Convenient synthesis of 3- and 6-deoxy-d-myo-inositol phosphate analogues from (+)-epi- and (-)-vibo-quercitols

Article abstract of DOI:10.1016/j.bmcl.2006.06.096

Starting from (+)-epi- and (-)-vibo-quercitols readily produced by bioconversion of myo-inositol, some biologically interesting phosphate and polyphosphate analogues, including the Ins(1,4,5)P₃ derivatives of 3-deoxy- and 6-deoxy-d-myo-inositol

Full text of DOI:10.1016/j.bmcl.2006.06.096

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5238–5243
Convenient synthesis of 3- and 6-deoxy-D-myo-inositol phosphate
analogues from (+)-epi- and (ꢀ)-vibo-quercitols
Seiichiro Ogawa* and Yoji Tezuka
Department of Biosciences and Informatics, Faculty of Science and Technology, Keio University,
Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan
Received 26 December 2005; revised 7 April 2006; accepted 26 June 2006
Available online 10 August 2006
Abstract—Starting from (+)-epi- and (ꢀ)-vibo-quercitols readily produced by bioconversion of myo-inositol, some biologically inter-
esting phosphate and polyphosphate analogues, including the Ins(1,4,5)P3 derivatives of 3-deoxy- and 6-deoxy-D-myo-inositol, could
be readily prepared in a conventional manner. In addition, chemical modification at C-2 of the 3-deoxy Ins(1,4,5)P3 provided 2-epi-
mer, and 2-deoxy and 2-deoxy-2-fluoro forms. Eight polyphosphate analogues obtained were assayed for biological activity against
PDH-Pase and PDH-K, and G6Pase, but none proved positive.
ꢀ 2006 Elsevier Ltd. All rights reserved.
In recent years, D-myo-inositol-1,4,5-trisphosphate
Ins(1,4,5)P3, as well as its bis and tetrakisphosphates,
have been demonstrated1 to play important roles as sec-
ond messengers which control many cellular processes
by generating internal calcium signals, which then dif-
fuse through the cytosol and bind to receptors on the
endoplasmic reticulum causing the release of calcium
ions (Ca2+) into the cytosol. Therefore, it is feasible that
inhibitors of enzymes of the phosphoinositide cascade,
involved in biosynthesis and degradation of Ins(1,4,5)P3,
could be of medicinal interest and also invaluable tools
to elucidate the individual roles of metabolites in the
regulation of cell function. In order to study biochemical
and medicinal properties of these polyphosphates, a
large number of analogues and derivatives have so far
been synthesized2 and their biological activity tested.
Recent findings3 of insulin-like and anti-inflammatory
properties have also stimulated us to develop means
for routine synthesis of these compounds (Fig. 1).
Bioconversion4 of myo-inositol readily provides some
inaccessible optically active deoxyinositols,5 such as
(+)-epi- and (ꢀ)-vibo-quercitols (1 and 2), in quantity,
Figure 1. Deoxyinositols 1 and 2, and Ins(1,4,5)P3 and its deoxy
derivatives.
which might allow their application as starting materials
Keywords: Inositol phosphates; myo-Inositol-1,4,5-trisphosphate;
Deoxyinositol trisphosphate analogues; myo-Inositol bioconversion.
for development of novel biologically active cyclitol
derivatives. In preceding papers, we reported the synthe-
sis of several anhydro5 and some C-(aminometh-
yl)deoxyinositols,6 and O-methyl-deoxyinosamines7 as
*
Corresponding author. Tel.: +81 45 566 1788; fax: +81 45 566
1789; e-mail
keio.ac.jp
addresses:
sogawa379@ybb.ne.jp;
ogawa@bio.
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.06.096

S. Ogawa, Y. Tezuka / Bioorg. Med. Chem. Lett. 16 (2006) 5238–5243
5239
potential candidate new glycosidase inhibitors. In this
letter, we describe convenient synthesis of a number of
mono-, tris-, and tetrakisphosphate derivatives of
3- and 6-deoxy-D-myo-inositols. In addition, chemical
modification at C-2 of the 3-deoxy Ins(1,4,5)P3 was car-
ried out to prepare 2-epimer, and 2-deoxy and 2-deoxy-
2-fluoro derivatives. Furthermore, in order to provide
certain trisphosphate mimics8 designed by analogy with
adenophostines,9 the most potent agonists of
Ins(1,4,5)P3 receptor, the 1-phosphate function of the
3-deoxy Ins(1,4,5)P3 was replaced with (phosphin-
yl)alkyloxyl groups (Schemes 1 and 2).
3 (24%), 4 (26%), and 5 (31%). These were treated with
dibenzylphosphoryl chloride in pyridine at room tem-
perature to give the respective dibenzylphosphates 6
(77%), 7 (86%), and 8 (87%), hydrogenolysis of which
with PtO2 in ethanol produced, after neutralization
with cyclohexylamine, the respective phosphates as
crystalline amine salts. Treatment of the salts with
Dowex 50 W · 2 (H+) resin gave the free phosphates
9 (88%), 10 (70%), and 11 (80%), isolated as bis-sodium
salts.
Furthermore, compounds 3 and 5 could be partially de-
O-isopropylidenated with TsOH in EtOH at 0 ꢁC to give
the triols 12 (70%) and 14 (78%), respectively. Possible
contamination of 12 and 14 due to acid-catalyzed migra-
tion of the cis-isopropylidene groups was not observed.
Compound 14 was similarly phosphorylated to give the
protected precursor 15 (60%) of 6-deoxy Ins(1,4,5)P3.11
The structure of 15 was indirectly confirmed by the fact
that 1H NMR spectrum of isomeric trisphosphate 13 ob-
tained for reference from 12 showed coupled signals (d
4.31, ddd, J = 5.6, 8.9, and 13.9 Hz) and (d 4.62,
J = 5.1, and 8.9 Hz) due to H-1 and H-2 bonded to car-
bon atoms of the O-isopropylidene group. Thus, five
biologically interesting phosphate analogues could be
demonstrated to be readily available from (+)-epi-querc-
itol (1).
Recently, synthesis of several polyphosphate derivatives
of 6-deoxy-D-myo-inositol10 (1) has been elaborated11
from precursors derived from D-galactose, and their bio-
logical activity assayed. 6-Deoxy Ins(1,4,5)P3 is recog-
nized by the highly selective 3-kinase,12 the kinetics of
its metabolism indicate that it is a substrate for this en-
zyme, with resultant competitive inhibition of phosphor-
ylation of Ins(1,4,5)P3.
We here describe a convenient preparative route from
(+)-epi-quercitol4 (1). Isopropylidenation6 of 1 with
an excess of 2,2-dimethoxypropane and TsOH in
DMF was carried out at room temperature to give
three readily separable di-O-isopropylidene derivatives
Scheme 1. Synthesis of some mono- and trisphosphates of 6-deoxy-D-myo-inositol. Reagents and conditions: (a) (MeO)2CMe2–DMF (1:2, v/v),
TsOH, 6 h, rt; (b) (PhO)2POCl (1.5 M equiv), pyridine, 1 h, rt; aqueous 80% AcOH; (c) H2, PtO2, EtOH, rt; C6H11NH2; Dowex 50 W · 2 (H+) resin,
1 M NaOMe, MeOH; (d) TsOH, EtOH, pH ꢁ 4, 0 ꢁC; (e) (PhO)2POCl, DMAP, pyridine, rt; aqueous 80% AcOH.

5240
S. Ogawa, Y. Tezuka / Bioorg. Med. Chem. Lett. 16 (2006) 5238–5243
and then with an excess of BnBr to give the benzyl ethers
18 and 19, which were partially de-O-isopropylidenated
under the influence of CSA in MeOH to afford, after
separation over a silica gel column, the 4- and 6-O-ben-
zyl derivatives 20 (55%) and 21 (42%). Compound 20
was treated with chloromethyl methyl ether and diiso-
propylethylamine to give the di-O-methoxymethyl deriv-
ative 22 (89%), de-O-isopropylidenation of which with
80% aqueous acetic acid gave the diol 23 (88%).
Treatment of 23 with dibutyltin oxide and tetrabutyl
ammonium bromide at 120 ꢁC, and subsequent similar
etherification, gave crude methoxymethyl ether 24
(87%). The structure of 24 could be fully characterized
1
with the H NMR spectrum of the O-acetyl derivative
25 (ꢁ100%) obtained. In addition, 24 was conventional-
ly benzylated to give the 2-O-benzyl derivative 26 (91%).
The methoxymethyl groups of 26 were removed by treat-
ment with 4 M hydrochloric acid, and the product was
subsequently acetylated to give the tri-O-acetyl deriva-
tive 27 (90% over-all yield). Compound 27 was treated
´
with methanolic sodium methoxide under Zemplen con-
ditions, and the resulting triol was phosphorylated with
dibenzyl diisopropylphosphoro-amidite in DMF, and,
then the reaction mixture was treated with mCPBA.
The product was isolated by chromatography on silica
gel to afford the 1,4,5-tris(dibenzylphosphate) 28 (93%
over-all yield). Hydrogenolysis of 28 in the presence of
10% Pd/C in aqueous ethanol at room temperature gave
the trisphosphate, which was purified by treatment with
cyclohexylamine to produce a crystalline salt. This com-
pound was deaminated by passage through a column
of Dowex 50 · 2 (H+) resin to afford the free phosphate
isolated as a bis-sodium salt 29 (97%).
The tetra-O-acetyl derivative 30 (98%), obtained similar-
ly from 20, was deacetylated and a crude alcohol was
dibenzylphosphorylated to give tetrakisbenzylphosphate
31 (70% over-all yield). It was deprotected and the prod-
uct was obtained as a bis-sodium salt15 32 (83%).
Oxidation of compound 24 with pyridinium chlorochro-
˚
mate in the presence of molecular sieves 4 A in CH2Cl2
Scheme 2. Synthesis of some tris- and tetrakisphosphates of 3-deoxy-
myo-inositol. Reagents and conditions: (a) CH2@C(OMe)CH3, TsOH
(0.1 M equiv), DMF, 3 h, rt; (b) NaH, BnBr, DMF; (c) CSA, MeOH,
pH ꢁ 4, rt; (d) MeOCH2Cl (4 M equiv), diisopropylethylamine, DMF;
(e) aqueous 80% AcOH; (f) Bu2SnO (2 M equiv), tetrabutylammonium
bromide; (g) Ac2O, pyridine; (h) NaH, BnBr, DMF, rt; (i) 4 M HCl,
50 ꢁC; Ac2O, pyridine; (j) NaOMe, MeOH; i-Pr2NP(OBn)2
(6 M equiv), DMF, rt; mCPBA (10 M equiv), rt; (k) H2, 10% Pd/C,
aqueous EtOH; C6H11NH2; Dowex 50 W · 2 (H+) resin, NaOMe,
MeOH; (l) aqueous 80% AcOH, 50 ꢁC.
gave the deoxyinosose derivative 33 (96%), which was
reduced with sodium borohydride to give the 2-epimer
34 (56%) as a major product. The structure of 34 was
1
confirmed by the H NMR spectrum of the 2-O-acetyl
derivative 35. Compound 34 was converted into the
2-epimer, 3-deoxy-scyllo Ins(1,4,5)P3 39, following the
standard sequence of reactions [!36 (95%) !37
(90%) !38 (56%) !39 (97%)].
Treatment of 24 with sulfuryl chloride in pyridine in the
presence of DAMP gave the chloride 40 (70%). This
compound was treated with tributyltin hydride in the
presence of AIBN to provide the 2-deoxy derivative 41
(94%). Starting from 41, 2,3-dideoxy Ins(1,4,5)(P3)14 44
was obtained [!42 (96%) !43 (86%) !44 (ꢁ100%)] in
a conventional manner. Fluorination of 24 with dimeth-
yl amino sulfur trifluoride (DAST) in CH2Cl2 afforded
the 2-deoxy-2-fluoro derivative 45 (70%), which was
converted into the tri-O-acetyl derivative 46 (95%). This
acetate 46 was deacylated and conventionally phosphor-
ylated to give the trisphosphate 47, which was deprotec-
3-Deoxy-D-myo-inositol-(1,4,5)P3 was first synthe-
sized13,14 from L-quebrachitol through a multi-step
sequence and shown to be a good substrate of Ins(1,4,5)P3-
5-phosphatase (Schemes 2 and 3).
Isopropylidenation5 of (ꢀ)-vibo-quercitol (2) was carried
out by treatment with 2-methoxypropene in the presence
of TsOH in DMF at room temperature. A mixture of the
1,2:4,5- 16 and 1,2:5,6-di-O-isopropylidene derivatives
17 was, without separation, treated with NaH in DMF

S. Ogawa, Y. Tezuka / Bioorg. Med. Chem. Lett. 16 (2006) 5238–5243
5241
Scheme 3. Chemical modification at C-2 of 3-deoxy Ins(1,4,5)P3. Synthesis of deoxy and deoxyfluoro derivatives. Reagents and conditions: (a) PCC,
˚
MS 4 A, CH2Cl2, rt; (b) NaBH4, EtOH; (c) Ac2O, pyridine; (d) NaH, BnBr, DMF; (e) MeOCH2Cl, diisopropylethylamime, DMF; Ac2O, pyridine;
(f) NaOMe, MeOH; dibenzyldiisopropylphosphoro-amidite, DMF; (g) H2, 10% Pd/C, aqueous EtOH; C6H11NH2, Dowex 50 W · 2 (H+) resin,
NaOMe, MeOH; (h) SOCl2, DAMP, pyridine; (i) Bu3SnH, AIBN, toluene, 120 ꢁC; (j) Ac2O, pyridine; (k) 4 M HCl, THF; phosphorylation; (l)
DAST, CH2Cl2, rt; (m) aqueous 80% AcOH; (PhO)2POCl.
ted to give the deoxyfluoro derivative15 48 (70% over-all
yield). Similarly, the alcohol 34 was transformed into the
deoxyfluoro derivative15 51 (80% over-all yield) via 49
and 50.
Compound 23 was treated with dibutyltin oxide in tol-
uene, and then after addition of allyl bromide, the
mixture was heated at reflux temperature to give the
allyl ether 52 (90%), which was characterized as the
acetate 53 (96%) as a syrup. Compound 52 was con-
verted into the benzyl ether (!54, 82%), which was
subjected to ozonolysis in CH2Cl2/MeOH, followed
by reduction with NaBH4 (!55a) and conventional
acetylation to give the 2-acetoxyethyl derivative 56a
(60% over-all yield). Hydrolysis of 56a with 4 M
Next, adenophosphine9 analogues of 3-deoxy
Ins(1,4,5)P3 were prepared. Several permeant analogues
of Ins(1,4,5)P3 have been synthesized8 and their ability
to cross the membrane tested with vasopressin cells
(Scheme 4).

5242
S. Ogawa, Y. Tezuka / Bioorg. Med. Chem. Lett. 16 (2006) 5238–5243
Scheme 4. Chemical modification of the C-2 function of 3-deoxy Ins(1,4,5)P3. Reagents and conditions: (a) Bu2SnO, TBAB, toluene, 120 ꢁC;
CH2@CHBr; (b) Ac2O, pyridine; (c) NaOMe, MeOH; NaH, BnBr, DMF; (d) (!55a) O3, CH2Cl2/MeOH, ꢀ78 ꢁC; NaBH4, MeOH; (!55b) BH3/
THF, THF; H2O2; (e) Ac2O, pyridine; (f) 4 M HCl, THF; Ac2O, pyridine; (g) NaOMe, MeOH; i-Pr2NP(OBn)2, 1H-Tetrazole, DMF; mCPBA; (h)
H2, 10% Pd/C, aqueous 30% EtOH; C6H11NH2, Dowex 50 W · 2 (H+) resin, 1 M NaOMe, MeOH.
hydrochloric acid in THF gave, after conventional
acetylation, the acetate 57a (80%). Conventional de-
O-acetylation of 57a gave the diol, which was treated
with 1H-tetrazole in DMF and then with dibenzyliso-
propylphosphoro-amidite to give the protected tris-
phosphate 58a (72% over-all yield). Hydrogenolysis
of 58a in aqueous ethanol in the presence of 10%
Pd/C gave the trisphosphate15 59a (ꢁ100%) isolated
as a syrupy sodium salt.
Acknowledgments
The authors sincerely thank Ms. Lili Zhao for perform-
ing elemental analyses, Drs. Atsushi Takahashi and
Kiyoshi Sato (Hokko Chemical Industry Co. Ltd, Atsu-
gi, Japan) for providing (+)-epi- and (ꢀ)-vibo-quercitols,
Drs. James W. Stygall and Khalid Gumaa (Rodaris
Pharmaceuticals Ltd, UK) for carrying out a part of
the biological assays, and Ms. Miki Kanto for her assis-
tance in preparing this manuscript.
Hydroboration of 54 was accomplished with boran–
THF complex in THF, and then the reaction was
quenched by addition of water, followed by treatment
with 35% aqueous hydroperoxide for 30 min (!55b).
The product was isolated as the acetate 56b (ꢁ100%)
and similarly converted in turn into the trisphosphate15
59b.
References and notes
1. (a) Streb, H.; Irvine, F. R.; Berridge, M. J.; Schulz, I.
Nature 1983, 306, 67; (b) Berridge, M. J.; Irvine, R. F.
Nature 1989, 341, 197; (c) Berridge, M. J. Nature 1993,
361, 315, and references cited therein.
2. (a) Potter, B. V. L.; Lampe, O. Angew. Chem., Int. Ed.
Engl. 1995, 34, 1933; (b) Irvine, R. F.; Schell, M. J. Nat.
Rev. Mol. Cell Biol. 2001, 2, 327.
3. (a) Cobb, J. E.; Johnson, M. R. Tetrahedron 1991, 47, 21;
(b) Cleophax, J.; Dubreuil, D.; Gero, S. D.; Loupy, A.;
Vieira de Almeida, M.; Adison, D.; Da Silva, I.; Vass, G.
Bioorg. Med. Chem. Lett. 1995, 5, 831.
None of the eight compounds 29, 32, 39, 44, 48, 51, and
59a,b activated pyruvate dehydrogenase phosphatase
(PDH-Pase), or inhibited pyruvate dehydrogenase ki-
nase (PDH-K) significantly. None of the compounds
tested inhibited glucose 6-phosphatase (G6Pase)
significantly.
4. Takahashi, A.; Kanbe, K.; Tamamura, T.; Sato, K.
Anticancer Res. 1999, 3807.
5. Ogawa, S.; Uetsuki, S.; Tezuka, Y.; Morikawa, T.;
Takahashi, A.; Sato, K. Bioorg. Med. Chem. 1999, 9,
1493.
Compounds 59a and 59b appeared to inhibit lipogen-
esis (Multi-Well Plate cell based assay). However, the
inhibition decreased with increasing concentrations of
the test compounds. When tested in vivo in streptozo-
tocin diabetic mice (a model of type I diabetes) at a
dose of 1 mg/kg, neither 59a nor 59b had any acute
lowering effect on blood glucose. Furthermore, when
administered chronically at a dose of 1 mg/kg/day
for 10 days, 59a did not significantly lower blood
glucose.
6. Ogawa, S.; Aoyama, H.; Tezuka, Y. J. Carbohydr. Chem.
2001, 20, 703.
7. Ogawa, S.; Asada, M.; Ooki, Y.; Mori, M.; Itoh, M.;
Korenaga, T. Bioorg. Med. Chem. 2005, 13, 4306.
8. Chretien, F.; Roussel, F.; Hilly, M.; Mauger, J.-P.;
Chapleur, Y. J. Carbohydr. Chem. 2005, 24, 549, and
references cited therein.

S. Ogawa, Y. Tezuka / Bioorg. Med. Chem. Lett. 16 (2006) 5238–5243
5243
21
9. (a) Takahashi, S.; Kinoshita, T.; Takahashi, M. J.
Antibiot. 1994, 47, 95; (b) Takahashi, M.; Tanzawa, K.;
Takahashi, S. J. Biol. Chem. 1994, 289, 369.
10. In this letter, the nomenclature of cyclitols follows
IUPAC-IUB 1973 Recommendations for Cyclitols [Pure.
Appl. Chem. 1974, 37, 285].
11. (a) Vieira de Almeida, M.; Dubreuil, D.; Cleophax, J.;
Verre-Sebrie, C.; Pipelier, M.; Prestat, G.; Vass, G.; Gero,
S. D. Tetrahedron 1999, 55, 7251, and references cited
therein; (b) Dubreuil, D.; Cleophax, J.; Vieira de Almeida,
M.; Pipelier, M.; Vass, G.; Gero, S. D. Tetrahedron 1999,
55, 7573, and references cited therein.
12. Safrany, S. T.; Wojcikiewicz, R. J. H.; Strupish, J.;
Nahorski, S. R.; Dubreuil, D.; Cleophax, J.; Gero, S.
D.; Potter, B. V. L. FEBS Lett. 1991, 278, 252.
13. Seewald, M. J.; Aksoy, I. A.; Powis, G.; Fauq, A. H.;
Kozikowski, A. P. J. Chem. Soc., Chem. Commun. 1990,
1638.
48 (in D2O); ½aꢂD +7.3 (c 2.5, H2O); d 4.53–4.29 (m, 1H,
H-5), 4.06–3.85 (m, 3H, H-1, H-2, H-4), 3.43 (dd, 1H,
J2,3 = 8.5 Hz, J3,4 = 8.8 Hz, H-3), 2.41–2.35 (m, 1H, H-
6eq), 1.75–1.65 (m, 1H, H-6ax). [MꢀH] m/z 405,
[M+Naꢀ2H] m/z 427, [Mꢀ2H]2 m/z 202; for compound
21
51 (in D2O); ½aꢂD +7.5 (c 1.4, H2O); d 4.90 (d, 1H,
J2,F = 48.6 Hz, H-2), 4.16–4.02 (m, 1H, H-4), 3.98–3.82
(m, 2H, H-1, H-5), 3.67 (dd, 1H, J1,6 = J5,6 = 9.5 Hz, H-
6), 2.41–2.30 (m, 1H, H-3eq), 1.60 (dddd, 1H,
J2,3ax = 2.2 Hz,
J3ax,4 = 12.9 Hz,
Jgem = 13.7 Hz,
J3ax,F = 46.9 Hz, H-3ax). [MꢀH] m/z 405, [M+Naꢀ2H]
23
m/z 427, [Mꢀ2H]2 m/z 202; for compound 59a (in
CDCl3); ½aꢂD +17 (c 2.4, CHCl3); d 7.34–7.01 (m, 40H,
8 · Ph), 5.08 and 4.44 (m, 18H, H-4, H-5, 8 · CH2Ph),
4.03–3.98 (m, 2H, 2 · H-20), 3.89 (dd, 1H, J3,4 = 9.4 Hz,
J2,3 = 9.8 Hz, H-3), 3.78 (br, 1H, H-1), 3.60–3.55 (m,
2H, 2 · H-10), 3.26 (dd, 1H, J1,2 = 2.7 Hz, J2,3 = 9.8 Hz,
H-2),
2.61
(ddd,
1H,
J1,6eq = J5,6eq = 4.4 Hz,
14. Kozikowski, A. P.; Ognyanov, V. I.; Fauq, A. H.;
Nahorski, S. R.; Wilcox, R. A. J. Am. Chem. Soc. 1993,
115, 4429.
Jgem = 13.9 Hz, H-6eq), 1.39–1.28 (m, 1H, H-6ax); and
26
for compound 59b (in CDCl3); ½aꢂD ꢀ129 (c 1.3,
CHCl3); d 7.36–7.01 (m, 40H, 8 · Ph), 5.08 and 4.56
(m, 17H, H-5, 8 · CH2Ph), 4.52 (dd, 1H, J4,5 = 9.0 Hz,
J3,4 = 9.3 Hz, H-4), 4.05–3.95 (m, 2H, 2 · H-30), 3.87
(dd, 1H, J3,4 = 9.3 Hz, J2,3 = 9.5 Hz, H-3), 3.76 (br, 1H,
H-1), 3.75–3.38 (m, 2H, 2 · H-10), 3.22 (dd, 1H,
J1,2 = 2.7 Hz, J2,3 = 9,5 Hz, H-2), 2.65 (ddd, 1H,
J1,6eq = J5,6eq = 4.4 Hz, Jgem = 14.4 Hz, H-6eq), 1.83–
1.79 (m, 2H, 2 · H-20), 1.39–1.30 (m, 1H, H-6ax).
15. 1H NMR (300 MHz) data for compound 32 (in D2O);
21
½aꢂD ꢀ19 (c 3.7, H2O); d 4.50–4.47 (m, 1H, H-2), 4.23–
4.12 (m, 1H, H-4), 3.90–3.81 (m, 2H, H-1, H-5), 3.70
(dd, 1H, J5,6 = 9.2 Hz, J1,6 = 9.6 Hz, H-6), 2.27 (ddd,
1H, J2,3eq = 4.4 Hz, J4,3eq = 4.7 Hz, Jgem = 14.4 Hz, H-
3eq), 1.52 (m, 1H, H-3ax). [MꢀH] m/z 482,
[M+Naꢀ2H] m/z 505, [Mꢀ2H]2 m/z 241; for compound