Cascade reactions involving formal [2+2] thermal cycloadditions: Total synthesis of artochamins F, H, I, and J

Article abstract of DOI:10.1002/anie.200702363

(Chemical Equation Presented) State of the artochamins: Total syntheses of artochamins F (1), H (2), I (3), and J (4; see scheme) have been achieved through a flexible and expedient strategy that features a cascade sequence involving two concurrent [3,3]

Full text of DOI:10.1002/anie.200702363

Angewandte
Chemie
DOI: 10.1002/anie.200702363
Cascade Reactions
Cascade Reactions Involving Formal [2+2] Thermal Cycloadditions:
Total Synthesis of Artochamins F, H, I, and J**
K. C. Nicolaou,* Troy Lister, Ross M. Denton, and Christine F. Gelin
Dedicated to Masakatsu Shibasaki on the occasion of his 60th birthday
The Artocarpus genus encompasses approximately 60 species
of trees that are distributed throughout the tropical regions of
Asia. Selected members of this family of plants have been
used as traditional folk medicines in Sri Lanka, Taiwan,
Thailand, and Indonesia.[1–4] A search for the bioactive
ingredients of these plants led to the isolation of several
bioactive prenylated flavanoids from the roots of Artocarpu-
s chama,[5] and more recently a number of weakly cytotoxic
prenylated stilbenes and their derivatives.[6] Artochamins F,
H, I, and J (1–4, Scheme 1) are among the most structurally
fascinating members of this group of compounds. Herein we
describe the total synthesis of all four natural products
through an expedient route that involves a cascade sequence
featuring a novel formal [2+2] thermal cycloaddition reac-
tion.
The unique bicyclo[3.2.0]heptane carbon frameworks of
artochamins H–J (2–4) would appear to be biogenetically
derived from artochamin F (1), or a derivative thereof,
through a formal [2+2] cycloaddition reaction between the
stilbene alkene and one of the prenyl groups.[6] Furthermore,
the racemic nature[7] of 2–4 could implicate a non-enzymatic
process for this transformation, given the pairwise enantio-
topic relationship between both the top and bottom faces of
the two prenyl moieties. We were intrigued by the possibility
of accomplishing the required cycloaddition under thermal
conditions[8] since we speculated that the realization of such a
reaction would lead to a concise synthetic approach to the
artochamins from stilbenes 5a and/or 5b through a cascade
sequence that involved, in addition to the key cyclobutane-
forming process, two consecutive Claisen rearrangements.
Furthermore, protecting-group design on the intermediates
was expected to modulate the cascade reaction and allow
selective pathways so as to deliver any of the four targeted
natural products (1–4).
The first objective was the stereoselective construction of
an appropriately functionalized stilbene derivative from
which the formal [2+2] cycloaddition reaction within the
projected cascade could be investigated. The Julia–Kocienski
olefination[9] was chosen[10] as the key step for this initial
construction as shown in Scheme 2.
Scheme 1. Structures of artochamins F (1) and H–J (2–4). Boc=tert-
butylcarbonate, TBS=tert-butyldimethylsilyl.
[*] Prof. Dr. K. C. Nicolaou, Dr. T. Lister, Dr. R. M. Denton, C. F. Gelin
Department of Chemistry and
The Skaggs Institute for Chemical Biology
The Scripps Research Institute
10550 North Torrey Pines Road, La Jolla, CA 92037 (USA)
Fax: (+1)858-784-2469
E-mail: kcn@scripps.edu
and
Department of Chemistry and Biochemistry
University of California, San Diego
9500 Gilman Drive, La Jolla, CA 92093 (USA)
Thus, the substituted phenyltetrazolesulfones 7a and 7b
were prepared from 3,4-dihydroxybenzaldehyde (6) in a
straightforward manner that involved protection (either as
the corresponding Boc derivative or silyl ether), reduction
with NaBH4, and Mitsunobu coupling with 1-phenyl-1H-
tetrazole-5-thiol, followed by molybdenum-catalyzed oxida-
tion of the resulting sulfides to the desired sulfones 7a (73%
overall yield from 6) and 7b (77% overall yield from 6). The
preparation of the other required coupling partner, aldehyde
10, began with methyl ester 8,[11] which was converted into the
corresponding bis-reversed prenylated ester 9 through a
copper-catalyzed etherification[12] with 1,1-dimethylpropynyl
carbonate and Lindlar hydrogenation (72% overall yield).
Chemoselective reduction of 9 with LiAlH4, followed by
oxidation with DMP led to aldehyde 10 in 92% overall yield
for the two steps. Finally, treatment of sulfone 7a with
[**] We wish to thank Dr. D. H. Huang, Dr. G. Siuzdak, and Dr. R. J.
Chadha for assistance with NMR spectroscopy, mass spectrometry,
and X-ray crystallography, respectively. Financial support for this
work was provided by grants from the National Institutes of Health
(USA) and the National Science Foundation (No. 06032187), the
Skaggs Institute for Chemical Biology, and a National Science
Foundation predoctoral fellowship (to C.F.G.).
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, 7501 –7505
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
7501

Communications
geometry of 5a could not be discerned from NMR spectro-
scopic data and was therefore determined by X-ray crystallo-
graphic analysis[13] (see ORTEP drawing, Scheme 2). Inter-
estingly, the Julia–Kocienski olefination between the bis(tert-
butyldimethylsilyl) sulfone 7b and aldehyde 10 delivered
stilbene 5b with similar efficiency (95% yield), but lower
stereoselectivity (E/Z ca. 2:1).
With an appropriately functionalized stilbene substrate in
hand, the projected cascade sequence could be investigated. It
was discovered that when 5a was subjected to microwave
heating[14] at 1808C in the presence of catalytic amounts of
Ph3PO in ortho-xylene for 20 minutes, the tetracyclic core 15
of the artochamins H, I, and J was formed in 55% overall
yield as shown in Scheme 3. A reasonable mechanism for this
remarkable transformation that generates two new rings and
three stereogenic centers (ca. d.r. 5:1 in favor of 15 with
respect to the benzylic stereocenter) is depicted in Scheme 3
(Table 1). Thus, rapid Claisen rearrangements[15] are followed
by collapse of the two Boc groups leading to 14. The stilbene
is then converted into 15 through a formal [2+2] cyclo-
addition reaction. The precise role of Ph3PO remains to be
determined,[16] as do the mechanistic details of the cyclo-
addition reaction itself (see below).
The completion of the synthesis of artochamins H, I, and J
from 15, which served as a common intermediate for all three
Scheme 2. Construction of the common intermediate stil-
benes 5a and 5b, and an ORTEP drawing of 5a with the
thermal ellipsoids at the 30% probability level. Reagents and
conditions: a) Boc2O (2.0 equiv), DMAP (0.05 equiv), iPr2NEt
(0.1 equiv), THF, 238C, 2 h; or TBSCl (2.0 equiv), imidazole
(2.0 equiv), DMF, 238C, 2 h; b) NaBH4 (1.0 equiv), EtOH,
238C, 30 min; c) 1-phenyl-1H-tetrazole-5-thiol (1.0 equiv),
DEAD (1.2 equiv), PPh3 (1.1 equiv), THF, 08C, 6 h; d) ammo-
nium molybdate tetrahydrate (0.1 equiv), H2O2, (20 equiv),
EtOH, 0!238C, 14 h, 73% for 7a, 77% for 7b over 4 steps;
e) methyl 1,1-dimethyl-2-propynyl carbonate (3.0 equiv),
CuCl2 (0.01 equiv), DBU (3.0 equiv), CH3CN, 08C, 14 h; f) H2
(balloon), Lindlar cat. (10% w/w), quinoline (0.5 equiv),
EtOAc, 238C, 3 h, 72% over 2 steps; g) LiAlH4 (1.2 equiv),
THF, 08C, 5 mins; h) DMP (1.2 equiv), CH2Cl2, 238C, 30 min,
92% over 2 steps; i) 7a or 7b (1.5 equiv), KHMDS
(1.55 equiv), THF, ꢀ788C, 30 min; then 10 (1.0 equiv), THF,
ꢀ78!238C, 3 h, 80% for 5a (E isomer only), 95% for 5b
(E/Z ca. 2:1). DBU=1,8-diaxabicyclo[5.4.0]undec-7-ene,
DEAD=diethylazodicarboxylate, DMAP=4-dimethylamino-
pyridine, DMF=N,N-dimethylformamide, DMP=Dess–
Martin periodinane; KHMDS=potassium bis(trimethylsilyl)-
amide.
KHMDS, followed by addition of aldehyde 10,
resulted in the formation of the desired (E)-stilbene
Scheme 3. Construction of the artochamin skeleton 15 through a microwave-
promoted cascade sequence. Reagents and conditions: a) ortho-xylene, Ph3PO
5a with complete stereoselectivity (80% yield). The (5% w/w), microwave, 1808C, 20 min, 55%.
7502
www.angewandte.org ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim Angew. Chem. Int. Ed. 2007, 46, 7501 –7505

Angewandte
Chemie
Table 1: Selected physical properties for 5a, 15, 18, and 20.
5a: Rf =0.27 (silica gel, 15% Et2O in hexanes; IR nmax (film) 2981w,
1767s, 1582w, 1506w, 1414w, 1370w, 1253s, 1156s, 1115m, 1072w cmꢀ1
1H NMR (400 MHz, CDCl3): d=7.37 (d, J=2.4 Hz, 1H), 7.32 (dd,
;
J=8.4, 2.4 Hz, 1H), 7.23 (d, J=8.4 Hz, 1H), 6.95 (s, 2H), 6.83 (s, 2H),
6.22 (dd, J=17.6, 10.8 Hz, 2H), 5.25 (dd, J=17.6, 0.8 Hz, 2H), 5.19 (dd,
J=10.8, 0.8 Hz, 2H), 1.56 (s, 9H), 1.55 (s, 9H) 1.53 ppm (s, 12H);
13C NMR (125 MHz, CDCl3) d=150.8, 150.7, 144.6, 144.4, 144.3, 144.3,
142.8, 141.9, 135.9, 135.5, 129.5, 127.4, 127.3, 124.6, 123.3, 120.9, 120.9,
120.8, 113.9, 113.8, 112.7, 83.9, 82.0, 27.8, 27.7, 27.0, 26.9 ppm; HRMS
(ESI TOF): m/z calcd for C34H43BrO8 [M+H]+: 659.2214; found
659.2197.
15: Rf =0.13 (silica gel, 5% MeOH in CH2Cl2); IR nmax (film) 3475br.s,
2956m, 2928m, 2855m, 1719w, 1600m, 1517m, 1441s, 1367m, 1285s,
1249s, 1188m, 1158m, 1105m, 1093m, 966w, 804w, 786w, 766w,
741w cmꢀ1; 1H NMR (500 MHz, CDCl3): d=6.82 (d, J=8.0 Hz, 1H),
6.76 (d, J=2.0 Hz, 1H), 6.67 (dd, J=8.0, 2.0 Hz, 1H), 5.48 (s, 1H), 5.25
(s, 1H), 5.15 (br.s, 1H), 5.00 (m, 1H), 3.95 (apt t, J=6.0 Hz, 1H), 3.13
(dd, J=16.5, 1.5 Hz, 1H), 3.09 (dd, J=15.5, 6.0 Hz, 1H), 3.01 (dd,
J=16.5, 9.5 Hz, 1H), 3.00 (m, 1H), 2.89 (d, J=6.0 Hz, 1H), 2.80 (m,
1H), 1.61 (s, 3H), 1.54 (s, 3H), 1.05 (s, 3H), 0.76 ppm (s, 3H);
13C NMR (150 MHz, CDCl3) d=149.8, 149.3, 146.6, 143.2, 141.6, 134.6,
133.5, 122.8, 121.9, 120.4, 115.6, 115.1, 115.0, 97.6, 57.4, 44.7, 44.6, 38.7,
30.5, 27.3, 27.0, 25.8, 25.7, 17.5 ppm; HRMS (ESI TOF): m/z calcd for
C24H27BrO4 [M+H]+: 459.1165; found 459.1145.
18: Rf =0.46 (silica gel, 20% Et2O in hexanes); IR nmax (film) 3483br.w,
2956m, 2929m, 2860m, 1716m, 1596w, 1505w, 1479w, 1462w, 1393w,
1367w, 1259w, 1226w, 1201w, 1120s, 1029w, 943w, 895w cmꢀ1; 1H NMR
(500 MHz, CDCl3): d=7.10 (m, 2H), 6.98 (m, 1H), 5.9 (d, J=10.2 Hz,
1H), 5.61 (s, 1H), 5.36 (d, J=10.2 Hz, 1H), 3.98 (apt t, J=6.0 Hz, 1H),
3.12 (dd, J=16.5, 2.0 Hz. 1H), 3.02 (d, J=6.0 Hz, 1H), 3.00 (dd,
J=16.5, 9.6 Hz, 1H), 2.85 (m, 1H), 1.38 (s, 3H), 1.36 (s, 9H), 1.35 (s,
9H), 1.35 (s, 3H), 1.09 (s, 3H), 0.77 ppm (s, 3H); 13C NMR (150 MHz,
CDCl3) d=176.1, 175.9, 149.3, 148.3, 144.2, 142.2, 140.7, 139.7, 135.7,
128.3, 125.5, 125.3, 123.0, 122.7, 122.4, 119.5, 110.1, 97.7, 56.7, 45.0,
43.4, 39.2, 39.2, 34.2, 30.4, 30.3, 29.7, 28.1, 27.5, 27.3, 27.3, 27.1, 25.8,
16.2 ppm; HRMS (ESI TOF): m/z calcd for C34H41BrO6 [M+H]+:
625.2159; found 625.2141.
Scheme 4. Total synthesis of 2, 3, and 4 from the common intermediate 15.
Reagents and conditions: a) PivCl (3.0 equiv), Et3N (6.0 equiv), CH2Cl2, 238C,
30 min, 75%; b) PivCl (2.0 equiv), py (6.0 equiv), CH2Cl2, 238C, 24 h, 64%;
c) PhSeCl (1.0 equiv), CH2Cl2, 238C, 1 h; then H2O2 (20 equiv), CH2Cl2, 08C,
1 h 85% over 2 steps; d) K2CO3 (5.0 equiv), MeI (6.0 equiv), DMF, 1008C,
1 h; e) LiAlH4 (6.0 equiv), THF, 238C, 14 h, 88% for 2, 91% for 3, and 88%
for 4 over 2 steps. Piv=trimethylacetyl, py=pyridine.
20: Rf =0.26 (silica gel, 50% EtOAc in hexanes); IR nmax (film) 3364br.m,
2956m, 2925m, 2856m, 2504br.w, 1697m, 1602m, 1516m, 1440m,
1365m, 1255m, 1193m, 1160m, 1109m, 1088s, 971m, 870m, 828m,
806m, 782m cmꢀ1; 1H NMR (600 MHz, [D6]acetone): d=6.77 (d,
J=7.8 Hz, 1H), 6.76 (d, J=1.8 Hz, 1H), 6.60 (dd, J=7.8, 1.8 Hz, 1H),
6.25 (s, 1H), 4.97 (m, 1H), 3.90 (apt t, J=6.0 Hz, 1H), 3.03–2.98 (m,
2H), 2.88–2.84 (m, 1H), 2.84 (dd, J=16.0, 9.0 Hz, 1H), 2.80 (d,
J=6.0 Hz, 1H), 2.72–2.68 (m, 1H), 1.49 (s, 3H), 1.38 (s, 3H), 1.00 (s,
3H), 0.71 ppm (s, 3H); 13C NMR (150 MHz, [D6]acetone) d=156.6,
153.5, 151.2, 146.5, 145.1, 135.4, 131.2, 127.0, 126.0, 123.4, 121.2, 117.1,
116.6, 103.0, 59.6, 46.8, 46.6, 40.0, 29.7, 28.5, 27.6, 27.2, 26.8, 18.8 ppm;
HRMS (ESI TOF): m/z calcd for C24H28O4 [M+H]+: 381.2060; found
381.2051.
regiochemistry of the two Claisen rearrangements and
directing the final functional group manipulation on the ring
on which it was located, was also removed under the reductive
conditions of the last step to furnish artochamin H (2, 88%
yield for the 2 steps). Alternatively, treatment of 15 with
PivCl in the presence of pyridine afforded the bis(pivalate) 17
(64% yield), which was dimethylated and its protecting
groups removed as described above to afford artochamin I( 3,
91% overall yield). Finally, selenoetherification[18] of bis-
(pivalate) 17 (PhSeCl), followed by oxidative removal of the
phenylselenyl group from the resulting product, afforded
benzopyran[17] derivative 18 (85% overall yield), which was
Scheme 5. Total synthesis of 1. Reagents and conditions: a) ortho-
xylene, microwave, 1808C, 20 min, 100%; b) AIBN (0.1 equiv),
nBu3SnH (2.0 equiv), benzene, reflux, 1 h; c) HF·Et3N (5.0 equiv), THF,
238C, 1 h, 92% over 2 steps; d) ortho-xylene, microwave, 1808C, 10
min, 82%. AIBN=azobisisobutylonitrile.
targets, is shown in Scheme 4. Thus, chemoselective protec-
tion of 15 with PivCl/Et3N afforded the triply protected
derivative 16 (75% yield),[17] which was methylated (K2CO3,
MeI) and fully deprotected by exposure to LiAlH4. The
bromine atom, having served its dual role of ensuring the
Angew. Chem. Int. Ed. 2007, 46, 7501 –7505
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7503

Communications
methylated and deprotected as described above to furnish
artochamin J (4, 88% overall yield).
tion with regard to its mechanism and scope because of its
novelty and potential in chemical synthesis.
The dependence of the thermal [2+2] cycloaddition
reaction on the hydroxy groups was then examined. To this
end, the bis(TBS) stilbene derivative 5b (Scheme 1) was
subjected to the same microwave conditions as those used for
5a, both in the presence and absence of Ph3PO. As shown in
Scheme 5, here the cascade stopped prior to the [2+2] cyclo-
addition reaction and delivered instead the stilbene derivative
19 in quantitative yield. The absence of any detectable
amounts of the corresponding cyclobutane derivative in this
reaction provided proof that the thermal formal [2+2] cyclo-
addition reaction required the presence of the unprotected
hydroxy groups on its aromatic nucleus. Debromination
(nBu3SnH, AIBN) followed by desilylation (HF·Et3N) of 19
gave artochamin F (1, 92% overall yield). The spectroscopic
data of synthetic 1, as well as those of synthetic 2, 3, and 4,
matched those reported for the natural products.[19] Subjec-
tion of synthetic artochamin F (1) to the microwave heating
conditions in the absence of Ph3PO provided the expected
tetracyclic skeleton 20 corresponding to artochamins H, I,
and J in 82% yield, thereby demonstrating that neither the
bromine atom nor Ph3PO are required for the thermal formal
[2+2] cycloaddition reaction in this instance.
Upon consideration of these experimental results and the
various mechanistic possibilities, we propose that artocha-
mins H–J (2–4) most likely originate in nature from artocha-
min F (1), or a derivative thereof, through a stepwise ionic
mechanism following an initial oxidation of the catechol
moiety (21!22!23!24!25) as shown in Scheme 6.[20] The
intervention of this redox/ionic mechanism in the microwave-
promoted cascade sequence of Scheme 3 and the conversion
of artochamin F (1) into 20 (Scheme 5) is also likely. [21]
The described total syntheses of artochamins F, H, I, and J
(1–4) demonstrate the power of cascade reactions in total
synthesis[20] and reveal the feasibility of the formal cyclo-
addition reaction between an electron-rich stilbene and a
prenyl group. This latter process warrants further investiga-
Received: May 30, 2007
Published online: September 4, 2007
Keywords: cycloaddition · domino reactions · microwaves ·
.
natural products · total synthesis
[1] C. Boonlaksiri, W. Oonanant, P. Kongsaeree, P. Kittakoop, M.
Tanticharoen, Y. Thebtaranonth, Phytochemistry 2000, 54, 415 –
417.
[2] C.-C. Chen, Y.-L. Huang, J.-C. Ou, J. Nat. Prod. 1993, 56, 1594 –
1597.
[3] E. H. Hakim, Asnizar, Yurnawilis, N. Aimi, M. Kitajima, H.
Takayama, Fitoterapia 2002, 73, 668 – 673.
[4] M. R. Fernando, S. M. D. Nalinie Wickramasinghe, M. I. Thab-
rew, P. L. Ariyananda, E. H. Karunanayake, J. Ethnopharmacol.
1991, 31, 277 – 282.
[5] Y. H. Wang, A. J. Hou, L. Chen, D. F. Chen, H. D. Sun, Q. S.
Zhao, K. F. Bastow, Y. Nakanish, X. H. Wang, K. H. Lee, J. Nat.
Prod. 2004, 67, 757 – 761.
[6] Y. H. Wang, A. J. Hou, D. F. Chen, M. Weiller, A. Wendel, R. J.
Staples, Eur. J. Org. Chem. 2006, 3457 – 3463.
[7] For another notable example of racemic natural products, see
the endiandric acids: isolation: a) W. M. Bandaranayake, J. E.
Banfield, D. St. C. Black, J. Chem. Soc. Chem. Commun. 1980,
19, 902 – 903; total synthesis: b) K. C. Nicolaou, N. A. Petasis,
R. E. Zipkin, J. Uenishi, J. Am. Chem. Soc. 1982, 104, 5555 –
5557; c) K. C. Nicolaou, N. A. Petasis, J. Uenishi, R. E. Zipkin, J.
Am. Chem. Soc. 1982, 104, 5557 – 5558; d) K. C. Nicolaou, R. E.
Zipkin, N. A. Petasis, J. Am. Chem. Soc. 1982, 104, 5558 – 5560;
e) K. C. Nicolaou, N. A. Petasis, R. E. Zipkin, J. Am. Chem. Soc.
1982, 104, 5560 – 5562.
[8] For related cyclobutane-forming reactions that involve a benzo-
pyran and a prenyl group, see: a) M. Mondal, V. G. Puranik, N. P.
Argade, J. Org. Chem. 2007, 72, 2068 – 2076; for a related formal
[2+2] cycloaddition reaction that involves an allylic cation and a
benzopyran, see: b) A. V. Kurdyumov, R. P. Hsung, J. Am.
Chem. Soc. 2006, 128, 6272 – 6273; for recent examples of
thermal [2+2] cycloaddition reactions between allenes and
alkenes, see: c) H. Ohno, T. Mizutani, Y. Kadoh, A. Aso, K.
Miyamura, N. Fujii, T. Tanaka, J. Org. Chem. 2007, 72, 4378 –
4389; d) M. Murakami, T. Mat-
suda in Modern Allene Chemis-
try, Vol. 2 (Eds.: N. Krause,
A. S. K. Hashmi), Wiley-VCH,
Weinheim, 2004, pp. 727 – 815.
[9] P. R. Blakemore, W. J. Cole, P. J.
Kocienski, A. Morley, Synlett
1998, 26 – 28.
[10] This choice was made after a
Wittig reaction produced a mix-
ture of stilbenes (E/Z ca. 1:1).
To the best of our knowledge,
this is the first application of the
Julia–Kocienski reaction to stil-
bene synthesis, and it is recom-
mended by virtue of its effi-
ciency and apparently higher
selectivity. For an alternative
approach to stilbenes, see: J. E.
Robinson,
R. J. K.
Taylor,
Scheme 6. Postulated mechanistic pathway for the generation of the tetracyclic artochamins from
Chem. Commun. 2007, 1617 –
stilbene precursors.
1619.
7504
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7501 –7505

Angewandte
Chemie
[11] a) Q. Wang , Q. Huang, B. Chen, J. Lu, H. Wang , X. She, X. Pan,
Angew. Chem. 2006, 118, 3733 – 3735; Angew. Chem. Int. Ed.
2006, 45, 3651 – 3653.
[12] J. D. Godfrey, Jr., R. H. Mueller, T. C. Sedergran, N. Soundar-
arajan, V. Colandrea, Tetrahedron Lett. 1994, 35, 6405 – 6408.
[13] CCDC-648446 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.ca-
m.ac.uk/data_request/cif.
[14] A Discovery System model number 908005 was used.
[15] TLC analysis indicated that the Claisen rearrangements proceed
faster than cleavage of the Boc group or the formal cyclo-
addition. Given the fact that hydroxy groups must be liberated
prior to formation of the cyclobutane, the sequence shown in
Scheme 3 is proposed.
[16] The effect of Ph3PO was recognized when stilbene 5a derived
from the Julia–Kocienski olefination failed to perform well in
this process, whilst the same substrate 5a derived from a Wittig
reaction (and therefore assumed to contain trace amounts of
Ph3PO) went through the cascade sequence smoothly; further
investigations into the precise role of the Ph3PO are currently
underway and full details will be reported in due course.
[17] The regiochemistry of this pivalate reaction was unambiguously
proven by a ROESY NMR experiment on synthetic artocha-
min H (2) which exhibited cross signals between the OCH3 and
the prenyl CH2 and vinyl CH protons.
Barluenga, H. J. Mitchell, J. Am. Chem. Soc. 2000, 122, 9939 –
9953.
[19] We are grateful to Dr. Ai-Jun Hou for kindly providing the H
and 13C NMR spectra of natural artochamins F, H, I, and J for
comparison purposes.
[20] We consider the stepwise mechanism shown in Scheme 6 to be a
more plausible alternative to that originally proposed, see
Ref. [6].
1
[21] We must also consider two further mechanistic alternatives for
the microwave-promoted reactions. The first, and least probable,
involves a [p2s + p2a] cycloaddition. The second involves a
stepwise diradical mechanism in which formation of the five-
membered ring precedes diradical recombination, see Ref. [8a].
Preliminary experiments with (E)- and (Z)-5a indicate that the
formal cycloaddition reaction is not stereospecific with respect
to the alkene geometry. On the basis of this observation a
stepwise radical mechanism, in which bond rotation occurs to
some extent before recombination of the diradical, or the redox
mechanism shown in Scheme 6 are currently favored. These
possibilities are under investigation and full details will be
disclosed in due course.
[22] For selected reviews on cascade reactions, see: a) K. C. Nic-
olaou, D. J. Edmonds, P. G. Bulger, Angew. Chem. 2006, 118,
7292 – 7344; Angew. Chem. Int. Ed. 2006, 45, 7134 – 7186;
b) K. C. Nicolaou, T. Montagnon, S. A. Snyder, Chem.
Commun. 2003, 5, 551 – 564; c) L. F. Tietze, Chem. Rev. 1996,
96, 115 – 136; d) L. F. Tietze, U. Beifuss, Angew. Chem. 1993, 105,
137 – 170; Angew. Chem. Int. Ed. Engl. 1993, 32, 131 – 163; e) H.
Pellissier, Tetrahedron 2006, 62, 1619 – 1665; f) H. Pellissier,
Tetrahedron 2006, 62, 2142 – 2173; g) R. A. Bunce, Tetrahedron
1995, 51, 13103 – 13159.
[18] a) K. C. Nicolaou, J. A. Pfefferkorn, G.-Q. Cao, Angew. Chem.
2000, 112, 750 – 755; Angew. Chem. Int. Ed. 2000, 39, 734 – 739;
b) K. C. Nicolaou, G.-Q. Cao, J. A. Pfefferkorn, Angew. Chem.
2000, 112, 755 – 759; Angew. Chem. Int. Ed. 2000, 39, 739 – 743;
c) K. C. Nicolaou, J. A. Pfefferkorn, A. J. Roecker, G.-Q. Cao, S.
Angew. Chem. Int. Ed. 2007, 46, 7501 –7505
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7505