Anti aldol selectivity in a synthetic approach to the C1-C 12 fragment of the tedanolides

Article abstract of DOI:10.1021/ol702729u

In a synthetic approach to the completely protected C₁ C ₁₂ fragment of the macrocyclic cytotoxic agent tedanolide 1, we carried out the tin-catalyzed Mukaiyama aldol reaction between the 2,3-dialkoxypropanal 5 and the silyl enol eth

Full text of DOI:10.1021/ol702729u

ORGANIC
LETTERS
2008
Vol. 10, No. 1
137-140
Anti Aldol Selectivity in a Synthetic
Approach to the C1−C12 Fragment of the
Tedanolides
Michael E. Jung* and Ting-hu Zhang
Department of Chemistry and Biochemistry, UniVersity of California,
Los Angeles, California 90095-1569
jung@chem.ucla.edu
Received November 10, 2007
ABSTRACT
In a synthetic approach to the completely protected C1−C12 fragment of the macrocyclic cytotoxic agent tedanolide 1, we carried out the
tin-catalyzed Mukaiyama aldol reaction between the 2,3-dialkoxypropanal 5 and the silyl enol ether 6 derived from the ketone 7, which gave,
unexpectedly, the anti aldol isomer, rather than the expected syn isomer 4, as the major diastereomer formed.
In 1984, Schmitz and co-workers1 isolated tedanolide 1 from
the Caribbean sponge Tedania ignis and reported that it
Scheme 1
showed very high cytotoxicity, with ED50 values of 250
pg/mL against human nasopharynx carcinoma and 16
pg/mL against in vitro lymphocytic leukemia. Seven years
later, Fusetani isolated 13-deoxytedanolide, which also dis-
played very potent cytotoxic effects.2 Owing to its structural
complexity and biological activity, tedanolide has generated
considerable synthetic interest,3 including two total syntheses
and significant synthetic work. Over the past few years, we
have employed the non-aldol aldol process4 in several
approaches to tedanolide and its analogues. By using a
straightforward retrosynthetic disconnection of the tedanolide
skeleton involving cleavage at the lactone moiety and scission
at the C12-C13 bond, we were able to generate the precursors
(1) Schmitz, F. J.; Gunasekera, S. P.; Yalamanchili, G.; Hossain, M. B.;
van der Helm, D. J. Am. Chem. Soc. 1984, 106, 7251.
(2) Fusetani, N.; Sugawara, T.; Matsunaga, S.; Hirota, H. J. Org. Chem.
1991, 56, 4971.
(3) (a) Total synthesis of tedanolide: Ehrlich, G.; Hassfeld, J.; Eggert,
U.; Kalesse, M. J. Am. Chem. Soc. 2006, 128, 14038. (b) Synthetic study
of tedanolide: Iwata, Y.; Tanino, K.; Miyashita, M. Org. Lett. 2005, 7,
2341. Ehrlich, G.; Kalesse, M. Synlett 2005, 655. Hassfeld, J.; Kalesse, M.
2 and 3 (Scheme 1). Recently, we reported two approaches
to the C1-C12 fragment of tedanolide 2, both of which used
the non-aldol aldol process and either a highly stereoselective
Synlett 2002, 2007. Roush, W. R.; Newcom, J. S. Org. Lett. 2002, 4, 4739.
Taylor, R. E.; Hearn, B. R.; Ciavarri, J. P. Org. Lett. 2002, 4, 2953. Loh,
T.-P.; Feng, L.-C. Tetrahedron Lett. 2001, 42, 6001. Loh, T.-P.; Feng,
L.-C. Tetrahedron Lett. 2001, 42, 3223. Smith, A. B.; Lodise, S. A. Org.
Lett. 1999, 1, 1249. Roush, W. R.; Lane, G. C. Org. Lett. 1999, 1, 95.
(4) (a) Jung, M. E.; Lee, C. P. Org. Lett. 2001, 3, 333. (b) Jung, M. E.;
Lee, C. P. Tetrahedron Lett. 2000, 41, 9719. (c) Jung, M. E.; Marquez, R.
Org. Lett. 2000, 2, 1669. (d) Jung, M. E.; Marquez, R. Tetrahedron Lett.
1999, 40, 3129. (e) Jung, M. E.; Karama, U.; Marquez, R. J. Org. Chem.
1999, 64, 663.
10.1021/ol702729u CCC: $40.75
© 2008 American Chemical Society
Published on Web 12/08/2007

syn aldol reaction or a stereoselective vinyllithium addition
coupled with a stereoselective hydroboration-protonation
scheme.5 While these routes are quite efficient, we neverthe-
less also investigated concurrently other possible routes to
prepare the same “top half” fragment of tedanolide since this
piece is a common intermediate for both tedanolide and 13-
deoxytedanolide. We now report an attempted synthesis of
this unit in which a novel anti-selective tin-catalyzed
Mukaiyama aldol reaction was revealed.6
ether 11,9 effected by treatment of ethyl phenyl ketone 10
with lithium diphenylamide and the silyl chloride, with the
aldehyde 5 under Mukaiyama conditions, namely in the
presence of stannic chloride in dichloromethane, afforded,
in 89% yield, mainly the desired syn aldol diastereomer 12
with essentially none of the anti isomer 13 (Scheme 3). This
Scheme 3
We thought that 2 could be prepared in a few steps from
the silyl ether 4, which could be formed by a route using as
the key constructive step the Mukaiyama aldol reaction
between the 2,3-dialkoxypropanal 5 and the silyl enol ether
6, the latter easily prepared from the ketone 7 (Scheme 2).
Scheme 2
result agrees well with the report of Reetz on a very similar
system.8b The structure of 12 was proven (Scheme 4) by
Scheme 4
reduction of the ketone with DIBAL to give a mixture of
diols which were cyclized to the acetonide 14, using 2,2-
dimethoxypropane (DMP) and camphorsulfonic acid (CSA),
which was purified by preparative TLC. The very small
coupling constants of Ha and Hc and Hb and Hc (J ) 2.0
Hz) in the proton NMR indicated that all three protons were
cis and that we had indeed obtained the syn isomer 12.10
However, when the known11 Z-trimethylsilyl enol ether
16, prepared from diethyl ketone 15 with trimethylsilyl iodide
and triethylamine, was reacted under the same conditions
with 5, a 93% yield of a >10:1 ratio favoring the anti product
18 over the syn 17 (Scheme 5) was obtained. Thus the two
analogous cases proceed to give the opposite diastereose-
lectivity. The structure of 18 was proven by the following
Thus the commercially available optically pure ester 9 was
converted in 13 steps to 8 by our earlier method. This ketone
was prepared from the bromo-tetrahydrofurfuryl aldehyde
8, the preparation of which we have reported previously,4c
by a four-step route using straightforward chemistry.7 The
ketone 7 has all the carbons from C4 to C11 and the three
correct chiral centers at C6, C7, and C10 of tedanolide. To
synthesize the top fragment 4 efficiently, a syn selective aldol
or Mukaiyama aldol reaction of the enolate or the enol ether
derived from 7, e.g., 6, with the aldehyde 5 was required.
The results of our study of that condensation follow.
Before proceeding with the reaction of the enol ether 6 of
the complex ketone 7 with the aldehyde 5, we first examined
the reaction of simpler ketones with 5. Literature reports of
such reactions of simple silyl enol ethers with R-alkoxy
aldehydes indicated that, depending on the exact case and
conditions, predominantly the syn aldol product was ob-
tained.8 Thus reaction of the known Z-trimethylsilyl enol
Scheme 5
(5) (a) Jung, M. E.; Yoo, D. Org. Lett. 2007, 9, 3543-3546. (b) Jung,
M. E.; Yoo, D. Tetrahedron Lett. In press.
(6) (a) Mukaiyama, T.; Banno, K.; Narasaka, K. J. Am. Chem. Soc. 1974,
96, 7503. For reviews, see: (b) Chan, T.-H. Formation and Addition
Reactions of Enol Ethers. In Comp. Org. Synth. 1991, Ch. 2.3, 595. (c)
Gennari, C. Asymmetric Synthesis with Enol Ethers. In Comp. Org. Synth.
1991, Ch. 2.4, 629.
(7) Ethylmagnesium bromide was added to the aldehyde 8 followed by
Dess-Martin oxidation to the ketone. Reductive ring-opening with zinc in
acetic acid and silylation of the diol with TBSCl gave 7.
138
Org. Lett., Vol. 10, No. 1, 2008

a 1:5 ratio favoring the anti isomer 24c in 82% yield while
the TBS enol ether 22d gave a higher 1:7 ratio again favoring
the anti isomer 24c.
Scheme 6
The structures of the two major diastereomeric products
24ac were proven via the coupling constant analysis of the
acetonides as described in detail above.
Interestingly, additional stereocenters in the ethyl ketone
fragment do not disturb the anti selectivity. Thus the known12
silyl enol ether 26, prepared from the ketone 25, reacted with
the aldehyde 5 to give an 88% yield of only the anti
diastereomer 28 (Scheme 8). Similarly treatment of the silyl
Scheme 8
two routes (Scheme 6). DIBAL reduction and acetonide
formation with TLC separation afforded the acetonide 19.
Proton coupling analysis indicated that protons Ha and Hc
were trans diaxial (J ) 10.5 Hz) while protons Hb and Hc
were cis (J ) 1.5 Hz). The differences in the NMR spectra
of 14 and 19 were very clear. The syn relationship between
the benzyloxy group and the alcohol was shown by preparing
the second acetonide 20 via first protection of the alcohol
as the TBS ether, reduction of the ketone with DIBAL, and
protection of the resulting alcohol as the methyl ether.
Removal of the two silyl groups (TBAF), acetonide forma-
tion, and finally separation afforded 20, in which protons
Ha and Hb showed a small coupling (J ) 1.5 Hz) therefore
indicating they were cis.
enol ether 29, derived from a two-step silylation of the ethyl
ketone 18, which was prepared as in Scheme 5, with the
aldehyde 5 afforded only the anti diastereomer 30 in 83%
yield (Scheme 9). The stereochemistry of 30 was easily
Next, we investigated the condensation of the silyl enol
ethers 22a-d prepared from a series of ethyl alkyl ketones
21 with the aldehyde 5 in the presence of stannic chloride
(Scheme 7). Whereas the TMS enol ether 22a gave relatively
Scheme 9
Scheme 7
assigned by silylation of the free alcohol to give the C2-
symmetric ketone 31.
Finally, even though it was apparent that anti selectivity
would be observed, nonetheless we tried the requisite key
step for the synthesis of 4, namely the Mukaiyama aldol
condensation of the silyl enol ether 6 (prepared from the
ketone 7) with the aldehyde 5 in the presence of stannic
chloride (Scheme 10). The condensation afforded, in 78%
isolated yield, the anti diastereomer 32 as the major product,
which is the epimer of the desired syn diasteromer 4 at C4.
The following reasons may be postulated for the observed
selectivity. Theoretical calculations indicate that the two Z
poor selectivity (1:2), the corresponding TBS enol ether 22b
gave a 1:9 ratio favoring the anti diastereomer 24a in 71%
yield. Likewise the cyclohexyl TMS enol ether 22c furnished
(8) (a) Heathcock, C. H.; Davidsen, S. K.; Hug, K. T.; Flippin, L. A. J.
Org. Chem. 1986, 51, 3027. (b) Reetz, M. T.; Kesseler, K.; Jung, A.
Tetrahedron 1984, 40, 4327.
(9) Xie, L. F.; Vanlandeghem, K.; Isenberger, K. M.; Bernier, C. J. Org.
Chem. 2003, 68, 641.
(10) Extensive decoupling experiments were carried out to ensure the
identity of the indicated protons in all the acetonides.
(11) Cazeau, P.; Duboudin, F.; Moulines, F.; Babot, O.; Dunogues, J.
Tetrahedron 1987, 43, 2088.
(12) Denmark, S. E.; Fujimori, S.; Pham, S. M. J. Org. Chem. 2005, 70,
10823.
Org. Lett., Vol. 10, No. 1, 2008
139

However the situation changes with the phenyl substituent:
the syn transition state D is calculated to have a lower ∆G
than that of the opposite anti transition state C by roughly
0.7 kcal/mol. Here, although the magnitude of the energy
difference does not correspond to the ratio seen (all syn),
the trend is at any rate in the right direction. Compared to
A, the syn TS B is disfavored, mainly due to repulsion
between the alkyl group R and the chiral C2 carbon of the
aldehyde. However, the reasons for the smaller difference
in energy between the phenyl substituted case, D and C, are
less obvious. From a careful examination of the transition
structures, it can be seen that in C, the silyl enol ether rotates
from a perfectly staggered conformation to avoid interaction
between the phenyl and a chloride on the tin, which
introduces steric repulsion between the methyl and again the
chiral C2 carbon of the aldehyde. This causes this normally
favored transition state now to be higher in energy than its
syn counterpart. Further theoretical studies will be needed
to gain more insight into this difference.
Scheme 10
transition states suggested by Heathcock,8a namely the Z-A3
and the Z-S1 transition states, have very different energies
depending on whether the substituent is an ethyl or a phenyl
group. Thus B3LYP/6-31G(d) level calculations13 give the
differences shown in Figure 1, specifically the anti transition
In conclusion, we have observed a novel change in the
diasteroselectivity of the tin-catalyzed Mukaiyama aldol
reaction between an R-alkoxy aldehyde and the Z trialkylsilyl
enol ether of ethyl ketones from completely syn when the
opposite group is phenyl to mainly or completely anti when
the opposite group is alkyl. This effect is seen even when
the opposite group is a relatively large alkyl unit and has
additional alkyl or oxygen stereocenters. Finally, we have
carried out theoretical calculations which provide some
rationale for this novel switch in the diastereoselectivity.
Further developments toward the total synthesis of tedanolide
will be published in due course.
Acknowledgment. We thank Professor Ken Houk and
Joann Um (UCLA) for the theoretical calculations. We thank
the NIH and NSF for generous support.
Supporting Information Available: Experimental pro-
cedures and spectral data (proton and carbon NMR, IR) for
all new compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
Figure 1. Energy Differences of Possible Transition States.
OL702729U
state A (Heathcock’s Z-A3) is calculated to have a lower
∆G than that of the opposite syn transition state B (Heath-
cock’s Z-S1) by roughly 2.4 kcal/mol. That energy difference
is in good agreement with the product ratios observed.
(13) These calculations were done on a simpler model system where
the enol silyl ether was replaced by an enol, the OTBDPS group by an
OH, and the OBn by an OMe group with the R group being ethyl (R )
Et).
140
Org. Lett., Vol. 10, No. 1, 2008