A new carbohydrate-based synthetic approach to trichothecenes. Synthesis of a bicyclic BC core of verrucarol from D-galactose

Article abstract of DOI:10.1016/S0040-4039(01)01366-1

An advanced bicyclic BC intermediate towards the total synthesis of verrucarol has been prepared from D-galactose via an intramolecular aldehyde-allylstannane reaction.

Full text of DOI:10.1016/S0040-4039(01)01366-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 6679–6682
A new carbohydrate-based synthetic approach to trichothecenes.
Synthesis of a bicyclic BC core of verrucarol from
D-galactose
Mikhail S. Ermolenko*
Institut de Chimie des Substances Naturelles du CNRS, Avenue de la Terrasse, 91198 Gif-sur-Yvette, France
Received 30 June 2001; revised 24 July 2001; accepted 25 July 2001
Abstract—An advanced bicyclic BC intermediate towards the total synthesis of verrucarol has been prepared from D-galactose via
an intramolecular aldehyde–allylstannane reaction. © 2001 Elsevier Science Ltd. All rights reserved.
The trichothecenes constitute a large family of metabo-
lites isolated from imperfect fungi and plants of the
genus Baccharis.1 They display a wide range of biologi-
cal activity, including antiviral, antibacterial, antifun-
gal, insecticidal and phytotoxic, and several members of
this family are listed among the most cytotoxic agents
known.2 Control programs for mycotoxins, implicated
in a number of serious threats to human and animal
health, and search for biologically active natural prod-
ucts led to isolation of almost two hundred tri-
chothecenes. With a few exceptions, the trichothecene
sesquiterpenes share a common trichothecane skeleton
(1), oxidized and esterified at different positions, with
the verrucarol nucleus 2 being the most frequently
found.
As potential anticancer agents, the trichothecenes have
been the subject of intense studies on both structure
modification2a,2b,3 and total synthesis.4 A large amount
of synthetic approaches to trichothecenes has been dis-
closed so far, and several representative naturally
occurring trichothecenes, trichodiene,5 12,13-epoxytri-
chotec-9-ene 3,6 trichodermol 4,7 trichodermin 5,8 ver-
rucarol 2,9 calonectrin 6,10 anguidine 7,11 and sporol,12
were synthesized, most of them in racemic form. In
view of the functionality pattern present, the annelation
of the C ring by an intramolecular aldol reaction
seemed to be a rational way to the tricyclic skeleton of
trichothecenes. Indeed, the first total synthesis of a
trichothecene, trichodermin 5, was accomplished, in
very low yield, by this ‘group 1 aldol’ approach,8a but
all attempts to extend this strategy to verrucarol 2
failed due to the easy isomerization of the required
axial aldehyde intermediate to the equatorial one via a
reversible Michael reaction under the basic reaction
conditions (Scheme 1)13,14
H
16
H
2
O
13
O
O
9
11
3
7
12
4
HO
H
5
15
HO
14
1
2 Verrucarol
As an alternative, an intramolecular aldehyde–allylstan-
nane reaction under neutral conditions, by thermolysis,
H
H
H
O
O
O
O
O
O
O
R = H
7%
R1
O
O
O
H
R2
R3
H
O
O
O
O
H
H
R
R
R
OH
R = OH
O
O
3 R1 = R2 =R3 = H
4 R1 = R3 = H, R2 = OH
O
H
5
R1 = R3 = H, R2 = OAc
HO
H
H
6 R2 = H, R1 = R3 = OAc
7 R1 = R2 = R3 = OAc
O-
HO
Roritoxin B
IG50 L1210 1.7 ng/ml
O
O
O
O
R
* Fax: +33-1 69 07 72 47; e-mail: mikhail.ermolenko@icsn.cnrs-gif.fr
Scheme 1.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01366-1

6680
M. S. Ermolenko / Tetrahedron Letters 42 (2001) 6679–6682
seemed plausible. This paper, inspired by some early
results from Fraser–Reid’s laboratory,14 describes the
first truly successful cyclization along the C4ꢀC5 bond
and the synthesis of an advanced bicyclic BC core of
ligand exchange18 with the tin cuprate followed by an
elimination to give the triene i‡ (Scheme 2). It was
found that the allylic substitution reaction (13“14) is
faster in diethyl ether, and upon dilution of the tin
verrucarol from
D
-galactose.
cuprate
reagent,
conveniently
prepared
from
Bu2CuLi·LiCN and Bu3SnH in THT,19 with ether the
gratifying yield of the required allylstannane 14 (87%
based upon recovered starting material (14%)) was
obtained after 18 h at −78°C and the quenching of the
reaction at the same temperature.
C-Allyl galactopyranoside 8, obtained as a mixture of
a/b isomers (9:1) by the published procedure,15 was
deacetylated and then selectively silylated on the pri-
mary hydroxyl group to give 9. Treatment of the latter
with trimethyl orthoacetate set a temporary orthoester
protection onto the cis diol and then, after the alkyla-
tion of a free hydroxyl group with para-methoxyben-
zyl (MPM) chloride and hydrolysis during the usual
aqueous acid workup, delivered the acetyl group to
the axial hydroxyl16 to give the selectively protected
C-allyl galactopyranoside 10, easily separable from the
minor stereo- and regio-isomers by chromatography.
Swern oxidation of 10 followed by the treatment with
MeMgCl afforded the diol 11 with the required axial
hydroxyl group at C5.† Oxidation of the secondary
hydroxyl group in 11 with the Dess–Martin periodi-
nane and protection of the tertiary one as the
methoxymethyl (MOM) ether gave the ketone 12. No
other reagent allowed oxidation of 11 in good yield;
however, the transformation of the diol 11 into the
ketone 12 by a four-step sequence (1. Ac2O/Py; 2.
MOMCl, i-Pr2NEt/CH2Cl2; 3. MeONa/MeOH; 4.
Swern oxidation) was also efficient (overall yield 70%).
Scheme 2.
With the strategic allylstannane functionality in place,
the delicate elaboration of a cyclization precursor, the
acid- and base-sensitive aldehyde 16,§ was next
addressed. In accordance with expectations, steric con-
straints and electronic20 factors prevailed over a con-
ventional order of olefin reactivity,21 and the ordinary
OsO4-catalyzed dihydroxylation of 14 in aqueous t-
BuOH proceeded at the terminal double bond exclu-
sively. Unfortunately, the reaction was very sluggish,
and over-catalytic amounts of OsO4 (0.2 equiv.) were
required for complete conversion of the starting 14 to
the diol 15 in a reasonable time (2 days). Because of
the fragile functionality present, no attempts were
made to liberate the diol from the residual osmate
ester during workup at the end of reaction. All the
above led to the significant loss of the valuable
product during the chromatographic purification of
the diol 15¶ that was absolutely necessary before its
periodate cleavage to the aldehyde 16. Eventually, the
reaction in the presence of quinuclidine21b (0.2 equiv.)
allowed to circumvent this drawback by keeping the
quantity of OsO4 as low as 0.05 equiv. at 24 h reac-
tion time with otherwise good yield of 86%. The pure
diol 15 was then cleaved by NaIO4 in aqueous MeOH,
and the crude aldehyde 16 thus obtained was refluxed
in xylene for 1 h to give the bicyclic product 17 in
72% overall yield from diol 15.
The Wittig methylenation of 12 to 13 was sluggish
despite the fast consumption of the starting ketone.
Under numerous conditions tried the yield of the
alkene 13 did not exceed 60%, while some highly polar
adduct persisted—presumably one of two Wittig
betains, either restricted in attaining the reacting con-
formation or intercepted by the proximate silyl group.
Under forcing conditions, the b-elimination of the
TBDPS-OH became also a problem. In any event, the
low efficiency of the reaction could definitely be
attributed to basicity and size of the Wittig reagent.
Indeed, the reaction of Me3SiCH2Li with the ketone
12 in the presence of CeCl3 gave a mixture of two
alcohols (45:55) which, upon treatment with KHMDS,
afforded the desirable alkene 13 in much better overall
17
yield. Careful preparation of anhydrous CeCl3 is
essential for success of the first step.
In accordance with prediction from a cyclic transition
state model,22 the configuration of the newly created
carbinol center in 17 proved to be unnatural and
required inversion. The problem of inversion of
configuration at C4 on the full-featured ABC tricyclic
trichothecene intermediates has already been addressed
in an early synthesis of verrucarol.
With the allyl ether 13 in hand, the elaboration of the
allylstannane 14 was then addressed. The reaction of
13 with a higher order cuprate, (Bu3Sn)2CuLi·LiCN,
in THF was found to be extremely temperature-depen-
dent rendering the usual process control methods (e.g.
TLC) inefficient. Very slow at −78°C, this reaction
went to completion within minutes at −20°C to give
the allylstannane 14, which underwent a consecutive
‡ Hexabutylditin was also found in this reaction.
§ No acid treatments should be applied to any allylstannane interme-
diates of the scheme to avoid the elimination of tin and MPMO
moieties (“i); besides, the aldehyde 16 undergo the fast isomeriza-
tion to an equatorial dead-end isomer under the mild basic condi-
tions, as mentioned above.
† The trichothecene numbering is used throughout the paper. For
stereoelectronic reasons, the axial orientation of this hydroxyl is
required for a later reaction 13“14. The structure of 11 was solved
by NMR, including NOE observed between the methyl group at C5
and the protons at C6 and C12.
¶ Theoretically, up to 2 mole/mole of osmium used.

M. S. Ermolenko / Tetrahedron Letters 42 (2001) 6679–6682
6681
As a consequence of the quaternary center C6 present,
the complex hydrides attack trichothec-9-en-4-ones
from the b-face exclusively,23 and the only solution to
the problem was found, in rather low yield, by a
nucleophilic substitution.9c,d On the contrary, the
flattened B ring of the bicyclic system obtained permit-
ted a beneficial a-side reagent approach. Indeed, the
reduction of the 4-keto derivative 18 with LiBHEt3
afforded the required isomeric alcohol as the sole
product and in high overall yield. The characteristic
NOE between the protons of the TBS and MPM
groups in 19 proved its natural configuration at C4.
possibility, the alkene 19 was subjected to OsO4-cata-
lyzed dihydroxylation to give a mixture of diols (ca.
1:1) which were protected by isopropylidene group (20).
Usually more robust, the TBDPS protection was selec-
tively removed in the presence of TBS group by treat-
ment with NaOH in aqueous DMPU,24 and a separable
mixture of alcohols 21 was oxidized to the aldehyde
22. The latter smoothly eliminated acetone upon heat-
ing with Ca(OH)2 in aqueous MeOH to give the lactol
23 with no open-chain g-hydroxyaldehyde present, as
evidenced by NMR spectra. Finally, oxidation of
23 afforded the a,b-unsaturated lactone 24.25 (Scheme
3).
The enantiomerically pure bicyclic product 19, obtained
in 15 steps and overall yield of 17%, provides a plethora
of possibilities for total synthesis of verrucarol 2 and
related trichothecenes. A very attractive one consists of
building the C ring of verrucarol by a ring-closing
olefin metathesis reaction (“25“2). To explore this
The pursuit of the total synthesis of verrucarol 2 by the
conjugate addition/ring-closing metathesis sequence
from the lactone 24 and its open-chain derivatives, or
by the cycloaddition onto 19, 23 and 24 will be reported
in due course.
O
OAc
OAc
O
O
O
AcO
AcO
d, e
AcO
AcO
TBDPSO
HO
TBDPSO
AcO
b
a
c
80%
87%
OAc
OH
85%
OMPM 90%
OAc
OAc
OH
OH
8
9
10
O
O
O
O
f, g
h, i
k
j
TBDPSO
HO
TBDPSO
TBDPSO
TBDPSO
Bu3Sn
12
84%
83%
87%
86%
OMPM
6
OMPM
O
OMPM
OMPM
5
OH
OMOM
OMOM
11
12
13
14
OTBDPS
O
OTBDPS
O
O
O
O
TBDPSO
Bu3Sn
OH
n, o
TBDPSO
Bu3Sn
l
m
OH
MPM
O
MPM
O
89%
72%
from 15
97%
OMPM
OMPM
HO
4
O
H
15
16
17
18
OTBDPS
O
OTBDPS
O
OH
O
CHO
O
H
O
H
O
O
O
O
s
p, q
r
m
O
O
MPM
O
MPM
O
MPM
O
MPM
79%
from 21
74%
87%
H
H
TBSO
TBSO
TBSO
TBSO
19
20
21
22
metathesis
OH
O
O
O
O
O
O
O
O
O
O
m
HO
MPM
MPM
MPM
O
86%
HO
H
H
H
H
HO
TBSO
TBSO
TBSO
23
24
25
2
Scheme 3. Reagents and conditions: (a) CH2ꢁCHCH2SiMe3 (3 equiv.), BF3·Et2O (5 equiv.)/CH3CN, 4°C, 48 h (Ref. 15); (b)
MeONa (cat)/MeOH, then TBDPSCl (1.2 equiv.), ImH (2.4 equiv.)/DMF; (c) CH3C(OMe)3 (2 equiv.), CSA (cat), rt; then Et3N
quench and concentration; then MPMCl (1.2 equiv.), NaH (2 equiv.), Bu4NI (0.05 equiv.)/THF, rt; aqueous acid workup; (d)
Swern oxidation; (e) MeMgCl (4 equiv.)/THF; −78°C to rt; (f) Dess–Martin periodinane (1.1 equiv.)/CH2Cl2, rt; (g) MOMCl (3
equiv.), i-Pr2NEt (4 equiv.)/CH2Cl2, D; (h) 12+CeCl3 (2 equiv.)/THF, rt, 0.5 h; then add Me3SiCH2Li/pentane (2 equiv.), −10°C
to rt; (i) (Me3Si)2NK (1.1 equiv.)/THF, rt; (j) (Bu3Sn)2CuLi·LiCN (1.6 equiv.)/THF–Et2O (1:4), −78°C, 18 h; (k) NMO (1.5
equiv.), OsO4 (0.05 equiv.), quinuclidine (0.2 equiv.)/t-BuOH–H2O (3:1), rt, 24 h; (l) NaIO4/MeOH–H2O (95:5), rt; then xylene,
,
140°C, 1 h; (m) NMO (1.5 equiv.), TPAP (0.05 equiv.), MS 4 A/CH3CN; (n) LiBHEt3 (1.2 equiv.)/THF, −78°C; (o) TBSCl (1.2
equiv.), ImH (2.4 equiv.)/DMF, rt; (p) NMO (1.5 equiv.), OsO4 (0.05 equiv.)/t-BuOH–H2O (3:1), rt; (q) DMP, CSA (cat.)/
CH2Cl2, rt; (r) 3 M NaOH–DMPU (1:9), 40°C, 3 h; (s) Ca(OH)2/MeOH–H2O, D.

6682
M. S. Ermolenko / Tetrahedron Letters 42 (2001) 6679–6682
1988, 53, 5586; (i) Ishihara, J.; Nonaka, R.; Terasawa, Y.;
References
Shiraki, R.; Yabu, K.; Kataoka, H.; Ochiai, Y.; Tadano,
K. Tetrahedron Lett. 1997, 38, 8311; (j) Ishihara, J.;
Nonaka, R.; Terasawa, Y.; Shiraki, R.; Yabu, K.;
Kataoka, H.; Ochiai, Y.; Tadano, K. J. Org. Chem. 1998,
63, 2679; (k) White, J. D.; Kim, N.-S.; Hill, D. E.;
Thomas, J. A. Synthesis 1998, 619.
1. (a) Ueno, Y. Trichothecenes-Chemical, Biological and
Toxicological Aspects; Elsevier: New York, 1983; (b)
Grove, J. F. Nat. Prod. Rep. 1988, 5, 187; (c) Grove, J. F.
Nat. Prod. Rep. 1993, 10, 429.
2. (a) Doyle, T. W.; Bradner, W. T. In Anticancer Agents
Based on Natural Product Models; Cassady, J. M.;
Douros, J. D., Eds.; Academic Press: New York, 1980,
pp. 43; (b) Jarvis, B. B.; Yatawara, C. S. J. Org. Chem.
1986, 51, 2906.
3. Jarvis, B. B.; Mazzola, E. P. Acc. Chem. Res. 1982, 15,
388.
4. McDougal, P. G.; Schmuff, N. R. In Fortschr. Chem.
Org. Naturst.; Zechmeister, L. et al. Eds.; Springer:
Wien–New York, 1985; Vol. 47, pp. 153.
5. (a) Welch, S. C.; Prakasa Rao, A. S. C.; Gibbs, C. G.;
Wong, R. Y. J. Org. Chem. 1980, 45, 4077; (b) Suda, M.
Tetrahedron Lett. 1982, 23, 427; (c) Schlessinger, R. H.;
Schultz, J. A. J. Org. Chem. 1983, 48, 407; (d) Harding,
K. E.; Clement, K. S. J. Org. Chem. 1984, 49, 3871; (e)
Gilbert, J. C.; Wiechman, B. E. J. Org. Chem. 1986, 51,
258; (f) Gilbert, J. C.; Kelly, T. A. J. Org. Chem. 1986,
51, 4485; (g) Kraus, G. A.; Thomas, P. J. J. Org. Chem.
1986, 51, 503; (h) Van Middlesworth, F. L. J. Org. Chem.
1986, 51, 5019; (i) Pearson, A. J.; O’Brien, M. K. J.
Chem. Soc., Chem. Commun. 1987, 11, 1445; (j) Snowden,
R. L.; Brauchli, R.; Sonnay, P. Helv. Chim. Acta 1989,
72, 570; (k) Gilbert, J. C.; Kelly, T. A. Tetrahedron Lett.
1989, 30, 4193; (l) Pearson, A. J.; O’Brien, M. K. J. Org.
Chem. 1989, 54, 4663; (m) Harding, K. E.; Clement, K.
S.; Tseng, C.-Y. J. Org. Chem. 1990, 55, 4403; (n)
Tanaka, M.; Sakai, K. Tetrahedron Lett. 1991, 32, 5581;
(o) Gilbert, J. C.; Seiliah, R. D. J. Org. Chem. 1993, 58,
6255.
6. (a) Fujimoto, Y.; Yokura, S.; Nakamura, T.; Morikawa,
T.; Tatsuno, T. Tetrahedron Lett. 1974, 2523; (b)
Masuoka, N.; Kamikawa, T. Tetrahedron Lett. 1976,
1691; (c) Hua, D. H.; Venkataraman, S.; Chan-Yu-King,
R.; Paukstelis, J. V. J. Am. Chem. Soc. 1988, 110, 4741.
See also Ref. 5l.
7. (a) Still, W. C.; Tsai, M.-Y. J. Am. Chem. Soc. 1980, 102,
3654; (b) O’Brien, M. K.; Pearson, A. J.; Pinkerton, A.
A.; Schmidt, W.; Willman, K. J. Am. Chem. Soc. 1989,
111, 1499. See also Ref. 51.
10. (a) Kraus, G. A.; Roth, B.; Frazier, K.; Shimagaki, M. J.
Am. Chem. Soc. 1982, 104, 1114; (b) Tomioka, K.; Sugi-
mori, M.; Koga, K. Chem. Pharm. Bull. 1987, 35, 906.
11. (a) Brooks, D. W.; Grothaus, P. G.; Palmer, J. T. J. Org.
Chem. 1982, 47, 2820; (b) Brooks, D. W.; Grothaus, P.
G.; Palmer, J. T. Tetrahedron Lett. 1982, 23, 4187; (c)
Brooks, D. W.; Grothaus, P. G.; Mazdiyasni, H. J. Am.
Chem. Soc. 1983, 105, 4472.
12. Ziegler, F. E.; Metcalif, III, C. A.; Schulte, G. Tetra-
hedron Lett. 1992, 33, 3117.
13. Colvin, E. W.; Malchenko, S.; Raphael, R. A.; Roberts,
J. S. J. Chem. Soc., Perkin Trans. 1 1978, 658.
14. Fraser-Reid, B.; Tsang, R. In Natural Product Chemistry
1984; Zalewski, R. I.; Skolik, J. J., Eds.; Elsevier:
Amsterdam, 1985, pp. 197.
15. Giannis, A.; Sandhoff, K. Tetrahedron Lett. 1985, 26,
1479.
16. Lemieux, R. U.; Drigues, H. J. Am. Chem. Soc. 1975, 97,
4069.
17. Dimitrov, V.; Kostova, K.; Genov, M. Tetrahedron Lett.
1996, 37, 6787.
18. Lipshutz, B. H.; Crow, R.; Dimock, S. H.; Ellsworth, E.
L.; Smith, R. A. J.; Behling, J. R. J. Am. Chem. Soc.
1990, 112, 4063.
19. Lipshutz, B. H.; Ellsworth, E. L.; Dimock, S. H.; Reuter,
D. C. Tetrahedron Lett. 1989, 30, 2065.
20. Solladie´, G.; Urbano, A.; Stone, G. Tetrahedron Lett.
1993, 34, 6489.
21. (a) Xu, D.; Crispino, G. A.; Sharpless, K. B. J. Am.
Chem. Soc. 1992, 114, 7570; (b) Andersson, P. G.; Sharp-
less, K. B. J. Am. Chem. Soc. 1993, 115, 7047.
22. Denmark, S. E.; Hosoi, S. J. Org. Chem. 1994, 59, 5133.
23. Adams, P. M.; Hanson, J. R. J. Chem. Soc., Perkin
Trans. 1 1972, 2283.
24. Smith, III, A. B.; Zhuang, L.; Brook, C. B.; Boldi, A. M.;
McBriar, M. D.; Moser, W. H.; Murase, N.; Nakayama,
K.; Verhoest, P. R.; Lin, Q. Tetrahedron Lett. 1997, 38,
8667.
8. (a) Colvin, E. W.; Raphael, R. A.; Roberts, J. S. J. Chem.
Soc., Chem. Commun. 1971, 858; (b) Colvin, E. W.;
Malchenko, S.; Raphael, R. A.; Roberts, J. S. J. Chem.
Soc., Perkin Trans. 1 1973, 1989.
9. (a) Schlessinger, R. H.; Nugent, R. A. J. Am. Chem. Soc.
1982, 104, 1116; (b) Trost, B. M.; Rigby, J. H. J. Org.
Chem. 1978, 43, 2938; (c) Trost, B. M.; McDougal, P. G.
J. Am. Chem. Soc. 1982, 104, 6110; (d) Trost, B. M.;
McDougal, P. G.; Haller, K. J. J. Am. Chem. Soc. 1984,
106, 383; (e) Roush, W. R.; D’Ambra, T. E. J. Org.
Chem. 1980, 45, 3929; (f) Roush, W. R.; D’Ambra, T. E.
J. Org. Chem. 1981, 46, 5045; (g) Roush, W. R.;
D’Ambra, T. E. J. Am. Chem. Soc. 1983, 105, 1058; (h)
Koreeda, M.; Ricca, D. J.; Luengo, J. I. J. Org. Chem.
25. All new compounds were characterized by analytical and
spectroscopic (including 2D NMR) methods. 24 had [h]D
+11.0 (c 1.0, CHCl3); 1H NMR (250 MHz, CDCl3): l
7.27 and 6.86 (4H, ArH
H-4), 4.72 and 4.65 (2H, 2d, J=16.6 Hz, 2×H-15), 4.61
and 4.51 (2H, 2d, J=11.8 Hz, OCH2Ar), 4.10 (1H, dd,
J=6.9, 2.5, H-2), 3.88 (1H, br.s, H-12), 3.81 (3H, s,
ArOCH3), 2.55 (1H, ddd, J=15.7, 6.9, 0.8 Hz, H-3 exo),
2.27 (1H, ddd, J=15.8, 5.8, 2.5 Hz, H-3 endo), 1.20 (3H,
s, CH3-14), 0.88 (9H, s, SiC(CH3)3), 0.047 and 0.041 (6H,
2d, Si(CH3 2
) ); 13C NMR (62.9 MHz, CDCl3): l 166.3,
6 ), 4.72 (1H, dd, J=5.8, 0.8 Hz,
6
6
6
6
6
159.1, 138.7, 137.1, 130.3, 128.5, 113.8, 83.8, 80.6, 78.2,
71.8, 65.7, 55.1, 48.2, 45.2, 25.6, 18.0, 10.6, −4.6, −5.0.