Chemo-enzymatic enantioconvergent synthesis of C4-building blocks containing a fully substituted chiral carbon center using bacterial epoxide hydrolases

Article abstract of DOI:10.1055/s-2001-9703

A highly efficient chemo-enzymatic asymmetric synthesis of chiral C₄-building blocks containing a fully substituted carbon center is reported. The key transformation consists of a deracemization based on an enantioconvergent asymmetric hydrolys

Full text of DOI:10.1055/s-2001-9703

LETTER
111
Chemo-Enzymatic Enantioconvergent Synthesis of C4-Building Blocks
Containing a Fully Substituted Chiral Carbon Center using Bacterial Epoxide
Hydrolases
Andreas Steinreiber, Helena Hellström, Sandra F. Mayer, Romano V. A. Orru,1 Kurt Faber
University of Graz, Department of Chemistry, Organic and Bio-organic Chemistry, Heinrichstrasse 28, A-8010, Graz, Austria
Fax +43-316-380-9840; E-mail: kurt.faber@kfunigraz.ac.at
Received 29 September 2000
tuted epoxides and their corresponding vic-diols were ob-
Abstract: A highly efficient chemo-enzymatic asymmetric synthe-
tained in excellent ee's.3 In order to overcome the inherent
drawbacks of kinetic resolution, deracemization methods
sis of chiral C4-building blocks containing a fully substituted carbon
center is reported. The key transformation consists of a deracemiza-
tion based on an enantioconvergent asymmetric hydrolysis of an ep-
have been developed which transform all of the racemic
oxide using combined bio- and chemo-catalysis leading to a single starting material into a single enantiomeric product.4
enantiomeric product in >98% e.e. A simple switch of steps leads to
In this letter, we wish to report a simple and cheap
kinetic resolution giving access to products of opposite configura-
chemoenzymatic protocol to produce both enantiomers of
tion.
acetonides 1 and 2 on a preparative scale with high selec-
Key words: biotransformation, epoxide hydrolase, deracemization,
tivity. The latter C4-units have been used in the asymmet-
C4-building block, enantioconvergent
ric synthesis of numerous natural products.5-9 In addition,
several structurally related chiral building blocks were
synthesized.
During the last few years, microbial epoxide hydrolases
From previous studies it was known that oxiranes bearing
have proven to be powerful catalysts for controlling the
polar functional groups (e.g. -OH) in the side chain were
stereochemistry of epoxides and their corresponding 1,2-
not accepted by bacterial epoxide hydrolases.10 However,
suitable protection (e.g. with an ether moiety) rendered
these compounds more lipophilic and thus made them ac-
ceptable for the enzymes.10 From these considerations,
benzyloxy oxirane ( )-5 (readily prepared from methallyl
alcohol10, Scheme 1) was the substrate of choice. As pre-
viously reported,11,12 the latter could be hydrolyzed on an
analytical scale by using whole cells of Rhodococcus
NCIMB 11216. Upon detailed investigation on a larger
scale, however, Rhodococcus SM 1789 proved to be
diol derivatives.2 Enzymes from bacterial sources are of
special interest since they seem to be the only biocatalysts
that selectively hydrolyze rather bulky oxiranes bearing a
fully substituted chiral carbon center.3 Bacterial epoxide
hydrolases are cofactor-independent and are available in
sufficient quantities by fermentation without the need for
enzyme induction. The preparative applicability of bacte-
rial epoxide hydrolases (usually employed as whole lyo-
philized cells) has been demonstrated on a range of
synthetically useful substrates: In particular, 2,2-disubsti-
Scheme 1
Synlett 2001 No. 1, 111–113 ISSN 0936-5214 © Thieme Stuttgart · New York

112
A. Steinreiber et al.
LETTER
superior for preparative-scale transformations in view of ucts.20b On the other hand, a recently reported one-pot pro-
reaction rate, enantioselectivity and reproducibility.13
tection-oxidation sequence21 worked rather well: Thus,
when (R)-6 was subsequently treated with 1 equiv. of n-
BuLi, di-tert-butylchlorosilane and another equiv. of n-
BuLi21 in the presence of N,N,N’,N’-tetramethylene-di-
amine (TMEDA), silyl derivative (R)-717a was isolated in
good yield (49%) and without racemization.22 The poten-
tial of silyloxy derivatives (R)-7, which could be oxidized
using PDC20c to the corresponding aldehyde,23 as chiral
intermediate is apparent, since both hydroxyl moieties are
differently protected which makes sequential chemical
modifications at each hydroxyl group possible.
Enzymatic hydrolysis of ( )-5 was shown to proceed via
attack at the less substituted carbon atom2b and, as a con-
sequence, the biotransformation took place with retention
of configuration at the stereogenic center to afford a 1:1
mixture of epoxide (R)-5 and diol (R)-6. Note that the ap-
parent inversion [(S)-5 Æ (R)-6] of the biohydrolysis is
only due to a switch in the CIP-priority and not due to in-
version at the stereogenic center. The scale-up of this re-
action was optimized using Rhodococcus SM 1789 as
follows: (i) In order to keep the reaction on a manageable
size and to improve recovery rates, the total volume of
buffer (Tris, 50 mM, pH 8) was considerably decreased
relative to the amount of epoxide and biocatalyst as com-
pared to analytical-scale reactions.10 (ii) The reaction tem-
perature was elevated to enhance the rate of
biohydrolysis. From previous studies it was known that
(even for epoxide hydrolases), epoxides exert a certain in-
hibition at elevated concentrations12 going in hand with
some deactivation at prolonged reaction times. The biohy-
drolysis came to a standstill at 50% conversion indicating
high enantioselectivity (E >100). Thus, ( )-5 could be ef-
ficiently resolved to give (R)-514 and (R)-6 within 80
hours.15a The optical purity of both the remaining epoxide
(R)-5 and the corresponding product diol (R)-6 was found
to be high (>98%)10 and the total recovery of materials
was optimal (>85%) when the reaction mixture was con-
tinuously extracted with CH2Cl2.
Table Chemical and Optical Yields.
a Determined by correlation with authentic material, see also
reference 10.
b Isolated yields calculated from methallyl alcohol 3.
c Determined by GC on a chiral stationary phase.10,15e
In conclusion, our synthesis of building blocks 1 and 2
Deracemization of ( )-5 was achieved by employing a shows the following advantages: (i) The method starts
protocol based on sequential bio- and chemo-hydrolysis.4 from cheap starting material and (ii) it gives access to both
Thus, the crude mixture of (R)-5 and (R)-6 resulting from stereo-isomers in high optical purity. (iii) Aldehyde (R)-2
biohydrolysis was directly treated with cat. 93% aq. may be obtained via deracemization, i.e. no ’unwanted’
H2SO4 in dioxane15b to effect hydrolysis of (R)-5 with enantiomer occurs as by-product. In addition, a useful si-
complete inversion. Thus, diol (R)-6 was isolated as the lylated synthon (R)-7 was prepared. In the latter com-
sole product in 75% chemical yield and >95% e.e. Protec- pound, the tert- and prim-hydroxyl group are differently
tion of the vic-diol as the acetonide [2,2-dimethoxypro- protected, which allows highly flexible synthetic transfor-
pane, cat. ion exchange resin IR 120 (H+-form)] followed mations. The application of these chiral synthons for the
by hydrogenolysis (Pd/C 10%, H2, 1 bar) of the benzyl
asymmetric synthesis of a physiologically active sesquit-
ether afforded the desired enantioenriched synthon (S)-117 erpenoid compound is currently in progress.
in 34% overall yield and >98% e.e. starting from methal-
lyl alcohol. Finally, carbaldehyde (R)-217 was obtained
Acknowledgement
without racemization via Dess-Martin oxidation18 follow-
ing a previously described procedure.
This research was performed within the Spezialforschungsbereich
Biokatalyse and was financed by the Fonds zur Förderung der wis-
senschaftlichen Forschung (project no. F-104) and the Austrian Mi-
nistry of Science.
In order to broaden the applicability of this method, access
to the opposite enantiomeric series [i.e. (R)-1 and (S)-2]
was achieved by a modification of the reaction sequence:
Thus, treatment of (R)-5 with aq. LiOH effected hydroly-
sis with complete retention to furnish diol (S)-6.
References and Notes
(1) New address: R. V. A. Orru, Division of Chemistry, Bio-
organic Chemistry, Vrije University Amsterdam, de
Boelelaan 1083, 1081 HV, Amsterdam, The Netherlands.
(2) a) Archer, I. V. J. Tetrahedron 1997, 53, 15617. b) Orru, R. V.
A.; Archelas, A.; Furstoss, R.; Faber, K. Adv. Biochem. Eng.
Biotechnol. 1999, 63, 145.
(3) Orru, R. V. A.; Faber, K. Curr. Opin. Chem. Biol. 1999, 3, 16
and references cited therein.
(4) Orru, R. V. A.; Mayer, S. F.; Kroutil, W.; Faber, K.
Tetrahedron 1998, 54, 859.
Direct oxidation of the primary alcohol in either enanti-
omer of diol 6 would give the corresponding a-hydroxy-
aldehyde, a useful intermediate for the preparation of
a-substituted a-hydroxy carboxylic acids.19 To our disap-
pointment, none of several oxidation methods attempted
proved to be successful;20 whereas Dess-Martin or Swern
conditions mainly led to condensation, chromium-based
oxidants, such as PDC or PCC gave rearranged prod-
Synlett 2001, No. 1, 111–113 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Chemo-Enzymatic Enantioconvergent Synthesis of C4-Building Blocks
113
(5) a) Wattenbach, C.; Maurer, M; Frauenrath, H. Synlett 1999,
303. b) Barner, R.; Hübscher, J. Helv. Chim. Acta 1983, 66,
880.
(6) Chen, S.-T.; Fang, J.-M. J. Org. Chem. 1997, 62, 4349.
(7) Wershofen, S.; Claßen, A.; Scharf, H.-D. Liebigs Ann. Chem.
1989, 9.
(8) For example, see: a) Williams, R. M.; Durham, C. A. Chem.
Rev. 1988, 88, 511. b) Williams, R. M.; Sabol, M. R.; Kim, H.-
D.; Kwast, A. J. Am. Chem. Soc. 1991, 113, 6621. c) Yamaura,
M.; Nakayama, T.; Hashimoto, H.; Shin, C.; Yoshimura, J. J.
Org. Chem. 1988, 53, 6035.
(9) Nicolaou, K. C.; Theodorakis, E. A.; Rutjes, F. P. J. T.; Sato,
M.; Tiebes, J.; Xiao, X.-Y.; Hwang, C.-K.; Duggan, M. E.;
Yang, Z.; Couladouros, E. A.; Sato, F.; Shin, J.; He, H.-M.;
Bleckman, T. J. Am. Chem. Soc. 1995, 117, 10239.
(10) Steinreiber, A.; Osprian, I.; Mayer, S. F.; Orru, R. V. A.;
Faber, K. Eur. J. Org. Chem. 2000, 3703.
c) Hydrolysis of (R)-5 to give (S)-6: To a stirred solution of
(R)-5 (150 mg, 0.84 mmol) in H2O (10 mL) and THF (3 mL)
LiOH (1.3 g, 54.3 mmol) was added and the mixture was
refluxed for 30 h. The reaction was quenched by addition of
H2O (50 mL) and sat. aq. NH4Cl (10 mL). Products were
extracted with EtOAc (5 ¥ 50 mL), the organic phase was
dried (Na2SO4) and evaporated to give (S)-6 (82 mg, 50%). No
racemization occurred.
d) Preparation of 1 and 2 from 6: 261 mg of (R)-6 (1.33
mmol, ee = 92%) were stirred in dimethoxypropane (5 mL,
40.7 mmol) under acidic conditions for 2 h (50 mg Amberlite
IR 120, H+ form). After filtration and evaporation, abs.
ethanol (5 mL) and Pd/C (10%, 5% w:w) were added. The
solution was stirred under an atmosphere of H2 to furnish
(after filtration and evaporation) (S)-1 (111 mg, 0.76 mmol,
57%, e.e. = 94%).
(S)-1 (45 mg, 0.31 mmol, ee = 97%) was stirred in CH2Cl2
(1.5 mL). To this solution Dess-Martin oxidant (periodinane,
0.25 g, 0.59 mmol)18 in CH2Cl2 (mL) was added. The solution
was diluted with ether (7 mL) and sat. NaHCO3 solution
(5 mL). The organic layer was dried (Na2SO4) and evaporated
to yield (R)-2 (29 mg, 0.20 mmol, 65%, ee = 98%).
e) Enantiomeric composition: Determined with Varian 3800
gas chromatograph equipped with FID, using a CP-Chirasil-
DEX CB column (25m ¥ 0.32mm, 0.25mm film, H2). 1:
isotherm 55 °C, 14.5 psi, 6.39 min (S) and 6.71 min (R); 2:
isotherm 90 °C, 14.5 psi, 4.00 min (S) and 4.26 min (R).
(16) The evident asymmetric Sharpless epoxidation/
dihydroxylation methods give less satisfactory results. See
Tietze, L. F.; Görlitzer, J. Synthesis 1998, 873.
(17) Spectroscopic and physical data were in full agreement with
those previously reported. For 1 see: Calinaud, P.; Gelas, J.
Bull. Soc. Chim. Fr. 1975, 1228. For 2 see: a) Barner, R.;
Schmid, M. Helv. Chim. Acta 1979, 62, 2384. b) Trost, B. M.;
Marrs, C. J. Org. Chem. 1991, 56, 6468.
(18) Dess, D. B.; Martin, J. C. J. Org. Chem. 1983, 48, 4155.
(19) Coppola, G. M.; Schuster, H. F. In a-Hydroxy Acids in
Enantioselective Syntheses; Verlag Chemie: Weinheim,
1997, p. 128, 145, 267, and 284.
(20) a) Swern oxidation of a model compound [(S)-6-bromo-2-
methyl-1,2-heptanediol] afforded the corresponding (S)-
aldehyde in reasonable yield (23%) and without racemization.
b) See also: Hudlicky, M. In Oxidations in Organic
Chemistry; Am. Chem. Soc.: Washington, DC, 1990, p. 128,
145, 267 and 284. c) Corey, E. J.; Schmidt, G. Tetrahedron
Lett. 1979, 5, 399.
(21) The second equivalent of n-BuLi acts as a nucleophile.
Tanino, K.; Shimizu, T.; Kuwahara, M.; Kuwajima, I. J. Org.
Chem. 1998, 63, 2422.
(11) Mischitz, M.; Kroutil, W.; Wandel, U.; Faber, K.
Tetrahedron: Asymmetry 1995, 6, 1261.
(12) Orru, R. V. A.; Osprian, I.; Kroutil, W.; Faber, K. Synthesis
1998, 1259.
(13) Biocatalyst preparation:
Rhodococcus ruber SM 1789 from the culture collection of
the Institute of Biotechnology (Graz University of
Technology) was grown in baffled shake-flask cultures on the
following medium: yeast extract (10.0 gL-1), peptone (10.0
gL-1), glucose (10.0 gL-1), NaCl (2.0 gL-1), MgSO4•7H2O
(0.147 gL-1), NaH2PO4 (1.3 gL-1), K2HPO4 (4.4 gL-1). Cells
were harvested at the late exponential growth phase by
centrifugation (3000 ¥ g), washed with Tris/HCl-buffer (0.05
M, pH 8.0) and lyophilized. The cells could be stored at 4 °C
for several months without a noticeable loss of activity. For
experimental details see ref. 12
(14) For example, epoxide 5 was used for the preparation of
a-methylamino acids. See: Lakner, F. J.; Hager, L. P.
Tetrahedron: Asymmetry 1997, 8, 3547 and references cited
therein.
(15) For the synthesis of ( )-5 see ref. 10.
a) Biohydrolysis of ( )-5. Epoxide ( )-5 (1.2 g, 6.73 mmol)
was added to a suspension of rehydrated lyophilized whole
cells of Rhodococcus SM 1789 (2.00 g) in Tris-buffer (30 mL,
0.05 M, pH 8). The mixture was agitated on a rotary shaker at
30∞ C and 130 rpm. After about 80 h, the reaction ceased at
50% conversion and the mixture was continuously extracted
with CH2Cl2 (100 mL, 20 h). The organic layer was washed
with brine (50 mL), dried (Na2SO4) and evaporated to furnish
a slightly orange oil (1.3 g). This residue can be directly
deracemized (see b). Alternatively, (R)-5 and (R)-6 can be
separated by flash chromatography (Merck silica gel 60,
petroleum ether/EtOAc 2:1) to afford 552 mg (3.1 mmol,
46%, ee = 99%) (R)-5 and 647 mg (3.3 mmol, 49%, ee = 99%)
(R)-6. The products were analyzed by GC on a chiral
stationary phase as described in reference 10.
(22) (R)-3-Benzyloxy-2-[n-butyl(di-tertbutyl)silyloxy]-2-
methyl-1-propanol (7): 1H NMR (360 MHz, CDCl3):
d = 0.73-1.02 (m, 22H), 1.28-1.33 (m, 8H), 3.38 -3.61 (m,
4H), 3.49-3.59 (q, J = 12 Hz, 2H), 7.27-7.38 (m, 5H); 13
C
b) Deracemization of ( )-5. The mixture of (R)-5 and (R)-6
obtained from the biohydrolysis (1.3 g) was dissolved in
dioxane (225 mL), and 93% (v/v) aq. H2SO4 (18 mL) was
added dropwise at 15 ∞C with stirring. After 15 min at r.t. the
acid was neutralized with sat. aq. NaHCO3. EtOAc (300 mL)
was added, and the resulting biphasic mixture was vigorously
stirred for an additional 30 min. The aq. layer was extracted
with EtOAc (3 ¥ 100 mL), the combined organic layers were
dried (Na2SO4) and the solvents were evaporated. Flash
chromatography of the oily residue (Merck silica gel 60,
petroleum ether/EtOAc 2:1) afforded (R)-6 as white crystals
in 0.70 g (60%) yield and 98% ee. Spectroscopic, physical and
optical data were in full agreement with those previously
reported: Tanner, D.; Somfai, P. Tetrahedron 1986, 42, 5985.
NMR (90 MHz, CDCl3): d = 28.71, 76.72, 77.08, 77.43,
127.59, 127.68, 128.42; elemental analysis [calcd. (found)]:
C: 69.84% (69.96), H: 10.71% (10.73). The enantiomeric
composition of (R)-7 was determined via HPLC using a
Chiralpak AD column (4.6mm ¥ 240mm, heptane/i-propanol
98/2, 0.6ml/min): (S)-7 10.39 min, (R)-7 11.28 min.
(23) 3-Benzyloxy-2-[n-butyl(di-tertbutyl)silyloxy]-2-methyl-1-
propanal: 1H NMR (360 MHz, CDCl3): d = 0.77-1.27 (m,
30H), 3.42-3.52 (m, 2H), 5.23 (s, 3H), 7.20-7.27 (m, 5H), 9.64
(s, 1H).
Article Identifier:
1437-2096,E;2001,0,01,0111,0113,ftx,en;G21000ST.pdf
Synlett 2001, No. 1, 111–113 ISSN 0936-5214 © Thieme Stuttgart · New York