M. T. Reetz and D. J. Opperman
We conclude that this study constitutes a step forward in
the engineering of practical Baeyer–Villiger monooxygenases
Table 2. Regio- and enantioselectivities of the oxidatively and thermosta-
ble mutants of CHMO. Primers used for site-directed and saturation muta-
genesis.
(BVMOs) for application in synthetic organic chemistry and bio-
technology. A pronounced increase in the oxidative stability of
the most commonly applied BVMO—cyclohexanone monooxy-
genase from Acinetobacter sp. NCIMB 9871 (CHMO)—has been
achieved through rational and combinatorial design, without
compromising activity, substrate scope, or stereoselectivity.
The results set the stage for further thermostabilization by
random or focused mutagenesis. User-friendly processes might
then emerge based on the in vitro use of isolated CHMO
mutants in combination with a practical NADPH-regeneration
system utilizing a robust secondary alcohol dehydrogenase
and isopropanol as the reductant.
[
a]
CHMO mutation
Primer (5’!3’)
M256L
M280L
M373V
M390I
M400L
M412L
M422L
M481A
C475V
C520V
M5I
M319V
M291L
C376L
C64V
M424L
C330L
C330NNK
C330S
CACAGTGCCAGCATTGAGCGTATCAG
CGGTTTCCGTTTCTTGTTTGAAACTTTCG
CGTTGAATTAGACGTGCTGATATGTGC
GCAACTATGTGCGCATAGACATTCAAGG
GCTTGGCCTTGAAAGACTACTGGAAAG
GAAGGTCCGTCGAGCTATTTGGGTGTC
CGTAAATAACTATCCAAACTTGTTCATGGTGC
CCAATATTGCGGAAGCGACCTTATTCC
GGACTCAAACTGTGGCCAATATTGCG
GTGCGCTAGCCAACGTCAAAAACCATG
GTCACAAAAACTGGATTTTGATGCTATCG
CCATTGCACAGAAGCTTGTGCCACAG
GGTGATATTGCCACCAATTTGGAAGC
CGTTGAATTAGACATGCTGATATTAGCCACAG
CAGAAACCCACCTCTACGTCTATTCTTGG
TGTTCTTGGTGCTTGGACCGAATGG
Experimental Section
Materials: All substrates were obtained from commercial sources
CAAAACGTCCGTTGTTAGACAGTGGTTACTAC
CAAAACGTCCGTTGNNKGACAGTGGTTACTAC
CAAAACGTCCGTTGTCTGACAGTGGTTACTAC
(
Sigma–Aldrich, Acros, and Fluka) and were used without any fur-
ther purification. NADPH was purchased from Sigma–Aldrich and
hydrogen peroxide from Merck.
[
a] Only forward primers are listed, with the reverse primers being the re-
Strains, vectors, and growth conditions: The CHMO wild type
and mutants were expressed from the pET22b(+) vector (Nova-
gen), which also served as template for site-directed mutagenesis.
E. coli strains were routinely grown in Luria–Bertani (LB) medium at
verse compliments of the tabulated primers.
(
1
800 rpm). Cells were harvested through centrifugation (2700g,
5 min) and pellets were resuspended in lysis buffer [350 mL, Tris-
3
78C with shaking (200 rpm). Expression of the CHMO wild type
ꢀ1
HCl (pH 8, 50 mm), lysozyme (1 gL ), EDTA-free Protease Inhibitor
and mutants was done in E. coli BL21-Gold (DE3) (Stratagene) in
(
Roche) and DNAse (1 h)]. Cells were broken with a single freeze-
thaw cycle, after which the cellular debris was removed through
[24]
ZYP5052 medium at 258C (200 rpm).
centrifugation (2700 g, 30 min). Supernatant was transferred either
Site-directed mutagenesis: Site-directed mutagenesis was per-
formed by the QuickChange (Stratagene) PCR method. PCR reac-
tion mixtures (50 mL) consisted of 10ꢁ KOD Hot Start Polymerase
buffer (5 mL), MgSO4 (1.5 mm), deoxynucleoside triphosphates
(
(
80 mL) to 96-well PCR plates (for thermal stability studies) or
50 mL) to 1.8 mL deep-well plates containing Tris-HCl buffer (pH 8,
2
50 mL) and various concentrations of hydrogen peroxide. Thermal
stability was assessed through incubation of the crude soluble frac-
tions at 378C for 1 h, and oxidative stability was determined by in-
cubating the crude extract with hydrogen peroxide (10–100 mm)
overnight at 48C. Residual activity was measured by determining
the percentage conversion of cyclohexanone (2 mm) at room tem-
perature after 15 min. Reaction mixtures (400 mL) contained equi-
molar concentrations of NADPH and were stopped/extracted
through the addition of equal volumes of ethyl acetate. The reac-
tion mixtures were centrifuged and the organic phases were trans-
ferred to 96-well glass plates for GC analysis.
(
(
0.2 mm each), KOD Hot Start DNA polymerase (1 U), plasmid DNA
20 ng), and both the forward and the reverse primers (Table 2,
0
.1 mm). The reaction conditions consisted of an initial denaturing
step at 958C for 2 min, followed by 18 cycles of denaturing at
58C (30 s), annealing at 618C (30 s), and elongation at 708C
4 min), with a final extension at 708C for 5 min. PCR products
were digested with DpnI (New England Biolabs, 0.4 U) at 378C for
h, to ensure removal of the template plasmid DNA. DpnI-digest-
9
(
3
ed products were purified with the aid of the QIAquick PCR purifi-
cation kit (Qiagen) and subsequently transformed into E. coli DH5a.
Random colonies were selected and grown overnight in LB
medium (5 mL) containing carbenicillin (100 mgmL ). Plasmid ex-
tractions were performed by use of the QIAprep Spin Miniprep kit
Biotransformations: The CHMO wild type and the mutants MUT15
and MUT16 were expressed as described above. Crude soluble
fractions, obtained after cell lysis and centrifugation, were incubat-
ed overnight at 48C with excess FAD. Samples were desalted and
unbound FAD was removed through passage of the extracts
through a PD-10 desalting column (GE Healthcare). Activity was de-
termined as the percentage conversion after 2 h by crude extract
ꢀ
1
(
Qiagen) and mutagenesis was confirmed through sequencing of
the CHMO gene (Eurofins MWG Operon, Germany). C330 site-di-
rected saturation mutagenesis was performed as described above,
with 96 random colonies inoculated into deep-well plate format.
Library quality was assessed through sequencing of a pooled li-
brary consisting of more than 100 colonies.
(50 mL) in a reaction volume of 500 mL containing substrate (2 mm)
and NADPH (3 mm) at room temperature [Tris-HCl (pH 8, 50 mm)].
Reaction mixtures were again extracted with equal amounts of
ethyl acetate. The organic phases were transferred to GC vials for
analysis. The selectivities of CHMO and its MUT15 and MUT16 mu-
tants were compared by determination of enantiomeric excesses
either after complete conversion of the 4-substituted cyclohexa-
nones or after approximately 30% conversion of the 2-substituted
cyclohexanones. These reaction mixtures were adjusted by increas-
ing the substrate concentrations to 4 mm and limiting the reaction
time and the NADPH concentration (to 1.5 mm).
Mutant screening: Initial analyses and screening of the mutants
were performed in 96-well plate format. CHMO mutants were
transformed into E. coli BL21-Gold (DE3) and plated on LB plates
ꢀ1
containing carbenicillin (100 mgmL ). Single colonies were inocu-
lated into deep-well plates (2.2 mL) containing LB medium (0.8 mL)
ꢀ1
and carbenicillin (100 mgmL ). Deep-well plates containing ZYP-
052 autoinduction medium (0.8 mL) were inoculated with over-
5
night-grown cultures of the CHMO mutants. Expression of the
recombinant proteins was allowed to proceed for 24 h at 258C
2
594
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemBioChem 2010, 11, 2589 – 2596