SEARCH

SEARCH BY CITATION

Keywords:

  • 2-hydroxypropiophenone;
  • acetoin;
  • acyloins;
  • asymmetric synthesis;
  • benzoin;
  • catalysis;
  • phenylacetylcarbinol;
  • protein engineering;
  • α,α′-dihydroxyketone;
  • α-hydroxyketone

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Enzymes from the DC family
  5. Engineering transketolases
  6. Conclusion
  7. Acknowledgements
  8. References

Thiamine diphosphate-dependent enzymes are broadly distributed in all organisms, and they catalyse a broad range of C–C bond forming and breaking reactions. Enzymes belonging to the structural families of decarboxylases and transketolases have been particularly well investigated concerning their substrate range, mechanism of stereoselective carboligation and carbolyase reaction. Both structurally different enzyme families differ also in stereoselectivity: enzymes from the decarboxylase family are predominantly R-selective, whereas those from the transketolase family are S-selective. In recent years a key focus of our studies has been on stereoselective benzoin condensation-like 1,2-additions. Meanwhile, several S-selective variants of pyruvate decarboxylase, benzoylformate decarboxylase and 2-succinyl-5-enolpyruvyl-6-hydroxy-3-cyclohexene-1-carboxylate (SEPHCHC) synthase as well as R-selective transketolase variants were created that allow access to a broad range of enantiocomplementary α-hydroxyketones and α,α′-dihydroxyketones. This review covers recent studies and the mechanistic understanding of stereoselective C–C bond forming thiamine diphosphate-dependent enzymes, which has been guided by structure–function analyses based on mutagenesis studies and from influences of different substrates and organic co-solvents on stereoselectivity.


Abbreviations
AHAS

acetohydroxy acid synthase

BAL

benzaldehyde lyase

BFD

benzoylformate decarboxylase

CDH

cyclohexane-1,2-dione hydrolase

DC family

decarboxylase family

FBA

formylbenzoic acid

HPA

β-hydroxypyruvate

HPP

hydroxypropiophenone, 1-phenyl-2-hydroxy-propan-1-one

KDH

α-ketoglutarate dehydrogenase

Kgd

α-ketoglutarate decarboxylase from Mycobacterium tuberculosis

MenD

2-succinyl-5-enolpyruvyl-6-hydroxy-3-cyclohexene-1-carboxylate synthase

PAC

phenylacetylcarbinol, 1-phenyl-1-hydroxypropan-2-one

PDC

pyruvate decarboxylase

PP

pyrophosphate

PYR

pyrimidine

SucA

α-ketoglutarate decarboxylase subunit of the α-ketoglutarate dehydrogenase from Escherichia coli

ThDP

thiamine diphosphate

TK

transketolase

wt

wild-type

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Enzymes from the DC family
  5. Engineering transketolases
  6. Conclusion
  7. Acknowledgements
  8. References

Thiamine diphosphate (ThDP) dependent enzymes are found in all organisms. Classified as lyases, transferases and oxidoreductases they catalyse a broad range of C–C, C–N, C–S, C–O ligase and cleavage reactions [1-4]. The currently known enzymes can be grouped into eight structural families with similar domain structures in each structural family (see below) but of very low sequence similarity. An overview of all annotated sequences can be found in the ThDP-dependent Enzyme Engineering Database (TEED, http://www.teed.uni-stuttgart.de) [5]. Among these, enzymes catalysing the cleavage and formation of C–C bonds have been most intensively studied [6-9]. Here, the potential to catalyse the formation of mixed carboligation products from a donor substrate and an acceptor substrate in a highly chemoselective and stereoselective manner is of special interest. Among the enzymes that have been most intensively studied are transketolase (TK) from Escherichia coli and Saccharomyces cerevisiae [10-16], acetohydroxy acid synthase (AHAS) isoenzymes I–III from E. coli [17-21], benzoylformate decarboxylase (BFD) from Pseudomonas putida [22-29], benzaldehyde lyase (BAL) from Pseudomonas fluorescens [22, 25-27, 29-39], pyruvate decarboxylase (PDC) from S. cerevisiae, Zymomonas mobilis and Acetobacter pasteurianus [24, 40, 41], branched chain ketoacid decarboxylase from Lactococcus lactis [42, 43], cyclohexane-1,2-dione hydrolase from Azoarcus sp. [44-46] and 2-succinyl-5-enolpyruvyl-6-hydroxy-3-cyclohexene-1-carboxylate (SEPHCHC) synthase (MenD) from E. coli [47-49]. In addition to MenD, two further α-ketoglutarate accepting enzymes were studied recently that represent the α-ketoglutarate decarboxylase part of the α-ketoglutarate dehydrogenase (KDH) complex and activity: SucA from E. coli and Kgd from Mycobacterium tuberculosis [49, 50].

Although the physiological activities of these enzymes are very broad (Table 1), all of them catalyse the non-physiological 1,2-addition of a donor (aldehyde, α-keto acid, β-hydroxy-α-keto acid, α-hydroxyketone) and an acceptor (aldehyde, α-keto acid). Recently, the first ThDP-dependent enzyme accepting ketones as acceptors was identified: YerE from Yersinia pseudotuberculosis [51, 52]. The reaction mechanism is shown in Scheme 1.

Table 1. ThDP-dependent enzymes described in this review. Assignment to structural families and domain organization of respective subunits are according to [5]. DC family, PYR-TH3-PP; KDH family, ODH-PP-PYR; TK family, PP-PYR-TKC; n.d., no data available.
Enzyme/EC numberOrganismPhysiological reactionStructural familyPDB entries (variant)Selected references
  1. a

    Crystal structure from Kgd from Mycobacterium smegmatis.

AHAS

Escherichia coli Biosynthesis of branched-chain amino acidsDC n.d. [55, 56]

BAL

Pseudomonas fluorescens UnknownDC2AG0, 2AG1, 3IAF (A28S), 3IAE (A28S), 2UZ1, 3D7K [57, 58]

BFD

Pseudomonas putida Mandelate catabolismDC1BFD, 1MCZ, 3FZN, 2FWN, 3FSJ, 1YNO, 3F6E, 3F6B, 1PO7 (E28A), 1Q6Z (E28A), 1PI3 (E28Q), 2V3W (L461A), 2FN3 (S26A) [59-61]

ZmPDC

Zymomonas mobilis Alcoholic fermentationDC1ZPD, 2WVH, 2WVG, 2WVA, 3OE1 (E473D) [62, 63]

ApPDC

Acetobacter pasteurianus Oxidative lactic acid metabolismDC 2VBI [41, 64]

MenD

Escherichia coli Biosynthesis of menaquinoneDC2JLC, 2JLA, 3FLM, 3HWX, 3HWW [65-68]
YerE Yersinia pseudotuberculosis Biosynthesis of yersiniose ADCn.d. [51]

CDH

Azoarcus sp.UnknownDC2PGN, 2PGO [45, 46]

SucA

Escherichia coli Citric cycleKDHn.d. [49]

Kgd

Mycobacterium tuberculosis Citric cycleKDH2XT6,a 2XT9,a 2XTA,a 2Y0P,a 2YIC,a 2YID a [50]
EcTK Escherichia coli Calvin cycle; pentose phosphate pathwayTK1QGD, 2R8O, 2R8P [69-71]

ScTK

Saccharomyces cerevisiae Calvin cycle; pentose phosphate pathwayTK1TRK, 1GPU, 1AY0, 1NGS, 1TKA, 1TKB, 1TKC, 1TRK [11, 72]
image

Scheme 1. Reaction mechanism of ThDP-dependent enzymes. The donor substrates bind covalently to C2 of the ThDP ylide and are subsequently deprotonated (for donor aldehydes) or decarboxylated (for the corresponding α-keto acids) to give the same ThDP-bound ‘activated aldehyde’ (carbanion-enamine) intermediate. The same intermediate is obtained by BAL-catalysed cleavage of 2-hydroxyketones, and has nucleophilic reactivity at the carbon atom that had been part of the carbonyl group in the donor substrate [53].

Download figure to PowerPoint

The donor substrates bind covalently to C2 of the thiazolium ring of ThDP and are subsequently deprotonated (for donor aldehydes) or decarboxylated (for the corresponding α-keto acids) to give the same ThDP-bound activated aldehyde intermediate, which now has nucleophilic reactivity at the carbon atom that had been part of the carbonyl group in the donor substrate [53].

All the enzymes mentioned above belong to three different structural families, each containing three different domains per subunit (Table 1). They all have the typical pyrimidine (PYR) binding domain and the pyrophosphate (PP) binding domain in common. Both domains, but from different subunits, bind one ThDP molecule via a divalent metal ion (usually Mg2+), which complexes the diphosphate moiety of the cofactor with conserved residues in the enzyme active site [2, 54]. The third domain (called TH3, ODH and TKC; for domain organization in the different structural classes see the caption of Table 1) is usually not directly involved in catalysis. Thus, the smallest functional unit of these enzymes is a dimer and the enzymes given in Table 1 are dimers or tetramers in their native state.

This review focuses on studies concerning the stereoselectivity of enzymes from the DC family and the TK family. Recent achievements in the design of enantiocomplementary enzyme variants are described based on a structure–function based mechanistic understanding and data from the use of different substrates and organic co-solvents.

Enzymes from the DC family

  1. Top of page
  2. Abstract
  3. Introduction
  4. Enzymes from the DC family
  5. Engineering transketolases
  6. Conclusion
  7. Acknowledgements
  8. References

Physiological activities

AHAS [18, 55, 56] and MenD [65, 73] catalyse a physiological C–C bond formation. In all cases the reaction starts with the initial decarboxylation of pyruvate (AHAS) or α-ketoglutarate (MenD), respectively. The resulting aldehyde moiety, activated by an umpolung through binding to the cofactor ThDP, is subsequently transferred to an acceptor.

In the case of AHAS a 1,2-addition of acetaldehyde to pyruvate or 2-keto butyrate results in the formation of (S)-acetolactate and (S)-acetohydroxybutyrate, respectively. MenD catalyses a 1,4-addition of succinylsemialdehyde, the decarboxylation product of α-ketoglutarate, to isochorismate (Scheme 2).

The natural activity of PDCs and BFD is the decarboxylation of pyruvate and benzoylformate to the respective aldehydes that are liberated upon protonation of the hydroxyethyl-ThDP or hydroxybenzyl-ThDP, respectively (Table 1). However, if another aldehyde is offered as the acceptor 1,2-addition yielding α-hydroxyketones occurs, which is discussed in more detail below (for further details see [3, 9]).

image

Scheme 2. (A) Selected physiological activities of MenD, YerE and CDH and (B) non-physiological activity of YerE.

Download figure to PowerPoint

In contrast, the physiological reaction of BAL is not yet known. Up to now, BAL with reasonable activity has only been found in P. fluorescens [58]. By comparison with all other enzymes mentioned, wild-type (wt) BAL does not catalyse the decarboxylation of α-keto acids to a significant extent [74, 75]. In addition to the 1,2-addition of aldehydes [37] BAL catalyses the cleavage of (R)-benzoins and araliphatic (R)-α-hydroxyketones [76].

Recently, Lehwald et al. presented the first example of an asymmetric intermolecular crossed aldehyde–ketone carboligation reaction using a ThDP-dependent enzyme. In the biosynthetic pathway of yersiniose A, which is a two-carbon branched-chain 3,6-dideoxyhexose isolated from Yersinia pseudotuberculosis, the ThDP-dependent flavoenzyme YerE catalyses the decarboxylation of pyruvate and the transfer of the ‘activated acetaldehyde’ to the carbonyl function of a 3,6-dideoxy-4-oxo-d-glucose (Scheme 2). Detailed investigation of the substrate range of the recombinant enzyme revealed that many non-physiological carbonyl compounds are accepted as substrates by YerE. The broad acceptor substrate range of the enzyme includes (besides aliphatic and aromatic aldehydes) cyclic and open-chain ketones, as well as diketones and α-keto esters [52].

Cyclohexane-1,2-dione hydrolase (CDH) from Azoarcus sp., like YerE, also accepts cyclohexane-1,2-dione as a substrate. In contrast to YerE, however, CDH catalyses a hydrolysis reaction, whereas an enantio-enriched tertiary alcohol was obtained with YerE [44, 45] (Scheme 2).

Non-physiological 1,2-carboligation

Despite their different physiological activities, all enzymes mentioned above catalyse a benzoin condensation-like 1,2-addition of two aldehydes yielding α-hydroxyketones. Inspired by the pioneering application of yeast PDC for the production of (R)-1-phenyl-1-hydroxypropan-2-one [(R)-PAC] in a fermentative process [77], different PDCs from bacteria and yeasts have been applied. Only recently the conversion of ZmPDC into an effective enantioselective carboligase was described. This was achieved by one single point mutation of a glutamate residue that is relevant for protonation of the carbanion/enamine intermediate [40, 78]. A variant of ScPDC with mutation at the corresponding glutamate residue (ScPDC-E477Q) was described earlier by Jordan and coworkers for stereoselective (R)-PAC synthesis [79].

BFD and BAL were thoroughly studied with regard to the carboligation of benzaldehyde and acetaldehyde (and their derivatives) to the respective α-hydroxyketones (Table 2).

Table 2. α-Hydroxyketones accessible by R- and S-selective 1,2-carboligations catalysed by ThDP-dependent enzymes.
ProductsR-specific enzyme (variant) [references]S-specific enzyme (variant) [references]
  1. a

    See Table 3 for details.

image_n/febs12496-gra-0001.png BAL [30, 34]

BFD-L476Q [23]

image_n/febs12496-gra-0002.png BAL [31, 34, 39, 87]BFD-L461A/Ga [28]
image_n/febs12496-gra-0003.png

ApPDC [41]

ZmPDC [88, 89]

ZmPDC-E473Q [40] ScPDC-E477Q [79] ScPDC-D28A [79] AHAS I + II [17, 90]

ApPDC-E469Ga [41]
image_n/febs12496-gra-0004.png MenD [47, 49]

MenD-I474A/F475Ga [91]

MenD-I474A/F475G/R395Ya [92]

image_n/febs12496-gra-0005.png

BFD-H281A [22, 30]

BAL (kinetic resolution) [76]
image_n/febs12496-gra-0006.png

ScPDC-E477Q [79]

ScPDC-D28A [79]

SucA [48]

BAL, BFD [25, 26]

PDHc-E1-E636A/Q [79]

CDH [48]

BAL (propioin) [25]

With the exception of BFD, all enzymes from the DC family are highly R-selective for this non-physiological reaction, if at least the donor or the acceptor, respectively, is aromatic (Table 2). BFD is S-selective for the formation of 2-hydroxypropiophenone (2-HPP), originating from benzaldehyde as the donor and acetaldehyde as the acceptor, but R-selective with acceptors larger than acetaldehyde [80-83].

In the case of aliphatic carboligation products only limited enantiomeric excesses (ees) could be obtained [25, 84, 85], which is in line with the fact that optimal stabilization of both donor and acceptor gives the best results with respect to enantioselectivity and conversion (as outlined below in more detail). The enzymatic formation of acetoin can occur via different routes that are discussed below and explain the usually low enantioselectivity for this carboligation product [48].

Although the sequence similarity among these enzymes is only about 20%, the highly conserved structures lead to similar active site architectures (Fig. 1). As demonstrated in Fig. 1, the enzymes vary in the size of both the donor binding site and the acceptor binding site. Concerning the latter, two orientations of the acceptor are possible yielding (R)- or (S)-α-hydroxyketones [7]. In most of the wt enzymes only the parallel arrangement of the acceptor relative to the donor is possible, because the antiparallel arrangement is sterically hindered by amino acid residues in the so-called S-pocket (Fig. 1A). This mechanistic model, which explains the stereoselectivity of ThDP-dependent enzymes of the DC family, was initially developed based on the R-selective BAL and BFD, which show S-selectivity only with acetaldehyde as the acceptor [32] (Fig. 1B). Since then, the model has been supported by structure-based site-directed mutagenesis studies on PpBFD [28], ApPDC [41] and MenD [91, 92] with several variants and different substrates (Fig. 1). The model is based on the assumption that the hydroxy group of the ThDP-bound donor and the carbonyl oxygen of the acceptor must point in the direction of amino acid side chains mediating proton transfer during the catalytic cycle (for details see [53] and Table 3). In the structural class of decarboxylases amino acid residues with respective charged side chains at an appropriate distance are rare, making the orientation of both oxygen functions in the same direction [both directed towards the same side chain(s)] most probable.

image

Figure 1. Active site architecture of different enzymes from the DC family that were studied with respect to structure–function aspects concerning stereoselectivity of the carboligation. The grey areas in the schematic presentations on the left represent the position of the protein backbone. Residues important for stereoselectivity are indicated with the respective sequence number (based on the wt enzyme) and the standard number (in parenthesis) according to the standard numbering scheme for ThDP-dependent decarboxylases [95]. In all cases the donor binding site is occupied with the respective aliphatic or aromatic donor bound to ThDP. The acceptor binding site is specifically shown for BAL (A, parallel orientation relative to the donor) and BFD (B, antiparallel orientation relative to the donor). Antiparallel arrangement of donor and acceptor prior to C–C bond formation is only possible if an S-pocket of sufficient size is available. In the wt enzymes, except BAL, the S-pockets are blocked by bulky side chains (red) and in case of ApPDC (C), MenD (D) and AHASII (E) bulky side chains are located at the entrance of the S-pocket (blue).

Download figure to PowerPoint

With the isoenzymes AHAS I and II from Ecoli analogous studies were performed based on a homology model [93] that was generated using the crystal structure of AHAS from S. cerevisiae [94]. The model revealed a potential S-pocket in both isoenzymes. However, the size of these pockets seemed to be too small to stabilize benzaldehyde binding optimally. This was supported by site-directed mutagenesis studies at respective positions of the two isoenzymes. Whereas AHAS I was inactivated by mutations in the S-pocket region, AHAS II proved to be much more robust towards mutations. By mutation of the neighbouring residues G459, M460 and V461 (Fig. 1E) the best results were obtained with the variant G459A/M460A/V461G yielding (R)-PAC with 48% ee demonstrating that benzaldehyde may react via the S-pathway but that the R-pathway is still preferred [93]. More detailed information about the residues lining the acceptor binding site of these five enzymes is given in Table 3.

Table 3. Amino acid residues lining the substrate channel and the active site of ThDP-dependent enzymes mentioned in this review. Residues in the same line are structurally similar with the same standard number [95], if not otherwise indicated. The respective areas are indicated in Fig. 1.
Standard number BAL BFDApPDCMenDAHAS IIb
  1. aResidues are located in a similar position but the respective side chains are oriented away from the S-pocket. b Based on homology modelling using AHAS from S. cerevisiae (1N0H) as template. c Red-marked position in Fig. 1. d Blue-marked position in Fig. 1. e Proton transfer may require a water molecule.

Residues lining the S-pocket
26H26P24A25P30P23
27G27G25G26G31G24
28A28S26D27S32G25
477cT481L461E469F475V461
480F484aF464I472aL478aW464
Entrance to the S-pocket
476dA480A460I468I474M460
Potential proton donors and acceptors (proton relay system)
28 S26D27e  
29H29e    
30 E28e   
73T73eH70T72eT78eT70e
114  H113  
115Q113 H114Q118eQ110e
292  Y290  
477  E469  

Apart from the S-pocket, a second binding site for the acceptor aldehyde was proposed in ApPDC [41]. However, this alternative pathway to S-specific carboligation is blocked by W388. Mutation of W388 by alanine opened the pathway and resulted in significant amounts of (S)-PAC.

Tailoring S-selective variants

Using a structure-based enzyme design approach, mutagenesis studies to enlarge the S-pocket of BFD and to open those of ApPDC, MenD and AHAS (Fig. 1B–E) were conducted (Table 4). For BFD an enlargement of the S-pocket by one mutation at position L461 (Fig. 1B, standard position 477) resulted in an improved ee for 2-HPP from 92% ee (S) observed with BFDwt to 98% ee (S) with the variants L461A and L461G, respectively. Furthermore, the larger S-pocket in these variants provided sufficient space also to properly bind the side chains of propanal and methoxyacetaldehyde yielding the respective (S)-2-hydroxyketones with 93%–97.5% ee (Table 4). Compared with BFD, ApPDC contains a larger potential S-pocket based on the position of the protein backbone; however, it is not accessible because it is blocked by a large glutamate residue. Opening the S-pocket in ApPDC for benzaldehyde by only one mutation of E469 (Fig. 1C, Table 4) provided access to (S)-PAC derivatives with larger aliphatic side chains, reaching ee values of 85–89% under standard conditions. In contrast to BFD and ApPDC, where one mutation was sufficient to almost completely invert stereoselectivity, one additional mutation was necessary in the case of EcMenD to achieve this goal. Here, in addition to the F475G exchange also Ile474 (standard number 476, Fig. 1D) at the entrance of the S-pocket must be substituted by glycine and alanine, respectively, in order to completely open the S-pocket for benzaldehyde derivatives [91]. With the double variant EcMenD-I474A/F475G moderate to very good (S)-enantiopurities were obtained for various benzaldehyde derivatives (Table 4), with meta-substituted benzaldehydes giving the best results. Further improvement of this variant was possible by targeted destabilization of the R-pathway. Therefore, saturation mutagenesis at position 395 (standard position 393) was performed in order to replace the arginine, which is assumed to stabilize benzaldehyde in parallel orientation by π-stacking effects [96]. From the mutant library MenD variants with the highest (S)-stereoselectivity contained a tyrosine in position 395. The positive effect of this mutation, which increased the stereoselectivity for the carboligation of benzaldehyde and (decarboxylated) α-ketoglutarate from 75% to 85% ee (S), can be explained as follows. R395 may stabilize parallel oriented benzaldehyde through π-stacking effects. Replacement by tyrosine abolishes such stabilizing effects, thus destabilizing the R-pathway and thereby enhancing the S-selectivity of the reaction [92].

Table 4. Current examples of S-selective variants of enzymes from the DC family.
VariantSelected reaction/main productReferences
BFD-L461A/G image_n/febs12496-gra-0007.png [28]
ApPDC-E469G image_n/febs12496-gra-0008.png [41]

MenD- I474A/F475G

MenD- I474A/F475G/R395Y

image_n/febs12496-gra-0009.png [91, 92]

Our results on engineering S-selectivity in the structural family of decarboxylases clearly demonstrated that stereoselectivity is predominantly controlled by steric effects. The first requirement to achieve S-selectivity is sufficient space for the respective side chain of the acceptor aldehyde in the S-pocket to allow its antiparallel arrangement relative to the donor (Fig. 1). The better the acceptor is stabilized in the S-pocket, the higher is the probability that carboligation occurs via the S-pathway. This stabilization can be further improved by selection of the appropriate substrate. The population of the S-pathway can be enhanced even more by targeted destabilization of the R-pathway. In many cases improved stereoselectivity was correlated with the highest conversion, indicating that optimal stabilization of the substrates has a positive effect also on the enzyme activity.

Aliphatic α-hydroxyketones

As mentioned above, stereoselectivity is often low if aliphatic donors and acceptors are used. This has been demonstrated for example in thorough studies with BAL and BFD [25], where acetaldehyde, propanal, n-butanal, n-pentanal and isovaleraldehyde were transformed to the respective acyloins. With the exception of propioin, which was formed with up to 53% ee for the S-enantiomer, both enzymes catalysed the formation of the R-enantiomers in excess, yielding highest enantioselectivity for the carboligation of isovaleraldehyde [89% ee (R)]. Besides, application of both enzymes in a gas-phase reactor yielded propioin with only very low ee [19% (S)] [26]. However, studies of Jordan and coworkers demonstrated that acetoin and acetolactate can be accessed in an enantiocomplementary fashion with a variant of ScPDC-E477Q yielding (R)-acetoin and a variant of the E1 subunit of the pyruvate dehydrogenase complex (PDHc-E1-E636A/Q) yielding the S-product [79] (Table 2).

The reasons for the often low optical purity of enzymatically formed acetoin with different enzymes were elucidated in detail. A combination of mechanistic studies of the different enzymes MenD, SucA, CDH and YerE resulted in (a) identification of a (non-enzymatic) α-hydroxy-β-keto acid rearrangement–decarboxylation of enzymatically formed acetolactate derivatives, and (b) a concise strategy for the enzymatic synthesis of highly enantio-enriched (R)- and (S)-acetoin [48].

Incubation of α-ketoglutarate with acetaldehyde and MenD led to the formation of (almost racemic) 5-hydroxy-4-oxohexanoate as the sole product (Scheme 3A). Incubation of α-ketoglutarate with pyruvate showed that the first formed intermediate is 2-hydroxy-2-methyl-3-oxohexanedioic acid. This, however, is transformed through a non-enzymatic rearrangement–decarboxylation to racemic 4-hydroxy-5-oxohexanoate (Scheme 3B). Thus, the change in regioselectivity induced by varying the substrate(s) in ThDP-dependent enzymatic transformations may not necessarily be due to swapping the acyl donor–acceptor roles of the substrates.

α-Keto acids, such as pyruvate in the above mentioned example, are frequently accepted as acyl acceptor substrates in reactions catalysed by different ThDP-dependent enzymes. Therefore, α-alkylated α-hydroxy-β-keto acids are often formed and decarboxylations similar to that described above can be expected. With this in mind, three further ThDP-dependent enzymes have been investigated.

image

Scheme 3. (Chemo)enzymatic formation of regioisomeric acyloins and enzymatic synthesis of highly enantio-enriched acetoin [48]. All biotransformations require the addition of ThDP and magnesium ions to the reaction buffer.

Download figure to PowerPoint

First, pyruvate was incubated with α-ketobutyrate in the presence of ThDP-dependent YerE. This transformation initially led to the formation of (S)-acetolactate (from two molecules of pyruvate) and (S)-acetohydroxybutyrate (from pyruvate as donor and α-ketobutyrate as acceptor). NMR data showed that α-ketobutyrate is the acceptor slightly preferred by the enzyme. When this reaction was followed for a longer time, it was found that, besides acetoin, two acyloin regioisomers were also formed in a 60:40 ratio as the final products (Scheme 3C).

Incubation of α-ketoglutarate and pyruvate with SucA led to similar results to those observed with MenD with respect to the regioselectivity. Additionally, SucA catalysed the formation of acetoin using either pyruvate or acetaldehyde as the sole substrate. Incubation of pyruvate with SucA led, via acetolactate, to almost racemic acetoin with only 8% ee of the (R)-isomer (Scheme 3D), but the ee increased to 90% when acetoin was prepared using acetaldehyde as substrate (Scheme 3E).

The third enzyme, CDH from Azoarcus sp., showed no formation of acetolactate as an intermediate, starting from pyruvate either as the sole substrate or in combination with acetaldehyde. Accordingly, highly enantio-enriched (S)-acetoin (up to 90% ee) was directly obtained using pyruvate as the sole substrate. This can be explained by enzymatic decarboxylation of pyruvate to give acetaldehyde, which then acts as the acyl acceptor substrate giving (S)-acetoin. Hence, CDH makes highly enantio-enriched acetoin from pyruvate as the sole substrate because it is incapable of producing acetolactate.

These studies highlight that using members of different enzyme subfamilies resulted in synthetically viable pathways to (R)- and (S)-acetoin, but also that the (non-enzymatic) α-hydroxy-β-keto acid rearrangement–decarboxylation of acetolactate derivatives identified will be of importance for all enzymatic carboligations with α-ketoacid substrates and/or acetolactate-like products [48].

Influence of reaction conditions on chemoselectivity and stereoselectivity

Apart from tailoring the active site, the influence of the reaction conditions should also be considered in stereoselectivity studies with enzymes. In the case of ThDP-dependent enzymes different effects on activity and stereoselectivities have been observed depending on the reaction system [25, 29, 97]. The positive effect of organic co-solvents such as dimethylsulfoxide and acetone on the enzyme activity was studied earlier with BAL and the BFD variant H281A concerning their influences on kinetic microscope reaction constants of benzoin formation [98].

In order to elucidate specifically the influence of organic co-solvents on a broader basis, six ThDP-dependent enzymes from the DC family were recently studied in the presence of 13 water-miscible organic solvents under equivalent reaction conditions [85]. These were used in the carboligation reaction of benzaldehyde and acetaldehyde to yield four different enantiomeric products in different ratios: acetoin, benzoin, PAC, 2-HPP. This study included the four wt enzymes ApPDC [84], BFD [84], BAL [99] and the branched-chain ketoacid decarboxylase from Lactococcus lactis (LlKdcA) [43], and two enzyme variants: the S-selective ApPDC variant E469G [41] and the BFD-H281A variant that was specifically designed for enhanced formation of (R)-benzoin [22]. The results were evaluated with respect to influences of different concentrations of these solvents on the chemoselectivity and stereoselectivity.

The influence of additives on the stereoselectivity was found to be very pronounced and followed a general trend. If the enzyme stereoselectivity in aqueous buffer was already > 99.9% ee, it was not influenced by any of the co-solvents tested. However, both stereoselectivity and chemoselectivity were strongly influenced by the co-solvents if the enzyme was rather unselective in aqueous buffer reactions. In all the reactions where an increased product concentration was detectable, dimethylsulfoxide was one of the improving solvents.

For the enzyme with the largest S-pocket (ApPDC-E469G), a general correlation of the solvent-excluded volume of the co-solvent molecule with the observed changes in stereoselectivity was demonstrated: the smaller the organic solvent molecule, the higher the impact of this additive. Further, a correlation of the solvent's log P with stereoselectivity was observed. Calculations revealed a direct interaction of organic solvent molecules with the active site, specifically the S-pocket, which can be used to explain why the S-selective ApPDC-E469G variant loses S-selectivity in the presence of organic solvents of an appropriate size to enter the S-pocket. This variant catalysed the synthesis of (S)-PAC with 87% ee in buffer. Addition of 0.5 vol% trichloromethane to the reaction mixture yielded predominantly (R)-PAC (49% ee). The same trend was observed with acetoin, which is formed with an ee of 85% (S) in buffer but with only 28% ee (S) in the presence of 30 vol% ethanol. This work demonstrated the potential of solvent engineering as a powerful additional tool for varying enzyme selectivity and thus engineering the product range of biotransformations [85].

Similar effects of the reaction system on the stereoselectivity of BAL-catalysed reactions were recently described by Demir's group who studied the self- and cross-1,2-carboligation of different aldehydes in a biphasic aqueous–organic reaction system. Best results were found with MOPS buffer (pH 7) and diisopropyl ether [37].

Similar pronounced effects on the activity and stereoselectivity of BAL were observed by the group of Ansorge-Schumacher, who studied this enzyme in supercritical fluids [100].

Engineering transketolases

  1. Top of page
  2. Abstract
  3. Introduction
  4. Enzymes from the DC family
  5. Engineering transketolases
  6. Conclusion
  7. Acknowledgements
  8. References

In vivo TK catalyses the reversible transfer of a C2-ketol unit from d-xylulose-5-phosphate to either d-ribose-5-phosphate or d-erythrose-4-phosphate with ThDP as a cofactor [69].

Use of the donor β-hydroxypyruvate (HPA) results in the formation of CO2 and an irreversible transformation [101], which has been utilized in many synthetic applications. In early work S. cerevisiae and spinach TK were used with α- and α,β-hydroxyaldehyde acceptors and HPA, highlighting the stereospecificity of TK for the α-(R)-hydroxyaldehyde and generation of the new (3S)-stereocentre [102-106]. The selectivity of TK for formation of the (S)-isomer is in contrast to the R-selective decarboxylase superfamily (see above). Several crystal structures have been reported including ones with and without a substrate bound from S. cerevisiae [11, 107], from E. coli [70] and more recently from human TK [108]. As with the other ThDP-dependent enzymes the first key mechanistic step is deprotonation at the C2 of the thiazolium ring and addition of the donor molecule. This results in formation of the enamine intermediate that then adds to the acceptor aldehyde, with subsequent release of the product to regenerate the ThDP cofactor.

The successful overexpression of the EcTK gene gave nearly 25% of the cell protein as TK after an overnight growth for use in biocatalytic conversions [109]. The recombinant E. coli strain overexpressing the TK and a more recently generated His-tagged version readily accepted α-hydroxylated acceptors such as glycolaldehyde [110, 111]. A range of non-α-hydroxylated aliphatic substrates were also tolerated as was the case in reports using spinach and yeast TK [10, 110]. The stereoselectivities of TK in early work with chiral acceptors established the newly formed stereocentre as (S). However, when using achiral acceptors stereoselectivities were reported in some cases but the ee was often not reported or the preferred stereochemistry was not given. In 2008 it was fully established that the (3S)-isomer was formed with the acceptor glycolaldehyde (> 95% ee) and propanal (58% ee), the latter via the synthesis of (3S)-α,α′-dihydroxyketone using Ender's chiral auxiliary methodology [13]. This moderate optical purity observed with propanal was then the starting point to engineer improved E. coli TK variants with enhanced stereoselectivities.

Early site-directed mutagenesis of selected active site residues in ScTK had given several variants to probe the role of mechanistically important residues [11]. Also an EcTK-D469E variant (D469 is the equivalent residue to D477 in ScTK) was formed; however, these early studies did not report any stereochemical studies [112].

More recently, with direct synthetic routes to ketodiol products and colorimetric and HPLC product assays in place, together with chiral GC, HPLC and Mosher's ester NMR methods to establish enantiomeric purities and absolute stereochemistries, it was possible to generate and screen numerous enzyme variants for bioconversion efficacy and stereoselectivities against non-natural acceptor aldehydes [13, 113-115]. For EcTK library generation, due to the relatively large size of the TK gene (1992 bp), a strategy with a manageable library size was considered. Accordingly, using a structural approach and available X-ray crystallographic structures, key active site residues in contact with the substrate were identified [11, 70]. In addition, a phylogenetic analysis of 66 TK sequences highlighted important residues within 10 Å of the substrate or cofactors. Using 19 residues identified by both structural and phylogenetic analyses, saturation mutagenesis was performed to generate EcTK variants. The non-phosphorylated acceptor glycolaldehyde and non-α-hydroxylated aldehyde propanal were used in initial screens, which focused on improved activities [12, 116]. For glycolaldehyde the variants that gave the greatest improvements in activity for the formation of l-erythrulose compared with EcTKwt included A29D, A29E, H461S and R520V (Table 5) [12]. Notably, the D469 library, a residue, which is important for determining the enantioselectivity of TK with α-hydroxylated aldehydes, gave some of the greatest increases in activity towards propanal compared with EcTKwt including variants D469T and R520V (5-fold) and D469A and D469Y (4-fold). In earlier work, when variants D477A (ScTK) and D469E (EcTK) had been used with α-hydroxylated aldehydes or formaldehyde, lower activities were observed [112, 117]. Further studies with EcTKs assessing activity data against propanal together with stereoselectivities for the formation of 1,3-dihydroxypentan-2-one indicated that the D469X and H26X libraries were particularly productive with variant D469E giving products in 90% ee [(S)-isomer] (8-fold rate enhancement) and H26Y in 88% ee [(R)-isomer] (4-fold enhancement) (Table 5) compared with EcTKwt 58% ee [(S)-isomer] [13]. The variant D469T produced 1,3-dihydroxypentan-2-one in a similar selectivity to EcTKwt [64% ee for the (3S)-isomer], while interestingly D469Y generated the product in 53% ee [(R)-isomer] reversing the stereoselectivity. The tendency for single-point active site mutants to dramatically reverse the ee was surprising.

Table 5. TK variants used with selected aliphatic acceptors.image_n/febs12496-gra-0010.png
 
SubstrateProductKey variants References
image_n/febs12496-gra-0011.png image_n/febs12496-gra-0012.png

EcTKwt 95% ee (S )

EcTKs: A29D, A29E, H461S, R520V (3 to 5 times greater activity compared with wt)

ScTK R526Q/S525T (4 times greater activity compared with wt)

[12, 15]
image_n/febs12496-gra-0013.png image_n/febs12496-gra-0014.png

EcTKwt 58% ee (S )

EcTKs: D469T 64% ee (S ); D469E 90% ee (S ); D469Y 53% ee (R ); H26Y 88% ee (R); D469T/R520Q 68% ee (S ); D469Y/R520V [85% ee (R )]

[13, 118]
image_n/febs12496-gra-0015.png image_n/febs12496-gra-0016.png

EcTKwt 84% ee (S )

EcTKs: D469E 97% ee (S ); H26Y 84% ee (R )

[119]
image_n/febs12496-gra-0017.png image_n/febs12496-gra-0018.png

EcTKwt 66% ee (S )

EcTKs: D469E 86% ee (S ); H26Y 83% ee (R )

[119]
image_n/febs12496-gra-0019.png image_n/febs12496-gra-0020.png

EcTKwt 0% ee

EcTKs: D469T 99% ee (S ); D469E >99% ee (1S); H26Y 30% ee (R )

[119]
image_n/febs12496-gra-0021.png image_n/febs12496-gra-0022.png

EcTKwt 0% ee

EcTKs: D469E > 97% ee (S); H26Y no reaction

[119]
image_n/febs12496-gra-0023.png image_n/febs12496-gra-0024.png

EcTKwt 64% ee (S )

EcTKs: D469E 87% ee (S ); H26Y 64% ee (R)

[120]
image_n/febs12496-gra-0025.png image_n/febs12496-gra-0026.png

EcTKwt 85% ee (S )

EcTKs: D469E 91% ee (S ); H26Y 60% ee (R)

[120]

Building upon this work the key variants D469E and H26Y were used with C3 to C8 linear acceptor aldehydes as well as cyclopropane, cyclopentane and cyclohexane carboxaldehydes. When using variant D469E the stereoselectivities of the products increased to at least 97% ee [(S)-isomer] for the C4 to C6 linear aldehydes, which are of a similar chain length to the in vivo substrates for EcTK (Table 5) [119]. For the cyclic aldehydes the stereoselectivities were also high [> 97% ee, (S)-isomer], although isolated yields were generally lower. The stereoselectivities were significantly higher than with EcTKwt for which the longer chain and cyclic aldehydes were particularly poor substrates. Only cyclopentane carboxaldehyde of the cyclic acceptors was accepted by variant H26Y with a low ee (30% ee, (R)-isomer). In contrast the linear series gave (R)-α,α′-dihydroxyketones with variant H26Y in 78–92% ee. High yielding applications of the EcTK-D469T with propanal have also been described in a two-step TK + transaminase procedure to generate (2S,3S)-2-aminopentane-1,3-diol on a preparative scale, and also a microfluidic reactor with an in-line filtration system [121, 122].

Recently, ScTK and EcTKwt together with the EcTK-D469E variant were used with a range of 20 hydroxylated and non-hydroxylated aldehydes and activities were compared with that with glycolaldehyde [120]. The results highlighted the similarities in the reactivity profiles of ScTK and EcTK, with 42% amino acid identity, while D469E with 99.8% similarity to the EcTKwt showed a preference towards 2-deoxygenated aldehydes such as 3-hydroxypropanal and 4-hydroxybutanal giving as previously the (S)-isomers and in approximately 90% ee (Table 5). These authors also generated the variant H26Y, which as before preferentially gave the (R)-products when using 3-hydroxypropanal and 4-hydroxybutanal in approximately 60% ee.

Single-point variants of EcTK, D469E, D469T and D469K previously used with aliphatic substrates, and an F434 variant which is adjacent to D469 in the first shell (F434A), were also used with aromatic acceptors. In general low conversion yields of < 5% were observed with aromatic aldehydes, although benzaldehyde was converted to the corresponding dihydroxyketone product in 10% isolated yield and 82% ee (R) when using EcTK-F434A, probably due to the removal of hydrophobic and steric interactions (Table 6) [123]. Despite these low yields it was the first time such aldehydes had generated products with TK and was promising for future developments. Higher conversion yields of up to 50% and stereoselectivities of 97% ee [(S)-isomer] were observed when using all four variants with phenylacetaldehyde, possessing a less hindered aldehyde moiety that had also previously been used with EcTKwt [124]. In addition, when using 2-phenylpropionaldehyde the dihydroxyketone product was generated in up to 40% yield where the major product was the (3S,4R)-isomer with the minor product the (3S,4S)-isomer in a ratio of 96:4. The use of several D469 variants and F434A has also recently been described with the substrates 3-formylbenzoic acid (3-FBA) and 4-formylbenzoic acid (4-FBA). Although no conversions were noted with EcTKwt, with variants D469Y and F434A remarkably a 65% conversion yield was reported for 3-FBA and 30% with 4-FBA [14]. It was rationalized that the benzoate on the benzaldehyde ring might interact positively with the phosphate binding residues in the active site, in a similar fashion to the natural substrate ribose-5-phosphate. This was supported by performing computational docking experiments with 3-FBA and 4-FBA into the structure of variant D469T (generated in silico), with the ThDP-enamine intermediate present. Application of such electrostatic and hydrogen bonding interactions with natural substrates to novel acceptors is an effective strategy to achieve high conversions.

Table 6. TK variants used with selected aromatic acceptors.
SubstrateProductKey variants References
  1. a

    Conversion yields.

image_n/febs12496-gra-0027.png image_n/febs12496-gra-0028.png

EcTKwt no reaction

EcTKs: D469T 70% ee (R ); D469K 82% ee (R ); F434A 82% ee (R )

[123]
image_n/febs12496-gra-0029.png image_n/febs12496-gra-0030.png

EcTKwt 93% ee (S )

EcTKs: D469T 96% ee (S ); D469K 95% ee (3S ); F434A 97% ee (S )

[123]
image_n/febs12496-gra-0031.png image_n/febs12496-gra-0032.png

EcTKwt 88% (3S,4R ), 12% (3S,4S )

EcTKs: D469T 96% (3S,4R ), 4% (3S,4S ); D469E 95% (3S,4R ), 5% (3S,4S )

[123]
image_n/febs12496-gra-0033.png image_n/febs12496-gra-0034.png

EcTKwt no reaction

EcTKs: D469T 67% yielda (kcat/KM 240 s−1·m−1); D469T/R520Q 67% yielda (kcat/KM 470 s−1·m−1)

[14]
image_n/febs12496-gra-0035.png image_n/febs12496-gra-0036.png

EcTKwt no reaction

EcTKs: D469T 30% yielda (kcat/KM 20 s−1·m−1); D469T/R520Q 13% yielda (kcat/KM 8 s−1·m−1)

[14]

Subsequent recent work has looked into the generation of EcTK double variants by recombining productive single variants active against glycolaldehyde and propanal. However, many of these showed lower specific activities against both substrates, together with poor protein stability or folding [118]. The application of statistical coupling analysis with TK sequences and closely related ThDP enzymes highlighted nine coevolved residues that as a network help maintain a functional protein. Effective mutations at D469 that resulted in improved TK performance were combined with natural variants in the same coevolved network (at neighbouring residues), and these were screened against propanal. Particularly interesting variants with enhanced stereoselectivities compared with the single-point variants were D469T/R520Q [68% ee (S)], D469Y/R520V [85% ee (R)] and D469Y/R520Q [65% ee (R)], all with > 6-fold increases in specific activity compared with EcTKwt (Table 5).

Building upon the high conversion yields noted with EcTKwt single-point variants and 3-FBA and 4-FBA, double and triple variants based on variant D469T were generated, using previously reported functional mutations and guided by computational modelling [14]. While the D469T/R520Q variant had a greater affinity for 4-FBA, the kcat decreased, resulting in a lower kcat/KM. Use of this same variant with 3-FBA gave a kcat/KM of 470 s−1·m−1 which was higher than that observed with EcTKwt and glycolaldehyde (Table 6). Other double and triple variants gave lower conversion yields with these substrates. In other recent studies to improve the activity of ScTK towards polyhydroxylated acceptors, single- and double-point variants were designed using docking experiments and iterative mutagenesis. A variant R526Q/S525T (R526 is the equivalent residue to R520 in EcTK) was identified with a four times greater activity towards glycolaldehyde and HPA and 2.6 times more active for ribose with HPA (Table 5). It was noted that a large number of productive mutations possessed a loss of the cationic charge necessary for binding to phosphorylated substrates in vivo, highlighting again the importance of electrostatic interactions in substrate affinity and turnover [15]. Overall, the generation of productive single and multiple mutations in TK variants against a structurally diverse range of substrates (Fig. 2) has highlighted strategies such as coevolved residue networks and the importance of electrostatic as well as hydrophobic interactions when designing variants.

image

Figure 2. Range of acceptor substrates accepted with engineered TKs.

Download figure to PowerPoint

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Enzymes from the DC family
  5. Engineering transketolases
  6. Conclusion
  7. Acknowledgements
  8. References

Since the first description of a carboligase reaction catalysed by ThDP-dependent PDC in the 1930s, over the last 20 years studies in this field have intensified and demonstrated that carboligase activity is a common feature of ThDP-dependent enzymes. The intrinsic property of the cofactor ThDP allows activation of carbonyl compounds which results in umpolung behaviour of the carbonyl reactivity. Subsequent reactions with different electrophilic acceptors, ranging from a proton to a large range of aldehydes, α-ketoacids, ketones or α,β-unsaturated aldehydes, results in a broad range of possible products. In addition, the ThDP-dependent enzyme broadly defines the donor and acceptor substrate range and the chemoselectivity and stereoselectivity of such reactions. Here, we highlight that modulation of the enzymes' active site, by targeted mutagenesis, is a powerful tool to enlarge the range and applicability of ThDP-dependent enzymes in biocatalysis. Specifically, we have reviewed changes to the stereoselectivity and substrate range, which was typically achieved by one to three targeted mutations. Expanding the scope and utility of such enzymes holds significant promise for wider applications in the future.

Factors guiding the stereoselectivity and chemoselectivity in the structural family of decarboxylases are already well understood and can be explained based on a model regarding the binding sequence and relative orientation of donor and acceptor prior to C–C bond formation. However, a similar mechanistic model [7, 32, 41] is not available for TK. Importantly, the significantly higher number of hydrophilic amino acid residues, specifically histidines, in the active site may result in different stabilizing effects not only on the substrates but also in guiding stereoselectivity. Furthermore, basic assumptions for the model derived for the DC family, such as the orientation of oxygen functions of both the ThDP-bound donor aldehyde and the acceptor to allow access to the same proton relay system, are unlikely to be valid for TK as the high number and location of available histidines would principally also allow an inverse orientation of the acceptor relative to the donor (with the oxygen functions pointing in opposite directions). To elucidate this in more detail, factors determing stereoselectivity in TK are currently under investigation in our laboratories.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Enzymes from the DC family
  5. Engineering transketolases
  6. Conclusion
  7. Acknowledgements
  8. References

Work described in this review was partially funded by the German Research Foundation (DFG) in the frame of the research group FOR 1296 and the research training group BioNoCo (GK 1166).

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Enzymes from the DC family
  5. Engineering transketolases
  6. Conclusion
  7. Acknowledgements
  8. References
  • 1
    Jordan F (2003) Current mechanistic understanding of thiamin diphosphate-dependent enzymatic reactions. Nat Prod Rep 20, 184201.
  • 2
    Frank RAW, Leeper FJ & Luisi BF (2007) Structure, mechanism and catalytic duality of thiamine-dependent enzymes. Cell Mol Life Sci 64, 892905.
  • 3
    Müller M, Sprenger GA & Pohl M (2013) C–C-bond formation using ThDP-dependent lyases. Curr Opin Chem Biol 17, 261270.
  • 4
    Brovetto M, Gamenara D, Saenz Méndez P & Seoane GA (2011) C–C bond-forming lyases in organic synthesis. Chem Rev 111, 43464403.
  • 5
    Widmann M, Radloff R & Pleiss J (2010) The Thiamine diphosphate dependent Enzyme Engineering Database: a tool for the systematic analysis of sequence and structure relations. BMC Biochem 11, 9.
  • 6
    Müller M, Gocke D & Pohl M (2009) Thiamin diphosphate in biological chemistry: exploitation of diverse thiamin diphosphate-dependent enzymes for asymmetric chemoenzymatic synthesis. FEBS J 276, 28942940.
  • 7
    Pohl M, Gocke D & Müller M (2009) Thiamin-based enzymes for biotransformations. In Handbook of Green Chemistry: Green Catalysis (Anastas PT, ed.), pp. 75114. Wiley-VCH, Weinheim.
  • 8
    Dünkelmann P & Müller M (2011) Enzymatic C–C coupling in the synthesis of fine chemicals. Speciality Chemicals Mag 31, 1618.
  • 9
    Pohl M, Dresen C, Beigi M & Müller M (2012) Enzymatic acyloin and benzoin condensations. In Enzyme Catalysis in Organic Synthesis (Drauz K, Gröger H & May O, eds), pp. 919946. Wiley-VCH, Weinheim.
  • 10
    Demuynck C, Bolte J, Hecquet L & Dalmas V (1991) Enzyme-catalysed synthesis of carbohydrates – synthetic potential of transketolase. Tetrahedron Lett 32, 50855088.
  • 11
    Nilsson U, Meshalkina L, Lindqvist Y & Schneider G (1997) Examination of substrate binding in thiamin diphosphate-dependent transketolase by protein crystallography and site-directed mutagenesis. J Biol Chem 272, 18641869.
  • 12
    Hibbert EG, Senussi T, Costelloe SJ, Lei W, Smith MEB, Ward JM, Hailes HC & Dalby PA (2007) Directed evolution of transketolase activity on non-phosphorylated substrates. J Biotechnol 131, 425432.
  • 13
    Smith MEB, Hibbert EG, Jones AB, Dalby PA & Hailes HC (2008) Enhancing and reversing the stereoselectivity of Escherichia coli transketolase via single-point mutations. Adv Synth Catal 350, 26312638.
  • 14
    Payongsri P, Steadman D, Strafford J, MacMurray A, Hailes HC & Dalby PA (2012) Rational substrate and enzyme engineering of transketolase for aromatics. Org Biomol Chem 10, 90219029.
  • 15
    Ranoux A, Karmee SK, Jin J, Bhaduri A, Caiazzo A, Arends IW & Hanefeld U (2012) Enhancement of the substrate scope of transketolase. ChemBioChem 13, 19211931.
  • 16
    Ranoux A & Hanefeld U (2013) Improving transketolase. Top Catal 56, 750764.
  • 17
    Engel S, Vyazmensky M, Geresh S, Barak Z & Chipman DM (2003) Acetohydroxyacid synthase: a new enzyme for chiral synthesis of R-phenylacetylcarbinol. Biotechnol Bioeng 83, 833840.
  • 18
    Chipman DM, Duggleby RG & Tittmann K (2005) Mechanisms of acetohydroxyacid synthases. Curr Opin Chem Biol 9, 475481.
  • 19
    Vinogradov M, Kaplun A, Vyazmensky M, Engel S, Golbik R, Tittmann K, Uhlemann K, Meshalkina L, Barak Z, Hübner G et al. (2005) Monitoring the acetohydroxy acid synthase reaction and related carboligations by circular dichroism spectroscopy. Anal Biochem 342, 126133.
  • 20
    Vyazmensky M, Steinmetz A, Meyer D, Golbik R, Ze Barak, Tittmann K & Chipman DM (2011) Significant catalytic roles for Glu47 and Gln 110 in all four of the C–C bond-making and -breaking steps of the reactions of acetohydroxyacid synthase II. Biochemistry 50, 32503260.
  • 21
    Belenky I, Steinmetz A, Vyazmensky M, Ze Barak, Tittmann K & Chipman DM (2012) Many of the functional differences between acetohydroxyacid synthase (AHAS) isozyme I and other AHASs are a result of the rapid formation and breakdown of the covalent acetolactate–thiamin diphosphate adduct in AHAS I. FEBS J 279, 19671979.
  • 22
    Dünkelmann P, Kolter-Jung D, Nitsche A, Demir AS, Siegert P, Lingen B, Baumann M, Pohl M & Müller M (2002) Development of a donor–acceptor concept for enzymatic cross-coupling reactions of aldehydes: the first asymmetric cross-benzoin condensation. J Am Chem Soc 124, 1208412085.
  • 23
    Lingen B, Kolter-Jung D, Dünkelmann P, Feldmann R, Grötzinger J, Pohl M & Müller M (2003) Alteration of the substrate specificity of benzoylformate decarboxylase from Pseudomonas putida by directed evolution. ChemBioChem 4, 721726.
  • 24
    Siegert P, McLeish MJ, Baumann M, Iding H, Kneen MM, Kenyon GL & Pohl M (2005) Exchanging the substrate specificities of pyruvate decarboxylase from Zymomonas mobilis and benzoylformate decarboxylase from Pseudomonas putida. Protein Eng Des Sel 18, 345357.
  • 25
    Domínguez de María P, Pohl M, Gocke D, Gröger H, Trauthwein H, Stillger T & Müller M (2007) Asymmetric synthesis of aliphatic 2-hydroxy ketones by enzymatic carboligation of aldehydes. Eur J Org Chem, 29402944.
  • 26
    Mikolajek R, Spiess AC, Pohl M, Lamare S & Büchs J (2007) An activity, stability and selectivity comparison of propioin synthesis by thiamine diphosphate-dependent enzymes in a solid/gas bioreactor. ChemBioChem 8, 10631070.
  • 27
    Cosp A, Dresen C, Pohl M, Walter L, Röhr C & Müller M (2008) α, β-Unsaturated aldehydes as substrates for asymmetric C–C bond forming reactions with thiamin diphosphate (ThDP)-dependent enzymes. Adv Synth Catal 350, 759771.
  • 28
    Gocke D, Walter L, Gauchenova E, Kolter G, Knoll M, Berthold CL, Schneider G, Pleiss J, Müller M & Pohl M (2008) Rational protein design of ThDP-dependent enzymes: engineering stereoselectivity. ChemBioChem 9, 406412.
  • 29
    Domínguez de María P, Stillger T, Pohl M, Kiesel M, Liese A, Gröger H & Trauthwein H (2008) Enantioselective C–C bond ligation using recombinant E. coli-whole-cell biocatalysts. Adv Synth Catal 350, 165173.
  • 30
    Demir AS, Sesenoglu O, Eren E, Hosrik B, Pohl M, Janzen E, Kolter D, Feldmann R, Dünkelmann P & Müller M (2002) Enantioselective synthesis of alpha-hydroxy ketones via benzaldehyde lyase-catalyzed C–C bond formation reaction. Adv Synth Catal 344, 96103.
  • 31
    Demir AS, Sesenoglu O, Dünkelmann P & Müller M (2003) Benzaldehyde lyase-catalyzed enantioselective carboligation of aromatic aldehydes with mono- and dimethoxy acetaldehyde. Org Lett 5, 20472050.
  • 32
    Knoll M, Müller M, Pleiss J & Pohl M (2006) Factors mediating activity, selectivity, and substrate specificity for the thiamin diphosphate-dependent enzymes benzaldehyde lyase and benzoylformate decarboxylase. ChemBioChem 7, 19281934.
  • 33
    Domínguez de María P, Stillger T, Pohl M, Wallert S, Drauz K, Gröger H, Trauthwein H & Liese A (2006) Preparative enantioselective synthesis of benzoins and (R)-2-hydroxy-1-phenylpropanone using benzaldehyde lyase. J Mol Catal B: Enzym 38, 4347.
  • 34
    Hildebrand F, Kühl S, Pohl M, Vasic-Racki D, Müller M, Wandrey C & Lütz S (2007) The production of (R)-2-hydroxy-1-phenyl-propan-1-one derivatives by benzaldehyde lyase from Pseudomonas fluorescens in a continuously operated membrane reactor. Biotechnol Bioeng 96, 835843.
  • 35
    Kühl S, Zehentgruber D, Pohl M, Müller M & Lütz S (2007) Process development for enzyme catalysed asymmetric C–C-bond formation. Chem Eng Sci 62, 52015205.
  • 36
    Shanmuganathan S, Natalia D, van den Wittenboer A, Kohlmann C, Greiner L & Dominguez de María P (2010) Enzyme-catalyzed C–C bond formation using 2-methyltetrahydrofuran (2-MTHF) as (co)solvent: efficient and bio-based alternative to DMSO and MTBE. Green Chem 12, 22402245.
  • 37
    Ayhan P, Simsek I, Cifci B & Demir AS (2011) Benzaldehyde lyase catalyzed enantioselective self and cross condensation reactions of acetaldehyde derivatives. Org Biomol Chem 9, 26022605.
  • 38
    Shanmuganathan S, Natalia D, Greiner D & Dominguez de Maria P (2012) Oxidation-hydroxymethylation-reduction: one-pot three-step biocatalytic synthesis of optically active α-aryl vicinal diols. Green Chem 14, 9497.
  • 39
    Müller CR, Pérez-Sanchéz M & Domínguez de María P (2013) Benzaldehyde lyase-catalyzed diastereoselective C–C bond formation by simultaneous carboligation and kinetic resolution. Org Biomol Chem 11, 20002004.
  • 40
    Meyer D, Walter J, Kolter G, Pohl M, Müller M & Tittmann K (2011) Conversion of pyruvate decarboxylase into an enantioselective carboligase with biosynthetic potential. J Am Chem Soc 133, 36093616.
  • 41
    Rother D, Kolter G, Gerhards T, Berthold Siöberg CL, Gauchenova E, Knoll M, Pleiss J, Müller M, Schneider G & Pohl M (2011) (S)-Selective mixed benzoin condensation by structure-based design of the pyruvate decarboxylase from Acetobacter pasteurianus. ChemCatChem 3, 15871596.
  • 42
    Smit BA, Vlieg J, Engels WJM, Meijer L, Wouters JTM & Smit G (2005) Identification, cloning, and characterization of a Lactococcus lactis branched-chain alpha-keto acid decarboxylase involved in flavor formation. Appl Environ Microbiol 71, 303311.
  • 43
    Gocke D, Nguyen CL, Pohl M, Stillger T, Walter L & Müller M (2007) Branched-chain ketoacid decarboxylase from Lactococcus lactis (KdcA), a valuable thiamin diphosphate-dependent enzyme for asymmetric C–C bond formation. Adv Synth Catal 349, 14251435.
  • 44
    Fraas S, Steinbach AK, Tabbert A, Harder J, Ermler U, Tittmann K, Meyer A & Kroneck PMH (2009) Cyclohexane-1,2-dione hydrolase: a new tool to degrade alicyclic compounds. J Mol Catal B: Enzym 61, 4749.
  • 45
    Steinbach AK, Fraas S, Harder J, Tabbert A, Brinkmann H, Meyer A, Ermler U & Kroneck PM (2011) Cyclohexane-1,2-dione hydrolase from denitrifying Azoarcus sp. strain 22Lin, a novel member of the thiamine diphosphate enzyme family. J Bacteriol 193, 67606769.
  • 46
    Steinbach A, Fraas S, Harder J, Warkentin E, Kroneck PM & Ermler U (2012) Crystal structure of a ring-cleaving cyclohexane-1,2-dione hydrolase, a novel member of the thiamine diphosphate enzyme family. FEBS J 279, 12091219.
  • 47
    Kurutsch A, Richter M, Brecht V, Sprenger GA & Müller M (2009) MenD as a versatile catalyst for asymmetric synthesis. J Mol Catal B: Enzym 61, 5666.
  • 48
    Beigi M, Loschonsky S, Lehwald P, Brecht V, Andrade SL, Leeper FJ, Hummel W & Müller M (2013) alpha-Hydroxy-beta-keto acid rearrangement-decarboxylation: impact on thiamine diphosphate-dependent enzymatic transformations. Org Biomol Chem 11, 252256.
  • 49
    Beigi M, Waltzer S, Fries A, Eggeling L, Sprenger GA & Muller M (2013) TCA cycle involved enzymes SucA and Kgd, as well as MenD: efficient biocatalysts for asymmetric C–C bond formation. Org Lett 15, 452455.
  • 50
    Wagner T, Bellinzoni M, Wehenkel A, O'Hare HM & Alzari Pedro M (2011) Functional plasticity and allosteric regulation of α-ketoglutarate decarboxylase in central mycobacterial metabolism. Chem Biol 18, 10111020.
  • 51
    Chen H, Guo Z & Liu H-W (1998) Biosynthesis of yersiniose: attachment of the two-carbon branched-chain is catalyzed by a thiamine pyrophosphate-dependent flavoprotein. J Am Chem Soc 120, 1179611797.
  • 52
    Lehwald P, Richter M, Röhr C, Liu HW & Müller M (2010) Enantioselective intermolecular aldehyde-ketone cross-coupling through an enzymatic carboligation reaction. Angew Chem Int Ed 49, 23892392.
  • 53
    Kluger R & Tittmann K (2008) Thiamin diphosphate catalysis: enzymic and nonenzymic covalent intermediates. Chem Rev 108, 17971833.
  • 54
    Costelloe SJ, Ward JM & Dalby PA (2008) Evolutionary analysis of the TPP-dependent enzyme family. J Mol Evol 66, 3649.
  • 55
    Chipman D, Barak Z & Schloss JV (1998) Biosynthesis of 2-aceto-2-hydroxy acids: acetolactate synthases and acetohydroxyacid synthases. Biochim Biophys Acta 1385, 401419.
  • 56
    Bar-Ilan A, Balan V, Tittmann K, Golbik R, Vyazmensky M, Hübner G, Barak Z & Chipman DM (2001) Binding and activation of thiamin diphosphate in acetohydroxyacid synthase. Biochemistry 40, 1194611954.
  • 57
    Mosbacher TG, Müller M & Schulz GE (2005) Structure and mechanism of the ThDP-dependent benzaldehyde lyase from Pseudomonas fluorescens. FEBS J 272, 60676076.
  • 58
    Gonzales B & Vicuna R (1989) Benzaldehyde lyase, a novel thiamine PPi-requiring enzyme from Pseudomonas fluorescens Biovar I. J Bacteriol 171, 24012405.
  • 59
    Hegeman GD (1970) Benzoylformate decarboxylase (Pseudomonas putida). Meth Enzymol 17A, 674678.
  • 60
    Hasson MS, Muscate A, McLeish MJ, Polovnikova LS, Gerlt JA, Kenyon GL, Petsko GA & Ringe D (1998) The crystal structure of benzoylformate decarboxylase at 1.6 A resolution: diversity of catalytic residues in thiamin diphosphate-dependent enzymes. Biochemistry 37, 99189930.
  • 61
    Polovnikova ES, McLeish MJ, Sergienko EA, Burgner JT, Anderson NL, Bera AK, Jordan F, Kenyon GL & Hasson MS (2003) Structural and kinetic analysis of catalysis by a thiamin diphosphate-dependent enzyme, benzoylformate decarboxylase. Biochemistry 42, 18201830.
  • 62
    Dawes EA, Ribbons DW & Large PJ (1966) The route of ethanol formation in Zymomonas mobilis. Biochem J 98, 795803.
  • 63
    Dobritzsch D, König S, Schneider G & Lu G (1998) High resolution crystal structure of pyruvate decarboxylase from Zymomonas mobilis. Implications for substrate activation in pyruvate decarboxylases. J Biol Chem 273, 2019620204.
  • 64
    Raj KC, Ingram LO & Maupin-Furlow JA (2001) Pyruvate decarboxylase: a key enzyme for the oxidative metabolism of lactic acid by Acetobacter pasteurianus. Arch Microbiol 176, 443451.
  • 65
    Jiang M, Cao Y, Guo Z-F, Chen M, Chen X & Guo Z (2007) Menaquinone biosynthesis in Escherichia coli: identification of 2-succinyl-5-enolpyruvyl-6-hydroxy-3-cyclohexene-1-carboxylate as a novel intermediate and re-evaluation of MenD activity. Biochemistry 46, 1097910989.
  • 66
    Dawson A, Fyfe PK & Hunter WN (2008) Specificity and reactivity in menaquinone biosynthesis: the structure of Escherichia coli MenD (2-succinyl-5-enolpyruvyl-6-hydroxy-3-cyclohexadiene-1-carboxylate synthase). J Mol Biol 384, 13531368.
  • 67
    Dawson A, Chen M, Fyfe PK, Guo Z & Hunter WN (2010) Structure and reactivity of Bacillus subtilis MenD catalyzing the first committed step in menaquinone biosynthesis. J Mol Biol 401, 253264.
  • 68
    Priyadarshi A, Kim EE & Hwang KY (2009) Structural and functional analysis of vitamin K2 synthesis protein MenD. Biochem Biophys Res Commun 388, 748751.
  • 69
    Racker E (1961) The Enzymes, Vol. 5. Academic Press, New York.
  • 70
    Littlechild JA, Turner NJ, Hobbs GR, Lilly MD, Rawas A & Watson H (1995) Crystallization and preliminary X-ray crystallographic data with Escherichia coli transketolase. Acta Crystallogr Biol Crystallogr 51, 10741076.
  • 71
    Asztalos P, Parthier C, Golbik R, Kleinschmidt M, Hübner G, Weiss MS, Friedemann R, Wille G & Tittmann K (2007) Strain and near attack conformers in enzymic thiamin catalysis: X-ray crystallographic snapshots of bacterial transketolase in covalent complex with donor ketoses xylulose 5-phosphate and fructose 6-phosphate, and in noncovalent complex with acceptor aldose ribose 5-phosphate. Biochemistry 46, 1203712052.
  • 72
    Lindqvist Y, Schneider G, Ermler U & Sundstrom M (1992) Three-dimensional structure of transketolase, a thiamine diphosphate dependent enzyme, at 2.5 A resolution. EMBO J 11, 23732379.
  • 73
    Fang M, Macova A, Hanson KL, Kos J & Palmer DRJ (2011) Using substrate analogues to probe the kinetic mechanism and active site of Escherichia coli MenD. Biochemistry 50, 87128721.
  • 74
    Kneen MM, Pogozheva ID, Kenyon GL & McLeish MJ (2005) Exploring the active site of benzaldehyde lyase by modeling and mutagenesis. Biochim Biophys Acta – Proteins Proteomics 1753, 263271.
  • 75
    Brandt GS, Kneen MM, Petsko GA, Ringe D & McLeish MJ (2010) Active-site engineering of benzaldehyde lyase shows that a point mutation can confer both new reactivity and susceptibility to mechanism-based inhibition. J Am Chem Soc 132, 438439.
  • 76
    Demir AS, Pohl M, Janzen E & Müller M (2001) Enantioselective synthesis of hydroxy ketones through cleavage and formation of acyloin linkage. Enzymatic kinetic resolution via C-C bond cleavage. J Chem Soc Perkin Trans 1, 633635.
  • 77
    Hildebrandt G & Klavehn W (1932) Verfahren zur Herstellung von 1-l-Phenyl-2-methylaminopropan-1-ol, Germany. Deutsches Reichspatent 548, 459.
  • 78
    Meyer D, Neumann P, Parthier C, Friedemann R, Nemeria N, Jordan F & Tittmann K (2010) Double duty for a conserved glutamate in pyruvate decarboxylase: evidence of the participation in stereoelectronically controlled decarboxylation and in protonation of the nascent carbanion/enamine intermediate. Biochemistry 49, 81978212.
  • 79
    Baykal A, Chakraborty S, Dodoo A & Jordan F (2006) Synthesis with good enantiomeric excess of both enantiomers of alpha-ketols and acetolactates by two thiamin diphosphate-dependent decarboxylases. Bioorg Chem 34, 380393.
  • 80
    Wilcocks R, Ward OP, Collins S, Dewdney NJ, Hong Y & Prosen E (1992) Acyloin formation by benzoylformate decarboxylase from Pseudomonas putida. Appl Environ Microbiol 58, 16991704.
  • 81
    Iding H, Dünnwald T, Greiner L, Liese A, Müller M, Siegert P, Grötzinger J, Demir AS & Pohl M (2000) Benzoylformate decarboxylase from Pseudomonas putida as stable catalyst for the synthesis of chiral 2-hydroxy ketones. Chem Eur J 6, 14831495.
  • 82
    Demir AS, Dünnwald T, Iding H, Pohl M & Müller M (1999) Asymmetric benzoin reaction catalyzed by benzoylformate decarboxylase. Tetrahedron: Asymmetry 10, 47694774.
  • 83
    Dünnwald T, Demir AS, Siegert P, Pohl M & Müller M (2000) Enantioselective synthesis of (S)-2-hydroxypropanone derivatives by benzoylformate decarboxylase catalyzed C–C bond formation. Eur J Org Chem, 21612170.
  • 84
    Gocke D, Graf T, Brosi H, Frindi-Wosch I, Walter L, Müller M & Pohl M (2009) Comparative characterisation of thiamin diphosphate-dependent decarboxylases. J Mol Catal B:Enzym 61, 3035.
  • 85
    Gerhards T, Mackfeld U, Bocola M, von Lieres E, Wiechert W, Pohl M & Rother D (2012) Influence of organic solvents on enzymatic asymmetric carboligations. Adv Synth Catal 354, 28052820.
  • 86
    Wilcocks R & Ward OP (1992) Factors effecting 2-hydroxypropiophenone formation by benzoylformate decarboxylase from Pseudomonas putida. Biotech Bioeng 39, 10581063.
  • 87
    Demir AS, Ayan P, Igdir AC & Guygu AN (2004) Enzyme catalyzed hydroxymethylation of aromatic aldehydes with formaldehyde. Synthesis of hydroxyacetophenones and (S)-benzoins. Tetrahedron 60, 65096512.
  • 88
    Bruhn H, Pohl M, Grötzinger J & Kula MR (1995) The replacement of Trp392 by alanine influences the decarboxylase/carboligase activity and stability of pyruvate decarboxylase from Zymomonas mobilis. Eur J Biochem 234, 650655.
  • 89
    Goetz G, Iwan P, Hauer B, Breuer M & Pohl M (2001) Continuous production of (R)-phenylacetylcarbinol in an enzyme-membrane reactor using a potent mutant of pyruvate decarboxylase from Zymomonas mobilis. Biotechnol Bioeng 74, 317325.
  • 90
    Engel S, Vyazmensky M, Berkovich D, Barak Z & Chipman DM (2004) Substrate range of acetohydroxy acid synthase I from Escherichia coli in the stereoselective synthesis of alpha-hydroxy ketones. Biotechnol Bioeng 88, 825831.
  • 91
    Westphal R, Waltzer S, Mackfeld U, Widmann M, Rother D, Pleiss J, Beigi M, Müller M & Pohl M (2013) (S)-Selective MenD variants from Escherichia coli provide access to new functionalized chiral α-hydroxy ketones. Chem Commun 49, 20612063.
  • 92
    Westphal R, Hahn D, Mackfeld U, Waltzer S, Beigi M, Widmann M, Vogel C, Pleiss J, Müller M & Pohl M (2013) Tailoring (S)-selectivity of MenD from Escherichia coli. ChemCatChem, doi:10.1002/cctc.201300318.
  • 93
    Schmitz C (2012) Untersuchung der Stereoselektivität und des Substratspektrums von AHAS I&II aus Escherichia coli. Fachhochschule Aachen, Jülich.
  • 94
    Pang SS, Duggleby RG & Guddat LW (2002) Crystal structure of yeast acetohydroxyacid synthase: a target for herbicidal inhibitors. J Mol Biol 317, 249262.
  • 95
    Vogel C, Widmann M, Pohl M & Pleiss J (2012) A standard numbering scheme for thiamine diphosphate-dependent decarboxylases. BMC Biochem 13, 24.
  • 96
    Blanco F, Kelly B, Alkorta I, Rozas I & Elguero J (2011) Cation–pi interactions: Complexes of guanidinium and simple aromatic systems. Chem Phys Lett 511, 129134.
  • 97
    Stillger T, Pohl M, Wandrey C & Liese A (2006) Reaction engineering of benzaldehyde lyase catalyzing enantioselective C–C bond formation. Org Proc Res Dev 10, 11721177.
  • 98
    Kokova M, Zavrel M, Tittmann K, Spiess AC & Pohl M (2009) Investigating the carboligase activity of thiamine diphosphate-dependent enzymes using kinetic modeling and NMR spectroscopy. J Mol Catal B:Enzym 61, 7379.
  • 99
    Janzen E, Müller M, Kolter-Jung D, Kneen MM, McLeish MJ & Pohl M (2006) Characterization of benzaldehyde lyase from Pseudomonas fluorescens – a versatile enzyme for asymmetric C–C-bond formation. Bioorg Chem 34, 345361.
  • 100
    Natalia D, Greiner L, Leitner W & Ansorge-Schumacher MB (2012) Stability, activity, and selectivity of benzaldehyde lyase in supercritical fluids. J Supercrit Fluids 62, 173177.
  • 101
    Srere P, Cooper JR, Tabachnick M & Racker E (1958) The oxidative pentose phosphate cycle. I. Preparation of substrates and enzymes. Arch Bioch Biophys 74, 295305.
  • 102
    Villafranca JJ & Axelrod B (1971) Heptulose synthesis from nonphosphorylated aldoses and ketoses by spinach transketolase. J Biol Chem 246, 31263131.
  • 103
    Bolte J, Demuynck C & Samaki H (1987) Utilisation of enzymes in organic chemistry – transketolase catalyzed synthesis of ketoses. Tetrahedron Lett 28, 55255528.
  • 104
    Myles DC, Andrulis PJ III & Whitesides GM (1991) A transketolase-based synthesis of (+)-exo-brevicomin. Tetrahedron Lett 32, 48354838.
  • 105
    Kobori Y, Myles DC & Whitesides GM (1992) Substrate-specificity and carbohydrate synthesis using transketolase. J Org Chem 57, 58995907.
  • 106
    Hecquet L, Bolte J & Demuynck C (1994) Chemoenzymatic synthesis of 6-deoxy-D-fructose and 6-deoxy-L-sorbose using transketolase. Tetrahedron 50, 86778684.
  • 107
    Nikkola M, Lindqvist Y & Schneider G (1994) Refined structure of transketolase from Saccharomyces cerevisiae at 2.0 A resolution. J Mol Biol 238, 387404.
  • 108
    Mitschke L, Parthier C, Schröder-Tittmann K, Coy J, Ludtke S & Tittmann K (2010) The crystal structure of human transketolase and new insights into its mode of action. J Biol Chem 285, 3155931570.
  • 109
    French C & Ward JM (1995) Improved production and stability of Escherichia coli recombinants expressing transketolase for large-scale biotransformation. Biotechnol Lett 17, 247252.
  • 110
    Hobbs GR, Lilly MD, Turner NJ, Ward JM, Willets AJ & Woodley JM (1993) Enzyme-catalyzed carbon carbon bond formation – use of transketolase from Escherichia coli. J Chem Soc Perkin Trans 1, 165166.
  • 111
    Aucamp JP, Martinez-Torres RJ, Hibbert EG & Dalby PA (2008) A microplate-based evaluation of complex denaturation pathways: structural stability of Escherichia coli transketolase. Biotechnol Bioeng 99, 13031310.
  • 112
    Schörken U, Sahm H & Sprenger G (1996) Substrate specificity, site-directed mutagenesis and modelling of the substrate channel and preliminary X-ray crystallographic data of E. coli transketolase. In Biochemistry and Physiology of Thiamine Diphosphate Enzymes. (Bisswanger H & Schellenberger A, eds), pp. 543554. Intemann, Prien.
  • 113
    Smith MEB, Smithies K, Senussi T, Dalby PA & Hailes HC (2006) The first mimetic of the transketolase reaction. Eur J Org Chem, 11211123.
  • 114
    Smith MEB, Kaulmann U, Ward JM & Hailes HC (2006) A colorimetric assay for screening transketolase activity. Bioorg Med Chem 14, 70627065.
  • 115
    Galman JL & Hailes HC (2009) Application of a modified Mosher's method for the determination of enantiomeric ratio and absolute configuration at C-3 of chiral 1,3-dihydroxy ketones. Tetrahedron: Asymmetry 20, 18281831.
  • 116
    Hibbert EG, Senussi T, Smith MEB, Costelloe SJ, Ward JM, Hailes HC & Dalby PA (2008) Directed evolution of transketolase substrate specificity towards an aliphatic aldehyde. J Biotechnol 134, 240245.
  • 117
    Nilsson U, Hecquet L, Gefflaut T, Guerard C & Schneider G (1998) Asp477 is a determinant of the enantioselectivity in yeast transketolase. FEBS Lett 424, 4952.
  • 118
    Strafford J, Payongsri P, Hibbert EG, Morris P, Batth SS, Steadman D, Smith MEB, Ward JM, Hailes HC & Dalby PA (2012) Directed evolution to re-adapt a co-evolved network within an enzyme. J Biotechnol 157, 237245.
  • 119
    Cazares A, Galman JL, Crago LG, Smith MEB, Strafford J, Rios-Solis L, Lye GJ, Dalby PA & Hailes HC (2010) Non-α-hydroxylated aldehydes with evolved transketolase enzymes. Org Biomol Chem 8, 13011309.
  • 120
    Yi D, Devamani T, Abdoul-Zabar J, Charmantray F, Helaine V, Hecquet L & Fessner W-D (2012) A pH-based high-throughput assay for transketolase: fingerprinting of substrate tolerance and quantitative kinetics. ChemBioChem 13, 22902300.
  • 121
    Smith MEB, Chen BH, Hibbert EG, Kaulmann U, Smithies K, Galman JL, Baganz F, Dalby PA, Hailes HC, Lye GJ et al. (2010) A multi-disciplinary approach toward the rapid and preparative scale biocatalytic synthesis of chiral amino alcohols: a concise transketolase/ω-transaminase-mediated synthesis of (2S,3S)-2-aminopentane-1,3-diol. Org Proc Res Dev 14, 99107.
  • 122
    O'Sullivan B, Al-Bahrani H, Lawrence J, Campos M, Cazares A, Baganz F, Wohlgemuth R, Hailes HC & Szita N (2012) Modular microfluidic reactor and inline filtration system for the biocatalytic synthesis of chiral metabolites. J Mol Catal B: Enzym 77, 18.
  • 123
    Galman JL, Steadman D, Bacon S, Morris P, Smith MEB, Ward JM, Dalby PA & Hailes HC (2010) α, α′-Dihydroxyketone formation using aromatic and heteroaromatic aldehydes with evolved transketolase enzymes. Chem Commun 46, 76087610.
  • 124
    Morris KG, Smith MEB, Turner NJ, Lilly MD, Mitra RK & Woodley JM (1996) Transketolase from Escherichia coli: a practical procedure for using the biocatalyst for asymmetric carbon–carbon bond synthesis. Tetrahedron: Asymmetry 7, 21852188.