All-carbon intramolecular conjugate displacement reactions: An effective route to carbocycles

Article abstract of DOI:10.1002/anie.200703022

(Chemical Equation Presented) Working together: Synergy between Michael addition and S_N2′ displacement allows stabilized carbanions or the nucleophilic carbon atoms of enamines to undergo intramolecular addition to an α,β-unsaturated ester unit bearing an allylic leaving group to generate unsaturated carbocycles (see scheme). The starting esters are available by a selenium-based alternative to the classical Baylis-Hillman reaction, and complex structures can be assembled.

Full text of DOI:10.1002/anie.200703022

Angewandte
Chemie
DOI: 10.1002/anie.200703022
Synthetic Methods
All-Carbon Intramolecular Conjugate Displacement Reactions: An
Effective Route to Carbocycles**
Bodhuri Prabhudas and Derrick L. J. Clive*
A general method was recently devised in our laboratory for
making cyclic amines along the lines shown in Scheme 1.[1,2]
The ring closure formally resembles both a conjugate addition
Scheme 1. ICD process for amines. EWG=electron-withdrawing
group; PG=protecting group.
Scheme 3. Selenium-basedroute to ICD substrates; LDA =lithium
diisopropylamide, DMAP=4-dimethylaminopyridine.
and an SN2’ displacement; consequently, the designation
intramolecular conjugate displacement (ICD) seems appro-
priate.[2] An analogous ICD route to carbocycles would
clearly be useful, and we report the development of such a
method based on the Baylis–Hillman[3] substructural unit 4,
where X is a group that can support an attached nucleophilic
carbon atom (Scheme 2).
affords products of the respective ICD reactions (Table 1),
usually in over 80% yield.[6]
The reactions shown in entries i–iii of Table 1 (which were
monitored by TLC at 5 min intervals) show the dependence
of the ICD process on the nature of the leaving group.
Acetate 11 reacts more rapidly than the corresponding silyl
ether 12, and in the absence of a leaving group (Table 1,
entry iii), cyclization does not take place, at least not under
the mild conditions that are effective with the other two
substrates. The allylic leaving group and the classical Michael
acceptor appear to be mutually reinforcing, and their
combination confers a greater degree of reactivity than
either feature alone. This phenomenon has been observed
in intermolecular reactions with the pivaloate of 2-nitro-2-
propen-1-ol.[7]
Scheme 2. The all-carbon ICD process. C=tetravalent carbon atom.
We find that suitable starting materials incorporating the
essential features of 4 are easily synthesized by the general
route illustrated in Scheme 3 for a particular case.[4,5] In the
present study, the electron-withdrawing group on the
acceptor double bond was kept constant as an ester, and the
other substrates 12–20 (Table 1) were also made by the
addition of a selenium-stabilized carbanion to an aldehyde
(except for 13) and subsequent selenoxide fragmentation.
Treatment of the resulting alkenes with a base, such as DBU,
The example in Table 1, entry iv, for which a secondary
amine is required, suggests that an enamine, generated in situ
from the aldehyde and the amine, is a suitable nucleophile for
the ICD reaction. Initially, we had tried pyrrolidine
(11 mol%, PhH, reflux, 18 h, 32%), but the yield is improved
using either proline or amine 21 (see Table 1).[8]
Entry vi in Table 1 illustrates an application to bridged
systems. With short reaction times (50 min), two products, 16a
and 16b, were isolated from reaction of 16 (R = Ac or
tBuCO) with DBU, and we found that 16b was always the
same 7:3 mixture of isomers, irrespective of the stereochem-
istry at C2 of 16. Formation of 16b is reversible, and complete
conversion of 16 to 16a is achieved after a longer reaction
time (150 min). With the inorganic base Cs2CO3, only 16a was
isolated (greater than 99% yield). When 16b itself was
treated with DBU (room temperature, 12 h), it was converted
into 16a (75%). Presumably, this conversion occurs by a
cascade of inter- and intramolecular conjugate displacements:
nucleophilic attack at C4 and then attack at C2, both by DBU,
and finally the normal ICD process.
[*] Dr. B. Prabhudas, Prof. Dr. D. L. J. Clive
Chemistry Department
University of Alberta
Edmonton, Alberta, T6G 2G2 (Canada)
Fax: (+1)780-492-8231
E-mail: derrick.clive@ualberta.ca
[**] We thank the NSERC for financial support, E. J. L. Stoffman for
assistance, andDr. R. McDonaldfor the X-ray determination.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 9295 –9297
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9295
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Take advantage of blue reference links
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

Communications
Table 1: Intramolecular conjugate displacements.[a]
Scheme 4. Proposedmechanism for the formation of 17a.
account for the fact that the individual acetates 16 (R = Ac),
which differ in stereochemistry only at C2, as well as the
corresponding pivaloates (R = tBuCO), each give the same
ratio (7:3) of isomeric products (16b)—an outcome readily
accommodated by initial intermolecular conjugate displace-
ment by DBU to remove the asymmetric center at C2.
The considerable power of the ICD reaction is apparent
from entries viii–x of Table 1, which illustrate the use of this
method to solve a difficult constructional problem. These
examples (18–20) afford the unusual and strained core
framework related to the ras farnesyltransferase inhibitors
CP-225,917 and CP-263,114.[11]
In the case of mesylate 18, treatment with DBU at room
temperature brought about the desired cyclization (96%), but
the ICD reactions of 19 and 20 occurred efficiently only on
heating in MeCNin the presence of DBU. The requirement
for a higher temperature in these cases, in which C8 carries a
non-hydrogen substituent, places a minor restriction on the
choice of leaving group at C7. The mesylate corresponding to
19 (Ms instead of Ac) underwent simple 1,2-elimination, but
less potent leaving groups, such as acetate or pivaloate, are
very satisfactory.
As previously indicated, the starting materials for ICD
reactions are readily available by the selenium-based method
shown in Scheme 3,[12] and even the complicated structures
18–20 are accessible using this method. In these examples, the
aldehydes 25a–c (Scheme 5) needed for reaction with 8, or
with the corresponding Pmb ester, were made in just eight
steps from 1,1-dimethoxy-2,3,4,5-tetrabromocyclopentadiene
(22).[12] The two key reactions in this route are the intra-
molecular Horner–Emmons–Wadsworth olefination (23!
24a–c) and the desilylation with fluoride ion, which causes
spontaneous strain-assisted fragmentation to release the
required aldehydes 25a–c.
[a] Unless statedotherwise, reactions were run in MeCN at room
temperature, using 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) as the
base, and yields refer to isolated products. [b] Mixed with the D5,6
E
isomer; D6,7/D5,6 =3:1. [c] Correctedfor recovered 14. [d] Mixture of keto
andenol tautomers. [e] Tentative stereochemical assignment. [f] Mixture
of three isomers (one enol as well as cis and trans keto forms), R=Ac or
tBuCO. [g] Depending on which isomer of 16 is used. [h] R=Ac or
tBuCO, only 16a isolated. [i] R=tBuCO (more polar isomer), only 16a
isolated. [j] Mixture of isomers (including keto and enol tautomers).
[k] Major andminor isomers, respectively. [l] Single isomer; Ms =meth-
anesulfonyl, PMB=p-methoxybenzyl. [m] Single stereochemistry at C7,
both epimers at C8 (85:15). [n] Single stereochemistry at C7, both
epimers at C8 (53:47).
The products of all the ICD reactions are themselves
classical Michael acceptors and are, therefore, appropriately
Treatment of 17 with base resulted in formation of a six-
membered ring and very little, if any, of the expected eight-
membered-ring product 17b. Our working hypothesis for the
mechanism (Scheme 4) is based on precedents for consec-
utive intermolecular conjugate displacements.[9,10] This path-
way accounts for the fact that 17, which is a 1:1 mixture of C2
epimers, gives 17a as a 1:5 epimeric mixture. A similar
mechanism may apply to the formation of 16b and would
Scheme 5. Synthesis of aldehydes 25a–c. For 24 and 25, R’’=H, Me,
CH2CH2OSiPh2tBu, for a, b, and c, respectively. Reagents andcon-
ditions: a) See the Supporting Information. b) (EtO)2P(O)CH(R’’)COCl,
Et3N, then NaH, THF, approximately 608C, 57% (24a), 76% (24b),
87% (24c). c) Bu4NF, AcOH, ꢀ10 to 08C, 95% (25a), 96% (25b),
75% (25c).
9296 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 9295 –9297
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Take advantage of blue reference links
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&

Angewandte
Chemie
set up for further useful manipulations. The present method[13]
is a metal-free process; hence it offers different opportunities
from palladium-catalyzed[14,15] cyclization of allylic acetates
and should tolerate the presence of palladium-sensitive
groups.
M. V. R. Reddy, J. Org. Chem. 2003, 68, 9310 – 9316; b) T.
Tsuda, T. Yoshida, T. Saegusa, J. Org. Chem. 1988, 53, 1037 –
1040.
[6] For examples of carbon-based intermolecular conjugate dis-
placements, see: a) A. Chamakh, H. Amri, Tetrahedron Lett.
1998, 39, 375 – 378; b) reference [3b].
The experiments summarized in Table 1 establish that the
ICD process is a general and high-yielding method for
constructing a broad range of carbocycles, and the conditions
required are mild because of the enhanced reactivity that is
characteristic[7] of the simultaneous presence of both a formal
Michael acceptor and an allylic leaving group.
[7] a) P. Knochel, D. Seebach, Nouv. J. Chim. 1981, 5, 75 – 77; b) P.
Knochel, D. Seebach, Tetrahedron Lett. 1981, 22, 3223 – 3226;
c) D. Seebach, P. Knochel, Helv. Chim. Acta 1984, 67, 261 – 283;
d) D. Seebach, G. Calderari, P. Knochel, Tetrahedron 1985, 41,
4861 – 4872.
[8] a) Y. Hayashi, H. Gotoh, T. Tamura, H. Yamaguchi, R. Masui,
M. Shoji, J. Am. Chem. Soc. 2005, 127, 16028 – 16029; b) In this
preliminary study, we have not evaluated the ee value of 14a. See
also: W. J. Yang, M. T. Hechavarria Fonseca, B. List, J. Am.
Chem. Soc. 2005, 127, 15036 – 15037.
[9] a) C.-W. Cho, J.-R. Kong, M. J. Krische, Org. Lett. 2004, 6, 1337 –
1339; b) S. E. Drewes, M. M. Horn, N. Ramesar, Synth.
Commun. 2000, 30, 1045 – 1055; c) Y. M. Chung, J. H. Gong,
T. H. Kim, J. N. Kim, Tetrahedron Lett. 2001, 42, 9023 – 9026;
d) Y. J. Im, J. M. Kim, J. H. Mun, J. N. Kim, Bull. Korean Chem.
Soc. 2001, 22, 349 – 350.
[10] We did not observe any transient intermediates but appreciate
that 17a might also be formed from 17b.
[11] Review on CP molecules: D. A. Spiegel, J. T. Njardarson, I. M.
McDonald, J. L. Wood, Chem. Rev. 2003, 103, 2691 – 2727.
[12] See the Supporting Information.
[13] For the rare radical analogue of an ICD reaction, see: I. W.
Harvey, H. H. Whitham, J. Chem. Soc. Perkin Trans. 1 1993,
191 – 196.
[14] Y.-G. Suh, J.-K. Jung, S.-A. Kim, D.-Y. Shin, K.-H Min,
Tetrahedron Lett. 1997, 38, 3911 – 3914.
Received: July 7, 2007
Revised: July 31, 2007
Published online: October 30, 2007
Keywords: carbocycles · cyclization · Michael addition ·
.
SN2’ displacement · synthetic methods
[1] D. L. J. Clive, M. Yu, Z. Li, Chem. Commun. 2005, 906 – 908.
[2] D. L. J. Clive, Z. Li, M. Yu, J. Org. Chem. 2007, 72, 5608 – 5617.
[3] Reviews of Baylis–Hillman reactions: a) D. Basavaiah, A. J.
Rao, T. Satyanarayana, Chem. Rev. 2003, 103, 811 – 891; b) K. Y.
Lee, S. Gowrisankar, J. N. Kim, Bull. Korean Chem. Soc. 2005,
26, 1481 – 1490.
[4] Under standard Baylis–Hillman conditions (1,4-diazabicyclo-
[2.2.2]octane (DABCO), room temperature, 2 days), 7 gave a
complex mixture. The selenium-based method is a reliable
alternative to the classical Baylis–Hillman procedure; see
a) reference [2]; b) S.-R. Sheng, Q. Wang, Q.-Y. Wang, L. Guo,
X. Huang, Synlett 2006, 1887 – 1890; c) I. Shiina, Y. Yamai, T.
Shimazaki, J. Org. Chem. 2005, 70, 8103 – 8106.
[15] a) B. M. Trost, Angew. Chem. 1989, 101, 1199 – 1219; Angew.
Chem. Int. Ed. Engl. 1989, 28, 1173 – 1192; b) A. Heumann, M.
Rꢀglier, Tetrahedron 1995, 51, 975 – 1015; c) B. M. Trost, J. D.
Oslob, J. Am. Chem. Soc. 1999, 121, 3057 – 3064; d) B. M. Trost,
S. J. Brickner, J. Am. Chem. Soc. 1983, 105, 568 – 575.
[5] For the use of vinylalanes to generate Baylis–Hillman adducts,
see: a) P. V. Ramachandran, M. T. Rudd, T. E. Burghardt,
Angew. Chem. Int. Ed. 2007, 46, 9295 –9297
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9297
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&
&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
&
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&
Take advantage of blue reference links
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&