ChemComm View Article Online
Published on 01 October 2013. Downloaded by University of Pittsburgh on 30/10/2014 05:17:36.
COMMUNICATION
Cite this: Chem. Commun., 2013, 49, 10914 Received 14th August 2013, Accepted 30th September 2013
View Journal | View Issue
Designer cells for stereocomplementary de novo enzymatic cascade reactions based on laboratory evolution† ´n Agudoab and Manfred T. Reetz*ab Rube
DOI: 10.1039/c3cc46229c www.rsc.org/chemcomm
Designer cells for a synthetic cascade reaction harnessing selective redox reactions were devised, featuring two successive regioselective P450-catalyzed CH-activating oxidations of 1-cyclohexene carboxylic acid methyl ester followed by stereoselective olefin-reduction catalysed by (R)- or (S)-selective mutants of an enoate reductase.
The benefits of performing sequential organic transformations without isolating intermediate products in a one-pot manner using synthetic reagents, catalysts or enzymes have been documented in numerous studies, such processes being termed ‘‘cascade’’, ‘‘domino’’ or tandem reactions.1,2 Nature orchestrates the buildup of structural complexity by enabling such reaction sequences in vivo in the cytosol of cells in which a multitude of enzymes function as selective catalysts. In the quest to produce complex natural products or biofuels researchers have used whole cells, state of the art metabolic engineering offering intriguing opportunities.3 These systems are considerably more complex than simple so-called designer cells4 such as engineered Escherichia coli cells harboring an alcohol dehydrogenase (ADH) and the necessary NAD(P)H regeneration enzyme as a platform for the asymmetric reduction of non-natural prochiral ketones.4a,b Alternatively, mixtures of isolated enzymes can be used, but this requires easy protein expression, chemical compatibility and sufficient stability under operating conditions.5 The de novo construction of any given non-natural reaction sequence that a synthetic organic chemist might envision in a designer cell is also challenging, especially for the conversion of achiral non-natural compounds into more complex value-added chiral products. Of particular interest are those cascade systems which include regioselective oxidative CH-activation, leading from simple starting materials to value-added intermediates for further elaboration. De novo construction of synthetic pathways on the basis of designer cells as described herein is an alternative to a
¨t Marburg, Hans-Meerwein Str., Fachbereich Chemie, Philipps-Universita 35032 Marburg, Germany. Fax: +49 (0)6421 28 25520; Tel: +49 (0)6421 28 25500 b ¨r Kohlenforschung, Kaiser-Wilhelm-Platz 1, Max-Planck-Institut fu ¨lheim an der Ruhr, Germany. E-mail: [email protected] 45470 Mu † Electronic supplementary information (ESI) available: Experimental details and chromatograms. See DOI: 10.1039/c3cc46229c
10914
Chem. Commun., 2013, 49, 10914--10916
modern synthetic organic techniques, especially if the use of enzyme mixtures in vitro is less efficient or fails completely. Related cases have been reported, but these require the use of chiral compounds which are consumed in the reaction sequence.6 Here we report the construction of two stereocomplementary designer cells which induce the following three-step transformations: 1 - (R)/(S)-2 - 3 - (R)-4 and 1 - (R)/(S)-2 - 3 - (S)-4, respectively (Scheme 1). The first two steps in each stereochemical strain require regioselective oxidative CH-activation catalysed by a cytochrome P450 enzyme,7 the choice being P450-BM3 from Bacillus megaterium.8 It consists of a heme-Fe-dependent monooxygenase and an electron-delivering NADPH-dependent diflavin-reductase. Since exploratory experiments showed that WT P450-BM3 fails in this respect, we counted on directed evolution9 in the quest to obtain a regio- and chemoselective mutant. The final stereoselective step was envisioned to be catalysed by the enoate reductase YqjM,10 which is a prominent member of the Old Yellow Enzyme family.11 We have previously evolved the necessary (R)- and (S)-selective YqjM mutants12 by iterative saturation mutagenesis.9i Compound 1 is not reduced under the experimental conditions. All three redox reactions require an NADPH regeneration system. Accordingly, glucose dehydrogenase (GDH) was chosen as the third enzyme with glucose being a cheap source of the necessary electrons. Recently we engineered an E. coli strain (BOU730) harboring GDH,13 which was used in all further elaboration. In order to achieve the envisioned multi-enzymatic cascade transformation using designer cells, several problems can be expected due to possible detrimental effects arising from potential metabolic burden.3 Three approaches were tested: (1) Use of two different engineered E. coli cells in a one-pot process, i.e., BOU730 cells containing P450-BM3 mutant genes and BOU730 cells containing the YqjM mutant genes. This enables a convenient control element over the multistep process, because the ratio of the two cells can be appropriately adjusted, while the second cells can be added strategically after the first have performed their function as revealed by reaction monitoring. (2) Use of a one-pot two plasmid system based on BOU730 cells transformed with two different plasmids that encode for P450-BM3 and YqjM mutants. This may lead to a metabolic This journal is
c
The Royal Society of Chemistry 2013
View Article Online
Published on 01 October 2013. Downloaded by University of Pittsburgh on 30/10/2014 05:17:36.
Communication
Scheme 1
ChemComm
Enzymatic cascade reactions envisioned for the construction of stereocomplementary designer cells harboring three different enzymes.
burden influencing bacterial growth rate and/or protein expression rates. (3) Use of engineered E. coli strains that harbor the YqjM gene inserted into the genome with P450-BM3 remaining in a plasmid. Problems similar to approach (2) may occur. In the present study we tested and compared all three options. Exploratory experiments regarding the regioselective formation of ketone 3 using wild-type (WT) P450-BM3 or the often employed ‘‘standard’’ mutant,7,8 F87A, led to poor results, meaning a mixture of compounds containing 10% of the desired product under screening-reaction conditions. Thus, the first task was protein engineering of a selective P450-BM3 mutant. We screened libraries produced by saturation mutagenesis at various sites at or near the binding pocket of P450-BM3 and also tested previous mutants,14 the total number of assayed variants amounting to about 4000. Maximum conversion of 1 to 3 within a defined time period served as the parameter of interest. The best variant proved to be V78L/A82F/F87A (55% yield of 3 under screening-reaction conditions). WT BM3 and mutants F87A and V78L/A82F/F87A are expressed equally well. Thus, the amount of protein present in each case is comparable (see ESI†). In optimization experiments following approach (1) at a 1.5 mM scale using BOU730 cells, the first (resting) cells harboring the P450BM3 mutant V78L/A82F/F87A designed for the two-step process 1 - 2 - 3 were incubated together with the starting substrate in a buffered solution containing glucose for one hour, followed by the addition of the second designer cells containing either (R)- or (S)-selective YqjM mutants. After a total reaction period of 75 minutes, this provided the final products (R)- or (S)-4 on an optional basis. In both cases respectable results were observed as revealed by GC analyses, namely 85% overall conversion to (R)-4 (99% ee) and (S)-4 (99% ee), respectively (60% yield in both cases). Upon upscaling the reaction to 7.3 mM of starting compound 1 in the case of the two-cell system with the (R)-selective YqjM mutant, 72% of the total products proved to be the desired compound (R)-4 in enantiomerically pure form (99% ee). In the case of the (S)-selective mutant, 75% of the total products turned out to be (S)-4 (99% ee) (69% yield in both cases). Importantly, when using both designer cells from the very beginning in the absence of a time-lag (1 hour), the results proved to be clearly inferior (less than 45% of 4). The two-step oxidation process (60 minutes) was found to be slower than the final reduction step (15 minutes). Thus, the use of two different designer cells at different stages of the cascade reaction This journal is
c
The Royal Society of Chemistry 2013
sequence in a one-pot procedure constitutes a convenient control element. We then turned to the second approach likewise involving a one-pot process, but using BOU730 cells transformed with two different plasmids, pRSF-P450BM3 and pACYC-YqjM, which encode for P450-BM3 and YqjM mutants, respectively, under regulation of bacteriophage T7 promoter. When IPTG is added to the medium, both proteins are expressed. Although full optimization was not attempted, reasonable results were observed both for (R)- and (S)-selective designer (resting) cells at a 1.5 mM scale, namely 48% regioselectivity as measured by consideration of all (side)products, with essentially complete stereoselectivity resulting as before (Z99% ee for (R)-4 and (S)-4, respectively) (40% yield in both cases). The reason for the less efficient performance of this system relative to approach (1) probably has to do with the lack of regulation of both proteins that are produced at the same time after IPTG induction. GC-MS analysis of the crude reaction product showed the presence of compounds resulting from over-oxidation of compounds 3 and 4 as shown by GC/MS spectra (see ESI†). The most difficult approach in terms of controlling all necessary parameters is option (3), in which a one-plasmid system encoding the P450-BM3 mutant necessary in the first two steps is utilized while the YqjM coding sequences of the (R)- or (S)-selective mutants are also inserted into the E. coli chromosome using BOU730 cells. The decision as to where in the E. coli genome to place the YqjM mutant gene may be a critical issue, since an exogenous piece of DNA inserted into the E. coli genome can adversely influence the metabolism of the strain.3 In this particular case, we chose to replace the endogenous NemA15 gene (that encodes for a flavoprotein belonging to the Old Yellow Enzyme family11 including YqjM10) for the corresponding YqjM mutant genes. Two cell strains were engineered in which a plasmid encoding the P450-BM3 mutant V78L/A82F/F87A was transformed, and the sequence from YqjM mutants were incorporated downstream to the T7 promoter into the E. coli genome by in vivo homologous recombination using the reliable Red–ET system.16 Addition of IPTG triggers overexpression of P450-BM3 and YqjM proteins. The results proved to be surprisingly good. In the case of the ‘‘(R)-strain’’, the desired overall regio- and chemoselectivity using resting cells amounted to 52% (at a 1.5 mM scale, 45% yield of 4), similar to the performance of the ‘‘(S)-strain’’ (55% regio- and chemoselectivity at a 1.5 mM scale; 50% yield). Thus, this approach provides results similar to option (2). Table S7 (ESI†) summarizes all results once more for convenient comparison. Chem. Commun., 2013, 49, 10914--10916
10915
View Article Online
Published on 01 October 2013. Downloaded by University of Pittsburgh on 30/10/2014 05:17:36.
ChemComm In all three options the integrity of the cells was maintained under operating conditions. A mixture of the involved enzymes in the form of lysates (e.g., from the two plasmid system) leads to poor results (o26% 4). The overall transformation is simple but not trivial using conventional synthetic reagents or catalysts, certainly not in a one-pot manner in which an oxidant and a reductant are both present, reagents which would not tolerate each other chemically. Compound rac-4 has been used in the synthesis of several therapeutic drugs in the pharmaceutical industry.17 It remains to be seen if the same concept can be put into practice when dealing with structurally more complex nonnatural starting materials and products which require longer multi-enzyme reaction sequences. Finally, we note that the use of mixtures of appropriately designed cells offers an alternative to molecular biological optimization necessary when constructing efficient single or multi-plasmid systems. Nevertheless, all three approaches featured here deserve attention in future studies.
Communication
5
6
7
Notes and references 1 Reviews of synthetic reagents and/or catalysts in multi-step cascade reactions: (a) K. C. Nicolaou, D. J. Edmonds and P. G. Bulger, Angew. Chem., 2006, 125, 7292 (Angew. Chem., Int. Ed., 2006, 45, 7134); (b) L. F. Tietze, G. Brasche and K. Gericke, Domino Reactions in ¨ller, Metal Organic Synthesis, Wiley-VCH, Weinheim, 2006; (c) T. J. Mu Catalyzed Cascade Reactions, Springer-Verlag, Heidelberg, 2006; (d) J. Zhu and H. Bienayme, Multicomponent Reactions, Wiley-VCH, Weinheim, 2005; (e) H. Pelissier, Adv. Synth. Catal., 2012, 354, 237; ¨rfert, Pure Appl. Chem., 2010, 82, 1375; ( f ) L. F. Tietze and A. Du ( g) C. Grondal, M. Jeanty and D. Enders, Nat. Chem., 2010, 2, 167. 2 Reviews and selected papers on enzymatic cascade reactions: (a) Multi-Step Enzyme Catalysis, ed. E. Garcia-Junceda, Wiley-VCH, Weinheim, 2008; (b) E. Ricca, B. Brucher and J. H. Schrittwieser, Adv. Synth. Catal., 2011, 353, 2239; (c) S. F. Mayer, W. Kroutil and K. Faber, Chem. Soc. Rev., 2001, 30, 332; (d) K. M. Koeller and C.-H. Wong, Chem. Rev., 2000, 100, 4465; (e) L. Cantarella, A. Gallifuoco, A. Malandra, L. Martinkova, F. Pascquarelli, A. Spera and M. Cantarella, Enzyme Microb. Technol., 2010, 47, 64; ¨hler and A. Schmid, Chem. Soc. ( f ) M. Schrewe, M. K. Julsing, B. Bu Rev., 2013, DOI: 10.1039/c3c60011d; ( g) A. Galkin, L. Kulakova, T. Yoshimura, K. Soda and N. Esaki, Appl. Environ. Microbiol., 1997, 63, 4651; (h) C. V. Voss, C. C. Gruber, K. Faber, T. Knaus, P. Macheroux and W. Kroutil, J. Am. Chem. Soc., 2008, 130, 13969; (i) J. K. Blum, A. L. Deaguero, C. V. Perez and A. S. Bommarius, ChemCatChem, 2008, 2, 987. 3 Selected reviews and papers on metabolic engineering: (a) F. LopezGallego and C. Schmidt-Dannert, Curr. Opin. Chem. Biol., 2010, 14, 174; (b) Y. Kung, W. Runguphan and J. D. Keasling, ACS Synth. Biol., 2012, 1, 498; (c) V. G. Yadav, M. De Mey, C. G. Lim, P. K. Ajikumar and G. Stephanopoulos, Metab. Eng., 2012, 14, 233; (d) E. A. Felnagle, A. Chaubey, E. L. Noey, K. N. Houk and J. C Liao, Nat. Chem. Biol., 2012, 8, 518; (e) J. W. Lee, D. Na, J. M. Park, J. Lee, S. Choi and S. Y. Lee, Nat. Chem. Biol., 2012, 8, 536; ( f ) J. Nielsen, C. Larsson, A. van Maris and J. Pronk, Curr. Opin. Biotechnol., 2013, 24, 398; ( g) A. Bar-Even and D. S. Tawfik, Curr. Opin. Biotechnol., 2013, 24, 310; (h) P. Carbonell, A.-G. Planson, D. Fichera and J.-L. Faulson, BMC Syst. Biol., 2011, 5, 122; (i) S. Kushnir, U. Sundermann, S. Yahiaoul, A. Brockmeyer, P. Janning and F. Schulz, Angew. Chem., Int. Ed., 2012, 51, 10664; ( j) C. M. Pirie, M. De Mey, K. L. Jones Prather and P. K. Ajikumar, ACS Chem. Biol., 2013, 8, 662. 4 Recent reports of designer cells as catalyst systems in the produc¨ger, F. Chamouleau, tion of (chiral) organic compounds: (a) H. Gro N. Orologas, C. Rollmann, K. Drauz, W. Hummel, A. Weckbecker and O. May, Angew. Chem., Int. Ed., 2006, 45, 5677; (b) T. Ema, S. Die, N. Okita and T. Sakai, Adv. Synth. Catal., 2008, 350, 2039; (c) K. Schroer, K. P. Luef, F. S. Hartner, A. Glieder and B. Pscheidt, Metab. Eng., 2010, 12, 8; (d) W. Szymanski, C. P. Postema, C. Tarabiono, F. Berthiol, L. Campbell-Verduyn, S. de Wildeman,
10916
Chem. Commun., 2013, 49, 10914--10916
8
9
10
11
12 13 14 15 16 17
J. G. de Vries, B. L. Feringa and D. B. Janssen, Adv. Synth. Catal., 2010, 352, 2111; (e) J.-W. Song, E.-Y. Jeon, D.-H. Song, H.-Y. Jang, U. T. Bornscheuer, D.-K. Oh and J.-B. Park, Angew. Chem., Int. Ed., 2013, 52, 2534. Key studies of the use of enzyme mixtures in a single reaction vessel: (a) C. A. Roessner, J. B. Spencer, N. J. Stolowich, J. Wang, G. P. Nayar, P. J. Santander, C. Pichon, C. Min, M. T. Holderman and A. I. Scott, Chem. Biol., 1994, 1, 119; (b) A. Bruggink, R. Schoevaart and T. Kiesboom, Org. Process Res. Dev., 2003, 7, 622; (c) Q. Cheng, L. Xiang, M. Izumikawa, D. Meluzzi and B. S. Moore, Nat. Chem. Biol., 2007, 3, 557; (d) C. J. Balibar, A. R. Howard-Jones and C. T. Walsh, Nat. Chem. Biol., 2007, 3, 584; (e) P. Pahari, M. K. Kharel, M. D. Shepherd, S. G. van Lanen and J. Rohr, Angew. Chem., Int. Ed., 2012, 51, 1216. (a) G. Fotheringham, N. Grinter, D. P. Pantaleone and R. F. Senkpeil, Bioorg. Med. Chem., 1999, 7, 2209; (b) C. U. Ingram, M. Bommer, M. E. B. Smith, P. A. Dalby, J. M. Ward, H. C. Hailes and G. J. Lye, Biotechnol. Bioeng., 2007, 96, 559; (c) W. Niu, M. N. Molefe and J. W. Frost, J. Am. Chem. Soc., 2003, 125, 12998. Selected reviews on P450 enzymes: (a) P. R. Ortiz de Montellano, Cytochrome P450: Structure, Mechanism, and Biochemistry, SpringerVerlag, Berlin, 3rd edn, 2005; (b) E. M. Isin and F. P. Guengerich, Biochim. Biophys. Acta, 2007, 1770, 314; (c) A. W. Munro, H. M. Girvan, A. E. Mason, A. J. Dunford and K. J. McLean, Trends Biochem. Sci., 2013, 38, 140; (d) M. C. Damsten, B. M. van VugtLussenburg, T. Zeldenthuis, J. S. de Vlieger, J. N. Commandeur and N. P. Vermeulen, Chem.-Biol. Interact., 2008, 171, 96; (e) C. J. C. Whitehouse, S. G. Bell and L.-L. Wong, Chem. Soc. Rev., 2012, ¨hler, S. Flitsch and N. J. Turner, Chem. 41, 1218; ( f ) E. O’Reilly, V. Ko Commun., 2011, 47, 2490. (a) L. O. Narhi and A. J. Fulco, J. Biol. Chem., 1986, 261, 7160; (b) A. W. Munro, D. J. Leys, K. J. McLean, K. R. Marshall, T. W. B. Ost, S. Daff, C. S. Miles, S. K. Chapman, D. A. Lysek, C. C. Page and P. L. Dutton, Trends BioChem. Sci., 2002, 27, 250; (c) T. Jovanovic, R. Farid, R. A. Friesner and A. E. McDermott, J. Am. Chem. Soc., 2005, 127, 13548; (d) K. H. Clodfelter, D. J. Waxman and S. Vajda, Biochemistry, 2006, 45, 9393; (e) U. Schwaneberg, A. Sprauer, C. Schmidt-Dannert and R. D. J. Schmid, J. Chromatogr., A, 1999, 848, 149. Recent reviews on directed evolution: (a) N. J. Turner, Nat. Chem. Biol., 2009, 5, 567; (b) M. B. Quin and C. Schmidt-Dannert, ACS Catal., 2011, 1, 1017; (c) E. M. Brustad and F. H. Arnold, Curr. Opin. ¨ckel and D. Hilvert, Curr. Opin. Chem. Biol., 2011, 15, 201; (d) C. Ja Biotechnol., 2010, 21, 753; (e) A. S. Bommarius, J. K. Blum and M. J. Abrahamson, Curr. Opin. Chem. Biol., 2011, 15, 194; ( f ) P. A. Dalby, Curr. Opin. Struct. Biol., 2011, 21, 473; ( g) R. M. P. Siloto and R. J. Weselake, Biocatal. Agric. Biotechnol., 2012, 1, 181; (h) N. U. Nair, C. A. Denard and H. Zhao, Curr. Org. Chem., 2010, 14, 1870; (i) M. T. Reetz, Angew. Chem., Int. Ed., 2011, 50, 138. (a) T. B. Fitzpatrick, N. Amrhein and P. J. Macheroux, Biol. Chem., 2003, 278, 19891; X-ray structure of YqjM: (b) K. Kitzing, T. B. Fitzpatrick, C. Wilken, J. Sawa, G. P. Bourenkov, P. Macheroux and T. J. Clausen, Biol. Chem., 2005, 280, 27904. Reviews on Old Yellow Enzymes (OYEs): (a) R. Stuermer, B. Hauer, M. Hall and K. Faber, Curr. Opin. Chem. Biol., 2007, 11, 203; (b) H. S. Toogood, J. M. Gardiner and N. S. Scrutton, ChemCatChem, 2010, 2, 892; (c) M. Hall and A. S. Bommarius, Chem. Rev., 2011, 111, 4088; (d) C. K. Winkler, G. Tasnadi, D. Caly, M. Hall and K. Faber, J. Biotechnol., 2012, 162, 381; (e) D. J. Bougioukou and J. D. Stewart, in Enzyme Catalysis in Organic Synthesis, ed. K. Drauz, ¨ger and O. May, Wiley-VCH, Weinheim, 2012, pp. 1111–1203. H. Gro D. J. Bougioukou, S. Kille, A. Taglieber and M. T. Reetz, Adv. Synth. Catal., 2009, 351, 3287. R. Agudo, G.-D. Roiban and M. T. Reetz, ChemBioChem, 2012, 13, 1465. S. Kille, F. E. Zilly, J. P. Acevedo and M. T. Reetz, Nat. Chem., 2011, 3, 738. (a) K. Miura, Y. Tomioka, H. Suzuki, M. Yonezawa, T. Hishinuma and M. Mizugaki, Biol. Pharm. Bull., 1997, 20, 110; (b) R. E. Williams and N. C. Bruce, Microbiology, 2002, 148, 1607. Y. Zhang, F. Buchholz, J. P. P. Muyrers and A. F. Stewart, Nat. Genet., 1998, 20, 123. (a) Bayer Healthcare AG, WO 2004/052839 A1, 2004; (b) GlaxoSmithKline, WO 2012/174342 A1, 2012; (c) Pfizer, WO 2005/080317 A2, 2005; (d) Merck Sharp & Dohme, WO 2011/106272 A1, 2011.
This journal is
c
The Royal Society of Chemistry 2013
Published on 01 October 2013. Downloaded by University of Pittsburgh on 30/10/2014 05:17:36.
COMMUNICATION
Cite this: Chem. Commun., 2013, 49, 10914 Received 14th August 2013, Accepted 30th September 2013
View Journal | View Issue
Designer cells for stereocomplementary de novo enzymatic cascade reactions based on laboratory evolution† ´n Agudoab and Manfred T. Reetz*ab Rube
DOI: 10.1039/c3cc46229c www.rsc.org/chemcomm
Designer cells for a synthetic cascade reaction harnessing selective redox reactions were devised, featuring two successive regioselective P450-catalyzed CH-activating oxidations of 1-cyclohexene carboxylic acid methyl ester followed by stereoselective olefin-reduction catalysed by (R)- or (S)-selective mutants of an enoate reductase.
The benefits of performing sequential organic transformations without isolating intermediate products in a one-pot manner using synthetic reagents, catalysts or enzymes have been documented in numerous studies, such processes being termed ‘‘cascade’’, ‘‘domino’’ or tandem reactions.1,2 Nature orchestrates the buildup of structural complexity by enabling such reaction sequences in vivo in the cytosol of cells in which a multitude of enzymes function as selective catalysts. In the quest to produce complex natural products or biofuels researchers have used whole cells, state of the art metabolic engineering offering intriguing opportunities.3 These systems are considerably more complex than simple so-called designer cells4 such as engineered Escherichia coli cells harboring an alcohol dehydrogenase (ADH) and the necessary NAD(P)H regeneration enzyme as a platform for the asymmetric reduction of non-natural prochiral ketones.4a,b Alternatively, mixtures of isolated enzymes can be used, but this requires easy protein expression, chemical compatibility and sufficient stability under operating conditions.5 The de novo construction of any given non-natural reaction sequence that a synthetic organic chemist might envision in a designer cell is also challenging, especially for the conversion of achiral non-natural compounds into more complex value-added chiral products. Of particular interest are those cascade systems which include regioselective oxidative CH-activation, leading from simple starting materials to value-added intermediates for further elaboration. De novo construction of synthetic pathways on the basis of designer cells as described herein is an alternative to a
¨t Marburg, Hans-Meerwein Str., Fachbereich Chemie, Philipps-Universita 35032 Marburg, Germany. Fax: +49 (0)6421 28 25520; Tel: +49 (0)6421 28 25500 b ¨r Kohlenforschung, Kaiser-Wilhelm-Platz 1, Max-Planck-Institut fu ¨lheim an der Ruhr, Germany. E-mail: [email protected] 45470 Mu † Electronic supplementary information (ESI) available: Experimental details and chromatograms. See DOI: 10.1039/c3cc46229c
10914
Chem. Commun., 2013, 49, 10914--10916
modern synthetic organic techniques, especially if the use of enzyme mixtures in vitro is less efficient or fails completely. Related cases have been reported, but these require the use of chiral compounds which are consumed in the reaction sequence.6 Here we report the construction of two stereocomplementary designer cells which induce the following three-step transformations: 1 - (R)/(S)-2 - 3 - (R)-4 and 1 - (R)/(S)-2 - 3 - (S)-4, respectively (Scheme 1). The first two steps in each stereochemical strain require regioselective oxidative CH-activation catalysed by a cytochrome P450 enzyme,7 the choice being P450-BM3 from Bacillus megaterium.8 It consists of a heme-Fe-dependent monooxygenase and an electron-delivering NADPH-dependent diflavin-reductase. Since exploratory experiments showed that WT P450-BM3 fails in this respect, we counted on directed evolution9 in the quest to obtain a regio- and chemoselective mutant. The final stereoselective step was envisioned to be catalysed by the enoate reductase YqjM,10 which is a prominent member of the Old Yellow Enzyme family.11 We have previously evolved the necessary (R)- and (S)-selective YqjM mutants12 by iterative saturation mutagenesis.9i Compound 1 is not reduced under the experimental conditions. All three redox reactions require an NADPH regeneration system. Accordingly, glucose dehydrogenase (GDH) was chosen as the third enzyme with glucose being a cheap source of the necessary electrons. Recently we engineered an E. coli strain (BOU730) harboring GDH,13 which was used in all further elaboration. In order to achieve the envisioned multi-enzymatic cascade transformation using designer cells, several problems can be expected due to possible detrimental effects arising from potential metabolic burden.3 Three approaches were tested: (1) Use of two different engineered E. coli cells in a one-pot process, i.e., BOU730 cells containing P450-BM3 mutant genes and BOU730 cells containing the YqjM mutant genes. This enables a convenient control element over the multistep process, because the ratio of the two cells can be appropriately adjusted, while the second cells can be added strategically after the first have performed their function as revealed by reaction monitoring. (2) Use of a one-pot two plasmid system based on BOU730 cells transformed with two different plasmids that encode for P450-BM3 and YqjM mutants. This may lead to a metabolic This journal is
c
The Royal Society of Chemistry 2013
View Article Online
Published on 01 October 2013. Downloaded by University of Pittsburgh on 30/10/2014 05:17:36.
Communication
Scheme 1
ChemComm
Enzymatic cascade reactions envisioned for the construction of stereocomplementary designer cells harboring three different enzymes.
burden influencing bacterial growth rate and/or protein expression rates. (3) Use of engineered E. coli strains that harbor the YqjM gene inserted into the genome with P450-BM3 remaining in a plasmid. Problems similar to approach (2) may occur. In the present study we tested and compared all three options. Exploratory experiments regarding the regioselective formation of ketone 3 using wild-type (WT) P450-BM3 or the often employed ‘‘standard’’ mutant,7,8 F87A, led to poor results, meaning a mixture of compounds containing 10% of the desired product under screening-reaction conditions. Thus, the first task was protein engineering of a selective P450-BM3 mutant. We screened libraries produced by saturation mutagenesis at various sites at or near the binding pocket of P450-BM3 and also tested previous mutants,14 the total number of assayed variants amounting to about 4000. Maximum conversion of 1 to 3 within a defined time period served as the parameter of interest. The best variant proved to be V78L/A82F/F87A (55% yield of 3 under screening-reaction conditions). WT BM3 and mutants F87A and V78L/A82F/F87A are expressed equally well. Thus, the amount of protein present in each case is comparable (see ESI†). In optimization experiments following approach (1) at a 1.5 mM scale using BOU730 cells, the first (resting) cells harboring the P450BM3 mutant V78L/A82F/F87A designed for the two-step process 1 - 2 - 3 were incubated together with the starting substrate in a buffered solution containing glucose for one hour, followed by the addition of the second designer cells containing either (R)- or (S)-selective YqjM mutants. After a total reaction period of 75 minutes, this provided the final products (R)- or (S)-4 on an optional basis. In both cases respectable results were observed as revealed by GC analyses, namely 85% overall conversion to (R)-4 (99% ee) and (S)-4 (99% ee), respectively (60% yield in both cases). Upon upscaling the reaction to 7.3 mM of starting compound 1 in the case of the two-cell system with the (R)-selective YqjM mutant, 72% of the total products proved to be the desired compound (R)-4 in enantiomerically pure form (99% ee). In the case of the (S)-selective mutant, 75% of the total products turned out to be (S)-4 (99% ee) (69% yield in both cases). Importantly, when using both designer cells from the very beginning in the absence of a time-lag (1 hour), the results proved to be clearly inferior (less than 45% of 4). The two-step oxidation process (60 minutes) was found to be slower than the final reduction step (15 minutes). Thus, the use of two different designer cells at different stages of the cascade reaction This journal is
c
The Royal Society of Chemistry 2013
sequence in a one-pot procedure constitutes a convenient control element. We then turned to the second approach likewise involving a one-pot process, but using BOU730 cells transformed with two different plasmids, pRSF-P450BM3 and pACYC-YqjM, which encode for P450-BM3 and YqjM mutants, respectively, under regulation of bacteriophage T7 promoter. When IPTG is added to the medium, both proteins are expressed. Although full optimization was not attempted, reasonable results were observed both for (R)- and (S)-selective designer (resting) cells at a 1.5 mM scale, namely 48% regioselectivity as measured by consideration of all (side)products, with essentially complete stereoselectivity resulting as before (Z99% ee for (R)-4 and (S)-4, respectively) (40% yield in both cases). The reason for the less efficient performance of this system relative to approach (1) probably has to do with the lack of regulation of both proteins that are produced at the same time after IPTG induction. GC-MS analysis of the crude reaction product showed the presence of compounds resulting from over-oxidation of compounds 3 and 4 as shown by GC/MS spectra (see ESI†). The most difficult approach in terms of controlling all necessary parameters is option (3), in which a one-plasmid system encoding the P450-BM3 mutant necessary in the first two steps is utilized while the YqjM coding sequences of the (R)- or (S)-selective mutants are also inserted into the E. coli chromosome using BOU730 cells. The decision as to where in the E. coli genome to place the YqjM mutant gene may be a critical issue, since an exogenous piece of DNA inserted into the E. coli genome can adversely influence the metabolism of the strain.3 In this particular case, we chose to replace the endogenous NemA15 gene (that encodes for a flavoprotein belonging to the Old Yellow Enzyme family11 including YqjM10) for the corresponding YqjM mutant genes. Two cell strains were engineered in which a plasmid encoding the P450-BM3 mutant V78L/A82F/F87A was transformed, and the sequence from YqjM mutants were incorporated downstream to the T7 promoter into the E. coli genome by in vivo homologous recombination using the reliable Red–ET system.16 Addition of IPTG triggers overexpression of P450-BM3 and YqjM proteins. The results proved to be surprisingly good. In the case of the ‘‘(R)-strain’’, the desired overall regio- and chemoselectivity using resting cells amounted to 52% (at a 1.5 mM scale, 45% yield of 4), similar to the performance of the ‘‘(S)-strain’’ (55% regio- and chemoselectivity at a 1.5 mM scale; 50% yield). Thus, this approach provides results similar to option (2). Table S7 (ESI†) summarizes all results once more for convenient comparison. Chem. Commun., 2013, 49, 10914--10916
10915
View Article Online
Published on 01 October 2013. Downloaded by University of Pittsburgh on 30/10/2014 05:17:36.
ChemComm In all three options the integrity of the cells was maintained under operating conditions. A mixture of the involved enzymes in the form of lysates (e.g., from the two plasmid system) leads to poor results (o26% 4). The overall transformation is simple but not trivial using conventional synthetic reagents or catalysts, certainly not in a one-pot manner in which an oxidant and a reductant are both present, reagents which would not tolerate each other chemically. Compound rac-4 has been used in the synthesis of several therapeutic drugs in the pharmaceutical industry.17 It remains to be seen if the same concept can be put into practice when dealing with structurally more complex nonnatural starting materials and products which require longer multi-enzyme reaction sequences. Finally, we note that the use of mixtures of appropriately designed cells offers an alternative to molecular biological optimization necessary when constructing efficient single or multi-plasmid systems. Nevertheless, all three approaches featured here deserve attention in future studies.
Communication
5
6
7
Notes and references 1 Reviews of synthetic reagents and/or catalysts in multi-step cascade reactions: (a) K. C. Nicolaou, D. J. Edmonds and P. G. Bulger, Angew. Chem., 2006, 125, 7292 (Angew. Chem., Int. Ed., 2006, 45, 7134); (b) L. F. Tietze, G. Brasche and K. Gericke, Domino Reactions in ¨ller, Metal Organic Synthesis, Wiley-VCH, Weinheim, 2006; (c) T. J. Mu Catalyzed Cascade Reactions, Springer-Verlag, Heidelberg, 2006; (d) J. Zhu and H. Bienayme, Multicomponent Reactions, Wiley-VCH, Weinheim, 2005; (e) H. Pelissier, Adv. Synth. Catal., 2012, 354, 237; ¨rfert, Pure Appl. Chem., 2010, 82, 1375; ( f ) L. F. Tietze and A. Du ( g) C. Grondal, M. Jeanty and D. Enders, Nat. Chem., 2010, 2, 167. 2 Reviews and selected papers on enzymatic cascade reactions: (a) Multi-Step Enzyme Catalysis, ed. E. Garcia-Junceda, Wiley-VCH, Weinheim, 2008; (b) E. Ricca, B. Brucher and J. H. Schrittwieser, Adv. Synth. Catal., 2011, 353, 2239; (c) S. F. Mayer, W. Kroutil and K. Faber, Chem. Soc. Rev., 2001, 30, 332; (d) K. M. Koeller and C.-H. Wong, Chem. Rev., 2000, 100, 4465; (e) L. Cantarella, A. Gallifuoco, A. Malandra, L. Martinkova, F. Pascquarelli, A. Spera and M. Cantarella, Enzyme Microb. Technol., 2010, 47, 64; ¨hler and A. Schmid, Chem. Soc. ( f ) M. Schrewe, M. K. Julsing, B. Bu Rev., 2013, DOI: 10.1039/c3c60011d; ( g) A. Galkin, L. Kulakova, T. Yoshimura, K. Soda and N. Esaki, Appl. Environ. Microbiol., 1997, 63, 4651; (h) C. V. Voss, C. C. Gruber, K. Faber, T. Knaus, P. Macheroux and W. Kroutil, J. Am. Chem. Soc., 2008, 130, 13969; (i) J. K. Blum, A. L. Deaguero, C. V. Perez and A. S. Bommarius, ChemCatChem, 2008, 2, 987. 3 Selected reviews and papers on metabolic engineering: (a) F. LopezGallego and C. Schmidt-Dannert, Curr. Opin. Chem. Biol., 2010, 14, 174; (b) Y. Kung, W. Runguphan and J. D. Keasling, ACS Synth. Biol., 2012, 1, 498; (c) V. G. Yadav, M. De Mey, C. G. Lim, P. K. Ajikumar and G. Stephanopoulos, Metab. Eng., 2012, 14, 233; (d) E. A. Felnagle, A. Chaubey, E. L. Noey, K. N. Houk and J. C Liao, Nat. Chem. Biol., 2012, 8, 518; (e) J. W. Lee, D. Na, J. M. Park, J. Lee, S. Choi and S. Y. Lee, Nat. Chem. Biol., 2012, 8, 536; ( f ) J. Nielsen, C. Larsson, A. van Maris and J. Pronk, Curr. Opin. Biotechnol., 2013, 24, 398; ( g) A. Bar-Even and D. S. Tawfik, Curr. Opin. Biotechnol., 2013, 24, 310; (h) P. Carbonell, A.-G. Planson, D. Fichera and J.-L. Faulson, BMC Syst. Biol., 2011, 5, 122; (i) S. Kushnir, U. Sundermann, S. Yahiaoul, A. Brockmeyer, P. Janning and F. Schulz, Angew. Chem., Int. Ed., 2012, 51, 10664; ( j) C. M. Pirie, M. De Mey, K. L. Jones Prather and P. K. Ajikumar, ACS Chem. Biol., 2013, 8, 662. 4 Recent reports of designer cells as catalyst systems in the produc¨ger, F. Chamouleau, tion of (chiral) organic compounds: (a) H. Gro N. Orologas, C. Rollmann, K. Drauz, W. Hummel, A. Weckbecker and O. May, Angew. Chem., Int. Ed., 2006, 45, 5677; (b) T. Ema, S. Die, N. Okita and T. Sakai, Adv. Synth. Catal., 2008, 350, 2039; (c) K. Schroer, K. P. Luef, F. S. Hartner, A. Glieder and B. Pscheidt, Metab. Eng., 2010, 12, 8; (d) W. Szymanski, C. P. Postema, C. Tarabiono, F. Berthiol, L. Campbell-Verduyn, S. de Wildeman,
10916
Chem. Commun., 2013, 49, 10914--10916
8
9
10
11
12 13 14 15 16 17
J. G. de Vries, B. L. Feringa and D. B. Janssen, Adv. Synth. Catal., 2010, 352, 2111; (e) J.-W. Song, E.-Y. Jeon, D.-H. Song, H.-Y. Jang, U. T. Bornscheuer, D.-K. Oh and J.-B. Park, Angew. Chem., Int. Ed., 2013, 52, 2534. Key studies of the use of enzyme mixtures in a single reaction vessel: (a) C. A. Roessner, J. B. Spencer, N. J. Stolowich, J. Wang, G. P. Nayar, P. J. Santander, C. Pichon, C. Min, M. T. Holderman and A. I. Scott, Chem. Biol., 1994, 1, 119; (b) A. Bruggink, R. Schoevaart and T. Kiesboom, Org. Process Res. Dev., 2003, 7, 622; (c) Q. Cheng, L. Xiang, M. Izumikawa, D. Meluzzi and B. S. Moore, Nat. Chem. Biol., 2007, 3, 557; (d) C. J. Balibar, A. R. Howard-Jones and C. T. Walsh, Nat. Chem. Biol., 2007, 3, 584; (e) P. Pahari, M. K. Kharel, M. D. Shepherd, S. G. van Lanen and J. Rohr, Angew. Chem., Int. Ed., 2012, 51, 1216. (a) G. Fotheringham, N. Grinter, D. P. Pantaleone and R. F. Senkpeil, Bioorg. Med. Chem., 1999, 7, 2209; (b) C. U. Ingram, M. Bommer, M. E. B. Smith, P. A. Dalby, J. M. Ward, H. C. Hailes and G. J. Lye, Biotechnol. Bioeng., 2007, 96, 559; (c) W. Niu, M. N. Molefe and J. W. Frost, J. Am. Chem. Soc., 2003, 125, 12998. Selected reviews on P450 enzymes: (a) P. R. Ortiz de Montellano, Cytochrome P450: Structure, Mechanism, and Biochemistry, SpringerVerlag, Berlin, 3rd edn, 2005; (b) E. M. Isin and F. P. Guengerich, Biochim. Biophys. Acta, 2007, 1770, 314; (c) A. W. Munro, H. M. Girvan, A. E. Mason, A. J. Dunford and K. J. McLean, Trends Biochem. Sci., 2013, 38, 140; (d) M. C. Damsten, B. M. van VugtLussenburg, T. Zeldenthuis, J. S. de Vlieger, J. N. Commandeur and N. P. Vermeulen, Chem.-Biol. Interact., 2008, 171, 96; (e) C. J. C. Whitehouse, S. G. Bell and L.-L. Wong, Chem. Soc. Rev., 2012, ¨hler, S. Flitsch and N. J. Turner, Chem. 41, 1218; ( f ) E. O’Reilly, V. Ko Commun., 2011, 47, 2490. (a) L. O. Narhi and A. J. Fulco, J. Biol. Chem., 1986, 261, 7160; (b) A. W. Munro, D. J. Leys, K. J. McLean, K. R. Marshall, T. W. B. Ost, S. Daff, C. S. Miles, S. K. Chapman, D. A. Lysek, C. C. Page and P. L. Dutton, Trends BioChem. Sci., 2002, 27, 250; (c) T. Jovanovic, R. Farid, R. A. Friesner and A. E. McDermott, J. Am. Chem. Soc., 2005, 127, 13548; (d) K. H. Clodfelter, D. J. Waxman and S. Vajda, Biochemistry, 2006, 45, 9393; (e) U. Schwaneberg, A. Sprauer, C. Schmidt-Dannert and R. D. J. Schmid, J. Chromatogr., A, 1999, 848, 149. Recent reviews on directed evolution: (a) N. J. Turner, Nat. Chem. Biol., 2009, 5, 567; (b) M. B. Quin and C. Schmidt-Dannert, ACS Catal., 2011, 1, 1017; (c) E. M. Brustad and F. H. Arnold, Curr. Opin. ¨ckel and D. Hilvert, Curr. Opin. Chem. Biol., 2011, 15, 201; (d) C. Ja Biotechnol., 2010, 21, 753; (e) A. S. Bommarius, J. K. Blum and M. J. Abrahamson, Curr. Opin. Chem. Biol., 2011, 15, 194; ( f ) P. A. Dalby, Curr. Opin. Struct. Biol., 2011, 21, 473; ( g) R. M. P. Siloto and R. J. Weselake, Biocatal. Agric. Biotechnol., 2012, 1, 181; (h) N. U. Nair, C. A. Denard and H. Zhao, Curr. Org. Chem., 2010, 14, 1870; (i) M. T. Reetz, Angew. Chem., Int. Ed., 2011, 50, 138. (a) T. B. Fitzpatrick, N. Amrhein and P. J. Macheroux, Biol. Chem., 2003, 278, 19891; X-ray structure of YqjM: (b) K. Kitzing, T. B. Fitzpatrick, C. Wilken, J. Sawa, G. P. Bourenkov, P. Macheroux and T. J. Clausen, Biol. Chem., 2005, 280, 27904. Reviews on Old Yellow Enzymes (OYEs): (a) R. Stuermer, B. Hauer, M. Hall and K. Faber, Curr. Opin. Chem. Biol., 2007, 11, 203; (b) H. S. Toogood, J. M. Gardiner and N. S. Scrutton, ChemCatChem, 2010, 2, 892; (c) M. Hall and A. S. Bommarius, Chem. Rev., 2011, 111, 4088; (d) C. K. Winkler, G. Tasnadi, D. Caly, M. Hall and K. Faber, J. Biotechnol., 2012, 162, 381; (e) D. J. Bougioukou and J. D. Stewart, in Enzyme Catalysis in Organic Synthesis, ed. K. Drauz, ¨ger and O. May, Wiley-VCH, Weinheim, 2012, pp. 1111–1203. H. Gro D. J. Bougioukou, S. Kille, A. Taglieber and M. T. Reetz, Adv. Synth. Catal., 2009, 351, 3287. R. Agudo, G.-D. Roiban and M. T. Reetz, ChemBioChem, 2012, 13, 1465. S. Kille, F. E. Zilly, J. P. Acevedo and M. T. Reetz, Nat. Chem., 2011, 3, 738. (a) K. Miura, Y. Tomioka, H. Suzuki, M. Yonezawa, T. Hishinuma and M. Mizugaki, Biol. Pharm. Bull., 1997, 20, 110; (b) R. E. Williams and N. C. Bruce, Microbiology, 2002, 148, 1607. Y. Zhang, F. Buchholz, J. P. P. Muyrers and A. F. Stewart, Nat. Genet., 1998, 20, 123. (a) Bayer Healthcare AG, WO 2004/052839 A1, 2004; (b) GlaxoSmithKline, WO 2012/174342 A1, 2012; (c) Pfizer, WO 2005/080317 A2, 2005; (d) Merck Sharp & Dohme, WO 2011/106272 A1, 2011.
This journal is
c
The Royal Society of Chemistry 2013
