Chemo-enzymatic synthesis of all isomers of 2-methylbutane-1,2,3,4-tetraol - Important contributors to atmospheric aerosols

Article abstract of DOI:10.1002/ejoc.200600873

By a combination of stereospecific osmium catalyzed oxidation of dimethyl citraconate and lipase catalysed enantioselective resolution of the formed dimethyl (2R*,3S*)-2,3-dihydroxy-2-methylbutanedioate, followed by reduction, (2R,3S)- and (2S,3R)-2-methylbutane-1,2,3,4-tetraol were isolated. Similar reactions starting with dimethyl mesaconate gave the isomers, (2R,3R)- and (2S,3S)-2-methylbutane-1,2,3,4-tetraol. Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

Full text of DOI:10.1002/ejoc.200600873

FULL PAPER
DOI: 10.1002/ejoc.200600873
Chemo-Enzymatic Synthesis of All Isomers of 2-Methylbutane-1,2,3,4-tetraol –
Important Contributors to Atmospheric Aerosols
Anders Riise Moen,[a] Kjersti Ruud,[a] and Thorleif Anthonsen*[a]
Keywords: Chemo-enzymatic synthesis / Tetraols / Isoprenoids / Carbohydrates
By a combination of stereospecific osmium catalyzed oxi-
isolated. Similar reactions starting with dimethyl mesaconate
dation of dimethyl citraconate and lipase catalysed enantio- gave the isomers, (2R,3R)- and (2S,3S)-2-methylbutane-
selective resolution of the formed dimethyl (2R*,3S*)-2,3-
dihydroxy-2-methylbutanedioate, followed by reduction,
(2R,3S)- and (2S,3R)-2-methylbutane-1,2,3,4-tetraol were
1,2,3,4-tetraol.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2007)
absolute configurations of the Amazonian tetritols have so
far not been possible to deduce. In spite of the fact that
they are claimed to be present in enormous quantities, they
were only detected by mass spectroscopy. These natural
products may be considered as carbohydrates and thus can
be named as suggested above (see Figure 1), but also as
isoprenoids. Synthesis of racemic 2-C-methylerythritol and
2-C-methylthreitol from dimethyl citraconate and dimethyl
mesaconate has been described previously.[1,5] The biosyn-
thesis of 1 from glyceraldehyde-3-phosphate and pyruvate
giving a mevalonate independent origin of isoprenoids, has
been discovered in bacteria and higher plants.[6] Since this
discovery, there is considerable interest in the synthesis of
1.[7–10]
1. Introduction
Thirty years ago 2-C-methyl--erythritol [(2S,3R)-1] was
isolated from Convolvulus glomeratus, a plant growing in
Pakistan.[1] The structure was deduced by NMR spec-
troscopy, synthesis of racemic derivatives of 2-C-methyl-
erythritol and 2-C-methyl-threitol and CD.[2] Finally the
enantiopure natural compound was synthesized by first
generation asymmetric synthesis starting with 2,3-O-isopro-
pylidene--glyceraldehyde.[3] Due to the presence of two ste-
reocenters there are four possible stereoisomers as shown in
Figure 1.
2. Results and Discussion
Dimethyl (2R*,3S*)-2,3-dihydroxy-2-methylbutanedioate
[(2R*,3S*)-2] was obtained from dimethylcitraconate [(Z)-
3] by stereospecific dihydroxylation (Scheme 1). The cis-di-
hydroxylations were performed using various osmium cata-
lysts in an “Upjohn” process with 4-methylmorpholine N-
oxide for reoxidation.[11] The best results were obtained in
acidic media with citric acid as additive which gave 85–95%
yield.[12] Oxidation with K2OsO2(OH)4 or osmium immobi-
lized in mesoporous materials[13] gave similar yields. The
latter can be easily separated from the product and reused.
Oxidations using potassium permanganate gave poor re-
sults. The resulting racemic dimethyl esters [(2R*,3S*)-2]
were resolved kinetically using lipase A from Candida ant-
arctica (CAL-A, Novozym 735) and vinyl butanoate to give
the dimethyl ester [(2S,3R)-2] and the monobutanoate
(2R,3S)-4. The enantioselectivity, the (E value, was low even
at 6 °C (E = 27) (Table 1) hence the resolution was per-
formed twice. Firstly the reaction was stopped at 67% con-
version and the remaining substrate, (2S,3R)-2, was isolated
Figure 1. The four stereoisomers of 2-methylbutane-1,2,3,4-tetraol
(1).
More recently it has been discovered that large amounts
of 2-C-methyltetritols are present in the air above the Ama-
zonian rain forest claimed to be due to photooxidation of
isoprene and moreover, that such compounds are hitherto
unknown.[4] Inspired by this statement we started to synthe-
size all four isomers of 2-methylbutane-1,2,3,4-tetraol.
Knowledge of the absolute configurations of the Ama-
zonian tetritols would indicate whether they were synthe-
sized as suggested, by achiral photooxidation, or by asym-
metric enzymatic catalysis in the plants, to give enantiopure
tetraols such as the Convolvulus compound. However, the
[a] Department of Chemistry, Norwegian University of Science
and Technology (NTNU),
7491 Trondheim, Norway
E-mail: Thorleif.Anthonsen@chem.ntnu.no
1262
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1262–1266

Important Contributors to Atmospheric Aerosols
FULL PAPER
Scheme 1.
Table 1. Resolutions of dimethyl (2R*,3S*)-2,3-dihydroxy-2-methylbutanedioate [(2R*,3S*)-2] and [(2R*,3R*)-2] catalyzed by Novozym
735 (immobilized lipase A from Candida antarctica) using vinyl butanoate as acyl donor in toluene at 6 °C. Two different reactions were
performed with (2R*,3S*)-2. One was stopped at 67% conversion to favour substrate enantiopurity (1st reaction) and one at 19%
conversion to obtain high product enantiopurity (2nd reaction), see also Figure 2. Optical rotations were measured in CH2Cl2.
E
c,%
eep,%
[α]2D0
ees,%
[α]2D0
(2R*,3S*)-2
(1st reaction)
(2R*,3S*)-2
(2nd reaction)
(2R*,3R*)-2
27
67
–
–
98.5
(2S,3R)-2
–
–22.7
c = 5.3
–
27
19
51
94
–16.0
c = 3.5
+8.3
(2R,3S)-4
94
(2S,3S)-4
272
Ͼ99.0
(2R,3R)-2
–23.6
c = 1.0
c = 1.0
with 98.5% ee. In a second reaction, the conversion was
stopped at 19% and the produced butanoate (2R,3S)-4 was
isolated with 94% ee. (Figure 2) Reduction of the unreacted
dimethyl ester (2S,3R)-2 gave the natural compound
(2R,3S)-1. The 3-butanoate, (2R,3S)-4, was reduced to give
the enantiomer (2S,3R)-1.
The two other stereoisomers were isolated by a similar
synthesis starting with dimethyl mesaconate [(E)-3]. Oxi-
dation gave a mixture of the enantiomers dimethyl
(2R*,3R*)-2,3-dihydroxy-2-methylbutanedioate [(2R*,3R*)-2]
which were resolved with an E value of 140 at 30 °C and
272 at 6 °C. In one single reaction both the remaining
(2R,3R)-2 and the butanoate (2S,3S)-4 were isolated with
high ee values. (Table 1) However, at the lower temperature
the reaction was much slower. To reach 50% conversion it
took 6 h at 30 °C and 3 days at 6 °C. Reductions of the
separated products gave (2S,3S)-1 and (2R,3R)-1 respec-
tively. In both esterifications only the secondary center was
esterified since tertiary centers are less reactive. Moreover,
the faster reacting enantiomer is expected to have the (3S)-
configuration. In particular when the (E value is high, this
is taken as proof of the absolute configurations based on
the knowledge of enantioselectivity of lipases.[14,15] Further-
more, based on the stereospecific reactions used to produce
the dihydroxy compounds, also the configurations at C-2
was consequently proven. In addition, Sakamoto et al. has
synthesized 2-C-methyl--threitol starting with -galac-
Figure 2. The enantiomeric excesses of the product fraction (I) and
the substrate fraction (II) in CAL-A catalyzed resolution at 6 °C
of
dimethyl
(2R*,3S*)-2,3-dihydroxy-2-methylbutanedioate
[(2R*,3S*)-2] as a function of the degree of conversion. The E value
is 27. Due to the relatively low E value, the product (2R,3S)-4 was
isolated after 19% conversion and the unreacted substrate after
67% conversion in two different reactions.
tose.[7] The optical rotation of 2S,3S-1 is almost identical shown in Figure 3 may be used to elucidate this. The E
with this product. value is the ratio of the rate constants for reactions with the
In spite of the similarity of the two racemic substrates, two enantiomers. Consequently, when the E value goes
they only differ in relative configuration, the E values differ down when resolution of 2R*,3R* is compared with the
a lot. Progress curves of conversion of each enantiomer resolution of 2R*,2S*, the question is: Is it the faster
Eur. J. Org. Chem. 2007, 1262–1266
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1263

A. R. Moen, K. Ruud, T. Anthonsen
FULL PAPER
rel) was from Novozymes, Denmark. The transesterification reac-
tions were performed at 30 °C in an Infors MINITRON Shaker
Incubator. The product and the remaining substrate were separated
by chromatography with Versaflash system from Supelco with ver-
saPak silica cartridge 40ϫ150 mm or 40ϫ75 mm. Optical rotations
were determined using a Perkin–Elmer 243B Polarimeter, concen-
trations are given in g/100 mL.
enantiomer that slows down or is it the slower enantiomer
that speeds up? In the present case it may be seen that the
faster reacting enantiomers are almost equally fast, but the
slow enantiomer 2S,3R has become relatively faster than
the other slow enantiomer 2R,3R in the resolution of the
other diastereomer. This may be explained in detail by mo-
lecular modeling of the interaction between the substrates
and CAL-A. However, the X-ray structure of this lipase has
not yet been published. Previously we have experienced that
CAL-A works better with more bulky substrates because it
has a wider active site than CAL-B. See ref.[16] and refer-
ences sited therein. For instance, CAL-B gave high E values
in resolution of 1-[4-(benzyloxy)-3-methoxyphenyl]ethanol,
but not for the corresponding more bulky 2-bromo and
2-chloro derivatives, then CAL-A was more efficient.
4.1.1. Analyses: Chiral analyses of the transesterification reactions
were performed on Varian 3400 gas chromatograph equipped with
a chiral CP Chirasil Dex CB column (25 m, 0.25 mm id and
0.32 µm film thickness), column pressure 8 psi, split flow 60 mL/
min.
Temperature program for analysis of products resulting from (E)-
3: 110–140 °C 4 °C/min, 140–180 °C 10 °C/min. Retention times of
the enantiomers of the alcohol were 6.4 min for (2R,3R)-2 and
7.5 min for (2S,3S)-2, and 9.3 min for the butanoate (2R,3R)-4,
9.4 min for (2S,3S)-4.
Temperature program for analysis of products resulting from (Z)-
3: 90 °C (10 min hold), 90–138 °C 2 °C/min, 138–180 °C 10 °C/min.
Retention times of the enantiomers of the alcohol were 15.6 min for
(2S,3R)-2 and 16.0 min for (2R,3S)-2, 23.1 min for the butanoate
(2R,3S)-4 and 23.4 min for (2S,3R)-4.
NMR spectra were recorded in CDCl3 solutions using Bruker DPX
300 and 400 instruments. Chemical shifts are given in ppm relative
to TMS and coupling constants in Hz. MS was performed by direct
infusion with a infusion pump for continues injection to a Agilent
LC-MSD Trap SL with positive APCI ionization. Enantiomeric
ratios (E) were calculated using the computer program E and K
calculator version 2.03.
4.1.2. Chemical Vapor Deposition of Os3(CO)12 on Al-MCM-41: Al-
MCM-41 was prepared as reported.[12] A glass ampoule containing
Al-MCM-41 (0.4 g) and Os3(CO)12 (50 mg, 0.055 mmol) was evac-
uated (P = 4·10–2 mbar), sealed and heated at 150 °C for 40 h be-
fore the ampoule was cooled to room temp. The ampoule was
broken, the powder was washed with water and methanol and cen-
trifuged. The dried powder (25 mg) was mixed in water (50 mL)
and stirred for 24 h. ICP-MS analysis of the filtrate showed that
leaching of Os from the matrix was less than 5 ppm.
Figure 3. Progress curves of conversion of each enantiomer at 30 °C
of (2R*,3S*)-2 and (2R*,3R*)-2. The E values were 20 and 140 as
depicted in the graphs. When the reactions were performed at 6 °C
the E values increased to 27 and 272, respectively.
4.2.
(2R*,3R*)-Dimethyl
2,3-Dihydroxy-2-methylbutanedioate
[(2R*,3R*)-2]: Dimethyl mesaconate [(E)-3] (0.5 g, 3.16 mmol) and
citric acid (0.46 g, 2.4 mmol) were dissolved in 5 mL of a 1:1 mix-
ture of tert-butyl alcohol/water. Potassium osmate, K2OsO2(OH)4,
(0.003 g, 0.26 mol-%) was added followed by 4-methylmorpholine
N-oxide (0.4 g, 3.4 mmol). The reaction was stirred at room temp
for 15 h before Na2SO3 (0.3 g) was added. The reaction mixture
was further stirred for 1 h and tert-butyl alcohol was evaporated
3. Conclusion
Combined stereospecific osmium-catalyzed oxidation of
dimethyl citraconate and lipase-catalysed enantioselective
resolution of the formed dimethyl (2R*,3S*)-2,3-dihydroxy- before the aqueous residue was extracted with Et2O (4ϫ10 mL).
The combined organic extracts were dried with MgSO4 and con-
centrated to give (2R*,3R*)-2 (0.58 g, 3.0 mmol, 94.9%) as an oil.
1H NMR: δ = 4.35 (s, 1 H), 3.83 and 3.82 (both s, 3 H, 2ϫOCH3),
1.49 (s, 3 H, CH3). 13C NMR: δ = 174.8, 171.6, 76.9, 75.2, 53.3,
52.8, 21.8. MS: 193.1 (M+1), 133.2 (100%), 87.
2-methylbutanedioate, followed by reduction, yields
(2R,3S)- and (2S,3R)-2-methylbutane-1,2,3,4-tetraol. The
enantioselectivity of the resolution is low (E = 27 at 6 °C).
Similar reactions starting with dimethyl mesaconate give
the isomers (2R,3R)- and (2S,3S)-2-methylbutane-1,2,3,4-
tetraol. In this case the E value is 272.
4.3.
(2R*,3S*)-Dimethyl
2,3-Dihydroxy-2-methylbutanedioate
[(2R*,3S*)-2]: Dimethyl citraconate [(Z)-3] (0.5 g, 3.16 mmol) and
citric acid (0.46 g, 2.4 mmol) were dissolved in 5 mL of a 1:1 mix-
ture of tert-butyl alcohol/water. Potassium osmate, K2OsO2(OH)4,
(0.003 g, 0.26 mol-%) was added followed by 4-methylmorpholine
N-oxide (0.4 g, 3.4 mmol). The reaction was stirred at room temp.
for 15 h before Na2SO3 (0.3 g) was added. The reaction mixture
was further stirred for 1 h. tert-Butyl alcohol was evaporated before
the aqueous residue was extracted with Et2O (4ϫ10 mL). The
4. Experimental Section
4.1. General: Osmium tetraoxide on poly(4-vinylpyridine), potas-
sium osmate dihydrate and tri-osmium dodecacarbonyl were pur-
chased from Sigma–Aldrich, Merck, and Fluka, respectively. Li-
pase A from Candida antarctica (CAL-A, Novozym 735 on Accu-
1264
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1262–1266

Important Contributors to Atmospheric Aerosols
FULL PAPER
combined organic extracts were dried with MgSO4 and concen-
stirred at room temp. for 6 h before it was poured into water
(25 mL). Solid material was removed by centrifugation. The solu-
tion was evaporated almost to dryness, mixed with methanol/water,
1:1 (40 mL) and solid carbon dioxide. The residue was filtered off
and the filtrate evaporated to dryness before methanol was added
and centrifuged. The liquid phase was again evaporated almost to
dryness. Acetone/methanol, 19:1 (20 mL) was added before the
mixture was filtered and the solvents evaporated to dryness to yield
trated to give (2R*,3S*)-2 as a solid (0.51 g, 2.66 mmol, 84.2%),
1
m.p. 93–94 °C. H NMR: δ = 4.37 (s, 1 H), 3.83 and 3.78 (both s,
3 H, 2ϫ OCH3), 3.47 (s, 1 H, OH), 3.09 (s, 1 H, OH), 1.52 (s, 3
H, CH3). 13C NMR: δ = 175.0, 172.0, 77.3, 75.7, 53.4, 53.1, 22.8
(ref.[1]).
In alternative procedures the potassium osmate was exchanged
with immobilized Os, [Os3(CO)12-Al MCM-41] or osmium tetra-
oxide on poly(4-vinylpyridine). The workup was started by filtering
off the osmium catalyst. Na2SO3 was not added when immobilized
osmium was used, The catalyst was reused.
1
an oil: 0.18 g (1.32 mmol, 58.4%) of (2R,3R)-1. H NMR (D2O):
δ = 3.67 (dd, J = 2.8 and 12 Hz, H-4b), 3.55 (dd, J = 2.8 and
12 Hz, H-3), 3.40–3.46 (3 H, H-1a, H-1b, H-4a), 1.0 (s, 3 H). 13C
NMR (D2O): δ = 75.4 (C-3), 74.4 (C-2), 66.4 (C-1), 62.2 (C-4), 19.4
(C-5). [α]2D0 = +13 (c = 0.5, MeOH).
4.4. Transesterification Reactions: Small scale reactions were car-
ried out in order to find the best conditions for high E values.
A typical reaction contained 20 mg of the substrate in hexane or
toluene (3 mL), immobilized lipase (20 mg) and vinyl butanoate
(3 equiv.). The reactions were monitored by chiral GLC-analyses.
4.6. 2-C-Methyl-L-threitol [(2S,3S)-1]: Reduction of (2R,3R)-2
(0.280 g, 1.46 mmol) and workup of (2S,3S)-1 (0.04 g, 0.29 mmol,
19.9%) were performed by the same procedure as above. 1H NMR
(D2O): δ = 3.67 (dd, J = 2.8 and 12 Hz, H-4b), 3.55 (dd, J = 2.8
and 12 Hz, H-3), 3.40–3.46 (3 H, H-1a, H-1b, H-4a), 1.0 (s, 3 H).
13C NMR (D2O): δ = 75.4 (C-3), 74.4 (C-2), 66.4 (C-1), 62.2 (C-
4), 19.4 (C-5). [α]2D0 = –12.5 (c = 0.5, MeOH), ref.[7] –11.7 (c = 0.5,
MeOH). Elemental analysis data also given.
4.4.1. Lipase catalyzed esterification of (2R*,3R*)-2: (2R*,3R*)-Di-
methyl 2,3-dihydroxy-2-methylbutanedioate [(2R*,3R*)-2] (1.0 g,
5.2 mmol), vinyl butanoate (3.05 g, 26.7 mmol) were dissolved in
toluene (50 mL) and Novozym 735 (0.1 g) was added. The reaction
was monitored by GLC and stopped at 51% conversion. The en-
zyme was filtered off and washed with Et2O (10 mL) and the com-
bined organic phases were concentrated under reduced pressure.
Ester and alcohol were separated by flash chromatography (EtOAc/
hexane, 1:1). Alcohol (2R,3R)-2 (0.335 g, 1.74 mmol, 33.5%), ee Ͼ
99%, butanoate: (2S,3S)-4 (0.649 g, 2.48 mmol, 47.7%), ee 94%.
Alcohol (2R,3R)-2: 1H NMR: δ = 4.35 (s, 1 H), 3.83 and 3.82 (both
s, 3 H, 2ϫ OCH3), 1.49 (s, 3 H, CH3). 13C NMR: δ = 174.8, 171.6,
76.9, 75.2, 53.3, 52.8, 21.8. MS: 193.1 (M+1), 133.2 (100%), 87.
[α]2D0 = –23.6 (c = 1.0, CH2Cl2).
4.7. 2-C-Methyl-D-erythritol [(2S,3R)-1]: Reduction of (2R,3S)-4
(0.35 g, 1.3 mmol) and workup of solid (2S,3R)-1 (0.12 g,
0.88 mmol, 67.7%) were performed by the same procedure as
1
above. H NMR (D2O): δ = 3.75 (dd, J = 2.3 Hz, 11.4 Hz, H-4b),
3.58 (dd, J = 2.3 and 8.5 Hz, H-3), 3.52 (dd, J = 8.5 and 11.4 Hz,
H-4a), 3.50 (d, J = 11.7 Hz, H-1b), 3.4 (d, J = 11.7 Hz, H-1b), 1.1
(s, 3 H). 13C NMR (D2O): δ = 76.6, 75.8, 67.9, 63.6, 20.0. [α]2D0
+16.5 (c = 0.5, MeOH), ref.[3] +15.7 (c = 1.42, MeOH).
=
4.8. 2-C-Methyl-L-erythritol [(2R,3S)-1]: Reduction of (2S,3R)-2
(0.22 g, 1.15 mmol) and workup of (2R,3S)-1 (0.109 g, 0.8 mmol,
69.6%) were performed by the same procedure above. 1H NMR
(D2O): δ = 3.75 (dd, J = 2.3 and 11.4 Hz, H-4b), 3.58 (dd, J = 2.3
and 8.5 Hz, H-3), 3.52 (dd, J = 8.5 and 11.4 Hz, H-4a), 3.50 (d, J
= 11.7 Hz, H-1b), 3.4 (d, J = 11.7 Hz, H-1b), 1.1 (s, 3 H). 13C
NMR (D2O): 76.6, 75.8, 67.9, 63.6, 20.0. [α]2D0 = –12 (c = 0.5,
MeOH).
Butanoate (2S,3S)-4: 1H NMR: δ = 5.12 (s, 1 H), 4.72 (s, 6 H,
OCH3), 1.50 (s, 3 H, CH3), ABMNX3 system: 2.30 (AB), 1.60
(MN), 0.90 (X3) (JAB, JMNX = 7.5). 13C NMR: δ = 174.1, 172.7,
167.4, 76.4, 75.6, 53.7, 53.0, 36.0, 22.4, 18.7, 13.9. MS: 263.1
(M+1), 245 (–18, 100%), 231, 203, 185, 131, 115. [α]2D0 = +8.3 (c
= 1.0, CH2Cl2).
4.4.2. Lipase-Catalyzed Esterification of (2R*,3S*)-2: Two parallel
reactions were prepared where (1.0 g, 5.2 mmol), vinyl butanoate
(3.02 g, 26.5 mmol) were dissolved in toluene (50 mL) before Novo-
zym 735 (0.1 g) was added. The reactions were monitored on a
chiral GLC-column and stopped at 19% conversion (eep = 94%)
in one reaction and at 67% conversion (ees = 98.5%) in the parallel
reaction. The enzyme was filtered off and washed with Et2O
(10 mL) before the combined organic solvents were removed under
reduced pressure. Ester and alcohol were separated by flash
chromatography (EtOAc/hexane; 1:1). The butanoate was isolated
from the first reaction and in the second reaction the alcohol was
isolated.
Acknowledgments
We are grateful to Novozymes, Bagsværd, Denmark for a gift of
Novozym 735 (Lipase A from Candida antarctica immobilized on
Accurel).
[1] T. Anthonsen, S. Hagen, M. A. Kazi, S. W. Shah, S. Tagar,
Acta Chem. Scand. 1976, B30, 91–93.
[2] S. W. Shah, S. Brandänge, D. Behr, J. Dahmén, S. Hagen, T.
Anthonsen, Acta Chem. Scand. Ser. B 1976, 30, 903.
[3] T. Anthonsen, S. Hagen, W. Lwande, Acta Chem. Scand. Ser.
B 1980, 34, 41–45.
[4] M. Claeys, B. Graham, G. Vas, W. Wang, R. Vermeylen, V.
Pashynska, J. Cafmeyer, P. Guyon, M. O. Andrae, P. Artaxo,
W. Maenhaut, Science 2004, 303, 1173–1176.
[5] T. Anthonsen, S. Hagen, M. A. E. Sallam, Phytochemistry
1980, 19, 2375–2377.
[6] M. Rohmer, Nat. Prod. Rep. 1999, 16, 565–574.
[7] I. Sakamoto, K. Ichimura, H. Ohrui, Biosci. Biotechnol. Bi-
ochem. 2000, 64, 1915–1922.
[8] K. Kis, J. Wungsintaweekul, W. Eisenreich, M. H. Zenk, A.
Bacher, J. Org. Chem. 2000, 65, 587–592.
[9] J.-L. Giner, W. V. Ferris, J. J. Mullins, J. Org. Chem. 2002, 67,
4856–4859.
Alcohol (2S,3R)-2: 1H NMR: δ = 4.37 (s, 1 H), 3.83 and 3.78 (both
s, 3 H, 2ϫ OCH3), 3.47 (s, 1 H, OH), 3.09 (s, 1 H, OH), 1.52 (s, 3
H, CH3). 13C NMR: δ = 175.0, 172.0, 77.3, 75.7, 53.4, 53.1, 22.8.
[α]2D0 = –22.7 (c = 5.3, CH2Cl2).
Butanoate (2R,3S)-4: 1H NMR: δ = 5.43 (s, 1 H), 3.80 and 3.65
(both s, 3 H, 2ϫ OCH3), 1.40 (s, 3 H, CH3), ABMNX3 system:
2.45 (AB), 1.65 (MN), 0.90 (X3) (JAB, JMNX = 7.3). 13C NMR: δ
= 177.0, 173.3, 167.8, 75.1, 74.5, 53.6, 52.6, 35.6, 22.9, 18.6, 13.8.
[α]2D0 = –16.0 (c = 3.5, CH2Cl2).
4.5. 2-C-Methyl-D-threitol [(2R,3R)-1]: A solution of (2S,3S)-4
(0.592 g, 2.26 mmol) in dry THF (5 mL) was added to a suspension
of LiAlH4 (0.707 g, 18.6 mmol) in THF (20 mL). The mixture was
Eur. J. Org. Chem. 2007, 1262–1266
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
1265

A. R. Moen, K. Ruud, T. Anthonsen
FULL PAPER
[10] M. Urbansky, C. E. Davis, J. D. Surjan, R. M. Coates, Org.
Lett. 2004, 6, 135–138.
[11] V. VanRheenen, R. C. Kelley, D. Y. Cha, Tetrahedron Lett.
1976, 1973–1976.
[12] P. Dupau, R. Epple, A. A. Thomas, V. V. Fokin, K. B.
Sharpless, Adv Synth. Catal. 2002, 344, 421–433.
[13] V. Caps, S. C. Tsang, Applied Catal. A: General 2003, 248, 19–
31.
[14] R. J. Kazlauskas, A. N. E. Weissfloch, A. T. Rappaport, L. A.
Cuccia, J. Org. Chem. 1991, 56, 2656–2665.
[15] B. H. Hoff, T. Anthonsen, Chirality 1999, 11, 760–767.
[16] E. Fuglseth, T. Anthonsen, B. H. Hoff, Tetrahedron: Asym-
metry 2006, 17, 1290–1295.
Received: October 4, 2006
Published Online: January 17, 2007
1266
www.eurjoc.org
© 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2007, 1262–1266