A short asymmetric synthesis of (+)-lyoniresinol dimethyl ether

Article abstract of DOI:10.1021/jo0497454

A short, efficient synthesis of the lignan (+)-lyoniresinol dimethyl ether is described. The synthesis is achieved by asymmetric photocyclization of an achiral dibenzylidenesuccinate to a chiral aryldihydronaphthalene. (-) -Ephedrine is used as a chiral a

Full text of DOI:10.1021/jo0497454

A Sh or t Asym m etr ic Syn th esis of (+)-Lyon ir esin ol Dim eth yl Eth er
Tokoure´ Assoumatine, Probal K. Datta, Timothy S. Hooper, Brigitte L. Yvon, and
J ames L. Charlton*
Department of Chemistry, University of Manitoba, Winnipeg, Manitoba, Canada R3T 2N2
charltn@cc.umanitoba.ca
Received February 12, 2004
A short, efficient synthesis of the lignan (+)-lyoniresinol dimethyl ether is described. The synthesis
is achieved by asymmetric photocyclization of an achiral dibenzylidenesuccinate to a chiral
aryldihydronaphthalene. (-)-Ephedrine is used as a chiral auxiliary to bias the atropisomeric
equilibrium in the dibenzylidenesuccinate prior to the photochemical reaction. The synthesis of
the title compound was accomplished in five steps, and the final product was recrystallized to
constant melting point and rotation.
SCHEME 1. P h otoch em istr y of
2,3-Diben zylid en esu ccin a tes
In tr od u ction
There are several papers in the literature on the use
of 2,3-dibenzylidenesuccinate derivatives, particularly
fulgides (the cyclic anhydrides of (E,E)-dibenzylidene-
succinic acids, 1, X ) O) and the corresponding lactones
(1, X ) 2H), as starting materials for the synthesis of
arylnaphthalene lignans (including dihydro and tetrahy-
dro derivatives) (Scheme 1).1-11 The syntheses involve
cyclization of the 2,3-dibenzylidenesuccinate derivative
to an unstable 1,8a-dihydronaphthalene (1,8a-DHN) fol-
lowed by isomerization to a 1,2- or 1,4-DHN from which
lignans can easily be prepared.
SCHEME 2. P h otocycliza tion of (E,E)-2,3-
Although an interesting methodology, there are no
examples yet of its use for the synthesis of optically active
lignans (asymmetric synthesis). In this paper, we de-
scribe a method for photochemically converting a 2,3-
dibenzylidenesuccinate precursor to an optically active
dihyronaphthalene (DHN) and subsequent transforma-
tion of the DHN into the optically pure lignan (+)-
lyoniresinol dimethyl ether.
Di(3,4,5-tr im eth oxyben zylid en e)Su ccin a te, 2
* To whom correspondence should be addressed. Tel: (204) 474-
9267. Fax: (204) 474-7608.
(1) Chang, S.-T.; Wang, S.-Y.; Su, U. C.; Hunag, S.-L.; Kuo, Y.-H.
Hozforschung 1999, 53, 142.
(2) Anjaneyulu, A. S. R.; Raghu, P.; Rao, K. V. R. Indian J . Chem.,
Sect. B 1980, 19B, 511.
(3) Anjaneyulu, A. S. R.; Raghu, P.; Rao, K. V. R. Indian J . Chem.,
Sect. B 1979, 18B, 535.
(4) Momose, T.; Nakamura, T.; Kanai, K. Chem. Pharm. Bull. 1978,
26, 3186.
Resu lts a n d Discu ssion
(5) Momose, T.; Kanai, K.; Hayashi, K. Chem. Pharm. Bull. 1978,
26, 3195.
(6) Momose, T.; Kanai, K.; Nakamura, T.; Kuni, Y. Chem. Pharm.
Bull. 1977, 25, 2755.
(7) Momose, T.; Nakamura, T.; Kanai, K. Heterocycles 1977, 6, 277.
(8) Momose, T.; Kanai, K.-I.; Nakamura, T. Heterocycles 1976, 4,
1481.
(9) Heller, H. G.; Strydom, P. J . J . Chem. Soc., Chem. Commun.
1976, 50.
(10) Ayres, D. C.; Carpenter, B. G.; Denney, R. C. J . Chem. Soc.
1965, 3578.
(11) Datta, P. K.; Yau C.; Hooper, T. S.; Yvon, B. L.; Charlton, J . L.
J . Org. Chem. 2001, 66, 8606.
We previously observed that simple diesters of (E,E)-
2,3-dibenzylidenesuccinates such as diethyl (E,E)-2,3-di-
(3,4,5-trimethoxybenzylidene)succinate (2) in ethanol
could be photochemically converted in modest yield to the
cis-1,2-dihydronaphthalene (cis-1,2-DHN, 3), accompa-
nied by smaller amounts of trans-1,2-DHN (4) (Scheme
2).11
It was also noted that the methylene protons of the
ethyl groups of the 2,3-dibenzylidenesuccinate 2 appeared
10.1021/jo0497454 CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/20/2004
4140
J . Org. Chem. 2004, 69, 4140-4144

Synthesis of (+)-Lyoniresinol Dimethyl Ether
SCHEME 3. Atr op isom er ic F or m s of
Diben zylid en esu ccin a tes
SCHEME 4. P r ep a r a tion of Ma n d elyl Ester 6
to be nonidentical (diastereotopic) in the NMR spectrum.
It was concluded, as had been noted previously,12-14 that
2,3-dibenzylidenesuccinates, such as 2, have a barrier to
rotation about the central single bond and that they exist
at room temperature as an equilibrium mixture of the
two enatiomeric atropisomers (Scheme 3).
SCHEME 5. Cyclic Am id e Ester 7
The fact that (E,E)-2,3-dibenzylidenesuccinates exist
as enantiomeric atropisomers suggested that their pho-
tocyclization might be stereoselective (pericyclic conro-
tatory cyclization) with the M conformation giving the
(1R)-1,8a-dihydronaphthalene and the P conformation
giving the (1S)-1,8a-dihydronaphthalene (Scheme 3). The
designation of the atropisomers of 2 as P or M is based
on the standard nomenclature for axially chiral molecules
where the point of view in this case is parallel to the
central single bond of the butadiene.15 The subsequent
rearrangement of the 1,8a-dihydronaphthalenes to either
1,2- or 1,4-DNPs should not affect the configuration at
the 1-center, and thus, the overall reaction of either the
M or P form of the 2,3-dibenzylidenesuccinate would
provide an asymmetric synthesis of a lignan precursor.
What remained was to find a way to bias the 2,3-
dibenzylidenesuccinate 2 to adopt only one of its two
axially chiral forms.
With the idea that a chiral ester corresponding to 2
might adopt a single atropisomeric form, the mandelate
ester 6 was prepared as shown (Scheme 4). The monoacid
5 was prepared using a previously described method.11
Reaction of the crude 5 with methyl R-bromophenyl-
acetate in the presence of potassium carbonate gave
chiral ester 6. The NMR spectrum of 6 in CDCl3 clearly
indicated that it was an approximately 50:50 mixture of
two diasteriomers, presumably rotamers about the cen-
tral bond of the diene. The two spectra collapsed to a
single average spectrum when the spectrum was acquired
in DMSO at 80 °C.
equilibrating diastereomers, and it was concluded that
it existed as a single rotameric form.
It was tentatively concluded that the preferred con-
formation of 7 was the M form shown in Scheme 5 as
this was the lowest energy conformation that could be
found using the Spartan molecular mechanics program
(computations were made on the nonmethoxylated com-
pound to save computing resources). The very large
rotation of 7 (-343°), a value too large to be accounted
for by the ephedrine chiral auxiliary alone, also suggested
that 7 was in an extreme conformation. (-)-Hexahelicene,
which has the same left-handed helical configuration as
shown for 7 in Scheme 5, also has a very large negative
rotation.16
Irradiation of 7 at 254 nm in 2-propanol followed by
crystallization of the major product from the irradiation
solution gave a 26% isolated yield of a product whose
structure was tentatively assigned to that shown as
structure 8 (Scheme 6). Interestingly, compound 8 had
a rotation of +476°, a very large optical rotation opposite
to that of 7.
Since the acyclic mandelate ester 6 was not at all
prejudiced to adopt a single rotameric form, attempts
were made to prepare cyclic chiral esters or amide esters.
We were unable to prepare the diethyl tartrate cyclic
diester but were eventually successful in preparing the
(-)-ephedrine cyclic amide ester 7 (Scheme 5). The NMR
spectrum of 7 showed no evidence for the presence of two
An NMR spectrum of the evaporated crude product
mixture before crystallization of 8 indicated that the
major compound, 8, made up approximately 60% of the
product mixture along with 40% of a second compound.
This second compound was isolated by column chroma-
(12) Cohen, M. D.; Kaufman, H. W.; Sinnreich, D.; Schmidt, G. M.
J . J . Chem. Soc., B 1970, 1035.
(13) Davidse, P. A.; Dillen, J . L. M.; Heyns, A. M.; Modro, T. A.;
van Rooyen, P. H. Can. J . Chem. 1990, 68, 741.
(14) Elbe, H.-L.; Kobrich, G. Chem. Ber. 1974, 107, 1654.
(15) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds;
J ohn Wiley & Sons: New York, 1994; p 1120.
(16) David A. Lightner, D. A.; Hefelfinger, D. T.; Thomas; Powers,
W.; Frank, G. W.; Trueblood, K. N. J . Am. Chem. Soc., 1972, 94, 3492-
3497.
J . Org. Chem, Vol. 69, No. 12, 2004 4141

Assoumatine et al.
SCHEME 7. F in a l Step s in th e Syn th esis of
SCHEME 6. Ir r a d ia tion of 7
(+)-Lyon ir esin ol Dim eth yl Eth er
tography of the mother liquors and was identified as the
1,4-dihydro isomer 9 (Scheme 6). Although the absolute
stereochemistry of 9 was not determined, it was pre-
sumed to be the same as in 8.
Reduction of 11 with hydrogen (Pd/C) gave the satu-
rated diester 12. Contrary to previous reductions of this
type that had given almost exclusively the all-trans
product,11 12 was formed as a 5:2 mixture of the 2,3-
trans/2,3-cis compounds. Attempts to improve the ste-
reoselectivity by changing the solvent (ethanol and
methanol), temperature (rt and reflux), and catalyst
(Pd/C and Ru/C) proved fruitless.
The gross structure of 8 was assigned by NMR spec-
troscopy. There was little or no coupling between the
allylic proton at 4.22 ppm and the benzylic proton at 4.84
ppm consistent with a 1,2-trans relative configuration
(either 1R,2S or 1S,2R). A similarly small coupling
constant was observed in 4.11 It is notable that the
irradiation of the acyclic diethyl ester 2 gave the cis
isomer 3 as a major product and the trans isomer 4 as a
minor product in contrast to the irradiation of 7, which
gave the trans isomer as the major product. If 7 initially
cyclizes directly to a 1,8a-DHN then its subsequent
rearrangement to trans-8 must involve solvent protona-
tion as a concerted 1,5-sigmatropic shift of hydrogen
(from position 8 to position 2) would be impossible. This
observation is consistent with earlier work by Heller who
had tentatively concluded that rearrangements of this
type do not involve a concerted sigmatropic shift.17 On
the other hand, if the double bond adjacent to the ester
group in 7 were to photoisomerize to the trans configu-
ration, a subsequent photocylization could then be fol-
lowed by an allowed, and sterically possible, 1,5-
sigmatropic shift to give the observed trans product 8.
No experiments were conducted to resolve this ambiguity
although the former seems more likely based on Heller’s
earlier work.
Reduction of 12 (mixture of stereoisomers) with LiAlH4
gave (+)-lyoniresinol dimethyl ether as a major product
and its 2,3-cis diastereomer as a minor product. Repeated
crystallization (three times) from methylene chloride/
hexanes produced the pure (+)-lyoniresinol dimethyl
ether of constant melting point 157-159 °C and rotation
[R]25 ) +41.2 (c 0.7, CHCl3). The melting point was
D
consistent with that reported previously (158-160 °C),19
and the NMR spectrum was consistent with previously
published data.20,21 The rotation was different from, but
not inconsistent with, earlier reports of +49.4 for the
originally isolated material22 and later reports of +21.5,23
+26.0,24 and +30.024 (the concentrations and solvents
used for these determinations were not all the same).
The important aspect of the rotation of the finally
synthesized material was that the sign of the rotation
was positive, confirming the assignment of the 1S,2R
configuration for compounds 8, 10, 11, and 12. This also
means that the photocyclization of 7 did occur by a
conrotatory closure from the M conformation of 7 as had
been predicted.
The regioselectivity of the photochemical closure of 7
to form 8 was confirmed by NMR spectroscopy using
heteronuclear multiple-bond correlation (HMBC) spec-
troscopy. The cyclization had occurred to the aryl ring
adjacent to the amide functionality.
The absolute stereochemistries at carbon centers 1 and
2 were still unconfirmed and could only be determined
absolutely by conversion to a compound whose configura-
tions at those two centers were known.
The synthesis of (+)-lyoniresinol was thus accom-
plished by the rather unique method of constraining
precursor (7) to adopt a single atropisomeric and chiral
form and using that atropisomeric form to synthesize an
optically active natural product.
Exp er im en ta l Section
Removal of the chiral auxiliary from compound 8 was
easily accomplished by refluxing in 3 M KOH in methanol
(Scheme 7). The resulting diacid 10 was re-esterified to
the diethyl ester 11 that had been reported earlier in its
racemic form.11,18
Eth yl Meth ylm a n d elyl (E,E)-Diben zylid en esu ccin a te
(6). Methyl R-bromophenylacetate (0.29 mL, 1.84 mmol) was
added to a stirred solution of bis(3,4,5-trimethoxybenzylidene)-
(19) Yasue, M.; Kato, Y. Yakugaku Zasshi 1960, 80, 1013-14.
(20) Ogawa, M.; Ogihara, Y. Chem. Pharm. Bull. 1976, 24 (9), 2102-
05.
(17) Crescente, O.; Heller, H. G.; Oliver, S. J . Chem. Soc., Perkin
Trans. 1 1979, 150.
(18) Cow, C.; Leung, C.; Charlton, J . L. Can. J . Chem. 2000, 78,
553.
(21) Wallis, A. F. A. Aust. J . Chem. 1973, 26, 1571-6.
(22) Kato, Y. Chem. Pharm. Bull. 1963, 11, 823.
(23) Hostettler, F. D.; Seikel, M. K. Tetrahedron 1969, 25, 2325.
(24) Chandel, R. S.; Rastog, M. K. Indian J . Chem. 1980, 19B, 279.
4142 J . Org. Chem., Vol. 69, No. 12, 2004

Synthesis of (+)-Lyoniresinol Dimethyl Ether
succinate monoacid ester 511 (0.841 g, 1.67 mmol) and K2CO3
(1.17 g, 8.46 mmol) in acetone (10 mL), and the solution was
stirred at reflux for 14 h. The reaction mixture was filtered,
the precipitate washed with acetone, and the filtrate evapo-
rated. The residue was dissolved in CH2Cl2, dried over anhy-
drous magnesium sulfate, and stripped of solvent under
reduced pressure to give an orange oil as crude product (0.986
g, 91%). The diester was purified by flash column chromatog-
raphy on silica gel (50 mL) using 30% EtOAc/hexanes as the
eluent to afford a cream-colored amorphous solid (0.367 g,
34%): 1H NMR (CDCl3) (a mixture of diasteromers) δ 1.00 (t,
3H, J ) 7.1), 1.11 (t, 3H, J ) 7.1), 3.59 (s, 3H), 3.63 (s, 3H),
3.69 (s, 6H), 3.71 (s, 6H), 3.74 (s, 12H), 3.81 (s, 12H), 4.07 (m,
2H), 4.17 (m, 2H), 5.91 (s, 1H), 5.98 (s, 1H), 6.68 (s, 2H), 6.73
(s, 2H), 6.76 (s, 2H), 6.79 (s, 2H), 7.32 (m, 10H), 7.83 (s, 1H),
7.87 (s, 1H), 7.88 (s, 1H), 7.89 (s, 1H); 13C NMR (CDCl3) (a
mixture of two diastereomers) δ 14.0 (CH3), 14.1 (CH3), 52.5
(CH3), 55.9 (CH3), 56.0 (CH3), 56.1 (CH3), 60.8 (CH3), 60.9
(CH3), 61.2 (CH2), 61.3 (CH2), 75.0 (2 × CH), 107.0 (CH), 107.3
(CH), 125.8 (C), 126.2 (2 × C), 127.4 (2 × CH), 128.7 (2 × CH),
129.1 (2 × CH), 129.9 (C), 130.0 (C), 130.1 (C), 133.8 (C), 139.4
(C), 139.5 (C), 139.7 (C), 139.8 (C), 142.8 (CH), 143.2 (CH),
143.5 (CH), 153.1 (2 × CH), 166.1 (C), 166.2 (C), 166.7 (C),
168.9 (C); mass spectrum m/z (relative intensity) 650 (M+, 11),
501 (21), 411 (37), 195 (100), 181 (72), 91 (52), 77 (18); HRMS
calcd for C35H38O12 650.2363, found 650.2356.
presence of two compounds in an approximately 6:4 ratio. The
combined irradiated solutions were concentrated and cooled
to 0 °C overnight. The precipitate was filtered to give colorless
crystals of the major product (202.4 mg, 0.34 mmol, 26%):
1
[R]25 ) +475.7 (c ) 0.2, CHCl3); H NMR (CDCl3) δ 1.23 (d,
D
3H, J ) 7.0 Hz) 2.50 (s, 3H) 3.77 (s, 3H) 3.80 (s, 3H) 3.84 (s,
6H) 3.86 (s, 3H) 3.88 (s, 3H), 4.22 (s, 1H), 4.38-4.48 (m,1H),
4.84 (s, 1H), 5.37 (d, 1H, J ) 4.0 Hz), 6.48 (s, 1H), 6.50 (s,1H),
6.69 (s, 2H), 7.28-7.36 (m, 5H); 13C NMR (CDCl3) δ 14.4, 29.4,
36.8, 52.3, 55.6, 56.0, 56.1 (2C), 60.7, 60.8 (2C) 78.8, 104.3 (2C),
107.4, 121.6, 124.8, 127.5 (2C), 127.6, 128.2 (2C), 128.7, 128.8,
133.6, 136.9, 140.1, 143.5, 151.4, 152.5, 153.2 (2C), 172.5,
172.7; MS-EI m/z (relative intensity) 603 (M+ 35), 557 (14),
502 (12), 454 (13), 441 (11), 412 (28), 369 (14), 245 (11), 181
(15), 168 (16), 105 (27), 56 (100); HRMS for C34H37NO9 calcd
603.2468, found 603.2475.
In a similar experiment carried out on a smaller amount of
7 (100 mg) the mother liquors recovered after crystallization
of 8 were evaporated and chromatographed (ethyl acetate/
hexanes) to give 16 mg (16%) of the minor compound 9 as an
oil: 1H NMR (CDCl3) δ 1.25 (d, 2H, J ) 7.0 Hz), 2.48 (s, 3H),
3.54 (dd, 1H, J ) 1.1, 22.2 Hz), 3.67 (s, 3H), 3.78 (s, 6H), 3.79
(s, 3H), 3.84 (s, 3H), 3.87 (s, 3H), 4.15 (dq, 1H, J ) 7.0, 3.7
Hz), 4.41 (dd, 1H, J ) 3.2, 22.2 Hz), 5.35 (d, 1H, J ) 3.7 Hz),
5.72 (dd, 1H, J ) 1.1, 3.2 Hz), 6.48 (s, 2H), 6.54 (s, 1H), 7.28
(m. 2H), 7.35 (m, 3H); 13C NMR (CDCl3) δ 13.9, 29.8, 36.9, 40.8,
55.3, 56.17, 56.22 (2C), 60.8, 60.9 (2C), 85.8, 105.4 (2C), 105.9,
123.7, 127.3 (2C), 128.4 (2C), 128.5, 128.8, 134.2, 134.7, 137.1,
138.4, 141.4, 145.3, 151.0, 153.0, 152.3 (2C), 169.8, 172.1; MS-
EI m/z (relative intensity) 603 (M+ 75), 441 (47), 436 (100),
425 (39), 412 (58), 381 (47), 329 (58), 289 (36), 195 (49), 181
(E,E)-2,3-Di(3,4,5-t r im et h oxxyb en zylid en e)su ccin a t e
(-)-Ep h ed r in e Cyclic Am id e Ester (7). To a solution of
(E,E)-2,3-di(3,4,5-trimethoxybenzylidene)succinic acid25 (0.5 g,
1.05 mmol) in CH2Cl2/DMF (45 mL, 3:1) cooled to 0 °C were
added TBTU (0.337 g, 1.05 mmol) and DIEA (0.55 mL, 3.16
mmol). The solution was allowed to stir under nitrogen at 0
°C for 30 min. A second solution of (1R,2S)-(-)-ephedrine
(0.174 g, 1.05 mmol) in CH2Cl2/DMF (25 mL, 3:1) was added
slowly and dropwise over 30 min. The resulting mixture was
stirred under nitrogen at 0 °C for 1 h 30 min and then at room
temperature overnight. After the mixture was cooled to 0 °C,
a further equivalent of TBTU (0.337 g, 1.05 mmol) was added
and the reaction stirred under nitrogen at 0 °C for 1 h 30 min
and for an additional 18 h at room temperature. CH2Cl2 was
evaporated, and the residue was mixed with water (50 mL)
and acidified with 10% aqueous HCl (25 mL). The crude
reaction product was extracted with EtOAc. The combined
organic layers were washed with brine and dried over mag-
nesium sulfate, filtered, and evaporated to dryness to give a
brown oil. The crude product was purified by flash column
chromatography (EtOAc/hexanes, 60:40) to afford an orange
(36), 168 (64), 148 (47), 147 (45), 91 (34); HRMS for C34H37
NO9 calcd 603.2468, found 603.2455.
-
(1S,2R)-1-(3,4,5-Tr im eth oxyp h en yl)-6,7,8-tr im eth oxy-
1,2-d ih yd r on a p h th a len e-2,3-d ica r boxylic Acid (10). To
compound 8 (202.4 mg, 0.34 mmol) was added a methanolic
solution of 3 M KOH (50 mL). The mixture was refluxed under
nitrogen for 3 h and the methanol then partially evaporated.
The solution was acidified with 10% aqueous HCl (150 mL)
and extracted with EtOAc. The organic layer was washed with
water, dried over MgSO4, and evaporated to dryness to give a
colorless oil that solidified (0.1832 g). The crude compound was
used in the next step without further purification: 1H NMR
(CDCl3) δ 3.63 (s, 3H), 3.70 (s, 6H), 3.76 (s, 3H), 3.88 (s, 3H),
3.89 (s, 3H), 4.02 (s, 1H), 5.02 (s, 1H), 6.22 (s, 2H), 6.73 (s,
1H), 7.73 (s, 1H).
Dieth yl (1S,2R)-1-(3,4,5-Tr im eth oxyp h en yl)-6,7,8-tr i-
m eth oxy-1,2-d ih yd r on a p h th a len e-2,3-d ica r boxyla te (11).
To crude 7 (183.2 mg, 0.39 mmol) was added anhydrous EtOH
(50 mL) followed by H2SO4 (concd, 1.5 mL) and the solution
refluxed under nitrogen overnight. The ethanol was evaporated
and the residue diluted with water and extracted with EtOAc.
Flash column chromatography of the ethyl acetate extract
(silica gel, EtOAc/hexanes, 60:40) gave 11 as a colorless oil
that solidified (111.4 mg, 0.21 mmol, 54%). This product was
spectroscopically identical to the racemic compound previously
reported,18 but with rotation [R]25D ) +129.7 (c ) 0.9, MeOH).
Dieth yl (1S,2R)-1-(3,4,5-Tr im eth oxyp h en yl)-6,7,8-tr i-
m eth oxy-1,2,3,4-tetr a h yd r on a p h th a len e-2,3-d ica r boxyl-
a te (12). Diester 11 (111.4 mg, 0.21 mmol) was added to
nitrogen-purged methanol (30 mL), 5% Pd/C (60 mg) was
added, and the solution was vigorously stirred under H2
atmosphere at room temperature overnight. The Pd/C was
filtered off using Celite. The methanol was evaporated to
dryness to give a colorless oil that solidified (97.7 mg, 0.18
mmol, 87%). The diastereomeric ratio (78:22) was determined
by 1H NMR of the crude product (δ 6.25 for major isomer, δ
6.22 for minor isomer). The product was used in the next step
with no further purification.
oil that solidified (0.277 g, 0.46 mmol) in 44% yield: [R]25
)
D
1
-342.9 (c ) 0.2, CHCl3); H NMR (CDCl3) δ 1.35 (d, 3H, J )
7.0 Hz), 2.90 (s, 3H), 3.81 (s, 3H), 3.82 (s, 6H), 3.86-3.88 (m,
4H), 3.90 (s, 6H) 6.29 (s, 1H), 6.79 (s, 2H) 7.07 (s, 2H) 7.30 (s,
1H) 7.32-7.43 (m, 5H); 7.63 (s, 1H); 13C NMR (CDCl3) δ 11.8,
35.8, 56.1 (2), 56.4 (2), 60.81, 60.84, 64.6, 81.2, 107.3 (2), 108.4
(2), 125.7 (2), 126.1, 128.4, 128.67, 128.74, 129.0, 129.5, 139.4,
139.7, 139.9, 146.3, 145.4, 153.1 (2), 153.2 (2), 168.0, 169.9;
MS-EI m/z (relative intensity) 603 (M+ 20), 557 (25), 500 (12),
454 (33), 411 (20), 392 (16), 289 (11), 245 (8), 181 (32), 168
(54), 146 (24), 56 (100); HRMS for C34H37NO9 calc 603.2458,
found 603.2468.
(1S,2R)-1-(3,4,5-Tr im eth oxyp h en yl)-6,7,8-tr im eth oxy-
1,2-dih ydr on aph th alen e-2,3-dicar boxylate (-)-Eph edr in e
Cyclic Am id e Ester (8) a n d (1S,2R)-1-(3,4,5-Tr im eth oxy-
p h en yl)-6,7,8-tr im eth oxy-1,4-d ih yd r on a p h th a len e-2,3-d i-
ca r boxyla te (-)-Ep h ed r in e Cyclic Am id e Ester (9). Com-
pound 7 (0.7712 g, 1.28 mmol) was dissolved in 2-propanol (700
mL), 100 mL aliquots of that solution were placed in a Vycor
cylinder and purged with nitrogen for 15 min, and the solution
was irradiated for 30 min (254 nm). NMR spectroscopy of a
small aliquot that was evaporated to dryness showed the
(+)-Lyon ir esin ol Dim eth yl Eth er (13). Dry THF (40 mL)
was introduced into a nitrogen-purged 100 mL round-bottom
flask containing the isomeric mixture of 12 (97.7 mg, 0.18
(25) Bambagotti-Alberti, M.; Coran, S. A.; Vincieri, F. F.; Mulinacci,
N.; Pieraccini, G. M. L. Heterocycles 1988, 27, 2185-96.
J . Org. Chem, Vol. 69, No. 12, 2004 4143

Assoumatine et al.
mmol) and LiAlH4 (80 mg, 2.11 mmol). The resulting slurry
was stirred at room temperature for 3 h. Workup was achieved
using Fieser’s method: consecutive slow adjunction of water
(0.24 mL), 15% aqueous NaOH (0.24 mL), and water (0.72 mL).
The granular precipitate was filtered off using Celite. The
filtrate was evaporated to dryness to give a colorless oil residue
that solidified (89.6 mg, 0.20 mmol). The diastereomeric ratio
(80:20) was determined by H NMR spectroscopy of the crude
product (δ 6.33 for major isomer, δ 6.19 for minor isomer).
Three recrystallizations from CH2Cl2/hexanes of the crude
C24H32O8 calcd 448.2097, found 448.2079. The compound was
observed to be spectroscopically identical to that previously
reported.19-23 The optical rotation was observed to be be-
low the rotation measured for the isolated natural product
([R]25D ) +49.4)21 and above all the values previously reported
for the synthetic product ([R]25 ) +21.5,22 +26.0,23 +30.023).
D
1
Ack n ow led gm en t. We are grateful to the Natural
Sciences and Engineering Research Council of Canada
and the University of Manitoba for financial support of
this project.
product gave the major isomer as the pure diastereoisomer
1
12: mp 157-159 °C; [R]25 ) +41.2 (c 0.7, CHCl3); H NMR
D
(CDCl3) δ 1.72-1.94 (m, 2H), 2.58-2.77 (m, 2H), 3.22 (s, 3H),
3.59-3.87 (m, 4H), 3.75 (s, 3H), 3.77 (s, 6H), 3.79 (s, 3H), 3.85
1
Su p p or tin g In for m a tion Ava ila ble: Copies of H NMR
(s, 3H), 3.97 (d, 1H, J ) 8.0 Hz), 6.30 (s, 2H), 6.46 (s, 1H); 13
C
spectra (compounds 6-13) and 13C NMR spectra (compounds
6-9 and 13); general experimental details. This material is
available free of charge via the Internet at http://pubs.acs.org.
NMR (CDCl3) δ 34.0, 40.2, 43.7, 49.4, 55.8, 56.2, 59.6, 60.5,
60.9, 63.7, 66.6, 105.8, 106.7, 125.0, 133.1, 136.2, 140.7, 143.3,
152.0, 152.1, 152.9; EI-MS m/z (relative intensity) 448 (M+ 38),
430 (100), 399 (28), 219 (45), 181 (93), 69 (63); HRMS for
J O0497454
4144 J . Org. Chem., Vol. 69, No. 12, 2004