Isolation and synthesis of antiproliferative eupolauridine alkaloids of Ambavia gerrardii from the Madagascar dry forest

Article abstract of DOI:10.1021/np200093n

Investigation of the Madagascan endemic plant Ambavia gerrardii for antiproliferative activity against the A2780 ovarian cancer cell line led to the isolation of the three new alkaloids 8-hydroxyeupolauridine (1), 9-methoxyeupolauridine 1-oxide (2), and 1

Full text of DOI:10.1021/np200093n

ARTICLE
pubs.acs.org/jnp
Isolation and Synthesis of Antiproliferative Eupolauridine Alkaloids of
Ambavia gerrardii from the Madagascar Dry Forest1
Ende Pan,† Shugeng Cao,†,‡ Peggy J. Brodie,† Martin W. Callmander,§ Richard Randrianaivo,§
Stephan Rakotonandrasana,^ Etienne Rakotobe,^ Vincent E. Rasamison,^ Karen TenDyke,||
Yongchun Shen,|| Edward M. Suh,|| and David G. I. Kingston*,†
†Department of Chemistry, M/C 0212, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061-0212,
United States
§Missouri Botanical Garden, B.P 3391, Antananarivo 101, Madagascar
^Centre National d’Application des Recherches Pharmaceutiques, B.P 702, Antananarivo 101, Madagascar
Next Generation Systems, Eisai Inc., 4 Corporate Drive, Andover, Massachusetts 01810, United States
S Supporting Information
b
ABSTRACT: Investigation of the Madagascan endemic plant Ambavia gerrardii for
antiproliferative activity against the A2780 ovarian cancer cell line led to the isolation
of the three new alkaloids 8-hydroxyeupolauridine (1), 9-methoxyeupolauridine
1-oxide (2), and 11-methoxysampangine (3) and the three known alkaloids 4ꢀ6.
The structures of 1 and 2 were confirmed by synthesis. Compounds 3, 4, and 6 showed
moderate to good antiproliferative activities, with IC50 values of 10.3, 3.5, and 0.60
μM, respectively, against the A2780 human ovarian cancer cell line and with IC50
values of 0.57, 1.77, and 0.58 μM, respectively, against the H460 human lung cancer
cell line.
n our continuing search for biologically active natural products
from tropical forests, as part of an International Cooperative
basis of their UV, IR, HRESIMS, and 1H NMR spectral data, but
the limited samples available combined with the absence of 13C
NMR data and HMBC correlations made it necessary to confirm
the structures by synthesis. The synthesis also provided material
for the evaluation of the biological activities of 1 and 2. Herein we
report the isolation and structure elucidation of compounds
1ꢀ3, the synthesis of compounds 1 and 2, and the antiprolifera-
tive activities of all isolates.
Eupolauridine N-oxide (4),11 eupolauridine (5),11 and sam-
pangine (6)12ꢀ14 were all obtained as yellow solids. Eupolaur-
idine (5) was previously isolated from the annonaceous plant
Cananga odorata (ylang ylang) collected in Madagascar15 and
from the eupomatiaceous plant Eupomatia laurina.16
I
Biodiversity Group (ICBG) program, we obtained an EtOH
extract from the roots of a plant identified as Ambavia gerrardii
(Baill.) Le Thomas (Annonaceae) from Madagascar. The extract
exhibited good antiproliferative activity against the A2780 human
ovarian cancer cell line with an IC50 value of 8.2 μg/mL. The
genus Ambavia consists of only two species, A. capuronii and A.
gerrardii, both of which are endemic to Madagascar, and no
phytochemical work has been reported on either species. Pre-
vious phytochemical studies of plant species in the Annonaceae
family have revealed the presence of cytotoxic acetogenins,2,3
miliusanes,4 styrylpyrones,5 polyacetylenes,6 diterpenoids,7,8
flavanones,9 and alkaloids.6,10 The extract was selected for
bioassay-directed fractionation to isolate its active components
on the basis of its activity and the absence of previous phyto-
chemical studies.
Compound 1 was obtained as a yellow solid. Its UV absorp-
tions in MeOH, with λmax (log ε) 220 (4.15), 240 (3.98), 287
(3.82), 351 (3.28), and 369 (3.35) nm, indicated the presence of
an extended aromatic chromophore similar to that of eupolaur-
idine 5. The IR spectroscopic data of compound 1, which showed
absorptions at 1599, 1580, 1398, 1378, 1202, 842, and 808 cmꢀ1
,
’ RESULTS AND DISCUSSION
confirmed the existence of the aromatic and CꢀO functions. The
positive ion HRESIMS of 1 revealed a pseudomolecular ion peak
at m/z 221.0713 [M þ H]þ corresponding to a molecular
NMR signals in CD3OD exhibited AB, A0B0, andA00B00X00 multiplets
at low field [δH 8.64 (1H, d, J = 6.1 Hz), 8.55 (1H, d, J = 6.1 Hz),
An EtOH extract of the roots of A. gerrardii was subjected to
liquidꢀliquid partitioning between hexanes, CH2Cl2, andMeOHto
give fractions with IC50 values of 14, 2.6, and >100 μg/mL,
respectively, in the A2780 assay. Fractionation of the active
CH2Cl2 fraction by C18 open column and high-performance
liquid chromatography (HPLC) yielded two new eupolauridine
derivatives (1 and 2), the new 11-methoxysampangine (3), and
eupolauridine N-oxide (4), eupolauridine (5), and sampangine
(6). The structures of compounds 1 and 2 were proposed on the
1
formula of C14H10N2O (calcd for C14H9N2O: 221.0715). Its H
Received: January 29, 2011
Published: April 19, 2011
r
Copyright
2011 American Chemical Society and
1169
dx.doi.org/10.1021/np200093n J. Nat. Prod. 2011, 74, 1169–1174
American Society of Pharmacognosy
|

Journal of Natural Products
ARTICLE
5-methoxy-1,3-indanedione as the starting material instead of the
1,3-indanedione of the published method. The key step to afford
7-methoxyonychin (9) involved the thermal rearrangement of an
oxime O-crotyl ether, and this gave a very low yield of product
(<10%) in our hands. We thus changed the synthetic strategy by
modifying Bracher’s synthesis of eupolauridine (5) (Scheme 1).18
Ethyl 3-(4-methoxyphenyl)-3-oxopropionate was deprotonated
by NaH, then underwent Michael addition with crotonaldehyde
to afford the intermediate R-(1-methyl-3-oxopropyl)-β-oxo-4-
methoxylbenzenepropanoic acetate, which reacted with hydro-
xylamine hydrochloride to give ethyl 2-(4-methoxy)phenyl-4-
methyl-3-pyridinecarboxylate (7). Although polyphosphoric
acid was reported to catalyze the formation of onychin from
ethyl 2-phenyl-4-methyl-3-pyridinecarboxylate in good yield,18 a
very low yield of product was obtained when 7 was used as a
substrate in the study. Compound 7 was thus hydrolyzed to the
corresponding nicotinic acid 8, which was then converted to its
acyl chloride by treatment with thionyl chloride. A Friedelꢀ
Crafts reaction of the acyl chloride in chlorobenzene under reflux
gave 7-methoxyonychin (9). The acidic methyl group of 9
reacted with dimethylformamide diethyl acetal to afford an
enamine intermediate, which yielded 8-methoxyeupolauridine
(10) through a ring-closing reaction of the intermediate in the
presence of ammonium acetate at 140 ꢀC. Cleavage of the methyl
ether of 10 in 48% HBr gave 8-hydroxyeupolauridine (1) in 1.7%
overall yield from 7. Comparison of the 1H NMR (Table 1) and
HRESIMS data of 8-hydroxyeupolauridine with the data of
compound 1 confirmed the proposed structure of the isolated
natural product.
7.78 (1H, d, J = 8.2 Hz), 7.61 (1H, d, J = 6.1 Hz), 7.51 (1H, d, J = 6.1
Hz), 7.42 (1H, d, J = 2.3 Hz), 6.86 (1H, dd, J = 8.2, 2.3 Hz)]. The
above data together with the fact that 1 and 5 (C14H10N2) had the
same unsaturation number (12) suggested that compound 1
was a hydroxylated eupolauridine. The position of the OH
group was assigned at C-8 from the coupling patterns of the
protons of the benzene ring of 1 [δH 7.78 (1H, d, J = 8.2 Hz),
6.86 (1H, dd, J = 8.2, 2.3 Hz), and 7.42 (1H, d, J = 2.3 Hz)] and
comparison of the 1H NMR spectroscopic data of 1 with those
of 5 [δH 8.71 (2H, d, J = 6.0 Hz), 8.01 (2H, m), 7.67 (2H, d, J =
6.0 Hz), 7.55 (2H, m)]. On the basis of the above data, we
proposed the structure of 8-hydroxyeupolauridine for com-
pound 1.
Compound 2 was also obtained as a yellow solid. It showed
strong UV absorptions (MeOH) at λmax (log ε) 225 (4.03), 254
(3.99), 292 (3.87), and 384 (3.36) nm. Its IR spectrum displayed
aromatic absorptions at 1611, 1593, 1569, 1451, 1423, 846, and
828 cmꢀ1, a CꢀO stretch absorption at 1022 cmꢀ1, and NꢀO
stretch absorptions at 1487 and 1380 cmꢀ1. The characteristic
alkyl CꢀH stretch absorptions were also observed at 2919 and
2851 cmꢀ1. The positive ion HRESIMS of 2 displayed a
pseudomolecular ion peak at m/z 251.0820 [M þ H]þ, corres-
ponding to a molecular formula of C15H11N2O2 (calcd for
C15H11N2O2: 251.0821). Its proton spectrum [AB, A0B0, and
A00B00X00 aromatic coupling pattern and one OCH3 group: δH
8.58 (1H, d, J = 6.0 Hz), 8.23 (1H, d, J = 7.1 Hz), 7.95 (1H, d,
J = 8.2 Hz), 7.92 (1H, d, J = 2.4 Hz), 7.79 (1H, d, J = 7.1 Hz),
7.57 (1H, d, J = 6.0 Hz), 7.09 (1H, dd, J = 8.2, 2.4 Hz)]
was very similar to that of 1 except for the presence of a
signal for an additional methyl group at δH 3.96 (3H, s). The
above spectroscopic data thus suggested that 2 was either
8-methoxyeupolauridine N1-oxide or 9-methoxyeupolauri-
dine N1-oxide, but the available data did not permit a distinc-
tion between these two structures. It was thus necessary to
synthesize 1 and 2 to confirm the structure of 1, to provide
additional material for bioassay, and to determine the position
of the N-oxide in 2.
Mono-oxidation of 8-methoxyeupolauridine (10) by one
equivalent of meta-chloroperoxybenzoic acid yielded the two
methoxyeupolauridine N-oxides in a 3.7:1 ratio in 2.7% overall
1
yield; the H NMR data for the major product matched the
corresponding data of the natural product 2 and differed from
those of the isomeric product 11. Comparison of the 1H NMR
data of 2 and 11 (Table 1) indicated that the proton signals of
H-10 for 2 [δH 7.95 (d, J = 8.2 Hz)] were less deshielded than
those of 11 [δH 8.21 (d, J = 8.4 Hz)]. On the other hand the
signals for H-7 of 2 [δH 7.92 (d, J = 2.4 Hz)] were more
deshielded than those of 11 [δH 7.61 (d, J = 2.3 Hz)]. These
differences enabled the structures of 2 and 11 to be assigned.
The larger deshielding of H-10 of 11 and of H-7 of 2 was
consistent with the major resonance structures of each com-
pound. For compound 2 the favored resonance structure is
expected to be the one that puts the negative charge on the
positively charged oxygenated nitrogen (structure 2a) rather
than a structure such as 2b, where the charge ends up on the
neutral nitrogen atom; this means that H-7 will be preferen-
tially deshielded. The situation is reversed for structure 11,
where the most stable resonance structure 11a results in
deshielding of H-10 and the less favored resonance structure
11b is the one that deshields H-7 (Figure 1). HMBC correla-
tions of 2 from both downfield shifted H-5 (δH 8.23) and H-7
(δH 7.92) to C-6a (δC 144.0) confirmed it to be 9-methox-
yeupolauridine-1-oxide.
11-Methoxysampangine (3) was obtained as a yellow solid
and had a molecular mass of 263.1, consistent with the
1
composition C16H11N2O2. Its H NMR signals in CDCl3
exhibited AB, A0B0, and A00B00C00 multiplets at low field: δH
8.88 (1H, d, J = 5.8 Hz), 7.71 (1H, d, J = 5.8 Hz), 7.86 (1H, d,
J = 5.8 Hz), 9.11 (1H, d, J = 5.8 Hz), 7.26 (1H, d, J = 8.0 Hz),
7.77 (1H, t, J = 8.0 Hz), and 8.56 (1H, d, J = 8.0 Hz). A triplet at
The synthesis of 8-hydroxyeupolauridine (1) initially fol-
lowed Wong’s method for synthesizing eupolauridine,17 using
1170
dx.doi.org/10.1021/np200093n |J. Nat. Prod. 2011, 74, 1169–1174

Journal of Natural Products
ARTICLE
Scheme 1. Synthesis of 1, 2, and 11
1
Table 1. H and 13C NMR Spectra of Compounds 1ꢀ3 and 1H NMR Spectra of Compounds 10 and 11a
1
2
3
10
11
position
1H (J, Hz)
13Cb
1H (J, Hz)
13Cb
1H (J, Hz)c
1H (J, Hz)
13C
1H (J, Hz)
1H (J, Hz)
2
8.55 d (6.1)
7.61 d (6.1)
150.3
117.9
137.0
122.7
119.7
150.5
163.1
143.0
112.3
8.58 d (6.0)
7.57 d (6.0)
149.6
117.8
130.4
125.0
124.4
145.8
144.0
136.3
113.5
8.88 d (5.8)
7.71 d (5.8)
8.87 d (5.7)
7.90 d (5.7)
148.9
120.6
140.6
8.57 d (5.7)
7.28 d (6.0)
8.16 d (7.1)
7.72 d (7.1)
3
3a
3b
4
7.51 d (6.1)
8.64 d (6.1)
7.79 d (7.1)
8.23 d (7.1)
7.86 d (5.8)
9.11 d (5.8)
8.12 d (5.7)
8.99 d (5.7)
125.1
148.9
149.5
7.35 d (5.7)
8.63 d (5.7)
7.65 d (5.8)
8.65 d (5.8)
5
6a
6b
7
7.42 d (2.3)
7.92 d (2.4)
182.1
139.2
119.0
137.4
116.7
7.49 d (1.4)
7.61 d (2.3)
7a
8
164.2
118.2
125.5
130.9
163.6
164.5
117.1
125.4
130.7
159.6
8.56 d (8.0)
7.77 t (8.0)
7.26 d (8.0)
8.52 d (8.0)
7.83 t (8.0)
7.42 d (8.0)
9
6.86 dd (8.2, 2.3)
7.78 d (8.2)
7.09 dd (8.2, 2.4)
7.95 d (8.2)
6.89 dd (8.2, 1.4)
7.81 d (8.2)
7.07 dd (8.4, 2.3)
8.21 d (8.4)
10
10a
10b
11
11a
11b
11c
OCH3
163.5
121.7
152.4
119.0
56.8
3.96 s
56.5
4.09
4.05
3.90 s
3.96 s
a In CD3OD unless otherwise stated: δ (ppm); multiplicities; J values (Hz) in parentheses. b The 13C NMR data on 1 and 2 were obtained on the
synthetic samples. c In CDCl3, δ (ppm); multiplicities; J values (Hz) in parentheses.
δH 7.77 with a coupling constant of 8.0 Hz indicated the presence
of a contiguous group of three aromatic protons and limited the
possible structures to either 8-methoxysampangine (12) or 11-
methoxysampangine (3). The 8-methoxysampangine structure
12 was initially favored, based on the absence of a 3J HMBC
correlation from the H-8 aromatic proton to the C-7 carbonyl
carbon, but a comparison of its 1H NMR spectrum with that of
synthetic 8-methoxysampangine prepared by Zjawiony et al.19
1171
dx.doi.org/10.1021/np200093n |J. Nat. Prod. 2011, 74, 1169–1174

Journal of Natural Products
ARTICLE
Figure 1. Favored resonance structures of 2 and 11.
indicated that the two compounds were not identical.
Although a sample for direct comparison was not available,
Dr. Zjawiony confirmed the structure of his synthetic com-
pound by a comparison of the NMR data of his intermediates
with data in the Supporting Information of a recent paper.20,21
These facts exclude the 8-methoxysampangine structure, and
so the structure of compound 3 is assigned as 11-methox-
ysampangine. The lack of sample regrettably precluded addi-
tional experiments to confirm this assignment.
The biosyntheses of these two types of alkaloids have been
proposed by Taylor.22 It was reported that sampangine (6) could
derive from a dihydroxyoxoaporphine. Then eupolauridine (5)
could arise from sampangine (6) through a ring-contraction
pathway with concurrent loss of carbon monoxide.
Previous research has shown eupolauridine to have antifungal
activity23 and sampangine derivatives to have cytotoxic, antima-
larial, and antifungal activities.24 8-Hydroxyeupolauridine (1),
9-methoxyeupolauridine 1-oxide (2), 8-methoxyeupolauridine
(10), and 8-methoxyeupolauridine 1-oxide (11), as well as 11-
methoxysampangine (3), eupolauridine N-oxide (4), eupolaur-
idine (5), and sampangine (6) were tested against the A2780
human ovarian cancer cell line. Among these eight compounds,
sampangine (6) was the most active against the A2780 cell line,
with an IC50 value of 0.60 μM, but sampangine derivative 3
showed much weaker activity, with an IC50 of 10.3 μM. Among
the eupolauridine analogues, compound 4 was the most active,
with an IC50 value of 3.5 μM.
When tested against the H460 human non-small-cell lung
cancer cell line, compounds 3 and 6 both exhibited strong
activity, with IC50 values of 0.57 and 0.58 μM, respectively.
Interestingly, only 4 among the eupolauridine analogues showed
antiproliferative activity against this cell line, with an IC50 value of
1.77 μM.
were obtained on an Agilent 6220 TOF mass spectrometer. HPLC was
performed on a Shimadzu LC-10AT instrument with a semipreparative
C18 or phenyl Varian Dynamax column (5 μm, 250 ꢁ 10 mm). Unless
otherwise noted, chemicals were obtained from commercial suppliers
and used without purification. Solvents were dried if necessary by
standard methods. All reactions were performed under air atmosphere
unless otherwise noted.
Plant Material. Roots of Ambavia gerradii were collected on July 15,
2005, in the Ambohibe dry forest near the village of Betsimiranja, Diana,
Antsiranana, Madagascar. The collection coordinates were 13ꢀ0204200 S,
049ꢀ0901100 E, and the elevation was 50 m. The plant sampled was a tree
of about 16 m height with a diameter at breast height of 35 cm, and with
green fruit; it occurs commonly in this area. The tree was identified by
one of the authors (R.R.), and its identity was confirmed by G. E. Schatz
(Missouri Botanical Garden). Voucher specimens with colllection
number Randrianaivo & al. 1196 have been deposited in herbaria at
the Parc Botanique and Zoologique de Tsimbazaza (TAN), at the
Centre National d’Application des Recherches Pharmaceutiques in
Antananarivo, Madagascar (CNARP), at the Missouri Botanical Garden
in St. Louis, Missouri (MO), and at the Musꢀeum National d’Histoire
Naturelle in Paris, France (P).
Extraction and Isolation. Dried roots of A. gerrardii (250 g) were
ground in a hammer mill, then extracted with EtOH by percolation for
24 h at room temperature to give the crude extract MG 3311 (6.4 g), of
which 3.0 g was available at Virginia Polytechnic Institute and State
University (VPISU) for evaluation. The extract MG 3311 (IC50 8.2 μg/mL,
1.3 g) was suspended in aqueous MeOH (MeOHꢀH2O, 9:1, 100 mL)
and extracted with hexanes (3 ꢁ 100 mL portions). The aqueous layer
was then diluted to 60% MeOH (v/v) with H2O and extracted with
CH2Cl2 (3 ꢁ 150 mL portions). The hexanes extract was evaporated in
vacuo to leave 138 mg with an IC50 value of 14 μg/mL. The residue from
the CH2Cl2 extract (354 mg) had an IC50 value of 2.6 μg/mL. The
aqueous MeOH extract (715 mg) was inactive. The CH2Cl2 extract was
selected for fractionation, and five fractions of 124, 44, 81, 26, and 38 mg
were collected from a C18 open column eluted with MeOHꢀH2O
(gradient from 60% to 100%). The first four fractions had IC50 values of
3.7, 1.9, 3, and 16 μg/mL, respectively, and the last fraction was inactive.
Fractions I and II were selected for further work. Separation of fraction I
by C-18 preparative HPLC (60% MeOHꢀH2O) yielded eight subfrac-
tions (IC50 19 μg/mL, inactive, inactive, 3.2 μg/mL, inactive, 2 μg/mL,
4.9 μg/mL, and inactive), and the most active subfractions, I-4 (IC50 3.2
μg/mL) and I-6 (IC50 2 μg/mL), were selected for further separation by
phenyl HPLC (60% MeOHꢀH2O). Compounds 1 (0.4 mg, tR 28.1
min) and 4 (2.0 mg tR 31.4 min) were isolated from subfraction I-4, and
’ EXPERIMENTAL SECTION
General Experimental Procedures. UV was measured on a
Shimadzu UV-1201 spectrophotometer. IR spectra of solid samples
were obtained directly on a MIDAC M-series FTIR spectrophotometer.
Melting points were obtained on a B-540 B€uchi melting-point apparatus.
NMR spectra were recorded in CDCl3, CD3OD, or D2O on either JEOL
Eclipse 500 or Bruker 600 spectrometers. The chemical shifts are given
in δ (ppm), and coupling constants (J) are reported in Hz. Mass spectra
1172
dx.doi.org/10.1021/np200093n |J. Nat. Prod. 2011, 74, 1169–1174

Journal of Natural Products
ARTICLE
compounds 2 (0.5 mg, tR 29.8 min) and 3 (1.0 mg, tR 36.0 min) were
isolated from subfraction I-6. Fraction II was applied to a C18 open
column. The most active subfraction (IC50 0.14 μg/mL) was subjected
to phenyl HPLC (70% MeOHꢀH2O) to give compounds 5 (1.9 mg, tR
25.3 min) and 6 (0.6 mg, tR 31.4 min).
Table 2. Antiproliferative Activities of Compounds 1ꢀ6, 10,
and 11 against the A2780 and H460 Cancer Cell Lines
IC50 (μM)
cancer
8-Hydroxyeupolauridine (1): yellow solid; UV (MeOH) λmax nm
(log ε) 220 (4.15), 240 (3.98), 287 (3.82), 351 (3.28), 369 (3.35); IR
νmax cmꢀ1 1638, 1599, 1580, 1398, 1378, 1289, 1243, 1202, 1091, 1060,
cell line
1
2
3
4
5
6
10
11
A2780
H460
12.7
54.3 10.3
>10
3.5
35.7 0.60
11.1
50.3
1
1016, 994, 842, and 808 cmꢀ1; H NMR (500 MHz, CD3OD), see
>10
0.57 1.77 >100
0.58 >10
>100
Table 1; HRESIMS m/z 221.0713 [M þ H]þ (calcd for C14H10N2O,
221.0715).
130.3, 126.7, 123.6, 115.0, 55.7, 49.0, 20.2; HRESIMS m/z 244.0958 [M
þ H]þ (calcd for C14H14NO3, 244.0974).
9-Methoxyeupolauridine 1-oxide (2): yellow solid; UV
(MeOH) λmax nm (log ε) 225 (4.03), 254 (3.99), 292 (3.87), 384
(3.36); IR νmax cmꢀ1 2919, 2851, 1611, 1593, 1487, 1451, 1423, 1380,
7-Methoxyonychin (9). A solution of 8 (93 mg, 0.38 mmol) in
2 mL of SOCl2 was refluxed 24 h under N2. After the solvent was
removed under vacuum, 51 mg (0.38 mmol) of AlCl3 and 2 mL of
chlorobenzene were added to the flask, and the mixture was refluxed
overnight under N2. The reaction was cooled and quenched by pouring
into 10 mL of saturated NaHCO3 solution. The resulting mixture was
extracted with CH2Cl2 (3 ꢁ 10 mL) and dried over K2CO3, and the
CH2Cl2 phase was evaporated. The residue was purified by silica gel
PTLC (hexanesꢀEtOAc, 6:4) to afford 7-methoxyonychin (9) (21 mg,
25%): yellow solid, mp 134ꢀ136 ꢀC; 1H NMR (500 MHz, CDCl3) δH
8.30 (1H, d, J = 5.2 Hz), 7.79 (1H, d, J = 8.2 Hz), 7.18 (1H, d, J = 2.3 Hz),
7.04 (1H, dd, J = 8.2, 2.3 Hz), 6.90 (1H, d, J = 5.2 Hz), 3.87 (3H, s), 2.59
(3H, s); 13C NMR (125 MHz, CDCl3) δC 192.4, 164.8, 162.5, 151.3,
148.2, 136.9, 134.5, 126.1, 125.0, 122.7, 120.4, 109.0, 55.8, 17.3;
HRESIMS m/z 226.0865 [M þ H]þ (calcd for C14H12NO2, 226.0868).
8-Methoxyeupolauridine (10). A solution of 33 mg (0.15 mmol)
of 9 plus N,N-dimethylformamide diethyl acetal in 1 mL of DMF was
stirred 2 h at 120 ꢀC under N2. NH4OAc (600 mg) was added to the
flask, the reaction mixture was stirred for another 30 min at 140 ꢀC and
quenched with 5 mL of water, and the solvent was extracted with EtOAc
(3 ꢁ 5 mL). The organic phase was combined, dried over K2CO3, and
concentrated to a residue, which was separated on a silica TLC plate
(hexanesꢀEtOAc, 6:4) to give 8-methoxyeupolauridine (10) (25.4 mg,
74%): yellow solid, mp 138ꢀ140 ꢀC; 1H NMR (500 MHz, CDCl3) δH
8.63 (1H, d, J = 5.7 Hz), 8.57 (1H, d, J = 5.7 Hz), 7.81 (1H, d, J = 8.2 Hz),
7.49 (1H, d, J = 1.4 Hz), 7.35 (1H, d, J = 5.7 Hz), 7.28 (1H, d, J = 5.7 Hz),
6.89 (1H, dd, J = 8.2, 1.4 Hz), 3.89 (3H, s); 13C NMR (125 MHz,
CDCl3) δC 162.8, 162.5, 162.2, 149.7, 149.6, 142.0, 134.9, 131.9, 123.7,
121.3, 117.8, 116.3, 115.5, 109.0, 55.7; HRESIMS m/z 235.0875 [M þ
H]þ (calcd for C15H11N2O, 235.0871).
1256, 1233, 1022, 973, 846, and 828 cmꢀ1 1H NMR (500 MHz,
.
CD3OD) and 13C NMR (125 MHz, CD3OD), see Table 1; HRESIMS
m/z 251.0820 [M þ H]þ (calcd for C15H11N2O2, 251.0821).
11-Methoxysampangine (3): yellow solid; LC-MS m/z 263.1
[M þ H]þ (calcd for C16H11N2O2, 263.1); 1H NMR (500 MHz,
CDCl3), see Table 1.
Eupolauridine N-oxide (4): yellow solid; LC-MS m/z 221.1
[M þ H]þ (calcd for C14H9N2O, 221.1); 1H NMR (500 MHz, CDCl3),
see Supporting Information
Eupolauridine (5): yellow solid; LC-MS m/z 205.1 [M þ H]þ
(calcd for C14H9N2, 205.1); 1H NMR (500 MHz, CDCl3), see
Supporting Information
Sampangine (6): yellow solid; LC-MS m/z 233.0 [M þ H]þ
(calcd for C15H9N2O, 233.1); 1H NMR (500 MHz, CDCl3), see
Supporting Information
Ethyl 2-(4-methoxy)phenyl-4-methyl-3-pyridinecarboxy-
late (7). To a solution of 3.84 g (17 mmol) of ethyl 3-(4-meth-
oxyphenyl)-3-oxopropionate in 14 mL of dioxane in a 100 mL flask was
added 48 mg (2 mmol) of NaH and then 1.68 g (24 mmol) of
crotonaldehyde dropwise in 6 mL of dioxane. After the reaction mixture
was stirred for another 30 min at room temperature, 4.9 g of H2NOH
3
HCl (70 mmol) and 20 mL of glacial AcOH were added. The reaction
mixture was stirred at 100ꢀ110 ꢀC for 90 min and then poured onto 100
g of ice, made basic with K2CO3, and extracted with ether (3 ꢁ 100 mL).
The combined organic phase was extracted with 2 N HCl (3 ꢁ 100 mL).
The combined acidic aqueous phase was neutralized with K2CO3 and
extracted with ether (3 ꢁ 100 mL), and the ether was dried over K2CO3
and evaporated under reduced pressure. The oily crude product was
purified by flash silica column chromatography (hexaneꢀEtOAc, 4:1) to
afford ethyl 2-(4-methoxy)phenyl-4-methyl-3-pyridinecarboxylate (7)
(1.47 g, 32%): yellow oil; 1H NMR (500 MHz, CDCl3) δH 8.53 (1H, d,
J = 5.1 Hz), 7.54 (2H, d, J = 8.8 Hz), 7.08 (1H, d, J = 5.1 Hz), 6.94 (2H, d,
J = 8.8 Hz), 4.17 (2H, q, J = 7.2 Hz), 3.82 (3H, s), 2.39 (3H, s), 1.08 (3H,
t, J = 7.2 Hz); 13C NMR (125 MHz, CDCl3) δC 168.8, 160.0, 156.0,
149.4, 145.4, 132.4, 129.6, 128.9, 123.1, 113.7, 61.3, 55.2, 19.3, 13.7;
HRESIMS m/z 272.1268 [M þ H]þ (calcd for C16H18NO3, 272.1287).
2-(4-Methoxyphenyl)-4-methyl-3-pyridinecarboxylic acid
(8). Compound 7 (670 mg, 2.5 mmol) was refluxed overnight in
aqueous NaOH (40%, 10 mL). The solution was extracted with CHCl3
to remove any unreacted starting material. The pH value of the mixture
was then adjusted to about 6, the solvent was removed under reduced
pressure, and the residue was extracted with hot MeOH (3 ꢁ 10 mL).
The MeOH extracts were combined and concentrated to a residue,
which was purified on a silica column eluted with MeOHꢀCH2Cl2 (3:1)
to give 2-(4-methoxy)phenyl-4-methyl-3-pyridinecarboxylic acid (8)
(403 mg, 66%): off-white solid, mp 198ꢀ200 ꢀC; 1H NMR (500
MHz, D2O) δH 8.43 (1H, d, J = 6.2 Hz), 7.78 (1H, d, J = 6.2 Hz),
7.62 (2H, d, J = 9.0 Hz,), 7.13 (2H, d, J = 9.0 Hz), 3.89 (3H, s), 2.59 (3H,
s); 13C NMR (125 MHz, D2O) δC 171.6, 161.8, 155.4, 147.6, 139.1,
8-Hydroxyeupolauridine (1). A solution of 10 (6 mg,
0.026 mmol) in 1 mL of 48% HBr was refluxed for 24 h. The mixture
was then cooled and evaporated in vacuo to give a residue, which was
purified by PTLC on silica gel (hexanesꢀEtOAc, 6:4). 8-Hydro-
xyeupolauridine (1) (2.5 mg, 43%) was collected as a yellow solid,
mp 278ꢀ280 ꢀC. Its 1H NMR and HRESIMS spectra were identical
to those of the isolated material; 13C NMR, see Table 1.
8-Methoxyeupolauridine 1-oxide (11) and 9-methoxyeu-
polauridine 1-oxide (2). A solution of 8-methoxyeupolauridine (10,
10 mg, 0.045 mmol) and meta-chloroperoxybenzoic acid (7.4 mg,
0.043 mmol) in 1 mL of CH2Cl2 was stirred at room temperature for
24 h. After the solvent was removed, the residue was separated by C-18
HPLC (60% MeOHꢀH2O) to afford 9-methoxyeupolauridine 1-oxide
(2) (1.6 mg, 15%) (mp 198ꢀ200 ꢀC) and 8-methoxyeupolauridine
1-oxide (11) (5.9 mg, 55%). The NMR and mass spectroscopic data for
2 were identical to those of the natural product; 13C NMR, see Table 1.
8-Methoxyeupolauridine 1-oxide (11): yellow solid, mp 218ꢀ
220 ꢀC; 1H NMR (600 MHz, CD3OD) δH 8.65 (1H, d, J = 5.8 Hz), 8.21
(1H, d, J = 8.5 Hz), 8.16 (1H, d, J = 7.1 Hz), 7.72 (1H, d, J = 7.1 Hz), 7.65
(1H, d, J = 5.8 Hz), 7.61 (1H, d, J = 2.3 Hz), 7.07 (1H, dd, J = 8.4,
2.4 Hz), 3.96 (3H, s); 13C NMR (150 MHz, CD3OD) δC 164.6, 159.3,
1173
dx.doi.org/10.1021/np200093n |J. Nat. Prod. 2011, 74, 1169–1174

Journal of Natural Products
ARTICLE
149.8, 145.6, 144.5, 141.1, 130.7, 128.4, 126.3, 125.3, 123.1, 119.3, 116.5,
111.2, 56.5; HRESIMS m/z 251.0807 [M þ H]þ (calcd for C15H11N2O2,
251.0821).
Antiproliferative Bioassays. The A2780 ovarian cancer cell line
assay was performed at Virginia Polytechnic Institute and State Uni-
versity as previously reported,25 except that the samples were added in
1 μL of 100% DMSO per well instead of 20 μL of 1:1 DMSOꢀH2O.
The A2780 cell line is a drug-sensitive human ovarian cancer cell line.26
Assays against the H460 NSCLC lung cancer cell line were carried out at
Eisai, Inc., as previously described.1
’ REFERENCES
(1) Biodiversity Conservation and Drug Discovery in Madagascar,
Part 47. For Part 46, see: Pan, E.; Harinantenaina, L.; Brodie, P.;
Callmander, M; Rakotonandrasana, S.; Rakotobe, E.; Rasamison,
V. E.; TenDyke, K.; Shen, Y.; Suh, E. M.; Kingston, D. G. I. Bioorg.
Med. Chem. 2011, 19, 422–428.
(2) McLaughlin, J. L. J. Nat. Prod. 2008, 71, 1311–1321.
(3) Nakanishi, Y.; Chang, F. R.; Liaw, C. C.; Wu, Y. C.; Bastow, K. F.;
Lee, K. H. J. Med. Chem. 2003, 46, 3185–3188.
(4) Zhang, H. J.; Ma, C. Y.; Van Hung, N.; Cuong, N. M.; Tan, G. T.;
Santarsiero, B. D.; Mesecar, A. D.; Soejarto, D. D.; Pezzuto, J. M.; Fong,
H. H. S. J. Med. Chem. 2006, 49, 693–708.
(5) Wang, S.; Zhang, Y. J.; Chen, R. Y.; Yu, D. Q. J. Nat. Prod. 2002,
65, 835–841.
’ ASSOCIATED CONTENT
(6) Li, C.; Lee, D.; Graf, T. N.; Phifer, S. S.; Nakanishi, Y.; Riswan, S.;
Setyowati, F. M.; Saribi, A. M.; Soejarto, D. D.; Farnsworth, N. R.;
Falkinham, J. O.; Kroll, D. J.; Kinghorn, A. D.; Wani, M. C.; Oberlies,
N. H. J. Nat. Prod. 2009, 72, 1949–1953.
Supporting Information. 1H and 13C NMR spectra of 1,
S
b
2, and 7ꢀ11 and 1H NMR spectra of 3ꢀ6. These materials are
available free of charge via the Internet at http://pubs.acs.org.
(7) Li, C.; Lee, D.; Graf, T. N.; Phifer, S. S.; Nakanishi, Y.; Burgess,
J. P.; Riswan, S.; Setyowati, F. M.; Saribi, A. M.; Soejarto, D. D.;
Farnsworth, N. R.; Falkinham, J. O.; Kroll, D. J.; Kinghorn, A. D.; Wani,
M. C.; Oberlies, N. H. Org. Lett. 2005, 7, 5709–5712.
(8) Chang, F. R.; Hwang, T. L.; Yang, Y. L.; Li, C. E.; Wu, C. C.; Issa,
H. H.; Hsieh, W. B.; Wu, Y. C. Planta Med. 2006, 72, 1344–1347.
(9) Khamis, S.; Bibby, M. C.; Brown, J. E.; Cooper, P. A.; Scowen, I.;
Wright, C. W. Phytother. Res. 2004, 18, 507–510.
(10) Lekphrom, R.; Kanokmedhakul, S.; Kanokmedhakul, K.
J. Ethnopharmacol. 2009, 125, 47–50.
(11) Waterman, P. G.; Muhammad, I. Phytochemistry 1985, 24, 523–
527.
’ AUTHOR INFORMATION
Corresponding Author
*Tel: (540) 231-6570. Fax: (540) 231-3255. E-mail: dkingston@
vt.edu.
Present Addresses
‡Department of Biological Chemistry and Molecular Pharmacol-
ogy, Harvard Medical School, Boston, MA 02115, USA.
(12) Rao, J. U. M.; Giri, G. S.; Hanumaiah, T.; Rao, K. V. J. J. Nat.
Prod. 1986, 49, 346–347.
(13) Peterson, J. R.; Zjawiony, J. K.; Liu, S.; Hufford, C. D.; Clark,
A. M.; Rogers, R. D. J. Med. Chem. 1992, 35, 4069–4077.
(14) Orabi, K. Y.; Li, E.; Clark, A. M.; Hufford, C. D. J. Nat. Prod.
1999, 62, 988–992.
(15) Leboeuf, M.; Cave, A. Lloydia 1977, 39, 459–460.
(16) Bowden, B.; Picker, K.; Ritchie, E.; Taylor, W. Aust. J. Chem.
1975, 28, 2681–2701.
(17) Tong, T. H.; Wong, H. N. C. Synth. Commun. 1992, 22, 1773–
1782.
(18) Bracher, F. Arch. Pharm. 1989, 322, 293–294.
(19) Zjawiony, J. K.; Srivastava, A. R.; Hufford, C. D.; Clark, A. M.
Heterocycles 1994, 39, 779–800.
(20) Jiang, C.; Xu, M.; Wang, S.; Wang, H.; Yao, Z.-J. J. Org. Chem.
2010, 75, 4323–4325.
(21) Zjawiony, J. K. Personal communication. January 13, 2011.
(22) Taylor, W. C. Aust. J. Chem. 1984, 37, 1095–1104.
(23) Hufford, C. D.; Liu, S.; Clark, A. M.; Oguntimein, B. O. J. Nat.
Prod. 1987, 50, 961–964.
(24) Muhammad, I.; Dunbar, D. C.; Takamatsu, S.; Walker, L. A.;
Clark, A. M. J. Nat. Prod. 2001, 64, 559–562.
(25) Cao, S.; Brodie, P. J.; Miller, J. S.; Randrianaivo, R.; Ratovoson,
F.; Birkinshaw, C.; Andriantsiferana, R.; Rasamison, V. E.; Kingston,
D. G. I. J. Nat. Prod. 2007, 70, 679–681.
(26) Louie, K. G.; Behrens, B. C.; Kinsella, T. J.; Hamilton, T. C.;
Grotzinger, K. R.; McKoy, W. M.; Winker, M. A.; Ozols, R. F. Cancer Res.
1985, 45, 2110–2115.
’ ACKNOWLEDGMENT
This International Cooperative Biodiversity Group project
was supported by the Fogarty International Center, the National
Cancer Institute, the National Institute of Allergy and Infectious
Diseases, the National Institute of Mental Health, the National
Institute on Drug Abuse, the National Heart Lung and Blood
Institute, the National Center for Complementary and Alter-
native Medicine, the Office of Dietary Supplements, the National
Institute of General Medical Sciences, the Biological Sciences
Directorate of the National Science Foundation, and the Office
of Biological and Environmental Research of the U.S. Depart-
ment of Energy under Cooperative Agreement U01 TW00313
with the International Cooperative Biodiversity Groups. This
project was also supported by the National Research Initiative of
the Cooperative State Research, Education and Extension Ser-
vice, USDA, Grant #2008-35621-04732. These supports are
gratefully acknowledged. Work at Virginia Polytechnic Institute
and State University was supported by the National Science
Foundation under Grant CHE-0619382 for the purchase of the
Bruker Avance 600 NMR spectrometer and Grant CHE-
0722638 for the purchase of the Agilent 6220 mass spectrometer.
We thank Mr. B. Bebout for obtaining the mass spectra, and Prof.
J. K. Zjawiony, University of Misisippi, for consulting with us on
the structure of his synthetic compound 12. Fieldwork essential
for this project was conducted under a collaborative agreement
between the Missouri Botanical Garden and the Parc Botanique
et Zoologique de Tsimbazaza and a multilateral agreement between
the ICBG partners, including the Centre National d’Application
des Recherches Pharmaceutiques. We gratefully acknowledge
courtesies extended by the Government of Madagascar (Ministꢁere
des Eaux et For^ets). We thank S. Randrianasolo, R. Rakoton-
drajaona, R. L. Andriamiarisoa, V. Benjara, and M. Velonjara for
assistance with the plant collection.
1174
dx.doi.org/10.1021/np200093n |J. Nat. Prod. 2011, 74, 1169–1174