An efficient] large-scale synthesis of (R)-(-)-mevalonolactone using simple biological and chemical catalysts

Article abstract of DOI:10.1055/s-1998-6108

Natural (R)-(-)-mevalonolactone (2) was synthesized in eight steps in 55% total yield and >99% ee employing an enantioconvergent chemoenzymatic route. In the key step, 2-benzyl-2-methyloxirane (±)-3 was deracemized on a large scale (10 g) using lyophilized cells of Nocardia EH1 and sulfuric acid. The product diol (S)-4 was isolated in 94% chemical yield and 94% optical purity.

Full text of DOI:10.1055/s-1998-6108

SYNTHESIS
September 1998
1259
An Efficient Large-Scale Synthesis of (R)-(–)-Mevalonolactone Using Simple Biological
and Chemical Catalysts
Romano V. A. Orru,* Ingrid Osprian, Wolfgang Kroutil, Kurt Faber
Institute of Organic Chemistry, Graz University of Technology, Stremayrgasse 16, A-8010 Graz, Austria
Fax: +43(316)8738740; E-mail: romano@orgc.tu-graz.ac.at
Received 22 December 1997; revised 9 February 1998
Abstract: Natural (R)-(−)-mevalonolactone (2) was synthesized
in eight steps in 55% total yield and >99% ee employing an enan-
tioconvergent chemoenzymatic route. In the key step, 2-benzyl-2-
methyloxirane (±)-3 was deracemized on a large scale (10 g)
using lyophilized cells of Nocardia EH1 and sulfuric acid. The
product diol (S)-4 was isolated in 94% chemical yield and 94%
optical purity.
Scheme 1
Key words: mevalonolactone, epoxide hydrolase, chemoenzym-
atic synthesis, enantioconvergent
scribe a chemoenzymatic route that does not incur these
drawbacks and yields (R)-(−)-mevalonolactone (2) effi-
ciently.
(R)-(−)-Mevalonic acid (1) is a key intermediate in a
broad spectrum of cellular biological processes and their
regulation.1 Its significance as a biosynthetic precursor for
terpenes, vitamins, and sterols has been clearly demon-
strated. The transformation of 3-hydroxy-3-methyl-
glutaryl CoA into mevalonate, catalyzed by HMG-CoA
reductase, is a crucial step in the biosynthesis of cholester-
ol.1 In medicine it is known that inhibition in the produc-
tion of cellular 1 leads to lowering of plasma cholesterol.2
This makes the biological formation of (R)-(−)-mevalonic
acid (1) an important control site.3a The level of (R)-1 in
plasma and/or urine has been used to study the mecha-
nisms and efficacy in the treatment of hyperlipidemias,
and the efficacy of lovastatin as an anticancer agent.3
Thus, both for diagnostic and therapeutic reasons, simple
and cheap access to natural mevalonic acid (1) is of signif-
icant value. As such 1 or, more particularly, its cyclic lac-
tone form 2 (Scheme 1) has been a synthetic target of con-
siderable interest. Folkers and co-workers isolated 2 and
elucidated its structure.4 The naturally occurring (−)-iso-
mer has the (R)-configuration which was shown by corre-
lation of the (+)-enantiomer with (−)-quinic acid.5 Corn-
forth and co-workers were the first to prepare both enan-
tiomers in essentially 100% ee from (+)- and (−)-linalool.6
Since then a number of asymmetric syntheses of (R)-(−)-
mevalonolactone (2) have been reported. Among the
chemical routes the most important employ the Sharpless
asymmetric epoxidation,7 which is a stereodifferentiating
reaction, as well as those routes exploiting starting mate-
rials from the chiral pool.8 A different and highly enanti-
oselective approach is based on transcription of chirality
from a chiral template.9 Biocatalysts have also been used
for the preparation of 2. (R)-(−)-Mevalonolactone (2) may
be produced via fermentation,10 via procedures involving
lipase-catalyzed kinetic resolutions,11 or via hydrolysis of
a prochiral diester employing esterases.12 A more recently
published method includes a chloroperoxidase-catalyzed
epoxidation of 3-methylbut-3-enoate.13
2,2-Disubstituted epoxides serve as suitable substrates for
a recently developed enantioconvergent chemoenzymatic
deracemization reaction.14 This process proceeds via a
socalled resolution-inversion sequence. Hydrolysis of
(±)-epoxides catalyzed by a bacterial epoxide hydrolase
proceeds via a classical kinetic resolution pattern.15 The
biotransformation reaction shows retention of configura-
tion at the stereogenic center and produces a mixture of
(R)-epoxide and (S)-diol.16 Directly following the biohy-
drolysis, the remaining (R)-epoxide is hydrolyzed selec-
tively by simple acid catalysis yielding the (S)-diol with
complete inversion. In this way racemic 2,2-disubstituted
oxiranes were selectively deracemized to give the corre-
sponding (S)-1,2-diols in high chemical and optical
yields.14
Substrates possessing a polar carboxy moiety relatively
close to the oxirane ring are not accepted by the bacteria
available, though an aryl functionality is lipophilic
enough to be tolerated by the epoxide hydrolase.17 Thus,
2-benzyl-2-methyloxirane [(±)-3], prepared from readily
available phenylacetone and trimethylsulfoxonium io-
dide,17 is selectively hydrolyzed by lyophilized cells of
Nocardia EH1 exhibiting strong epoxide hydrolase activ-
ity.14 At this point we envisaged that the phenyl moiety
might serve as a masked carboxylate functionality for a
total synthesis of (R)-(−)-mevalonolactone (2) (Scheme
2).
However, most of these methods make use of expensive
reagents, are environmentally unsound, or are not applica-
ble on a large scale. In the present paper we wish to de-
Scheme 2

SYNTHESIS
Papers
1260
In general, oxiranes are believed to be toxic to living
cells.17 For example the Nocardia sp. dies within 72 hours
upon incubation with (±)-3. Furthermore, compound 3 is
not very soluble in the buffer system required for opti-
mum activity and selectivity of the biocatalyst. This was
believed to be problematic for a large-scale biotransfor-
mation of 3.18 To enhance solvation of the substrate which
would promote the biohydrolysis reaction, a range of or-
ganic cosolvents at different concentrations were tested.19
To our disappointment the enzymatic activity was lost and
no hydrolysis products were detected at all. Apparently
the epoxide hydrolase is deactivated in the solvent sys-
tems tested. At this point we decided to perform the large-
scale biohydrolysis in plain buffer without any organic co-
solvent. In order to keep the reaction manageable, the
buffer volume relative to the amount of substrate and bio-
catalyst was decreased considerably as compared to the
analytical scale reaction.17 Furthermore, the enzymatic
hydrolysis was performed at a slightly elevated tempera-
ture in order to speed up the reaction. In this way (±)-3
could be resolved with high selectivity (E = 123)20 into
(R)-3 and (S)-4 on a large scale (10 g) within 48 hours.
These results support a recent semiempirical model for
enzymatic suspension reactions,21 which suggests that the
process time in such cases can only be accelerated by in-
creasing the substrate solubility through raising the tem-
perature and not through the addition of organic cosol-
vents.
Scheme 3
Direct coupling of the acid-catalyzed inversion reaction to
the biocatalytic resolution went smoothly. Accordingly, a
mixture of (R)-3 and (S)-4 was treated with 93% aqueous
H2SO4 as catalyst in dioxane from which the diol (S)-4
could be isolated in 94% yield and 94% ee. An efficient
procedure for the large-scale preparation of the key inter-
mediate had been established, and the synthesis of natural
mevalonolactone could be completed (Scheme 3).
addition of a catalytic amount of fresh ruthenium trichlo-
ride. It should be emphasized that the ruthenium catalyst
was added, at once, to a refluxing biphasic water/aceto-
nitrile/CCl4 (2:1:1) solvent mixture containing the aryl
compound and sodium metaperiodate. Immediate workup
(to prevent overoxidation) gave 72% of pure carboxylic
acid (R)-10. Simple saponification followed by acidic lac-
tonization afforded enantiopure (>99% ee) natural (R)-(−)-
mevalonolactone (2), which was isolated in 55% overall
yield starting from racemic 3.
The primary hydroxy group of (S)-4 was tosylated (TsCl,
pyridine) and after crystallization enantiopure (S)-5 was
obtained in 95% yield. Chain elongation was achieved
quantitatively by treatment of the tosylate with KCN in
EtOH/H2O. Carboxylic acid (S)-7 was then formed in
high yield from nitrile (S)-6 using strongly basic H2O2.
Alternatively, the same transformation could be per-
formed under much milder (neutral) conditions employ-
ing a chemoselective biocatalyst with nitrile-hydrolyzing
activity (Rhodococcus R312).22 The carboxylate function
of (S)-7 was reduced with LiAlH4 in refluxing THF to
give diol (S)-8 in 93% isolated yield. Subsequent acetyla-
tion employing neat acetic anhydride and a catalytic
amount of 4-dimethylaminopyridine (DMAP) afforded
quantitatively the diacetate (S)-9. From literature it be-
came clear that the oxidation of the aryl moiety to a car-
boxylate using ruthenium tetroxide8a,23 was to be prefered
over ozonolysis23b (poor yields). However, in our hands
the excellent yields under the conditions reported
elsewhere8a could not be reproduced. After closely moni-
toring the course of the reaction it was found that the for-
mation of (R)-10 was complete within 5 minutes after the
This represents the first application of a bacterial epoxide
hydrolase in a large-scale total synthesis. The key inter-
mediate was formed in ca. 10 g in an enantioconvergent
way via simple combination of bio- and acid catalysis. As
such, this procedure provides a new tool which can be
used in the syntheses of chiral intermediates and/or opti-
cally active natural products.
Melting points are uncorrected. 1H and 13C NMR spectra were record-
ed in CDCl3 , unless otherwise noted, on a Bruker MSL 300 at 300
and 75.47 MHz, respectively. Chemical shifts are relative to TMS (δ
= 0.00) with CHCl3 as an internal standard [δ = 7.23,1H and 76.90,
13C]. 13C NMR multiplicities were determined by using a DEPT pulse
sequence. MS were recorded at 70 eV via DI-EI on a KRATOS-pro-
file spectrometer. FT-IR spectra were recorded on a Bomem-Michel-
son M100 spectrometer as a neat film on a NaCl disc. Optical rotation
values were measured on a Perkin-Elmer 341 polarimeter at 589 nm
(Na line) in a 1 dm cuvette at 20°C. The reaction course was routinely
monitored by TLC (Merck 60 F254) and the products were visualized

SYNTHESIS
September 1998
1261
by spraying with vanillin/H2SO4, Mo-reagent, or KMnO4-reagent.
Carboxylic acids were visualized using a bromocresol green/brom-
ophenol blue/KMnO4 spray.24 Alternatively, reactions were followed
by GC analysis carried out on a Shimadzu GC-14A equipped with
FID and a RSL 1701 capillary column (30m, 0.25 mm, 0,25 µm film,
N2). Enantiomeric excesses were analyzed on the same gas chro-
matograph using a CP-Chirasil-DEX CB column (25m, 0.32 mm,
0.25 µm film, H2). Flash chromatography was performed on silica gel
Merck 60 (230–400 mesh). Petroleum ether had a boiling range of
60–90°C.
the combined organic layers were dried and evaporated. Flash chro-
matography (petroleum ether/EtOAc, 1:2) gave 6 as a white solid;
yield: 5.01 g (~100%); mp 101–102°C (hexane/EtOAc, 1:9); [α]D
+8.3 (c = 0.5, 90% EtOH).
IR: ν = 3436, 2977, 2922, 2254, 1493, 1450, 1112, 744, 702 cm–1.
1H NMR: δ = 1.38 (s, 3 H), 2.19 (s, 1 H), 2.47 (s, 2 H), 2.92 (s, 2 H),
7.22–7.39 (m, 5 H).
13C NMR: δ = 26.94 (CH3), 30.65 (CH2), 47.28 (CH2), 71.21 (C),
117.78 (C), 127.33 (CH), 128.73 (2 × CH), 130.41 (2 × CH), 135.75
(C).
MS: m/z (%) = 175 (M+, 4), 160 (2), 142 (8), 135 (18), 91 (100), 84
(9), 65 (22), 43 (31).
Solvents were dried and freshly distilled by standard techniques. For
anhydrous reactions, flasks were dried overnight at 150°C and
flushed with dry argon just before use, and reactions were carried out
under argon. Organic extracts were dried over Na2SO4, and then the
solvent was evaporated under reduced pressure. The bacterial strain
of Nocardia EH1 was a gift from J. de Bont (Wageningen, The Neth-
erlands) and C. Syldatk (Stuttgart, Germany), and was grown as pre-
viously described.17
HRMS: m/z calc. for C11H13NO (M+) 175.09971, found 175.09909.
(S)-(−)-3-Hydroxy-3-methyl-4-phenylbutanoic acid (7):
Nitrile 6 was hydrolyzed with basic H2O2 (Method A) or with Rhodo-
coccus R312 (Method B):
Method A: A mixture of NaOH (100 mL, 3 M) and H2O2 (40 mL,
30%) was added to the hydroxy nitrile 6 (3.50 g, 18.0 mmol). The
reaction mixture was heated at 65°C for 1 h and then refluxed for an
additional hour, after which the temperature of the mixture was al-
lowed to lower to r.t. and then further cooled to 0°C. HCl (6 M) was
then added until a pH < 3 was reached. The resulting suspension was
extracted with Et2O (5 × 25 mL) and the combined Et2O layers were
dried and concentrated to afford pure 7 as a strong smelling oil which
solidified below 20°C; yield: 3.36 g (96%); [α]D −2.4 (c = 5, abs
EtOH).
(S)-(−)-2-Methyl-3-phenylpropane-1,2-diol (4):
Epoxide (±)-3 (10.0 g, 67.7 mmol) was added to a suspension of re-
hydrated lyophilized whole cells of Nocardia EH1 (8.00 g)25 in Tris-
buffer (250 mL, 0.05 M, pH 7.5). The mixture was agitated at 35°C
and 125 rpm. At exactly 50% conversion (48 h) the mixture was con-
tinuously extracted with CH2Cl2 (500 mL, 36 h). The organic layer
was washed with brine (100 mL), dried, and evaporated. To a stirred
solution of the resulting bright orange oil (12.3 g) in dioxane
(225 mL) was added dropwise 93% aq H2SO4 (18 mL) at 0°C. The
reaction mixture was stirred for 15 min at r.t. after which the acid was
neutralized with satd aq NaHCO3 solution. EtOAc (300 mL) was add-
ed, and the resulting biphasic mixture was stirred vigorously for an
additional 30 min. The aqueous layer was extracted with EtOAc (3 ×
100 mL) and the combined organic layers were dried and concentrat-
ed. Flash chromatography of the residue (petroleum ether/EtOAc,
2:1) afforded 4 as white crystals; yield: 10.5 g (94%); ee 94%. Spec-
troscopic, physical and optical data of 4 were in full agreement with
those previously reported.14b,17
IR: ν = 3605−2885, 1709, 1600, 1494, 1452, 1214, 744, 702 cm–1.
1H NMR (acetone-d6): δ = 1.25 (s, 3 H), 1.41 (s, 1 H), 2.45 (s, 2 H),
2.89 (s, 2 H), 7.17–7.31 (m, 5 H), 7.51–7.94 (br s, 1 H).
13C NMR (acetone-d6): δ = 27.44 (CH3), 45.22 (CH2), 48.49 (CH2),
72.01 (C), 127.22 (CH), 128.81 (2 × CH), 131.68 (2 × CH), 138.75
(C), 174.35 (C).
MS: m/z (%) = 194 (M+, 0.5), 143 (3), 135 (12), 103 (51), 92 (100),
85 (43), 65 (22), 43 (63).
HRMS: m/z calc. for C11H14O3 (M+) 194.09429, found 194.09429.
Method B: Nitrile 6 (1.00 g, 5.72 mmol) was added to a suspension of
rehydrated lyophilized whole cells of Rhodococcus R31222 (1.25 g) in
Tris-buffer (25 mL, 0.05 M, pH 7.5). The mixture was agitated at
30°C and 120 rpm until the starting material was consumed (24 h).
Then the pH was adjusted to just below 3 with HCl (6 M) and the
remaining cell material was precipitated by centrifugation at 10000 g
for 15 min. The aqueous suspension was decanted and extracted with
Et2O (4 × 20 mL). The combined organic phases were dried and evap-
orated. The resulting orange oil was chromatographed (EtOAc) on a
silica column treated with AcOH (5%) to give pure 7; yield: 0.74 g
(67%); [α]D −2.2 (c = 5, abs EtOH).
(S)-(+)-2-Methyl-3-phenylpropane-1,2-diol 1-(p-Toluenesul-
fonate) (5):
To a stirred solution of diol 4 (5.00 g, 30.1 mmol) in pyridine
(250 mL) was added TsCl (9.00 g, 47.0 mmol). The solution was
stirred for 18 h at r.t. and concentrated. Aq CuSO4 solution (20%
w/v, 150 mL) and EtOAc (200 mL) were then added and the mixture
was stirred for 30 min at r.t. The aqueous layer was extracted with
EtOAc (3 × 75 mL) and the combined organic phases were washed
with brine (50 mL), dried and evaporated. Crystallization (diisoprop-
yl ether) of the crude material gave 5; yield: 9.15 g (95%); mp 134°C
(dec); [α]D +9.2 (c = 5, CHCl3).
IR: ν = 3497, 3029, 2956, 1596, 1361, 973, 824, 737, 702 cm–1.
1H NMR: δ = 1.10 (s, 3 H), 2.01 (s, 1 H), 2.45 (s, 3 H), 2.75 (d, 1 H,
J = 13.5 Hz), 2.83 (d, 1 H, J = 13.5 Hz), 3.81 (s, 2 H), 7.10-7.28 (m,
5 H), 7.36 (d, 2 H, J = 8.6 Hz) 7.82 (d, 2 H, J = 8.2 Hz).
13C NMR: δ = 21.73 (CH3), 23.80 (CH2), 44.49 (CH3), 71.42 (CH2),
75.06 (C), 126.92 (CH), 128.09 (2 × CH), 128.41 (2 × CH), 130.04
(2 × CH), 130.45 (2 × CH), 132.32 (C), 135.90 (C), 145.17 (C).
MS: m/z (%) = 229 (M+–91, 66), 155 (77), 135 (22), 91 (100), 65 (18),
43 (21).
(S)-(−)-3-Methyl-4-phenyl-1,3-butanediol (8):
A solution of acid 7 (3.00 g, 15.5 mmol) in anhyd THF (75 mL) was
added slowly to a refluxing suspension of LiAlH4 (0.75 g, 19.7 mmol)
in anhyd THF (15 mL). The mixture was refluxed for 1 h, then cooled
to 0°C. Excess LiAlH4 was quenched with satd aq NH4Cl solution
(20 mL), the salts were dissolved by adding HCl (5 mL, 6 M) and the
aqueous layer was extracted with EtOAc (6 × 25 mL). The combined
organic extracts were washed with satd aq NaHCO3 (25 mL), dried
and concentrated. Flash chromatography (petroleum ether/EtOAc,
gradient from 3:1 to 1:1) gave 8 as a clear oil; yield: 2.59 g (93%);
[α]D −2.0 (c = 5, abs EtOH).
HRMS: m/z calc. for C10H13SO4 (M+–91) 229.05346, found
229.05294.
IR: ν = 3476–3027, 2937, 1602, 1493, 1452, 1113, 1054, 761,
701 cm–1.
(S)-(+)-3-Hydroxy-3-methyl-4-phenylbutanonitrile (6):
1H NMR: δ = 1.20 (s, 3 H), 1.61–1.89 (s, 2 H), 2.75 (d, 1 H, J =
13.1 Hz), 2.86 (d, 1 H, J = 13.1 Hz), 2.96 (br s, 1 H), 3.43 (br s, 1 H),
3.88 (dd, 2 H, J = 5.3, 6.3 Hz), 7.18–7.31 (m, 5 H).
The tosylate 5 (9.15 g, 28.6 mmol) was dissolved in a mixture of
EtOH/H2O (60% v/v, 150 mL) and stirred at 0°C. KCN (4.00 g,
61.5 mmol) was added and the solution was allowed to warm to r.t.
After stirring for 18 h the solution was concentrated, brine (100 mL)
was added, the mixture was extracted with CH2Cl2 (5 × 50 mL), and
13C NMR: δ = 26.47 (CH3), 41.68 (CH2), 48.93 (CH2), 59.65 (CH2),
73.76 (C), 126.58 (CH), 128.23 (2 × CH), 130.68 (2 × CH), 137.29 (C).

SYNTHESIS
Papers
1262
MS: m/z (%) = 162 (M+–18, 4), 135 (26), 117 (12), 91 (62), 71 (32),
65 (17), 43 (100).
13C NMR: δ = 29.74 (CH3), 35.89 (CH2), 44.72 (CH2), 66.14 (CH2),
68.19 (C), 170.42 (C).
HRMS: m/z calc. for C11H14O (M+-18) 175.09971, found 175.09909.
MS: m/z (%) = 130 (M+, 4), 115 (6), 85 (10), 71 (92), 58 (51), 43
(100).
HRMS: m/z calc. for C6H10O3 (M+) 130.06299, found 130.06268.
(S)-(−)-1,3-Diacetoxy-3-methyl-4-phenylbutane (9):
Ac2O (100 mL) containing DMAP (0.75 g) was added to diol 8
(2.50 g, 13.8 mmol). The stirred solution was heated at 95°C and after
the reaction was complete (4 h) poured into ice water (250 mL). The
mixture was stirred for another 2 h and then extracted with CH2Cl2
(5 × 50 mL). The combined organic extracts were washed with satd
aq NaHCO3 solution (3 × 50 mL) and brine (50 mL), dried, and con-
centrated (eventually at 60°C/0.09 Torr). The resulting strong smell-
ing brown oil was flash chromatographed (petroleum ether/EtOAc,
3:1) to give 9 as a clear viscous oil; yield: 3.67 g (~100%); [α]D −8.5
(c = 5, abs EtOH).
The authors would like to express their thanks to A. de Raadt and
D.V. Johnson for critically proofreading the manuscript, H.J. Weber
and C. Illaszewicz for recording the NMR spectra, and C. Mirtl for
measuring the MS spectra. This research was performed within the
Spezialforschungsbereich Biokatalyse (SFB-A4) and was financed by
the Fonds zur Förderung der wissenschaflichen Forschung (Austrian
Ministry of Science, Vienna F104) and the European Commission
(BIO4-CT-0005).
IR: ν = 2941, 1734, 1451, 1368, 1240, 1031, 734, 703 cm–1.
1H NMR: δ = 1.44 (s, 3 H), 2.00 (s, 3 H), 2.05 (s, 3 H), 2.17-2.39 (m,
2 H), 3.06 (d, 1 H, J = 13.7 Hz), 3.23 (d, 1 H, J = 13.7 Hz), 4.20 (t, 2
H, J = 7.0 Hz), 7.15-7.36 (m, 5 H).
(1) Stryer, L. Biochemistry; W.H. Freeman and Company: New
York, 1981, Chap. 20, p 465.
(2) Brown, M. S.; Goldstein, J. L. J. Lipid Res. 1980, 21, 505.
(3) (a) Yamamoto, A. V.; Sudo, H.; Endo, A. Athero. 1980, 35, 259.
(b) Spencer, T. A.; Clark, D. S.; Johnson, G. A.; Erickson, S. K.;
Curtiss, L. K. Bioorg. Med. Chem. 1997, 5, 873, and references
cited therein.
(4) Wolf, D. E.; Hoffman, C. H.; Aldrich, P. E.; Skeggs, H. R.;
Wright, L. D.; Folkers, K. J. Am. Chem. Soc. 1956, 78, 4499.
(5) Eberle, M.; Arigoni, D. Helv. Chim. Acta 1960, 43, 1508.
(6) Cornforth, R. H.; Cornforth, J. W.; Popjak, G. Tetrahedron,
1962, 18, 1351.
13C NMR: δ = 21.05 (CH3), 22.49 (CH3), 23.92 (CH3), 36.71 (CH2),
44.68 (CH2), 60.63 (CH2), 82.79 (C), 126.71 (CH), 128.17 (2 × CH),
130.61 (2 × CH), 136.60 (C), 170.66 (C), 171.08 (C).
MS: m/z (%) = 205 (M+–59, 11), 173 (10), 144 (98), 129 (61), 91 (62),
71 (50), 43 (100).
HRMS: m/z calc. for C8H13O4 (M+–91) 173.08138, found 173.08148
and calc. for C11H12 (M+–120) 144.09390, found 144.09235.
(R)-(−)-3,5-Diacetoxy-3-methylbutanoic Acid (10):
(7) (a) Bonadies, F.; Rossi, G.; Bonini, C. Tetrahedron Lett. 1984,
25, 5431.
The diacetate 9 (2.20 g, 8.33 mmol) was dissolved in a mixture of
MeCN (50 mL), CCl4 (50 mL) and distilled water (100 mL). The vig-
orously stirred mixture was heated to reflux temperature. Subsequent-
ly, NaIO4 (56.0 g, 260 mmol) and fresh RuCl3 hydrate (0.25 g,
1.20 mmol)26 were added. The colour changed from black via orange
to yellow within 5 min, after which the heterogenous mixture was
poured into a mixture of CH2Cl2 (150 mL) and ice water (150 mL).
After stirring for 30 min, the pH of the mixture was adjusted to 10
with aq NaOH (3 M) and the mixture was extracted with CH2Cl2
(2 × 50 mL). The aqueous layer was acidified with concd HCl (pH <
3) and extracted with EtOAc (10 × 50 mL). The combined EtOAc ex-
tracts were dried and evaporated to give pure 10 as a strong smelling
yellow oil; yield: 1.39 g (72%); [α]D −3.6 (c = 1, abs EtOH).
IR: ν = 3675−2685, 1734, 1722, 1430, 1370, 1243, 1026 cm–1.
1H NMR: δ = 1.59 (s, 3 H), 2.01 (s, 3 H), 2.04 (s, 3 H), 2.08–2.22 (m,
1 H), 2.32–2.46 (m, 1 H), 2.95 (d, 1 H, J = 15.0 Hz), 3.12 (d, 1 H, J =
15.1 Hz), 4.15–4.23 (m, 2 H), 8.11−8.94 (br s, 1 H).
(b) Mori, K.; Okada, K. Tetrahedron 1985, 41, 557.
(c) Schneider, J. A.; Yoshihara, K. J. Org. Chem. 1986, 51,
1077.
(d) Ray, N. C.; Raveendranath, P. C.; Spencer, T. A. Tetrahe-
dron 1992, 48, 9427.
(8) (a) Frye, S. V.; Eliel, E. L. J. Org. Chem. 1985, 50, 3402.
(b) Mash, E. A.; Arterburn, J. B. J. Org. Chem. 1991, 56, 885.
(9) Kishida, M.; Yamauchi, N.; Sawada, K.; Ohashi, Y.; Eguchi, T.;
Kakinuma, K. J. Chem. Soc.,Perkin Trans 1 1997, 891.
(10) Koike, A.; Murahawa, S.; Endo, A. J. Ferment. Technol. 1989,
68, 58.
(11) (a) Sugai, T.; Kakeya, H.; Ohta, H. Tetrahedron 1990, 46, 3463.
(b) Ferraboschi, P.; Grisenti, P.; Casati, S.; Santaniello, E. Syn-
lett 1994, 754.
(c) Mizuguchi, E.; Suzuki, T.; Achiwa, K. Synlett 1996, 743.
(12) Huang, F.-C.; Lee, L. F. H.; Mittal, R. S. D.; Ravikumar, P. R.;
Chan, J. A.; Sih, C. J.; Caspi, E.; Eck, C. R. J. Am. Chem. Soc.
1975, 97, 4144.
(13) Lakner, F. J.; Hager, L. P. J. Org. Chem. 1996, 61, 3923.
(14) (a) Orru, R. V. A.; Kroutil, W.; Faber, K. Tetrahedron Lett.
1997, 38, 1753.
13C NMR: δ = 20.99 (CH3), 22.23 (CH3), 24.43 (CH3), 37.03 (CH2),
42.54 (CH2), 60.22 (CH2), 79.90 (C), 170.62 (C), 171.14 (C), 175.57
(C).
MS: m/z (%) = 217 (M+–15, 0.5), 173 (2), 155 (4), 145 (7), 112 (28),
87 (12), 71 (20), 43 (100).
HRMS: m/z calc. for C8H13O4 (M+–59) 173.08138, found 173.08102
and calc. for C6H9O4 (M+–87) 145.05008, found 145.05008.
(b) Orru, R. V. A.; Mayer, S. F.; Kroutil, W. Faber, K. Tetrahe-
dron 1998, 54, 859.
(15) For recent reviews see: (a) Faber, K.; Mischitz, M.; Kroutil, W.
Acta Chem. Scand. 1996, 50, 249.
(R)-(−)-Mevalonolactone (2):
(b) Archer, I. V. J. Tetrahedron 1997, 53, 15617.
(16) Mischitz, M.; Mirtl, C.; Saf, R.; Faber, K. Tetrahedron: Asym-
metry 1996, 7, 2041.
(17) Osprian, I.; Kroutil, M.; Mischitz, M.; Faber, K. Tetrahedron:
Asymmetry 1997, 8, 65.
(18) Klibanov, A. M. Acc. Chem. Res. 1990, 23, 114.
(19) Epoxide (±)-3 (15 mg) was added to a suspension of rehydrated
lyophilized whole cells of Nocardia EH1 (15 mg) in Tris-buffer
(0.05 M, pH 7.5, 1 mL) containing an organic cosolvent in
amounts of 1, 5, 10, and 20% (v/v), respectively. The mixture
was agitated at 25°C with 125 rpm. The following organic sol-
vents were tested: cyclohexane, benzene, diisopropyl ether,
To a stirred solution of 10 (1.30 g, 5.60 mmol) in MeOH (125 mL)
was added K2CO3 (5.00 g, 35.0 mmol). The mixture was refluxed
overnight, then the MeOH was evaporated and H2O (200 mL) was
added. The resulting solution was cooled to 0°C, acidified to pH 1
with concd HCl, and continuously extracted with CHCl3 (250 mL,
18 h). The organic phase was dried, evaporated and flash chromato-
graphed (EtOAc) to give pure 2; yield: 0.71 g (97%); ee > 99%; [α]D
−23.7 (c = 4, abs EtOH) [Lit.6 [α]D −23 (c = 6, abs EtOH)].
IR: ν = 3392, 2944, 1717, 1396, 1252, 1126, 1069 cm–1.
1H NMR: δ = 1.38 (s, 3 H), 1.87–1.94 (m, 2 H), 2.38−2.64 (br s, 1 H),
2.50 (d, 1 H, J = 17.8 Hz), 2.66 (d, 1 H, J = 17.6 Hz), 4.28–4.39 (m,
1 H), 4.53–4.66 (m, 1 H).

SYNTHESIS
September 1998
EtOAc, t-BuOH, THF, dioxane, isopropyl alcohol, acetone,
1263
(b) Chakraborti, A. K.; Ghatak, U. R. Synthesis 1983, 746.
(c) Gosh, S.; Ghatak, U. R. Tetrahedron 1992, 48, 7289.
(24) Pásková, J.; Munck, V. J.; J. Chromatogr. 1960, 4, 241.
(25) Large amounts of lyophilized whole cells of this biocatalyst
have become available by easy fermentation (70 g from 700 L),
see: Kroutil, W.; Genzel, Y.; Pietsch, M.; Syldatk, C.; Faber, K.
J. Biotechnol. 1998, in press.
(26) The quality of the RuCl3 hydrate is critical. The reaction was
performed immediately after the reagent was received from
Fluka.
acetonitrile, DMF, DMSO, MeOH, and ethylene glycol.
(20) Interestingly, the same reaction performed on an analytical
scale showed only modest selectivity (E = 42), see also 17.
(21) Wolff, A.; Zhu, L.; Wong, Y. W.; Straathof, A. J. J.; Jongejan,
J. A.; Heijnen, J. J. Biotechnol. Bioeng. 1998, in press.
(22) Formerly known as Brevibacterium R312. A 10 L fermentation
procedure which gives easy access to large amounts of this bac-
terium has been developed by us and will be published else-
where.
(23) (a) Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B.
J. Org. Chem. 1981, 46, 3936.