Biomimetic total syntheses of cassiarins A and B

Article abstract of DOI:10.1021/jo801017b

(Chemical Equation Presented) Total syntheses of cassiarins A and B have been efficiently accomplished using a common strategy with biomimetic considerations. Key reactions involved in this synthesis include a Negishi-type coupling, a Ag(I)-promoted formation of the tricyclic 8H-pyrano[2,3,4-de] chromen-8-one core, and a sequential amine-condensation and cyclization. Three new analogues of cassiarin A bearing different substituents at the C-11 position were synthesized in parallel from the same intermediate. In addition, two other transformations to the key tricyclic cores and cassiarins A and B were achieved from corresponding chemically equivalent precursors.

Full text of DOI:10.1021/jo801017b

Biomimetic Total Syntheses of Cassiarins A and B
Yuan-Shan Yao and Zhu-Jun Yao*
State Key Laboratory of Bioorganic and Natural Products Chemistry, Shanghai Institute of Organic
Chemistry, Chinese Academy of Sciences, 354 Fenglin Road, Shanghai 200032, China, and Joint
Laboratory of Green Chemistry, Department of Chemistry, UniVersity of Science and Technology of China,
Hefei, Anhui 230026, China
yaoz@mail.sioc.ac.cn
ReceiVed May 13, 2008
Total syntheses of cassiarins A and B have been efficiently accomplished using a common strategy with
biomimetic considerations. Key reactions involved in this synthesis include a Negishi-type coupling, a
Ag(I)-promoted formation of the tricyclic 8H-pyrano[2,3,4-de]chromen-8-one core, and a sequential amine-
condensation and cyclization. Three new analogues of cassiarin A bearing different substituents at the
C-11 position were synthesized in parallel from the same intermediate. In addition, two other
transformations to the key tricyclic cores and cassiarins A and B were achieved from corresponding
chemically equivalent precursors.
Introduction
Development of new antimalarial drugs has been neglected
for decades until multiple-drug-resistant strains of malaria
parasites were discovered in recent years.1 The search for new
therapeutic agents against malaria is thus pursued from the
natural resources and synthetic compounds. Cassiarins A (1,
0.0008% isolated yield) and B (2, 0.0017% isolated yield, Figure
1) have been isolated by Morita and co-workers in 2007 from
the leaves of Cassia siamea (Leguminosae), which have been
widely used in traditional medicine particularly for the treatment
of periodic fever and malaria.2 Their structures, with an
unprecedented tricyclic skeleton (Figure 1A,B), were elucidated
and assigned by spectroscopic analysis. Cassiarins A and B
exhibit very potent antiplasmodial activity against Plasmodium
falciparum with an IC50 0.005 and 6.9 µg/mL, respectively.
FIGURE 1. Structures of cassiarins A (1) and B (2), their tricyclic
cores A and B, and a proposed biogenetic precursor (3a).
Thus, these two natural products would be very attractive targets
for organic synthesis. Very recently, the first total synthesis of
cassiarin A was completed in eight steps from a known
compound, methyl 2,4-dihydroxybenzoate, by Honda and Mori-
(1) For recent reviews, see: (a) Schlitzer, M. Arch. Pharm. (Weinheim) 2008,
341, 149–163. (b) Nosten, F.; White, N. J. Am. J. Trop. Med. Hyg. 2007, 77,
181–192.
(2) Morita, H.; Oshimi, S.; Hirasawa, Y.; Koysma, K.; Honda, T.; Ekasari,
W.; Indrayanto, G.; Zaini, N. C. Org. Lett. 2007, 9, 3691–3693.
10.1021/jo801017b CCC: $40.75
Published on Web 06/21/2008
2008 American Chemical Society
J. Org. Chem. 2008, 73, 5221–5225 5221

Yao and Yao
SCHEME 1. Synthesis of the Common Acetylene
Intermediate 6b
the literature conditions,6 5,7-dihydroxy-2-methyl-4H-chromen-
4-one (8) was prepared in a large scale from the commercially
available material 2,4,6-trihydroxyacetophenone (9). With as-
sistance of the intramolecular H-bonding between C-5 hydroxyl
and C-4 ketone carbonyl,7 selective protection of C-7 hydroxyl
group of chromenone 8 was achieved using MOMCl and DIPEA
in CH2Cl2, and gave MOM ether 10 in 91% yield. Then, the
remaining phenol hydroxyl of chromenone 10 was converted
to the corresponding triflate 7. Initially, attempts to prepare
triflate 7 all failed using combinations of Tf2O with different
bases. Finally, treatment of 10 with PhN(Tf)2 in the presence
of NaH8 at 0 °C gave the desired triflate 7 in 99% yield. Negishi-
type coupling of triflate 7 with commercially available 1-pro-
pynylmagnesium bromide (1 M, THF solution) was accom-
plished in the presence of ZnCl2 (1 equiv) and Pd(PPh3)4 (5
mol %),9 affording acetylene 6b in up to 73% yield.
FIGURE 2. Retrosynthetic analysis of cassiarins A (1) and B (2).
ta.3 However, their attempts failed to convert cassiarin A to
cassiarin B. In this paper, we report highly efficient biomimetic
total syntheses of cassiarins A and B, as well as expanded
syntheses of three new analogues of cassiarin A with similar
chemistries.
Retrosynthetic Analysis. As mentioned, cassiarins A and B
have a common tricyclic core, pyrano[2,3,4-ij]isoquinolin-8-ol
(A) or its tautomer (B) (Figure 1). A biogenetic pathway has
been proposed for both of them, starting from a common
precursor, 5-acetonyl-7-hydroxy-2-methylchromone (3a).2 With
considerations of the plausible biogenetic pathway2 and our
previous experience on isochromenylium salts chemistry,4 an
active salt intermediate 4 and its neutral equivalents 3 and 5
(Figure 2) were therefore regarded as key common intermedi-
ate(s) in this synthesis. Salt 4 could be generated in situ from
an acetylene precursor 6 using the acidic or Ag(I)-catalyzed
reaction conditions,4,5 and it was supposed to react with the
corresponding amines (ammonia and methyl 4-aminobutyrate),
affording the desired tricyclic compounds, respectively. Acety-
lene chromenone 6 could be synthesized from a triflate precursor
7 by cross-coupling with propyne. The further precursor, 5,7-
dihydroxy-2-methyl-4H-chromen-4-one (8), is known and can
be easily prepared from a commercially available material, 2,4,6-
trihydroxyacetophenone (9).
Total Syntheses. With acetylene 6b in hand, elaboration to
the tricyclic salt intermediate 4b was then explored (Scheme
2). This reaction worked very well in 1,2-dichloroethane at -24
°C with a catalytic amount of AgNO3 and 3 equiv of TFA.
However, our attempts to separate and characterize salt 4b failed.
Finally, a successive operation was adopted. Directly bubbling
dry ammonia into the dichloroethane solution of the previous
reaction containing crude salt 4b afforded a tricyclic compound
1
11, which was isolated and its structure was confirmed by H
NMR and MS. Further treatment of 11 with 2 M aqueous HCl
in MeOH under reflux gave cassiarin A (1) (60% overall yield
from 6b). In parallel, cassiarin B (2, 52% yield from 6b) was
synthesized from the same precursor 6b using methyl 4-ami-
nobutyrate (to replace ammonia in the previous case).
According to our previous retroanalysis (Figure 2), other two
pathways to cassiarins were also explored using chemical
equivalents of salt 4. One is based on the neutral tricyclic
compound 5, which is chemically equal to salt 4a (R ) H,
Figure 2) and offers another potential entrance to cassiarins.
Compound 5 could be easily prepared from the acetylene 6b
using either a one-step procedure (2 N aq HCl, MeOH, reflux)
or a two-step alternative (Lewis acid based deprotection
followed by Ag+-catalyzed cyclization) (Scheme 3). Exactly
as previously proposed, reaction of 5 with 3 equiv of NH4Cl in
MeOH under reflux (sealed tube) for 8 h gave cassiarin A (1)
Results and Discussion
Synthesis of Common Precursor 6b. The common alkyne
precursor 6b was synthesized as shown in Scheme 1. Under
(3) Rudyanto, M.; Tomizawa, Y.; Morita, H.; Honda, T. Org. Lett. 2008,
10, 1921–1922.
(4) (a) Wei, W. G.; Yao, Z. J. J. Org. Chem. 2005, 70, 4585–4590. (b)
WeiW.-G.; Qian, W.-J.; Zhang, Y.-X.; Yao, Z.-J. Tetrahedron Lett. 2006, 47,
4171–4174. (c) Qian, W.-J.; Wei, W.-G.; Zhang, Y.-X.; Yao, Z.-J. J. Am. Chem.
Soc. 2007, 129, 6400–6401.
(6) Murti, S. Proc. Indian Acad. Sci. Sect. A 1949, 30, 107–112.
(7) Kelly, T. R.; Kim, M. H. J. Org. Chem. 1992, 57, 1593–1597.
(8) (a) Arcadi, A.; Cacchi, S.; Fabrizi, G.; Manna, F.; Pace, P. Synlett 1998,
446–448. (b) Zhong, B. F.; McDonald, F. E. Org. Lett. 2005, 7, 3617–3620.
(9) (a) King, A. O.; Okukado, N.; Negishi, E.-i. J. Chem. Soc., Chem. Commun.
1977, 683–684. (b) Negishi, E.-i.; Kotora, M.; Xu, C. D. J. Org. Chem. 1997, 62,
8957–8960. (c) Wang, K. K.; Wang, Z. G.; Gannett, P. J. Am. Chem. Soc. 1996,
118, 10783–10791.
(5) (a) Zhu, J. L.; Grigoriadis, N. P.; Lee, J. P.; Porco, J. A., Jr J. Am. Chem.
Soc. 2005, 127, 9342–9343. (b) Zhu, J. L.; Germain, A. R.; Porco, J. A., Jr
Angew. Chem., Int. Ed. 2004, 43, 1239–1243. (c) Tovar, J. D.; Swager, T. M. J.
Org. Chem. 1999, 64, 6499–6504.
5222 J. Org. Chem. Vol. 73, No. 14, 2008

Biomimetic Total Syntheses of Cassiarins A and B
SCHEME 2. Total Syntheses of Cassiarins A and B via Salt
Intermediate 4
SCHEME 4. Total Syntheses of Cassiarins A and B via
Diketone Intermediates 3a and 3b
TABLE 1. Summary of Key Information from NOESY and
HMBC Experiments of Synthesized Cassiarinsa
SCHEME 3. Total Syntheses of Cassiarins A (1) and B (2)
via Neutral Intermediate 5
a 2D NMR experiments were performed in CD3OD-CDCl3 (1:1).
in 85% yield. The other parallel reaction of 5 with methyl
4-aminobutyrate hydrochloride in MeOH under reflux for 12 h
afforded cassiarin B (2) in 68% yield.
suspected that the reported solvent ratio (CDCl3-CD3OD, 1:1)
for NMR measurements might not be very accurate. Thus, a
number of mixed solvents with different ratios around 1:1
(CDCl3 vs CD3OD, in ratios of 40:60, 45:55, 48:52, 50:50, 52:
5-Acetonyl-7-hydroxy-2-methylchromone (3a, Figure 2) is
one more potential biogenetic precursor for both cassiarins.2
Hydrolysis of salt 4b (prepared from 6b, Scheme 2) with
aqueous NaOAc afforded diketone 3b (70%). Further acidic
deprotection of MOM ether 3b gave the desired diketone 3a
(78%). To our delight, both diketones 3b and 3a could be
converted to cassiarins A (1, 52%) and B (2, 65%) after their
reactions with NH4Cl and methyl 4-aminobutyrate hydrochlo-
ride, respectively (Scheme 4).
1
48, 55:45, 60:40) were examined by H NMR experiments of
1
our synthesized cassiarin B (2). Unfortunately, that H NMR
spectrum reported for natural cassiarin B could not be repro-
duced, and the closest one is that measured in CD3OD-CDCl3
1
(1:1). Even in a single solvent CD3OD, both the H and 13C
NMRs of synthetic cassiarin B could not match the standard
spectra of natural sample (its concentration unknown).10
These results suggested to us that such irreproducibility of
NMR spectra might be caused by some unusual solution
behaviors, which are usually sensitive to the sample concentra-
Confirmation of Structures. The synthesized cassiarin A
(1) is in full agreement with all aspects of natural product.2
Surprisingly, synthetic cassiarin B (2) shows significant differ-
1
tions. With such suspicion, H NMRs of synthetic cassiarin B
1
ences in both H NMR and 13C NMR (CD3OD-CDCl3, 1:1)
(in CD3OD) were examined using samples within different
concentrations. Tuning sample concentrations of cassiarin B
resulted in slow movement of the chemical shifts of some
protons (Table 2). As sample concentration increases, chemical
shifts of protons H-6, H-8, H-3, and H-10 move to the higher
fields, and other protons H-17, H-15, H-12, and H-9 are
by comparison with those spectra of natural product, though it
gives similar peak patterns. Our further H-H COSY, HMBC,
and NOESY studies confirmed that both synthetic cassiarins
have correct structures as proposed (Table 1).2 In order to
explain the causes of spectral difference, extensive studies were
conducted. At first, the chemical shifts of lower field protons
of cassiarin B (especially for H-6, H-8, H-3, H-10) were found
to be very sensitive to the ratio of the mixed solvent
(CD3OD-CDCl3) used for NMR experiments. We initially
(10) The 1H and 13C NMR spectra of natural cassiarins A and B (CD3OD,
400 MHz) were recorded and provided by Prof. Hiroshi Morita on April 3, 2008
(see the Supporting Information).
J. Org. Chem. Vol. 73, No. 14, 2008 5223

Yao and Yao
TABLE 3. Parallel Syntheses of New Analogues of Cassiarin A
1
TABLE 2. Summary for H NMR Comparison of Synthetic
Cassiarin B (2) in CD3OD
chemical shifts (ppm)
sample
no.
wta
H-10 H-3 H-8 H-6 H-13 H-17 H-15 H-12 H-9 H-14
1
2
3
4
5
6
7
8
0.5 mg 6.90 6.83 6.61 6.49 4.17 3.72 2.60 2.54 2.45 2.01
1.0 mg 6.89 6.82 6.59 6.47 4.16 3.72 2.60 2.54 2.45 2.01
1.5 mg 6.88 6.81 6.58 6.46 4.15 3.72 2.60 2.53 2.44 2.01
2.0 mg 6.86 6.80 6.57 6.45 4.14 3.72 2.60 2.53 2.44 2.00
4.0 mg 6.82 6.76 6.53 6.41 4.11 3.72 2.59 2.52 2.43 1.99
6.0 mg 6.80 6.74 6.50 6.38 4.11 3.72 2.59 2.51 2.42 1.95
10.0 mg 6.75 6.70 6.45 6.33 4.06 3.72 2.59 2.49 2.41 1.96
no. 4 + 6.83 6.76 6.53 6.41 4.12 3.72 2.60 2.52 2.43 1.99
no. 6
9b natural 6.89 6.83 6.61 6.50 4.15 3.72 2.60 2.53 2.45 2.01
a Samples 1-7 were dissolved in 0.4 mL of CD3OD (see the
Supporting Information for the recorded spectra). b The 1H NMR
spectrum of natural cassiarin B in CD3OD was recorded and provided
by Prof. H. Morita on April 3, 2008, and its concentration is unknown.
relatively not very sensitive to such changes. For example, the
spectra of samples 4 (2 mg of 2 in 0.4 mL CD3OD, Table 2),
5 (4 mg of 2 in 0.4 mL CD3OD), and 6 (6 mg of 2 in 0.4 mL
CD3OD) are quite different from each other. However, an
additional experiment by mixing samples 4 and 6 afforded the
same spectrum as sample 5. The above information proposes
that the densely conjugated benzoquinone moiety in cassiarin
B might be responsible for the supramolecular interactions in
solution observed in NMR studies. We believe supramolecular
interactions of cassiarin B in solution are more complicated,
and further investigation will be needed for a full explanation.
To our delight, a very dilute sample (0.5 mg of synthetic
a The cross coupling was carried out with 10 mol % of Pd(PPh3)4, 30
mol % of CuI, and 150 mol % of n-Bu4NI in DMF/Et3N (1/2, v/v)
under N2 at rt.11 b Pd(PPh3)2Cl2 was used to replace Pd(PPh3)4. c All
reactions were carried out under the same conditions as described in
Scheme 2.
1
cassiarin B in 0.4 mL of CD3OD) finally gave an H NMR
spectrum identical to that of the natural product.10
Synthesis of New Analogues. Because of the highly potent
antiplasmodial activity exhibited by cassiarin A, structural
expansion upon its unique tricyclic core is of extreme impor-
tance. As one advantage of our developed protocols, such a task
could be easily achieved by simply following the steps in
Scheme 2. Three new analogues of cassiarin A (1c, 1d, and 1e)
bearing different substituents at the C-11 position were prepared
in parallel, commonly starting from the same triflate 7 (Table
3).
Experimental Section
5-Hydroxy-7-(methoxymethoxy)-2-methyl-4H-chromen-4-
one (10). To a stirred solution of 5,7-dihydroxy-2-methyl-chromen-
4-one (8, 2.70 g, 14.0 mmol) in CH2Cl2 (40 mL) was added DIPEA
(1.3 mL, 17.0 mmol) at 0 °C under N2. After the solution was stirred
for 10 min, MOMCl (1.27 mL, 17.0 mmol) was added at the same
temperature. The reaction mixture was allowed to warm to room
temperature and stirred for 1.5 h until completion of the reaction
(TLC). The reaction was quenched with satd aq NaHCO3. The
solvent was removed under reduced pressure and water was added
and then extracted with ethyl acetate. The combined organic phases
were successively washed with 1 M HCl, water, and brine, dried
with anhydrous Na2SO4, and concentrated in vacuo. Purification
with column chromatography (petroleum ether/ethyl acetate ) 2/1)
gave a pale yellow solid (10, 3.07 g, 91%): mp 109-111 °C; IR
Conclusion
In summary, the efficient biomimetic total syntheses of
cassiarins A (1) and B (2) have been accomplished in four
operations from 5,7-dihydroxy-2-methyl-4H-chromen-4-one (8)
in 39% and 34% overall yield, respectively, according to the
first route (Schemes 1 and 2). A Negishi-type coupling, a Ag(I)-
catalyzed formation of key salt intermediate 4b, and an
aminolysis of salt 4 and following condensations successfully
served as key steps in these syntheses. Conversions of other
two precursors 3 and 5 to cassiarins A and B were also achieved.
Both the synthesized cassiarins were confirmed by their 1D
NMR and 2D NMR experiments, as well as corresponding
spectral comparisons. A preliminary structural expansion based
on tricyclic core of cassiarin A was explored, and three new
analogues of cassiarin A were prepared from the same inter-
mediate 7. Further applications of the developed chemistry and
biological study of new cassiarin analogues are underway and
will be reported in due course.
1668, 1625, 1594, 1338, 1158, 1020 cm-1; H NMR (300 MHz,
1
CDCl3) δ 12.67 (s, 1H), 6.53 (d, J ) 2.4 Hz, 1H), 6.45 (d, J ) 2.1
Hz, 1H), 6.04 (s, 1H), 5.22 (s, 2H), 3.49 (s, 3H), 2.35 (s, 3H);
ESI-MS (m/z) 237 (M + H+), 259 (M + Na+). Anal. Calcd for
C12H12O5: C, 61.01; H, 5.12. Found: C, 60.68; H, 5.09.
7-(Methoxymethoxy)-2-methyl-4-oxo-4H-chromen-5-yl Tri-
fluoromethanesulfonate (7). To a stirred solution of 10 (5.00 g,
21.2 mmol) in THF (90 mL) was added NaH (60%, 0.97 g, 24.3
mmol) by portions at 0 °C under N2. The suspended yellow solution
was stirring for 10 min, and then a solution of PhNTf2 (9.08 g,
25.4 mmol) in THF (25 mL) was added. The resulting mixture was
allowed to warm to room temperature after it became clear. After
being stirred for an additional 1 h, the reaction was quenched with
satd aq NH4Cl. THF was evaporated, and the residue was diluted
with water and extracted with ethyl acetate three times. The
combined organic phases were washed with water and brine, dried
over anhydrous Na2SO4, and concentrated. The crude product was
purified by flash chromatography (petroleum ether/ethyl acetate )
(11) Dai, W.-M.; Guo, D.-S.; Sun, L.-P. Tetrahedron 2001, 42, 5275–5278.
5224 J. Org. Chem. Vol. 73, No. 14, 2008

Biomimetic Total Syntheses of Cassiarins A and B
3/1) to give 7 (7.30 g, 99%): mp 109-110 °C; IR 1668, 1623,
1425, 1389, 1214, 1138, 1024 cm-1; 1H NMR (300 MHz, CDCl3)
δ 7.11 (d, J ) 2.1 Hz, 1H), 6.85 (d, J ) 1.8 Hz, 1H), 6.10 (s, 1H),
5.27 (s, 2H), 3.51 (s, 3H), 2.35 (s, 3H); ESI-MS (m/z) 369 [M +
H+], 391 (M + Na+). Anal. Calcd for C13H11F3O7S: C, 42.40; H,
3.01. Found: C, 42.18; H, 3.07.
MHz) δ 20.8, 23.6, 102.0, 104.3, 105.0, 112.8, 115.0, 140.7, 151.4,
152.2, 157.9, 163.4, 166.4. ESI-MS (m/z) 214 [M + H+]; HRMS
(ESI) calcd for C13H12NO2 [M + H+] 214.0863, found 214.0867.
Cassiarin B (2) (from 6b). To a solution of 6b (117 mg, 0.426
mmol) and AgNO3 (3.6 mg, 5 mol %) in (CH2Cl)2 (4 mL) was
dropwise added TFA (100 µL, 1.35 mmol) at -24 °C under
nitrogen atmosphere. The orange solution was stirred for 1 h at
same temperature. After the cooling bath was removed, methyl
γ-aminobutyrate hydrochloride (85 mg, 0.54 mmol) and Et3N (0.3
mL, 2.25 mmol) were added. The resulting mixture was stirred
overnight at room temperature. After removal of solvent, 2 N aq
HCl (3 mL) and MeOH (7 mL) were added. The mixture was
refluxed for 30 min before cooling to room temperature. Solvent
was removed, and water was added. The aqueous layer was adjusted
to pH 9 with 2 N aq Na2CO3 and extracted with CHCl3. The organic
phase was washed with water and brine, dried over anhydrous
Na2SO4, and concentrated. The crude product was purified by flash
chromatography (CH2Cl2/MeOH ) 10/1) to give 2 (74 mg, 52%):
7-(Methoxymethoxy)-2-methyl-5-(prop-1-ynyl)-4H-chromen-
4-one (6b). To a 50 mL flask were added 7 (2.00 g, 5.40 mmol),
Pd(PPh3)4 (314 mg, 0.27 mmol), and LiCl (420 mg, 10.8 mmol).
The mixture was degassed and kept under nitrogen atmosphere at
room temperature, and then 15 mL of freshly distilled THF was
added. In the other flask, 1-propynylzinc chloride was prepared by
treatment of 1-propynylmagnesium bromide (0.5 M in THF, 16.2
mL) with a solution of anhydrous ZnCl2 in THF (1.0 M, 8.1 mL)
at rt. The prepared solution of 1-propynylzinc chloride was
transferred via syringe into the first flask at room temperature. The
resulting red solution was stirred for 20 h until water was added to
quench the reaction. THF was evaporated, and the residue was
diluted with water and extracted with ethyl acetate. The combined
organic phases were successively washed with water and brine,
dried over anhydrous Na2SO4, and concentrated. Purification by
flash chromatography (petroleum ether/ethyl acetate ) 4/1) gave
6b (0.97 g, 75% based on 93% conversion of starting material):
mp 80 °C dec; IR 1733, 1654, 1597, 1457, 1437, 1197, 1170 cm-1
;
1H NMR (500 MHz, CDCl3/CD3OD ) 1/1) δ 6.78 (s, 1H), 6.73
(s, 1H), 6.61 (d, J ) 1.8 Hz, 1H), 6.46 (d, J ) 1.8 Hz, 1H), 4.14
(t, J ) 8.5 Hz, 2H), 3.81 (s, 3H), 2.65 (t, J ) 6.4 Hz, 2H), 2.56 (s,
1
3H), 2.49 (s, 3H), 2.05 (m, 2H); H NMR (0.5 mg in 0.4 mL
mp 109-110 °C; IR 1654, 1601, 1391, 1338, 1156, 1078 cm-1
;
CD3OD, 300 MHz) δ 6.90 (s, 1H), 6.83 (s, 1H), 6.61 (d, J ) 1.8
Hz, 1H), 6.49 (d, J ) 1.8 Hz, 1H), 4.17 (t, J ) 8.1 Hz, 2H), 3.72
(s, 3H), 2.60 (t, J ) 6.6 Hz, 2H), 2.54 (s, 3H), 2.45 (s, 3H),
1.99-2.05 (m, 2H); 13C NMR (125 MHz, CDCl3/CD3OD ) 1/1)
δ 20.4, 21.3, 24.2, 30.5, 47.9, 52.6, 97.0, 106.2, 108.7, 109.0, 116.8,
136.4, 140.9, 148.2, 157.0, 167.4, 174.5, 177.9; 13C NMR (10.0
mg in 0.4 mL CD3OD, 100 MHz) δ 20.1, 20.8, 24.1, 30.6, 48.3,
52.3, 97.5, 105.6, 108.5, 109.1, 116.9, 137.0, 142.1, 148.8, 157.2,
168.0, 174.9, 176.8; ESI-MS (m/z) 314 [M + H+]; HRMS (ESI)
calcd for C18H20NO4 [M + H+] 314.1387, found 314.1389.
1H NMR (300 MHz, CDCl3) δ 7.08 (d, J ) 1.8 Hz, 1H), 6.93 (d,
J ) 2.1 Hz, 1H), 6.03 (s, 1H), 5.21 (s, 2H), 3.47 (s, 3H), 2.28 (s,
3H), 2.16 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 5.0, 20.0, 56.4,
78.6, 92.6, 94.3, 103.3, 111.2, 118.6, 120.7, 124.3, 158.5, 159.7,
164.2, 177.1; ESI-MS (m/z) 259 [M + H+]. Anal. Calcd for
C15H14O4: C, 69.76; H, 5.46. Found: C, 69.41; H, 5.52.
Cassiarin A (1) (from 6b). To a solution of 6b (105 mg, 0.415
mmol) and AgNO3 (3.5 mg, 5 mol %) in (CH2Cl)2 (5 mL) was
added dropwise TFA (92 µL, 1.25 mmol) at -24 °C under nitrogen
atmosphere. The resulting orange solution was stirred for 1 h at
this temperature. After removal of the cooling bath, anhydrous
ammonia was bubbled into the reaction mixture for 20-30 min,
and the whole mixture was stirred overnight at room temperature.
The solvent was removed, and 2 N aq HCl (3 mL) and MeOH (7
mL) were added. The mixture was refluxed for 30 min. The reaction
was cooled to room temperature, and MeOH was removed in vacuo.
The aqueous layer was adjusted to pH 9 with 2 N aq Na2CO3 and
then extracted with a mixture of CH2Cl2 and MeOH (20/1). The
combined organic phases were washed with water and brine, dried
over anhydrous Na2SO4, and concentrated. The crude product was
purified by flash chromatography (CH2Cl2/MeOH ) 30/1) to give
1 (52 mg, 60%): mp 280-282 °C; IR 1663, 1618, 1567, 1394,
Acknowledgment. This work is financially supported by
NSFC (20672128, 20432020, 20425205, and 20621062), the
Chinese Academy of Sciences (KJCX2-YW-H08), and the
Shanghai Municipal Committee of Science and Technology
(06QH14016). We thank Prof. H. Morita for providing the
1
copies of H and 13C NMR spectra of natural cassiarins in
CD3OD.
Supporting Information Available: Experimental proce-
dures and characterization data; 1H NMR and 13C NMR spectra
of compounds 6b, 6a, 5, 3b, 3a, 6c, 1c, 1d, 6e, 1e, 1, and 2; 1H
NMR spectra of compounds 10, 11, 7, and 6d; NOESY, HMBC,
HMQC, and H-H COSY spectra of synthetic cassiarin A (1);
NOESY and HMBC spectra of synthetic cassiarin B (2); and
1H NMR spectral comparison of cassiarin B (2). This material
is available free of charge via the Internet at http://pubs.acs.org.
1
1189, 1170 cm-1; H NMR (500 MHz, CDCl3/CD3OD ) 1/1) δ
6.71 (s, 1H), 6.48 (d, J ) 1.9 Hz, 1H), 6.46 (d, J ) 1.9 Hz, 1H),
1
6.03 (s, 1H), 2.35 (s, 3H), 2.20 (s, 3H); H NMR (CD3OD, 500
MHz) δ 6.79 (s, 1H), 6.50 (s, 2H), 6.09 (d, J ) 1.0 Hz, 1H), 2.38
(s, 3H), 2.25 (d, J ) 1.0 Hz, 3H); 13C NMR (125 MHz, CDCl3/
CD3OD ) 1/1) δ 20.0, 22.9, 100.8, 102.9, 103.9, 111.7, 113.7,
139.0, 149.9, 150.8, 156.4, 161.4, 164.6; 13C NMR (CD3OD, 125
JO801017B
J. Org. Chem. Vol. 73, No. 14, 2008 5225