An unexpected access to 5-epi-cyclophellitol: a new cyclitol member

Article abstract of DOI:10.1016/j.tetasy.2007.08.002

A thorough reinvestigation of the hydroboration-oxidation of methylene epoxycyclohexane 3 has revealed the formation of a direct precursor of the hitherto unreported 5-epi-cyclophellitol. Hydroboration of the starting precursor 3 takes place with total an

Full text of DOI:10.1016/j.tetasy.2007.08.002

Tetrahedron: Asymmetry 18 (2007) 1971–1974
An unexpected access to 5-epi-cyclophellitol: a new cyclitol member
a
a
a,b,
*
Pedro Serrano,a,b Meritxell Egido-Gabas, Amadeu Llebaria and Antonio Delgado
´
aResearch Unit on Bioactive Molecules (RUBAM), Department of Biological Organic Chemistry, Chemical and
Environmental Research Institute of Barcelona (IIQAB-CSIC), Jordi Girona 18-26, 08034 Barcelona, Spain
bUniversity of Barcelona, Faculty of Pharmacy, Unit of Pharmaceutical Chemistry (CSIC Associated Unit), Avda. Joan XXIII,
s/n, 08028 Barcelona, Spain
Received 19 July 2007; accepted 2 August 2007
Abstract—A thorough reinvestigation of the hydroboration–oxidation of methylene epoxycyclohexane 3 has revealed the formation of a
direct precursor of the hitherto unreported 5-epi-cyclophellitol. Hydroboration of the starting precursor 3 takes place with total and
opposite stereocontrol to that previously described in the literature. The stereochemistry of this new cyclophellitol isomer has been unam-
biguously confirmed by comparison with reference compounds obtained by independent methods.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
2. Results and discussion
(+)-Cyclophellitol 1a (Fig. 1)1 is a b-glucosidase inhibitor
isolated from the mushroom Phellinus sp.,2 which has also
been postulated as an HIV inhibitor as well as a potential
antimetastatic.3 From a structural standpoint, 1a can be
considered a carbasugar analogue of D-glucopyranose with
a characteristic b-epoxide ring. Several syntheses of 1a, as
well as those of analogues differing on the stereochemistry
of one or several stereogenic centers, have been described
in the literature.4 Over the course of our current research
on aminocyclitol derivatives as modulators of glicosphin-
golipid metabolism,5–7 we required the synthesis of 2,3,4-
tri-O-benzylcyclophellitol 1b for a subsequent epoxide
opening with nucleophiles.
Among the different approaches to the cyclophellitol skel-
eton described in the literature, we followed the sequence
described by Uzan et al.8,9 for the synthesis of 1b from
enone 2, obtained via a multi-step sequence from methyl-
a-D-glucopyranoside.10 The approach relied on the nucleo-
philic epoxidation of 2, followed by Wittig olefination and
stereoselective hydroboration–oxidation of intermediate
methylene epoxycyclohexane 3 (Scheme 1). However, con-
trary to what was described by the above authors, the epi-
meric cyclophellitol derivative 4 was obtained instead.
Compound 4 was in agreement with the spectroscopic data
described by Uzan et al. for this transformation.9,11 How-
1
ever, a comparison of its H NMR spectrum and physical
data with those described for 1b by other authors12–14
shows significant and striking differences. The most notice-
able ones were observed in the region between 2 and 4 ppm
OH
OH
8
8
1
4
3
6
4
3
6
RO
RO
HO
HO
in the H NMR spectrum, as well as an upfield shift of
5
2
5
2
O
4.2 ppm for C5 in compound 4 in the 13C NMR spectrum
(CDCl3).12,15 In addition, regio- and stereoselective lithium-
promoted nucleophilic opening of perbenzylated epoxide
7
7
O
1
1
OR
OH
1c: (+)-5-epi-cyclophellitol
5
with NaN3, following our previously described
1a: R=H; (+)-cyclophellitol
1b: R=Bn
protocol,16,17 afforded azido alcohol 6 (Scheme 2), whose
stereochemistry was assigned by spectroscopic methods.
Thus, the small J values observed for vicinal couplings
C5(H)–C4(H) and C5(H)–C6(H) are in agreement with a
C(5)H equatorial disposition in the presumably most stable
conformation (see Scheme 2). In addition, the large J1–6
Figure 1. Structures of (+)-cyclophellitol 1a, the 2,3,4-tri-O-benzyl deriv-
ative 1b, and (+)-5-epi-cyclophellitol 1c.
*
Corresponding author. Fax: +34 932045904; e-mail: adcqob@cid.csic.es
0957-4166/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2007.08.002

1972
P. Serrano et al. / Tetrahedron: Asymmetry 18 (2007) 1971–1974
OH
8
4
3
6
BnO
BnO
v
5
2
7
O
1c
O
1
BnO
BnO
BnO
BnO
OBn
OH
i, ii
iii, iv
8
O
this study
4
4
3
6
BnO
BnO
5
7
O
OBn
OBn
2
1
2
3
1b
OBn
Scheme 1. Reagents and conditions: (i) t-BuOOH, Triton B, CH2Cl2; (ii) Ph3P@CHBr, BuLi, THF; (iii) 1 M BH3ÆTHF, THF; (iv) 30% H2O2, 2 N NaOH,
THF–H2O (1:1) (72%); (v) 1 M BCl3, heptane–CH2Cl2 (ꢀ78 ꢁC).
OBn
OBn
H
H
OBn
4
6
4
BnO
BnO
N3
H
6
BnO
BnO
6
i
ii
J1-6= 10.1 Hz
J5-6= 5.4 Hz
J4-5= 5.0 Hz
5
OBn
BnO
HO
5
4
O
1
H
H
H
OH
5
1
1
N3
OBn
OBn
OBn
5
6
OH
OH
OBn
HCl
4
6
4
5
6
BnO
BnO
BnO
BnO
NHCH3
OH
HO
HO
NHCH3
5
iii
iv
v
CHO
OBn
OH
1
1
OBn
OH
7
8·HCl
Scheme 2. Reagents and conditions: (i) BnBr, NaH, DMF; (ii) NaN3, CH3CN, LiClO4; (iii) 2 M CH3NH2 (2 M in THF), CH3CN, LiClO4; (iv) BCl3 (1 M
in heptane), CH2Cl2, ꢀ78 ꢁC; (v) Ref. 20.
coupling constant also observed is consistent with the
expected trans-diaxial opening of the starting epoxide
under the above conditions.16,17 These data are clearly dif-
ferent from those described in the literature for the epimer
of 6 at C(5),18 the compound, which should be expected
from 1b under otherwise identical reaction conditions.19
current repertoire of cyclitol derivatives4 and will allow a
deeper insight into the influence of the stereochemistry of
the hydroxymethyl group on the interaction of these types
of systems with biological targets. Studies along this line, as
well those addressed at the evaluation of amino cyclitol
derivatives arising from the stereocontrolled epoxide open-
ing of 1c, are currently underway in our laboratory and will
be reported in due course.
In addition, the stereochemistry of 4 was unambiguously
confirmed by its regio and stereoselective conversion into
amino alcohol 8 (Scheme 2), and comparison of its spectro-
scopic data with those described in the literature for the
same compound obtained by an alternative synthesis.20,21
4. Experimental
4.1. General methods
Solvents were distilled prior to use and dried by standard
methods. Melting points are uncorrected. FT-IR spectra
3. Conclusion
are reported in cmꢀ1 1H and 13C NMR spectra were
.
In light of the above results, we can conclude that the ste-
reochemical course of the borane reduction of 3 affords
2,3,4-tri-O-benzyl-5-epi-cyclophellitol 4 following an ste-
reochemical course opposite to that described in the litera-
ture for this process.8 The reaction outcome can be
rationalized by assuming that the hydroboration of 3 takes
place with strict stereocontrol from the face opposite to the
bulky benzyloxy group at C4.22 Debenzylation of 4 in the
obtained in CDCl3 solutions at 500 MHz or 300 MHz or
1
200 MHz (for H) and 125 or 75 or 50 MHz (for 13C),
unless otherwise indicated. Chemical shifts were reported
in delta (d) units, parts per million (ppm) relative to the sin-
glet at 7.24 ppm of CDCl3 for 1H and in ppm relative to the
center line of a triplet at 77.0 ppm of CDCl3 for 13C. ESI/
HRMS spectra were recorded on a Waters LCT Premier
Mass spectrometer.
23
presence of BCl3 afforded the hitherto unreported (+)-
5-epi-cyclophellitol 1c (Fig. 1), a new cyclophellitol isomer
of the cyclitol family of natural products. In conclusion, a
close investigation of a described protocol has shown the
opportunity to gain access to a new isomeric cyclophellitol
isomer. This finding represents a valuable addition to the
4.2. 2,3,4-Tri-O-benzyl-5-epi-cyclophellitol 4
A solution of 536 mg (1.25 mmol) of 3 in freshly distilled
THF (5.5 mL) under Ar was treated with a BH3ÆTHF

P. Serrano et al. / Tetrahedron: Asymmetry 18 (2007) 1971–1974
1973
complex (3.8 mL of a 1 M solution in THF). The mixture
was stirred for 15 min at rt, cooled to 5 ꢁC in an ice bath
and diluted with 5.5 mL of a 1:1 THF–H2O mixture, after
which it was treated with a solution of 2 M NaOH (4.0 mL)
and 30% H2O2 (2.2 mL). After stirring for 30 min at rt, the
reaction mixture was diluted with H2O (20 mL) and
extracted with Et2O (4 · 50 mL). Usual work-up followed
by flash chromatography (hexanes–EtOAc from 9:1 to
3:2) afforded 402 mg (72% yield) of 4, as a colorless oil.
IR (film, cmꢀ1): 3446, 3029, 2879, 1496, 1453, 1365; 13C
NMR (125 MHz, CDCl3, assignments based on gHMQC):
39.8 (C5), 53.6 (C1), 55.5 (C6), 61.1 (C7), 73.0, 73.9, 74.9
(CH2OPh), 77.2 (C4), 79.4 (C2), 80.1 (C3), 127.6–128.6
(CHAr), 137.7, 137.9, 138.6 (CAr); 1H NMR (500 MHz,
CDCl3, assignments based on gCOSY and gHMQC):
2.83 (m, 1H, H5), 3.20 (d, 1H, J1–6 = 3.6 Hz, H1), 3.30
(m, 1H, H6), 3.65–3.85 (m, 3H, H3, H4, H7), 3.95 (m,
1H, H2), 4.10 (m, 1H, H70), 4.60–4.90 (m, 6H,
3 · CH2OPh), 7.20–7.30 (m, 15H, HAr). [a]D = +3.7 (c 1,
CHCl3); lit.:1 [a]D = +4.0 (c 1, CHCl3); mp: 89–91 ꢁC;
lit.:1 mp: 44–47.5 ꢁC.
73.3, 75.1, 75.4, 79.9, 82.4, 84.1, 127.4, 127.7, 127.8,
127.9, 128.1, 128.3, 129.4, 138.2, 138.3, 1398.7. H NMR
1
(CDCl3, 300 MHz): 2.45 (m, 1H, H5), 3.35 (t,
J = J0 = 10.1 Hz, 1H, H1), 3.52 (dd, J = 10.1, J0 = 5.4 Hz,
1H, H6), 3.65 (dd, J = 9.5, J0 = 5.0 Hz, 1H, H4), 3.71 (dd,
J = 14.5, J0 = 5.5 Hz, 1H, H7), 3.75 (dd, J = 14.5,
J0 = 5.5 Hz, 1H, H70), 4.05–4.30 (m, 2H, H2, H3), 4.60–
5.0 (m, 8H), 7.2–7.4 (m, 20H). HRMS: m/z calcd for
C35H37N3NaO5 (M+23): 602.2631; found: 602.2643.
4.5. (1R,2S,3S,4R,5S,6S)-2,3,4-Tris(benzyloxy)-5-
(hydroxymethyl)-6-(methylamino)cyclohexanol 7
Following the same protocol described for 6, treatment of
epoxide 4 (40 mg, 0.089 mmol) with MeNH2 (0.5 mL of a
2 M solution in THF) afforded 7 (39 mg, 92% yield) after
flash chromatography on 9:1 CH2Cl2–MeOH. 13C NMR:
34.8, 47.7, 60.5, 60.9, 70.6, 72.7, 75.8, 77.1, 80.5, 82.2,
84.8, 127.9–128.9, 138.1–138.6. 1H NMR: 2.54 (s, 3H,
CH3), 2.79 (m, 1H, H5), 3.36 (t, 1H, J = J0 = 9.7 Hz,
H1), 3.59 (dd, 1H, J = 9.7 Hz, J0 = 5.1 Hz, H6), 3.68 (t,
1H, J = J0 = 9.7 Hz, H2), 3.89–3.97 (m, 4H, H3, H4, H7,
H70), 4.56–5.00 (m, 6H, 3 · CH2OPh), 7.27–7.34 (m,
15H, CHAr). HRMS: m/z calcd for C29H34NNaO5
(M+23): 499.2335; found: 499.2346.
4.3. 2,3,4,8-Tetra-O-benzyl-5-epi-cyclophellitol 5
Benzyl bromide (1.1 mL, 9.28 mmol) was added dropwise
to a suspension of 95% NaH (0.54 g, 22.5 mmol) in DMF
(25 mL) at rt. The resulting mixture was cooled to 5 ꢁC
(ice-water bath) and treated with a solution of 4 (3.6 g,
8.08 mmol) in DMF (20 mL). The cooling bath was
removed and the reaction allowed to stir for 3 h, after
which it was quenched with H2O (15 mL) and extracted
with Et2O (3 · 50 mL). The usual work-up afforded a
syrup, which was purified by flash chromatography
(hexanes–EtOc 9:1) to afford 3.76 g (87% yield) of 5.
4.6. (1R,2S,3S,4R,5S,6S)-5-(Hydroxymethyl)-6-(methyl-
amino)cyclohexane-1,2,3,4-tetraol hydrochloride 8ÆHCl
A solution of 7 (24 mg, 0.05 mmol) in CH2Cl2 (2 mL) un-
der nitrogen at ꢀ78 ꢁC, was treated with 1 M BCl3 solution
in heptane (0.4 mL). After stirring for 2 h at ꢀ78 ꢁC, the
reaction mixture was allowed to warm to 0 ꢁC and stirred
for additional 24 h. The mixture was cooled to ꢀ78 ꢁC,
quenched with methanol (1 mL) and evaporated under
reduced pressure. The resulting residue was taken up in a
1:1 MeOH–H2O mixture, filtered through a small pad of
charcoal, and lyophilized to afford 8ÆHCl in 95% yield.
IR (film): 3034, 3003, 2910, 1496. 13CNMR (CDCl3,
75 MHz): 38.3 (C6), 53.9 (C1), 56.7 (C6), 67.3 (C7), 72.9,
73.9, 75.0 (CH2Ph), 75.9 (C2), 79.8 (C4), 80.7 (C3), 127–
129 (aromatic), 138–149 (aromatic). 1H NMR (CDCl3,
300 MHz, assignments based on gHMQC): 2.88 (m, 1H,
H5), 3.21(d, J = 5.4 Hz, 1H, H1), 3.45(m, 1H, H6), 3.65
(dd, J = 8.1 Hz, J0 = 5.4 Hz, 1H, H4), 3.73–3.90 (m, 4H,
H2, H3, H7, H70), 4.55–4.88 (m, 8H), 7.2–7.4 (m, 15H).
[a]D = +31.5 (c 2.8, CHCl3). HRMS: m/z calcd for
C35H36NaO5 (M+23): 559.2460; found: 559.2468.
13C NMR (125 MHz, DMSO-d6, assignments based on
gHSQC): 32.0 (CH3), 40.4 (C5), 55.9 (C7), 60.6 (C6), 69.8
(C1), 70.4 (C4), 73.1 (C3), 76.4 (C2); lit.: see Ref. 2 13C
NMR (125 MHz, D2O): 32.2 (CH3), 38.7 (C5), 57.2 (C7),
1
61.0 (C6), 69.5 (C1), 70.3 (C4), 72.8 (C3), 75.8 (C2). H
NMR (500 MHz, D2O, assignments based on gCOSY):
2.65 (m, 1H, H5), 2.70 (s, 3H, CH3), 3.05 (dd, J = 11.2,
J0 = 5.5 Hz, 1H, H6), 3.15 (app t, 1H, J = J0 = 11.2 Hz,
H2), 3.33 (app t, 1H, J = J0 = 11.2 Hz, H3), 3.48–3.60
(m, 3H, H4, NH2þ), 3.72 (app t, 1H, J = J0 = 11.0 Hz,
H70), 3.78 (app t, 1H, J = J0 = 11.2 Hz, H1), 3.85 (dd,
J = 11.0, J0 = 5.5 Hz, 1H, H7); lit (in DMSO-d6, D2O):
see Ref. 20 HRMS: m/z calcd for C8H16NNaO5 (M+23):
229.0926; found: 299.0915.
4.4. (1R,2S,3S,4R,5S,6S)-2-Azido-4,5,6-tris(benzyloxy)-3-
(benzyloxymethyl)cyclohexanol 6
A solution of the starting epoxide 5 (500 mg, 1.2 mmol) in
CH3CN (10 mL) was slowly added dropwise under argon
at rt over previously dried LiClO4 (2.87 g, 27.0 mmol). A
solution of 780 mg (12 mmol) of NaN3 in CH3CN
(2.5 mL) was then added and the reaction mixture was stir-
red at 80 ꢁC under argon. After 18 h, the reaction mixture
was cooled to rt, quenched with H2O (10 mL), extracted
with CH2Cl2 (3 · 20 mL), and dried over anhydrous
Na2SO4. Filtration and evaporation afforded the crude azi-
do alcohol, which was purified by filtration through a plug
of silica and elution with hexanes–EtOAc (9:1 to 8:2). IR
(film): 3405, 3062, 3029, 2867, 2103, 1496, 1453, 1269.
13C NMR (CDCl3, 75 MHz): 40.8, 62.3, 63.5, 72.1, 72.7,
4.7. 5-epi-Cyclophellitol; (1S,2R,3S,4R,5S,6R)-5-hydroxy-
methyl-7-oxabicyclo[4.1.0]heptane-2,3,4-triol 1c
Following the same protocol as described above for the
synthesis of amino alcohol 8, 22.5 mg of epoxide 4
(0.05 mmol) afforded 8.5 mg (98% yield) of 5-epi-cyclo-
phellitol 1c. 1H NMR (500 MHz, D2O): 2.50 (m, 1H),
3.10 (m, 1H), 3.35–3.45 (m, 2H), 3.50 (dd, 1H), 3.65 (m,

1974
P. Serrano et al. / Tetrahedron: Asymmetry 18 (2007) 1971–1974
2H), 3.80 (dd, 1H). 13C NMR (125 MHz, D2O): 39.5, 55.2,
55.9, 56.8, 65.7, 69.9, 71.4. [a]D = +65.9 (c 0.4, H2O)
HRMS: m/z calcd for C7H12O5Na (M+23): 199.0582;
found: 199.0581.
5. Egido-Gabas, M.; Serrano, P.; Casas, J.; Llebaria, A.;
Delgado, A. Org. Biomol. Chem. 2005, 3, 1195–1201.
6. Serrano, P.; Casas, J.; Zucco, M.; Emeric, G.; Egido-Gabas,
M.; Llebaria, A.; Delgado, A. J. Comb. Chem. 2007, 9, 43–52.
7. Egido-Gabas, M.; Canals, D.; Casas, J.; Llebaria, A.;
Delgado, A. ChemMedChem 2007, 5, 992–994.
`
8. Letellier, P.; Ralainirina, R.; Beaupere, D.; Uzan, R. Tetra-
Acknowledgments
hedron Lett. 1994, 35, 4555–4558.
`
9. Letellier, P.; Ralainairina, R.; Beaupere, D.; Uzan, R.
Financial support from the Ministerio de Ciencia y Tec-
Synthesis 1997, 925–930.
´
nologıa, Spain (Project CTQ2005-00175/BQU), the Minis-
´
10. Jaramillo, C.; Chiara, J. L.; Martın-Lomas, M. J. Org. Chem.
terio de Sanidad y Consumo (Spain) (Project PIO40767),
Fondos Feder (EU), Generalitat de Catalunya
(2005SGR01063), and CSIC is acknowledged. P.S. and
1994, 59, 3135–3141.
11. The stereochemistry of the new stereogenic center at C5 was
assigned by the authors on the basis of the magnitude of the
coupling constant J(H4–H5) = 7.4 Hz for H4 in CDCl3.
However, we were unable to measure this coupling constant,
since, in our case, this diagnostic proton overlapped with the
H3 and H8 signals. Unfortunately, attempts to carry out the
structural assignment of 4 by spectroscopic methods (2D
experiments) were inconclusive.
´
M.E.-G. are acknowledged to the Ministerio de Educacion
y Ciencia (Spain) and CSIC, respectively, for pre-doctoral
fellowships.
References
12. Ziegler, E.; Yizshe, W. J. Org. Chem. 1998, 63, 426–427.
13. Trost, B. M.; Patterson, D. E.; Hembre, E. J. Chem. Eur. J.
2001, 7, 3768–3775.
14. Ishikawa, T.; Shimizu, Y.; Kudoh, T.; Saito, S. Org. Lett.
2003, 5, 3879–3882.
1. The systematic nomenclature for (+)-cyclophellitol is (1S,2R,
3S,4R,5R,6R)-5-hydroxymethyl-7-oxabicyclo[4.1.0]heptane-
2,3,4-triol. This numbering will be used through this work.
However, (+)-cyclophellitol can also be named as the inositol
derivative as 1,2-anhydro-3-deoxy-3-(hydroxymethyl)-D-myo-
inositol. 5-epi-Cyclophellitol, [(1S,2R,3S,4R,5S,6R)-5-hydroxy-
methyl-7-oxabicyclo[4.1.0]heptane-2,3,4-triol], can be also
named as 3-epi-cyclophellitol, as the inositol derivative.
2. Atsumi, S.; Umezawa, K.; Iinuma, H.; Naganawa, H.;
Nakamura, H.; Iitaka, Y.; Takeuchi, T. J. Antibiot. 1990,
43, 49–53.
15. Moreover, differences in [a]D and mp values between 1b and
4
were also significant; 1b (Ref. 12): [a]D = +71.0 (c
0.9, CHCl3); mp: 92–93 ꢁC; 1b (Ref. 13): [a]D = +71.7 (c
1.62, CHCl3); mp: 101 ꢁC; 1b (Ref. 14): [a]D = +71.2 (c 1.62,
CHCl3); mp: 102–103.5 ꢁC; 4 (Ref. 9): [a]D = +4.0 (c 1.0,
CHCl3); mp: 44.6–45.7 ꢁC; 4 (this work): [a]D = +3.7 (c 1.0,
CHCl3); mp: 89–91 ꢁC.
16. Serrano, P.; Llebaria, A.; Vazquez, J.; de Pablo, J.; Anglada,
J. M.; Delgado, A. Chem. Eur. J. 2005, 11, 4465–4472.
17. Serrano, P.; Llebaria, A.; Delgado, A. J. Org. Chem. 2005, 70,
7829–7840.
18. Nakata, M.; Chong, C.; Niwata, Y.; Toshima, K.; Tatsuta,
K. J. Antibiot. 1993, 46, 1919–1922.
19. Coupling constants for (5-epi)6, as described in Ref. 18:
J1–6 = 9.8 Hz; J5–6 = 10.8 Hz; J4–5 = 10.8 Hz.
20. Peet, N. P.; Huber, E. W.; Farr, R. A. Tetrahedron 1991, 47,
7537–7550.
3. Atsumi, S.; Jinuma, H.; Nosaka, C.; Umezawa, K. J.
Antibiot. 1990, 43, 1579–1585.
4. For total syntheses of (+)-cyclophellitol, see: (a) Arjona, O.;
Gomez, A. M.; Lopez, J. C.; Plumet, J. Chem. Rev. 2007, 107,
1919–2036, references therein, Ref. 13 and; (b) Kireev, A. S.;
Breithaupt, A. T.; Collins, W.; Nadein, O. N.; Kornienko, A.
J. Org. Chem. 2005, 70, 742–745; Synthesis of cyclophellitol
stereoisomers: (ꢀ)-cyclophellitol: Refs. 4b and 13; (1R,6S)-
cyclophellitol: (c) Tai, V. W. F.; Fung, P. H.; Wong, Y. S.;
Shing, T. K. M. Biochem. Biophys. Res. Commun. 1995, 213,
175–180; (d) Jung, M. E.; Choe, S. W. T. J. Org. Chem. 1995,
60, 3280–3281; (e) McDevitt, R. E.; Fraserreid, B. J. Org.
Chem. 1994, 59, 3250–3252; (f) Tatsuta, K.; Niwata, Y.;
Umezawa, K.; Toshima, K.; Nakata, M. J. Antibiot. 1991, 44,
456–458; (1R,2S,6S)-cyclophellitol: (g) Shing, T. K. M.; Tai,
V. W. F. J. Chem. Soc., Perkin Trans. 1 1994, 2017–2025; (h)
Tatsuta, K.; Niwata, Y.; Umezawa, K.; Toshima, K.;
Nakata, M. J. Antibiot. 1991, 44, 912–914, and Ref. 4c;
(2S,3R,4S) and (2S,3R)-cyclophellitol: (i) Roberts, S. M.;
Sutton, P. W. J. Chem. Soc., Perkin Trans. 1 1995, 1499–1503
(2S)-cyclophellitol: Ref. 4f.
21. 13C NMR spectrum in DMSO-d6 was superimposable to that
described in the literature (see Ref. 20) and clearly different
from that of the epimer of 8 at C(5). In addition, C(6)H
proton (D2O, 45 ꢁC) was of diagnostic value. Thus, one large
(J = 11.2 Hz) and one small (J = 5.5 Hz) coupling constant
were observed for this proton. This is consistent with its axial
disposition, in a conformation equivalent to that postulated
for azido alcohol 6 (see Scheme 2).
22. In our hands, formation of 1b was never detected.
23. Williams, D. R.; Brown, D. L.; Benbow, J. W. J. Am. Chem.
Soc. 1989, 111, 1923–1925.