation reactions are currently being investigated for the selective
formation of carbon–carbon bonds using CO 2 as a raw material for organic
synthesis [26].
The methodology concerning the employment of enzymes in nonaqueous
organic solvents with respect to enzyme activity is well understood and highly
predictable. Thus, the synthesis of esters, lactones, amides, peptides, and peracids
by using enzymes is standard methodology in industry. On the other hand, the
influence of the nature of organic solvents on an enzyme’s selectivity is still
insufficiently understood.
4 State of the Art and Outlook
409