A total synthesis of (+)-Goniodiol using an anomeric oxygen-to-carbon rearrangement

Article abstract of DOI:10.1039/a806584e

A new route to (+)-Goniodiol 1, a potent and selective cytotoxin, is described, using a diastereoselective oxygen-to-carbon rearrangement of an anomerically linked silyl enol ether as the key step.

Full text of DOI:10.1039/a806584e

View Article Online / Journal Homepage / Table of Contents for this issue
A total synthesis of (؉)-Goniodiol using an anomeric oxygen-to-
carbon rearrangement
Darren J. Dixon, Steven V. Ley* and Edward W. Tate
Department of Chemistry, University of Cambridge, Lensfield Rd., Cambridge, UK CB2 1EW
Received (in Cambridge) 21st August 1998, Accepted 27th August 1998
A new route to (؉)-Goniodiol 1, a potent and selective
cytotoxin, is described, using a diastereoselective oxygen-
to-carbon rearrangement of an anomerically linked silyl
enol ether as the key step.
O
a
b
Ph
O
OH
O
O
H
H
H
OTBDPS
OTBDPS
O
OH
Studies on natural products isolated from Asian trees of the
genus Goniothalamus have led to the discovery of several classes
of compounds with interesting biological properties, including
acetogenins, alkaloids and styrylactones. For example, (ϩ)-
Goniodiol† 1 was isolated from petroleum ether extracts of the
2
4
3
c
O
O
OH
+
O
Ph
O
Ph
H
H
H
H
O
O
Ph
H
OTBDPS
OH
OH
OTBDPS
OH
6
Goniodiol 1
5
leaves and twigs of Goniothalamus sesquipedalis,1 and shown to
have potent and selective cytotoxic activity against A-549
human lung carcinoma.2 Closely related derivatives have since
been found in a number of other Goniothalamus species.3
We have recently communicated a general method for the
introduction of carbon linked substituents adjacent to the
heteroatom in pyran ring systems via Lewis acid mediated
oxygen-to-carbon rearrangements of a variety of different
anomerically linked carbon centred nucleophiles.4a–c For the
total synthesis of (ϩ)-Goniodiol5 reported here we anticipated
that an anomeric rearrangement of this type, using a silyl enol
ether as the nucleophile, could be used to introduce important
elements of the functionality present in the target molecule. We
envisaged using a protected hydroxymethyl group opposite to
the anomeric position to control the stereochemistry at C-5 in
the rearrangement step and we hoped for some degree of con-
current diastereocontrol at C-6, similar to that seen in previous
examples.4c Furthermore, we expected that the protected
hydroxymethyl group could be efficiently converted to the
lactone present at C-1 of (ϩ)-Goniodiol in the final stages of
the synthesis.
d
Ph
Ph
e
HOOC
O
O
O
O
H
H
H
H
O
OTBDPS
O
7
8
f
Ph
OH
g
O
O
Ph
O
O
O
H
H
OH
O
1
9
The synthesis begins from commercially available S-(Ϫ)-
glycidol 2 (Scheme 1). Treatment with tert-butyldiphenylsilyl
chloride in the presence of Et3N gave the protected alcohol in
86% yield. Subsequent addition of 1.2 equivalents of but-3-
enylmagnesium bromide in the presence of 0.1 equivalents of
dilithium copper() chloride6 proceeded with exclusive attack
at the less substituted end of the epoxide to afford the corre-
sponding alkenol in 99% yield. Reductive ozonolysis of this
material afforded lactol 3 in 99% yield. Alkylation of 3 with
α-bromo-N-methyl-N-methoxyacetamide in the presence of
KHMDS afforded 81% yield, at 84% conversion, of the cis
anomerically-linked amide. Subsequent treatment with phenyl-
magnesium bromide in THF at Ϫ30 ЊC led directly to the
phenyl ketone 4 in 95% yield.7,8
With gram quantities of 4 in hand, we were in a position to
examine the key oxygen-to-carbon rearrangement step. Treat-
ment of 4 with 1.4 equivalents of Et3N followed by 1.2 equiv-
alents of trimethylsilyl triflate at 0 ЊC afforded the TMS enol
ether exclusively as the Z-isomer.9 On exposure to 0.1 equiv-
alents of TMSOTf at Ϫ30 ЊC this was smoothly converted to
the exclusively trans α-hydroxy ketones 5 and 6 (5:6, dr 1:1), as
Scheme 1 Reagents and conditions:‡ (a) i. TBDPSCl, Et3N, CH2Cl2
(86%); ii. 1.2 eq. but-3-enylmagnesium bromide, 0.1 eq. CuLi2Cl2, THF,
Ϫ30 ЊC, 5 min (99%); iii. O3, CH2Cl2, Ϫ78 ЊC, 10 min, then PPh3, rt, 12 h
(99%); (b) i. 0.5 M KHMDS in toluene, BrCH2CON(OMe)Me, THF,
Ϫ78 ЊC, 2 h (81% ϩ 16% returned 3); ii. PhMgBr, THF, Ϫ30 ЊC, 2 min
(95%); (c) i. 1.4 eq. Et3N then 1.2 eq. TMSOTf, CH2Cl2, 0 ЊC, 30 min; ii.
0.1 eq. TMSOTf, CH2Cl2, Ϫ30 ЊC, 5 min (88% combined yield over two
steps from 4); (d) i. 2 eq. NaBH4, MeOH, 0 ЊC, 5 min; ii. CH3C-
(OMe)2CH3, acetone, cat. CSA, rt, 30 min (95% over two steps from 6);
(e) i. 1 M TBAF in THF, rt, 4 h (96%); ii. DMSO, (ClCO)2, Ϫ78 ЊC, 30
t
min then Et3N, rt, 1 h (93%); iii. NaO2Cl, BuOH, H2O, KHPO4, 2-
methylbut-2-ene, rt, 10 min; (f) i. Pb(OAc)4, py, THF, rt, 1 h (68% over
two steps); ii. 0.5 eq. NaOMe, MeOH, rt, 30 min; iii. TPAP, NMO,
CH2Cl2, 4 Å sieves, rt, 10 min (97% over two steps); (g) i. 3 eq. LDA,
THF then 3 eq. PhSeCl, Ϫ78 ЊC, 1 h; ii. 30% H2O2, CH2Cl2, 0 ЊC (82%
over two steps from 9); iii. 50% aq. AcOH, 80 ЊC, 30 min (97%).
a separable mixture, in 88% overall combined yield from 4.10
Somewhat surprisingly, unlike our previous study,4c no control
is observed at the position adjacent to the ring.
The stereochemistry present at C-7 of (ϩ)-Goniodiol was
J. Chem. Soc., Perkin Trans. 1, 1998, 3125–3126
3125

View Article Online
4 (a) M. F. Buffet, D. J. Dixon, G. L. Edwards, S. V. Ley and E. W.
Tate, Synlett, 1997, 1055; (b) M. F. Buffet, D. J. Dixon, S. V. Ley and
E. W. Tate, Synlett, in press; (c) D. J. Dixon, S. V. Ley and E. W. Tate,
Synlett, in press.
5 For previous total syntheses of this molecule, see (a) M. Tsubuki,
K. Kanai and T. Honda, J. Chem. Soc., Chem. Commun., 1992,
1640; (b) J.-P. Surivet, J. Goré and J.-M. Vatèle, Tetrahedron
Lett., 1996, 37, 371; (c) C. Mukai, S. Hirai and M. Hanaoka, J. Org.
Chem., 1997, 62, 6619.
6 R. J. K. Taylor, Organocopper Reagents, Oxford University Press,
Oxford, 1994, and references cited therein.
7 Direct treatment of the lactol 3 with α-bromoacetophenone in
the presence of a variety of bases gave extensive decomposition of
the starting material.
introduced via a highly diastereoselective reduction of the
ketone moiety of 6 (>95% de) using 2 equivalents of NaBH4
in MeOH at 0 ЊC. Subsequent reaction with 2,2-dimethoxy-
propane in acetone with catalytic camphorsulfonic acid gave
the protected diol 7 in 95% yield from 6. The sequence to con-
vert the tert-butyldiphenylsilyl protected alcohol of 7 into the
α,β-unsaturated lactone of the natural product was initiated by
treatment with TBAF to release the free alcohol in 96% yield.
Oxidation to the aldehyde using Swern’s protocol11 in 93%
yield, was followed by exposure to NaO2Cl, KHPO4 and
2-methylbut-2-ene in 1:2 water–tBuOH12 to give acid 8, which
was used without further purification.
Exposure of acid 8 to lead tetraacetate13 in the presence of
pyridine in THF at room temperature afforded the anomeric
acetate in 68% yield, as a 2:1 mixture of anomers. Deacetyl-
ation using 0.5 equivalents of NaOMe in MeOH was followed
by oxidation with tetra n-propylammonium perruthenate14
(TPAP) to give the lactone 9 in 97% overall yield. Introduction
of the α,β-unsaturation was achieved via α-selenation followed
by oxidative elimination with H2O2 (82% from 9). Final depro-
tection of the C-6, C-7 diol with 50% aqueous AcOH at 80 ЊC
for 30 minutes gave the natural product (ϩ)-Goniodiol in 97%
8 S. M. Weinreb and S. Nahm, Tetrahedron Lett., 1981, 22, 3815.
9 The configuration of the silyl enol ether was tentatively assigned by
analogy with previous work.
10 Typical experimental procedure for the conversion of 4 into 5 and 6:
to a stirred solution of 4 (410 mg, 0.84 mmol) in CH2Cl2 (4.0 mL) at
0 ЊC was added Et3N (0.17 mL, 1.18 mmol) followed by TMSOTf
(0.18 mL, 0.10 mmol). After 30 min the reaction mixture was
quenched by the rapid addition of saturated NaHCO3(aq) (10 mL)
and extracted with CH2Cl2 (3 × 10 mL). Drying (anhydrous
Na2SO4), filtration and evaporation of the combined organic
extracts in vacuo gave the crude TMS enol ether which was dissolved
in CH2Cl2 (1.0 mL) and cooled to Ϫ30 ЊC. To this stirred solution
was added TMSOTf (0.015 mL, 0.084 mmol) and after 5 min at
Ϫ30 ЊC the reaction mixture was quenched by the addition of sat-
urated NaHCO3(aq) (5 mL). Extraction with CH2Cl2 (3 × 10 mL)
was followed by drying (anhydrous MgSO4), filtration and con-
centration in vacuo, to leave a yellow oil. The product ratio of 5
and 6 was determined to be 1:1 by the integration of signals at 5.17
1
yield. The H NMR, 13C NMR, IR and mass spectra of this
synthetic sample were all in excellent agreement with previously
published data.1,3 The specific rotation, [α]D30 = ϩ71.4Њ (c 0.74,
CHCl3), was also in good agreement with that reported for the
natural product, [α]D22 = ϩ74.4Њ (c 0.3, CHCl3).3
The route to (ϩ)-Goniodiol described above illustrates the
utility of the anomeric oxygen-to-carbon rearrangement in
natural product synthesis. It provides rapid and diastereo-
selective access to a densely functionalised molecule, starting
from a commercially available starting material, which was sub-
sequently converted to the desired product via a short reaction
sequence.
1
(CHOH in 5), and 4.90 (CHOH in 6) in the H NMR (600 MHz;
CDCl3) spectrum of the crude product. Purification of this oil by
medium pressure liquid chromatography (MPLC) on a Biotage
FLASH 40S column, eluting with 15% ethyl acetate–40/60
petroleum ether isolated 5 (179 mg, 44%) and 6 (182 mg, 44%) as
yellow oils. Selected spectroscopic data for 5: δH (600 MHz; CDCl3):
7.88–7.33 (15H, m, Ph), 5.17 (1H, dd, J 7.1 and 4.2, CHOH), 4.04–
4.01 (1H, m, OCH2CH), 3.89 (1H, dt, J 8.6 and 4.2, CHCHOH),
3.68 (1H, dd, J 10.4 and 6.2, OCHH), 3.63 (1H, d, J 7.1, OH), 3.57
(1H, dd, J 10.4 and 6.7, OCHH), 1.75–1.32 (6H, m, CH2CH2CH2),
1.02 (9H, s, (CH3)3Si). Selected spectroscopic data for 6: δH (600
MHz; CDCl3): 7.80–7.33 (15H, m, Ph), 4.90 (1H, dd, J 6.2 and 3.3,
CHOH), 3.93–3.90 (1H, m, CHCHOH), 3.86 (1H, m, OCH2CH),
3.73 (1H, d, J 6.2, OH), 3.46 (1H, dd, J 10.1 and 7.5, OCHH), 3.26
(1H, dd, J 10.1 and 5.3, OCHH), 1.79–1.51 (6H, m, CH2CH2CH2),
0.97 (9H, s, (CH3)3Si).
Acknowledgements
We thank the EPSRC (EWT and DJD), the Novartis Research
Fellowship (SVL) and Pfizer inc., Groton, USA for financial
support.
Notes and References
† IUPAC name: 6-(1,2-dihydroxyphenethyl)-5,6-dihydro-2-pyrone.
‡ Satisfactory acurate mass and/or microanalysis data was obtained for
all new compounds.
11 S. L. Huang, K. Omura and D. Swern, Synthesis, 1978, 297.
12 For example: J. E. Baldwin, A. K. Forrest, S. Ko and L. N.
Sheppard, J. Chem. Soc., Chem. Commun., 1987, 81.
13 L. S. Jeong, R. F. Schinazi, J. W. Beach, H. O. Kim, S. Nampalli,
K. Shanmuganathan, A. J. Alves, A. McMillan, C. K. Chu and
R. Mathis, J. Med. Chem., 1993, 36, 181.
14 S. V. Ley, J. Norman, W. P. Griffith and S. P. Marsden, Synthesis,
1994, 639.
1 B. Talapatra, S. K. Talapatra, D. Basu, T. Deb and S. Goswami,
Indian J. Chem., Sect. B, 1985, 24, 29.
2 X.-P. Fang, J. E. Anderson, C.-J. Chang, J. L. McLaughlin and
P. E. Fanwick, J. Nat. Prod., 1991, 54, 1034.
3 For example, Y.-C. Wu, C.-Y. Duh, F.-R. Chang, G.-Y. Chang,
S.-K. Wang, J.-J. Chang, D. R. McPhail, A. T. McPhail and K.-H.
Lee, J. Nat. Prod., 1991, 54, 1077.
Communication 8/06584E
3126
J. Chem. Soc., Perkin Trans. 1, 1998, 3125–3126