PEDS Advance Access originally published online on December 19, 2007
Protein Engineering Design and Selection 2008 21(1):29-35; doi:10.1093/protein/gzm074
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Laboratory evolution of P450 BM3 for mediated electron transfer yielding an activity-improved and reductase-independent variant
Jacobs University, Campus Ring 8, 28759 Bremen, Germany
1To whom correspondence should be addressed. E-mail: u.schwaneberg{at}jacobs-university.de
 |
Abstract |
|---|
 Top Abstract IntroductionResultsDiscussionMaterials and methodsFundingReferences |
|---|
One of the main obstacles in employing P450 monooxygenases for preparative chemical syntheses in cell-free systems is their requirement for cofactors such as NAD(P)H. In order to engineer P450 BM3 from Bacillus megaterium for cost-effective process conditions in vitro, a validated medium throughput screening system based on cheap Zn dust as an electron source and Cobalt(III)sepulchrate (Co(III)sep) as a mediator was reported. In the current study, the alternative cofactor system Zn/Co(III)sep was used in a directed evolution experiment to improve the Co(III)sep-mediated electron transfer to P450 BM3. A variant, carrying five mutations (R47F F87A V281G M354S D363H, Table I), P450 BM3 M5 was identified and characterized with respect to its kinetic parameters. P450 BM3 M5 achieved for mediated electron transfer a 2-fold higher kcat value and a 3-fold higher catalytic efficiency compared with the starting point mutant P450 BM3 F87A (kcat: 62 min–1 compared with 28 min–1; kcat/Km: 62 µM–1min–1 compared to 19 µM–1min–1). For obtaining first insights on electron transfer contributions, three reductase-deficient variants were generated. The reductase-deficient mutant of P450 BMP M5 exhibited a catalytic efficiency of 69% and a kcat value of 89% of the values obtained for P450 BM3 M5.
|
Keywords: cofactor/directed evolution/high-throughput screening/mediator/monooxygenase
|
Introduction |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingReferences |
|---|
The ability of the heme containing ‘superfamily’ of P450 monooxygenases to oxygenate unactivated C–H bonds is one of the most attractive reactions from a catalysis point of view. The need of P450 monooxygenases for expensive cofactors, such as NADPH or NADH, limits their applications in cell-free organic syntheses. The use of whole cells for cofactor regeneration is efficient (Wolberg et al., 2000
; Appel et al., 2001; Stampfer et al., 2002), but restricted in terms of solvent choices and to substrates able to translocate through cell membranes without affecting membrane integrity. In addition, cofactor and substrate concentrations have to be balanced to achieve high catalytic efficiencies, because many P450s are inhibited in the presence of cofactors and absence of substrate (Murataliev et al., 1997).
Approaches to provide reduction equivalents to P450s comprise reports regenerating NAD(P)H (Adlercreutz, 1996; Wolberg et al., 2000; Appel et al., 2001; Drauz et al., 2002; Hollmann et al., 2002; Stampfer et al., 2002; van der Donk and Zhao, 2003), the use of peroxides as alternative electron sources (Joo et al., 1999a, 1999b; Cirino and Arnold, 2003), mediators for shuttling electrons (Estabrook et al., 1996; Reipa et al., 1997, 2002; Mayhew et al., 2000; Shumyantseva et al., 2000; Gilardi et al., 2002) and direct electron transfer on electrodes (Lvov et al., 1998; Munge et al., 2003; Bistolas et al., 2004; Fantuzzi et al., 2004; Shumyantseva et al., 2004, 2007; Shukla et al., 2005; Matsumura et al., 2006; Udit et al., 2006). NAD(P)H regenerations has been achieved electrochemically (Hollmann et al., 2002), enzymatically (Adlercreutz, 1996; Drauz et al., 2002; van der Donk and Zhao, 2003) and with in vivo recycling systems (Wolberg et al., 2000; Appel et al., 2001; Stampfer et al., 2002). The use of peroxides as alternative electron sources (Joo et al., 1999a, 1999b; Cirino and Arnold, 2003) is scientifically interesting, but suffered from low tolerance of monooxygenases toward peroxides, which resulted in rapid inactivation (Cirino and Arnold, 2003).
Previously, we reported an alternative cofactor system based on cheap zinc dust as an electron source and Cobalt(III)sepulchrate (Co(III)sep) as a mediator (Nazor and Schwaneberg, 2006). The alternative cofactors system has been shown successful in driving the monooxygenase P450 BM3 from Bacillus megaterium (Estabrook et al., 1996; Schwaneberg et al., 2000) and a Zn/Co(III)sep-driven screening system based on the Schwaneberg pNCA assay (Schwaneberg et al., 1999a) was developed and downscaled to the 96-well microtiter plate format (Nazor and Schwaneberg, 2006).
P450 BM3 has, from an application point of view, several beneficial properties compared with other P450s: (i) reductase and heme domains are on a single polypeptide chain (Narhi and Fulco, 1987; Ruettinger et al., 1989), (ii) kcat values are often 10–1000 times higher compared with other fatty acid hydroxylases (Zimmer et al., 1995), (iii) production and purification in gram scale has been achieved (Schwaneberg et al., 1999b) and (iv) validated high-throughput screening systems to improve properties exist, for instance, to broaden substrate spectra (Schwaneberg et al., 2001) and to improve organic solvent resistance (Wong et al., 2004).
Here, we report a first directed evolution experiment of a monooxygenase (P450 BM3) for mediated electron transfer (Co(III)sep) in which a whole P450 BM3 gene was randomized using error-prone polymerase chain reaction (epPCR). Individual contributions of P450 BM3’s reductase to the Co(III)sep-mediated electron transfer rates were further quantified by generating three reductase-deficient variants (P450 BMP; Table I) in which a stop codon was introduced after the heme domain by site-directed mutagenesis.
|
Results |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingReferences |
|---|
Table I summarizes the P450 BM3 variants used in this study, which have been generated by saturation mutagenesis, random mutagenesis and site-directed mutagenesis.
The P450 BM3 mutant F87A showed a more than four times higher kcat than the wild-type for the 12-pNCA substrate (Schwaneberg et al., 1999a). The differences in kcat resulted for Zn/Co(III)sep microtiter plate screening system in significantly different coefficient of variations (9.8% for F87A; 19.2% wild type; Nazor and Schwaneberg, 2006), so that the mutein F87A was selected finally for directed evolution.
Saturation mutagenesis of P450 BM3 M1 at position M354
In a previous study, the P450 BM3 M1 (Table I) variant was identified by screening a saturation mutagenesis library at position R47 of P450 BM3 F87A. P450 BM3 M1 and has a two times increased catalytic efficiency, for mediated electron transfer compared with the starting point mutant P450 BM3 F87A. The increase in catalytic efficiency could mainly be attributed to lowered Km values (Nazor and Schwaneberg, 2006). In order to have a good starting point for directed evolution of P450 BM3 F87A for mediated electron transfer, further saturation mutagenesis libraries (R47, L188, P329, M354S and L437) were screened using the developed microtiter plates (Nazor and Schwaneberg, 2006). Screening of a saturation mutagenesis library of P450 BM3 M1 at position M354 yielded eight mutants with improved activity under screening conditions. Sequencing results showed that all eight mutants have the M354S substitution, with a codon change from ATG to TCT. This mutant has been named P450 BM3 M2 (Table I). Position M354 is located within a 12 Å radius hemisphere from the position R47 and lies on a β-sheet formed by three β-strands (residues 47–53, 329–335 and 350–356; Ravichandran et al., 1993). Position M354 is one of the residues involved in the primary contacts between a fatty acid substrate and P450 BM3 (Li and Poulos, 1997).
cpPCR mutagenesis of P450 BM3 M2 and library screening
P450 BM3 M2 was taken as a template for a directed evolution experiment. Random mutations were inserted into the whole gene of P450 BM3 M2 under PCR conditions designed to introduce one to two amino acid changes per gene. In the first round of screening (
1920 variants), five mutants with increased activity were discovered and used as template for a second round of mutagenic PCR under identical conditions. Screening of another 1920 clones resulted in a clone named P450 BM3 M5 (Table I) with three times improved activity under the screening conditions. P450 BM3 M5 was analyzed in detail. Sequencing results revealed that the P450 BM3 M5 mutant contained an additional two mutations which resulted in two amino acid substitutions: V281G (GTG to GGG) and D363H (CAC to GAC).
Site-directed mutagenesis studies
The P450 BM3 mutants obtained in the first round of epPCR were not sequenced; the individual amino acid exchanged from P450 BM3 M5 were therefore introduced subsequently to P450 BM3 M2 (V281G: P450 BM3 M3; D363H: P450 BM3 M4) for elucidating their individual influence on the kinetic parameters. Reductase-deficient variants were generated by site-directed mutagenesis from P450 BM3 F87A, P450 BM3 M2 and P450 BM3 M5. These reductase-deficient variants were used for quantifying the reductase contribution of the Zn/Co(III)sep-mediated electron transfer.
Characterization of P450 BM3 mutants by 12-pNCA conversion employing NADPH and Zn/Co(III)sep as ‘electron’ donors
Figure 1 shows the Km, kcatvalues and kcat/Km of the starting point P450 BM3 F87A and all P450 BM3 variants which have been improved for the alternative cofactor system Zn/Co(III)sep. For activity measurements, the 12-pNCA assay and NADPH (Fig. 1, left part) or Zn/Co(III)sep (Fig. 1, right part) were employed as an ‘electron’ donor.
|
Employing the natural reduction equivalent, NADPH resulted in an increase of Km values for all P450 BM3 mutants compared with the starting point mutant, P450 BM3 F87A (4.0 to 22–39 µM). kcat values of P450 BM3 mutants dropped from 170 min–1 (P450 BM3 F87A) and 171 min–1 (P450 BM3 M1) to 124 and 100 min–1 for P450 BM3 M2 and P450 BM3 M5. The catalytic efficiencies of the P450 BM3 mutants dropped progressively compared with P450 BM3 F87A (42 µM–1 min–1 to 5.0, 4.3 and 4.5 for P450 BM3 M1, M2 and M5; Fig. 1, left part).
Opposite results were obtained when using the Zn/Co(III)sep electron donor system. All P450 BM3 mutants had lower Km values; the Km values ranged from 1.6 µM (P450 BM3 F87A) to 0.6, 1.0 and 1.1 µM (P450 BM3 M1, M2 and M5). The kcat values of the starting point mutant, P450 BM3 F87A, and of the double and triple mutants, P450 BM3 M1 and P450 BM3 M2, were in the same range (28 min–1 for P450 BM3 F87A, 19 min–1 for P450 BM3 M1 and 37 min–1 for P450 BM3 M2). The kcat value of the P450 BM3 M5 mutant increased by 2.3-fold compared with the starting point mutant P450 BM3 F87A (28–62 min–1). Notably, only one of the quadruple mutants, P450 BM3 M3, had the same level of activity as P450 BM3 M5. The level of activity of the other quadruple mutant, P450 BM3 M4, was comparable with that of the triple mutant P450 BM3 M2. The catalytic efficiencies for all P450 BM3 mutants, except the starting point P450 BM3 F87A, were higher when switching from NADPH to the Zn/Co(III)sep electron donor system. Catalytic efficiencies values were in the same range, for the starting point mutant (P450 BM3 F87A, 19 µM–1 min–1), the double mutant (P450 BM3 M1, 30 µM–1 min–1) and the triple mutant (P450 BM3 M2, 26 µM–1 min–1). For the P450 BM3 M5 and the quadruple mutant P450 BM3 M3, a 3.0- and 2.3-fold higher catalytic efficiency was achieved.
When switching from NADPH to the Zn/Co(III)sep, one can observe a catalytic efficiency which is 12.9-fold (P450 BM3 M5) and 14.8-fold (P450 BM3 M3) higher for the Zn/Co(III)sep electron donor system (Fig. 1).
Figure 2 shows the Km, kcatvalues and kcat/Km of reductase-deficient variants (P450 BMP F87A, P450 BMP M2 and P450 BMP M5). For activity measurements, the 12-pNCA assay and Zn/Co(III)sep were employed as an ‘electron’ donor.
|
The kinetic parameters of reductase-deficient P450 BMP variants could not be determined when employing NADPH as reduction equivalent. This is quite expectable since the NADPH-binding site is located in the removed reductase domain of P450 BM3.
On the other hand, all of the reductase-deficient P450 BMP variants were active when employing the Zn/Co(III)sep system as an electron source. The Km values of all the reductase-deficient mutants were in the same range: 1.7 µM (P450 BMP F87A), 1.2 µM (P450 BMP M2) and 1.3 µM (P450 BMP M5). The kcat values increased from 28 min–1 (P450 BMP F87A) to 31 min–1 and to 57 min–1 (P450 BMP M2 and M5). Catalytic efficiencies also increased progressively from 14 µM–1 min–1 (P450 BMP F87A) to 26 µM–1 min–1 and to 43 µM–1 min–1 (P450 BMP M2 and M5).
|
Discussion |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingReferences |
|---|
Biocatalysts are by natural design not optimized for applications in bioelectrocatalysis since random electron transfer would in living organisms result in wasting energy and radical formation, which reduces the fitness of living organisms. Redox-active centers are therefore often embedded deeply in an ‘insulating’ protein shell and sophisticated control mechanisms, such as the thermodynamic switch mechanism in P450 BM3 (Daff et al., 1997
), regulate electron transfer upon substrate binding from FMN to the heme. Successful protein engineering strategies, for improving mediated electrical communication between biocatalyst and the electrode, are, to our best knowledge, rational design approaches that have been summarized in a review (Wong and Schwaneberg, 2003). One exception is the recent improvement of the mediated electron transfer between glucose oxidase from Aspergillus niger employing ferrocenemethanol as mediator (Zhu et al., 2006). Fundamental design principles for efficiently redesigning biocatalyst for bioelectrocatalytical applications are not understood yet. Directed protein evolution allows improving biocatalyst properties without understanding targeted properties and thereby offers opportunities to generate hypotheses for redesigning biocatalysts by studying improved biocatalyst variants.
We improved mediated electron transfer properties by directed evolution of P450 BM3 for an alternative cofactor system, Zn/Co(III)sep. P450 BM3- and reductase-deficient P450 BMP variants that have been generated in this study are summarized in Table I. P450 BM3 M1 was found in a previous research effort (Nazor and Schwaneberg, 2006). In the current study, the screening of further saturation mutagenesis libraries resulted in P450 BM3 M2 which was subsequently subjected to two rounds of random mutagenesis using the classical MnCl2 epPCR method. In two rounds of random mutagenesis and screening, two additional amino acid substitutions were obtained (position 281 and 363; P450 BM3 M5).
On the basis of the results for P450 BM3 M1, we hypothesized in the initial report on the Zn/Co(III)sep-mediated screening system (Nazor and Schwaneberg, 2006) that favorable association of Co(III)sep at the entrance of the substrate access channel might improve electron transfer rates and/or substrate binding because R47 plays a key role in modulating electron transfer rates and/or substrate binding when Co(III)sep is present. Therefore, several amino acid positions in close vicinity of the substrate access channel (L188, P329, M354S and L437) were saturated and screened for improved activity. Only saturation of position M354 in P450 BM3 M1 and analysis of the eight most active clones resulted in amino acid substitution from M354S (eight out of eight sequenced clones). Position 354 is located within a 12 Å radius from the position R47 on a conformationally flexible structure element (Fig. 3; Ravichandran et al., 1993). The lack of finding improved variants, except the M354S, makes it quite unlikely that Co(III)sep can enter inside the substrate access channel of P450 BM3 and that amino acid substitutions in the binding pocket might orient or hold Co(III)sep in a favorable orientation for electron transfer. The role of the M345S might therefore lie in a relay function associated with increased flexibility of the 2-2–1-3 or 1-3–K' loop.
|
The P450 BM3 M2 mutant was subjected to two rounds of epPCR. After screening a total of
3840 colonies, it was found that P450 BM3 M5 has 2.3-fold higher kcat and 3-fold higher catalytic efficiency (Fig. 1). Comparing the catalytic efficiencies toward the natural cofactor and the alternative cofactor system, P450 BM3 M5 shows a 12.9-fold increased catalytic efficiency for the Zn/Co(III)sep system. P450 BM3 M5 contained two additional amino acid substitutions at position V281 (V–G) and D363 (D–H). To differentiate which one of these mutations played a more crucial role in enhancing activity, two quadruple mutants (containing one of the mutations each) were generated and characterized kinetically. From the two quadruple mutants, it was found that the kinetic constants of only P450 BM3 M3 was close to that of P450 BM3 M5, whereas the kinetic constant of P450 BM3 M4 was comparable with that of P450 BM3 M2. The D363H substitution increases, in contrast to the amino acid substitution, V281G, the kcat for 12-pNCA conversion.
Position D363 is located on the K'–L loop of the heme domain and its side-chain carboxyl group is hydrogen bonded to the backbone amide group of Q55 which is located on the B helix (Fig. 3, enlarged). The hydrogen bond between Q55 and D363 holds the two structural elements in close vicinity to each other and has been shown to remain stable to an extent of 80% in a 15 ns simulation of the wild-type P450 BM3 and the P450 BM3 F87A mutant (Roccatano et al., unpublished results). Replacing D363 with a hystidine residue could lead to the weakening of a loop stabilizing interactions. By visual inspection of the crystal structure of the complex between the FMN and the heme binding domains (pdb code: 1bvy), 50 
-bonds were identified that could provide a pathway for electron transfer from FMN to heme (via indol ring conjugated 
-orbitals of W574 to the stretch of amino acids N381-Q387 located on the heme-binding loop). This stretch of amino acids is located on the same K'–L loop as D363 (Fig. 4). In the light of this proposed electron transfer mechanism (Sevrioukova et al., 1999), the weakening of D363–Q55 interactions might alter the structure of the region. Hence, an altered loop orientation could expose the N381–Q387 peptide stretch to the solvent, making it more accessible for the mediator Co(III)sep and thereby leading to an enhanced electron transfer rate.
|
The comparison of reductase-deficient variants (P450 BMP F87A, P450 BMP M2 and P450 BMP M5) with the P450 BM3 M5 mutant shows that the reductase-deficient variants are not active with NADPH. For P450 BMP M5 using Zn/Co(III)sep as an electron source, a kcat value and catalytic efficiency of 89 and 69% were achieved compared with P450 BM3 M5 (Figs. 2 and 1, black part). These results show that the electron transfer from the reductase to the heme contributes only to a minor extent to the overall electron transfer rates of the Co(III)sep-mediated electron transfer. In the light of the electron transfer mechanism proposed by Servioukova et al., one could hypothesize that Co(III)sep interfaces in P450 BM3 M5 with the proposed electron pathway from FMN to the N381–Q387 peptide stretch. Saturation mutagenesis and focused randomized mutant libraries in the K'–L loop (Fig. 4) are currently being generated to understand possible interfacing mechanisms. These studies are complemented by crystallization efforts in the presence of Co(III)sep.
Out of curiosity, the total turnover numbers (TTNs) for Zn/Co(III)sep were determined under non-optimized conditions to be 936 for P450 BM3 M5 and 670 for P450 BMP M5. These TTN are still one to two orders of magnitude below any industrial consideration. However, directed evolution of P450 BM3 for Zn/Co(III)sep-mediated electron transfer demonstrates that it is possible to improve bioelectrocatalytical properties that were never required for biological function. Directed evolution raises, therefore, hopes that redox proteins in general can be engineered for bioelectrochemical applications in industrial biotechnology through introducing or redesigning interfaces between biocatalyst and an electrochemical setup.
|
Materials and methods |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingReferences |
|---|
All chemicals used were of analytical reagent grade or higher quality and purchased from Sigma-Aldrich (Steinheim, Germany), AppliChem (Darmstadt, Germany) and Carl Roth (Karlsruhe, Germany). Zn dust of p.a. grade was obtained from Riedel-de Haen (Seelze, Germany), and the 12-pNCA substrate was synthesized as described previously (Schwaneberg et al., 1999a
). Thermal cycler (Mastercyler® gradient; Eppendorf, Hamburg, Germany) and thin-wall PCR tubes (Mµlti-Ultra tubes; 0.2 ml; Carl Roth, Germany) were used in all PCRs. The PCR volume was always 50 µl. The amount of DNA in cloning experiments was quantified using a NanoDrop photometer (NanoDrop Technologies, DE, USA).
Saturation mutagenesis of P450 BM3 M1 at position M354
The saturation mutagenesis library of P450 BM3 M1 at position M354 was generated using the vector pCYTEXP1 (Belev et al., 1991). The saturation mutagenesis library was subsequently cloned from pCYTEXP1 into pCWori (Barnes et al., 1991) for high-level expression in microtiter plates using the restriction sites BamHI and EcoRI. For saturation of position M354, 35 ng of P450 BM3 M1 template DNA was employed. A volume of 17.5 pmol of each mutagenic primer were used: forward primer: 5'-TGCGGGACACGAAANNNCAAGTGGTCTTTTA-3'; reverse primer: 5'-TAAAAGACCACTTGNNNTTTCG TGTCCCGCA-3' under the following PCR conditions: 94°C for 4 min 1 cycle; 94°C for 1.5 min/55°C for 1 min/68°C for 17 min 20 cyles; 68°C for 20 min 1 cycle, employing 2.5 U Pfu polymerase (Fermentas, St. Leon-Rot, Germany) and 200 nM dNTP. Subsequently, 40 U of DpnI (New England Biolabs, Frankfurt am Main, Germany) were added after the PCR and incubated for 3 h at 37°C. The PCR product was purified using NucleoSpin Extract kit (Machery-Nagel, Düren, Germany) and transformed into Escherichia coli DH5
. The mutant library was subsequently isolated from pooled agar plate clones, digested and cloned, for screening, into pCWori as described previously (Wong et al., 2005). The improved variant P450 BM3 M2 containing the amino acid substitution M354S was used for randomized mutant library generation.
Error-prone mutant libraries of P450 BM3 M2
Error-prone mutagenesis libraries of P450 BM3 M2 were generated as described previously (Wong et al., 2004). In all PCRs, 25 pmol of each primer were used: forward primer: 5'-TATTATGATTCTTAATGATGATGATGATGATGCCCAG-3'; reverse primer: 5'-ATTATTATCCATGACAATTAAAGAAATGCCTC-3' and 80 ng of P450 BM3 M2 template DNA, under the following PCR conditions: 94°C for 1 min 1 cycle; 94°C for 1.5 min/50°C for 1 min/68°C for 4 min 30 cycles; 94°C for 10 min 1 cycle, employing 5 U Taq polymerase (Fermentas), 40 nM MnCl2 and 200 nM dNTP. The PCR product was digested with BamHI (Fermentas) and EcoRI (Fermentas). The mutant library of P450 BM3 was subsequently ligated into the pCWori vector and transformed into E. coli DH5 for high-level expression in microtiter plates (Wong et al., 2005). Two identical rounds of epPCR were performed; in the second round, five improved variants of the first round were used as template DNA for diversity generation.
Site-directed mutagenesis was employed to generate reductase-deficient variants of the P450 BM3 M5 mutant (P450 BMP F87A, P450 BMP M2 and P450 BMP M5), and two quadruple P450 BM3 mutants (P450 BM3 M3 and P450 BM3 M4). The quadruple mutants were generated by site-directed mutagenesis using the triple mutant, P450 BM3 M2, as template. P450 BMP M5 was generated by insertion of a stop codon in the gene, upstream of the SacI restriction site. In all cases, a modified Quick-Change method was carried out (Zheng et al., 2004).
For insertion of the stop codon, 35 ng of P450 BM3 M5 template DNA was employed and 17.5 pmol of each mutagenic primer were used: forward primer: 5'- TACAAACTACGAGCTCGATATTAAAGAAACTCA-3'; reverse primer: 5'-AGCTCGTAGTTTGTATGATCTTCAAAGTCAAAG-3', under the following PCR conditions: 94°C for 4 min 1 cycle; 94°C for 1.5 min/55°C for 1 min/68ºC for 17 min 20 cyles; 68°C for 20 min 1 cycle, employing 2.5 U Pfu polymerase (Fermentas) and 200 nM dNTP. P450 BMP F87A and P450 BMP M2 were prepared using the same procedure.
For inserting the V281G and D363H mutations, 50 ng of P450 BM3 M2 template DNA and 17.5 pmol of each mutagenic primer were used: forward primer: 5'-GCTGTATTTCTTAGGCAAAAATCCACATGTATTACA-3'; reverse primer: 5'-ATTTTTGCCTAAGAAATACAGCGCAAATGATAAAAG-3' (in case of V281G mutation) and forward primer: 5'-CTCAGCTTCACCGTCACAAAACAATTTGGGGA-3'; reverse primer: 5'-GTTTTGTGACGGTGAAGCTGAGGAATCAGAACCA-3' (in case of D363H mutation), under the following PCR conditions: 94°C for 4 min 1 cycle; 94°C for 1.5 min/55°C for 1 min/68°C for 17 min 20 cyles; 68°C for 20 min 1 cycle, employing 2.5 U Pfu polymerase (Fermentas) and 200 nM dNTP.
Subsequently, 40 U of DpnI (New England Biolabs) were supplemented to each PCR mix and incubated for 2–3 h at 37°C. The PCR products were purified using NucleoSpin Extract kit (Machery-Nagel). Subcloning of the mutated P450 BMP genes from pCYTEXP1 to pCWori and transformation in E. coli DH5 was carried out as described previously (Wong et al., 2005).
Cultivation and expression in 96-well plates
Colonies grown on LBamp agar plates were transferred, using toothpicks, into 96-deep-well microtiter plates (2.2 ml polypropylene plates; Brand GmbH, Wertheim, Germany), containing 150 µl of LB media supplemented with 15 µg ampicillin per well. After growing overnight in a microtiter plate shaker (Multitron II, Infors GmbH, Einsbach, Germany; 37°C, 900 rpm, 70% humidity), 5 µl culture volume per well was transferred with the System Duetz tool (Kühner, Birsfelden, Switzerland) into 2 ml deep-well plates containing 600 µl of enriched TBamp medium (isopropyl-β-D-thiogalactoside IPTG 100 µM; 
-aminolevulinic acid anhydride, ALA, 0.5 mM) and separately autoclaved trace element solution (0.6 µl in 600 µl enriched TB) consisting of 0.5 g CaCl2 · 2H2O, 0.18 g ZnSO4 · 7H2O, 0.10 g MnSO4 · H2O, 20.1 g Na-EDTA, 16.7 g FeCl3 · 6H2O, 0.16 g CuSO4 · 5H2O, 0.18 g CoCl2 · 6H2O dissolved in 1 l of dH2O). Clones were cultivated in a Multitron II shaker (Infors GmbH; at 30°C, 500 rpm, 70% humidity, 24 h) and harvested after centrifugation (Eppendorf Centrifuge 5810R, Hamburg, Germany; 4°C, 3220 g, 15 min) by discarding the supernatant.
Zn/Co(III)sep assay in 96-well plate format
The harvested cells in the deep-well plates were resuspended in 230 µl of a buffer mixture (Tris–HCl 50 mM, KH2PO4/K2HPO4 50 mM, KCl 0.25 M, pH 8.0), permeabilized by supplementing Polymyxin B sulfate (59.4 µM; final concentration) and 12-pNCA substrate dissolved in DMSO (247.5 µM; final concentration). After 5 min of incubation, 600 U of catalase was added into every well. Twenty milligrams of zinc dust were subsequently supplemented per well using a solid dispenser (Resin Dispenser, Mettler-Toledo Bohdan Inc., Vernon Hills, USA). Reactions were finally initiated by Co(III)sep addition (396.0 µM; final concentration) in a total reaction volume of 253 µl. The reaction mixture was incubated for 35 min at 1400 rpm (Miniature Shaker KM2, Edmund Bühler, Tübingen, Germany) and enzymatic conversions were stopped by the addition of 50 µl NaOH (1 M). Deep-well plates were centrifuged (Eppendorf 5810R; 4°C, 3220 g for 10 min) and 200 µl of the clarified supernatants were transferred using an automated liquid-handling machine (MultimekTM 96 Beckman, Krefeld, Germany) into 96-well filtration plates (MultiScreen-FC glass filter plates, Millipore, Eschborn, Germany). Filtration using a vacuum manifold (MultiScreen Vacuum Manifold, Millipore) resulted in a clear filtrate. Owing to inhomogeneous filtrate volumes, 100 µl of filtrate was transferred from recipient plate (96-well, polystyrene, flat-bottom microtiter plate; Greiner Bio-One GmbH, Frickenhausen, Germany) into an identical 96-well flat-bottom plate. The conversion of 12-pNCA substrate was measured as described previously (Schwaneberg et al., 1999a) at 410 nm using a microtiter plate reader (FlashSCAN S12, Analytik Jena AG, Jena, Germany).
Expression and purification of P450 BM3 mutants
Shaking flasks (1 l) containing 250 ml of TBamp media, supplemented with 250 µl trace element solution, were inoculated with a 1:100 dilution of an LBamp-overnight culture (E. coli DH5 harboring pCWori). After reaching an OD578 value of 0.8–1 during cell cultivation (35°C, 250 rpm, Multitron II), ALA (0.5 mM, final concentration) was added and expression was induced by supplementing IPTG (100 µM; final concentration). E. coli cells were harvested, after 20 h of expression by centrifugation (Eppendorf 5810 R 4°C, 3220 g, 15 min) and resuspended in phosphate buffer (25 ml; KH2PO4/K2HPO4, 25 mM, pH 7.5). E. coli cells were subsequently lysed using a high-pressure homogenizer (1500 bar, two cycles; Avestin Emulsiflex, Mannheim, Germany). The lysate was centrifuged (Eppendorf 5417R, 4°C, 16 000 g, 15 min) and further clarified by filtration through a low protein binding filter (0.45 µm; Celtron 30/0 Syringe-driven filter unit; Schleicher & Schuell, Dassel, Germany). All kinetically characterized P450 BM3 mutants were purified by anion exchange chromatography as described previously (Schwaneberg et al., 1999b). Monooxygenase concentrations were determined by CO-difference spectra as reported by Omura and Sato using 
= 91 mM–1 cm–1 (Omura and Sato, 1964).
Determination of Km and kcat of P450 BM3 mutants
NADPH consumption assays and the Zn/Co(III)sep activity assays were performed as described previously (Schwaneberg et al., 1999a, 2000) except that the buffer was changed in the NADPH consumption assays (Tris–HCl 50 mM, KH2PO4/K2HPO4 50 mM, KCl 0.25 M, pH 8.0) and an Eppendorf Thermomixer comfort (Eppendorf AG, Hamburg, Germany) was used for incubating 1.5 ml Eppendorf tubes at 1200 rpm, in the Zn/Co(III)sep activity assay. The amount of P450 BM3 variants used in each experiment was 0.17 nmol and 12-pNCA substrate concentrations varied between 0.6 µM and 114 µM. Kinetic constants were calculated by fitting the experimental results into a hyperbolic function using the Origin 7.0 software and by Eadie-Hofstee plots.
Determination of TTNs of P450 BM3 mutants
Total turnover numbers were determined for 0.1 nmol of P450 BM3 variants with 75 µM 12-pNCA in a total volume of 1 ml. The conversions were preformed in 1.5 ml Eppendorf tubes as described previously (Schwaneberg et al., 2000), except that the reactions were continued for 13 h. Product generation and TTNs were calculated using the following calibration curve: y (µM) = 0.00848x + 0.03819 (r2 = 0.9982), which was obtained by measuring the absorbance at 410 nm of p-nitrophenole concentrations ranging from 0.03 to 0.3 mM.
|
Funding |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingReferences |
|---|
Deutsche Forschungsgemeinschaft (SPP1170 Gelenkte Evolution; award SCHW-790/2-1).
|
Footnotes |
|---|
Edited by Eva Petersen
|
References |
|---|
|
TopAbstractIntroductionResultsDiscussionMaterials and methodsFundingReferences |
|---|
Adlercreutz P. Biocatal. Biotransformation (1996) 14:1–30.[CrossRef]
Appel D., Lutz-Wahl S., Fischer P., Schwaneberg U., Schmid R.D. J. Biotechnol. (2001) 88:167–171.[CrossRef][Web of Science][Medline]
Barnes H.J., Arlotto M.P., Waterman M.R. Proc. Natl Acad. Sci. USA (1991) 88:5597–5601.
Belev T.N., Singh M., McCarthy J.E.G. Plasmid (1991) 26:147–150.[CrossRef][Web of Science][Medline]
Bistolas N., Christenson A., Ruzgas T., Jung C., Scheller F.W., Wollenberger U. Biochem. Biophys. Res. Commun. (2004) 314:810–816.[CrossRef][Web of Science][Medline]
Cirino P.C., Arnold F.H. Angew. Chem. Int. Edit. (2003) 42:3299–3301.[CrossRef]
Daff S.N., Chapman S.K., Turner K.L., Holt R.A., Govindaraj S., Poulos T.L., Munro A.W. Biochemistry (1997) 36:13816–13823.[CrossRef][Web of Science][Medline]
Drauz K., Eils S., Schwarm M. Chimica Oggi (2002) 20:15–21.
Estabrook R.W., Shet M.S., Faulkner K., Fisher C.W. Endocr. Res. (1996) 22:665–671.[Web of Science][Medline]
Fantuzzi A., Fairhead M., Gilardi G. J. Am. Chem. Soc. (2004) 126:5040–5041.[CrossRef][Web of Science][Medline]
Gilardi G., Meharenna Y.T., Tsotsou G.E., Sadeghi S.J., Fairhead M., Giannini S. Biosens. Bioelectron. (2002) 17:133–145.[CrossRef][Web of Science][Medline]
Hollmann F., Witholt B., Schmid A. J. Mol. Catal. B Enzym. (2002) 19:167–176.[CrossRef]
Joo H., Arisawa A., Lin Z.L., Arnold F.H. Chemis. Biol. (1999) a 6:699–706.[CrossRef]
Joo H., Lin Z.L., Arnold F.H. Nature (1999) b 399:670–673.[CrossRef][Medline]
Li H.Y., Poulos T.L. Nat. Struct. Biol. (1997) 4:140–146.[CrossRef][Web of Science][Medline]
Lvov Y.M., Lu Z.Q., Schenkman J.B., Rusling J.F. Abstr. Am. Chem. Soc. (1998) 215:U486–U486.
Matsumura H., Wiwatchaiwong S., Nakamura N., Yohda M., Ohno H. Electrochem. Commun. (2006) 8:1245–1249.[CrossRef]
Mayhew M.P., Reipa V., Holden M.J., Vilker V.L. Biotechnol. Progress (2000) 16:610–616.[CrossRef][Medline]
Munge B., Estavillo C., Schenkman J.B., Rusling J.F. Chembiochem (2003) 4:82–89.[CrossRef][Web of Science][Medline]
Murataliev M.B., Klein M., Fulco A., Feyereisen R. Biochemistry (1997) 36:8401–8412.[CrossRef][Web of Science][Medline]
Narhi L.O., Fulco A.J. J. Biol. Chem. (1987) 262:6683–6690.
Nazor J., Schwaneberg U. Chembiochem (2006) 7:638–644.[CrossRef][Web of Science][Medline]
Omura T., Sato R. J. Biol. Chem. (1964) 239:2379–2385.
Ravichandran K.G., Boddupalli S.S., Hasemann C.A., Peterson J.A., Deisenhofer J. Science (1993) 261:731–736.
Reipa V., Mayhew M.P., Holden M.J., Vilker V.L. Chem. Commun. (2002) 318–319.
Reipa V., Mayhew M.P., Vilker V.L. Proc. Natl Acad. Sci. USA (1997) 94:13554–13558.
Ruettinger R.T., Wen L.P., Fulco A.J. J. Biol. Chem. (1989) 264:10987–10995.
Schwaneberg U., Appel D., Schmitt J., Schmid R.D. J. Biotechnol. (2000) 84:249–257.[CrossRef][Web of Science][Medline]
Schwaneberg U., Otey C., Cirino P.C., Farinas E., Arnold F.H. J. Biomol. Screening (2001) 6:111–117.
Schwaneberg U., Schmidt-Dannert C., Schmitt J., Schmid R.D. Anal. Biochem. (1999) a 269:359–366.[CrossRef][Web of Science][Medline]
Schwaneberg U., Sprauer A., Schmidt-Dannert C., Schmid R.D. J. Chromatogr. A (1999) b 848:149–159.[CrossRef][Web of Science][Medline]
Sevrioukova I.F., Li H.Y., Zhang H., Peterson J.A., Poulos T.L. Proc. Natl Acad. Sci. USA (1999) 96:1863–1868.
Shukla A., Gillam E.M., Mitchell D.J., Bernhardt P.V. Electrochem. Commun. (2005) 7:437–442.[CrossRef]
Shumyantseva V.V., Bulko T.V., Rudakov Y.O., Kuznetsova G.P., Samenkova N.F., Lisitsa A.V., Karuzina I.T., Archakov A.I. J. Inorg. Biochem. (2007) 101:859–865.[CrossRef][Web of Science][Medline]
Shumyantseva V.V., Bulko T.V., Schmid R.D., Archakov A.I. Biofizika (2000) 45:1013–1018.[Medline]
Shumyantseva V.V., Ivanov Y.D., Bistolas N., Scheller F.W., Archakov A.I., Wollenberger U. Anal. Chem. (2004) 76:6046–6052.[Medline]
Stampfer W., Kosjek B., Moitzi C., Kroutil W., Faber K. Angew. Chem. Int. Ed. (2002) 41:1014–1017.[CrossRef]
Udit A.K., Hill M.G., Gray H.B. Langmuir (2006) 22:10854–10857.[CrossRef][Web of Science][Medline]
Donk van der W.A., Zhao H.M. Curr. Opin. Biotechnol. (2003) 14:421–426.[CrossRef][Web of Science][Medline]
Wolberg M., Hummel W., Wandrey C., Muller M. Angew. Chem. Int. Ed. (2000) 39:4306–4308.[CrossRef]
Wong T.S., Arnold F.H., Schwaneberg U. Biotechnol. Bioeng. (2004) 85:351–358.[CrossRef][Web of Science][Medline]
Wong T.S., Schwaneberg U. Curr. Opin. Biotechnol. (2003) 14:590–596.[CrossRef][Web of Science][Medline]
Wong T.S., Wu N., Roccatano D., Zacharias M., Schwaneberg U. J. Biomol. Screening (2005) 10:246–252.
Zheng L., Baumann U., Reymond J.L. Nucleic Acids Res. (2004) 32.
Zhu Z.W., Momeu C., Zakhartsev M., Schwaneberg U. Biosens. Bioelectron. (2006) 21:2046–2051.[Web of Science][Medline]
Zimmer T., Kaminski K., Scheller U., Vogel F., Schunck W.H. DNA Cell Biol. (1995) 14:619–628.[Web of Science][Medline]
Received August 26, 2007; revised October 30, 2007; accepted November 15, 2007.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||