Comparative Effects of Conjugated and Deconjugated Isomeric Enones on the Transannular Diels-Alder Reaction

Article abstract of DOI:10.1021/ol035791z

(Equation Presented) Two isomeric cyclic trienones 2 and 3 (with four methyl esters at positions 3 and 10) underwent transannular Diels-Alder (TADA) reaction at very different temperatures. This drastic difference could be traced to the capacity of the en

Full text of DOI:10.1021/ol035791z

ORGANIC
LETTERS
2003
Vol. 5, No. 25
4799-4802
Comparative Effects of Conjugated and
Deconjugated Isomeric Enones on the
Transannular Diels−Alder Reaction
,†
,‡
Franck Caussanel,† Pierre Deslongchamps,* and Yves L. Dory*
Laboratoire de Synthe`se Organique and Laboratoire de Synthe`se Supramole´culaire,
De´partement de Chimie, Institut de Pharmacologie, UniVersite´ de Sherbrooke,
3001, 12e AVenue Nord, Sherbrooke, Que´bec J1H 5N4, Canada
pierre.deslongchamps@usherbrooke.ca
Received September 16, 2003
ABSTRACT
Two isomeric cyclic trienones 2 and 3 (with four methyl esters at positions 3 and 10) underwent transannular Diels−Alder (TADA) reaction at
very different temperatures. This drastic difference could be traced to the capacity of the enone dienophile to be conjugated or unconjugated
at the transition state. Molecular modeling could confirm this explanation. The calculated enone proved to be very distorted in transition state
ts2, and the TADA reaction temperature was accordingly much higher than the one corresponding to ts3 in which the enone was flat.
The transannular environment provides a very powerful way
to control the trajectories and the shapes of two reactive
partners facing each other across macrocycles.1 We could
take advantage of this approach to design and carry out
various syntheses of natural products, the reaction being a
transannular Diels-Alder (TADA).2 To further understand
the rules governing this powerful reaction, we designed two
isomeric macrocyclic trienones 2 and 3 (Figure 1) on the
basis of their unactivated parent 1. All these macrocycles
have CTC (cis trans diene and cis dienophile) geometry, for
the triene would yield TADA tricycle adducts having all the
same CST (cis syn trans) framework.3 Earlier studies of
dienophile activation confirmed that an enone leads to a tight
and a loose forming bond, at the transition state (TS), at the
â and R positions of the enone, respectively.4 It was then
shown that bulky groups placed across the tight bond hinder
the approach of the diene and dienophile, resulting in partial
deactivation of the TADA reaction.5 We now wish to
demonstrate that the enone must adopt a flat shape at the
TS for activation to take place.6 Thus, macrocycles 2 and 3,
which differ only from each other by the relative position
of the ketone group, would respectively fight or favor
† Laboratoire de Synthe`se Organique.
‡ Laboratoire de Synthe`se Supramole´culaire.
(1) Dory, Y. L.; Hall, D. G.; Deslongchamps, P. Tetrahedron 1998, 54,
12279-12288.
(2) (a) Germain, J.; Deslongchamps, P. J. Org. Chem. 2002, 67, 5269-
5278. (b) Toro´, A.; Deslongchamps, P. J. Org. Chem. 2003, 68, 6847-
6852. (c) Marsault, EÄ .; Toro´, A.; Nowak, P.; Deslongchamps, P. Tetrahedron
2001, 57, 4243-4260 (report 565).
(3) Lamothe, S.; Ndibwami, A.; Deslongchamps, P. Tetrahedron Lett.
1988, 29, 1641-1644.
Figure 1. CTC macrocyclic trienes 1-3.
10.1021/ol035791z CCC: $25.00 © 2003 American Chemical Society
Published on Web 11/20/2003

conjugation of the enone at the critical TS level. The results
of this investigation constitute the topic of this letter.
Macrocycle 1 has already been prepared.7 The two other
macrocycles 2 and 3 were synthesized according to a similar
general sequence of 13 linear steps starting from 3-butyn-
1-ol 4 (Scheme 1).
Scheme 2a
Scheme 1a
a Reaction conditions: (a) DHP, PTSA, CH2Cl2, 0 °C. (b)
TBDMSCl, imidazole, CH2Cl2. (c) (MeCN) 2PdCl2, DMF. (d) 2,4,6-
Collidine, MsCl, LiCl, DMF, 0 °C.
into a chloride group to finally obtain the diene 17 in 80%
yield. Having protected the alcohol 14 as its silyl ether 15
(93%), the same sequence was then applied to the vinyl
iodide 12 and the stannane 15 to get the other required diene
19 in 64% overall yield for the two steps.
a Reaction conditions: (a) DHP, PTSA, CH2Cl2, 0 °C. (b) (i)
n-BuLi, THF, -78 °C; (ii) ClCH2CHO, THF, -60 °C. (c) TIPSOTf,
2,6-lutidine, THF, 0 °C. (d) H2, Lindlar, Et2O. (e) PTSA, MeOH.
(f) MsCl, Et3N, DMAP, CH2Cl2, 0 °C. (g) NaCH(CO2Me)2, KI,
THF, DMF, 80 °C.
The malonate 11 was either alkylated with the chloride
17 or 19 (Scheme 3) to yield the trienes 20 and 21,
Thus, 4 was protected as its corresponding THP ether 5
(86%).8 The anion of 5 was prepared with n-butyllithium
and treated with freshly prepared chloro-acetaldehyde (ozo-
nolysis of allyl chloride) to yield the chlorohydrin 6 (60%),
whose alcohol group was protected as a TIPS silyl ether in
98% yield. The resulting alkyne 7 was hydrogenated with
Lindlar catalyst and hydrogen to afford the cis alkene 8
(98%).9 The THP ether was cleaved with PTSA (75%), and
the alcohol 9 was activated as its mesylate 10 with a yield
of 98%. Subsequent coupling with dimethyl malonate anion
gave the dienophile precursor 11 (80%) that was used to
finally obtain both 2 and 3.
Scheme 3a
The diene 17 is a known compound,10 but we developed
a new and more convergent synthesis (Scheme 2) from the
allylic alcohol 12.11 Thus, the vinyl iodide 13, obtained by
THP protection of the alcohol 12 (95%), was coupled with
the stannane 1412 by means of Pd(II)13 to yield the desired
diene 16 (yield: 85%) whose alcohol group was transformed
(4) (a) Birney, D. M.; Houk, K. N. J. Am. Chem. Soc. 1990, 112, 4127-
4133. (b) Fortin, S.; Barriault, L.; Dory, Y. L.; Deslongchamps, P. J. Am.
Chem. Soc. 2001, 123, 8210-8216.
(5) Bourque, EÄ .; Dory, Y. L.; Deslongchamps, P. J. Org. Chem. 2003,
68, 2390-2397.
(6) Barriault, L.; Ouellet, S. G.; Deslongchamps, P. Tetrahedron 1997,
53, 14937-14956.
(7) Cantin, M.; Xu, Y.-C.; Deslongchamps, P. Can. J. Chem. 1990, 68,
2144-2152.
(8) Roberts, C. D.; Schuetz, R.; Leumann, C. J. Synlett 1999, 6, 819-
821.
(9) (a) Lindlar, H. HelV. Chim. Acta 1952, 35, 446. (b) McEwen, A. B.;
Guttieri, M. J.; Maier, W. F.; Laine, R. M.; Shvo, Y. J. Org. Chem. 1983,
48, 4436-4438.
(10) Lavoie, R.; Ouellet, S. G.; Dallaire, C.; Dory, Y. L.; Toro´, A.;
Deslongchamps, P. Tetrahedron 2000, 56, 5509-5522.
(11) Beruben, D.; Marek, I.; Normant, J.-F.; Plazer, N. J. Org. Chem.
1995, 60, 2488-2501.
(12) Jung, M. E.; Light, L. A. Tetrahedron Lett. 1982, 23, 3851-3854.
(13) (a) Stille, J. F.; Groh, B. L. J. Am. Chem. Soc. 1987, 109, 813-
817. (b) Sheffy, F. K.; Godschalx, J. P.; Stille, J. F. J. Am. Chem. Soc.
1984, 106, 4833-4840.
a Reaction conditions: (a) NaH, THF, DMF. (b) PTSA, MeOH.
(c) Ph3P, HCA, THF, -20 °C. (d) NaH, KI, CH2E2, THF, DMF,
80 °C. (e) TBAF, THF, -20 °C. (f) Dess-Martin periodinane,
DIPEA, CH2Cl2. (g) CsI, Cs2CO3, MeCN.
4800
Org. Lett., Vol. 5, No. 25, 2003

Scheme 4a
Figure 2. TSs of triene models M1-3.
In principle, all CTC macrocycles such as 1-3 should only
give the CST adducts 32, 33, and 35.17 However, the ketones
33 and 35 can easily epimerize to give tricycles 34 and 36,
having TAT (trans anti trans) and CAC (cis anti cis)
geometries, respectively. Under the TADA reaction condi-
tions, partial epimerization occurred, yielding 1:1 mixtures
of epimers in both cases. By treating these mixtures with
PTSA in toluene at reflux, we proved that 34 and 36 were
more stable than their epimers. This latter experiment helped
us to obtain pure TAT tricycle 34, from which we could
clarify its identity by NMR spectrometry. The tricycle 35
was crystallized and its geometry proven by X-ray crystal
analysis.
Ab initio calculations (3.21G)18 were carried out on simple
models M1-3 (Figure 2) representing the macrocycles 1-3,
respectively, without their four methyl esters. In each case,
the lowest energy conformation for the macrocycle was
characterized together with the corresponding TSs (Table 1,
a Reaction conditions: (a) PTSA, toluene, reflux.
respectively (yields: 87 and 96%). The THP ether 20 was
cleaved to yield the alcohol 22 (75%) that was transformed
into the corresponding allylic chloride 23 by means of
hexachloroacetone in 98% yield.14 Dimethylmalonate was
monoalkylated with the freshly prepared chloride 23 to afford
the tetraester 24 (90%). Its silyl ether was cleaved with TBAF
(80%), and the resulting alcohol 25 oxidized to the trienone
26 using Dess and Martin conditions (85% yield).15 In
summary, the triene 20 was transformed into the trienone
26 with an overall yield of 45%. The same five-step sequence
was applied to the triene 21 to get the isomeric trienone 31
with an overall yield of 47%. Both chloroketones 26 and 31
were macrocyclized by slowly injecting them into a suspen-
sion of cesium carbonate in acetonitrile.16 The corresponding
desired macrocycles 2 and 3 were finally obtained with the
rather poor yields of 30 and 17%, respectively.
The trienone 2 had to be heated at 220 °C for 120 min to
produce tricycles 33 and 34 (Scheme 4). It is worth noting
that these conditions are almost as drastic as the ones used
to get the triene 1 to react (250 °C, 90 min). These results
suggest that the activation of the dienophile by a ketone in
macrocycle 2 is almost nonexistent or at best very weak.
On the contrary, the isomeric trienone 3 shows marked
activation of the dienophile, since it reacts at a much lower
temperature (160 °C, 120 min) yielding tricycles 35 and 36.
Table 1. Summary of Calculationsa
M1
M2
33°
51.56
2.30 Å
2.18 Å
138°
49.51
2.45 Å
2.07 Å
4°
M3
37°
44.10
2.16 Å
2.33 Å
9°
42.94
2.10 Å
2.36 Å
15°
macrocycle
TS cbc
θ (enone)
activation
bond cis
bond trans
θ (enone)
activation
bond cis
50.90
2.27 Å
2.22 Å
TS bbc
50.78
2.26 Å
2.20 Å
bond trans
θ (enone)
a θ: torsion of the enone. Energy (italics) in kcal/mol.
no zero point energy corrections were applied). On the basis
that a boat tether next to a trans ring junction is known to
be disfavored,19 we only considered TSs having either cbc
(chair boat chair) or bbc (boat boat chair) conformations for
the CST (cis syn trans) ABC ring systems under scrutiny. It
was first observed that the bbc TSs were consistently more
stable than their cbc competitors.
It was then possible to estimate and compare the reaction
energy barriers (assuming that the entropy of reaction would
be similar for all three systems).20 The calculations proved
(14) (a) Romo, D.; Meyer, S.; Johnson, D. D.; Schreiber, S. J. Am. Chem.
Soc. 1993, 115, 7906-7907. (b) Magid, R. M.; Fruchey, O. S.; Johnson,
W. L.; Allen, T. G. J. Org. Chem. 1979, 44, 359-363.
(15) Dess, D. B.; Martin, J. C. J. Am. Chem. Soc. 1991, 113, 7277-
7287.
(16) Barriault, L.; Deslongchamps, P. Bull. Soc. Chim. Fr. 1997, 134,
969-980.
(17) Lamothe, S.; Ndibwami, A.; Deslongchamps, P. Tetrahedron Lett.
1988, 29, 1639-1640.
(18) Binkley, J. S.; Pople, J. A.; Hehre, W. J. J. Am. Chem. Soc. 1980,
102, 939-947.
(19) (a) Dory, Y. L.; Soucy, P.; Drouin, M.; Deslongchamps, P. J. Am.
Chem. Soc. 1995, 117, 518-529. (b) Coe, J. W.; Roush, W. R. J. Org.
Chem. 1989, 54, 915-930.
Org. Lett., Vol. 5, No. 25, 2003
4801

TS. Again, calculations agree with the experimental data
since the activation energy for M3 is now as low as 42.94
kcal/mol, a decrease of around 7 kcal/mol with regards to
its isomeric trienone M2.
Inspection of the various TS geometries revealed interest-
ing hints to explain the marked difference of reactivity of
the isomeric enones 2 and 3 (Figure 3). The enone part of
M3 takes an almost flat cisoid shape both in its cbc (9°)
and bbc (15°) conformations, hence its strong activation
effect. On the contrary, the enone in M2 cannot adopt such
a flat arrangement in its cbc TS; in fact, its transoid
conformation deviates substantially from planarity (42° )
180° - 138°). The activation energy corresponding to this
cbc TS is accordingly high due to the lack of activation. In
the bbc TS, the enone can assume a cisoid flat conformation.
However, the energy remains rather high because the flat
enone shape can only be reached through costly backbone
strain, a fact clearly demonstrated by the overall conformation
of the TS: the forming bonds are very asymmetric (2.45
and 2.07 Å instead of 2.30 and 2.18 Å for the cbc competing
TS) and do not belong to the same plane, leading then to a
very distorted Diels-Alder TS geometry.
Figure 3. Stereoviews of TSs. (a) M2 cbc, (b) M2 bbc, (c) M3
bbc. The twisted transoid enone in the M2 cbc TS and the distorted
M2 bbc TS geometry are clearly visible.
Acknowledgment. We thank NSERC Canada and FQRNT
Que´bec for financial support.
Supporting Information Available: Experimental pro-
cedures and full characterizations for new compounds, except
for 33 and 36 (partial 1H NMR assignments from mixtures),
all coordinates (xyz) for M1-3 (macrocycles and TSs), and
a CIF file for 35. This material is available free of charge
via the Internet at http://pubs.acs.org.
to be very successful for modeling the three reactions as
primarily shown by their energy barriers. An energy of 50.78
kcal/mol was associated with the no-activation case M1, very
close to the energy calculated for M2 (49.51 kcal/mol).
Accordingly, the effect of the carbonyl in 2 should be
marginal, and this is indeed the case as indicated by the
temperatures of reaction: 2 reacts at 220 °C, only 30 °C
lower than 1. On the other hand, 3, which reacts at 160 °C,
must experience a strong activation from its carbonyl at the
OL035791Z
(20) Houk, K. N.; Li, Y.; Evanseck, J. D. Angew. Chem. 1992, 104,
711-739; Angew. Chem., Int. Ed. Engl. 1992, 31, 682-708.
4802
Org. Lett., Vol. 5, No. 25, 2003