Iron-catalyzed rearrangements and cycloaddition reactions of 2h-chromenes

Article abstract of DOI:10.1021/ol202772k

Iron(III) salts catalyze the tandem rearrangement/hetero-Diels-Alder reaction of 2H-chromenes to yield tetrahydrochromeno heterocycles. The process can occur as a homodimerization and cycloaddition process using electron-rich dienophiles. Deuterium labeli

Full text of DOI:10.1021/ol202772k

ORGANIC
LETTERS
2011
Vol. 13, No. 24
6480–6483
Iron-Catalyzed Rearrangements and
Cycloaddition Reactions of
2H-Chromenes
Yi Luan, Huan Sun, and Scott E. Schaus*
Department of Chemistry and Center for Chemical Methodology and Library
Development (CMLD-BU), Life Sciences and Engineering Building, Boston University,
24 Cummington Street, Boston, Massachusetts 02215, United States
seschaus@bu.edu
Received October 14, 2011
ABSTRACT
Iron(III) salts catalyze the tandem rearrangement/hetero-DielsꢀAlder reaction of 2H-chromenes to yield tetrahydrochromeno heterocycles. The
process can occur as a homodimerization and cycloaddition process using electron-rich dienophiles. Deuterium labeling and mechanistic studies
revealed a hydride shift and ortho-quinone methide cycloaddition reaction pathway.
The tetrahydrochromeno polycyclic core is a character-
istically unique structure among the large number of bio-
logically active phenolic compounds isolated from Nature.1
Natural product examples include mulberrofuran G,2
australisine A,3 and sorocenol E,4 bearing a quaternary
ketal carbon (Figure 1). Mulberrofuran, australisine, and
sorocenol natural products possess hypotensive effects and
cytotoxicities against human cancer cell lines.2,3 The dime-
ric flavonoid dependensin isolated from the antimalarial
uvarza dependens also contains a benzopyranobenzopyran
polycyclic ring structure.5 We envisaged synthetic access
to these natural product core structures via tetrahydro-
chromeno and benzopyrano starting materials. Herein, we
describe an FeCl3 6H2O-catalyzed rearrangement and
3
hetero-DielsꢀAlder reaction of 2H-chromenes with dieno-
philes to access these unique heterocyclic structures.6
We have been interested in the synthesis and utility of
2H-chromenes (40-methoxyflav-3-enes).7 Homodimeriza-
tion of small molecules is a highly efficient way toconstruct
complex dimeric structures,8 and recently, Kumar and
co-workers demonstrated the homodimerization of 2H-
chromenes under Brønsted acidic conditions.9
The homodimerization of 2H-chromenes provides rapid
access to ketal polycyclic or benzopyranobenzopyran core
structures but, as reported, did not readily provide discrete
access to heterodimeric stuctures, similar to those found
in Nature. During the course of our study on dimerization
of 40-methoxyflav-3-ene, we discovered that iron salts,
(1) (a) Nomura, T.; Fukai, T.; Hano, Y. In Studies in Natural
Products Chemistry; Rahman, A., Eds.; Elsevier: 2003. (b) Bhandari, P.;
Crombie, L.; Harper, M. F.; Rossiter, J. T.; Sanders, M. D.; Whiting, A.
J. Chem. Soc., Perkin Trans. 1 1992, 1685–1697. (c) Reisch, J.; Adebajo,
A. C.; Kumar, V.; Aladesanmi, A. J. Phytochemistry 1994, 36, 1073–
1076.
(2) (a) Fukai, T.; Hano, Y.; Hirakura, K.; Nornura, T.; Uzawa, J.;
Fukushima, K. Chem. Pharm. Bull. 1985, 33, 3195. (b) Fukai, T.; Hano,
Y.; Hirakura, K.; Nornura, T.; Uzawa, J.; Fukushima, K. Heterocycles
1984, 22, 1007–1011.
(3) Zhang, Q. J.; Tang, Y. B.; Chen, R. Y.; Yu, D. Q. Chem.
Biodiversity 2007, 4, 1533–1540.
(4) Hano, Y.; Yamanaka, J.; Momose, Y.; Nomura, T. Heterocycles
1995, 41, 2811–2821.
(5) Nkunya, M. H. H.; Waibel, R.; Achenbach, H. Phytochemistry
(6) (a) Bolm, C.; Legros, J.; Paih, J. L.; Zani, L. Chem. Rev. 2004, 104,
6217–6254. (b) Rao, W. D.; Chan, P. W. H. Org. Biomol. Chem. 2010, 8,
4016–4025. (c) Iovel, I.; Mertins, K.; Kischel, J.; Zapf, A.; Beller, M.
Angew. Chem., Int. Ed. 2005, 44, 3913–3917.
(7) Moquist, P. N.; Kodama, T.; Schaus, S. E. Angew. Chem., Int. Ed.
2010, 49, 7096–7100.
(8) (a) Ibrahim, A. A.; Wei, P. H.; Harzmann, G. D.; Kerrigan, N. J.
J. Org. Chem. 2010, 75, 7901–7904. (b) Guidi, V.; Sandoval, S.;
McGregor, M. A.; Rosen, W. Tetrahedron Lett. 2010, 51, 5086–5090.
(c) Chapman, O. L.; Engel, M. R.; Springer, J. P.; Clardy, J. C. J. Am.
Chem. Soc. 1971, 93, 6696–6698.
(9) Devakarama, R.; Black, D. StC.; Kumar, N. Tetrahedron Lett.
2010, 51, 3636–3638.
1993, 34, 853–856.
r
10.1021/ol202772k
Published on Web 11/18/2011
2011 American Chemical Society

Table 1. Acid-Catalyzed Homodimerization of 1aa
entry
catalyst
solvent
drb
2:1
yieldc
1
FeCl3 6H2O
PhCH3
CH3OH
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH3OH
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
40%
0%
3
2
FeCl3 6H2O
3
3
FeCl3 6H2O
2:1
3:1
2:1
1:1
1:1
81%
85%
22%
5%
3
4d
5e
6
FeCl3 6H2O
3
FeCl3
Figure 1. Representative natural products.
HCl(aq)
HCl(aq)
HCl in ether
7f
8
34%
0%
especially FeCl3 6H2O, served as an efficient catalyst
for the homodimerization of 2H-chromene 1a (Table 1,
entries 1ꢀ4). Anhydrous FeCl3 performedpoorly, suggest-
9
Fe(OTs)3 6H2O
2:1
2:1
2:1
2:1
80%
85%
14%
12%
3
3
10
11
12
Fe(OTf)3
ZnCl2
AlCl3
ing the hydrate as FeCl3 6H2O is essential for catalytic
3
a Reaction conditions: 0.25 mmol of 1a, 10 mol % catalyst, 0.2 M in
activity (entry 5). To eliminate the possibility that a trace
amount of HCl was acting as the catalytic species, catalytic
quantities of aqueous HCl was evaluated in the reaction to
afford a low yield of 2a in 1:1 dr (entriy 6). The yield was
slightly improved using anhydrous methanol as the solvent
but in no diastereoselectivity (entry 7).9 Dry HCl failed to
provide product from the reaction (entry 8). Similar yields
and diastereoselectivities were achieved using alterna-
tive iron(III) salts (entries 9ꢀ10), while other metal chloride
salts gave little or no yield (entries 11ꢀ12). We chose
the solvent for 12 h at room temperature. b Ratio was determined by 1H
NMR analysis. c Isolated yield. d 15 mol % FeCl3 6H2O. e Anhydrous
3
conditions. f 0.32 mmol of 1a in 20 mL of methanol, with 10 drops of
HCl, then refluxed at 65 °C for 12 h. 9
Table 2. FeCl3 6H2O Catalyzed Dimerizationa
3
FeCl3 6H2O as the catalyst to further investigate in the
3
reaction.
The optimized reaction conditions proved general for
a range of 2H-chromenes investigated (Table 2). The re-
action proceeded smoothly with 6-methyl chromene 1b,
affording chromeno[2,3-b]chromene dimer 2b in good
yields (entry 2). Electron-deficient chromenes 1cꢀ1e un-
derwent dimerization in a similar fashion, providing 2cꢀ2e
in moderate to good yields, but ultimately requiring longer
reaction times (entries 3ꢀ5). The relative stereochemistry
of both the major and minor diastereomers was deter-
mined by single crystal X-ray analysis of the dimer 2d.10
A change in chemoselectivity was observed in the dimerization
of 7-methoxy chromene 1f. Pyran 1f afforded isomer 3f as
the major product in good yield at rt; only a trace amount
(<5%) of isomer 2f was identified (entry 6). The chemos-
electivity reverted to the hetero-DielsꢀAlder chromeno-
[2,3-b]chromene dimer product with the methoxy group in
the 6-position (40,6-dimethoxy-3-flavene 1g, Table 2, entry 7).
Electron-rich chromene 1h did lead to the formation of iso-
mer 3h in good yield (entry 8). Alternatively, use of the di-
methyl carbamate protecting group attenuated the reactivity
of 7-oxygenated arene 1i to yield the chromeno[2,3-b]chromene
entry
2H-chromene (1)
drb
2:3
yieldc
1
R1 = H, R2 = H (1a)
3:1
3:1
3:1
3:1
>99:1 85% (2a)
>99:1 89% (2b)
>99:1 82% (2c)
>99:1 86% (2d)
2
R1 = CH3, R2 = H (1b)
R1 = H, R2 = Cl (1c)
3
4d
5e
6
R1 = Br, R2 = H (1d)
R1 = NO2, R2 = H (1e)
R1 = H, R2 = OCH3 (1f)
R1 = OCH3, R2 = H (1g)
R1, R2 = 6,7-(OCH2O) (1h)
R1 = H, R2 = OCON(CH3)2 (1i)
1.5:1 >99:1 72% (2e)
>99:1 5:95 86% (3f)
7
2:1
>99:1 3:97 66% (3h)
3:1 >99:1 70% (2i)
>99:1 87% (2g)
8
9
a 0.25 mmol of 1, 15 mol % FeCl3 6H2O, 0.2 M in CH2Cl2 for 12 h at
3
rt. b 1H NMR analysis of 2 or 3. c Isolated yield. d 24 h. e 72 h.
dimer 2i (entry 9). The more electron-rich chromene reac-
tion partners appeared to have the appropriate reactivity
profile to react with the heterodiene as it was being formed
in the reaction mixture.
(10) CCDC 812201 and 812202 contain the supplementary crystal-
lographic data for this paper.
Org. Lett., Vol. 13, No. 24, 2011
6481

Scheme 1. Synthesis of (()-Dependensin
Scheme 3. Proposed Mechanism for Cycloaddition
The Fe-catalyzed reaction was utilized in the synthesis
of dependensin (5) via the homodimerization of 5,7,8-
trimethoxyflav-3-ene 4, easily prepared in three steps
from commercially available starting materials. Depen-
densin has been previously synthesized11 in a similar
fashion with our route demonstrating a reduction in reac-
tion sequence steps in high stereocontrol (Scheme 1).
Subjecting 5,7,8-trimethoxy-3-flavene 4 to the optimized
FeCl3-dimerization conditions afforded (()-dependensin
(5) as a single diastereomer in 29% overall yield from
1,2,3,5-tetramethoxybenzene.
E-conformer under the reaction conditions (E)-7-D.14 The
second 2H-chromene undergoes a hydride shift to yield the
40-methoxy-2-flavene 6-D as the reactive dienophile.15
Cycloaddition of the E-configured vinyl oQM (E)-7-D
and 6-D in an inverse electron demand [4 þ 2] fashion
furnishes the dimer 2-D. The chemoselectivity observed in
the 40,7-dimethoxy-3-flavene 1f dimerization (Table 2, en-
try 7) is rationalized by a [4 þ 2] cycloaddition of a highly
electron-rich olefin with the oQM at a faster rate than the
hydride shift.
We proposed further experiments based on our mechan-
istic hypothesis aimed at the heterodimerization process.
We first postulated that 40-methoxy-2-flavene 6 would
react with the oQM generated in situ from 2H-chromene
1hunder the Fe-promoted conditions (Scheme 4, reaction 1).
The reaction proceeded well yielding the corresponding
heterodimer 9h in high yield and 4:1 diastereoselectivity,
with only trace amounts (<5%) of homodimer 3h formed.
We demonstrated the intermediacy of the oQM by making
(E)-8 and reacting it with dienophile 6 under the same
reaction conditions (Scheme 4, reaction 2).16 The product
was isolated in comparable yield and the same diastereos-
electivity to provide further evidence for the intermediacy
of both reaction partners in the [4 þ 2] cycloaddition
pathway.
We envisaged a cyloaddition processes to occur with
electron-rich dienophiles utilizing 2H-chromenes as oQM
precursors based on our preliminary studies of the
reaction.17 Inspired by natural products such as the mul-
berrofurans and australisine we first evaluated hetero-
DielsꢀAlder reactions between 2H-chromenes 1 and 40-
methoxy-2-flavenes 6 under the Fe-promoted conditions.
The reactions proceeded well with 6-methyl and 1,3-
benzodioxol substituted chromenes 1b and 1j (Table 3,
entries 1 and 9). Electron-deficient 2H-chromenes 1cꢀ1e
were also able to participate, but at a slower rate (entries 2ꢀ4).
Scheme 2. Deuterium-Labeled 1a in the Homodimerization
We designed experiments to gain insight into the reac-
tion sequence leading to the dimerization product. We
postulated use of 2-deutero-40-methoxyflav-3-ene 1-D in
the dimerization reaction would illustrate the isomeriza-
tion of the chromene to the dienophile. The chromene 1-D
was synthesized via boronate addition onto the deuterated
2-ethoxy-2H-chromene. Subjecting 1-D to the dimeriza-
tion conditions led to the formation of dimer 2-D, epimeric
at the benzylic positions (Scheme 2). Under UV irradia-
tion, oxa-6π rearrangement of chromenes12 results in the
formation of the corresponding ortho-quinone methide
(oQM).13 We propose the similar oxa-6π rearrangement
of 2H-chromene could occur under Fe-catalyzed condi-
tions. Illustrated in Scheme 3, the Fe-catalyst is responsible
for promoting the hydride shift to yield the dienophile and
the ring-opening reaction.
The ring-opening process can lead to the formation of
Z-ortho-quinone methide that rapidly equilibrates to the
(15) For similar hydride transfer, see: (a) McQuaid, K. M.; Long,
J. Z.; Sames, D. Org. Lett. 2009, 11, 2972–2975. (b) McQuaid, K. M.;
Sames, D. J. Am. Chem. Soc. 2009, 131, 402–403.
(11) Deodhar, M.; Black, D. StC.; Kumar, N. Tetrahedron 2007, 63,
5227–5235.
(16) Jurd, L.; Roitman, J. N.; Wong, R. Y. Tetrahedron 1979, 35,
1041–1054.
(12) (a) Delbaere, S.; Micheau, J. C.; Vermeersch, G. J. Org. Chem.
2003, 68, 8968–8973. (b) Padwa, A.; Lee, G. A. J. Chem. Soc., Chem.
Commun. 1972, 795–796.
(13) Pathak, T. P.; Sigman, M. S. J. Org. Chem. 2011, 76, 9210–9215.
(14) (a) Bishop, L. M.; Winkler, M.; Houk, K. N.; Bergman, R. G.;
Trauner, D. Chem.;Eur. J. 2008, 14, 5405–5408. (b) Jurd, L. Tetra-
hedron 1977, 33, 163–168.
(17) (a) Van de Water, R. W.; Pettus, T. R. R. Tetrahedron 2002, 58,
5367–5405. (b) Wu, K. L.; Mercado, E. V.; Pettus, T. R. R. J. Am. Chem.
Soc. 2011, 133, 6114–6117. (c) Marsini, M. A.; Huang, Y.; Lindsey, C.;
Wu, K. L.; Pettus, T. R. R. Org. Lett. 2008, 10, 1477–1480. (d) Jensen,
K. H.; Webb, J. D.; Sigman, M. S. J. Am. Chem. Soc. 2010, 132, 17471–
17482. (e) Lumb, J. P.; Choong, K. C.; Trauner, D. J. Am. Chem. Soc.
2008, 130, 9230–9231.
6482
Org. Lett., Vol. 13, No. 24, 2011

Scheme 4. Cycloaddition Reactions of 40-Methoxy-2-flavene 6
Table 4. Cycloaddition Reactions Using the ortho-Quinone
Methide Precursor 40,7-Dimethoxy-3-flavene 1f
Table 3. Cycloaddition Reactions Using 40-Methoxyflav-2-ene
as the Dienophile
entry
2H-chromene (1)
dr yielda
1
2
3
4b
5
6
7
8
9
R1 = CH3, R2 = H, R3 = 4-OCH3 (1b)
R1 = H, R2 = Cl, R3 = 4-OCH3 (1c)
R1 = Br, R2 = H, R3 = 4-OCH3 (1d)
R1 = NO2, R2 = H, R3 = 4-OCH3 (1e)
R1 = H, R2 = OCH3, R3 = 4-OCH3 (1f)
R1 = OCH3, R2 = H, R3 = 4-OCH3 (1g)
R1, R2 = 6,7-(OCH2O), R3 = 4-OCH3 (1h)
3:1 87% (9b)
3:1 87% (9c)
2:1 72% (9d)
3:1 65% (9e)
3:1 87% (9f)
2:1 83% (9g)
4:1 81% (9h)
a Isolated yield. Ar = PMP (para-methoxyphenyl).
R1 = H, R2 = OCON(CH3)2, R3 = 4-OCH3 (1i) 4:1 89% (9i)
R1 = R2 = H, R3 = 3,4-(OCH2O) (1j)
2:1 90% (9j)
smoothly with dihydropyran 10f to produce the corre-
sponding acetal (entry 6).
a Isolated yield. b 24 h, another 0.186 mmol of 6 and 40 mol %
FeCl3 6H2O were added after first 12 h.
In conclusion, we have developed an iron-catalyzed
tandem rearrangement/hetero-DielsꢀAlder approach to
access tetrahydrochromeno heterocycles. Our studies have
revealed an oQM involved cycloaddition reaction mechan-
ism. Further mechanistic evidence, including trapping of
the in situ generated oQM and use of the oQM (E)-8,
supports the proposed multistep process. Ongoing studies
include expansion of the scope and synthetic utility of the
reaction sequence.
3
Electron-rich 2H-chromenes 1fꢀ1h underwent a hetero-
DielsꢀAlder reaction in good yields, providing selectively
functionalized 9fꢀ9h with similar diastereoselectivities
(entries 5ꢀ7). Attenuating the electron-donating group
with a dimethyl carbamate did not affect the yield or selec-
tivity of the reaction (entry 8).
We further explored the scope of the cycloaddition
reaction pathway using the in situ generated oQM of chro-
mene 1fwithdienophiles10(Table4). Underthe optimized
reaction conditions, p-methoxystyrene and its derivatives
10aꢀ10c successfully afforded chromans 11aꢀ11c in good
yields (Table 4, entries 1ꢀ3). Furthermore, dihydrona-
phthalene 10d yielded the [4 þ 2] adduct 11d as a single
diastereomer (entry 4). Electron-rich indene 10e was also a
suitable dienophile (entry 5), and the reaction proceeded
Acknowledgment. This research was supported by the
NIH (R01 GM078240, S.E.S. and Y.L.) and BU Under-
graduate Research Opportunities Program (H.S.).
Supporting Information Available. Experimental pro-
cedures and characterization data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
Org. Lett., Vol. 13, No. 24, 2011
6483