Chemoenzymatic synthesis of (2S,3S,4S)-form, the physiologically active stereoisomer of dehydroxymethylepoxyquinomicin (DHMEQ), a potent inhibitor on NF-κB

Article abstract of DOI:10.1016/j.tet.2010.07.013

A new route for (2S,3S,4S)-form, the physiologically active stereoisomer of dehydroxymethylepoxyquinomicin (DHMEQ), a potent NF-κB inhibitor, was established by chemoenzymatic approach. Elaboration on the asymmetric epoxidation of a p-benzoquinone monoket

Full text of DOI:10.1016/j.tet.2010.07.013

Tetrahedron 66 (2010) 7083e7087
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Chemoenzymatic synthesis of (2S,3S,4S)-form, the physiologically active
stereoisomer of dehydroxymethylepoxyquinomicin (DHMEQ), a potent inhibitor
on NF-kB
Manabu Hamada a, Yukihiro Niitsu b, Chihiro Hiraoka c, Ikuko Kozawa a, Toshinori Higashi b,
,
b,
*
Mitsuru Shoji b, Kazuo Umezawa a , Takeshi Sugai
*
a Department of Applied Chemistry, Keio University, 3-14-1 Hiyoshi, Yokohama 223-8522, Japan
b Faculty of Pharmacy, Keio University, 1-5-30 Shibakoen, Minato-ku, Tokyo 105-8512, Japan
c Department of Chemistry, Keio University, 3-14-1 Hiyoshi, Yokohama 223-8522, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 June 2010
Received in revised form 6 July 2010
Accepted 7 July 2010
Available online 24 July 2010
A new route for (2S,3S,4S)-form, the physiologically active stereoisomer of dehydroxymethylepoxy-
quinomicin (DHMEQ), a potent NF-kB inhibitor, was established by chemoenzymatic approach. Elabo-
ration on the asymmetric epoxidation of a p-benzoquinone monoketal with benzylcinchonidinium
tert-butylhydroperoxide yielded an epoxyenone, in 79.8% ee and 57% yield in reproducible manner. By
way of the transformation of this key intermediate to enantiomerically pure (2S,3S,4S)-DHMEQ, the
contaminating undesired enantiomer could be effectively removed by applying Burkholderia cepacia
lipase-catalyzed hydrolysis of diacylated precursor. The above integrated combination of chemical
asymmetric synthesis and enzyme-catalyzed kinetic resolution enabled us to prepare active DHMEQ in
a large-scale.
Keywords:
DHMEQ
Asymmetric epoxidation
Lipase
Hydrolysis
Ó 2010 Elsevier Ltd. All rights reserved.
Kinetic resolution
1. Introduction
First, according to the reported procedure,7 amino group of
commercially available 2,5-dimethoxyaniline 4a was protected by
DHMEQ (dehydroxymethylepoxyquinomicin, 1a) is a newly
designed and highly active NF-
B inhibitor,1 and has shown potent
Boc group and the resulting hydroquinone derivative 4b was sub-
sequently dehydrogenated by the action of PhI(OAc)2 in methanol
with a concomitant ketalization. The convergence of 5 to desired 3a
was important, because diketal 5, the direct precursor for 3a,
behaves as the troublesome contaminant at the stage of purifica-
tion. Then, the crude mixture was treated with 2 M hydrochloric
acid according to the original procedure,4 but such operation
caused an unexpected decomposition of the materials to result in
k
anti-inflammatory and anticancer activities in animals.2 Through
extensive studies on structure-activity relationships,3 it was
revealed that (2S,3S,4S)-1a is more than ten times as physiologi-
cally active as its enantiomer (Fig. 1). So far, the chromatographic
separation of racemate by means of a chiral stationary phase has
only been the way to approach enantiomerically pure form of
(2S,3S,4S)-1a.4 Herein we report a new asymmetric synthesis of
(2S,3S,4S)-1a.
2. Results and discussion
The synthetic plan of (2S,3S,4S)-1a is shown in Scheme 1.
Enantiomerically enriched form of 2a is a known building block
developed by Taylor5,6 by an asymmetric epoxidation of the pre-
cursor aminoquinine monoketal 3a with the action of chiral qua-
ternary ammonium tert-butylhydroperoxide.
* Corresponding authors. Tel./fax: þ81 3 5400 2665; e-mail address: sugai-tk@
pha.keio.ac.jp (T. Sugai).
Figure 1. The enantiomers of dehydroxymethylepoxyquinomicin (DHMEQ, 1a).
0040-4020/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2010.07.013

7084
M. Hamada et al. / Tetrahedron 66 (2010) 7083e7087
necessity of neutral to slightly acidic environment, even sacrificing
the nucleophilicity of tert-butylhydroperoxide.
Scheme 3. (a) N-Benzylcinchonidinium chloride, TBHP, 6 M NaOH aq soln, toluene,
40%, 52.4% ee; (b) TFA, CH2Cl2, quant.
Accordingly, a new substrate, bis-Boc derivative 3c,8 bearing no
acidic proton on enamine nitrogen was designed. Under the con-
ditions with cinchonidinium salt and free base in 1:1, the epoxi-
dation proceeded smoothly at 25 ꢀC in a conversion of 74%. The
crude mixture was directly treated with trifluoroacetic acid, and
the desired epoxyenamine (2S,3R)-2a was obtained in 30% yield
over two steps. Since an unidentified byproduct was detected on
TLC during the epoxidation, the reaction temperature was lowered
to 0 ꢀC, with increasing the equivalence of hydroperoxide from two
to five, in order to compensate the low reactivity. After extensive
optimization, we found that the equivalence of cinchonidinium salt
and NaOH should be two and one, respectively, under carefully
controlled reaction conditions.
Scheme 1. Synthetic plan of (2S,3S,4S)-1a.
3a in as low as 21%. This observation promoted us to switch
hydrochloric acid into weaker aqueous citric acid, and the depro-
tection proceeded in a reproducible manner to give 3a in 48% yield.
Moreover, careful examination in regard to oxidizing agent, the
yield of the dehydrogenation of Boc derivative 4b was improved as
high as 80%, by applying PhI(OPiv)2 (Scheme 2).
Finally, it was concluded that the absorption of any trace of
ꢀ
water with MS 4 A and trapping sodium cation with 15-crown-5
further were important. In this way, the yield involving epoxidation
and deprotection was improved to be 43% (over two steps) from
29% in original procedure and ee of the product was constantly over
75% (Scheme 4).
Scheme 2. (a) Boc2O, THF, 99%; (b) PhI(OPiv)2, MeOH; (c) citric acid aq soln, 80% over
two steps.
In our hand, Taylor’s asymmetric epoxidation of 3a was not
entirely reproducible,5,6 in terms of either enantioselectivity or
conversion. Even under the optimal conditions, the results were not
satisfactory. For example, the yield and ee of the product fluctuated
between 30 and 40%, and between 43.8 and 52.4%, respectively.
This asymmetric epoxidation initiates with an enantiofacially
preferential 1,4-attack of chiral N-benzylcinchonidinium salt of
tert-butylhydroperoxide (TBHP) to prochiral enone 3a. The epoxide
2b was obtained only as an inseparable mixture with the starting
material 3a. Removal of Boc group in 2b provided 2a in pure form,
however, the remaining enone 3a decomposed via 3b to an in-
tractable waste. In this point of view, higher conversion in epoxi-
dation is desirable. We improved the yield, by stepwise
examination as follows: (1) the modification of the substrate
structure; (2) careful adjustment of the equivalence of reagents;
(3) additives, which suppress the undesired species, which would
cause non-asymmetric epoxidation. The detail is shown below.
Under Taylor’s procedure,5 the epoxidation had been carried out
in buffered conditions with a ratio of 4:1 between acidic hydro-
chloride and basic hydroxide of cinchonidine. An equilibration
between 3a and enolate 6 by the deprotonation of tert-butylcar-
bamate group under basic conditions (Scheme 3) can explain the
Scheme 4. (a) Boc2O, DMAP, THF, 89%; (b) N-benzylcinchonidinium chloride, TBHP,
NaOH, 15-crown-5, toluene, MS 4 A, 57%; (c) TFA, anisole, CH2Cl2, 76%, 79.8% ee.
ꢀ
Although enantiomerically pure 2a was obtained by re-
crystallization from ethyl acetate, the recovery was onlyas lowas 8%.
This situation prompted us to use complementary enzyme-cata-
lyzed kinetic resolution to remove undesired enantiomer on
DHMEQ itself or intermediates toward target molecule (Scheme 5),
which had been established in the synthesis of racemate of DHMEQ.
After several trials, we turned our attention to the application
of lipase-catalyzed hydrolysis on the diacylated form 1b and 1c
of DHMEQ, whose acyl chains were introduced on both hydroxyl
groups in secondary alcohol and phenol. Between two candi-
dates, dihexanoate 1c with two medium-length acyl chains was
advantageous because of its stability on silica gel column chro-
matography, was much superior to that of diacetate 1b. At
this stage, diastereomerically pure 1c free from (2S,3S,4R)-
or (2R,3R,4S)-1c0 became in hand. Two commercially available
lipases were compared, and Burkholderia cepacia lipase (Amano
PS-IM) showed higher activity than that of Candida antarctica

M. Hamada et al. / Tetrahedron 66 (2010) 7083e7087
7085
Scheme 5. (a) Acetylsalicyloyl chloride, t-BuOLi, THF; (b) K2CO3, aq MeOH, 67% over
two steps; (c) BF3$OEt2, CH2Cl2, 42%; (d) NaBH(OAc)3, MeOH.
lipase B (Novozym 435). The hydrolysis of both acyl chain pro-
ceeded very smoothly in the case of (2S,3S,4S)-isomer. By virtue
the unique properly of DHMEQ, the poor solubility in neither
water nor organic solvent except for tetrahydrofuran, the puri-
fication protocol became very simple. First, the crude residue
after evaporation of the solvent was triturated in tetrahydrofu-
ran to recover all of organic materials. Then, the residue after
concentration of the above extract was alternatively triturated in
diisopropyl ether. Unreacted recovery and hexanoic acid resulted
by the hydrolysis were removed, and pure DHMEQ 1a was
obtained as precipitated solid in 69% yield (Scheme 6). All of the
spectral and physical properties were identical with those
reported previously. The enantiomeric excess (>99.9% ee) was
confirmed by HPLC analysis of 1c, which was prepared by re-
acylation of 1a.
The lipase-catalyzed hydrolysis was really advantageous, as
(2S,3S,4S)-isomer was readily transformed to 1a, while the un-
desired (2R,3R,4R)-isomer stayed at the stage of monohexanoate
1d. Any attempts for the isolation of (2R,3R,4R)-1d were un-
successful, as the monohexanoate was labile on silica gel and is
prone to isomerize into two components of 1a and 1c even on
the occasion of TLC analysis. Thanks to the lipophilic property of
1d, it remained in supernatant at the stage of trituration with
diisopropyl ether, and silica gel chromatography could be
avoided.
We refrain from discussing the enantioselectivity in a quantita-
tive manner, for example, as E value, because precise evaluation of
neither the conversion of the reaction nor the ee of the slow-
reacting isomer (1d) was available. A very high enantioselectivity,
however, is believed, as even from racemate of 1c, enantiomerically
pure (2S,3S,4S)-1a was obtained.
Scheme 6. (a) (C5H11CO)2O, DMAP, THF, 90%; (b) B. cepacia lipase (Amano PS-IM),
acetone, water, 69%; (c) trituration in THF; (d) trituration in diisopropyl ether; (e) silica
gel chromatography.
3. Conclusion
We established an efficient chemoenzymatic route to (2S,3S,4S)-
DHMEQ 1a. First, bis-Boc derivative 3c was designed, and the
desired epoxy ketone (79.8% ee) became available in a reproducible
manner. In the final step, contaminated (10%) undesired enantio-
mer could be easily removed by lipase-catalyzed kinetic resolution.
This achievement for large-scale and practical synthesis would
promote the clinical trials of DHMEQ itself. Moreover, the stable,
lipophilic acylated derivative is promising as the clue to design and
synthesize pro-drugs.

7086
M. Hamada et al. / Tetrahedron 66 (2010) 7083e7087
4. Experimental
4.6. Direct approach to (2S,3R)-5-amino-2,3-epoxy-4,4-
dimethoxycyclohexa-5-en-1-one (2a)
4.1. Material and methods
In a similar manner as described above, to a soln of 3a (2.50 g,
9.30 mmol), N-benzylcinchonidinium chloride (5.85 g, 13.9 mmol,
1.5 equiv) and TBHP (2.0 M in toluene, 1.6 mL, 15.0 mmol, 1.6 equiv)
Merck silica gel 60 F254 thin-layer plates (1.05744, 0.5 mm thick-
ness) and silica gel 60 (spherical and neutral; 100e210 mm,
37,560e79)from KantoChemicalCo., wereusedforpreparative thin-
layer chromatography and column chromatography, respectively.
Under anhydrous reaction conditions, molecular sieves were pre-
dried in vacuo for 1 h at 100 ꢀC.
in toluene (150 mL) were added NaOH aq soln (6 M, 750 mL) at
ꢁ10 ꢀC. The similar workup as above yielded a mixture of (ꢁ)-2b
and 3a (total 2.71 g) as yellow oil.
Then, it was treated with TFA (25 mL, 27.0 mmol) according to
the reported procedure6 to give 2a (528 mg, 30%, 43.8% ee) as dark
brown oil. This was further purified by recrystallization from
4.2. Analytical methods
EtOAc to give (ꢁ)-2a (141 mg, 8.0% recovery, 99.0% ee) as colorless
23
All mps are uncorrected. IR spectra were measured as films for
oils or KBr disks for solids on a Jasco FT/IR-410 spectrometer or ATR
on a Jeol FT-IR SPX60 spectrometer. 1H NMR spectra were measured
in CDCl3 at 270 MHz on a Jeol JNM EX-270 or 400 MHz on a VARIAN
400-MR spectrometer, and 13C NMR spectra were measured in
CDCl3 at 100 MHz on a VARIAN 400-MR spectrometer. Optical
rotation values were recorded on a Jasco P-1010 polarimeter. High
resolution mass spectra were recorded on a Jeol JMS-700 MStation
spectrometer. HPLC data were recorded on Jasco MD-2010 multi-
channel detectors and SHIMADZU SPD-M20A diode array detector.
needles. Mp 162e163 ꢀC [lit.9 mp 161e162 ꢀC]; [
a
]
ꢁ106 (c 0.60,
D
MeOH) [lit.9
[
a
]
ꢁ91.0 (c 1.0, MeOH)]. Its IR and 1H NMR spectra
D
were identical with those reported previously.9 The product 2a was
analyzed by HPLC [CHIRALCELÒ OD-H; hexanee2-propanol (2:1);
flow rate 0.5 mL/min detected at 300 nm]: tR (min)¼17.1 (99.5%),
28.2 (0.5%).
4.7. 3-N,N-Bis-(tert-butoxycarbonyl)amino-4,4-
dimethoxycyclohexa-2,5-dien-1-one (3c)
To a soln of 3a (538 mg, 2.0 mmol) in anhydrous THF (6 mL)
were added DMAP (122 mg, 1.0 mmol) and di-tert-butyl dicar-
bonate (874 mg, 4.0 mmol) at 25 ꢀC. The mixture was stirred at
70 ꢀC under argon atmosphere for 6 h. The reaction was quenched
by the addition of water, and then extracted with EtOAc. The or-
ganic layer was washed with brine, dried over Na2SO4, and con-
centrated in vacuo. The residue was purified by silica gel column
chromatography (10 g). Elution with hexaneeEtOAc (4:1) afforded
3c (561 mg, 89%) as white solid. Mp 117e118 ꢀC; IR nmax 2981, 1753,
4.3. N-tert-Butoxycarbonyl-2,5-dimethoxyaniline (4b)
A soln of 4a (5.01 g, 32.6 mmol) and di-tert-butyl dicarbonate
(8.55 g, 39.2 mmol) in THF (20 mL) was heated under reflux for 2
days. After removal of volatile materials in vacuo, the residue was
purified by silica gel column chromatography (100 g) with hex-
aneeEtOAc (10:1) to afford 4b (8.16 g, 99%) as a colorless oil. Its 1H
NMR spectrum was identical with that reported previously.8
1718, 1674, 1340, 1234, 1122 cmꢁ1 1H NMR (400 MHz)
; d: 1.47 (s,
4.4. 3-tert-Butoxycarbonylamino-4,4-dimethoxycyclohexa-
2,5-dien-1-one (3a)
18H), 3.29 (s, 6H), 6.28 (d, J¼2.4 Hz, 1H), 6.36 (dd, J¼2.4, 10.4 Hz,
1H), 6.80 (d, J¼10.4 Hz, 1H); 13C NMR (100 MHz)
d: 185.7, 150.8,
149.4, 142.1, 130.8, 130.6, 94.8, 83.3, 51.3, 27.8; MS (FAB): m/e (%
relative intensity) 370.30 (33) [Mþ1]þ, 392.29 (22) [Mþ23]þ;
HRMS (FAB): calcd for C18H27NO7: [Mþ1]þ: 370.1866; found:
m/z¼370.1864.
To a soln of 4b (3.20 g, 12.6 mmol) in anhydrous MeOH (45 mL)
at 0 ꢀC under N2 were added MS 4 A (1.2 g), and then portionwise
ꢀ
PhI(OPiv)2 (6.16 g, 15.2 mmol). The mixture was stirred for 24 h at
0 ꢀC. The mixture was diluted with EtOAc (25 mL) and saturated
citric acid aq soln (10 mL) was added to organic layer. After stirring
for 30 min, the mixture was extracted with EtOAc, and the com-
bined organic layer was washed with saturated NaHCO3 aq soln,
water, and brine. The organic phase was dried over Na2SO4 and
concentrated in vacuo. The residue was purified by recrystallization
from EtOAcehexane to give 3a (1.90 g, 60%) as colorless needles.
The residue from mother liquor was further purified by silica gel
column chromatography (90 g) with hexaneeEtOAc (1:1) to afford
3a (0.63 g, 20%). Mp 130e131 ꢀC [lit.7 mp 130e131 ꢀC]. Its 1H NMR
spectrum was identical with that reported previously.7
4.8. Asymmetric epoxidation of 3c: (2S,3R)-5-N,N-bis-(tert-
butoxycarbonyl)amino-2,3-epoxy-4,4-dimethoxycyclohexa-6-
en-1-one (2c)
To
a mixture of N-benzylcinchonidinium chloride (30.0 g,
ꢀ
70.5 mmol, 2.0 equiv) and MS 4 A (20.0 g) in anhydrous toluene
(130 mL) were added TBHP (2.0 M in toluene, 88 mL, 176 mmol,
5.0 equiv), NaOH (1.40 g, 35.2 mmol, 1.0 equiv), and 15-crown-5
(9.50 g, 8.50 mL, 39.0 mmol, 1.1 equiv) at 10 ꢀC. The resulting sus-
pension was stirred for 30 min, and then to the mixture was added
3c (13.0 g, 35.2 mmol) at 10 ꢀC under argon atmosphere. After
stirring for 24 h, the mixture was quenched with saturated Na2S2O3
4.5. (2S,3R)-5-tert-Butoxycarbonylamino-2,3-epoxy-4,4-
dimethoxycyclohexa-5-en-1-one (2b)
ꢀ
aq soln MS 4 A and other insoluble materials were removed by
According to the reported procedure,6 3a (50.0 mg, 0.19 mmol)
was treated with N-benzylcinchonidinium chloride (117 mg,
0.28 mmol), TBHP (32 mL, 0.30 mmol), and NaOH aq soln (6 M,
filtration, and the filtrate was extracted with CHCl3. The organic
layer was washed with brine, dried over Na2SO4, and concentrated
in vacuo to give a mixture of 2c and 3c (total 10.3 g) as oil. As the
product 2c and the starting material 3c showed same Rfs on silica
gel, the yield was estimated to be 57%, based on 1H NMR with an
internal standard [ethyl benzoate, Tokyo Kasei Co., B0069, analyt-
15
m
L) in toluene (3 mL) for 1 h at ꢁ10 ꢀC, and then allowed to
warm to room temperature and stirred for further 48 h to afford 2b
(21.2 mg, 40%, 52.4% ee) as a colorless needle. Mp 118e119 ꢀC [lit.6
mp 116e117 ꢀC]; [
a]
24 ꢁ103 (c 0.92, CHCl3) [lit.6
[a
]
D ꢁ189.4 (c 1.0,
ically pure grade, standard signal at
d
¼4.36 (dd, J¼6.8, 7.2 Hz, 2H)].
D
CHCl3)]. Its 1H NMR spectrum was identical with that reported
previously.5 The product 2b was analyzed by HPLC [column, Daicel
CHIRALCELÒ OD-H, 0.46 cmꢂ25 cm; hexanee2-propanol (15:1);
flow rate 0.5 mL/min; detected at 300 nm]: tR (min)¼18.8 (76.2%),
28.8 (23.8%).
Then, the crude product in CH2Cl2 (120 mL) was treated with
TFA (30 mL, 32.4 mmol) and anisole (12 mL), to give (2S,3R)-2a
(2.90 g, 46% for two steps, 79.8% ee) as solid. HPLC analysis was
performed in the same manner as above: tR (min)¼17.1 (89.9%),
28.1 (10.1%).

M. Hamada et al. / Tetrahedron 66 (2010) 7083e7087
7087
4.9. (2S,3R)-5-N-[(2-Acetoxybenzoyl)amino]-2,3-epoxy-4,4-
dimethoxycyclohexa-6-en-1-one (7a)
(d, J¼7.8 Hz, 1H), 7.35 (dd, J¼7.8, 7.8 Hz, 1H), 7.54 (ddd, J¼1.5, 7.8,
7.8 Hz, 1H), 7.67 (dd, J¼1.5, 7.8 Hz, 1H), 7.99 (br s, 1H); 13C NMR
(100 MHz) d: 192.2, 173.3, 172.3, 164.3, 147.7, 133.0, 129.6,127.9, 126.6,
To a soln of above-mentioned 2a (2.80 g, 15.1 mmol) in THF
(140 ml) at ꢁ78 ꢀC were added the lithium tert-butoxide (Aldrich,
398185, 16.6 mL, 16.6 mmol), and subsequently acetylsalicyloyl
chloride (3.50 g, 18.1 mmol) in portionwise. After stirring for 1 h at
ꢁ78 ꢀC, the mixture was quenched with saturated NH4Cl aq soln and
extracted with EtOAc. The organic layer was washed with brine, dried
over Na2SO4, and concentrated in vacuo. The residue was purified by
silica gel column chromatography (240 g). Elution with hex-
aneeEtOAc (2:1) afforded 7a (3.50 g) as white solid. Mp 84e85 ꢀC; IR
nmax 3367, 2920, 2848, 1772, 1670, 1597, 1495, 1456, 1363, 1273, 1234,
123.5, 109.9, 66.5, 52.8, 34.2, 34.0, 31.2, 24.5, 22.3, 13.9, 13.8; IR nmax
3372, 3284, 3256, 3217, 3148, 3102, 3043, 2956, 2930, 2871,1752,1699,
1642, 1520, 1282, 1241, 1214, 1155, 1099, 1033, 905, 781 cmꢁ1. Anal.
Calcd for C25H31NO7: C, 65.6; H, 6.83; N, 3.06. Found: C, 65.6; H, 6.73;N,
3.10. The product 1c was analyzed by HPLC [CHIRALCELÒ AD-H,
0.46ꢂ25 cm; hexanee2-propanol (7:1); flow rate 0.5 mL/min;
detected at 290 nm]: tR (min)¼28.3 (10.1%), 30.9 (89.9%).
4.13. B. cepacia lipase-catalyzed hydrolysis of (2S,3S,4S)-1c
1178,1109,1045 cmꢁ1; 1H NMR (400 MHz)
d: 2.37 (s, 3H), 3.29 (s, 3H),
To a soln of (2S,3S,4S)-1c (46.0 mg, 79.8% ee, 0.10 mmol) in ace-
tone (0.7 mL) and water (0.7 mL) was added B. cepacia lipase (Amano
PS-IM, 70 mg). After stirring for 24 h at room temperature, the mix-
ture was concentrated in vacuo for 1 h to dryness and then mixed
with THF (10 mL). Ultrasonic vibration (200 W) was applied for the
resulted suspension for 10 min, and the mixture was filtered through
a pad of Celite to remove insoluble materials. The filtrate was con-
centrated in vacuo and the residue was mixed with i-Pr2O (10 mL). To
the mixture was again applied the ultrasonic vibration (200 W) for
10 min, to wash out monohexanoyl DHMEQ 1d. After filtration, the
residue on the filter paper was charged on a silica gel column
(300 mg), and the elution with THF afforded (2S,3S,4S)-1a (18.0 mg,
69%) as white solid. Mp 187 ꢀC [lit.4 mp 185 ꢀC]. As the solubility of
DHMEQ was very low in methanol, the comparison of the sign of
rotations between that of present sample and authentic datum
seemed not to be reliable enough, then the ee as well as the absolute
configuration of the present DHMEQ was confirmed by derivation to
the corresponding dihexanoate (2S,3S,4S)-1c again. Colorless oil;
3.55(dd, J¼2.0, 4.4 Hz,1H), 3.67 (s, 3H), 3.83(d, J¼4.4 Hz,1H), 7.15 (dd,
J¼1.2, 8.4 Hz, 1H), 7.31 (d, J¼2.0 Hz, 1H), 7.33 (ddd, J¼1.2, 7.6, 8.4 Hz,
1H), 7.54 (ddd, J¼1.6, 7.6, 8.0 Hz,1H), 7.85 (dd, J¼1.6, 8.0 Hz,1H), 8.73
(br s,1H); 13C NMR (100 MHz)
d: 192.7,168.7,164.1,148.0,145.1,133.2,
130.5, 126.9, 126.6, 123.5, 109.7, 95.7, 52.1, 51.6, 51.5, 50.8, 20.9. This
was employed for the next step without further purification.
4.10. (2S,3R)-2,3-Epoxy-5-N-[(2-hydroxybenzoyl)amino]-4,4-
dimethoxycyclohexa-6-en-1-one (7b)
To a soln of 7a (3.50 g) in MeOH (42 mL) and H2O (14 mL) was
added K2CO3 (1.40 g, 10.1 mmol). After stirring for 30 min at room
temperature, the mixture was diluted with H2O (20 mL), and then
extracted with EtOAc. The organic layer was washed with saturated
NaHCO3 aq soln and brine, dried over anhydrous Na2SO4, and
concentrated in vacuo to afford 7b (3.10 g, 67% for two steps) as
white solid. Mp 117e118 ꢀC. Its IR and 1H NMR spectra were iden-
tical with those reported previously.4 This was employed for the
next step without further purification.
[
a
]
23 ꢁ126 (c 0.66, CHCl3). HPLC analysis was performed in the same
D
mannerasabove:tR (min)¼30.9(singlepeak).Thisretentiontimewas
in good accordance with that of an authentic specimen of (2S,3S,4S)-
1c, and there was no peak at 28.3 min ascribable to (2R,3R,4R)-1c.
4.11. (2S,3R)-2,3-Epoxy-5-N-[(2-hydroxybenzoyl)amino]-
cyclohexa-6-en-1,4-dione (8)
Acknowledgements
Accordingtothereportedprocedure,4 ketal7b(2.90 g, 9.50 mmol)
was treated with BF3$OEt2 (3.74 mL, 29.0 mmol) in CH2Cl2 (15 mL) for
1 h at ꢁ20 ꢀC, and then allowed to warm to room temperature and
stirred for further 7 h to afford 8 (1.03 g, 42%) as yellow solid. Its 1H
NMR spectrumwas identical with that reported previously.4 This was
employed for the next step without further purification.
The authors thank Dr. Yoshihiko Hirose of Amano Enzyme Inc.
for generous gift of lipase PS-IM and Dr. Yoichi Suzuki of Novo-
zymes Japan for Novozym 435. This work was supported both by
a Grant-in-Aid for Scientific Research and ‘High-Tech Research
Center’ Project for Private Universities: matching fund subsidy
2006e2011 from the Ministry of Education, Culture, Sports, Science
and Technology, Japan, and acknowledged with thanks.
4.12. (2S,3S,4S)-2,3-Epoxy-4-hexanoyloxy-5-N-[(2-
hexanoyloxybenzoyl)amino]-cyclohexa-6-en-1-one
(dihexanoyl DHMEQ, 1c)
Supplementary data
According to the reported procedure,4 diketone
8 (1.0 g,
Supplementary data associated with this article can be found in
the online version at doi:10.1016/j.tet.2010.07.013.
3.90 mmol) was treated with NaBH(OAc)3 (1.70 g, 7.80 mmol) in
MeOH (100 mL) for 20 min at 0 ꢀC, and then the mixture was
allowed to warm to room temperature and stirred for further 4 h to
afford crude mixture (2.7 g) containing 1a.
The residuewas suspended inTHF (100 mL) and tothatwere added
hexanoic anhydride (2.80 mL) and DMAP (16.0 mg). After stirring for
30 min at room temperature, the mixture was quenched with ice-
water (20 mL), then stirred for 20 min at room temperature and
extractedwithEtOAc. Theorganiclayerwaswashedwithhydrochloric
acid (0.5 M), saturated NaHCO3 aq soln and brine, dried over Na2SO4,
and concentrated in vacuo. The residue was purified by silica gel col-
umn chromatography (190 g). Elution with hexaneeEtOAc (4:1) to
References and notes
1. Matsumoto, N.; Ariga, A.; To-e, S.; Nakamura, H.; Agata, N.; Hirano, S.; Inoue, J.;
Umezawa, K. Bioorg. Med. Chem. Lett. 2000, 10, 865e869.
2. Umezawa, K. Cancer Sci. 2006, 97, 990e995.
3. Chaicharoenpong, C.; Kato, K.; Umezawa, K. Bioorg. Med. Chem. 2002, 10,
3933e3939.
4. Suzuki, Y.; Sugiyama, C.; Ohno, O.; Umezawa, K. Tetrahedron 2004, 60,
7061e7066.
5. Macdonald, G.; Alcaraz, L.; Lewis, N. J.; Taylor, R. J. K. Tetrahedron Lett. 1998, 39,
5433e5436.
6. Alcaraz, L.; Macdonald, G.; Ragot, J. P.; Lewis, N.; Taylor, R. J. K. J. Org. Chem. 1998,
63, 3526e3527 and referenced cited therein.
7. Taylor, R. J. K.; Alcaraz, L.; Kapfer-Eyer, I.; Macdonald, G.; Wei, X.; Lewis,
N. Synthesis 1998, 775e790.
8. Sunitha, S.; Kanjilal, S.; Reddy, P. S.; Prasad, R. B. N. Tetrahedron Lett. 2008, 49,
2527e2532 and referenced cited therein.
afford (2S,3S,4S)-1c (1.30 g, 72%, 79.8% ee) as yellow oil. [
a
]
23 ꢁ82.0 (c
D
1.00, CHCl3); 1H NMR (270 MHz)
d: 0.85 (t, J¼6.8 Hz, 3H), 0.91 (t,
J¼6.8 Hz, 3H),1.30 (m, 4H),1.34 (m, 4H),1.70 (m, 4H), 2.53 (t, J¼7.8 Hz,
2H), 2.56 (t, J¼7.8 Hz, 2H), 3.50 (dd, J¼2.0, 3.9 Hz,1H), 3.91 (dd, J¼2.9,
3.9 Hz,1H), 5.85 (dd, J¼1.5, 2.9 Hz,1H), 7.02 (dd, J¼1.5, 2.0 Hz,1H), 7.10
9. Alcaraz, L.; Macdonald, G.; Ragot, J.; Lewis, N. J.; Taylor, R. J. K. Tetrahedron 1999,
55, 3707e3716.