Synthesis of the spirocyclic cyclohexadienone ring system of the schiarisanrins.

Article abstract of DOI:10.1021/ol991238z

[structure: see text] Studies on the synthesis of the spirocyclic cyclohexadienone ring system 2 of the schiarisanrin family of natural products 1 are described and were based on the Lewis acid-promoted C-alkylation of the corresponding phenolic precursor.

Full text of DOI:10.1021/ol991238z

ORGANIC
LETTERS
2000
Vol. 2, No. 3
277-280
Synthesis of the Spirocyclic
Cyclohexadienone Ring System of the
Schiarisanrins
Robert S. Coleman,* Jason M. Guernon, and Jason T. Roland
Department of Chemistry, The Ohio State UniVersity, 100 West 18th AVenue,
Columbus, Ohio 43210-1185
coleman@chemistry.ohio-state.edu
Received November 9, 1999
ABSTRACT
Studies on the synthesis of the spirocyclic cyclohexadienone ring system 2 of the schiarisanrin family of natural products 1 are described and
were based on the Lewis acid-promoted C-alkylation of the corresponding phenolic precursor.
Schiarisanrins A-D (1a-d) are a structurally unique family
of homolignan natural products isolated from the fruit of the
Taiwanese medicinal plant Schizandra arisanensis (Schi-
zandraceae)1 and are related to the dibenzocyclooctadiene
lignans (Figure 1).2,3 Schiarisanrin C exhibits interesting
cedented 2,4-cyclohexadienone-6-spiro-3′-(2′,3′-dihydroben-
zo[b]furan) substructure 2. This spirocyclic ring system could
arise biosynthetically by oxidation of the methyl ether
precursor 4 to the oxonium ion 3 (or its equivalent radicaloid
species4), which undergoes spirocyclization by nucleophilic
attack of the adjacent phenolate to afford the substructure 2
(Scheme 1).
Scheme 1. Formation of the Spirocyclic Ring System
Figure 1. Structures of the schiarisanrins.
Taking this biosynthetic clue from nature, we set out to
investigate the ionic version of this bond formation strategy.
cytotoxicity against several standard cell lines at µg/mL
levels. A truly unique feature of these agents is the unpre-
(3) For recent synthetic work on dibenzocyclooctadiene lignans, see:
Dhal, R.; Landais, Y.; Lebrun, A.; Lenain, V.; Robin, J.-P. Tetrahedron
1994, 50, 1153-1164. Carroll, A. R.; Taylor, W. C. Aust. J. Chem. 1994,
47, 937-941. Warshawsky, A. M.; Meyers, A. I. J. Am. Chem. Soc. 1990,
112, 8090-8099.
(1) Kuo, Y.-H.; Kuo, L.-M. Y.; Chen, C.-F. J. Org. Chem. 1997, 62,
3242-3245.
(2) Ayres, D. C.; Loike, J. D. Lignans: Chemical, biological and clinical
properties; Cambridge University Press: England, 1990.
10.1021/ol991238z CCC: $19.00 © 2000 American Chemical Society
Published on Web 01/14/2000

Formally, our synthetic strategy amounts to an intramolecular
C-alkylation of a phenolate ion by way of a 5-endo-trig ring
closure. C-Alkylation of phenols is not common in the
literature, and the most relevant example of an intramolecular
version was found in the recent work from Mukherjee and
co-workers in their synthesis of (()-isolongifolene.5 In their
system, there is no possibility of competing O-alkylation.
2,4-Cyclohexadienones bearing 6,6-dialkyl substitution are
largely unknown.6
substituents served as surrogates of the four-carbon chain
of the natural products and provided an impediment to
dimerization of the cyclohexadienone product.7
A variety of Lewis acids promoted this process effectively,
many of them in high yield (Table 1). For example, AlCl3
Table 1. Yields of Spirocyclization Reaction of 7
yield of
In our initial examination of this bond construction
pathway (Scheme 2), we prepared simple 3,3′,5,5′-tetra-
Lewis acid
reaction conditions
6 (%)
AlCl3
EtAlCl2
Et2AlCl
1.25 equiv, CH2Cl2, 25 °C, 5 min
1.25 equiv, CH2Cl2, 25 °C, 24 h
1.25 equiv (5 × 0.25 equiv at 4 h intervals),
CH2Cl2, reflux, 30 h
95
91
65
Scheme 2. C- versus O-Alkylation of Phenolate
Me3SiOTf 1.25 equiv, CH2Cl2, 25 °C, 48 h
74
56
53
SnCl4
FeCl3
1.25 equiv, CH2Cl2, 25 °C, 36 h
1.25 equiv, CH2Cl2, 25 °C, 30 min
in CH2Cl2 afforded 6 in 95% yield, and the cyclization was
equally successful with EtAlCl2 although with a decreased
rate of cyclization.
With Et2AlCl, the cyclization was significantly slower, and
under carefully controlled reaction conditions with incre-
mental addition of Et2AlCl, the yields of 6 were modest.
Me3SiOTf, SnCl4, and FeCl3 were less than optimal. Protic
acids such as camphorsulfonic acid and the Lewis acids BF3‚
OEt2, TiCl4, LiClO4, and LiBF4 did not promote the
cyclization.
methyl-2,2′-biphenol systems 5 with one phenolic oxygen
bearing an oxonium ion precursor (e.g., X ) Cl, S(O)Ph,
OCH3, OCH2CH2OCH3). In all cases, once a phenolic oxygen
or the corresponding phenolate anion was liberated from the
appropriately protected precursor, rapid O-alkylation via
pathway b proceeded to afford the dibenzodioxepin 7, at the
expense of the desired C-alkylation pathway a to spirocycle
6. No evidence was observed for formation of the cyclo-
hexadienone product 6 with X ) Cl (basic conditions),
S(dO)Ph (AcCl), OCH3 (Lewis acid), or OCH2CH2OCH3
(Lewis acid).
1
The H NMR of 6 was indicative of the rigid spirocyclic
structure, and the diastereotopic furan methylene protons
were observed as a characteristic AB pattern. Irradiation of
the cyclohexadienone C5-methyl group that is proximal to
the furan resulted in a reciprocal NOE of the syn-methylene
hydrogen. Single-crystal X-ray analysis confirmed the spiro-
bicyclic structure (Figure 2).
Assuming that the O-alkylation could be reversible but
that the corresponding C-alkylation pathway would not be,
we examined the Lewis acid-promoted conversion of diben-
zodioxepin 7 to the presumed ring-opened oxonium ion,
which could cyclize by a 5-endo-trig pathway to the desired
spirocyclic product 6 (Scheme 3). We examined this reaction
pathway with the model system 7, where the aryl methyl
Scheme 3. Lewis Acid-Promoted Acetal Cleavage/
C-Alkylation
Figure 2. X-ray crystal structure of spirobicyclic cyclohexadienone
6 (hydrogens added for clarity).
The presence or absence of the aryl methyl groups was
found to exert a significant influence on the success or failure
278
Org. Lett., Vol. 2, No. 3, 2000

of the spirocyclization (Scheme 4). With two methyl groups
ortho to the biaryl bond (8; R ) CH3, R′ ) H), the
cyclization proceeded with modest yield (60%) to afford 11
substituent was used to block the p-phenoxy radical coupling
pathway11 in the dimerization of 14.
The other spirocyclization substrates were prepared as
illustrated in Scheme 6 starting from the corresponding
Scheme 4. Effect of Substituents on Cyclization
Scheme 6. Synthesis of Dibenzodioxepin Precursors
and was accompanied by small amounts of the corresponding
bis-phenol. With two methyl groups para to the biaryl bond
(9, R ) H, R′ ) CH3), no cyclization to 12 was observed
but only small amounts of the bis-phenol were present; the
multiple products formed were unidentifiable. In the absence
of all four methyl groups (10, R ) R′ ) H), the cyclization
to 13 failed completely, and the corresponding bis-phenol
was isolated in 25% yield, among numerous unidentifiable
products. The origin of this effect may arise either from
decreased stability of less substituted cyclohexadienone
products as in 12 or from an effect similar to the Thorpe-
Ingold effect where the ortho methyl groups preorganize the
system for cyclization as in 8 f 11.
Construction of the tetramethyldibenzodioxepin system 7
used in these studies (Scheme 5) relied on the Cu-promoted
nitroaromatic systems 17 and 18. Nitro to amine reduction
to the anilines12 followed by diazotization and Sandmeyer
reaction to the iodide13 afforded aryl iodides 1914 and 20 in
84% and 78%, respectively, for the two steps. Copper bronze
coupling15 was performed in the absence of solvent, and
although the yields to biaryls 21 and 22 were quite modest
(15-20%), suitable quantities of these biaryl systems could
be isolated for subsequent studies. Demethylation in the
presence of boron tribromide (91% for 21; 85% for 22)
afforded the corresponding bis-phenols16 and methylenation
with CH2BrCl and NaH afforded the dibenzodioxepins 8
Scheme 5. Synthesis of Tetramethylbenzodioxepin Precursor
(7) Katayama, S.; Watanabe, T.; Yamauchi, M. J. Pharm. Soc. Jpn. 1993,
41, 439-444. Dewar, P. S.; Forrester, A. R.; Thomson, R. H. J. Chem.
Soc. C 1971, 3950-3959. Curtin, D. Y.; Stein, A. R. Org. Synth. 1966, 46,
115-119.
(8) Kantam, M. L.; Santhi, P. L. Synth. Commun. 1996, 26, 3075-3079.
(9) Tashiro, M.; Fukata, G.; Oe, K. Org. Prep. Proc. Int. 1975, 7, 183-
188.
(10) Simpson, J. E.; Daub, G. H.; Hayes, F. N. J. Org. Chem. 1973, 38,
1771-1772.
phenoxy radical coupling8 of commercially available phenol
14 (70%) followed by removal of the aryl chlorides of 15
with Raney nickel9 to afford 16 (96%). Methylene introduc-
tion with CH2BrCl and NaH afforded 7 (64%).10 The 4-chloro
(11) Kende, A. S.; Ebetino, F. H.; Ohta, T. Tetrahedron Lett. 1985, 26,
3063-3066.
(12) Kno¨lker, H. J.; Bauermeister, M. J. Chem. Soc., Chem. Commun.
1990, 664-665. McKenzie, T. C.; Hassen, W.; Macdonald, S. J. F.
Tetrahedron Lett. 1987, 28, 5435-5436. Flaugh, M. E.; Crowell, T. A.;
Clemens, J. A.; Sawyer, B. D. J. Med. Chem. 1979, 22, 63-69.
(13) Lavastre, O.; Cabioch, S.; Dixneuf, P. H.; Vohlidal, J. Tetrahedron
1997, 53, 7595-7604.
(4) Green, S. P.; Whiting, D. A. J. Chem. Soc., Perkin Trans. 1 1998,
193-201.
(5) Das, S.; Karpha, T. K.; Ghosal, M.; Mukherjee, D. Tetrahedron Lett.
1992, 33, 1229-1232.
(14) dos Santos, M. L.; de Magalhaes, G. C.; Braz Filho, R. J.
Organomet. Chem. 1996, 526, 15-19.
(6) Bodajla, M.; Jones, G. R.; Ramsden, C. A. Tetrahedron Lett. 1997,
38, 2573-2576. Chamberlin, S.; Wulff, W. D. J. Am. Chem. Soc. 1992,
114, 10667-10669. Becker, A. R.; Richardson, D. J.; Bruice, T. C. J. Am.
Chem. Soc. 1977, 99, 5058-5068.
(15) Kelly, T. R.; Xie, R. L. J. Org. Chem. 1998, 63, 8045-8048.
(16) Moorlag, H.; Meyers, A. I. Tetrahedron Lett. 1993, 34, 6993-6996.
Barrett, A. G. M.; Itoh, T.; Wallace, E. M. Tetrahedron Lett. 1993, 34,
2233-2234.
Org. Lett., Vol. 2, No. 3, 2000
279

(53%) and 9 (73%). These yields were not optimized. The
parent dibenzo[d,f][1,3]dioxepin (10) was prepared by simple
phenoxy radical coupling followed by methylenation as
above.
Acknowledgment. This work was supported by a grant
from the NIH (CA 65875). R.S.C. was a Research Fellow
of the Alfred P. Sloan Foundation (1995-1998). We thank
Professor John Swenton for helpful discussions and Dr.
Judith Gallucci for X-ray structural analysis.
We have described a synthetic approach to the spirobi-
cyclic ring system of the schiarisanrin family of homolignan
natural products that was based on a Lewis acid-promoted
intramolecular C-alkylation of a phenol. Cyclization yields
were near quantitative when methyl groups were present on
the aromatic systems ortho to the biaryl bond.
Supporting Information Available: Experimental pro-
cedures and spectral characterization of synthetic intermedi-
ates and products. This information is available free of charge
via the Internet at http://pubs.acs.org.
OL991238Z
280
Org. Lett., Vol. 2, No. 3, 2000