摘要:
J. CHEM. SOC. PERKIN TRANS. 1 1993 Enzyme-catalysed Carbon-Carbon Bond Formation: Use of Transketolase from Escherichia coli Gordon R. Hobbs,8 Malcolm D. Lilly,b Nicholas J. Turner,**8 John M. Ward,= Andrew J. Willetsdand John M. Woodley” a Department of Chemistry, University of Exeter, Stocker Road, Exeter EX4 4QD, UK Advanced Centre for Biochemical Engineering, Department of Chemical and Biochemical Engineering, University College London, Torrington Place, London WClE 7J€, UK Department of Biochemistry and Molecular Biology, University College London, Gower Street, London WCIE SST, UK Department of Biological Sciences, University of Exeter, Prince of Wales Road, Exeter EX4 4QG, UK Transketolase has been obtained in greater quantities from an over-expressed E.coli transformant carrying the transketolase gene. Crude extracts of this organism are suitable for use in small scale biotransformations to provide mmol quantities of product. Initial results indicate that the transketolase from E. coli is relatively non-specific for the aldehyde component of the reaction. In uiuo transketolase (TK) (E.C. 2.2.1.1) catalyses a vital step in the pentose phosphate pathway, namely the transfer of a two carbon ketol group from ~-xylulose-5-phosphate to ~-ribose-5- phosphate thereby generating ~-sedoheptulose-7-phosphate.’ Consequently the enzyme plays an important role in the biosynthesis of aromatic compounds of industrial interest.’ The enzyme requires magnesium(I1) ions and thiamine pyro- phosphate (TPP) as cofactors.However, the isolated enzyme also has synthetic potential. For this the reaction can be made more efficient by the use of hydroxypyruvate 2 as the ketol donor (Scheme 1). The resultant release of carbon dioxide ensures that the reaction is irreversible leading to trihydroxy- lated compounds 3 containing the D-threo configuration. OH 0 0 OH 1 2 3 Scheme 1 Until recently it was believed that only a-hydroxyaldehydes of the D-configuration were possible aldehyde acceptors 1 but a recent report described the use of some simple a-unsubstituted aldehydes. Thus the low specificity of TK for the aldehyde component 1 together with the high stereoselectivity for the carbon-carbon bond forming step combine to make this enzyme a potentially useful catalyst for asymmetric synthesis.With this in mind we have initiated a programme whose aim is to study the fundamental principles required to develop an integrated process for carbon-carbon bond formation on a large scale using transketolase as the catalyst. This communi- cation describes our initial work aimed at developing an efficient and reliable source of the biocatalyst. Transketolase can be obtained from two sources (i) Sac- charomyces cereuisiae (commercially available from Sigma) and (ii) ~pinach.~ However neither of these sources yields large quantities of the enzyme although the synthetic potential of these sources has been demonstrated on a small scale.6 * The growth medium contained the following: Na,HP04.12 HzO (15 g dm-3), KHzP04 (3 g dm-3), NaCl (0.5 g dm-j NH,Cl (1 g dm-3), MgS04.7H,0 (0.25 g dm-3), shikimic acid (0.04 g dm-3), thiamine (0.001 g dm-3), ampicillin (0.1g dm-3) and either glucose (2.5 g dm-3) or glycerol (2.5 g dm-3). Therefore we sought to develop an organism that could be used to produce greater quantities of TK.A recent paper described the cloning of the gene for TK from E. coli back into E. coli BJ 502’ and this proved to be a useful starting point for over- expression of the protein. The fragment of DNA (5Kb) encoding the TK gene (2Kb) was excised from the piasmid of E. coli (BJ502/pKD112A) and reintroduced into a high copy plasmid vector (pUC18) in strain JM107, generating two new trans- formants (JM107/pQR182 and JM107/pQR183), differing by the orientation of the gene with respect to the lac promoter sited on pUCl8.These two new transformants showed increased specific activity as measured by assay of the crude cell free extracts (Table 1). Both JM107/pQR182 and JM107/pQR183 produced TK with approximately four fold higher specific activity than BJ502/pKD112A. Examination of polyacrylamide gels run under denaturing conditions reveal TK to be the dominant protein. Thus by the simple expedient of introducing the TK gene into a high copy plasmid we were able to overexpress the protein in E. coli. Having established a source of substantial quantities of TK we next examined its substrate specificity with respect to the aldehyde acceptor.All experiments were carried out using JM107/pQR183 and the results obtained are shown in Table 2. It can be seen that the enzyme shows a relatively low specificity for the aldehyde substrate. Although a-hydroxyaldehydes were the best substrates, simple aldehydes (e.g. propionaldehyde, pyruvaldehyde) also reacted and could be converted into the corresponding products on a preparative scale (see below). Cyclic aldehydes were poor substrates presumably due to unfavourable steric interactions at the active site. Of particular interest is the similarity of the substrate specificity profile to that reported for yeast and may indicate a similarity between TK from the two different sources. We are currently carrying out additional experiments to characterise the E.coli TK more fully. In summary, we have established a new source of trans- ketolase that is able to produce the enzyme in large quantities suitable for preparative biotransformations. Experimental Preparation of Biocata1yst.-Starter cultures were prepared by inoculating sterile media* (50 cm3) with single colonies of the organism from an agar plate and incubated for 16 h on an orbital shaker (150 rpm) at 37°C. Aliquots (5 cm3) of these cultures were transferred to flasks containing sterile batch J. CHEM. SOC. PERKIN TRANS. 1 1993 Table 1 Activities of transketolase in cell-free extracts * TK activity Protein Specific activity Transformant Carbon source Ua/cm3 mg~m-~ Ua/mg BJ502/pKD112A glucose 42.0 5.0 8.4 BJ502/pKD112A glycerol 31.0 3.0 10.3 JM lO7/pQR182 glycerol 218.4 5.3 41.2 JM 107/pQR183 glycerol 334.0 8.1 43.0 Table 2 Relative rates (Vre,) of aldehydes in their reaction" with Preparative TK Catalysed Reactions.-To a solution of hydroxypyruvate using crude extracts of E.cofi JM107/pQR183 magnesium chloride (1 mg, 6 mmol), thiamine pyrophosphate (1 1.5 mg, 0.025 mmol) and transketolase (50 U) in glycylglycine buffer (0.5 em3, 100 mmol dm-3, pH 7.6)was added a solution of hydroxypyruvic acid (105 mg, 1 mmol) and the aldehyde (3 mmol, 3 equiv.) in glycylglycine buffer (20 em3, 100mmol drn-,, pH 7.6).The pH of the solution was then adjusted to pH 7.6 with sodium hydroxide (0.1 mol drn-,). The reaction was allowed to proceed at 37OC and monitored by TLC (ethyl acetate-methanol = 95 :5).After the reaction was complete, silica (ca. 4 g) was added to the reaction mixture and the solvent ""ql-i89 removed under reduced pressure. Flash chromatography (ethyl OH 0 acetate-petrol = 1 :1) afforded the product. (a) Using propionaldehyde the reaction yielded (3S)-1,3- dihydroxypentan-2-one (27 mg, 23%); 6,(250 MHz; D20)4.70 (1 H, d, J 19.0, HCHOH), 4.59 (1 H, d, J 19.0, HCHOH), 4.47 MehH 18 " A H 56 (1 H, dd, J7.5 and 4.5, CHOH), 2.05-1.88 (1 H, m, HCHCH3), 0 OH 0 1.91-1.70 (1 H, m, HCHCH,) and 1.07 (3 H, t, J 7.5, CH,); (CH) and 213.7 (GO).Treatment with benzoyl chloride in ac(62.9 MHz; D2O) 7.9 (CH,), 25.7 (CH,), 64.6 (CH,), 75.3 pyridine yielded the corresponding dibenzoate ester (25 mg, 45%); [u]D = -5.6 (C 0.7,CHC1,).' O q H 14 (b) Using pyruvaldehyde the reaction yielded (3s)- 1,3-di- OH OH 0 hydroxypentan-2,4-dione (2 1 mg, 16%); 6,(250 MHz; D20) 4.05-3.94 [l H, m, CH(OH)], 3.66 (1 H, dd, J 4.3 and 11.6, HCHOH), 3.56 (1 H, dd, J 7.8 and 11.6, HCHOH) and 1.26 (3 QfH0 H O G H H, d, J 6.5, CH3CO); 6,(62.9 MHz; D,O) 17.5 (CH,), (CH,), 67.4 (CH) and 217 (2 x C=O). 66.2 OH OH 0 Acknowledgements We thank Professor J. W. Frost for a gift of the strains E. coli BJ502 and BJ502/pKD112A and the SERC for the award of a 0 studentship to G.R. H. a The reaction mixture contained the following expressed as final References concentration: TPP (0.2 mmol dm-3), MgCl,*6H,O (0.9 mmol dm-3), 1 E. Racker, in The Enzymes, ed.P. D. Boyer, H. Lardy and K. aldehyde (100 mmol dm-3), hydroxypyruvate (7.5 mmol dm-3), Myrzback, vol. 5, p. 397, Academic Press, New York, 1961. glycylglycine buffer (70 mmol dm-3, pH 7.6), transketolase (3U) in a 2 K. M. Draths, D. L. Pompliano, D. L. Conley, J. W. Frost, A. Berry, final volume of 1 cm3. TK was added to initiate the reaction and aliquots G. L. Disbrow, R. J. Staversky and J. C. Lievense, J. Am. Chem. Soc., (30 mm3) were taken at 4 min intervals, deproteinised immediately and 1992,114,3956. assayed for hydroxypyruvate.' 3 C. Demuynck, J. Bolte, L. Hecquet and V. Dalmas, Tetrahedron Lett., 1991,32,5085. 4 A. Datta and E. Racker, J. Bid. Chem., 1961,236,617. 5 J. Villafranca and B. Axelrod, J. Bid. Chem., 1971,246,3126.medium (50 em3 each).The batch cultures were incubated as 6 D. C. Myles, P. J. Andrulis and G. M. Whitesides, Tetrahedron Lett., above and then used to inoculate a fermenter (2 dm3) 1991,32,4835. containing the identical medium as the batch culture at 30 "C 7 K. M. Draths and J. W. Frost, J. Am. Chem. Soc., 1990,112,1657. with an aeration rate of 6 dm3 min-'. Growth was allowed to 8 Transketolase assayed as described by C. P. Heinrick, K. Noack and proceed for 16 h after which the cells were harvested by centri- 0.Wiss, Biochem. Biophys. Res. Commun., 1972,49, 1427. fugation (15 min, 3800 rpm) and resuspended in glycylglycine 9 Hydroxypyruvate assayed as described by A. W. Holldorf, in Methods in Enzymology, vol. 3, Academic Press, New York, 1966, 578. buffer (0.1 mol dmP3, pH 7.6, 20 em3). The cell free extract was prepared by sonication (3 x 1 min sonication followed by 1 min Paper 2/05949E interval) and centrifugation (15 min, 3800 rpm) to remove cell Received 6th November 1992 debris. Accepted 20th November 1992
ISSN:1472-7781
DOI:10.1039/P19930000165
出版商:RSC
年代:1993
数据来源: RSC