Marrying iterative synthesis to cascading radical cyclization: 6-endo/5-exo radical cascade across bis-vinyl ethers

Article abstract of DOI:10.1021/ol202286q

Application of iterative protocols to the synthesis of functionally and stereochemically complex small molecules is an emerging area of research with the potential to create new efficiencies in complex molecule synthesis. Similarly, the discovery of tande

Full text of DOI:10.1021/ol202286q

ORGANIC
LETTERS
2011
Vol. 13, No. 20
5552–5555
Marrying Iterative Synthesis to Cascading
Radical Cyclization: 6-endo/5-exo Radical
Cascade across Bis-Vinyl Ethers
Katherine A. Davies and Jeremy E. Wulff*
Department of Chemistry, University of Victoria, Victoria, British Columbia,
Canada, V8W 3V6
wulff@uvic.ca
Received August 23, 2011
ABSTRACT
Application of iterative protocols to the synthesis of functionally and stereochemically complex small molecules is an emerging area of research
with the potential to create new efficiencies in complex molecule synthesis. Similarly, the discovery of tandem or cascade reactions can aid in the
rapid generation of new structures for biological screening programs. This report describes a cascading 6-endo-trig/5-exo-trig radical cyclization
across bis-vinyl ether substrates, which are themselves iteratively synthesized from simple building blocks.
The introduction of solid phase synthesis techniques in
the 1960s1 stimulated an explosion in the development of
efficient iterative protocols for the synthesis of polynucleic
acids (i.e., DNA and RNA)2 and polyamino acids (i.e.,
peptides or proteins).3 Ready access to these synthetic
materials revolutionized biochemistry and molecular biol-
ogy and inspired the subsequent development of related
approaches to oligosaccharides.4
Historically, the application of iterative techniques to
the synthesis of secondary metabolite natural products, or
other functionally and stereochemically complex small
molecules, has been less actively pursued;notwithstand-
ing some excellent contributions to the syntheses of
polyketides,5 “ladder”-type polyethers,6 and other select
classes of oligomeric natural products.7 More recently, the
invention of protective groups for aryl or vinyl boranes has
led to the development of powerful iterative Suzuki coup-
ling strategies for small molecule synthesis.8
As is the case for iterative synthesis, the use of tandem or
cascade reactions in the assembly of complex molecules
can significantly enhance the efficiency of the overall
(1) Merrifield, R. B. Angew. Chem., Int. Ed. Engl. 1985, 24, 799–810.
(2) Caruthers, M. H. Science 1985, 230, 281–285.
(3) Atherton, E.; Sheppard, R. C. Solid-Phase Peptide Synthesis:
A Practical Approach; Oxford Univ. Press: Oxford, U.K., 1989.
(4) For leading reviews, see: (a) Weishaupt, M.; Eller, S.; Seeberger,
P. H. Methods Enzymol. 2010, 478, 463–484. (b) Sears, P.; Wong, C.-H.
Science 2001, 291, 2344–2350. (c) Wong, C.-H. J. Org. Chem. 2005, 70,
4219–4225.
(5) (a) Paterson, I.; Scott, J. P. J. Chem. Soc., Perkin Trans. 1 1999,
1003–1014. (b) Kesavan, S.; Su, Q.; Shao, J.; Porco, J. A., Jr.; Panek, J. S.
Org. Lett. 2005, 7, 4435–4438. (c) Paterson, I.; Gottschling, D.; Menche,
D. Chem. Commun. 2005, 3568–3570.
(7) For a recent review, see: Snyder, S. A.; ElSohly, A. M.; Kontes, F.
Nat. Prod. Rep. 2011, 28, 897–924.
(8) (a) Gillis, E. P.; Burke, M. D. J. Am. Chem. Soc. 2007, 129, 6716–
6717. (b) Noguchi, H.; Hojo, K.; Suginome, M. J. Am. Chem. Soc. 2007,
129, 758–759. (c) Lee, S. J.; Anderson, T. M.; Burke, M. D. Angew.
Chem., Int. Ed. 2010, 49, 8860–8863. (d) Gillis, E. P.; Burke, M. D.
Aldrichimica Acta 2009, 42, 17–27. (e) Manabe, K.; Ohba, M.; Matsushima,
Y. Org. Lett. 2011, 13, 2436–2439.
(6) (a) Evans, P. A.; Roseman, J. D.; Garber, L. T. J. Org. Chem.
€
1996, 61, 4880–4881. (b) Marmsater, F. P.; West, F. G. J. Am. Chem.
Soc. 2001, 123, 5144–5145. (c) Kadota, I.; Takamura, H.; Sato, K.;
Yamamoto, Y. J. Org. Chem. 2002, 67, 3494–3498. (d) Trost, B. M.;
Frederiksen, M. U.; Rudd, M. T. Angew. Chem., Int. Ed. 2005, 44, 6630–
6666. (e) Simpson, G. L.; Heffron, T. P.; Merino, E.; Jamison, T. F.
J. Am. Chem. Soc. 2006, 128, 1056–1057.
(9) For recent reviews, see: (a) Nicolaou, K. C.; Chen, J. S. Chem.
ꢀ
Soc. Rev. 2009, 38, 2993–3009. (b) Toure, B. B.; Hall, D. G. Chem. Rev.
2009, 109, 4439–4486. (c) Bur, S. K.; Padwa, A. Adv. Heterocycl. Chem.
2007, 94, 1–105.
r
10.1021/ol202286q
Published on Web 09/15/2011
2011 American Chemical Society

approach by reducing the number of steps requiring
independent optimization.9
Table 1. Iterative Synthesis of Vinyl Ethersa
In principle, combining the inherently complementary
strategies of iterative synthesis and cascade methodology
could reduce the development time associated with the
synthesis of certain types of architecturally complex struc-
tures. Specifically, we propose the use of short iterative
protocols to prepare reactive oligomers, followed by the
application of a cascade cyclization to access a (poly)cyclic
target with good regiochemical and stereochemical con-
trol. In this way, the synthesis of structural analogues
should only require input of an array of simple monomeric
building blocks, while the core structures of the (poly)cycles
themselves might be altered through the use of chemically
orthogonal cascades.
In this Letter, we describe an iterative conjugate addi-
tion/reduction sequence leading to a family of oligo-vinyl
ethers (e.g., 4À10, Table 1). These highly functionalized
structures (or related congeners) should be amenable to
cascade cyclization through a variety of orthogonal trans-
formations; as a proof of concept, we demonstrate here the
conversion of bis-vinyl ethers 6 to functionalized hexahy-
dro-2H-furo[3,4-b]pyrans 15 through an unusual 6-endo-
trig/5-exo-trig radical cascade. To the best of our knowl-
edge, this is the first radical cascade reported for a bis-vinyl
ether substrate.
Inthe first iterative step, reaction of1 equiv ofalcohol (1,
4, 6, or 8, Table 1) with 1 equiv of alkyne 2 proceeded
smoothly under the influence of 10 mol % of PMe310 to
afford addition products (3, 5, 7, or 9) with generally
high yield and good selectivity for the E isomer.11 With
more hindered electrophiles (particularly those bearing
a trifluoromethyl group), the E:Z selectivity was sub-
stantially eroded; in these cases (3f and 5i) the two
isomers were separated chromatographically before
proceeding. Depending on the substitution at the R1
and R2 positions, bis-vinyl ethers 5 were occasionally
prone to unwanted Claisen rearrangements. These were
particularly problematic when the R2 substituent was an
electron-withdrawing group (unless R1 was also an
electron-withdrawing group, as in the case of 5i), or an
aromatic group. We therefore focused our methodology
on substrates incorporating alkyl groups or protons at
the R2 position. After reduction with diisobutylalumi-
num hydride (DIBAL), rearrangement was no longer
observed.
Products 3À10 were purified over triethylamine-treated
silica. However, both the addition and reduction steps
yielded relatively pure material for most substrates (ca.
90% purity by NMR) even in the absence of column
chromatography; simple filtration through basic alumina
was sufficient to remove impurities in the addition reaction,
a Conditions: (i) CH2Cl2, 0 to 23 °C, 24 h. (ii) Et2O, À78 to À40 °C,
4 h. b For compounds with more than one vinyl ether, the E:Z ratio refers
the ratio of the all-E product to all other adducts. c Pure E isomer,
separated from 59% of the corresponding Z isomer. d Isolated along
with 22% of a transesterification product. e Pure E isomer, separated
from 58% of the corresponding Z isomer.
(10) Inanaga, J.; Baba, Y.; Hanamoto, T. Chem. Lett. 1993, 241–244.
(11) A similar sequence of reactions was recently used by our group to
prepare a family of juvenile hormone mimics; see: Davies, K. A.; Kou,
K. G. M.; Wulff, J. E. Tetrahedron Lett. 2011, 52, 2302–2305.
while aqueous workup removed most of the aluminum-based
byproducts following DIBAL reduction. This lack of
required chromatographic purification steps suggests
that the iterative conjugate addition/reduction sequence
ꢀ
(12) (a) Lubineau, A.; Auge, J.; Bellanger, N.; Caillebourdin, S.
J. Chem. Soc., Perkin Trans. 1 1992, 1631–1636. (b) Krompiec, S.;
ꢀ
Kuznik, N.; Urbala, M.; Rzepa, J. J. Mol. Catal. A: Chem. 2006, 248,
198–209. Also see ref 11.
Org. Lett., Vol. 13, No. 20, 2011
5553

described here could form the basis for an automated or
semiautomated assembly of oligo-vinyl ethers.
Barring a few reports of Claisen rearrangements,12 the
chemistry of bis-vinyl ethers like 6 is unknown. We opted
to first explore their reactivity in radical cascade reactions,
since this would provide an opportunity to directly com-
pare the behavior of bis-vinyl ethers with that of the
analogous polyene systems, for which numerous radical
cascades have been described previously.13 To better un-
derstand the reactivity of these substrates, we also briefly
investigated the radical cyclization of monovinyl ethers 4a
and 4b. Following methylation14 of the alcohol in 4a (to
provide 11, Scheme 1), treatment with triphenyltin hydride
and azobisisobutyronitrile (AIBN) in refluxing benzene
afforded a good yield of 12, consistent with results for
related systems.15 No 5-exo product was observed. The
methyl ether of 4b also provided mostly the corresponding
6-endo product, but with less selectivity. Subsequent reac-
tions were exclusively carried out with alkyl or aryl sub-
stituents at the R1 position.
Scheme 1. Radical Cyclization onto a Mono-Vinyl Ether
Table 2. Radical Cyclization across Bis-Vinyl Ethersa
Moving on to the cascading cyclization of bis-vinyl
ethers (6, and alkylated derivatives thereof), we found that
many such substrates engaged in a highly selective 6-endo-
trig/5-exo-trig reaction to provide 15.16 Not surprisingly,
the reaction leading to 15 was most efficient for substrates
in which R2 was a proton (Table 2, entries 1À5), presum-
ably owing to the reduced steric hindrance for the final
bond-forming step.
Substrates bearing alkyl groups at the R2 position were
more resistant to bicycle formation (Table 2, entries 6À9).
Nonetheless, the use of high dilution (Table 2, entry 7)
provided an acceptable yield of 15f (R1 = Me, R2 = Me),
containing two adjacent quaternary centers. Further in-
creasing the size of the substituents (Table 2, entries 8 and
9) blocked the formation of 15, such that only monocyclic
products 16g and 16h were isolated cleanly. The presence
of a free hydroxyl group (Table 2, entry 10) or a benzyl
function in place of the terminal methyl ether (entry 11)
did not adversely affect the yield, but in both cases the
diastereoselectivity was eroded.
Assignment of the relative stereochemistry for 15 was
complicated by the presence of minor isomers visible in the
NMR spectra. To simplify the assignment, representative
structures 15a and 15f were subjected to protodestannyla-
tion with aqueous HCl (Scheme 2). The structures of
the products were fully assigned by 2D NMR methods
(see Supporting Information); assignments of the major
a Conditions: (i) Ph3SnH (1.5 equiv), AIBN (0.1 equiv) added over
6 h to a refluxing solution of 14 in benzene. Heating was continued
for 1.5 h prior to workup. Major diastereomers are shown for products
15 and 16. b Reported values refer to the ratio between the major
observed product and the next most abundant species; refer to the
Supporting Information for more details. c Product was not observed
in g10% yield.
products in Table 2 were made by analogy. Destannylation
of 15a (as a 4:1 ratio of major isomers, each of which could
be further resolved by NMR into a mixture of secondary
isomers) afforded both possible diastereomers of product
(17a and 18a, again in a 4:1 ratio), while destannylation of
15f (as a 5:1 mixture of isomers) provided a single diaster-
eomer (17f) in excellent yield. These results indicate that
the bicyclic structure of 15f is formed as a single diaster-
eomer and that the minor isomer visible spectroscopically
corresponds to a second geometric isomer at the double
bond. By contrast, the less-substitued analogues (15a À 15e,
(13) (a) Handa, S.; Pattenden, G. J. Chem. Soc., Perkin Trans. 1 1999,
843–845. (b) Handa, S.; Nair, P. S.; Pattenden, G. Helv. Chim. Acta
2000, 83, 2629–2643.
(14) See Supporting Information for alkylation conditions.
(15) (a) Munt, S. P.; Thomas, E. J. J. Chem. Soc., Chem. Commun.
1989, 480–482. (b) Maguire, R. J.; Munt, S. P.; Thomas, E. J. J. Chem.
Soc., Perkin Trans. 1 1998, 2853–2864.
(16) The methyl ether of 6i (R1 = R2 = CF3) reacted through a
different process that we have tentatively assigned as a 5-exo-trig/β-
scission, suggesting the possibility that all of the substrates in Table 2
arise through initial 5-exo cyclization, followed by rearrangement to the
more stable radical intermediate, as has been previously suggested for
other 6-endo-trig reactions. Further mechanistic investigations are
underway.
5554
Org. Lett., Vol. 13, No. 20, 2011

as well as 15i and 15j) are formed as mixtures of epimers at
the carbon bearing R2, as well as geometric isomers at the
double bond. No evidence of trans-fused products was
observed.
Scheme 3. Proposed Mechanism for the Radical Cascade
Scheme 2. Protodestannylations
2H-pyran core during the final cyclization event (as repre-
sented by 20a, Scheme 3).20 This picture is consistent with
both the increase in selectivity for 15f over 15a (a methyl
group in the R2 position is more easily accommodated anti
to the R1 substituent) and the decrease in selectivity for 15e
(in which the bulky protecting group in the R1 position
may present an obstacle to the allylic ether).
Hexahydrofuropyrans of the type described here (and
oxidized forms thereof) comprise the core structures of
several natural products21 and have also been utilized in
medicinal chemistry programs.22 Their selective formation
through a high-yielding radical cascade reaction across
iteratively synthesized vinyl ether precursors provides a
compelling validation of the potential efficiency that may
be achieved by uniting iterative synthesis and cascade
cyclization methodologies.
As described in a series of papers by Pattenden, polyenes
that are structurally related to 6a (but which lack polarized
double bonds) react in radical cascades through a series of
6-endo-trig cyclizations, a fact that has been attributed to
steric control.13,17 Thus, the first ring-closing event in the
synthesis of 15 (i.e., 19 f 20, Scheme 3) is in keeping with
the reactivity observed for the corresponding olefinic
system (and is, likewise, probably due largely to steric
factors),16 while the second ring closure takes the opposing
5-exo pathway. We attribute this regiochemical switch to
the electronic properties of intermediate 20 (Scheme 3);
due to electronic donation from the adjacent oxygen atom
to the radical-bearing carbon, the energy of the SOMO is
increased, thereby enhancing the nucleophilicity of the
radical.18 Although a second 6-endo-trig cyclization (to
afford 22) would be favored on steric grounds (at least for
derivatives of 6a) this pathway is disfavored electronically,
since it would require the nucleophilic radical to add to the
nucleophilic end of the vinyl ether function. Addition to
the more hindered, but less nucleophilic, end of the vinyl
ether group in a 5-exo-trig cyclization19 (to provide 21) is
therefore favored on electronic grounds.
Acknowledgment. We are grateful to NSERC (Canada),
Merck Frosst, and the University of Victoria for financial
support.
Supporting Information Available. Experimental pro-
cedures including spectroscopic and analytical data for
new compounds. This material is available free of charge
via the Internet at http://pubs.acs.org.
Regarding the relatively high levels of diastereoselectiv-
ity observed in the cascade reaction (particularly in the
formation of 15f), we hypothesize that the allylic ether side
chain is preferentially oriented away from the tetrahydro-
(21) (a) Shibano, M.; Okuno, A.; Taniguchi, M.; Baba, K.; Wang,
N.-H. J. Nat. Prod. 2005, 68, 1445–1449. (b) Li, C.-Y.; Wu, T.-S. J. Nat.
Prod. 2002, 65, 1452–1456. (c) Seaforth, C. E.; Mohammed, S.; Maxwell,
A.; Tinto, W. F.; Reynolds, W. F. Tetrahedron Lett. 1992, 33, 4111–
4114. (d) Oulad-Ali, A.; Guillaume, D.; Jiang, Y.; Weniger, B.; Anton,
R. Nat. Prod. Lett. 1993, 2, 203–207. (e) Krohn, K.; Biele, C.; Drogies,
(17) Less substituted 1,5-dienes (where both alkenes were disubstituted)
were reported to engage in tandem6-endo-trig/5-exo-trig cyclization, but
the presence of additional substituents favored the 6-endo pathway; see:
Batsanov, A.; Chen, L.; Gill, G. B.; Pattenden, G. J. Chem. Soc., Perkin
Trans. 1 1996, 45–55.
€
K.-H.; Steingrover, K.; Aust, H.-J.; Draeger, S.; Schulz, B. Eur. J. Org.
(18) Fleming, I. Molecular Orbitals and Organic Chemical Reactions;
John Wiley & Sons: West Sussex, U.K., 2010; pp 369À400.
€
Chem. 2002, 2331–2336. (f) Qin, S.; Krohn, K.; Florke, U.; Schulz, B.;
ꢀ
Draeger, S.; Pescitelli, G.; Salvadori, P.; Antus, S.; Kurtan, T. Eur. J.
(19) 5- and 6-exo radical cyclizations onto vinyl ethers bearing
electron-withdrawing directing groups are well-known; see for example:
(a) Lee, E. Pure Appl. Chem. 2005, 77, 2073–2081. (b) Leeuwenburgh,
Org. Chem. 2009, 3279–3284. (g) Gao, X.; Nakadai, M.; Snider, B. B.
Org. Lett. 2003, 5, 451À454 and references therein.
(22) See for example: (a) Stewart, A. M.; Meier, K.; Schulz, B.;
Steinert, M.; Snider, B. B. J. Org. Chem. 2010, 75, 6057–6060.
(b) Ghosh, A. K.; Chapsal, B. D.; Baldridge, A.; Steffey, M. P.; Walters,
D. E.; Koh, Y.; Amano, M.; Mitsuya, H. J. Med. Chem. 2011, 54, 622–
634. (c) Fathi, A. R.; Krautheim, A.; Kaap, S.; Eger, K.; Steinfelder,
H. J. Bioorg. Med. Chem. Lett. 2000, 10, 1605–1608.
ꢀ
M. A.; Litjens, R. E. J. N.; Codee, J. D. C.; Overkleeft, H. S.; van der
Marel, G. A.; van Boom, J. H. Org. Lett. 2000, 2, 1275–1277.
(20) Anomeric radicals generally react through axial approach of the
radical acceptor; see for example: Abe, H.; Shuto, S.; Matsuda, A. J. Am.
Chem. Soc. 2001, 123, 11870À11882 and references therein.
Org. Lett., Vol. 13, No. 20, 2011
5555