(±)-Diinsininone: made nature's way

Article abstract of DOI:10.1016/j.tet.2006.01.109

We report the synthesis of diinsininone (33), the aglycone of (±)-diinsinin (2). Thereby, we complete the first construction of a proanthocyanidin (PA) type-A compound incorporating a [3.3.1]-bicyclic ketal as its characteristic core. Our strategy utilizes a coupling between a benzopyrilium salt and a flavanone that proves applicable to other PA type-A compounds. During this undertaking, treatment of naringenin (9) with 2-iodoxybenzoic acid (IBX) followed by reductive work-up affords eriodictyol (10). This reactivity mirrors that of catechol hydroxylase (F3H) found in the flavonoid pathway. Other interesting transformations include the formation of flavonoids through an ortho-quinone methide (o-QM) cycloaddition-oxidation sequence and regioselective β-glycosidations of several unprotected flavanones suggesting a likely synthesis of 2 from the aglycone 33.

Full text of DOI:10.1016/j.tet.2006.01.109

Tetrahedron 62 (2006) 5298–5307
( )-Diinsininone: made nature’s way
*
Carolyn Selenski and Thomas R. R. Pettus
University of California, Santa Barbara, Department of Chemistry & Biochemistry, Santa Barbara, CA 93106, USA
Received 16 December 2005; revised 2 January 2006; accepted 3 January 2006
Available online 5 April 2006
Abstract—We report the synthesis of diinsininone (33), the aglycone of (ꢀ)-diinsinin (2). Thereby, we complete the first construction of
a proanthocyanidin (PA) type-A compound incorporating a [3.3.1]-bicyclic ketal as its characteristic core. Our strategy utilizes a coupling
between a benzopyrilium salt and a flavanone that proves applicable to other PA type-A compounds. During this undertaking, treatment
of naringenin (9) with 2-iodoxybenzoic acid (IBX) followed by reductive work-up affords eriodictyol (10). This reactivity mirrors that of
catechol hydroxylase (F3H) found in the flavonoid pathway. Other interesting transformations include the formation of flavonoids through
an ortho-quinone methide (o-QM) cycloaddition–oxidation sequence and regioselective b-glycosidations of several unprotected flavanones
suggesting a likely synthesis of 2 from the aglycone 33.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
these compounds in an efficient manner, it might fashion
a single flavan component using genomics, and then cause
this entity to dimerize, forming bonds i and ii, whereupon
oxidases and lipases would complete its creation. By mim-
icking this purported pathway, we hoped to reveal some of
nature’s more closely guarded secrets regarding chemo-
selectivity, diastereoselectivity and enantioselectivity during
the assembly process leading to 1 and 2.
Chemists should ponder probable biosynthetic pathways
leading to their target, because few things are more hum-
bling to one’s best-laid plans than nature’s simpler solution.
Nature has been conducting and optimizing chemical experi-
ments for millions of years; therefore, matching or exceed-
ing nature’s accumulated wisdom is a daunting challenge.
However, the precise sequence used by nature to assemble
its products is rarely known. In these instances, synthetic
chemists are empowered with the ability to illuminate
nature’s hidden pathways by determining the reactivity
inherent to its intermediates.
In 1996, the compounds 1 and 2 were discovered in the rhi-
zome of Sarcophyte piriei Hutch, a parasitic plant that grows
on the roots of the Acacia species of tropical shrubs and
trees.1 A decoction of its tuber is used in Somalian folk
medicine; an activity that led to their eventual isolation
and evaluation.2 Diinsininol (1) and diinsinin (2) inhibit
prostaglandin synthesis (IC50 values of 9.20 mM and
13.14 mM, respectively) and platelet activating factor-
induced exocytosis (IC50 values of 49 mM and 39 mM, re-
spectively). These biological activities are common among
drugs that inhibit cyclooxygenase-2 (COX-2) and protect
cells from tumor growth; tumors with high COX-2 levels
are usually larger and metastasize faster than those without.3
Consider the [3.3.1]-bicyclic compounds diinsininol (1) and
diinsinin (2) shown in Figure 1. In nature’s quest to build
OR
7
5
OH
HO
O
HO
R'
OH
ii
i
O
O
2
The blacked skeleton
reveals a dimeric
2. Biological relevance
5O
O
benzopyran backbone.
H
diinsininol (1): R= β-glucose, R'=OH
diinsinin (2): R= β-glucose, R'=H
The flavonoid carbon skeleton encompasses a range of struc-
tural and biological diversity and is widely distributed
throughout the plant kingdom.4 Because of its abundance,
it often proves useful as a starting material for synthesis.
Its structural diversity on the other hand enables evaluation
of an assortment of pharmacological targets. More than
5000 flavonoids have been isolated varying in the hydroxyl-
ation, methoxylation, glycosylation, and acylation patterns
Figure 1. PA type-A structures, diinsininol (1) and diinsinin (2).
Keywords: Proanthocyanidin; [3.3.1]-Bicyclic ketal; Benzopyrilium;
o-Quinone methide; Benzopyran; IBX; Catechol; Flavonoid.
* Corresponding author. Tel.: +1 805 637 5651; fax: +1 805 893 5690;
e-mail: pettus@chem.ucsb.edu
0040–4020/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2006.01.109

C. Selenski, T. R. R. Pettus / Tetrahedron 62 (2006) 5298–5307
5299
of the carbon backbone.5 This arsenal of structures evolved
as a defense to oxidative stress and hungry herbivores.6
betes.14 In 2004, several powerful antihyperglycemic com-
pounds were identified in cinnamon and determined to be
trimer and tetramers of PA type-A compounds displaying
the characteristic [3.3.1]-bicyclic ketal core.15
Flavonoids protect organisms from oxidative stress by de-
stroying reactive oxygen species that would have otherwise
caused cell damage. It has long been known that foods high
in antioxidants lower the risk of many diseases, including
cancer, cataract, heart disease, and rheumatoid arthritis.7
The so-called ‘French Paradox’ stems from the observation
that French people consuming as much as 1200 mg of flavo-
noids per liter of red wine with structures similar to 1 and 2
display lower incidences of coronary disease despite a higher
intake of saturated fat.8–12
3. Relevance of the speculative biosynthetic pathways
The biosyntheses of 1 and 2 initially follows the flavonoid
pathway, which proceeds from phenylalanine through
trans-cinnamate, chalcones, flavanones, anthocyanidins, to
flavans. It is among the best studied of biosynthetic pathways
in nature.16–18 The optically active compounds 1 and 2 sug-
gest a double union between two products emerging from the
flavonoid pathway. It is reasonable to speculate that an or-
ganism would utilize an identical path for these structurally
related natural products. Paths A and B have been proposed
(Scheme 1). In our minds, these proposed routes are pro-
foundly different and suggest different consequences for
reactivity and stereochemistry leading to the [3.3.1]-bicyclic
ketal. Path B, which has been shown to afford the singly
linked type-B PA structures, was proposed by Nonaka to
In addition to these suppositious medical benefits, com-
pounds 1 and 2 resemble antihyperglycemic components
found within cinnamon, which have been studied for their
ability to potentiate insulin, maintain normal blood sugar
levels, and assist in fat metabolism.13 Currently insulin resis-
tance, which is referred to as metabolic syndrome X, affects
50 million people. Syndrome X is linked to an unhealthy
lifestyle, genetic factors, and the progression to type-II dia-
Phenylpropanoid Pathway
OH
OH
+
NH3
H
PAL
C4H
4CL
CoA-S
-O2C
L-phenylalanine
-O2C
trans-cinnamate
-O2C
O
p-coumarate
coenzyme-A ester
CHS
OH
OH
OH
OH
OH
CoA-S
O
HO
O
O
HO
O
O
HO
OH
O
O
CHI
F3H
OH
OH
OH
O
O
FR
chalcone
eriodictyol
flavanone
polyketide
DFR
OH
R
OH
R
OH
OH
R
HO
O
HO
O
HO
O
O
O
R
ANS
ANR
OH OH
flavanol
OH
OH
flavan
OH
anthocyanidin
p-quinone methide
Path A
diastereoselective
regioselective
Path B
R
OH
O
R
OH
HO
O
oxidation
HO
O
R
HO
O
O
OH
OH
HO
OH
O
OH
O
OH
HO
R
R
OH
O
OH
R
X
X
X
OH
stereochemical equilibration;
FGT, UDP-glucose
HO
PA Type-B structures
singly linked
proanthocyanidin
stereochem.
equilibrates
PA Type-A structures
diinsininol (1)
+
diinsinin (2)
doubly linked
proanthocyanidin
Scheme 1. Known (solid arrows) and speculative (dashed arrows) biosynthetic pathways leading to 1 and 2. PAL: phenylalanine ammonia lyase; C4H:
cinnamate 4-hydroxylase; 4CL: 4-coumarate: CoA ligase; CHS: chalcone synthase; CHI: chalcone flavanone isomerase; F3H: flavanone 3-hydroxylase;
FR: flavanone reductase; DFR: dihydroxyflavanone reductase; ANS: anthocyanidin synthase; ANR: anthocyanidin reductase.

5300
C. Selenski, T. R. R. Pettus / Tetrahedron 62 (2006) 5298–5307
yield the [3.3.1]-bicyclic ketal structure found in type-A PA
compounds by an oxidative cyclization. A catechin type-B
dimer has been transformed into the corresponding type-A
structure with picrylhydrazyl radical and hydrogen perox-
ide.19 Path B appears to entail the chemoselective addition
of a flavanone or flavan nucleophile to an electrophile arising
from the para-quinone methide or its corresponding cation.
This initial coupling should be inherently diastereoselective.
However, if a Path B-oxidation sequence does result in the
biosyntheses of 1 and 2, it is remarkable because the original
stereocenters ultimately become compromised when carried
to the optically active natural products. Path A, on the other
hand, suggests that a benzopyrylium salt can undergo a net
[3+3]-cycloaddition in an enantioselective fashion chaper-
oned by an unknown enzyme,20 or perhaps proceeds in a
diastereoselective fashion with the directing stereochemistry
again compromised when forming the final natural products.
is available regarding the likelihood of a successful Path A
coupling. However, Jurd reports a simple benzopyrylium
salt undergoes coupling with phloroglucinol to result in
a similar [3.3.1]-bicyclic ketal in 25% yield.27
6. Our synthetic studies
We began surveying these problems by constructing the
flavans, flavanones, and dihydroxyflavanones, which might
arise in the putative biosynthetic pathway, and then pro-
ceeded to investigate the inherent ability of these structures
to form type-A and B PA adducts (Scheme 2). Recently, our
group had shown that o-QMs generated from o-Boc salicyl-
alcohols would undergo cycloadditions with electron-rich
olefins forming chroman adducts that resemble compound
5.22 However, our original two-pot process, involving reduc-
tion and coupling proved unfruitful with aldehyde 3. This
substrate mandates the use of an in situ reduction–cyclo-
addition protocol. The tris-o-Boc salicylaldehyde 3 (0.10 M
The oxidation state of the components at the time of cou-
pling remains unclear as does the timing of the oxidation
that distinguishes between the two natural products. Flava-
none 3-hydroxylase (F3H) may catalyze hydroxylation of
(2S)-flavanone to eriodictyol. A combination of di-hydrox-
ylated and mono-hydroxylated materials leads to diinsini-
none (33), while a combination of di-hydroxylated systems
provides diinsininolone (34). Alternatively, F3H or some de-
rivative of this enzyme may establish divergence in oxida-
tion states (catechol vs phenol) in the final or penultimate
step of the biosynthesis. Similarly, glycosylation may occur
as the final or penultimate step of the biosynthesis. Diver-
gence among the pathways for similar components seems
energetically wasteful for an organism.
ꢂ
diethyl ether, ꢁ78 C), styrene 4 (40 equiv), and magnesium
bromide dietherate (1.0 equiv) are subjected to lithium
aluminum hydride (1.0 equiv) and thereby produce the
flavan 5 in a 45% isolated yield.
Ot-Bu
LiAlH4,
BocO
OBoc
BocO
O
MgBr2•OEt2,
45%
4
CHO
Ot-Bu
3
5
OBoc
OBoc
4
DDQ (2 equiv),
H2O, CH2Cl2,
88%
OH
ZnBr2,
HO
O
0.01M CH2Cl2,
75%
4. Synthetic relevance
flavan 6
OH
Ot-Bu
To the best of our knowledge, there are no previous synthetic
endeavors addressing diinsininol (1), diinsinin (2), or similar
type-A PA compounds in the literature. While there are
numerous methods for the construction of chromans, the
enantioselective construction of flavonoids is a surprisingly
difficult task and it is difficult to improve upon their biosyn-
thetic pathway.21 Recently, our group devised a strategy for
the enantioselective assembly of non-natural flavonoids by
a [4+2]-cycloaddition between an o-QM and a chiral enol
ether. We utilized this process in the first total synthesis of
the flavonoid (+)-mimosifoliol.22 We initially envisioned ex-
tending this strategy to the chroman functionality embedded
within diinsininol (1) and diinsinin (2). The additional
hydroxyl residue that distinguishes 1 from 2 could be intro-
duced using our IBX oxidation–reduction sequence for the
construction of o-quinols from phenols.28
K2CO3, MeI,
acetone,
90%
BocO
O
O
OMe
MeO
O
BocO
ZnBr2,
0.01M CH2Cl2,
82%
8
7
OMe
OH
OH
OH
HO
O
HO
O
IBX, DMSO;
Na2S2O4,
80%
OH
O
OH
O
eriodictyol (10)
flavanone 9
Ag2CO3,
quinoline, 11
R
OH
AcO
AcO
AcO
O
O
O
AcO
O
OAc
AcO
5. Related path A and B synthetic endeavors
AcO
12: R=OH, 72%
13: R=H, 85%
AcO
11
Br
OH
O
Previous syntheses of the singly linked PA type-B structures
have followed the putative biosynthetic route, Path B.23–26
However, unless other stereochemical substituents adjoin
the site of reactivity and direct the process Path B couplings
appear to result in a complex diastereomeric mixture of
products. Moreover, regioselectivity and chemoselectivity
prove problematic for most systems. Far less information
Scheme 2. Synthesis of flavan, flavanone, and catechol.
C(4)-oxidation of flavan 5 (0.10 M methylene chloride)
occurs with 2,3-dichloro-5,6-dicyano-1,4-benzoquinone
(2 equiv) in the presence of water (1.0 equiv) to afford the
flavanone 8 in 88% yield. Cleavage of the t-butyl residues

C. Selenski, T. R. R. Pettus / Tetrahedron 62 (2006) 5298–5307
5301
OR2
found in compounds 5 and 8 (0.01 M methylene chloride) is
accomplished with zinc bromide (10 equiv) to afford the fla-
van 6 and the flavanone 9, respectively. Per-methylation of 6
(0.10 M acetone) using methyl iodide (5.0 equiv) and potas-
sium carbonate (4.0 equiv) affords tris-methylated flavan
ether 7 and facilitates its structural identification.
K2CO3, MeI, acetone, 65%
R1O
O
O
or
9
i. K2CO3, MeI, acetone
ii. TBSOTf, DMF, imid.
iii. Boc2O, Hunig's base,
CH2Cl2, 75%
OR3
14: R1=R2=R3=–Me
15: R1=Me, R2= –TBS,
R3=–Boc
Next, we used our one-pot two-step chemical process that
mimics the enzyme F3H found in the flavan biosynthesis.28
The unprotected flavanone 9 (0.03 M dimethyl sulfoxide) is
subjected to 2-iodoxybenzoic acid (2.0 equiv) and forms
a highly reactive o-quinone intermediate, which is reduced
during work-up with sodium sulfite (3.0 equiv) to provide
eriodictyol (10) in an isolated 80% yield. As previously ob-
served, phenols bearing electron-withdrawing groups such
as –C(O)R, –CHO, and –NO2 fail to undergo oxidation
with 2-iodoxybenzoic acid.28 This reaction can be viewed
as electrophilic aromatic substitution.29 The –C(O)R substit-
uent of flavanone 9 deactivates the A-ring and effectively
prevents its oxidation. With the phenol 9 and catechol 10
in hand, methods for their regioselective glycosidation
were explored. The insolubility of substrates 9 and 10 se-
verely limited the conditions that could be employed. After
some experimentation, we found that phenol 9 and catechol
10 (0.10 M quinoline) undergo regioselective b-glycosida-
tion with acetobromo-a-D-glucose 11 (1.50 equiv) and silver
carbonate (1.50 equiv) to produce compounds 13 and 12 in
good yield.30 The following order of reactivity was observed
for flavanone 9 and appears to follow the relative acidity of
each phenol: C(5)–OH<c(40)–OH<c(7)–OH. Given this
result, we expect the glycosidation of the aglycones of 1 and
2 to be fairly straightforward.
Li
OTBS
BnO
OBn
17
MeO
O
OBn
PhMgBr; LiBH4
THF, 0 °C,
72%
THF, –78 °C,
52% (3:2 ratio)
OH
[o-QM]
16
OMe
OMe
MeO
O
MeO
O
EtOMgBr,
benzene,
6 h
19
OBn
18
OBn
OMe
BnO
OMe
BnO
OBn
OBn
a) debenzylation
b) ketalization
[3.3.1]-ketal
Scheme 3. o-QM reaction with flavanone backbone.
desired [3.3.1]-bicyclic system. However, treatment of the
debenzylated adduct with strong acid results in loss of the
phloroglucinol fragment from adduct 19.
Next, we turned our attention to the formation of the C(4)–
C(8) bond following a putative Path B biosynthesis. Given
our experiences with diastereoselective additions to o-
QMs, we decided to explore this reaction within the context
of a flavanone (Scheme 3). The glycosidation reaction re-
vealed that the hydroxyl residues of flavanones would also
undergo regioselective alkylation. We decided to exploit
this inherent reactivity for selective protection. Sequential
treatment of flavanone 9 with methyl iodide, tert-butyldi-
methylsilyl triflate, and di-tert-butyl dicarbonate and the
appropriate bases affords compound 15 in good yield (75%).
With compound (ꢀ)–7 in hand, we considered other strate-
gies mimicking Path B (Scheme 4).32 Treatment of the flavan
(ꢀ)–7 (0.10 M methylene chloride) with 2,3-dichloro-5,6-
dicyano-1,4-benzoquinone (0.5 equiv) affords a mixture of
more than four stereoisomers of the dimer 22 in a combined
52% yield. Given this result, we expected optically pure 7 to
give more than two diastereomers; there must be atropiso-
merism among the trans and/or cis adducts. To partially
circumvent this stereochemical problem, we examined the
oxidation of 7 (0.1 M methylene chloride) with 2,3-
dichloro-5,6-dicyano-1,4-benzoquinone (1.0 equiv) and in
Reaction of the o-OBoc flavanone 15 (0.05 M tetrahydro-
furan) with phenylmagnesium bromide (1.0 equiv) followed
by lithium borohydride (1.0 equiv) results in flavan 16 for-
mation in a 52% yield and a 3:2 ratio of diastereomers. We
presume that the reaction proceeds through an o-QM inter-
mediate. The lack of stereoselectivity was somewhat trou-
bling and did not bode well for applying any of these Path
B strategies to the syntheses of 1 and 2. Therefore, we sought
another method for carbon–carbon bond formation. The fla-
vanone 9 was per-methylated to afford compound 1ꢂ4. Treat-
ment of flavanone 14 (0.2 M tetrahydrofuran, ꢁ78 C) with
the phloroglucinol lithium 17 (3.0 equiv) affords the chro-
mene 18 by spontaneous elimination of the tertiary alcohol
intermediate. Isomerization of thꢂe double bond within com-
pound 18 (0.07 M benzene, 65 C) occurs upon treatment
with ethoxymagnesium bromide (2.0 equiv) to form the
enol ether 19.31 We envisioned that debenzylation of com-
pound 19 and subsequent ketalization would afford the
OMe
DDQ, CH2Cl2, 0 °C
MeO
O
7
BnO
OBn
DDQ (0.5
equiv),
3 equiv
20
60%: 3:2 ratio
OBn
OMe
BnO
OBn
CH2Cl2,
0 °C, 52%
OMe
21
OBn
MeO
O
22
OMe
OMe
MeO
O
OMe
Scheme 4. DDQ dimerization and coupling.

5302
C. Selenski, T. R. R. Pettus / Tetrahedron 62 (2006) 5298–5307
the presence of tribenzylated phloroglucinol (20) (3 equiv).
The reaction proceeds in a 60% yield and affords the type-
B PA structure 21 in a 3:2 ratio favoring the trans diastereo-
mer. In addition to the stereochemistry issues raised by this
strategy, there are numerous protecting group issues that
would have to be resolved as well. Therefore, we began to
investigate a Path A inspired synthesis.
this extended reaction as a single diastereomer. However,
the similar conditions proved entirely unproductive for the
combination of the benzopyrilium 27 with the hydroxylated
flavanone 10.
OH
OH
OH
A Path A strategy requires rapid and easy access to the cor-
responding benzopyrylium salts. We initially considered
methods to construct these systems from 7, 9, and 10. How-
ever, careful scrutiny of the literature reveals oxidative
methods are plagued by low yields, crude mixtures.33 In-
stead, we opted to explore a new protocol recently reported
by Mas (Scheme 5).34 2,4,6-Triacetoxybenzaldehyde (24)
undergoes condensation with hydroxyacetophenones 25
and 26 in a saturated hydrochloric acid solution in methanol
to result in the benzopyrylium salts 27 and 28 in good yields.
HO
O
O
O
HO
6 (6 equiv),
32
150 °C, 1 hr,
45%
OH
OH
OH
OH
O
HO
HO
( )-9 (6 equiv),
27
O
O
O
OAc
OH
120 °C, 20 hr,
32%
K2CO3, Ac2O,
Et2O
diinsininone (33)
OH
OH
OAc
AcO
HO
OH
24
23
H
O
O
H
OH
(7 equiv)
R
( )-10 (6 equiv),
120 °C, 20 hr
HCl (g),
MeOH.
15 min
OH
O
O
HO
O
25: R=H
26: R=OH
HO
HO
OH
OH
OH
O
Cl-
R
diinsininolone (34)
O
HO
HO
HO
O
O
O
MeOH,
pH 5.8, 2 h,
150 o
OH
OH
Scheme 6. Reactions of the benzopyrilium salt 27 with flavan and flavanone.
C
OH
HO
OH
(3 equiv)
29
R
The formation of a single diastereomer of diinsininone (33)
is explained by the equilibria shown in Scheme 7. It should
be noted that we had originally imagined degrading naringin
(35) to afford optically active (S)-C(2) flavanone 9. How-
ever, in our hands, hydrolysis of the sugar results in
30: R=OH, 80%
OH 31: R=H, 83%
27: R=OH, 65%
28: R=H, 70%
OH
Scheme 5. Mas’s procedure provides benzopyrylium salts that readily
undergo addition with phloroglucinol.
We find that addition of an aqueous buffered solution of
phloroglucinol 29 (3 equiv) to the respective benzopyry-
liums 27 and 28 (0.20 M methanol) affords the ketals 30
and 31, respectively, in outstanding yields. Optimal condi-
tionsꢂ employ a Emrys microwave reactor with heating to
150 C for 120 min. Isolation of non-ketalized products
from reactions conducted at lower temperatures reveals
that the carbon–carbon bond formation precedes the oxy-
gen–carbon bond formation. While we entertained the possi-
bility of desymmetrizing the meso compounds 30 and 31, we
first examined the addition of flavans and flavanones to the
benzopyrilium salts 27 and 28.
OH
HO
O–
(E)
OH
O
B-
OH
OH
O
HO
HO
O
O
(S)
HCl, MeOH,
70%
O
( )-9
O
HO
OH
O
HO
OH
naringin (35)
The benzopyrilium salt 27 (0.10 M methanol) combines
with flavan 6 (6 equiv) using the previously described micro-
wave equipment and results in the regioselective formation
of the type-A PA compound 32 in 42% yield (Scheme 6).
However, a 1:1 mixture of diastereomers is observed.
H+
H+
OH
O
H
O
H
OH
ꢂ
(Z)
Upon lowering the temperature to 120 C and lengthening
the reaction time to 20 h for the subsequent combination
of the benzopyrilium 27 with the flavanone 9, diinsininone
(33) forms. To our delight, compound 33 emerges from
O
O
Scheme 7. Acid and base-catalyzed racemization of a C(2)-stereocenter.

C. Selenski, T. R. R. Pettus / Tetrahedron 62 (2006) 5298–5307
5303
ꢂ
formation of racemic 9 as determined by a Chiralcel OD-H
column.35
to ꢁ78 C and LiAlH4 (4 mg, 1.00 equiv) was added. The
reaction was warmed to rt over 3 h then quenched with
1 M NaHCO3 and extracted with Et2O. The ether layer
was washed with brine, dried over Na2SO4, filtered, and con-
centrated in vacuo. The crude mixture was then chromato-
graphed through silica gel, eluting with petroleum ether/
EtOAc (95:5) to afford the desired product in 45% yield.
1H NMR (400 MHz, CDCl3): d 7.08 (d, 2H, J¼9 Hz), 6.70
(d, 2H, J¼9 Hz), 6.39 (s, 1H), 6.34 (s, 1H), 5.20–4.90 (m,
1H), 2.60–2.20 (m, 4H), 1.45 (s, 9H), 1.40 (s, 18H).
7. Conclusions
One day, a better understanding of the PA biosynthetic path-
way may allow the genetic engineering of plants for the pro-
duction of compounds coveted for their antioxidant and
antihyperglycemic activities. Until that day arrives, chemi-
cal synthesis remains the best method for the construction
of these compounds and enables the exploration of their
biological activities. The preceding synthetic studies have
illuminated inherent reactivity of several intermediates
belonging to the flavonoid pathway. These studies suggest
that compounds 1 and 2 most likely arise from an identical
biosynthetic pathway proceeding through a Path A benzo-
pyrilium coupling. Whether the oxidative divergence among
these structures arise as the ultimate or penultimate step re-
mains to be determined. Related natural products belonging
to the Dragon’s Blood family of flavanones, a name that de-
rives from the red resin plant exudates, can also be accessed
by appropriately adapting the above strategy (Fig. 2).32b
8.3. Flavan 6
A flame-dried flask equipped with stir-bar and nitrogen line,
was charged with flavan 5 (20.0 mg, 0.01 M CH2Cl2) and
ZnBr2 (87.5 mg, 10 equiv). The solution was stirred at rt
for 6 h. The reaction was quenched with 1 M HCl and
extracted twice with EtOAc. The ether layer was washed
with brine, dried over Na2SO4, filtered, and concentrated
in vacuo. The crude mixture was then chromatographed
through silica gel, eluting with petroleum ether/EtOAc
(50:50) to afford the desired product in 75% yield. 1H
NMR (400 MHz, CD6CO): d 7.27 (d, 2H, J¼9 Hz), 6.85
(d, 2H, J¼9 Hz), 6.00 (s, 1H), 5.87 (s, 1H), 4.89–4.86 (m,
1H), 2.72–2.66 (m, 2H), 2.11–1.88 (m, 2H).
OR1
OR1
8.4. Tris-methylated chroman 7
R2O
O
R2O
O
O
O
O
O
R3
R3
A round-bottom flask equipped with a stir-bar was charged
with flavan 6 (1.67 g, 0.2 M acetone), K2CO3 (340 mg,
4 equiv), and MeI (0.19 mL, 5 equiv). The reaction was
stirred overnight and filtered. The reaction was diluted
with H2O and extracted twice with EtOAc. The organic layer
was washed with brine, dried over MgSO4, filtered, and con-
centrated in vacuo. The crude mixture was then chromato-
graphed through silica gel, eluting with petroleum ether/
EtOAc (80:20) to afford the desired product in 90% yield.
1H NMR (400 MHz, CDCl3): d 7.36 (d, 2H, J¼9 Hz), 6.93
(d, 2H, J¼9 Hz), 6.13 (s, 1H), 6.09 (s, 1H), 4.93–4.92
(m, 1H), 3.83 (s, 3H), 3.80 (s, 3H), 3.77 (s, 3H), 2.80–2.60
(m, 4H), 2.20–1.97 (m, 4H).
OMe
OMe
36: R1=H;R2=Me;R3=OH
37: R1=Me;R2=Me;R3=OH
38: R1=Me;R2=Me;R3=OAc
39: R1=H;R2=Me;R3=H
40: R1=Me;R2=Me;R3=H
41: R1=Ac;R2=Me;R3=H
42: R1=R2=R3=H
43: R1=R2=R3=Me
Figure 2. Natural products of the Dragon’s Blood family.
8. Experimental
8.1. Aldehyde 3
8.5. Flavanone 8
A flame-dried flask equipped with a stir-bar and a nitrogen
line containing 2,4,6-trihydroxybenzaldehyde (2 g, dehy-
drated with P2O5, 0.05 M CH2Cl2), Hunig’s Base (1.36 mL,
0.6 equiv), DMAP (four crystals), and Boc2O (7.74 g,
3.2 equiv) was stirred overnight. The reaction was quenched
with 1 M NH4Cl and extracted twice with CH2Cl2. The
organic layer was washed with water, brine, dried over
MgSO4, and concentrated in vacuo. The crude mixture was
then chromatographed on silica gel, eluting with petroleum
ether/EtOAc (95:5) to afford the desired product in 75%
Flavan 5 (30 mg, 1 equiv, 0.01 M CH2Cl2), H2O (1 mL,
1 equiv), and DDQ (recrystallized from hot CHCl3,
26.5 mg, 2 equiv) were added to a round-bottom flask equip-
ped with a stir-bar, a septum and stirred under N2 (g) at rt
overnight. The reaction was diluted with H2O and extracted
twice with EtOAc. The organic layer was washed with brine,
dried over MgSO4, filtered, and concentrated in vacuo. The
crude mixture was then chromatographed through silica gel,
eluting with petroleum ether/EtOAc (70:30) to afford the
desired product in 88% yield. 1H NMR (400 MHz,
CDCl3): d 7.08 (d, 2H, J¼9 Hz), 6.70 (d, 2H, J¼9 Hz),
6.54 (s, 1H), 6.48 (s, 1H), 5.55–5.50 (m, 1H), 3.38–3.13
(m, 2H), 1.56 (s, 18H), 1.40 (s, 9H).
1
yield. H NMR (400 MHz, CDCl3): d 10.24 (s, 1H), 6.88
(s, 2H), 1.46 (s, 18H), 1.42 (s, 9H).
8.2. Tris-protected flavan 5
8.6. 40,5,7-Trihydroxyflavanone (9)
A flame-dried flask equipped with stir-bar and nitrogen
line, was charged with aldehyde 3 (50 mg, 0.1 M Et2O),
1-tert-butoxy-4-vinylbenzene (0.19 mL, 40 equiv), and
MgBr2$OEt2 (28.4 mg, 1.00 equiv). The solution was cooled
A flame-dried flask equipped with stir-bar and nitrogen line,
was charged with flavanone 8 (20.0 mg, 0.01 M CH2Cl2) and
ZnBr2 (85.2 mg, 10 equiv). The solution was stirred at rt for

5304
C. Selenski, T. R. R. Pettus / Tetrahedron 62 (2006) 5298–5307
6 h. The reaction was quenched with 1 M HCl and extracted
twice with EtOAc. The ether layer was washed with brine,
dried over Na2SO4, filtered, and concentrated in vacuo.
The crude mixture was then chromatographed through silica
gel, eluting with petroleum ether/EtOAc (50:50) to afford
the desired product in 82% yield. 1H NMR (400 MHz,
CD6CO): d 12.19 (s, 1H), 9.59 (s, 1H), 8.55 (s, 1H), 7.39
(d, 2H, J¼9 Hz), 6.90 (d, 2H, J¼9 Hz), 5.95 (s, 1H), 5.94
(s, 1H), 5.48–5.44 (m, 1H), 3.22–3.15 (m, 2H), 2.75–2.70
(m, 2H); 13C NMR (100 MHz, CD6CO): d 197.4, 167.3,
165.4, 164.5, 158.8, 130.9, 129.1, 116.3, 103.3, 96.8, 95.9,
80.0, 43.6, 30.5; HRMS (ESI-TOF) (M++H+) m/z calcd for
C15H12O5 273.0685, found 273.0765.
8.9. Naringenin glucoside 1330
A solution of flavanone 9 (9.4 mg, 0.14 M quinoline),
bromide 11 (21 mg, 1.50 equiv), and Ag2CO3 (13.8 mg,
1.5 equiv) was stirred for 6 h at rt. The reaction was poured
into methanol, filtered through a short pad of Celite, and
evaporated in vacuo. The residue was dissolved into EtOAc,
washed with 1 N HCl, washed with brine, and dried over
MgSO4. After evaporation, the resulting crude product was
purified by flash chromatography eluting with hexane/
EtOAc (1:1) to afford the desired product in 85% yield.
The product was an inseparable mixture of diastereomers.
1H NMR (400 MHz, DMSO): d 12.03 (s, 1H), 9.65 (s, 1H),
7.31 (d, 2H, J¼9 Hz), 6.79 (d, 2H, J¼9 Hz), 6.16–6.09
(m, 2H), 5.67–5.63 (m, 1H), 5.53–5.50 (m, 1H), 5.37–5.34
(m, 1H), 5.06 (m, 2H), 4.30–4.26 (m, 1H), 4.15–4.05
(m, 2H), 3.40–3.33 (m, 1H), 2.76–2.70 (m, 1H), 2.00–1.95
(m, 12H); (ESI-TOF) (M++Na) m/z calcd for C29H30O14
625.1636, found 625.1528.
8.7. Eriodictyol (10)28
To a solution of flavanone 9 (1.011 g, 0.03 M DMSO) was
added IBX36 (2.135 g, 2 equiv) and stirred at rt for 5 h. As
the reaction progressed, the color changed from a translucent
yellow solution to an opaque brown solution. Na2S2O4
(1.94 g, 3 equiv) was added to the reaction and stirred for
an additional 4 h. The reaction was quenched with water
(300 mL) and extracted with EtOAc (3ꢃ100 mL). The com-
bined organic extracts stirred with pH 8.5 buffer solution to
remove o-iodobenzoic acid. The organic layer was washed
with brine, dried with MgSO4, filtered, and concentrated.
The crude mixture was then chromatographed through silica
gel, eluting with petroleum ether/EtOAc (50:50) to afford
the desired product in 80% yield. 1H NMR (400 MHz,
CD6CO): d 7.03 (s, 1H), 6.87 (s, 2H), 5.96–5.94 (m, 2H),
5.41–5.38 (m, 1H), 3.18 (m, 1H), 2.75 (m, 1H); 13C NMR
(100 MHz, CD6CO): d 167.7, 137.8, 135.7, 134.8, 116.8,
116.5, 102.0, 89.7, 86.5, 85.2, 73.6, 67.2, 66.3, 50.4, 14.0;
HRMS (ESI-TOF) (M++Na+) m/z calcd for C15H12O6
311.0633, found 311.0525.
8.10. Trimethylated naringenin 14
Naringenin (9, 1 g, 0.2 M acetone), K2CO3 (3.05 g, 6 equiv),
and MeI (1.83 mL, 8 equiv) were stirred overnight. The re-
action was filtered and concentrated in vacuo. The resulting
crude product was purified by flash chromatography eluting
with petroleum ether/EtOAc (70:30) to afford the desired
1
product in 65% yield. H NMR (400 MHz, CDCl3): d 7.47
(d, 2H, J¼9 Hz), 6.98 (d, 2H, 9 Hz), 6.18 (s, 1H), 6.16
(s, 1H), 5.45–5.40 (m, 1H), 3.85 (s, 3H), 3.82 (s, 3H), 3.81
(s, 3H), 3.03–2.85 (m, 2H), 2.64–2.58 (m, 2H).
8.11. Tris-protected naringenin 15
Naringenin (9, 162 mg, 0.1 M DMF), Na2CO3 (69 mg,
1.10 equiv), and MeI (0.4 mL, 1.10 equiv) were stirred at
rt overnight. The reaction was quenched with H2O and ex-
tracted with EtOAc. The organic layer was washed repeat-
edly with H2O, washed with brine, and dried over MgSO4.
The resulting crude product was purified by flash chroma-
tography (petroleum ether/EtOAc 50:50) to afford mono-
protected naringenin. 1H NMR (400 MHz, CD6CO):
d 12.15 (s, 1H), 8.57 (s, 1H), 7.40 (d, 2H, J¼8 Hz), 6.89
(d, 2H, J¼8 Hz), 6.04 (s, 1H), 6.03 (s, 1H), 5.39–5.35 (m,
1H), 3.82 (s, 3H), 3.10–3.03 (m, 2H), 2.78–2.70 (m, 2H).
mono-Protected narigenin (66 mg, 0.1 M THF), 2,6-lutidine
(0.028 mL, 1.05 equiv), and TBSOTf (0.056 mL,
1.05 equiv) were stirred at rt for 4 h. The reaction was
quenched with H2O, extracted with EtOAc, and dried with
Na2SO4. The resulting crude product was purified by flash
chromatography (petroleum ether/EtOAc 80:20) to afford
bis-protected naringenin. 1H NMR (400 MHz, CD6CO):
d 12.15 (s, 1H), 7.47 (d, 2H, J¼8 Hz), 6.96 (d, 2H,
J¼8 Hz), 6.07 (s, 1H), 6.05 (s, 1H), 5.55–5.51 (m, 1H),
3.86 (s, 3H), 3.27–3.19 (m, 2H), 2.81–2.77 (m, 2H), 1.00
(s, 9H), 0.33 (s, 6H). The bis-protected compound (0.1 M
CH2Cl2), DMAP (one crystal), Hunig’s Base (0.037 mL,
1.0 equiv), and Boc2O (100 mg, 1.0 equiv) were stirred at
rt for 1 h. The reaction was quenched with H2O, extracted
with EtOAc, and dried with Na2SO4. The resulting crude
product was purified by flash chromatography (petroleum
ether/EtOAc 90:10) to afford tris-protected naringenin 15
8.8. Acetobromo-a-D-glucose (11)
D-Glucose (2.05 g, 0.40 M pyridine) and Ac2O (12 mL,
0.44 M) were stirred at rt overnight. The reaction was
quenched with H2O, extracted with CDCl3, washed with
NaHCO3, dried with Na2SO4, and afforded pentaacetyl-
glucose. 1H NMR (400 MHz, CDCl3): d 5.72–5.71 (m,
1H), 5.28–5.24 (m, 1H), 5.16–5.11 (m, 2H), 4.32–4.28 (m,
1H), 4.13–4.10 (m, 1H), 3.86 (m, 1H), 2.12 (s, 3H), 2.09
(s, 3H), 2.04 (s, 3H), 2.03 (s, 3H), 2.02 (s, 3H); 13C NMR
(100 MHz, CDCl3): d 170.8, 170.3, 169.6, 169.5, 169.2,
91.9, 73.0, 72.9, 70.4, 67.9, 61.6, 21.0, 20.9, 20.8, 20.7;
HRMS (ESI-TOF) (M++Na+) m/z calcd for C16H22O11
413.1162, found 413.1057. When the pentaacetoglucose
(2 g, 0.57 M acetic acid) was dissolved, HBr (0.64 mL,
48% in acetic acid) was added drop-wise. The reaction
was stirred overnight, poured over an ice/H2O slurry,
extracted with CHCl3, neutralized with NaHCO3, washed
with H2O, dried with Na2SO4, and afforded acetobromo-a-
1
D-glucose (11). H NMR (400 MHz, CDCl3): d 6.62–6.61
(m, 1H), 5.59–5.55 (m, 1H), 5.19–5.15 (m, 1H), 4.86–4.83
(m, 1H), 4.36–4.30 (m, 2H), 4.15–4.12 (m, 1H), 2.11 (s,
3H), 2.10 (s, 3H), 2.06 (s, 3H), 2.05 (s, 3H); 13C NMR
(100 MHz, CDCl3): d 170.7, 170.1, 170.0, 169.7, 86.8,
72.3, 70.8, 70.4, 67.4, 61.1, 20.9, 20.9, 20.8, 20.7; HRMS
(ESI-TOF) (M++Na+) m/z calcd for C14H19O9Br 433.0212,
found 433.0126.
1
in 75% yield. H NMR (400 MHz, CD6CO): d 7.50–7.48

C. Selenski, T. R. R. Pettus / Tetrahedron 62 (2006) 5298–5307
5305
(d, 2H, J¼8 Hz), 6.98–6.95 (d, 2H, J¼8 Hz), 6.50 (s, 1H),
6.37 (s, 1H), 5.52–5.49 (s, 1H), 3.90 (s, 3H), 3.08–3.05
(m, 2H), 2.85 (m, 2H), 1.54 (s, 9H), 1.01 (s, 9H), 0.25
(s, 6H).
The reaction was stirred for 3 h, quenched with H2O, ex-
tracted with EtOAc, the organic layer was washed with
brine, dried over MgSO4, filtered, and concentrated in vacuo.
The crude mixture was then chromatographed through silica
gel, eluting with petroleum ether/EtOAc (70:30) to afford
the desired product in 60% yield.
8.12. Compound 16
A flame-dried flask equipped with stir-bar and nitrogen line,
was charged with flavanone 15 (8.0 mg, 1 equiv, 0.05 M
8.16. Dimer 22
ꢂ
THF) and cooled to ꢁ78 C. Phenylmagnesium bromide
To a flame-dried flask equipped with a stir-bar and a nitrogen
line containing flavan 7 (30 mg, 0.10 M CH2Cl2) stirring at rt
was added DDQ (recrystallized from hot CHCl3, 11.2 mg,
0.5 equiv). The reaction was stirred for 3 h, quenched with
H2O, extracted with EtOAc. The organic layer was washed
with brine, dried over MgSO4, filtered, and concentrated
in vacuo. The crude mixture was then chromatographed
through silica gel, eluting with petroleum ether/EtOAc
(40:60) to afford the desired product in 52% yield.
(8.0 mL, 1.50 equiv, 3 M THF) was added drop-wise and
stirred for 1 h. LiBH4 (1.0 mg, 1.00 equiv) was added to
the solution, warmed to rt, quenched with H2O, and ex-
tracted with EtOAc. The ether layer was washed with brine,
dried over Na2SO4, filtered, and concentrated in vacuo. The
crude mixture was then chromatographed through silica gel,
eluting with petroleum ether/EtOAc (90:10) to afford the
desired product in 52% yield.
8.13. Chromene 18
8.17. 2,4,6-Triacetoxybenzaldehyde (24)
Bromotribenzylated phloroglucinol (65 mg, 0.3 M THF,
ꢂ
A flame-dried flask equipped with a stir-bar and a nitrogen
line containing phenol 23, dehydrated with P2O5, (2.58 g,
1 equiv, 0.15 M Et2O), K2CO3 (8.41 g, 3.6 equiv), and
Ac2O (16.8 mL, 11 equiv) was stirred overnight. The reac-
tion was filtered and concentrated to afford the desired
product in 80% yield. 1H NMR (400 MHz, CDCl3):
d 10.13 (s, 1H), 6.97 (s, 2H), 2.38 (s, 6H), 2.31 (s, 3H);
13C NMR (100 MHz, CDCl3): d 185.6, 168.7, 167.9,
155.2, 1530.0, 118.3, 114.9, 21.4, 21.0; d IR (CH2Cl2, ymax
cmꢁ1) 1778, 1697, 1616, 1580, 1433, 1370, 1298, 1180,
1122, 1051, 1022; HRMS (ESI-TOF) (M++Na) m/z calcd
for C13H12O7 303.2302, found 303.0484.
3 equiv) was cooled to ꢁ78 C, n-BuLi (0.052 mL, 2.5 M
THF, 3.00 equiv) was added drop-wise, and the reaction
stirred until complete by TLC (eluting with petroleum
ether/EtOAc 90:10). This solution was added drop-wise to
a stirꢂring solution of flavan 15 (13.7 mg, 0.2 M THF) at
ꢁ78 C. After 1 h, the reaction was quenched with H2O, ex-
tracted EtOAc, washed with brine, and dried with MgSO4.
The reaction was purified by flash chromatographed with
EtOAc/hexane (70:30) to afford the desired product in
1
72% yield. H NMR (400 MHz, CDCl3): d 7.53–7.22 (m,
17H), 6.61–6.59 (m, 2H), 6.35–6.30 (m, 2H), 6.14–6.12
(m, 1H), 5.92–5.91 (m, 1H), 4.91 (d, 1H, J¼5 Hz), 4.76
(d, 1H, J¼5 Hz), 5.05–4.96 (m, 6H), 3.73 (s, 3H), 3.72 (s,
3H), 3.30 (s, 3H); HRMS (ESI-TOF) (MꢁH)+ m/z calcd
for C45H40O7 691.7900, found 691.2711.
8.18. General method for benzopyrilium salt
formation34
8.14. Enol ether 1931
A stream of anhydrous gaseous HCl was first passed through
a solution of hydroxyacetophenone (1.6 g, 7 equiv, 0.12 M
MeOH) at 0 C, until the concentration of HCl rose to
ꢂ
15% by mass. A solution of aldehyde 24 (0.43 g, 1 equiv,
0.06 M MeOH) was then added drop-wise at 0 C. At the
A flame-dried flask equipped with a stir-bar and a nitrogen
line containing EtOH (2 equiv, 0.5 M THF) was cooled to
ꢂ
ꢂ
0 C. Phenylmagnesium bromide (2 equiv) was added to
the solution warmed to rt, then evaporated to dryness. The
chromene 18 (1 equiv, 0.7 M benzene) was added to the
EtOMgBr, equipped with a reflux condenser, stir-bar, nitro-
gen line, and refluxed for 6 h. The reaction was quenched
with H2O, extracted EtOAc, washed with brine, and dried
with MgSO4. The reaction was purified by flash chromato-
graphy eluting with EtOAc/hexane (70:30) to afford the
desired product in 65% yield. 1H NMR (400 MHz,
CDCl3): d 7.53–7.05 (m, 15H), 7.03 (d, 2H, J¼8 Hz), 6.66
(d, 2H, J¼8 Hz), 6.47 (s, 1H), 6.36 (d, 1H, J¼10 Hz),
6.28–6.27 (m, 1H), 6.00 (s, 1H), 4.94 (s, 1H), 4.98 (s, 2H),
4.88–4.83 (m, 4H), 4.66 (d, 1H, J¼10 Hz), 3.74 (s, 3H),
3.72 (s, 3H), 3.33 (s, 3H).
end of the addition, the reaction mixture was kept for
a few minutes at rt and evaporated to dryness. The resulting
residue was vigorously stirred in Et2O, which afforded the
crude product after filtration and washing with the same sol-
vent. The pure product was obtained by recrystallization in
MeOH–HCl (95:5) in 65 and 70% yields for compounds
27 and 28, respectively. Luteolinidin Chloride (27). 1H
NMR (400 MHz, CD3OD–TFA 99:1): d 8.95 (d, 1H,
J¼9 Hz), 7.94 (d, 1H, J¼9 Hz), 7.78 (dd, 1H, J¼9, 2 Hz),
7.65 (d, 1H, J¼2 Hz), 6.98 (d, 1H, J¼9 Hz), 6.87 (d, 1H,
J¼2 Hz), 6.63 (d, 1H, J¼2 Hz); 13C NMR (100 MHz,
CD3OD–TFA 99:1): d 172.4, 171.9, 160.2, 159.6, 156.1,
149.0, 148.0, 125.3, 121.6, 117.9, 115.9, 113.5, 110.6,
103.3, 96.1. Apigeninidin Chloride (28). 1H NMR
(400 MHz, CD3OD–TFA 99:1): d 9.06 (dd, 1H, J¼9,
1 Hz), 8.27 (d, 2H, J¼9 Hz), 8.01 (d, 1H, J¼9 Hz), 7.06
(d, 2H, J¼9 Hz), 6.92 (dd, 1H, J¼2, 1 Hz), 6.64 (d, 1H,
J¼2 Hz); 13C NMR (100 MHz, CD3OD–TFA 99:1):
d 172.7, 172.4, 167.3, 160.6, 160.0, 149.5, 133.1, 121.3,
118.4, 113.8, 110.3, 103.2, 96.1.
8.15. Chroman 21
To a flame-dried flask equipped with a stir-bar and a nitrogen
line containing flavan 7 (20 mg, 0.10 M CH2Cl2) and tri-
benzylphloroglucinol (33.6 mg, 3 equiv) was added DDQ
(recrystallized from hot CHCl3, 15.8 mg, 1 equiv) at rt.

5306
C. Selenski, T. R. R. Pettus / Tetrahedron 62 (2006) 5298–5307
8.19. General method for the condensation of benzo-
pyrilium salt with phloroglucinol27
8.21. Compound 32
Benzopyrilium salt 27 (8.5 mg, 0.6 M MeOH/pH 5.8 buffer
(1:1) and flavan 6 (43 mg, 6 equiv) were heated at 150 C for
ꢂ
Benzopyrilium salt 27 (20 mg, 0.6 M MeOH/pH 5.8 buffer
(1:1)ꢂand phloroglucinol (26.2 mg, 3 equiv) were heated at
150 C for 2 h in an Emrys microwave reactor. The reaction
was concentrated to dryness on silica gel and purified by
flash chromatographed with EtOAc/hexane (80:20) to afford
2 h in an Emrys microwave reactor. The reaction was con-
centrated to dryness and submitted to pyridine and Ac2O
overnight. The reaction was quenched with 1 M HCl, repeat-
edly extracted with EtOAc, and dried with MgSO4. The de-
sired product was purified by gradient chromatography with
EtOAc/hexane (60:40–80:20) and the desired product was
isolated in 45% yield. 1H NMR (400 MHz, CDCl3):
d 7.58–7.50 (m, 4H), 7.46–7.38 (m, 4H), 7.16–7.08 (m,
4H), 6.75–6.68 (m, 2H), 6.53–6.51 (m, 2H), 6.29–6.27 (m,
2H), 5.17–4.92 (m, 2H), 4.14 (br s, 2H), 2.97 (m, 8H),
2.44–2.31 (m, 40H); HRMS (ESI-TOF) (M++Na) m/z calcd
for C42H36O15 803.2054, found 803.1935.
1
the desired product in 80% yield. Compound 30. H NMR
(400 MHz, CD6CO): d 7.19 (d, 1H, 2 Hz), 7.05 (dd, 1H,
J¼2, 8 Hz), 6.88 (d, 1H, J¼2 Hz), 6.06 (s, 4H), 4.33 (t,
1H, J¼3 Hz), 2.22 (d, 3H, J¼3 Hz); 13C NMR (100 MHz,
CD6CO): d 158.1, 154.7, 154.3, 146.4, 145.6, 134.9,
118.4, 115.8, 114.2, 107.5, 99.4, 97.3, 96.7, 34.5, 21.4;
HRMS (ESI-TOF) (M++Na) m/z calcd for C21H16O8
419.0843, found 419.0758. An alternative method for easier
purification is as follows. The reaction is concentrated to
dryness and submitted to pyridine (0.3 mL) and Ac2O
(0.5 mL) overnight. The reaction is quenched with 1 M
HCl, repeatedly extracted with EtOAc, and dried with
MgSO4. The reaction was purified with flash chromato-
graphy eluting with EtOAc/hexane (60:40) and the ketal
and non-ketal compounds were isolated. Acetylated com-
8.22. Diinsininone (33)
Benzopyrilium salt 27 (8.9 mg, 0.6 M MeOH/pH 5.8 buffer
(1:1)ꢂ and flavanone 9 (47 mg, 6 equiv) were heated at
120 C for 20 h in an Emrys microwave reactor. The reac-
tion was concentrated to dryness and submitted to pyridine
and Ac2O overnight. The reaction was quenched with 1 M
HCl, repeatedly extracted with EtOAc, and dried with
MgSO4. The desired product was purified by gradient chro-
matography with EtOAc/hexane (60:40–80:20) and the
desired product was isolated in 32% yield. 1H NMR
(400 MHz, CDCl3): d 7.57–7.48 (m, 8H), 7.24–7.19 (m,
6H), 6.74–6.72 (m, 2H), 6.51–6.50 (m, 4H), 5.60–5.51 (m,
2H), 4.47 (br s, 1H), 4.44 (br s, 1H), 3.09–3.00 (m, 2H),
2.78–2.27 (m, 2H), 2.39–2.27 (m, 40H); HRMS (ESI-
TOF) (M++Na) m/z calcd for C42H34O16 817.1847, found
817.1593.
1
pound 30. H NMR (400 MHz, CDCl3): d 7.56–7.27 (m,
3H), 6.74 (d, 2H, J¼2 Hz), 6.54 (d, 2H, J¼2 Hz), 4.28 (t,
1H, J¼3 Hz), 2.42 (s, 6H), 2.33 (s, 3H), 2.32 (s, 3H), 2.29
(s, 3H, J¼3 Hz), 2.26 (s, 6H); 13C NMR (100 MHz,
CDCl3): d 169.2, 169.0, 168.3, 153.7, 150.2, 147.8, 142.9,
142.1, 139.0, 124.3, 123.7, 121.6, 115.7, 109.8, 108.5,
98.4, 33.0, 22.9, 21.7, 21.3, 20.9, 20.8; IR (CH2Cl2, ymax
cmꢁ1) 1772, 1618, 1369, 1198, 1222, 1024; HRMS (ESI-
TOF) (M++Na) m/z calcd for C33H28O14 671.1479, found
671.1378. Enol ether 30. 1H NMR (400 MHz, CDCl3):
d 7.53–7.22 (m, 3H), 6.90–6.82 (m, 4H), 5.32 (d, 1H,
J¼4 Hz), 4.90 (d, 1H, J¼4 Hz), 2.48–2.25 (m, 21H);
HRMS (ESI-TOF) (M++Na) m/z calcd for C35H30O15
713.1585, found 713.1511.
Acknowledgments
Research support from the University of California Coordi-
nating Committee on Cancer Research in the form of two
awards (19990641) and (SB010064) are greatly appreciated
along with additional support of NSF (CHE-9971211), PRF
(PRF34986-G1), NIH (GM-64831), the UC-AIDS Initiative
(K00-SB-039) and Research Corporation (R10296). We are
also thankful for the support of our M.S. facility funded in
part by DAAD19-00-1-0026.
8.20. Acetylated compound of 31
Benzopyrilium salt 28 (30 mg, 0.6 M MeOH/pH 5.8 buffer
(1:1)ꢂand phloroglucinol (39.0 mg, 3 equiv) were heated at
150 C for 2 h in an Emrys microwave reactor. The reaction
was concentrated to dryness and submitted to pyridine
(0.3 mL) and Ac2O (0.5 mL) overnight. The reaction is
quenched with 1 M HCl, repeatedly extracted with EtOAc,
and dried with MgSO4. The crude product was purified by
flash chromatography with EtOAc/hexane (60:40) and the
1
ketal and non-ketal product were isolated in 83% yield. H
References and notes
NMR (400 MHz, CDCl3): d 7.68 (d, 2H, J¼9 Hz), 7.18 (d,
2H, J¼9 Hz), 6.75 (d, 2H, J¼2 Hz), 6.55 (d, 2H, J¼2 Hz),
4.27 (t, 1H, J¼3 Hz), 2.43 (s, 6H), 2.29 (d, 3H, J¼3 Hz),
2.33 (s, 3H), 2.27 (s, 6H); 13C NMR (100 MHz, CDCl3):
d 169.5, 169.2, 169.0, 153.8, 151.4, 150.2, 147.8, 137.8,
127.3, 121.8, 115.8, 109.7, 108.5, 98.9, 33.1, 23.0, 21.7,
21.4, 21.3; d IR (CH2Cl2, ymax cmꢁ1) 1768, 1620, 1434,
1369, 1201, 1184, 1130, 1116, 1072, 1022, 904; HRMS
(ESI-TOF) (M++Na) m/z calcd for C31H26O12 613.1424,
found 613.1337. Enol ether 31. 1H NMR (400 MHz,
CDCl3): d 7.63 (d, 2H, J¼9 Hz), 7.10 (d, 2H, J¼9 Hz),
6.90–6.60 (m, 4H), 5.30 (d, 1H, J¼4 Hz), 4.91 (d, 1H,
J¼4 Hz), 2.38–2.13 (m, 18H); HRMS (ESI-TOF) (M++Na)
m/z calcd for C33H28O13 655.1530, found 655.1423.
1. Ogundaini, A.; Farah, M.; Perera, P.; Samuelsson, G.; Lars, B.
J. Nat. Prod. 1996, 59, 587–590.
2. Samuelsson, G.; Farah, M. H.; Claeson, P.; Hagos, M.; Thulin,
M.; Hedberg, O.; Warfa, A. M.; Hassan, A. O.; Elmi, A. H.;
Abdulrahman, A. D.; Elmi, A. S.; Adbi, Y. A.; Alin, M. J.
J. Ethnopharmacol. 1991, 35, 25–63.
3. Sheehan, K. M.; Sheahan, K.; O’Donoghue, D. P.; Mac-
Sweeney, F.; Conroy, R. M.; Fitzgerald, D. J.; Murray, F. E.
JAMA 1999, 282, 1254–1257.
4. Haborne, J. B. Phytochemistry 1966, 5, 589–600.
5. Pietta, P.; Gardana, C.; Pietta, A. Flavanoids in Herbs. In Flavo-
noids in Health and Disease, 2nd ed.; Rice-Evans, C. A.,

C. Selenski, T. R. R. Pettus / Tetrahedron 62 (2006) 5298–5307
5307
Packer, L., Eds.; Marcel Dekker: New York, NY, 2003; pp 43–
69.
6. Winkel-Shirley, B. Curr. Opin. Plant Biol. 2002, 5, 218–
223.
23. (a) Steynberg, P. J.; Nel, R. J. J.; van Rensburg, H.; Bezuiden-
houdt, B. C. B.; Ferreira, D. Tetrahedron 1998, 54, 8153–8158;
(b) Botha, J. J.; Ferreira, D.; Roux, D. G. J. Chem. Soc., Perkin
Trans. 1 1981, 1235–1245; (c) Steenkamp, J. A.; Ferreira, D.;
Roux, D. G. Tetrahedron Lett. 1985, 26, 3045–3048; (d)
Delcour, J. A.; Ferreira, D.; Roux, D. G. J. Chem. Soc., Perkin
Trans. 1 1983, 1711–1717; (e) Botha, J. J.; Ferreira, D.; Roux,
D. G. J. Chem. Soc., Perkin Trans. 1 1981, 1235–1245.
7. Flavonoids in Health and Disease; Rice-Evans, C. A.,
Packer, L., Eds.; Marcel Dekker: New York, NY, 2003.
8. Renaud, S.; De Lorgeril, M. Lancet 1992, 339, 1523–1526.
9. Simonetti, P.; Gardana, C.; Pietta, P. Caffeic Acid as Biomarker
of Red Wine Intake. In Methods in Enzymology; Packer, L.,
Ed.; Academic: San Diego, CA, 2001; Vol. 335, pp 122–130.
10. (a) Frankel, E. N.; Waterhouse, A. L.; Kinsella, J. E. Lancet
1993, 341, 454–457; (b) Soleas, G. I.; Diamandis, E. P.; Gold-
berg, D. M. J. Clin. Lab. Anal. 1997, 11, 287–313.
11. Frei, B. Crit. Rev. Food Sci. Nutr. 1995, 35, 83–98.
12. Ursini, F.; Rapuzzi, I.; Toniolo, R.; Tubaro, F.; Bontempelli, G.
Characterization of Antioxidant Effect of Procyanidins. In
Methods in Enzymology; Packer, L., Ed.; Academic: San
Diego, CA, 2001; Vol. 335, pp 338–360.
13. (a) McCarty, M. F. Med. Hypotheses 2005, 64, 151–158; (b)
Imparl-Radosevich, J.; Deas, S.; Polansky, M. M.; Baedke,
D. B.; Ingebritsen, T. S.; Anderson, R. A.; Graves, D. Horm.
Res. 1998, 50, 177–182; (c) Khan, A.; Safdar, M.; Ali Kahn,
M. M.; Kattak, K. N.; Anderson, R. A. Diabetes Care 2003,
26, 3215–3218.
24. (a) Arnaudinaud, V.; Nay, B.; Nuhrich, A.; Deffieux, G.;
´
Merillon, J.-M.; Monti, J.-P.; Vercauteren, J. Tetrahedron
Lett. 2001, 42, 1279–1281; (b) Arnaudinaud, V.; Nay, B.;
´
Sarah, V.; Nuhrich, A.; Deffieux, G.; Merillon, J.-M.; Monti,
J.-P.; Vercauteren, J. Tetrahedron Lett. 2001, 42, 5669–5671.
25. (a) Saito, A.; Nakajima, N.; Tanaka, A.; Ubukata, M. Tetra-
hedron 2002, 58, 7829–7837; (b) Saito, A.; Nakajima, N.;
Tanaka, A.; Ubukata, M. Tetrahedron Lett. 2003, 58, 5449–
5452; (c) Saito, A.; Emoto, M.; Tanaka, A.; Doi, Y.; Shoji,
K.; Mizushina, Y.; Ikawa, H.; Yoshida, H.; Matsuura, N.;
Nakajima, N. Tetrahedron 2004, 60, 12043–12049; (d) Saito,
A.; Doi, Y.; Tanaka, A.; Matsuura, N.; Ubukata, M.; Nakajima,
N. Bioorg. Med. Chem. 2004, 12, 4783–4790; (e) Saito, A.;
Tanaka, A.; Ubukata, M.; Nakajima, N. Synlett 2004, 1069–1073.
26. (a) Tuckmantel, W.; Kozikowski, A. P.; Romanczyk, L. J., Jr.
¨
J. Am. Chem. Soc. 1999, 121, 12073–12081; (b) Kozikowski,
14. Kraja, A. T.; Rao, D. C.; Weder, A. B.; Mosley, T. H.; Turner,
S. T.; Hsiung, C. A.; Quertermous, T.; Cooper, R.; Curb, J. D.;
Province, M. A. Nutr. Metab. 2005, 2, 29–42.
15. Anderson, R. A.; Broadhurst, C. L.; Polansky, M. M.; Schmidt,
W. F.; Khan, A.; Flanagan, V. P.; Schoene, N. W.; Graves, D. J.
J. Agric. Food Chem. 2004, 52, 65–70.
A. P.; Tuckmantel, W.; Hu, Y. J. Org. Chem. 2001, 66, 1287–
¨
1296; (c) Kozikowski, A. P.; Tuckmantel, W.; Bottcher, G.;
¨
¨
Romanczyk, L. J., Jr. J. Org. Chem. 2003, 68, 1641–1658.
27. (a) Jurd, L.; Waiss, A. C., Jr. Tetrahedron 1965, 21, 1471–1483;
(b) Jurd, L. J. Heterocycl. Chem. 1981, 18, 429–430; (c) Jurd,
L.; Lundin, R. Tetrahedron 1968, 24, 2653–2661.
28. Magdziak, D.; Rodriguez, A. A.; van de Water, R. W.; Pettus,
T. R. R. Org. Lett. 2002, 4, 285–288.
16. Marles, M. A. S.; Ray, H.; Gruber, M. Y. Phytochemistry 2003,
64, 367–383.
17. Winefield, C. S.; Lewis, D. H.; Swinny, E. E.; Zhang, H.;
Arathoon, H. S.; Fischer, T. C.; Halbwirth, H.; Stich, K.; Gosch,
C.; Forkmann, G.; Davies, K. M. Physiol. Plant. 2005, 124,
419–430.
18. Lo, S-Z. C.; Nicholson, R. L. Plant Physiol. 1998, 116, 979–
989.
29. Huang, Y.; Zhang, J.; Pettus, T. R. R. Org. Lett. 2005, 7, 5841–
5844.
30. Oyama, K.-I.; Kondo, T. Tetrahedron 2004, 60, 2025–2034.
31. (a) Caisraghi, G.; Casnati, G. Chem. Commun. 1970, 321; (b)
Caisraghi, G.; Casnati, G.; Salerno, G. J. Chem. Soc. C 1971,
2546–2548.
19. (a) Kondo, K.; Kurihara, M.; Fukuhara, K.; Tanaka, T.; Suzuki,
T.; Miyata, N.; Toyoda, M. Tetrahedron Lett. 2000, 41, 485–
488; (b) Nonaka, G.-I.; Morimoto, S.; Kinjo, J.-E.; Nohara,
T.; Nishioka, I. Chem. Pharm. Bull. 1987, 35, 149–155.
20. The Handbook of Natural Flavonoids, Vol. 1; Harborne, J. B.,
Baxter, H., Eds.; Wiley: New York, NY, 1999, pp vii–xvi.
21. Marais, J. P. J.; Ferreira, D.; Slade, D. Phytochemistry 2005, 66,
2145–2176.
22. (a) Selenski, C.; Pettus, T. R. R. J. Org. Chem. 2004, 69, 9196–
9203; (b) Selenski, C.; Mejorado, L. H.; Pettus, T. R. R. Synlett
2004, 1101–1104; (c) Jones, R. M.; Selenski, C.; Pettus, T. R.
R. J. Org. Chem. 2002, 67, 6911–6915; (d) Selenski, C.; Pettus,
T. R. R. o-Quinone Methides. In Science of Synthesis; Gries-
beck, A. G., Ed.; Georg Thieme: Stuttgart, in press; (e) Van
De Water, R.; Magdziak, D.; Chau, J.; Pettus, T. R. R. J. Am.
Chem. Soc. 2000, 122, 6502–6503.
32. (a) Camarda, L.; Merlini, L.; Nasini, G. Flavonoids from
Natural Red Resins. In Flavonoids and Bioflavonoids, 1981;
´
Farkas, L., Kallay, F., Gabor, M., Wagner, H., Eds.; Akademiai
Kiado: Budapest, 1982; pp 311–320; (b) Arnone, A.; Nasini,
´
G.; de Pava, O. V. J. Nat. Prod. 1997, 60, 971–975.
33. (a) Sweeney, J. G.; Iacobucci, G. A. Tetrahedron 1981, 37,
1481–1483; (b) Sweeney, J. G.; Iacobucci, G. A. Tetrahedron
1977, 33, 2927–2932; (c) Robinson, G. M.; Robinson, R.;
Todd, A. R. J. Chem. Soc. 1934, 809–812; (d) Leon, A.;
Robinson, R. J. Chem. Soc. 1931, 2732–2737; (e) Pratt, D.
D.; Robinson, R. J. Chem. Soc. 1925, 1128–1138.
34. Mas, T. Synthesis 2003, 12, 1878–1880.
35. Caccamese, S.; Caruso, C.; Parrinello, N.; Savarino, A.
J. Chromatogr. 2005, 1076, 155–162.
36. Frigerio, M.; Santagostino, M.; Sputore, S. J. Org. Chem. 1999,
64, 4537–4538.