Constituents of the stems of Macrococculus pomiferus and their inhibitory activities against cyclooxygenases-1 and -2

www.elsevier.com/locate/phytochem

Phytochemistry 65 (2004) 2861–2866

PHYTOCHEMISTRY

Constituents of the stems of Macrococculus pomiferus andtheir inhibitory activities against cyclooxygenases-1 and -2

Bao-Ning Su a,1, William P. Jones a,1, Muriel Cuendet a, Leonardus B.S. Kardono b,Rachman Ismail c, Soedarsono Riswan c, Harry H.S. Fong a, Norman R. Farnsworth a,

John M. Pezzuto a,2, A. Douglas Kinghorn a,*,1

a Program for Collaborative Research in the Pharmaceutical Sciences and Department of Medicinal Chemistry and Pharmacognosy, College

of Pharmacy, University of Illinois at Chicago, Chicago, IL 60612, USAb Research Center for Chemistry, Indonesian Institute of Science, Serpong 15310, Tangerang, Indonesia, and Herbarium Bogoriense, Indonesia

c Research Center for Biology, Indonesian Institute of Science, 16122 Bogor, Indonesia

Received 31 March 2004; received in revised form 29 June 2004

Available online 23 September 2004

Dedicated to Prof. Dr. Kurt Hostettmann on the occasion of his 60th birthday

Abstract

As part of our program directed towards the discovery of new cancer chemopreventive agents from plants, the EtOAc-soluble

extract of the stems of M. pomiferus was found to inhibit the enzyme cyclooxygenase-2 (COX-2). Bioassay-directed fractionation of

this extract led to the isolation of two dibenzylbutyrolactone lignans, (8R,8 0R)-3 0-O-demethyl-5-hydroxymatairesinol (1) and

(8R,8 0R)-3 0-O-demethyl-5-methoxymatairesinol (2), as well as seven known compounds, (�)-5 0-methoxyyatein (3), blumenol A,

(�)-deoxypodophyllotoxin (anthricin), (�)-deoxypodorhizone, 2,6-dimethoxyhydroquinone, 4-hydroxybenzaldehyde, and b-sito-sterol glucoside. The structures of compounds 1 and 2 were determined using spectroscopic data (1D and 2D NMR, and HREIMS),

and the 8R and 8 0R absolute stereochemistry was established for both 1 and 2 on the basis of their CD spectra. All isolates obtained

in the present study were evaluated for their inhibitory effects with both COX-1 and -2. Of these, only 5 0-methoxyyatein (3) showed

weak activity against COX-2, while all other compounds isolated were inactive. The COX-2 inhibitory activity of the EtOAc extract

was also traced to the presence of several common fatty acids by LC–MS.

� 2004 Elsevier Ltd. All rights reserved.

Keywords: Macrococculus pomiferus; Menispermaceae; (8R,8 0R)-3 0-O-Demethyl-5-hydroxymatairesinol; (8R,80R)-3 0-O-Demethyl-5-methoxyma-

tairesinol; Cyclooxygenases-1 and -2; LC–MS analysis

0031-9422/$ - see front matter � 2004 Elsevier Ltd. All rights reserved.

doi:10.1016/j.phytochem.2004.08.007

* Corresponding author. Tel.: +1 614 247 8094; fax: +1 614 247

8081.

E-mail address: [email protected] (A. Douglas Kinghorn).1 Present address: College of Pharmacy, The Ohio State University,

Columbus, OH 43210, USA.2 Present address: Heine Pharmacy Building, Purdue University,

West Lafayette, IN 47907, USA.

1. Introduction

An increase in prostaglandin (PG) synthesis may

influence tumor growth in both humans and experimen-tal animals (Karmali, 1980), and numerous studies have

illustrated the effect of PG synthesis on carcinogen

metabolism, tumor cell proliferation, and metastatic

potential (Marnett, 1992; Levy, 1997). As a result, inhi-

bition of PG synthesis has been investigated as a means

of preventing tumor development (Levy, 1997). PGs

2862 B.-N. Su et al. / Phytochemistry 65 (2004) 2861–2866

produced by cyclooxygenases-1 and -2 (COX-1 and

COX-2) are represented by a large series of compounds

which enhance mainly cancer development and progres-

sion, acting as carcinogens or tumor promoters, with

profound effects on carcinogenesis (Lupulescu, 1996).

As part of our program directed towards the discoveryof new cancer chemopreventive agents from plants, the

EtOAc-soluble extract of the stems of Macrococculus

pomiferus Becc. (Menispermaceae) was found to inhibit

the enzyme cyclooxygenase-2 (COX-2). Bioassay-di-

rected fractionation of this extract led to the isolation

of two new dibenzylbutyrolactone lignans, (8R,8 0R)-3 0-

O-demethyl-5-hydroxymatairesinol (1) and (8R, 8 0R)-

3 0-O-demethyl-5-methoxymatairesinol (2), as well asseven known compounds, (�)-5 0-methoxyyatein (3),

blumenol A, (�)-deoxypodophyllotoxin (anthricin),

(�)-deoxypodorhizone, 2,6-dimethoxyhydroquinone, 4-

hydroxybenzaldehyde, and b-sitosterol glucoside. The

structures of compounds 1 and 2 were determined using

physical and spectroscopic methods. All isolates were

evaluated for their inhibitory effects against both

COX-1 and COX-2.

O

O

OHOMeRO

HOHO

H

H1

4

9

1'

5'

7' 9'

O

O

OMeOMeMeO

3

MeO

H

H1

4

9

1'

5'

7' 9'

1 R = H2 R = Me

O

OA

B6

2. Results and discussion

Compound 1 was isolated as a colorless oil, ½a�23D�27.5� (c 0.10, MeOH). A molecular formula ofC19H20O7 was determined for this compound from the

molecular ion peak at m/z 360.1220 [M]+ (calc. for

C19H20O7, 360.1209) obtained in the HREIMS. The1H NMR spectrum (Table 1) of compound 1 displayed

the characteristic signals of a 1,2,4-trisubstituted aro-

matic ring at dH 6.66 (1H, d, J = 8.0 Hz, H-5 0), 6.53

(1H, d, J = 2.0 Hz, H-2 0) and 6.41 (1H, dd, J = 8.0, 2.0

Hz, H-6 0), a 1,2,3,5-tetrasubstituted aromatic ring atdH 6.34 (1H, d, J = 2.0 Hz, H-6) and 6.25 (1H, d,

J = 2.0 Hz, H-2), an oxygenated methylene at dH 4.09

(1H, dd, J = 8.8, 7.4 Hz, H-9 0a) and 3.87 (1H, dd,

J = 8.8, 8.3 Hz, H-9 0b), and one methoxy group at dH3.80 (3H, s). In addition to the signal typical for a meth-

oxy substituent group, the 13C and DEPT 135 NMR

spectra (Table 1) of 1 exhibited the signals of two aro-

matic rings and six other carbons (dC 181.7, s, C-9;

72.9, t, C-9 0; 47.9, d, C-8; 42.5, d, C-8 0; 38.8, t, C-7 0;

and 35.6, t, C-7), comprising two C6–C3 units. In the

HMBC spectrum of 1, the correlations from the signals

of the oxygenated methylene at dH 4.09 (H-9 0a) and 3.87

(H-9 0b) to the carbonyl carbon at dC 181.7 (C-9) wereobserved. These 1D and 2D NMR spectral data sug-

gested that compound 1 is a dibenzylbutyrolactone lig-

nan (Ikeda et al., 1998). Further interpretation of the

HMBC spectrum of 1 revealed the correlations from

H-2 0, H-6 0 and H2-90 to C-7 0, and from both H-2 and

H-6 to C-7. These correlations indicated that rings A

and B were tri- and tetra-substituted, respectively. By

considering the determined molecular formula and theabove-mentioned 1D NMR spectroscopic data, com-

pound 1 was assigned with four hydroxyl groups and

one methoxy group. The location of the methoxy group

was determined as C-3 on the basis of the observed

HMBC correlations from dH 6.25 (H-2) and 3.80

(OMe) to dC 149.7 (C-3), which was confirmed by the

observed NOESY correlation between dH 3.80 (OMe)

and 6.25 (H-2). The further consideration of the struc-ture of 1 required comparison with the structure of the

closely related compound 2.

Compound 2 was also isolated as colorless oil, ½a�23D�6.0� (c 0.12, MeOH). A molecular formula of

C20H22O7 was provided from the molecular ion peak

at m/z 374.1389 [M]+ (calc. for C20H22O7, 374.1365)

in the HREIMS of 2. Both the 1H and 13C NMR

spectra of compound 2 (Table 1) were closely compa-rable to those of 1, and suggested this compound is

also a dibenzylbutyrolactone lignan. In the 1H NMR

spectrum of 2, the signals for a 1,2,4-trisubstituted

aromatic ring at dH 6.67 (1H, d, J = 8.0 Hz, H-5 0),

6.54 (1H, d, J = 2.0 Hz, H-2 0), and 6.42 (1H, dd,

J = 8.0, 2.0 Hz, H-6 0) were evident, similar to those

in 1. However, a two-proton aromatic singlet at dH6.38 (H-2 and H-6) and a six-proton methoxy singletat dH 3.80 in the 1H NMR spectrum of 2 replaced

the two meta-coupled aromatic signals and the three-

proton methoxy singlet of the unsymmetrical 1,2,3,5-

tetrasubstituted aromatic ring of 1. In the 13C and

DEPT NMR spectra of 2, the signals of an aromatic

methine at dC 107.8 (C-2 and C-6) and a methoxy

group at dC 56.8 were more intense than the reso-

nances for the other methines and methylenes. These1D NMR observations suggested the presence of a

symmetrical 1,2,3,5-tetrasubstituted aromatic ring in

the molecule of compound 2. This, in combination

with the determined molecular formula, suggested that

a methoxy group occurs at C-5 in 2 instead of the

hydroxy group as in 1. The structure of 2 was sup-

ported by the observed HMBC correlations from dH6.38 (H-2 and H-6) and 3.80 (OMe) to dC 149.3 (C-3 and C-5) and the NOESY correlation between dH3.80 (OMe) and 6.38 (H-2 and H-6).

Table 11H and 13C NMR spectroscopic data of compounds 1 and 2 in CD3ODa

Position 1 2

dH dC dH dC

1 130.1 s 130.0 s

2 6.25, d (2.0) 105.9 d 6.38 s 107.8 d

3 149.7 s 149.3 s

4 134.0 s 135.4 s

5 146.6 s 149.3 s

6 6.34, d (2.0) 111.1 d 6.38, s 107.8 d

7 2.80, br d (5.9) 35.6 t 2.86, dd (12.2, 6.4) 35.6 t

2.81, dd (12.2, 6.4)

8 2.63, td (8.7, 5.9) 47.9 d 2.68, td (8.2, 6.4) 48.0 d

9 181.7 s 181.7 s

1 0 131.6 s 131.6 s

2 0 6.53, d (2.0) 116.9 d 6.54, d (2.0) 117.0 d

3 0 146.4 s 146.5 s

4 0 145.0 s 145.1 s

5 0 6.66, d (8.0) 116.5 d 6.67, d (8.0) 116.5 d

6 0 6.41, dd (8.0, 2.0) 121.1 d 6.42, dd (8.0, 2.0) 121.1 d

7 0 2.47–2.55, m; 2.24, m 38.8 t 2.46–2.51, m 38.8 t

8 0 2.47–2.55, m 42.5 d 2.46–2.51, m 42.3 d

9 0 4.09, dd (8.8, 7.4) 72.9 t 4.15, dd (8.8, 7.3) 72.9 t

3.87, dd (8.8, 8.3) 3.88, dd (8.8, 8.2)

OMe 3.80, s 56.7 q 3.80, s 56.8 q

a Spectra taken at 500 and 125 MHz for proton and carbon, respectively; chemical shift values were assigned based on the observed 2D NMR

correlations and presented in ppm with TMS as the internal standard; J values given in Hz in parentheses.

B.-N. Su et al. / Phytochemistry 65 (2004) 2861–2866 2863

The resonance signals of H2-70 and H-8 0 were unre-

solved in the 1H NMR spectra of both compounds 1

(dH 2.47–2.55) and 2 (dH 2.46–2.51), so it was not con-

sidered reliable to determine the relative configurations

of H-8 and H-8 0 using NOESY correlations. However,

it has been reported that the two protons of the oxygen-

ated methylene (C-9 0) of dibenzylbutyrolactone lignan

are almost magnetically equivalent when H-8 and H-8 0

are in a cis-relationship, and therefore H-9 0a and H-

9 0b are displayed as an overlapping broad singlet in

the 1H NMR spectrum (Lopes et al., 1983). In contrast,

the 1H NMR signals for H-9 0a and H-9 0b would be ob-

served as separated doublet of doublets when H-8 and

H-8 0 are in a trans-relationship. Therefore, H-8 and H-

8 0 were assigned with a trans-relationship for com-

pounds 1 and 2, based on the observed splitting patternsand chemical shifts of H-9 0a and H-9 0b (Table 1). Simi-

lar CD spectral data were obtained for compounds 1

and 2, which permitted the assignment of the same abso-

lute stereochemistry, 8R and 8 0R, for both of these two

new isolates (Tanoguchi et al., 1987). Accordingly, the

structures of 1 and 2 were assigned, in turn, as

(8R,8 0R)-3 0-O-demethyl-5-hydroxymatairesinol and

(8R, 8 0R)-3 0-O-demethyl-5-methoxymatairesinol.Seven known compounds, (�)-5 0-methoxyyatein (3)

(Richomme et al., 1985), blumenol A (Galbraith and

Horn, 1972), (�)-deoxypodophyllotoxin (anthricin)

(Ikeda et al., 1998), (�)-deoxypodorhizone (Ikeda

et al., 1998), 2,6-dimethoxyhydroquinone (Freire et al.,

2002), 4-hydroxybenzaldehyde (Achenbach et al.,

1988), and b-sitosterol glucoside (Masan et al., 1978),

were isolated along with compounds 1 and 2. The struc-

tures for these known compounds were identified by

comparison of their physical and spectral data with lit-

erature values.

The in vitro inhibitory activity was evaluated against

both COX-1 (Cuendet and Pezzuto, 2000; Waffo-Teguo

et al., 2001) and COX-2 (Cuendet and Pezzuto, 2000;Waffo-Teguo et al., 2001) for the isolates obtained in

the present study. Generally, a compound is considered

to be inactive against COX-1 and -2 if the percentage

inhibition is less than 50 at 100 lg/mL. Among the test

compounds, only 5 0-methoxyyatein (3) showed weak but

selective activity against COX-2 (57% inhibition for

COX-2 and 17% for COX-1 at 100 lg/mL, respectively),

while all the other compounds were demonstrated to beinactive. An IC50 value of 247 lM for 5 0-methoxyyatein

(3) was determined in the COX-2 assay [positive control,

resveratrol, IC50 1.1 lM (COX-1), 1.3 lM (COX-2)].

The initial petroleum ether- and EtOAc-soluble ex-

tracts of M. pomiferus stems showed activity against

COX-2 (the % inhibition was 86 for both extracts

at 100 lg/mL). Previous communications (Ringbom

et al., 2001; Su et al., 2002; Henry et al., 2002; Janget al., 2004) have demonstrated that fatty acids, especially

certain unsaturated fatty acids, are inhibitory towards

COX-2. However, some of these in vitro active fatty

acids have been found to be totally inactive when tested

by our group in a follow up ex vivo mouse mammary or-

gan culture assay (Mehta and Moon, 1991) (data not

2864 B.-N. Su et al. / Phytochemistry 65 (2004) 2861–2866

shown). During the isolation and identification of cancer

chemopreventive compounds directed using COX-1

and/or -2 assays over the past few years, we have found

a number of plant petroleum ether-soluble extracts to be

COX-1 and COX-2 inhibitory active. However, the ma-

jor active compounds then obtained by further purifica-tion are often fatty acids (Su et al., 2002; Jang et al.,

2004). Identification of individual components in active

fractions, a process often termed ‘‘dereplication’’, typi-

cally employs LC–MS, because of the efficiency of

HPLC separations, and the wide range of compound

polarities that can be ionized using atmospheric pressure

ionization interfaces (Wolfender et al., 1992; Cordell

and Shin, 1999). Thus, for routine dereplication in ourlaboratory, plant extracts have been analyzed using

LC–MS conditions that allow detection of a variety of

active plant metabolites (Constant and Beecher, 1995;

Cordell and Shin, 1999). Although GC–MS is the most

often used method for identification of free fatty acids,

negative-ion electrospray LC–MS has also been used

as a convenient alternative for analysis of fatty acids

in mixtures (Carrier and Parent, 2001). In the presentstudy, linoleic acid, oleic acid, and stearic acid were

demonstrated as the main fatty acid components of

the petroleum ether-soluble extract of M. pomiferus

based on the NMR and LC–MS data (Section 3). There-

fore, only the EtOAc-soluble extract was chosen for fur-

ther purification in this study. The percentage

inhibitions in the COX-2 assay of the six fractions

(F001–F006) obtained from the silica gel column chro-matography of this extract were 84, 65, 53, 4, 0, and 1,

respectively. The most active fraction was the most

non-polar fraction, F001, and the main constituents of

this fraction were also indicated to be fatty acids,

namely, linoleic acid, linolenic acid, oleic acid and stea-

ric acid based on the observed 1D NMR and LC–MS

data. Therefore, the pure compounds obtained in this

study were isolated from fractions F002 and F003.Accordingly, to avoid unnecessary time-consuming

activity-guided fractionation, it is recommended that

fatty acids are removed from plant extracts by defatting

with a suitable solvent prior to their evaluation in the

COX-1 and COX-2 bioassays, and these non-polar ex-

tracts are then subjected to LC–MS dereplication.

3. Experimental

3.1. General

Melting points were determined on a Fisher–Johns

melting point apparatus and are uncorrected. Optical

rotations were measured with a Perkin–Elmer 241 auto-

matic polarimeter. UV spectra were obtained with aBeckman DU-7 spectrometer. IR spectra were run on

an ATI Mattson Genesis Series FT-IR spectrophotome-

ter. CD measurements were performed using a JASCO-

710 CD spectropolarimeter. NMR spectral data were

recorded at room temperature on a Bruker Avance

DPX-300 or DRX-500 MHz spectrometer with tetra-

methylsilane (TMS) as an internal standard. Standard

pulse sequences were employed for the measurement of2D NMR spectra (1H–1H COSY, HMQS, HMBC, and

NOESY). HREIMS and EIMS were performed on a

Finnigan/MAT 90/95 sector-field mass spectrometer.

LC–MS analyses were carried out using an Agilent (Palo

Alto, CA) G1946A LCMSD quadrupole mass spec-

trometer equipped with an 1100 HPLC and an XTerra

MS C18 column (5 lm, 150 · 2.1 mm, Waters, Milford,

MA, USA) fitted with a Sentry C18 guard column (5 lm,10 · 2.1 mm, Waters). The reference compounds, elaidic

acid, linoleic acid, linolenic acid, oleic acid, petroselinic

acid, and stearic acid were purchased from Sigma–Ald-

rich (Milwaukee, WI, USA). Column chromatography

was carried out with Si gel G (Merck, 70–230 or 230–

400 mesh). Analytical thin-layer chromatography

(TLC) was performed on 250 lm thickness Merck Si

gel 60 F254 aluminum plates, while preparative thin-layer chromatography was performed on precoated

500 or 1000 lm thick Merck Si gel 60 F254 glass plates.

3.2. Plant material

The stems (400 g) of M. pomiferus were collected in

Indonesia, in October 2000, and identified by S.R. A

voucher specimen (Accession No. ISNG-3) has beendeposited at the Herbarium Bogoriense, Indonesia Insti-

tute of Science, Bogor, Indonesia.

3.3. Cyclooxygenases-1 and -2 inhibitory assays

Inhibition assays against both cylooxygenase-1 and -

2 were performed by the methods described previously

(Cuendet and Pezzuto, 2000; Waffo-Teguo et al., 2001).

3.4. Extraction and isolation

The dried and milled stems (400 g) were extracted by

maceration with MeOH three times (3 · 4 L) at room

temperature, for two days each. After filtration and

evaporation of the solvent under reduced pressure, the

combined crude methanolic extract was suspended inH2O (500 mL) to yield an aqueous MeOH solution

(�95%), which was then partitioned in turn with petro-

leum ether (3 · 500 mL) and EtOAc (3 · 500 mL), to af-

ford dried petroleum ether-(3.81 g), EtOAc-(2.67 g), and

H2O-soluble (ca. 12 g) extracts. The percentage inhibi-

tion of these three extracts against COX-2 at 10 lg/mL were 86, 86, and 69, respectively.

The EtOAc-soluble extract was chromatographedover a Si gel column (5.0 · 45 cm, 350 g 70–230 mesh

Si gel), and eluted with gradient mixtures of CHCl3–

B.-N. Su et al. / Phytochemistry 65 (2004) 2861–2866 2865

MeOH (from 50:1 to 4:1), to afford six fractions (F001–

F006). The percentage inhibitions at 10 lg/mL of these

fractions against COX-2 assay were 84, 65, 53, 4, 0,

and 1, respectively. The fatty acids, linoleic acid, lino-

lenic acid, oleic acid, and stearic acid were demonstrated

to be the major compounds of the first fraction (F001)by NMR and LC–MS data (see below).

Fraction F002, eluted with CHCl3–MeOH (40:1), was

subjected to Sephadex LH-20 column chromatography

(2.8 · 55 cm) and eluted with MeOH, to give four sub-

fractions (F00201–F00204). The subfraction F00202,

was chromatographed over a Si gel column (2.8 · 30

cm) and eluted with CHCl3–MeOH (40:1), to give a mix-

ture of lignans (identified by 1H and 13C NMR spectralmeasurements), 2,6-dimethoxyhydroquinone (8 mg),

and 4-hydroxybenzaldehyde (7 mg), in order of polarity.

The lignanmixture was then purified by preparative TLC

(Merck 60 A Si gel, 20 · 20 cm, 500 lm), developed with

n-hexane-EtOAc twice, in the ratios of 3:1 and 2:1,

respectively, to afford (�)-deoxypodorhizone (Rf =

0.68; 9.0 mg), (�)-deoxypodophyllotoxin (anthricin)

(Rf = 0.65; 4.4 mg), and (�)-5 0-methoxyyatein (3,Rf = 0.57; 0.5 mg). Fraction F003, eluted with CHCl3–

MeOH (30:1), was chromatographed over a Si gel

column (3.8 · 45 cm) and eluted with petroleum ether–

acetone mixtures of increasing polarity (from 3:1 to

1:1), to give three subfractions (F00301–F00303). Blume-

nol A (5.0 mg) was obtained by purifying F00302 using

preparative TLC (Merck 60 A Si gel, 20 · 20 cm, 500

lm), developed with n-hexane-EtOAc (1:1; Rf = 0.47).Subfraction F00303 was finally purified by preparative

TLC (Merck 60 A Si gel, 20 · 20 cm, 1000 lm), devel-

oped with CHCl3–MeOH (12:1), to afford compounds

1 (Rf = 0.52; 1.3 mg) and 2 (Rf = 0.47; 1.2 mg). The yel-

lowish amorphous powder, b-sitosterol glucoside (58

mg), was obtained from a CHCl3–MeOH (�5:1) solution

of F005, eluted with CHCl3–MeOH (10:1).

3.5. (�)-3 0-O-Demethyl-5-hydroxymatairesinol (1)

½a�23D �27.5� (c 0.10, MeOH). UV (MeOH) kmax (log e)209 (3.86), 281 (3.42) nm; IR tmax (film) cm�1 2928,

1768, 1592, 1488, 1240, 1126; CD (MeOH, c 67 mM) knm (De) 237 (�3.03), 276 (�0.65); EIMS m/z (rel. int.):

360 [M]+ (71), 237 (10), 210 (8), 153 (100), 123 (33), 77

(9); HREIMS m/z : 360.1220 (calc. for C19H20O7,360.1209); for the 1H and 13C NMR data see Table 1.

3.6. (�)-3 0-O-Demethyl-5-methoxymatairesinol (2)

½a�23D �6.0� (c 0.12, MeOH); UV (MeOH) kmax (log e)205 (3.82), 281 (3.43) nm; IR tmax (film) cm�1 2912,

1756, 1614, 1512, 1452, 1107; CD (MeOH, c 54 mM) knm (De) 238 (�3.12), 275 (�0.45); EIMS m/z (rel. int.):374 [M]+ (74), 248 (6), 224 (7), 167 (100), 137 (5), 123

(21), 69 (8), 44 (11); HREIMS m/z : 374.1389 (calc. for

C20H22O7, 374.1365); for the 1H and 13C NMR data

see Table 1.

3.7. NMR data and LC–MS analysis of the petroleum

ether extract and a non-polar fraction eluted from the

EtOAc extract of M. pomiferus

That the major constituents of the petroleum ether

extract (D001) and the first fraction (F001) from the col-

umn chromatography of the EtOAc extract are fatty

acids was determined by their 1H and 13C NMR data

(Gunstone et al., 1977; Su and Takaishi, 1999; Su

et al., 2002). The major 1H NMR data in CDCl3 (300

MHz, TMS): d 11.66 (broad singlet, –COOH), 5.28–5.37 (m, olefinic protons), 2.75 (t, J = 5.9 Hz, double

allylic methylene protons), 2.34 (t, J = 7.5 Hz, a-methyl-

ene protons), 2.02–2.06 (m, allylic methylene protons),

1.60–1.65 (m, b-methylene protons), 1.26–1.32 (strong

singlet, other methylene protons), 0.86–0.89 (m, terminal

methyl protons); The major 13C NMR data in CDCl3(75 MHz, TMS): d 180.7 (s; –COOH), 128.0–130.2 (d;

olefinic carbons), 22.8–34.2 (t; other methylenes), 14.1(q; terminal methyls).

In order to determine the compositions of the fatty acid

mixtures, D001 and F001 were analyzed by LC–MS. The

mobile phase for the LC–MS analyses consisted of 0.1%

aqueous formic acid (eluent A) and 0.1% formic acid in

acetonitrile (eluent B). During the LC–MS analysis, the

flow rate was 200 lL/min. After each chromatographic

run, the columnwas equilibrated for 10min at the startingconditions. Negative-ionization ESIMS was operated

with a nitrogen nebulizer pressure of 20 psi, a nitrogen

drying gas temperature of 275 �C at a rate of 4.5 L/min,

a capillary voltage of�2500 V, and a fragmentor voltage

of 60 V. For the initial LC–MS analysis, a mobile phase

system of solvent B in A was used as follows: 0–3 min,

5% B; 3–8 min, linear gradient from 5 to 50% B; 8–20

min, linear gradient from 50 to 95% B; 20–30 min, 95%B. Negative ion mass spectra were recorded scanning

fromm/z 250 to 750. For subsequent LC–MS comparison

of D001 and F001 with fatty acid standards, a 20 min lin-

ear gradient from 80% solvent B to 95% solvent B, fol-

lowed by 5 min at 95% solvent B, was employed, with

solvent A as a counter-solvent. After a 3-min delay, neg-

ative ion mass spectra were recorded for 22 min scanning

fromm/z 200 to 450. Following each analysis, the columnwas flushed with 100% acetonitrile for 5 min at a flow rate

of 250 lL/min.

The initial LC–MS analyses indicated F001 to be a

relatively simple mixture, comprised mainly of four

non-polar compounds forming negative ions at m/z

277, 279, 281, and 283. Coupled with the above-men-

tioned NMR observations, this suggested that the major

constituents in this sample were a series of 18-carbonfatty acids with masses of 278, 280, 282, and 284, respec-

tively. To test this supposition, reference samples of

2866 B.-N. Su et al. / Phytochemistry 65 (2004) 2861–2866

several fatty acids with these molecular weights (elaidic

acid, M = 282; linoleic acid, M = 280; linolenic acid,

M = 278; oleic acid, M = 282; petroselinic acid,

M = 282; and stearic acid, M = 284) were analyzed by

LC–MS using a variation on the method reported by

Carrier and Parent (2001). The fatty acid constituentsin F001 were found to be identical with linoleic, lino-

lenic, oleic, and stearic acids. Based on peak area, the

relative proportions of the fatty acids in F001 were

determined. The main constituents were linoleic and

oleic acids in equal proportions, with linolenic and stea-

ric acids being present at about a 10-fold lower

concentration.

Acknowledgements

This work was supported by Program ProjectP01CA48112, funded by the National Cancer Institute,

NIH, Bethesda, MD. We are grateful to the Research

Resources Center, UIC, for the provision of certain

spectroscopic equipment used in this investigation.

References

Achenbach, H., Stoecker, M., Constenla, M.A., 1988. Constituents of

tropical medicinal plants. Part 31. Flavonoid and other constitu-

ents of Bauhinia manca. Phytochemistry 27, 1835–1841.

Carrier, A., Parent, J., 2001. Liquid chromatography–mass spectrom-

etry determination of free fatty acids in phospholipid-based

formulations. J. Liq. Chromatogr. Rel. Technol. 24, 97–107.

Constant, H.L., Beecher, C.W.W., 1995. A method for the dereplica-

tion of natural product extracts using electrospray HPLC/MS. Nat.

Prod. Lett. 6, 193–196.

Cordell, G.A., Shin, Y.G., 1999. Finding the needle in the haystack.

The dereplication of natural product extracts. Pure Appl. Chem.

71, 1089–1094.

Cuendet, M., Pezzuto, J.M., 2000. The role of cyclooxygenase and

lipoxygenase in cancer chemoprevention. Drug Metab. Drug

Interact. 17, 109–157.

Gunstone, F.D., Pollard, M.R., Scrimgeour, C.M., Vedanayagam,

H.S., 1977. Fatty acids. Part 50. 13C Nuclear magnetic resonance

studies of olefinic fatty acids and esters. Chem. Phys. Lipids 18,

115–129.

Freire, C.S.R., Silvestre, A.J.D., Pascoal Neto, C., Cavaleiro, J.A.S.,

2002. Lipophilic extractives of the inner and outer barks of

Eucalyptus globulus. Holzforschung 56, 372–379.

Galbraith, M.N., Horn, D.H.S., 1972. Structures of natural products

blumenols A, B, and C. J. Chem. Soc., Chem. Commun., 113–114.

Henry, G.E., Momin, R.A., Nair, M.G., Dewitt, D.L., 2002. Antioxi-

dant and cyclooxygenase activities of fatty acids found in food. J.

Agric. Food Chem. 50, 2231–2234.

Ikeda, R., Nagao, T., Okabe, H., Nakano, Y., Matsunaga, H.,

Katano, M., Mori, M., 1998. Antiproliferative constituents in

Umbelliferae plants. III. Constituents in the root and the ground

part of Anthriscus sylvestris Hoffm. Chem. Pharm. Bull. 46, 871–

874.

Jang, D.S., Cuendet, M., Fong, H.H.S., Pezzuto, J.M., Kinghorn,

A.D., 2004. Constituents of Asparagus officinalis evaluated for

inhibitory activity against cyclooxygenase-2. J. Agric. Food Chem.

52, 2218–2222.

Karmali, R.A., 1980. Review: prostaglandins and cancer. Prostaglan-

dins Med. 5, 11–28.

Levy, G.N., 1997. Prostaglandin H synthases, nonsteroidal anti-

inflammatory drugs, and colon cancer. FASEB J. 11, 234–247.

Lopes, L.M.X., Yoshida, M., Gottlieb, O.R., 1983. Dibenzylbutyro-

lactone lignans from Virola sebifera. Phytochemistry 22, 1516–

1518.

Lupulescu, A., 1996. Leukotrienes essent. Fatty Acids 54, 83–94.

Marnett, L.J., 1992. Aspirin and the potential role of prostaglandins in

colon cancer. Cancer Res. 52, 5575–5589.

Masan, C., Woitke, H.D., Hiller, K., Franke, P., 1978. Components of

several herbs. Part 31. Isolation of b-sitosterol-O-DD-glucoside from

Astrantia major L. Pharmazie 33, 382.

Mehta, R.G., Moon, R.C., 1991. Characterization of effective chem-

opreventive agents in mammary gland in vitro using an initiation-

promotion protocol. Anticancer Res. 11, 593–596.

Richomme, P., Bruneton, J., Cave, A., 1985. Study of Hernandiaceae.

VII. 50-Methoxyyatein and 5 0-methoxypodorhizol. New lignans

isolated from Hernandia cordigera Viell. Heterocycles 23, 309–312.

Ringbom, T., Huss, U., Stenholm, A., Flock, S., Skattebøl, L., Perera,

P., Bohlin, L., 2001. COX-2 Inhibitory effects of naturally

occurring and modified fatty acids. J. Nat. Prod. 64, 745–749.

Su, B.-N., Cuendet, M., Farnsworth, N.R., Fong, H.H.S., Pezzuto,

J.M., Kinghorn, A.D., 2002. Activity-guided fractionation of the

seeds of Ziziphus jujuba using a cyclooxygenase-2 inhibitory assay.

Planta Med. 68, 1125–1128.

Su, B.-N., Takaishi, Y., 1999. Morinins H–K, four novel phenylpro-

panol ester lipid metabolites from Morina chinensis. J. Nat. Prod.

62, 1325–1327.

Tanoguchi, M., Arimoto, M., Saika, H., Yamaguchi, H., 1987. Studies

on the constituents of the seeds of Hernandia ovigera L. VI.

Isolation and structural determination of three lignans. Chem.

Pharm. Bull. 35, 4162–4168.

Waffo-Teguo, P., Lee, D., Cuendet, M., Merillon, J.-M., Pezzuto,

J.M., Kinghorn, A.D., 2001. Two new stilbene dimer glucosides

from grape (Vitis vinifera) cell cultures. J. Nat. Prod. 64, 136–138.

Wolfender, J.-L., Maillard, M., Marston, A., Hostettmann, K., 1992.

Mass spectrometry of underivatized naturally occurring glycosides.

Phytochem. Anal. 3, 193–214.