Baeyer-Villiger Monooxygenase-Mediated Synthesis of Esomeprazole As an Alternative for Kagan Sulfoxidation

Article abstract of DOI:10.1021/acs.joc.8b00468

A wild-type Baeyer-Villiger monooxygenase was engineered to overcome numerous liabilities in order to mediate a commercial oxidation of pyrmetazole to esomeprazole, using air as the terminal oxidant in an almost exclusively aqueous reaction matrix. The developed enzyme and process compares favorably to the incumbent Kagan inspired chemocatalytic oxidation, as esomeprazole was isolated in 87% yield, in >99% purity, with an enantiomeric excess of >99%.

Full text of DOI:10.1021/acs.joc.8b00468

Article
pubs.acs.org/joc
Cite This: J. Org. Chem. XXXX, XXX, XXX−XXX
Baeyer−Villiger Monooxygenase-Mediated Synthesis of
Esomeprazole As an Alternative for Kagan Sulfoxidation
Yong Koy Bong,† Shiwei Song,† Jovana Nazor,†,‡ Michael Vogel,† Magnus Widegren,† Derek Smith,†
Steven J. Collier,‡ Rob Wilson,†,‡ S. M. Palanivel,† Karthik Narayanaswamy,† Ben Mijts,‡
Michael D. Clay,‡ Ryan Fong,‡ Jeff Colbeck,‡ Amritha Appaswami,‡ Sheela Muley,‡ Jun Zhu,‡
Xiyun Zhang,‡ Jack Liang,‡ and David Entwistle*,†,‡
†Codexis Laboratories Singapore Pte Ltd., The Galen, 61 Science Park Road, Singapore 117525
‡Codexis, Inc., 200 Penobscot Drive, Redwood City, California 94070, United States
S
* Supporting Information
ABSTRACT: A wild-type Baeyer−Villiger monooxygenase was engineered to overcome numerous liabilities in order to
mediate a commercial oxidation of pyrmetazole to esomeprazole, using air as the terminal oxidant in an almost exclusively
aqueous reaction matrix. The developed enzyme and process compares favorably to the incumbent Kagan inspired
chemocatalytic oxidation, as esomeprazole was isolated in 87% yield, in >99% purity, with an enantiomeric excess of >99%.
control, and chemo-selectivity.7 However, several aspects of
INTRODUCTION
■
oxidative biocatalysis need to be addressed for commercial
Chiral sulfoxides are gaining interest as design elements in drug
utilization. First of all, wild-type enzymes are typically less
discovery as well as chiral auxiliaries in chemical synthesis.1
stable under the desired reaction conditions that do not
The “prazole” proton pump inhibitors for the treatment of
necessarily reflect the native conditions the enzymes originally
gastric reflux are possibly the most significant examples of
evolved under. Besides that, enzymes identified in the nature,
sulfoxide-containing active pharmaceutical ingredients (API).
generally, are limited in substrate range and do not possess
Esomeprazole, the S-enantiomer of omeprazole, is the API of
sufficient activity or selectivity to the substrate of interest.
the blockbuster drug Nexium.
Furthermore, the soluble expression of enzymes, especially
Esomeprazole is produced commercially using variations of
oxidative enzymes, through heterologous hosts can be
the Kagan−Sharpless−Pitchen sulfoxidation2 of the parent
challenging. Nevertheless, with CodeEvolver directed evolu-
sulfide, pyrmetazole (Scheme 1).3 However, the enantiose-
tion technology, many of the above-mentioned concerns have
lectivity of the titanium-catalyzed asymmetric oxidation is
been addressed.8 For esomeprazole, a whole cell biocatalytic
sensitive to the amount of water present.4 Over oxidation of
reaction has been reported, but in low yield.9 Therefore, a
the sulfoxide to sulfone also poses a problem as the
process using specifically engineered isolated sulfoxdiation
downstream removal of sulfone is challenging.5 In addition,
enzyme as the oxidation catalyst and a suitable enzyme for
waste treatment, process safety, and environmental impact are
cofactor recycling is attractive.
also important aspects to consider for the reaction.
Controlled asymmetric oxidations are uncommon in tradi-
tional syntheses, and the development of safe scalable
oxidations in general has been identified as a key requirement
Cyclohexanone monooxygenase (CHMO) was first identi-
fied in Acinetobacter sp. strain NCIB 9871.10 This flavoprotein
monoxygenenase, also referred to as a Baeyer−Villiger
monoxygenase (BVMO), catalyzes the Baeyer−Villiger
oxidation of cyclohexanone to the ring expanded lactone.11
if more oxidations are to be used in the synthesis of active
pharmaceutical ingredients.6 Asymmetric oxidations that
BVMOs are known to catalyze other oxidative reactions such
proceed near room temperature in water using molecular
oxygen as the oxidant are particularly desirable. A biocatalytic
Special Issue: Organic and Biocompatible Transformations in
approach can be considered for asymmetric sulfoxidation
Aqueous Media
which would combine these highly desirable attributes with
other potential benefits in terms of milder processing
Received: February 26, 2018
conditions, minimal organic solvent waste, better chirality
© XXXX American Chemical Society
A
DOI: 10.1021/acs.joc.8b00468
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 1. Kagan−Sharpless−Pitchen-Based Oxidative Formation of Esomeprazole
Scheme 2. BVMO-FAD Catalytic Cycle and NADPH Recycle
Table 1. Key BVMO and Reaction Performance Parameters over the Course of BVMO Evolution
a
fold improvement in productivity
ee (S)
sulfone (%)
NADP+ (g/L)
BVMO WT
round 2 hit
round 3 hit
round 7 hit
round 13 hit
round 19 hit
1
800
610
Nd
−95%
92%
96%
98%
Nd
Nd
1
0.5
0.5
0.1
0 (an equivalent NADPH was used)
2
2
0.5
0.5
0.1
∼6600
∼85000
∼140000
>99%
a
Fold improvement was calculated by using the ratio of productivity (gP/(gE·hr) of the evolved variants with respect to wild-type.
as hydroxylation, epoxidation, amine oxidation, halogenations,
and, of specific interest in this context, sulfoxidation.12 BVMO
requires NADPH to reduce the oxidized flavin adenosine
nucleotide (FAD) cofactor generated per catalytic cycle
(Scheme 2). The cost of NADPH prohibits stoichiometric
use, and therefore a recycling system is used. Numerous
recycling systems are available, but for this reaction a
ketoreductase (KRED)-mediated propan-2-ol (IPA) oxidation
was selected to efficiently enable NADPH cofactor regener-
ation.
Through Codexis’ CodeEvolver directed evolution technol-
ogy, a BVMO was improved to allow a synthesis of
esomeprazole in high chemical and enantiomeric purity with
a catalytic efficiency suitable for use in commercial production.
The variants produced in the directed evolution program were
also found to be useful for enantioselective sulfoxidaton of
other prazole precursor sulfides.13
RESULTS AND DISCUSSION
■
Directed Evolution of BVMO. The wild-type BVMO
from Acinetobacter (Uniprot accession number P12015)
showed only a trace detectable activity on pyrmetazole, the
sulfide precursor to esomeprazole. With such barely detectable
activity using LC-MS, the BVMO enzyme was evolved to
improve four critical factors namely, the enzyme productivity
(gram of product formed per gram enzyme per hour), the
enantioselectivity, the chemoselectivity (sulfoxidation versus
sulfone formation), and cofactor loading. In this approach,
libraries of thousands of monooxygenase variants incorporating
programmed mutations were tested for activity, stereo-
selectivity, and chemoselectivity for the desired target in
rounds of evolution. Variants were identified with improved
function, and algorithms used to assign predicted coefficients
of the contributions of individual mutations to the overall
fitness of the catalyst. Beneficial mutations were recombined in
iterative rounds of directed evolution until biocatalysts were
produced that met the targeted catalytic performance. In this
work, tens of thousands of variants were produced and tested,
B
DOI: 10.1021/acs.joc.8b00468
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Scheme 3. Overall Reaction Scheme
and the final biocatalyst was improved over 140,000-fold over
the initial wild-type enzyme.
catalysis, but as yet no defining structural motifs have been
identified.
Biocatalytic Process Development. Despite having a
productive enzyme, the process development for esomeprazole
synthesis was challenging. As demonstrated in Scheme 3, the
reaction is a three-phase three-enzyme system with numerous
potential side reactions, and the product isolation is further
complicated by the presence of esomeprazole polymorphs.15
Molecular oxygen is the oxygen donor in the reaction.
However, the aqueous solubility of oxygen is low (∼8 mg/L,
∼0.2 mM at 25 °C).16 The mass transfer rate of oxygen
between the gaseous and aqueous phases was frequently found
to be reaction rate determining during process development.
The use of oxygen blanket was found to deliver superior
reaction rate as compared to air. On the other hand, the low
aqueous solubility of substrate pyrmetazole (<1 g/L/) in the
reaction matrix creates another phase transfer process in the
reaction. The low concentration of the reactants, oxygen and
pyrmetazole, suggests that the enzyme possesses a very high
catalytic efficiency, enabling the reaction to proceed at a
practically useful rate.
A third phase transfer process is related to the poorly soluble
product esomeprazole in the reaction system. At pH 9, the
esomeprazole is mainly in the free acid form. The
esomeprazole continuously precipitates out from the reaction
solution as the reaction progresses, giving rise to a “slurry to
slurry” process. Numerous esomeprazole polymorphs have
been reported. In this reaction, the presence of different
polymorphs was suspected to result in a viscous reaction
medium. The viscosity posed a challenge to the reactor design,
as efficient mixing is crucial to ensure the reaction is not
limited by oxygen mass transfer rate. Additionally, the solid
particles in the reaction medium tended to clump together at
high solid loadings but in an unpredictable manner, and
product polymorphism was suspected to be the culprit. A
seeding strategy was implemented that controlled the variation
in reaction slurry viscosity.
Stability of the products and safety of the reaction are key
concerns. Esomeprazole is unstable at neutral or low pH which
sets a window of operation for the reaction pH and
temperature.17 The reaction is kept above pH 8 to ensure
product stability, and the process temperature is kept at 25 °C.
The reaction is also protected from excessive ambient light as
light exposure is known to accelerate the rate of esomeprazole
impurity formation.18 One of the significant side reactions is
the overoxidation of the product to give omeprazole sulfone.
The esomeprazole USP specification for sulfone impurity is
stringent (<0.2%),19 but the sulfone is difficult to remove. As a
result, it is desirable to minimize the sulfone formation at
source in the oxidative reaction. The biocatalytic chemo-
selectivity can be effectively controlled though directed enzyme
Table 1 indicates the improvements made against the four
key measures at various points through the engineering
program. Various evolution library design strategies were
employed, ranging from random mutation, recombination of
diversity from related homologous enzyme, to protein-model
guided design. As evolution progressed through successive
rounds, the high-throughput assay was adjusted to apply the
appropriate evolutionary pressure on the mutants, such as the
substrate loading, cofactor loading, %IPA cosolvency, as well as
reaction time, pH, and temperature.
As demonstrated in Table 1, the activity toward pyrmetazole
was significantly improved by round 2. However, the variant
showed undesirable R-enantioselectivity. The enantioselectivity
was then immediately targeted, reversed in round 3. The
enantioselectivity was subsequently improved to 96.5% ee (S),
in addition a further 8-fold increase in productivity in round 7.
With continued evolution, the productivity was further
enhanced, but the enantioselectivity was less than ideal with
significant overoxidation to the sulfone. This level of sulfone
was highly undesirable as the downstream removal of sulfone
was challenging. In addition, a cost analysis suggested a
significant positive impact could be made with a reduction in
NADP cofactor used. As a result, the subsequent evolution
effort was aimed at the fine-tuning of enantioselectivity,
suppression of overoxidation, and reduction in cofactor
loading. The best round 19 hit was approximately 140,000-
fold improved in productivity, over the starting wild-type
enzyme. A review of the mutations introduced to the final
variant, unsurprisingly, showed a significant number in the
substrate- and cofactor-binding regions. These mutations are
believed to be influential in enantioselectivity improvement,
sulfone suppression, cofactor loading reduction, as well as
improvement in enzyme productivity. Interestingly, most of
the mutations in the substrate-binding region were guided by
the protein homology model (e.g., identifying regions for
saturation mutagenesis through examination of protein
model), while random and homology-based mutations
accounted for most of the mutations in the nonsubstrate
binding region.
During the course of evolution, the BVMO’s native
capability of catalyzing Baeyer−Villiger oxidation was
diminished, signaling a fundamental change in the underlying
catalytic mechanism. The catalysis for Baeyer−Villiger
oxidation is likely to proceed through nucleophilic attack of
the substrate by the FAD-peroxide intermediate species, while
the sulfoxidation is likely to involve electrophilic attack at sulfur
by hydroperoxide intermediate.14 The mutations around the
active site have apparently changed the dominant mode of
C
DOI: 10.1021/acs.joc.8b00468
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
evolution, although a tendency for more sulfone formation at
high pH set a new boundary for the reaction pH. The
uncoupled reaction, where inefficient binding of substrate in
the active site results in enzyme turnover with release of
hydrogen peroxide, is a known side reaction for BVMO.20
Hydrogen peroxide is highly detrimental to the reaction as it
oxidatively inactivates enzymes and can also oxidize
pyrmetazole in an uncatalyzed fashion to the achiral sulfoxide
and further to sulfone (Scheme 4).
process. At the same time, the overall enzyme loading needed
to be kept to the minimum to reduce the overall process cost.
A search through Codexis’ in-house KRED and catalase
collections identified the suitable enzymes for the process.
At the end of the enzyme and process developmental efforts,
the optimal biocatalytic process involved the use of 50 g/L
pyrmetazole, 1 g/L BVMO variant, 0.5 g/L KRED CDX-019,
0.1 g/L NADP+, 0.2% v/v catalase CAT-101, 2.5 g/L
esomeprazole seed, and 4% v/v IPA, in 50 mM phosphate
buffer pH 9.0, under oxygen blanket, at 25 °C overnight.
Downstream reaction workup through extraction into isobutyl
methyl ketone (MiBK) and filtration afforded esomeprazole in
87% isolated yield, 99% purity (by HPLC), 99.9% ee, and
<0.1% sulfone. As compared to Kagan−Sharpless−Pitchen
oxidation process, this biocatalytic process serves as an
alternative practical process with good productivity at
attractive cost. The reaction is mainly aqueous based, with
only minimal organic solvent, IPA. The downstream process
workup is high yielding due to the high enantiopurity of the
product with minimal sulfone impurity.
Scheme 4. Uncoupled Reaction, Formation of Hydrogen
Peroxide
In order to minimize decoupling, a BVMO with high
binding affinity for pyrmetazole is needed, which was
addressed through directed evolution. Additionally, operating
the reaction where substrate mass transfer from solid to
solution phase is not rate limiting and will also minimize the
effects of decoupling. Lastly, the detrimental oxidative harm by
hydrogen peroxide is removed by introduction of catalase,
which degrades the hydrogen peroxide into water and oxygen.
In terms of safety, the use of volatile and flammable cosolvents
in the process should be kept to a minimum and at a
concentration such that the flash point is well in excess of the
desired reaction temperature. In the final process, 4% IPA (3.4
mol equiv) was used in tandem with an additional
ketoreductase (KRED) to effect necessary regeneration of
the expensive NADPH cofactor. The flash point of 5%v/v IPA
in water is reported to be 50 °C, which was considered a
suitable excess over the reaction temperature of 25 °C.21
NADPH is a significant cost contributor in the process.
Hence, a NADPH recycling system is needed to enable
practical synthesis of esomeprazole. Both glucose dehydrogen-
ase (GDH)/glucose-dependent and KRED/IPA-dependent
cofactor recycling systems were evaluated. Both systems have
pros and cons in various applications, but in this instance the
KRED/IPA system was selected as it offered a more atom
economical and simpler process without the need for constant
pH adjustment that the GDH system required due to the
formation of gluconic acid (Scheme 5).
Enantioselective Sulfoxidation of Other Sterically
Demanding Sulfides. The BVMO variants produced in the
evolution program described were found to be active on other
sterically demanding sulfides, such as the other proton pump
inhibitors, “prazole” sulfides and the modafinil sulfide.22 This
demonstrated the broader applicability of the evolved BVMO
as catalyst for enantioselective sulfoxidation. Although the
enantioselectivity and productivity might not be ready for
commercial manufacturing of those materials, further directed
evolution could be employed to improve the biocatalytic
process. This further exemplifies the usefulness of BVMO-
based biocatalytic approaches as a practical alternative to
Kagan−Sharpless−Pitchen oxidation.
CONCLUSION
■
Kagan−Sharpless−Pitchen oxidation remains the method of
choice for many chemists when considering enantioselecitve
sulfoxidation. However, this report offers a practical alternative
process using an evolved BVMO for enantioselective
sulfoxidation which gives high catalytic efficiency with much
superior chemo- and enantioselectivity. Through both directed
evolution and process development, high enantioselectivity is
demonstrated with minimal side products formation, on top of
good productivity, cost effectiveness, and environmental
sustainability. The enzyme variants produced in the course of
this evolution effort were also useful for the enantioselective
sulfoxidation of other bulky−bulky compounds, such as
armodafinil and rabeprazole.
The envisioned three enzyme (BVMO, KRED, and catalase)
process that also had several phase transitions required careful
optimization. The KRED and catalase loadings were selected,
so that these enzymes were not the limiting enzyme for the
For esomeprazole, specific physical challenges were encoun-
tered, such as mass transfer challenges, reaction medium
viscosity, and product instability. Within the boundaries set by
these physical limitations, directed evolution technology was
employed to improve the enzyme performance and character-
istics. The biocatalytic sulfoxidation process parameters can be
first adjusted to address the physical mass transfer limitations,
and then directed evolution technology can be used to adapt
the enzyme to the new process conditions, instead of having to
further restrict the reaction parameters to accommodate
enzyme limitations.
Scheme 5. Comparison of KRED and GDH-Mediated
NADPH Recycling Systems
EXPERIMENTAL SECTION
Laboratory Scale Sulfoxidation. Pyrmetazole (30 g, 0.091 mol,
ex Sinojie (HK) Ltd.) and esomeprazole (1.5 g) were charged to a
■
D
DOI: 10.1021/acs.joc.8b00468
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
jacketed reactor vessel, equipped with baffle and an anchor-shaped
agitator, containing potassium phosphate buffer (0.05 M, pH 9.0, 517
mL). This reaction mixture was agitated at 150 rpm at 25 °C for 10
min in order to obtain a well-suspended slurry, and the vessel was
filled with oxygen via three cycles of evacuation and oxygen refill. The
reaction vessel was held under positive oxygen pressure with an
oxygen filled balloon. Propan-2-ol (HPLC grade, 24 mL, 0.31 mol, 3.4
mol equiv), NADP solution (60 mg in 4 mL 0.05 M phosphate pH
9.0 buffer), ketoreductase CDX-019 solution (300 mg in 15 mL 0.05
M phosphate pH 9.0 buffer), BVMO round 19 hit solution (600 mg
in 40 mL 0.05 M phosphate pH 9.0 buffer), and 1.2 mL of catalase
(1.2 mL, from Sigma-Aldrich) were added to the mixture which was
stirred at 25 °C for 48 h. The initial stir rate of 300 rpm was gradually
increased stepwise to 450 rpm throughout the reaction. The course of
the reaction was followed by taking periodic samples from the
reaction mixture which were diluted 100-fold with MeOH and
analyzed using HPLC method 1.
ACKNOWLEDGMENTS
■
Codexis wishes to thank the many Codexis colleagues who
contributed to this work.
REFERENCES
■
(1) (a) Carreno, M. C. Applications of Sulfoxides to Asymmetric
Synthesis of Biologically Active Compounds. Chem. Rev. 1995, 95 (6),
1717−1760. (b) Kagan, H. B. Chapter 1. Asymmetric Synthesis of
Chiral Sulfoxides. In Organosulfur Chemistry in Asymmetric Synthesis;
Toru, T., Bolm, C., Eds.; Wiley-VCH Verlag GmbH & Co. KGaA:
Weinheim, Germany, 2008; p 129.
(2) Pitchen, P.; Kagan, H. An Efficient Asymmetric Oxidation of
Sulfides to Sulfoxides. Tetrahedron Lett. 1984, 25, 1049−1052.
(3) (a) Larsson, M. E.; Stenhede, U. J.; Sorensen, H.; Von Unge, S.
P. O.; Cotton, H. K. Process for Synthesis of Substituted Sulfoxides.
US patent 5948789, September 7, 1999. (b) Cotton, H.; Elebring, T.;
Upon reaction completion, methyl isobutyl ketone (165 mL) was
added to the vessel with stirring at 300 rpm, and the jacket
temperature was increased to 48 °C. After complete dissolution
(approximately 25 min), the stirring was stopped, and the phases
allowed to separate (approximately 20 min). The aqueous turbid layer
was drained and collected. The brown organic layer was subsequently
drained and submitted to a warm filtration over Celite applying
vacuum (the temperature of the jacket filter was adjusted to 45 °C).
The Celite cake was retained for later use, the aqueous phase was
transferred back to the heated vessel, and further methyl isobutyl
ketone (45 mL) was added and stirred at 300 rpm for 30 min.
Approximately 20 min after the agitation was stopped, the yellow
turbid aqueous layer was drained and discarded. The brown organic
layer was drained and filtered over the retained Celite filter cake used
previously. The organic phases when combined released a small
amount of water which was separated. The organic phase was
transferred back to the vessel, the temperature was adjusted to 15 °C,
and the agitator was set to 150 rpm for 1 h. The product precipitated
from the solution as a dense slurry. The temperature was further
reduced to 10 °C, and the slurry stirred at 150 rpm for 30 min. Finally
the temperature was adjusted to 5 °C, and the mixture stirred at 150
rpm for 30 min. In order to increase the mobility of the slurry, n-
heptane (240 mL) was added slowly in 30 mL portions (over 5 min).
The slurry was drained and filtered off, and the filter cake was dried in
a vacuum oven (3−10 mmHg) at 25 °C for 72 h. A total of 28.7 g of
esomeprazole was isolated (87% isolated yield) with a chemical purity
of 99% by HPLC (method 4, see Supporting Information) and >99%
ee (method 5, see Supporting Information).
̈
Larsson, M.; Li, L.; Sorensen, H.; von Unge, S. Asymmetric Synthesis
of Esomeprazole. Tetrahedron: Asymmetry 2000, 11, 3819−3825.
́
(4) Seenivasaperumal, M.; Federsel, H.-J.; Szabo, K. J. Mechanism of
the Asymmetric Sulfoxidation in the Esomeprazole Process: Effects of
the Imidazole Backbone for the Enantioselection. Adv. Synth. Catal.
2009, 351, 903−919 and the references therein.
(5) Ratios of sulfoxide to sulfone reported between 35:1 and 79:1 in
ref 3b.
(6) Constable, D. J. C.; Dunn, P. J.; Hayler, J. D.; Humphrey, G. R.;
Leazer, J. L., Jr.; Linderman, R. J.; Lorenz, K.; Manley, J.; Pearlman, B.
A.; Wells, A. S.; Zaks, A.; Zhang, T. Y. Key Green Chemistry Research
AreasA Perspective from Pharmaceutical Manufacturers. Green
Chem. 2007, 9, 411.
(7) Bornscheuer, U. T.; Huisman, G. W.; Kazlauskas, R. J.; Lutz, S.;
Moore, J. C.; Robins, K. Engineering the Third Wave of Biocatalysis.
Nature 2012, 485, 185−194.
(8) (a) Modukuru, N.; Sukumaran, J.; Collier, S. J.; Chan, S. A.;
Gohel, A.; Huisman, G. W.; Keledjian, R.; Narayanaswamy, K.;
Novick, S.; Palanivel, S.; Smith, D.; Wei, Z.; Wong, B.; Yeo, W. L.;
Entwistle, D. A. Development of a Practical, Biocatalytic Reduction
for the Manufacture of (S)-Licarbazepine Using an Evolved
Ketoreductase. Org. Process Res. Dev. 2014, 18 (6), 810−815. (b)
Nazor, J.; Osborne, R.; Liang, J.; Vroom, J.; Zhang, X.; Entwistle, D.;
Voladri, R.; Garcia, R. D.; Moore, J. C.; Grosser, S.; Kosek, B.;
Truppo, M. Biocatalysts and Methods for Hydroxylation of Chemical
Compounds. Patent application WO2017213758, December 14,
2017. (c) Savile, C. K.; Janey, J. M.; Mundorff, E. C.; Moore, J. C.;
Tam, S.; Jarvis, W. R.; Colbeck, J. C.; Krebber, A.; Fleitz, F. J.; Brands,
J.; Devine, P. N.; Huisman, G. W.; Hughes, G. J. Biocatalytic
Asymmetric Synthesis of Chiral Amines from Ketones Applied to
Sitagliptin Manufacture. Science 2010, 329, 305−309. (d) Ma, S. K.;
Gruber, J.; Davis, C.; Newman, L.; Gray, D.; Wang, A.; Grate, J.;
Huisman, G. W.; Sheldon, R. A. A Green-by-Design Biocatalytic
Process for Atorvastatin Intermediate. Green Chem. 2010, 12, 81−86.
(e) Liang, J.; Mundorff, E.; Voladri, R.; Jenne, S.; Gilson, L.; Conway,
A.; Krebber, A.; Wong, J.; Huisman, G. W.; Truesdell, S.; Lalonde, J.
Highly Enantioselective Reduction of a Small Heterocyclic Ketone:
Biocatalytic Reduction of Tetrahydrothiophene-3-one to the Corre-
sponding (R)-Alcohol. Org. Process Res. Dev. 2010, 14 (1), 188−192.
(9) Holt, R.; Lindberg, P.; Reeve, C.; Taylor, S. Preparation of
Pharmaceutically Active Compounds by Biooxidation. US patent
5840552, November 24, 1998.
1H NMR (400 MHz, MeOH-d4): δ 8.09 (s,1H), 7.51 (d, J = 8.8
Hz, 1H), 7.08 (d, J = 2.4 Hz, 1H), 6.96 (dd, J = 8.8, 2.4 Hz, 1H), 4.73
(m, 2H), 3.83 (s, 3H), 3.68 (s, 3H), 2.22 (s, 3H), 2.15 (s, 3H). 13C
NMR (100 MHz, MeOH-d4): δ 166.2 (s), 159.3 (s), 150.5 (s), 149.8
(s), 129.1 (s), 128.2 (s), 115.8 (s), 61.4 (s), 60.6 (s), 56.3 (s), 13.4
(s), 11.7 (s). LCMS (ESI+) m/z 346.4.
ASSOCIATED CONTENT
* Supporting Information
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.joc.8b00468.
Selected NMR data, analytical methods, HPLC data, and
HPLC/LCMS chromatograms (PDF)
(10) Donoghue, N. A.; Norris, D. B.; Trudgill, P. W. The
Purification and Properties of Cyclohexanone Oxygenase from
Nocardia globerula CL1 and Acinetobacter NCIB 9871. Eur. J.
Biochem. 1976, 63, 175−192.
AUTHOR INFORMATION
Corresponding Author
*E-mail: david.entwistle@codexis.com.
ORCID
David Entwistle: 0000-0001-6268-1399
Notes
The authors declare no competing financial interest.
■
(11) Mihovilovic, M. D.; Muller, B.; Stanetty, P. Monooxygenase-
̈
Mediated Baeyer-Villiger Oxidations. Eur. J. Org. Chem. 2002, 2002,
3711−3730.
(12) (a) Light, D. R.; Waxman, D. J.; Walsh, C. Studies on the
Chirality of Sulfoxidation Catalyzed by Bacterial Flavoenzyme
Cyclohexanone Monooxygenase and Hog Liver Flavin Adenine
E
DOI: 10.1021/acs.joc.8b00468
J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Dinucleotide Containing Monooxygenase. Biochemistry 1982, 21,
2490−2498. (b) Branchaud, B. P.; Walsh, C. T. Functional Group
Diversity in Enzymatic Oxygenation Reactions Catalyzed by Bacterial
Flavin-Containing Cyclohexanone Oxygenase. J. Am. Chem. Soc. 1985,
107, 2153−2161 and the references therein.
(13) (a) Bong, Y. K.; Clay, M. D.; Collier, S. J.; Mijts, B.; Vogel, M.;
Zhang, X.; Nazor, J.; Smith, D.; Song, S. Synthesis of Prazole
Compounds. US patent 8895271, November 25, 2014. (b) Wilson, R.
Evolving BVMOs for Increasing the Yield and Purity of Enzyme
Catalyzed Oxidation Reactions. Chim. Oggi 2015, 33, 50−52.
(14) (a) Ryerson, C. C.; Ballou, D. P.; Walsh, C. Mechanistic
Studies on Cyclohexanone Oxygenase. Biochemistry 1982, 21, 2644−
2655. (b) Sheng, D.; Ballou, D. P.; Massey, V. Mechanistic Studies of
Cyclohexanone Monooxygenase: Chemical Properties of Intermedi-
ates Involved in Catalysis. Biochemistry 2001, 40, 11156−11167.
(15) (a) Bohlin, M.; Hovarth, K. Crystalline Form of the S-
Enantiomer of Omeprazole. US patent 6162816, December 19, 2000.
̈
̈
(b) Cotton, H.; Kronstrom, A.; Mattson, A.; Moller, E. Form of S-
Omeprazole. US patent 6369085, April 9, 2002.
(16) Truesdale, G. A.; Downing, A. L. Solubility of Oxygen in Water.
Nature 1954, 173, 1236 and the references therein.
̈
(17) Brandstrom, A.; Lindberg, P.; Bergman, N.-Å.; Alminger, T.;
Ankner, K.; Junggren, U.; Lamm, B.; Nordberg, N.; Erickson, M.;
Grundevik, I.; Hagin, I.; Hoffmann, K.-J.; Johansson, S.; Larsson, S.;
̈
Lofberg, I.; Ohlson, K.; Persson, B.; Skånberg, I.; Tekenbergs-Hjelte,
L. Chemical Reactions of Omeprazole and Omeprazole Analogues. I.
A Survey of the Chemical Reactions of Omeprazole and its Analogues.
Acta Chem. Scand. 1989, 43, 536−548.
(18) Iuga, C.; Bojita, M. Stability Study of Omeprazole. Farmacia
(Bucharest, Rom.) 2010, 58, 203−210.
(19) https://www.accessdata.fda.gov/drugsatfda_docs/nda/2013/
021153Orig1s045.pdf (accessed May 16, 2018).
(20) Balke, K.; Beier, A.; Bornscheuer, U. T. Hot Spots for the
Protein Engineering of Baeyer-Villiger Monooxygenases. Biotechnol.
Adv. 2018, 36, 247−263.
(21) Astbury, G. R.; Bugand-Bugandet, J.; Grollet, E.; Stell, K. M.
Flash Points of Aqueous Solutions of Flammable Solvents. Institution
of Chemical Engineers Symposium Series 2004, 150, 489−506. https://
w w w . i c h e m e . o r g / c o m m u n i t i e s / s u b j e c t _ g r o u p s /
safety%20and%20loss%20prevention/resources/hazards%20archive/
~/media/Documents/Subject%20Groups/Safety_Loss_Prevention/
Hazards%20Archive/XVIII/XVIII-Paper-32.pdf (accessed May 16,
2018).
(22) Ang, E. L.; Alvizo, O.; Behrouzian, B.; Clay, M.; Collier, S.;
Eberhard, E.; Fu, F. J.; Song, S.; Smith, D.; Widegren, M.; Wilson, R.;
Xu, J.; Zhu, J. Biocatalysts and Methods for the Synthesis of
Armodafinil. US patent 9267159, February 23, 2016.
F
DOI: 10.1021/acs.joc.8b00468
J. Org. Chem. XXXX, XXX, XXX−XXX