Synthesis and structures of regioisomeric hydnocarpin-type flavonolignans

Article abstract of DOI:10.1021/np000166d

Flavonolignans represent natural compounds whose biosynthesis presumes a radical coupling of a ring B catecholic flavonoid with a molecule of coniferyl alcohol or an analogue. Many natural flavonolignans can exist as regioisomers, depending on how the coupled coniferyl alcohol moiety orients to the flavonoid. These regioisomers are often difficult to separate and have virtually identical NMR spectra. Structural assignments for some have changed with time or have been given without proof. We here report syntheses of both regioisomers of the flavonolignan hydnocarpin and one isomer of a plant isolate previously known as 5'-methoxyhydnocarpin. This isomer, here renamed 5'-methoxyhydnocarpin-D, was recently shown to be a potent inhibitor of a Staphylococcus aureus multidrug resistant efflux pump.

Full text of DOI:10.1021/np000166d

1140
J . Nat. Prod. 2000, 63, 1140-1145
Syn th esis a n d Str u ctu r es of Regioisom er ic Hyd n oca r p in -Typ e F la von olign a n s
Nathan R. Guz and Frank R. Stermitz*
Department of Chemistry, Colorado State University, Fort Collins, Colorado 80523
Received April 14, 2000
Flavonolignans represent natural compounds whose biosynthesis presumes a radical coupling of a ring
B catecholic flavonoid with a molecule of coniferyl alcohol or an analogue. Many natural flavonolignans
can exist as regioisomers, depending on how the coupled coniferyl alcohol moiety orients to the flavonoid.
These regioisomers are often difficult to separate and have virtually identical NMR spectra. Structural
assignments for some have changed with time or have been given without proof. We here report syntheses
of both regioisomers of the flavonolignan hydnocarpin and one isomer of a plant isolate previously known
as 5′-methoxyhydnocarpin. This isomer, here renamed 5′-methoxyhydnocarpin-D, was recently shown to
be a potent inhibitor of a Staphylococcus aureus multidrug resistant efflux pump.
In 1968, what was described as the first flavonolignan
(silybin) was reported as an isolate from fruits of the
medicinal plant Silybum marianum and identified on the
basis of spectroscopic analysis as either 1a or 2a .1a It was
stated that the data did not distinguish between these
structures and that structure 1a was used only “for
convenience” when the data were discussed. It was later
shown1b-d that silybin was actually 1b and that another
isolate, the regioisomer isosilybin, was 2b. The second
flavonolignan reported, identified in 1973 as an isolate from
Hydnocarpus wightiana, was given the name hydnocarpin
and on the basis of spectroscopy and chemical reactions
was assigned structure 3, “based on analogy with silymarin
and also on reactivity considerations”.2 Silymarin was an
alternate name for silybin.3 The spectral data reported for
hydnocarpin could not distinguish between 3 and the
alternate structure 4, and there was no discussion of what
the “reactivity considerations” were for the assignment. In
1974 the same research group reported on the isolation of
two minor components from H. wightiana, isohydnocarpin4
(a structural, not regiochemical isomer of hydnocarpin) and
what was termed 5′-methoxyhydnocarpin (5).5 NMR and
mass spectral data for the latter isolate as well as for
derivatives were presented, and the structure was given
as 5 based on these data and “the similarity in the spectra
[of the isolate and that of hydnocarpin]”. Hydnocarpin was
also reported6 in 1977 from seeds of Cassia absus and
compared to the H. wightiana isolate by spectral data,
mixed melting point, co-TLC, and co-IR. (The hydnocarpins
were devoid of optical activity and hence are scalemic
isolates. The stereochemistry at C-12/C-13 was trans in all
these flavonolignans. Structures 1-5 (Chart 1) do not imply
absolute configurations.)
We recently isolated9 a compound whose 1H and 13C
NMR, mass, and UV spectra were essentially identical with
those reported for 5′-methoxyhydnocarpin5 and that was
a potent inhibitor of a Staphylococcus aureus multidrug
resistance efflux pump.10 To provide more material for
testing and to begin to probe bacterial MDR pump inhibi-
tory structure-activity relationships among flavonolig-
nans, we elected to use a synthetic approach. We could find
no literature reports on the synthesis of 5′-methoxyhyd-
nocarpin and only a single report11 on that of hydnocarpin,
via coupling of luteolin and coniferyl alcohol in a cell-free
suspension culture from fruit of S. marianum.
Resu lts a n d Discu ssion
Hyd n oca r p in a n d Hyd n oca r p in -D. The literature
suggested that horseradish peroxidase11-initiated coupling
of the flavone luteolin with the lignan coniferyl alcohol in
a biomimetic-type reaction might yield hydnocarpin (re-
gioisomer 4). On the other hand, the same coupling with
Ag2O1d,12 or Ag2CO3 should provide regioisomer 3, an
13
unknown compound here denoted hydnocarpin-D. These
preparations were indeed achieved (Scheme 1), although
the resulting regioselectivities were less than expected
based upon literature results for somewhat analogous
systems.9,12,13 In our hands, 3 and 4 were not separable by
usual HPLC methods, although extensive solvent and
column variations were not attempted. This parallels
difficulties experienced with similar separations in the
silybin/isosilybin case.1d,14 The 3/4 ratios in the crude
preparations were, however, easily determinable from the
ratios of the two 1H NMR doublets for H-13 (in 3) and H-12
(in 4), which appeared at δ 4.97 and 5.05, respectively, in
DMSO-d6. These ratios showed the crudes to be 9:1 3/4 for
the Ag2CO3-catalyzed reaction and 3:2 4/3 for the HRP
reaction. 13C NMR resonances for 3 and 4 were essentially
identical (all within 0.2 ppm), with the exception of those
for C-12 and C-13. For 3, C-12 was at δ 78.0 and C-13 at
δ 76.0, while for 4 C-13 was at δ 78.5 and C-12 at δ 75.9.
Although small, these resonance differences were some-
times distinguishable in a mixture of the two. It is
Shortly thereafter, americanin-A was isolated from Phy-
tolacca americana and, on the basis of spectroscopic
evidence, given structure 6.7 It was recognized that spec-
troscopic evidence alone could not distinguish between 6
and its regioisomer, and hence definitive proof for 6 was
established by correlation of degradation products of 6
derivatives with the same derivatives available from silybin
(1b) and isosilybin (2b). One of the americanin-A degrada-
tion products was used to synthesize a trimethyl ether of
isolated hydnocarpin, and their identity resulted8 in a final
proof for the structure of hydnocarpin as 4.
1
debatable if either the H or 13C NMR spectra alone would
differentiate the two compounds unless both were avail-
able.
5′-Meth oxyh ydn ocar pin -D. Because regiochemical mix-
tures were obtained in the above oxidative coupling reac-
tions, we elected first to attempt a regiospecific synthesis
of 5′-methoxyhydnocarpin. In a series of reactions on
* To whom correspondence should be addressed. Tel: 970-491-5158.
Fax: 970-491-5610. E-mail: frslab@lamar.colostate.edu.
10.1021/np000166d CCC: $19.00
© 2000 American Chemical Society and American Society of Pharmacognosy
Published on Web 07/28/2000

Regioisomeric Hydnocarpin-Type Flavonolignans
J ournal of Natural Products, 2000, Vol. 63, No. 8 1141
Ch a r t 1
substituted catechols such as 3,4-dihydroxytoluene and 3,4-
dihydroxybenzaldehyde, it was suggested12 that in the
presence of an electron-donating group (such as the methyl
on 3,4-dihydroxytoluene) the coupling reaction with co-
niferyl alcohol would preferentially result in the regioiso-
mer with the pendent aromatic ring down (as in 3, for
example). On the other hand the presence of an electron-
withdrawing group (such as the carboxaldehyde on 3,4-
dihydroxybenzaldehyde) would result in the opposite re-
gioisomer (as in 4, for example). We therefore coupled
coniferyl alcohol with 3,4-dihydroxy-4-methoxybenzalde-
hyde (12) and then further elaborated the flavonolignan
via chalcone 15 (Scheme 2). A single regioisomeric flavono-
lignan was indeed obtained, but an X-ray diffraction crystal
structure of intermediate 7 showed that the final product
had to be 5 and not the expected12 isomer 9. The NMR
spectra were identical with those of the purported 5′-
methoxyhydnocarpin.2,10
Although the NMR spectra of synthesized 5 and our
isolated compound were apparently identical, our experi-
ence with 3 and 4 suggested that one could not be certain
which was on hand if only one isomer was available. We
hence needed independent evidence for the structure of the
isolated flavonolignan. This was achieved by HMBC NMR
with the coupling constant optimized to 1.6 Hz, a technique
similar to the “selective heteronuclear decoupling” experi-
ment used to prove the regiochemistry of coumarinolig-
noids.15 As a model, compound 7 (whose structure was
known from X-ray) showed the following diagnostic HMBC
correlations: H-3 (δ 5.0 resonance) to C-2 (79.1), C-12
(109.7), and C-4a (144.9); H-10 (δ 3.6 resonance) to C-3
(76.2) and C-8a (138.7). The experiment was then repeated
on the isolated9 flavonolignan. Although the same entire
array of correlations were not seen, the key H-13 to C-3′
correlation (structure 5) was indeed seen. Thus, the natural
5′-methoxy compound2,9,10 does not have the same regio-
chemistry as hydnocarpin, and hence we have designated
it as 5′-methoxyhydnocarpin-D.
Since 3, 4, and 5 as well as some additional analogues
were found to be inhibitors of the S. aureus NorA MDR
pump (unpublished results), we desired a sample of true
5′-methoxyhydnocarpin (9) for structure-activity relation-
ship studies. By analogy with the syntheses of 3 and 4,
radical-initiated coupling of coniferyl alcohol to the rare
flavone selgin might yield a mixture of 5 and 9. Selgin (8),
also known as selagine and selagin, was first reported from
Huperzia selago and its structure proven by spectroscopic
analysis and synthesis.15 We prepared 8 by a much more
efficient method (Scheme 3) and then coupled it with
coniferyl alcohol using Ag2CO3 as well as horseradish
peroxidase. Unlike these reactions in the preparation of 3
and 4, only a single flavonolignan, again 5, resulted.
Regiochemical product ratios from these coupling reactions
are probably a result of the relative stability of radicals
10 and/or 11, where 10 would lead to 5 and 11 would lead
to 9. The methoxy group ortho to the phenoxide radical in
10 may provide extra stabilization which is unavailable in
the case of 11. Where the methoxy group is absent, as in
the precursor radicals to 3 and 4, there is less difference
in the radical stabilities and mixtures arising from both
radicals are seen.
Exp er im en ta l Section
Gen er a l Exp er im en ta l P r oced u r es. 1H and 13C NMR
spectra were recorded on a Varian Inova spectrometer at 400
and 100 MHz, respectively, using CDCl3, acetone-d,6 or DMSO-
d6 as the solvent and internal reference. Melting points were
determined on
a Laboratory Device’s Mel-Temp and are
uncorrected. All solvents were distilled prior to use. THF and
1,4-dioxane were freshly distilled from benzophenone-ketyl and
benzene was freshly distilled from CaH2. ACS acetone was
stored over 4 Å molecular sieves. All nonaqueous reactions

1142 J ournal of Natural Products, 2000, Vol. 63, No. 8
Guz and Stermitz
Sch em e 1. Biomimetic Radical Coupling Syntheses of Hydnocarpin (4) and Hydnocarpin-D (3)
Sch em e 2. Synthesis of 5′-Methoxyhydnocarpin-D (5) via an Intermediate (7) of Known (X-ray Diffraction) Regiochemistry
were performed in dry glassware under an argon atmosphere.
All starting materials were used as received. Luteolin was
purchased from Indofine Chemical Co., and all other reagents
were purchased from Aldrich Chemical Co.
Hyd n oca r p in -D (3). To a 250 mL three-neck round-bottom
flask was added 0.360 g (1.26 mmol) of luteolin and 0.227 g
(1.26 mmol) of coniferyl alcohol with 50 mL of dry benzene
and 25 mL of dry acetone. The reaction vessel was placed in a
60 °C oil bath and let stir for 10 min. Next, 0.347 g (1.26 mmol)
of Ag2CO3 was added and the reaction solution stirred vigor-
ously for 36 h. The reaction was then allowed to cool and
using 95:5 CHCl3/MeOH to yield 0.220 g of a 9:1 (by NMR)
mixture of 3 and 4 plus other minor impurities. The sample
was recrystallized from 9:1 MeOH/H2O to yield 0.125 g (21%)
of pure 3 and 4. This mixture was acetylated using standard
acetic anhydride/pyridine conditions to yield an off-white
powder, which was washed with acetone until bright white. A
standard deprotection using K2CO3/MeOH was used to remove
the acetyl groups and yielded 0.040 g (7%, yellow-white
powder) of pure 3: 1H NMR (DMSO-d6) δ 3.37 (dd, J ) 12.6,
4.8 Hz, H-11a), 3.57 (dd, J ) 12.6, 2.8 Hz, H-11b), 4.31 (m,
H-12), 4.97 (d, J ) 8.0 Hz, H-13), 6.19 (d, J ) 2.0 Hz, H-6),
6.51 (d, J ) 2.0 Hz, H-8), 6.82 (dd, J ) 8.0, 1.6 Hz, H-5′′), 6.87
(s, H-3), 6.89 (dd, J ) 8.0, 1.6 Hz, H-6′′), 7.05 (d, J ) 1.6 Hz,
H-2′′), 7.12 (d, J ) 8.4 Hz, H-5′), 7.63 (dd, J ) 8.8, 2.0 Hz,
filtered through
a Buchner funnel, and the solvent was
removed by rotary evaporation to yield a yellow powder. The
yellow powder was subjected to column chromatography (CC)

Regioisomeric Hydnocarpin-Type Flavonolignans
J ournal of Natural Products, 2000, Vol. 63, No. 8 1143
Sch em e 3. Biomimetic Synthesis of 5′-Methoxyhydnocarpin-D (5) via the Flavone Selgin (8)
H-6′), 7.67 (d, J ) 2.0 Hz, H-2′), 9.18 (br s), 10.85 (br s), 12.90
(br s); 13C NMR (DMSO-d6) δ 55.7 (OMe), 60.0 (C-11), 75.9
(C-13), 78.5 (C-12), 94.1 (C-8), 98.9 (C-6), 103.8 (C-3), 103.8
(C-10), 111.8 (C-2′′), 115.0 (C-2′), 115.3 (C-5′′), 117.4 (C-5′),
120.1 (C-6′), 120.6 (C-6′′), 123.4 (C-1′), 127.0 (C-1′′), 144.0 (C-
4′), 146.9 (C-3′), 147.1 (C-4′′), 147.6 (C-3′′), 157.3 (C-9), 161.4
(C-5), 162.9 (C-7), 164.2 (C-2), 181.8 (C-4). Peracetate: 1H NMR
(CDCl3) δ 2.08 (s), 2.34 (s), 2.35 (s), 2.44 (s), 3.88 (s, OMe),
4.03 (dd, J ) 12.4, 4.0 Hz), 4.34 (m), 4.41 (dd, J ) 12.0, 3.0
Hz), 5.00 (d, J ) 8 Hz), 6.57 (s), 6.84 (d, J ) 2.1 Hz), 6.98 (d,
J ) 1.8 Hz), 7.02 (m), 7.11 (d, J ) 8.7 Hz), 7.12 (d, J ) 7.8
Hz), 7.32 (d, J ) 2.1), 7.45 (dd, J ) 8.4, 2.1), 7.53 (d, J ) 2.1);
13C NMR (CDCl3) δ 20.7, 21.1, 21.1, 21.2, 56.0, 62.4, 75.9, 76.3,
107.5, 108.9, 110.9, 113.5, 115.3, 117.8, 119.7, 120.2, 123.3,
124.4, 133.9, 140.6, 143.6, 146.0, 150.0, 151.6, 153.7, 157.4,
161.8, 167.9, 168.6, 169.3, 170.2, 176.2; mp 233-234 °C; anal.
C 63.00%, H 4.11%, calcd for C33H28O13, C 62.66%, H 4.46%.
Hyd n oca r p in (4). To a 100 mL three-neck round-bottom
flask was added 0.300 g (1.05 mmol) of luteolin and 0.189 g
(1.05 mmol) of coniferyl alcohol. Next, 20 mL of dry ACS
acetone and 5 mL of a 0.2 M citric acid/phosphate buffer were
added to the reaction flask, and the flask was cooled to 0 °C.
Two drops of 30% H2O2 and 1 mL of a horseradish peroxidase
solution (1.5 mg HRP (1100 U/mg)/3 mL water) were added.
The HRP solution (1 mL) was added every 15 min thereafter,
and the reaction then allowed to stir at 0 °C for 7 h. The
reaction solution was allowed to warm to room temperature,
washed with brine, and extracted with EtOAc. The EtOAc was
dried with anhydrous MgSO4, filtered, and removed by rotary
evaporation to yield a brown-orange solid. This solid was
subjected to CC using 95:5 CHCl3/MeOH to yield 0.187 g (40%)
of a white-yellow solid, 3:2 ratio of 4 to 3 by 1H NMR. This
solid was further purified by reversed-phase (C-18) vacuum-
liquid chromatography (75:25 H2O/MeOH to 20:80 H2O/MeOH)
to yield 0.087 g of a white-yellow solid that was recrystallized
from 9:1 MeOH/H2O to yield 0.025 g (5%, white-yellow powder,
mp 258-259 °C) of pure 4. 1H and 13C NMR spectra were
essentially identical to the literature.16
the solvent was removed by rotary evaporation to yield a brown
microcrystalline solid. This solid was subjected to CC using
4:6 hexanes/EtOAc to yield 0.687 g (72%) of a near pure sample
of 7. This solid was recrystallized from 7:3 EtOH/H2O to yield
0.344 g (36%, white microcrystalline solid, mp 103-105 °C) of
pure 7: 1H NMR (CDCl3) δ 3.60 (dd, J ) 12.6, 4.0 Hz), 3.93 (s,
OMe), 3.94 (m), 3.97 (s, OMe), 4.10 (m), 4.99 (d, J ) 8.0 Hz),
6.94 (br s, 1H), 6.97 (m, 2 H), 7.13 (d, J ) 2.0 Hz), 7.16 (d, J
) 2.0 Hz), 9.80 (s); 13C NMR (CDCl3) δ 56.0, 56.3, 61.2, 76.1,
79.1, 103.0, 109.6, 114.6, 114.8, 120.8, 127.3, 129.4, 138.7,
144.4, 146.5, 147.0, 149.5, 190.8; anal. C 62.55%, H 5.00%,
calcd for C18H16O7, C 62.61%, H 4.96%.
2-H yd r oxym e t h yl-8-m e t h oxy-3-(3-m e t h oxy-4-m e t h -
oxym eth oxyp h en yl)-2,3-d ih yd r oben zo[1,4]d ioxin e-6-ca r -
ba ld eh yd e (13). To a 100 mL round-bottom flask was added
0.196 g (0.569 mmol) of 7 and 35 mL of dry THF. The flask
was placed into an ice-water bath, and 0.017 g (0.683 mmol)
of NaH (washed prior with hexanes) was added. The solution
was stirred 1 h, then 0.055 g (0.683 mmol) of MOMCl was
added, and the reaction was allowed to warm to room tem-
perature. The reaction was stirred for 6 h at room temperature,
poured into a saturated solution of NaHCO3, and extracted
with EtOAc. The EtOAc was dried with anhydrous MgSO4,
filtered, and removed by rotary evaporation to yield an off-
white microcrystalline solid. This solid was subjected to CC
using 3:7 hexanes/EtOAc to yield 0.187 g (85%, white micro-
crystalline solid, mp 67-68 °C) of pure 13: 1H NMR (CDCl3)
δ 3.52 (s), 3.60 (dd, J ) 12.0, 3.3 Hz), 3.91 (s, OMe), 3.93 (m),
3.95 (s, OMe), 4.10 (m), 5.03 (d, J ) 8.1 Hz), 5.25 (s), 6.98 (d,
J ) 1.8 Hz), 6.99 (dd, J ) 8.7, 1.8 Hz), 7.12 (d, J ) 1.8 Hz),
7.15 (d, J ) 1.8 Hz), 7.20 (d, J ) 8.7 Hz), 9.79 (s); 13C NMR
(CDCl3) δ 30.9, 56.0, 56.2, 61.2, 75.9, 78.9, 95.3, 103.0, 110.5,
114.5, 116.2, 120.0, 129.3, 129.5, 138.6, 144.2, 147.1, 149.4,
149.9, 190.6.
3-[2-Hyd r oxym eth yl-8-m eth oxy-3-(3-m eth oxy-4-m eth -
oxym eth oxyp h en yl)-2,3-d ih yd r oben zo[1,4]d ioxin -6-yl]-1-
(2,4,6-tr ism eth oxym eth oxyp h en yl)p r op en on e (14). To a
25 mL round-bottom flask was added 0.098 g (0.252 mmol) of
13 and 0.076 g (0.252 mmol) of 2,4,6-tris(methoxymethoxy)-
acetophenone, 10 mL of EtOH, and 0.339 g of crushed solid
KOH. The blocked acetophenone was prepared from reaction
of the bis-protected acetophenone18 with NaH (1.3 equiv) and
MOMCl (1.3 equiv) in THF at 0 °C for 4 h (86%). This was
stirred at room temperature for 17 h, washed with brine, and
extracted with EtOAc. The EtOAc was dried with anhydrous
MgSO4, filtered, and removed by rotary evaporation to yield a
yellow oil, which was subjected to CC 2:8 using hexanes/EtOAc
3-(4-H yd r oxy-3-m et h oxyp h en yl)-2-h yd r oxym et h yl-8-
m eth oxy-2,3-d ih yd r oben zo[1,4]d ioxin e-6-ca r ba ld eh yd e
(7). To a 250 mL three-neck round-bottom flask was added
0.486 g (2.77 mmol) of 12 and 0.500 g (2.77 mmol) of coniferyl
alcohol with 100 mL of dry benzene and 20 mL of dry acetone.
This solution was allowed to stir 20 min in a 60 °C oil bath,
and then 0.765 g (2.77 mmol) of Ag2CO3 was added and the
reaction stirred vigorously for 7 h. The reaction was then
allowed to cool and filtered through a Buchner funnel, and

1144 J ournal of Natural Products, 2000, Vol. 63, No. 8
Guz and Stermitz
to yield 0.115 g (68%, yellow solid) of pure 14: 1H NMR
(acetone-d6) δ 3.36 (s), 3.46 (s), 3.52 (dd, J ) 12.8, 3.6 Hz),
3.83 (m), 3.85 (s, OMe), 3.89 (s, OMe), 4.14 (m), 5.06 (d, J )
7.6 Hz), 5.14 (s), 5.19 (s), 5.22 (s), 6.88 (d, J ) 2.0 Hz), 6.88 (d,
J ) 16.0 Hz), 6.98 (d, J ) 2.0 Hz), 7.03 (dd, J ) 8.4, 2.0 Hz),
7.15 (d, J ) 8.4 Hz), 7.17 (d, J ) 2.0 Hz), 7.24 (d, J ) 16.0
Hz); 13C NMR (acetone-d6) δ 56.3, 56.3, 56.3, 56.4, 56.4, 56.5,
61.6, 76.8, 79.6, 95.3, 95.3, 95.4, 95.4, 96.3, 97.9, 105.2, 111.4,
112.7, 118.1, 121.1, 128.1, 128.6, 132.2, 136.8, 145.5, 148.0,
150.5, 151.5, 156.6, 160.3, 193.6.
3-[2-Hyd r oxym eth yl-8-m eth oxy-3-(4-h yd r oxy-3-m eth -
oxyp h en yl)-2,3-d ih yd r oben zo[1,4]d ioxin -6-yl]-1-(2,4,6-tr i-
h yd r oxyp h en yl)p r op en on e (15). To a 25 mL round-bottom
flask was added 0.042 g of 14, 5 mL of MeOH, and 2 drops of
concentrated HCl. This solution was stirred at room temper-
ature for 17 h, poured into a saturated NaHCO3 solution, and
extracted with EtOAc. The EtOAc was dried with anhydrous
MgSO4, filtered, and removed by rotary evaporation to yield a
brown-orange oil that was subjected to CC using 92:8 CHCl3/
MeOH to yield 0.022 g (72%, brown-orange microcrystalline
solid, mp 175-177 °C) of pure 15: 1H NMR (acetone-d6) δ 3.52
(dd, J ) 12.8, 3.6 Hz), 3.83 (m), 3.88 (s, OMe), 3.91 (s, OMe),
4.13 (m), 5.03 (d, J ) 8.0 Hz), 5.97 (s), 6.89 (d, J ) 8.0 Hz),
6.93 (d, J ) 2.0 Hz), 6.94 (d, J ) 2.0 Hz), 6.98 (dd, J ) 8.0, 2.0
Hz), 7.14 (d, J ) 2.0 Hz), 7.70 (d, J ) 15.6 Hz), 8.13 (d, J )
15.6 Hz); 13C NMR (acetone-d6) δ 56.4, 56.5, 61.8, 77.1, 79.8,
96.2, 105.8, 106.1, 110.8, 112.0, 115.9, 121.7, 126.7, 128.8,
129.2, 136.9, 143.4, 145.8, 148.2, 148.6, 150.4, 165.6, 165.8,
193.2; anal. C 56.10%, H 5.42%, calcd for C26H24O10.3.5H2O,
C 55.81%, H 5.58%.
5,7-Dih yd r oxy-2-[3-(4-h yd r oxy-3-m et h oxyp h en yl)-2-
h yd r oxym eth yl-8-m eth oxy-2,3-d ih yd r oben zo[1,4]d ioxin -
6-yl]ch r om a n -4-on e (16). To a 50 mL round-bottom flask was
added 0.021 g (0.043 mmol) of 15, 15 mL of MeOH, and 0.035
g (0.43 mmol) of NaOAc. This solution was heated at reflux
for 3 h, poured into a saturated NaHCO3 solution, and
extracted with EtOAc. The EtOAc was dried with anhydrous
MgSO4, filtered, and removed by rotary evaporation to yield
an off-white solid. This solid was subjected to CC using 1:9
hexanes/EtOAc to afford 0.019 g (90%, white powder) of 16 as
a 1:1 (1H NMR) diastereomeric mixture (from integration of
the singlets at δ 12.170 and δ 12.173): 1H NMR (CDCl3) δ
3.19 (d, J ) 12.4 Hz), 3.23 (d, J ) 12.8 Hz), 3.52 (m), 3.81 (m),
3.87 (s, OMe), 3.88 (s, OMe), 4.07 (m), 5.01 (d, J ) 7.6 Hz),
5.45 (dd, J ) 12.4, 2.4 Hz), 5.89 (m), 5.95 (m), 6.76 (m), 6.83
(m), 6.88 (d, J ) 8.0 Hz), 6.98 (dd, J ) 8.0, 2.0 Hz), 7.13 (m),
7.76 (bs), 9.62 (bs), 12.17 (bs); HRFAB+ 497.1430 (calcd for
was dried with anhydrous Mg2SO4, filtered, and removed by
rotary evaporation to yield a bright yellow oil. The oil was
subjected to CC using 1:1 hexanes/EtOAc to yield 0.154 g (75%,
yellow solid) of pure 18: 1H NMR (CDCl3) δ 3.49 (s), 3.52 (s),
3.54 (s), 3.63 (s), 3.90 (s, OMe), 5.19 (s), 5.20 (s), 5.25 (s), 5.30
(s), 6.26 (d, J ) 2.4 Hz), 6.33 (d, J ) 2.4 Hz), 6.86 (d, J ) 2.0
Hz), 7.17 (d, J ) 2.0 Hz), 7.71 (d, J ) 15.6 Hz), 7.87 (d, J )
15.6 Hz); 13C NMR (CDCl3) δ 56.0, 56.2, 56.5, 56.9, 57.2, 94.0,
94.6, 95.0, 95.3, 97.5, 98.4, 106.9, 107.4, 108.6, 127.0, 131.6,
137.3, 142.4, 151.4, 153.5, 159.9, 163.5, 167.4, 192.6; anal. C
58.11%, H 6.16%, calcd for C24H30O11, C 58.29%, H 6.12%.
5-Hydr oxy-2-(3-m eth oxy-4,5-bis-m eth oxym eth oxyph en -
yl)-7-m eth oxym eth oxych r om en -4-on e (19). To a 25 mL
round-bottom flask was added 0.095 g (0.192 mmol) of 18,
0.109 g (0.480 mmol) of DDQ, and 10 mL of dry 1,4-dioxane.
This solution was heated at reflux for 48 h and allowed to cool,
and solvent was removed by rotary evaporation to yield a dark
brown solid. This solid was subjected to CC using 97:3 CH2-
Cl2/Me2CO to yield 0.056 g (65%, pale yellow powder, mp 122-
123 °C) of pure 19: 1H NMR (CDCl3) δ 3.52 (s), 3.55 (s), 3.63
(s), 3.97 (s, OMe), 5.23 (s), 5.27 (s), 5.29 (s), 6.50 (d, J ) 2.4
Hz), 6.63 (s), 6.69 (d, J ) 2.4 Hz), 7.14 (d, J ) 2.4 Hz), 7.34 (d,
J ) 2.4 Hz); 13C NMR (CDCl3) δ 56.3, 56.4, 56.4, 57.3, 94.1,
94.3, 95.4, 98.4, 100.2, 104.4, 105.8, 106.3, 107.8, 127.0, 138.8,
151.4, 153.8, 157.6, 162.0, 163.0, 163.8, 182.4; anal. C 58.83%,
H 5.27%, calcd for C22H24O10 C 58.93%, H 5.39%.
Selgin (8). To a 25 mL round-bottom flask was added 0.092
g (0.291 mmol) of 19, 3 mL of 3 N HCl, and 10 mL of MeOH.
This solution was heated at reflux for 20 min, allowed to cool,
washed with a saturated NaHCO3 solution, and extracted with
EtOAc. The EtOAc was dried with anhydrous MgSO4, filtered,
and removed by rotary evaporation to yield 0.047 g (72%, pale
yellow powder) of pure 8. The 1H NMR spectrum was es-
sentially the same as in the literature,15 but the 13C NMR
spectrum was not previously reported: 1H NMR (DMSO-d6) δ
3.87 (s, OMe), 6.20 (d, J ) 2.0 Hz), 6.48 (d, J ) 2.0 Hz), 6.83
(s), 7.15 (d, J ) 2.0 Hz), 7.17 (d, J ) 2.0 Hz), 9.25 (bs), 9.41
(bs), 10.85 (bs), 12.98 (bs); 13C NMR (DMSO-d 6) δ 56.3, 93.2,
98.8, 102.4, 103.3, 103.7, 107.5, 120.4, 138.6, 146.0, 148.6,
157.3, 161.5, 163.9, 164.2, 181.8.
5′-Meth oxyh yd n oca r p in -D (5). To a 50 mL three-neck
round-bottom flask was added 0.011 g (0.035 mmol) of 8, 0.007
g (0.035 mmol) of coniferyl alcohol, 15 mL of dry benzene, and
7.5 mL of dry acetone. The flask was placed in a 60 °C oil bath
and stirred for 20 min. Next, 0.010 g of Ag2CO3 was added
and the reaction solution stirred vigorously for 7 h. The
reaction was then allowed to cool and filtered through a
Buchner funnel, and the solvent was removed by rotary
evaporation to yield a yellow powder. The powder was sub-
jected to CC using 8:2 CH2Cl2/Me2CO to yield 0.006 g (33%,
C
26H25O10 497.1447).
5′-Meth oxyh yd oca r p in -D (5). To a 25 mL round-bottom
flask was added 0.016 g (0.032 mmol) of 16, 0.018 g of DDQ
(0.081 mmol), and 7 mL of dry 1,4-dioxane. The solution was
heated at reflux for 36 h and allowed to cool, and the solvent
was removed by rotary evaporation to yield a dark brown solid.
The solid was chromatographed on 95:5 CHCl3/MeOH to yield
0.012 g of pure 5 (72%, pale yellow powder): HRFAB+ m/z
495.1310 (calcd for C26H23O10 495.1291); 1H and 13C NMR
spectral data were identical with those of the isolate.9
1-(2-H yd r oxy-4,6-b is-m e t h oxym e t h oxyp h e n yl)-3-(3-
m eth oxy-4,5-bis-m eth oxym eth oxyph en yl)pr open on e (18).
To a 100 mL three-neck round-bottom flask was added 0.202
g (1.20 mmol) of 12, 1.75 g (16.22 mmol) of K2CO3 and 35 mL
of dry acetone. This solution was stirred for 10 min, then 0.242
g (3.00 mmol) of MOMCl was added. The solution was heated
at reflux for 15 min, poured into a saturated NaHCO3 solution,
and extracted with EtOAc. The EtOAc was dried with anhy-
drous MgSO4, filtered, and removed by rotary evaporation to
yield a brown, viscous oil. The oil was subjected to CC using
1:1 hexanes/EtOAc to yield a near pure sample of 17 (0.262 g,
85%, clear oil that partially solidified upon standing), which
was carried directly to the next step. To a 50 mL round-bottom
flask was added 0.129 g (0.505 mmol) of 17, 0.129 g (0.505
mmol) of 2-hydroxy-4,6-bis(methoxymethoxy)acetophenone,18
20 mL of EtOH, and 0.680 g (12.12 mmol) of crushed solid
KOH. This was stirred at room temperature for 17 h, brought
to pH 7 using 1 N HCl, and extracted with EtOAc. The EtOAc
1
pale yellow powder) of pure 1. H and 13C spectral data were
identical with those of 5 prepared regiospecifically (see above)
and with those of the isolate.9
Ack n ow led gm en t. This research was supported by the
Colorado State University Agricultural Experiment Station
and by Grant CHE 9619213 from the National Science
Foundation.
Su p p or tin g In for m a tion Ava ila ble: X-ray diffraction data for
7. This material is available free of charge via the Internet at http://
pubs.acs.org.
Refer en ces a n d Notes
(1) (a) Pelter, A.; Ha¨nsel, R. Tetrahedron Lett. 1968, 2911-2916. (b)
Pelter, A.; Ha¨nsel, R. Chem. Ber. 1975, 108, 790-802. (c) Ha¨nsel, R.;
Schulz, J .; Pelter, A.; Chem. Ber. 1975, 108, 1482-1501. (d) Arnone,
A.; Merlini, L.; Zanarotti, A. J . Chem. Soc., Chem. Commun. 1979,
696-697.
(2) Raganathan, K. R.; Seshadri, T. R. Tetrahedron Lett. 1973, 3481-
3482.
(3) Wagner, H.; Ho¨rhammer, L.; Mu¨nster, R. Arzneim.-Forsch. 1968, 18,
698.
(4) Raganathan, K. R.; Seshadri, T. R. Indian J . Chem. 1974, 12, 888-
889.
(5) Raganathan, K. R.; Seshadri, T. R. Indian J . Chem. 1974, 12, 993.
(6) Kostova, I.; Rangaswami, S. Indian J . Chem. 1977, 15b, 764-765.

Regioisomeric Hydnocarpin-Type Flavonolignans
J ournal of Natural Products, 2000, Vol. 63, No. 8 1145
(7) Woo, W. S.; Kang, H.; Wagner, H.; Chari, V. M. Tetrahedron Lett.
1978, 3239-3242.
(14) Tittel, G.; Wagner, H. J . Chromatography 1977, 135, 499-501.
(15) Ray, A. B.; Chattopadyay, S. K.; Kumar, S. Tetrahedron 1985, 41,
209-214
(16) Voirin, B.; J ay, M.; Hauteville, M. Phytochemistry 1976, 15, 840-
841.
(17) Afifi, M. S. A.; Ahmed, M. M.; Pezzuto, J . M.; Kinghorn, A. D.
Phytochemistry 1993, 34, 839-841.
(18) Sherif, E. A.; Islam, A.; Krishnamurti, M. Ind. J . Chem. 1982, 21b,
(8) Antus, S.; Seligmann, O.; Wagner, H. Liebigs Ann. Chem. 1986, 647-
654.
(9) Stermitz, F. R.; Tawara-Matsuda, J .; Lorenz, P.; Mueller, P.; Ze-
newicz, L.; Lewis, K. J . Nat. Prod. 2000, 1146-1149.
(10) Stermitz, F. R.; Lorenz, P.; Tawara, J .; Zenewicz, L. A.; Lewis, K.
Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 1433-1437.
(11) Schrall, R.; Becker, H. Planta Med. 1977, 32, 27-32.
(12) Merlini, L.; Zanarotti, A.; Pelter, A.; Rochefort, M. P.; Ha¨nsel, R. J .
Chem. Soc., Perkin Trans. 1 1980, 775.
478-479.
(13) She, X.; Gu, W.; Wu, T.; Pan, X. J . Chem. Res. 1999, 100-102.
NP000166D