Biocatalytic desymmetrization of an atropisomer with both an enantioselective oxidase and ketoreductases

Article abstract of DOI:10.1002/anie.201002580

All roads lead to ... atropisomeric diaryl ethers containing a benzylic hydroxy group and an aldehyde unit. The isomers were synthesized in enantiomerically enriched form by desymmetrizing atroposelective enzymatic oxidation (with a galactose oxidase (GOa

Full text of DOI:10.1002/anie.201002580

Communications
DOI: 10.1002/anie.201002580
Atroposelectivity
Biocatalytic Desymmetrization of an Atropisomer with both an
Enantioselective Oxidase and Ketoreductases**
Bo Yuan, Abigail Page, Christopher P. Worrall, Franck Escalettes, Simon C. Willies,
Joseph J. W. McDouall, Nicholas J. Turner,* and Jonathan Clayden*
Atropisomeric ligands have found numerous powerful appli-
cations in catalysis,[1] and the atropisomeric biaryl bisphos-
phine binap played an important role in the award of a Nobel
Prize to Noyori in 2001.[2] Enantiomerically pure atropisom-
ers commonly employed as chiral ligands are generally made
by resolution: there are still relatively few effective methods
for direct asymmetric coupling to form single enantiomers.[3]
Kinetic resolution[4] and dynamic resolution[5] under kinetic[5a]
or thermodynamic[5b] control are particularly appealing given
the possibility offered by atropisomerism for thermal race-
mization of the less reactive enantiomer. The use of desym-
metrization for the synthesis of single atropisomers is rare.[6]
Following the early example of enantioselective lithiation
reported by Raston and co-workers,[6b] the research groups of
Hayashi[6c] and Harada[6d] also reported chemical methods for
desymmetrizing biphenyl compounds. A single example of
the enzymatic desymmetrization of a biaryl compound with a
lipase was reported by Matsumoto et al.[6e]
Herein, we report two novel and complementary biocat-
alytic approaches to the enantioselective synthesis of atro-
pisomers by the desymmetrization of appropriate achiral
substrates containing a pair of enantiotopic functional groups.
The atropisomer in question is the diaryl ether 2, which may
be formed either by enantioselective oxidation of the sym-
Scheme 1. Enantioselective transformations of atropisomeric and pro-
atropisomeric diaryl ethers. GOase=galactose oxidase. KRED=keto-
reductase.
metrical diol 1 or by the corresponding reduction of the
symmetrical dialdehyde 3 (Scheme 1). The enzymes we
employed for these transformations were 1) a variant of
galactose oxidase (GOase) which had been previously
evolved to accept chiral benzylic alcohols as substrates with
high enantioselectivity (1!2)[7] and 2) a family of ketoreduc-
tases that are known to possess good activity and enantiose-
lectivity for the asymmetric reduction of benzylic ketones
(3!2).[8]
Atropisomeric diaryl ethers[9] form part of the structure of
vancomycin[10] and are promising scaffolds for the construc-
tion for new chiral ligands.[11] Dialdehyde 3 and diol 1 were
made by our published route.[9] In an initial screen, we
attempted enantioselective acetylation by incubating diol 1
with Candida antarctica lipase B and vinyl acetate. Slow
acylation of 1 was observed with approximately 50%
conversion after 24 h to the monoacetate 4 and modest
enantioselectivity (60% ee). In contrast, when diol 1 was
incubated with the previously reported M3–5 variant of
GOase,[7] rapid oxidation to the monoaldehyde (P)-2 resulted
in 80% conversion after 24 h to material with 94% ee.
[*] B. Yuan, Dr. F. Escalettes, Dr. S. C. Willies, Prof. N. J. Turner
School of Chemistry, University of Manchester
Manchester Interdisciplinary Biocentre
131 Princess Street, Manchester M1 7DN (UK)
Fax: (+44)161-275-1311
E-mail: nicholas.turner@manchester.ac.uk
During the oxidation of 1 to 2, rapid formation of the
product (P)-2 with approximately 88% ee (see below for
assignment of the absolute configuration) was observed after
1 h, followed by a slower increase in enantiomeric purity to a
maximum ee value of 94% (Figure 1). This increase in the
ee value, along with the formation of the dialdehyde 3 (14%
after 24 h), suggested that the minor enantiomer (M)-2
produced in the enantioselective oxidation of 1 was removed
preferentially by a secondary oxidation process to the
dialdehyde 3. Thus, the final enantiomeric purity of (P)-2
resulted from a combination of enantioselective desymmet-
A. Page, C. P. Worrall, Dr. J. J. W. McDouall, Prof. J. Clayden
School of Chemistry, University of Manchester
Oxford Road, Manchester M13 9PL (UK)
E-mail: clayden@man.ac.uk
[**] We are grateful to GlaxoSmithKline and the EPSRC for an Industrial
CASE studentship (to A.P.) and to Croda for a Dorothy Hodgkin
Award (to B.Y.). We thank Dr. James Raftery for the X-ray data on 3.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201002580.
7010
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 7010 –7013

Angewandte
Chemie
Figure 3. CD spectra of 2. Solid line: spectrum observed for the less
reactive enantiomer of 2; dashed line: spectrum modeled for the
lowest-energy conformer of (P)-2.
Figure 1. Enantioselective desymmetrization of diol 1 with the GOase
~
&
M3–5 variant. Composition: diol 1, monoaldehyde 2, ꢀ dialdehyde
^
3; ee value of 2.
and 353 nm (modeled),
a negative maximum at 316
rization and kinetic resolution, as has been noted before for
other systems.[6a]
To investigate the enantioselectivity of the kinetic reso-
lution process, we incubated racemic (ꢀ )-2 with the GOase
M3–5 variant. We observed 89% conversion after 24 h
(Figure 2). As expected, the remaining atropisomer was of
(observed) and 313 nm (modeled), and a shoulder in both
observed and modeled spectra at 280 nm. The match confirms
that the slower-reacting enantiomer of 2, and hence also the
product of the desymmetrization of 1, has the P configuration
(Scheme 1).
The active site of GOase is characterized by a bowl-
shaped cleft near the surface of the protein.[15] Modeling of
the two enantiomers of 2 into the binding site of GOase
revealed that the bulky tert-butyl group of the substrate is
forced to occupy a position in which it is pointing away from
the cleft and towards the surface of the protein. In such a
binding mode, the hydroxymethyl group of (M)-2 is placed
close to the copper-containing reactive center and poised for
oxidation to the aldehyde (Figure 4). This model therefore
Figure 2. Kinetic resolution of (ꢀ)-2 with the GOase M3–5 variant.
&
^
Composition: monoaldehyde 2; ee value of 2.
P configuration with an ee value of 99%. The resolution
process proceeds more slowly than the desymmetrization
reaction. The kinetic resolution under these conditions had a
selectivity (E) value of approximately 4.[12,13]
The configurational stability of the sample of enantio-
merically enriched 2 obtained by kinetic resolution was
established by incubation in heptane at 908C. A slow first-
order decay in the ee value from 62 to 34% was observed over
a period of several days (see the Supporting Information).
From this result, we deduced the half-life for racemization at
this temperature to be 120 h, and estimated the barrier to
enantiomerization to be greater than 130 kJmolꢁ1 at 908C.[14]
The absolute sense of the enantioselective oxidation was
established by comparison of experimental and modeled
circular dichroism spectra of the starting material remaining
after kinetic resolution, (+)-(P)-2 (see the Supporting Infor-
mation). The modeled spectrum (Figure 3, dotted line)
matches closely the experimental CD spectrum (Figure 3,
continuous line), with positive maxima aligning at 234
(observed) and 235 nm (modeled) and at 348 (observed)
Figure 4. Model of the more reactive enantiomer (M)-2 in the active
site of galactose oxidase.
predicts that (M)-2 is the faster-reacting enantiomer in the
kinetic resolution process, an observation which is in agree-
ment with the assignment of absolute configuration on the
basis of circular dichroism.
We also examined an alternative approach for the syn-
thesis of enantiomerically enriched monoaldehyde 2: asym-
metric reduction of the symmetrical dialdehyde 3 with
ketoreductases (KREDs). KREDs have been widely applied
to the asymmetric reduction of ketones, particularly benzylic
Angew. Chem. Int. Ed. 2010, 49, 7010 –7013
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7011

Communications
ketones,[16] but to our knowledge have not previously been
[1] R. Noyori, Asymmetric Catalysis in Organic Synthesis, Wiley,
New York, 1994; Y.-M. Li, F.-Y. Kwong, W.-Y. Yu, A. S. C. Chan,
Coord. Chem. Rev. 2007, 251, 2119; M. Berthod, G. Mignani, G.
Woodward, M. Lemaire, Chem. Rev. 2005, 105, 1801.
[2] R. Noyori, Angew. Chem. 2002, 114, 2108; Angew. Chem. Int. Ed.
2002, 41, 2008; binap = 2,2’-bis(diphenylphosphanyl)-1,1’-
binaphthyl.
used for the desymmetrization of pro-atropisomeric sub-
strates. We screened 17 different KREDs[17] and determined
the conversion and ee value of the product after 24 h
(Table 1). All enzymes were active, and many showed high
Table 1: Reduction of 3 with ketoreductases (KREDs).
[3] a) G. Bringmann, A. J. P. Mortimer, P. A. Keller, M. J. Gresser, J.
Garner, M. Breuning, Angew. Chem. 2005, 117, 5518; Angew.
Chem. Int. Ed. 2005, 44, 5384; b) A. N. Cammidge, K. V. L.
Crꢀpy, Chem. Commun. 2000, 1723; J. Yin, S. L. Buchwald, J.
Am. Chem. Soc. 2000, 122, 12051.
[4] T. Itoh, J. Org. Chem. 1995, 60, 4968; O. Kitagawa, M. Fujita, M.
Kohriya, H. Hasegawa, T. Taguchi, Tetrahedron Lett. 2000, 41,
8539; T. Watanabe, K. Kamikawa, M. Uemura, Tetrahedron Lett.
1995, 36, 6695; R. Rios, C. Jimeno, P. J. Carroll, P. J. Walsh, J.
Am. Chem. Soc. 2002, 124, 10272; J. Clayden, H. Turner,
Tetrahedron Lett. 2009, 50, 3216; D. Pamperin, H. Hopf, C.
Syldatk, M. Pietzsch, Tetrahedron: Asymmetry 1997, 8, 319; D.
Pamperin, C. Schulz, H. Hopf, C. Syldatk, M. Pietzsch, Eur. J.
Org. Chem. 1998, 1441.
[5] a) G. Bringmann, R.-M. Pfeifer, P. Schreiber, K. Hartner, M.
Schraut, M. Breuning, Tetrahedron 2004, 60, 4349; G. Bring-
mann, D. Menche, Acc. Chem. Res. 2001, 34, 615; G. Bringmann,
J. Hinrichs, T. Pabst, P. Henschel, K. Peters, E.-M. Peters,
Synthesis 2001, 155; A. Bracegirdle, J. Clayden, L. W. Lai,
Beilstein J. Org. Chem. 2008, 4, 47; b) J. Clayden, P. Johnson, J. H.
Pink, M. Helliwell, J. Org. Chem. 2000, 65, 7033; J. Clayden, D.
Mitjans, L. H. Youssef, J. Am. Chem. Soc. 2002, 124, 5266; J.
Clayden, L. W. Lai, M. Helliwell, Tetrahedron 2004, 60, 4399; J.
Clayden, C. P. Worrall, W. Moran, M. Helliwell, Angew. Chem.
2008, 120, 3278; Angew. Chem. Int. Ed. 2008, 47, 3234; J.
Clayden, J. Senior, M. Helliwell, Angew. Chem. 2009, 121, 6388;
Angew. Chem. Int. Ed. 2009, 48, 6270; J. Clayden, S. P. Fletcher,
J. J. W. McDouall, S. J. M. Rowbottom, J. Am. Chem. Soc. 2009,
131, 5331.
[6] a) E. Garcꢁa-Urdiales, I. Alfonso, V. Gotor, Chem. Rev. 2005,
105, 313; b) L. M. Engelhardt, W.-P. Leung, C. L. Raston, G.
Salem, P. Twiss, A. H. White, J. Chem. Soc. Dalton Trans. 1988,
2403; c) T. Hayashi, S. Niizuma, T. Kamikawa, N. Suzuki, Y.
Uozumi, J. Am. Chem. Soc. 1995, 117, 9101; T. Kamikawa, Y.
Uozumi, T. Hayashi, Tetrahedron Lett. 1996, 37, 3161; d) T. M. T.
Tuyet, T. Harada, K. Hashimoto, M. Hatsuta, A. Oku, J. Org.
Chem. 2000, 65, 1335; T. Harada, T. M. T. Tuyet, A. Oku, Org.
Lett. 2000, 2, 1319; e) T. Matsumoto, T. Konegawa, T. Nakamura,
K. Suzuki, Synlett 2002, 122.
[7] F. Escalettes, N. J. Turner, ChemBioChem 2008, 9, 857 – 860;
M. D. Truppo, F. Escalettes, N. J. Turner, Angew. Chem. 2008,
120, 2679 – 2681; Angew. Chem. Int. Ed. 2008, 47, 2639 – 2641.
[8] G. W. Huisman, J. Liang, A. Krebber, Curr. Opin. Chem. Biol.
2010, 14, 122 – 129.
[9] M. S. Betson, J. Clayden, C. P. Worrall, S. Peace, Angew. Chem.
2006, 118, 5935; Angew. Chem. Int. Ed. 2006, 45, 5803.
[10] K. C. Nicolaou, C. N. C. Boddy, S. Brꢂse, N. Winssinger, Angew.
Chem. 1999, 111, 2230; Angew. Chem. Int. Ed. 1999, 38, 2096.
[11] For uses of non-biaryl atropisomers in catalysis, see: A. Honda,
K. M. Waltz, P. J. Carroll, P. J. Walsh, Chirality 2003, 15, 615; W.-
M. Dai, K. K. Y. Yeing, J.-T. Liu, Y. Zhang, I. D. Williams, Org.
Lett. 2002, 4, 1615; W.-M. Dai, K. K. Y. Yeung, Y. Wang,
Tetrahedron 2004, 60, 4425; W.-M. Dai, Y. Li, Y. Zhang, Z. Yue,
J. H. Wu, Chem. Eur. J. 2008, 14, 5538; S. Brandes, B. Niess, M.
Bella, A. Prieto, J. Overgaard, K. A. Jørgensen, Chem. Eur. J.
2006, 12, 6039.
Entry
Enzyme[a]
Conversion[b] [%]
ee [%]
Configuration[c]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
KRED101 (R)
KRED103
KRED104
KRED105
KRED106
KRED108 (S)
KRED109
KRED110
KRED114
KRED115 (R)
KRED116
53
6
3
2
3
39
3
1
95
33
23
7
91
99
22
84
3
24
22
75
99
11
78
54
99
40
9
79
79
77
71
87
61
99
P
P
M
M
P
M
M
M
M
M
M
M
P
KRED117
KRED118 (S)
KRED119 (S)
KRED120 (S)
KRED121 (R)
KRED124 (R)
P
M
M
M
[a] The typical selectivity of the KRED in the reduction of nonsymmetrical
diaryl ketones is shown in parentheses (when known).[16] [b] Conversion
after 24 h. [c] Major enantiomer produced by reduction.
enantioselectivity (Table 1, entries 4, 8, and 17); however,
conversion was often low. Notable success was achieved with
KRED118 (91% conversion, 77% ee (P enantiomer)) and
KRED121 (84% conversion, 61% ee (M enantiomer);
Table 1, entries 13 and 16). These KREDs were previously
classified as either S- or R-selective on the basis of their
stereoselectivity in the reduction of nonsymmetrical diaryl
ketones.[16] Interestingly, this trend was not always followed
for the reduction of 3; that is, enantiomerically enriched (P)-2
was obtained from both R-selective KRED101 and S-
selective KRED118 (Table 1, entries 1 and 13). Similarly,
(M)-2 was obtained from both R- (KRED121) and S-selective
(KRED108) enzymes (Table 1, entries 16 and 6). These
results suggest that dialdehyde 3 binds at the active site of
the ketoreductases in a different mode to that adopted by
more conventional benzylic ketones.
Our results suggest that biocatalytic redox reactions may
be more widely applicable to the asymmetric synthesis of
atropisomers by desymmetrization. We are currently explor-
ing further opportunities in this area.
Received: April 29, 2010
Published online: August 16, 2010
Keywords: asymmetric synthesis · atropisomerism ·
biocatalysis · diaryl ethers · enzymes
[12] W. Kroutil, A. Kleewein, K. Faber, Tetrahedron: Asymmetry
1997, 8, 3263 – 3274.
.
7012
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 7010 –7013

Angewandte
Chemie
[13] Attempted kinetic resolution with GOase of a less hindered
compound with an isopropyl group in place of the tert-butyl
group gave the racemic compound, probably as a result of
racemization on the timescale of the reaction: compound 2 is
resolvable by HPLC, but related compounds have low barriers to
racemization (see Ref. [9]).
[15] S. J. Firbank, M. S. Rogers, C. M. Wilmot, D. M. Dooley, M. A.
Halcrow, P. F. Knowles, M. J. McPherson, S. E. Phillips, Proc.
Natl. Acad. Sci. USA 2001, 98, 12932.
[16] M. D. Truppo, D. Pollard, P. Devine, Org. Lett. 2007, 9, 335 – 338.
[17] All ketoreductases (KREDs) were obtained from Codexis Inc.,
200 Penobscot Drive, Redwood City, CA 94063, USA.
[14] See: A. Ahmed, R. A. Bragg, J. Clayden, L. W. Lai, C. McCarthy,
J. H. Pink, N. Westlund, S. A. Yasin, Tetrahedron 1998, 54, 13277.
Angew. Chem. Int. Ed. 2010, 49, 7010 –7013
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7013