New gastroprotective ferruginol derivatives with selective cytotoxicity against gastric cancer cells

Article abstract of DOI:10.1055/s-2008-1074532

The diterpene ferruginol has shown a strong protective effect in animal gastric ulcer models. In the present work, we report the gastroprotective effect and cytotoxicity of 16 new semisynthetic ester derivatives of ferruginol. The gastroprotective effect of these compounds was assessed with the HCl/EtOH-induced gastric lesions model in mice and the cytotoxicity was measured using MRC-5 fibroblasts, gastric adenocarcinoma (AGS) and liver hepatoma Hep G2 cells. The compounds were tested for a gastroprotective effect at a single oral dose of 20 mg/kg. The best gastroprotective effect was elicited by ferruginyl nicotinate (13), reducing the lesion index by 71 %, while the derivatives ferruginyl chloroacetate (2), ferruginyl palmitate (6), ferruginyl oleate (7), ferruginyl 3,5-dinitrobenzoate (11), ferruginyl 3-methyl-benzofuran-2-carbonyl ester (12), ferruginyl in-doleacetate (14), ferruginyl indolebutyrate (15) and ferruginyl pthalate (16) reduced the lesions by 49-66%. The most promising compounds were 11, 13 and 14, presenting a gastroprotective effect higher or similar to that of ferruginol but with a high selectivity towards the tumor AGS cells. Among the three products, the most selective towards AGS cells was 14, followed by 13, and 11 (IC₅₀ values of 12, 22 and 29 μM, respectively). The isobutyrate 4, inactive as a gastroprotective agent, showed selective cytotoxicity against AGS and Hep G2 cells (IC₅₀ values of 60 and 39.2 μM, respectively). The cytotoxicity of the above cited compounds towards fibroblasts was >1000, μM. Considering the aliphatic esters of ferruginol, the best gastroprotective activity was found in the C₁₆ and C₁₈ derivatives but tended to decrease with increasing aliphatic chain unsaturation. For short-chain esters, the gastroprotective effect could be observed when the chain contained a chlorine atom. For aromatic esters, the presence of nitro groups or a nitrogen atom in the aromatic ring enhanced the gastroprotective activity. The compounds with the best gastroprotective effect and the highest selectivity against tumor cells bear an amino group (indo-leacetate and nicotinate) or nitro group (3,5-dinitrobenzoate).

Full text of DOI:10.1055/s-2008-1074532

802 Original Paper
New Gastroprotective Ferruginol Derivatives with
Selective Cytotoxicity against Gastric Cancer Cells
Author
Marlene Espinoza1, Leonardo S. Santos2, Cristina Theoduloz1, Guillermo Schmeda-Hirschmann3, Jaime A. Rodríguez1
1
Affiliation
Departamento de Ciencias Básicas Biomédicas, Facultad de Ciencias de la Salud, Universidad de Talca, Talca, Chile
Laboratorio de Síntesis Asimétrica, Instituto de Química de Recursos Naturales, Universidad de Talca, Talca, Chile
Laboratorio de Química de Productos Naturales, Instituto de Química de Recursos Naturales, Universidad de Talca, Talca,
2
3
Chile
Key words
Abstract
the above cited compounds towards fibroblasts
was >1000 μM. Considering the aliphatic esters
of ferruginol, the best gastroprotective activity
was found in the C16 and C18 derivatives but tend-
ed to decrease with increasing aliphatic chain
unsaturation. For short-chain esters, the gastro-
protective effect could be observed when the
chain contained a chlorine atom. For aromatic es-
ters, the presence of nitro groups or a nitrogen
atom in the aromatic ring enhanced the gastro-
protective activity. The compounds with the best
gastroprotective effect and the highest selectivity
against tumor cells bear an amino group (indo-
leacetate and nicotinate) or nitro group (3,5-dini-
trobenzoate).
"
● Podocarpus nubigena
!
"
● Prumnopitys andina
The diterpene ferruginol has shown a strong pro-
tective effect in animal gastric ulcer models. In
the present work, we report the gastroprotective
effect and cytotoxicity of 16 new semisynthetic
ester derivatives of ferruginol. The gastroprotec-
tive effect of these compounds was assessed
with the HCl/EtOH-induced gastric lesions model
in mice and the cytotoxicity was measured using
MRC-5 fibroblasts, gastric adenocarcinoma (AGS)
and liver hepatoma Hep G2 cells. The compounds
were tested for a gastroprotective effect at a sin-
gle oral dose of 20 mg/kg. The best gastroprotec-
tive effect was elicited by ferruginyl nicotinate
(13), reducing the lesion index by 71%, while the
derivatives ferruginyl chloroacetate (2), ferrugin-
yl palmitate (6), ferruginyl oleate (7), ferruginyl
3,5-dinitrobenzoate (11), ferruginyl 3-methyl-
benzofuran-2-carbonyl ester (12), ferruginyl in-
doleacetate (14), ferruginyl indolebutyrate (15)
and ferruginyl pthalate (16) reduced the lesions
by 49–66%. The most promising compounds
were 11, 13 and 14, presenting a gastroprotective
effect higher or similar to that of ferruginol but
with a high selectivity towards the tumor AGS
cells. Among the three products, the most selec-
tive towards AGS cells was 14, followed by 13,
and 11 (IC50 values of 12, 22 and 29 μM, respec-
tively). The isobutyrate 4, inactive as a gastropro-
tective agent, showed selective cytotoxicity
against AGS and Hep G2 cells (IC50 values of 60
and 39.2 μM, respectively). The cytotoxicity of
"
● Podocarpaceae
"
● Diterpenes
"
● Ferruginol
"
● gastroprotective effect
"
● cytotoxicity
Abbreviations
!
AGS cells:
human gastric adenocarcinoma
cells
DMSO:
FBS:
dimethyl sulphoxide
fetal bovine serum
received December 12, 2007
revised April 3, 2008
accepted April 7, 2008
HEP G2 cells: human liver hepatocellular
carcinoma cells
Bibliography
MEM:
minimum essential Eagle's
medium
DOI 10.1055/s-2008-1074532
Planta Med 2008; 74: 802–808
© Georg Thieme Verlag KG
Stuttgart · New York
Published online May 21, 2008
ISSN 0032-0943
MRC-5 cells: human lung fibroblasts
NRU: neutral red uptake
Supporting information available online at
http://www.thieme-connect.de/ejournals/toc/
plantamedica
Correspondence
Prof. Dr. Jaime A. Rodríguez
Departamento de Ciencias
Básicas Biomédicas
Facultad de Ciencias de la
Salud
Universidad de Talca
Casilla 747
Talca
Introduction
[1]. Recently, it has been well established that
chronic gastric ulcer might lead to the develop-
ment of gastric and pancreatic cancers [2], [3].
Abietane diterpenes have been found to be a
promising source of bioactive compounds [4].
Ferruginol is an abietane diterpene occurring in
plants belonging to the families Podocarpaceae
!
Around 8–10% of the population living in indus-
trialized countries are affected by gastric and du-
odenal ulcers. In 2002, stomach cancer and pep-
tic ulcer diseases represented 1.5 and 0.5% of the
total causes of death in the world, respectively
Chile
Tel.: +56-71-200-262
Fax: +56-71-200-276
jrodrig@utalca.cl
Espinoza Met al. New Gastroprotective Ferruginol… Planta Med 2008; 74: 802–808

Original Paper 803
[5], Cupressaceae [6], Lamiaceae [7], [8] and Verbenaceae [9],
among others [10]. Ferruginol exhibits different biological activ-
ities including antioxidative [9], miticidal [11], antifungal and
antibacterial [12], cardioactive [13], antiplasmodial [14] and an-
titumoral [15]. Recently, it was reported that ferruginol displays
a gastroprotective effect in animal models of induced gastric le-
sions. Furthermore, the diterpene accelerates the gastric ulcer
healing process after subchronic ulcer induction in animals
[16]. These effects have been related with the capacity of ferru-
ginol to increase the gastric prostaglandin content in vitro, a
stimulating effect on cell proliferation and the antioxidant prop-
erties of the compound [16]. In spite of these valuable properties
as a gastroprotective agent, ferruginol proved to be a cytotoxic
compound with IC50 values ranging from 24 to 26 μM towards
both human normal and tumor cells [16]. In order to obtain bet-
ter activity/cytotoxicity ratios, Areche et al. [17] prepared 18 fer-
ruginol derivatives by semisynthesis. Some of the semisynthetic
compounds were more active as gastroprotective agents than
the parent compound. In addition, some of them proved to be
less cytotoxic than ferruginol against normal fibroblasts and
gastric cancer cells [17]. Interestingly, the derivative 12-
(2,3,4,6-tetra-O-acetyl-β-D-galactopiranosyloxy)-abieta-8,11,13-
triene, that proved to be as active as ferruginol in protecting the
gastric mucosa, showed a selective cytotoxicity against gastric
cancer cells (at least four times higher compared to normal fi-
broblasts) [17].
um specimens (N° 3008 and 3009, respectively) were deposited
at the Herbarium of the Universidad de Talca. For the prepara-
tion of derivatives 2–11 as well as 13–17, 150 mg of ferruginol
were used. The amount of 1,3-dicyclohexylcarbodiimide (DCC)
and 4-dimethylaminopyridine (DMAP) employed in all the men-
tioned reactions was 140 mg and 13 mg, respectively. Esterifica-
tion of ferruginol was performed using chloroacetic (111 mg), io-
doacetic (127 mg), isobutyric (60 mg), lauric (137 mg), palmitic
(175 mg), oleic (296 mg) and linoleic (319 mg) acids (Sigma
Chemical Co.) with (DCC/DMAP) in CH2Cl2 as solvent for 24 h at
room temperature [18] affording the corresponding esters in
86%, 35%, 89%, 54%, 67%, 61% and 63% yields for compounds
2–8, respectively. Treatment of ferruginol with salicylic
(94 mg), phenylacetic (93 mg) and 3,5-dinitrobenzoic (145 mg)
acids (Sigma Chemical Co.) using the same protocol as described
above afforded compounds 9–11 in 32%, 99% and 10% yields, re-
spectively. Esterification of ferruginol (100 mg) using 3-methyl-
benzofuran-2-carbonyl chloride (204 mg) (Sigma Chemical Co.)
in pyridine at 0°C under an inert atmosphere afforded the ester
12 in 7% yield. Treatment of ferruginol with nicotinic (84 mg),
indoleacetic (110 mg), indolebutyric (139 mg), phthalic (57 mg)
and succinic (81 mg) acids (Sigma Chemical Co.) and DCC/
DMAP in CH2Cl2 gave the corresponding esters in 99%, 73%,
67%, 20% and 10% yields, for compounds 13–17, respectively.
The purity of all the derivatives was over 98% as assessed by
1H-NMR spectroscopy.
In an attempt to obtain compounds with better antiulcer activity
and selective cytotoxicity, we prepared 16 new ferruginol deriv-
atives 2–17, ● Fig. 1. The gastroprotective effects as well as the
Animals
"
Male Swiss albino mice from the Instituto de Salud Pública de
Chile, weighing 30 3 g were used. The animals were fed on cer-
tified Champion diet with free access to water under standard
conditions of 12 h dark-light period, 50% relative humidity and
22°C room temperature.
cytotocity towards normal human fibroblasts, human gastric
cancer cells and human hepatoma cells are presented in this re-
port.
Materials and Methods
HCl/ethanol-induced gastric lesions
!
Mice were randomly distributed into groups of seven or eight
animals each and fasted for 24 h with free access to water prior
to the experiment. For the comparison of the ferruginol deriva-
tives, a single oral dose of 20 mg/kg was selected because in a
previous experiment we had determined that the lesion index
was reduced by about 50% by ferruginol at 20 mg/kg [16], [17].
Compounds were suspended in a 12% solution of the non-ionic
detergent Tween 80 (Sigma Chemical Co.) at a concentration of
2 mg/mL. Fifty minutes after oral administration of the com-
pounds, or the positive control lansoprazole [2-({[3-methyl-4-
General procedure
Melting points were determined on a Köffler hot stage apparatus
(Electrothermal 9100) and were uncorrected. Optical rotations
were obtained for solutions in CHCl3 (concentrations expressed
in g/100 mL) on a Jasco DIP 370 polarimeter. IR spectra were re-
corded on a Nicolet Nexus FT-IR instrument. 1H-NMR spectra
were recorded at 400 MHz, and 13C-NMR data were obtained at
100 MHz on a Bruker Avance spectrometer (and expressed in δ).
Mass spectra were obtained on an MAT 95XP, Thermo Finnigan
spectrometer using electron ionization and perfluorotributyl-
amine (Fluorinert FC-43) (Sigma Chemical Co.) as a reference
and are presented as m/z (rel. int.%). Silica gel 60 (Merck, 63–
200 μm particle size) was used for column chromatography, pre-
coated silica gel plates (Merck, Kieselgel 60 F254, 0.25 mm) were
used for TLC analysis. Gel permeation was performed on Sepha-
dex LH-20 (Pharmacia). TLC spots were visualized by spraying
the chromatograms with p-anisaldehyde-ethanol-acetic acid-
H2SO4 (2:170:20:10 v/v) and heating at 110°C. All reactions
were carried out under an inert dry nitrogen atmosphere.
(2,2,2-trifluroethoxy)-2-pyridyl]methyl}sulfinyl)
benzimida-
zole; Sigma Chemical Co.], > 98% purity by HPLC (20 mg/kg), or
the vehicle 12% Tween 80 (10 mL/kg), all groups were orally
treated with 0.2 mL of a solution containing 0.3 M HCl/60% etha-
nol (HCl/EtOH) for gastric lesion induction. Animals were sacri-
ficed 1 h after the administration of HCl/ethanol, and the stom-
achs were excised and inflated by injection of saline (1.0 mL).
The ulcerated stomachs were fixed in 5% formalin for 30 min
and opened along the greater curvature. Gastric damage visible
to the naked eye was observed in the gastric mucosa as elonga-
ted black-red lines, parallel to the long axis of the stomach simi-
lar to the HCl/EtOH-induced lesions in rats. The length (mm) of
each lesion was measured, and the lesion index was expressed as
the sum of the length of all lesions. The percentage of gastropro-
tective effect was calculated as follows:
Preparation of derivatives
Ferruginol was isolated from the wood of Podocarpus nubigena
Lind. and from the stem bark of Prumnopitys andina (Poepp. ex
Endl.) de Laub. (Podocarpaceae) as described previously [17].
The plants were identified by Dr. Patricio Peñaillillo, Departa-
mento de Botánica, Universidad de Talca, and voucher herbari-
% lesion reduction = (1 – lesion index of sample/lesion index of
control)×100
Espinoza Met al. New Gastroprotective Ferruginol… Planta Med 2008; 74: 802–808

804 Original Paper
The protocols were approved by the Universidad de Talca Insti-
tutional Animal Care and Use Committee that follows the rec-
ommendations of the Canadian Council on Animal Care [19].
iance (ANOVA) and when the F value was significant, post hoc
differences were determined by the Dunnett's multiple compar-
ison tests. The level of significance was set at P < 0.01. All statis-
tical analyses were performed using the software Statistica 5.1
(StatSoft, Inc.) and Statistical Package S-Plus 2000.
MRC-5 cell culture
Human normal lung fibroblasts MRC-5 (ATCC CCL-171) were
grown as monolayers in minimum essential Eagle's medium
(MEM), with Earle's salts, 2.0 mM L-glutamine (Sigma Chemical
Co.) and 2.2 g/L sodium bicarbonate (Sigma Chemical Co.), sup-
plemented with 10% heat-inactivated fetal bovine serum (FBS),
100 IU/mL penicillin and 100 μg/mL streptomycin in a humidi-
fied incubator with 5% CO2 in air at 37°C. Culture media, antibi-
otics and fetal bovine serum were obtained from Invitrogen
Corp. Cell passage was maintained between 10 and 16. The me-
dium was changed every 2 days.
Compounds
Ferruginol (1): White powder; m.p. 50–53°C; [α]2D0: + 38.0 (c 0.5,
CHCl3); HR-MS (EI): m/z = 286.2297 (calcd. for C20H30O:
286.2297). For NMR data see [21] and [17].
Ferruginyl chloroacetate (2): Colorless powder; m.p.128–130°C;
[α]2D0: + 51.8 (c 0.06, CHCl3); IR (film): νmax = 2916, 2844, 1752,
1166, 1130 cm–1
;
HR-MS (EI): m/z = 362.20119 (calcd. for
C
22H31ClO2 :362.20126); EI-MS: m/z (rel. int.%) = 364 (20)
(C22H3137ClO2), 362 (58) (C22H3135ClO2), 349 (38), 347 (100), 319
(6), 305 (12), 286 (23), 277 (31), 265 (73), 250 (34), 201 (22),
187 (27), 147 (38), 77 (30), 69 (98).
AGS cell culture
Human gastric adenocarcinoma cells AGS (ATCC CRL-1739) were
grown as monolayers in Ham F-12 medium containing 1.0 mM L-
glutamine and 1.5 g/L sodium bicarbonate, supplemented with
10% heat-inactivated FBS, 100 IU/mL penicillin and 100 μg/mL
streptomycin in a humidified incubator with 5% CO2 in air at
37°C. The cell passage was maintained between 42 and 48. The
medium was changed every 2 days.
Ferruginyl iodoacetate (3): Colorless powder; m.p.101–102°C;
[α]2D0: + 38.2 (c 0.06, CHCl3); IR (film): νmax = 2964, 2920, 1744,
1501, 1242, 1162, 1083 cm–1; HR-MS (EI): m/z = 454.13408 (calcd.
for C22H31IO2 :454.13693); EI-MS: m/z (rel. int.%) = 454 (16), 439
(8), 356 (6), 327 (4), 286 (100), 271 (40), 229 (6), 201 (12), 175
(16), 147 (18), 69 (29).
Ferruginyl isobutyrate (4): Colorless crystals; m.p. 76–77°C;
[α]2D0: + 52.2 (c 0.09, CHCl3); IR (film): νmax = 2960, 2868, 1752,
1465, 1126, 772 cm–1; HR-MS (EI): m/z = 356.27193 (calcd. for
C24H36O2 :356.27155); EI-MS: m/z (rel. int%) = 356 (11), 286
(100), 271 (34), 229 (3), 201 (8), 189 (9), 175 (9), 159 (4), 149 (3),
147 (9), 71 (15), 69 (10), 55 (5).
Ferruginyl laurate (5): Colorless oil; [α]2D0: + 33.9 (c 0.06, CHCl3);
IR (film): νmax = 2916, 2852, 1756, 1465, 1166, 1142 cm–1; HR-MS
(EI): m/z = 468.39614 (calcd. for C32H52O2 :468.39673); EI-MS:
m/z (rel. int.%) = 468 (2), 286 (100), 2719 (22), 201 (6), 189 (7),
183 (8), 175 (7), 149 (8), 147 (7), 71 (5), 69 (14), 57 (15), 55 (14).
Ferruginyl palmitate (6): Oily solid; [α]2D0: + 36.2 (c 0.07, CHCl3);
IR (film): νmax = 2924, 2848, 1760, 1166, 1134 cm–1; HR-MS (EI):
m/z = 524.45718 (calcd. for C36H60O2 :524.45935); EI-MS: m/z
(rel. int.%) = 524 (2), 427 (0.5), 286 (100), 271 (17), 239 (5), 201
(5), 189 (5), 175 (5), 149 (8), 69 (9), 57 (9).
Hep G2 cell culture
Human liver hepatocellular carcinoma cells Hep G2 (ATCC HB-
8065) were grown as monolayers in minimum essential Eagle's
medium (MEM), with Earle's salts, 2.0 mM L-glutamine and
2.2 g/L sodium bicarbonate, supplemented with 10% heat-inacti-
vated FBS, 100 IU/mL penicillin and 100 μg/mL streptomycin in a
humidified incubator with 5% CO2 in air at 37°C. Cell passage
was maintained between 79 and 82. The medium was changed
every 2 days.
Cytotoxicity assay
Confluent cultures of MRC-5, AGS or Hep G2 cells were treated
with medium containing the diterpene derivatives as well as
with the reference compounds at concentrations of 0, 200, 400,
600, 800 and 1000 μM. The products were first dissolved in
DMSO (Sigma Chemical Co.) and then in the corresponding cul-
ture medium supplemented with 2% FBS. The final content of
the solvent DMSO in the test medium and controls was 1%. Cells
were exposed for 24 h to the test medium with or without the
compound (control). Stability of the compounds in the culture
medium was confirmed by means of TLC. Cells treated with lan-
soprazole or vinblastine (Sigma Chemical Co.) ≥ 97% purity by
TLC, were used as positive controls for cytotoxicity. Each concen-
tration was tested in quadruplicate together with the control
and repeated three times in separate experiments. At the end of
the incubation, the neutral red uptake (NR) (Sigma Chemical Co.)
assay was carried out [20]. To calculate the IC50 values (concen-
tration that produces a 50% inhibitory effect on the evaluated
parameter), the results were transformed to percentage of con-
trols assuming that they represent the 100% of cell viability.
The IC50 values were graphically obtained from the dose-re-
sponse curves.
Ferruginyl oleate (7): Colorless oil; [α]2D0: + 23.7 (c 0.08, CHCl3); IR
(film): νmax = 2924, 2852, 1760, 1461, 1162 cm–1; HR-MS (EI): m/z
= 550.47177 (calcd. for C38H62O2 :550.47498); EI-MS: m/z (rel.
int.%) = 550 (1), 453 (0.25), 286 (100), 271 (20), 201 (7), 175 (9),
149 (10), 83 (11), 69 (31), 55 (36).
Ferruginyl linoleate (8): pale yellow oil; [α]2D0: + 30.4 (c 0.26,
CHCl3); IR (film), νmax = 2932, 2848, 1756, 1461, 1162, 1138 cm–1
HR-MS (EI): m/z = 548.45747 (calc. for C38H60O2: 548.4593).
;
Ferruginyl salicylate (9): Colorless crystals; m.p.101–102°C;
[α]2D0: + 58.1 (c 0.09, CHCl3); IR (film): νmax = 3410, 2964, 2850,
1684, 1298, 1154, 760 cm–1; HR-MS (EI): m/z = 406.25049 (calcd.
for C27H34O3 :406.25079); EI-MS: m/z (rel. int.%) = : 406 (17), 286
(100), 271 (29), 245 (10), 189 (12), 147 (9), 121 (79), 93 (10), 69
(12), 65 (13).
Ferruginyl phenylacetate (10): Colorles crystals; m.p. 98–99°C;
[α]2D0: + 51.4 (c 0.07, CHCl3); IR (film): νmax = 2960, 2868, 1752,
1489, 1234, 1122 cm–1; HR-MS (EI): m/z = 404.27005 (calcd. for
C
28H36O2 :404.27155); EI-MS: m/z (rel. int.%) = 404 (6), 286 (85),
271 (28), 201 (9), 189 (10), 187 (7), 175 (10), 147 (14), 119 (6), 91
(100), 69 (14).
Statistical analysis
Results were expressed as the mean S.D. In all experiments,
statistical differences between several treatments and their re-
spective control were determined by one-way analysis of var-
Ferruginyl 3,5-dinitrobenzoate (11): Pale yellow crystals;
m.p.190–192°C; [α]D20: + 51.9 (c 0.05, CHCl3); IR (film): νmax
=
2928, 2888, 1732, 1549, 1346, 1166, 772 cm–1; HR-MS (EI): m/z
Espinoza Met al. New Gastroprotective Ferruginol… Planta Med 2008; 74: 802–808

Original Paper 805
= 480.22196 (calcd. for C27H32 N2O6 :480.22601); EI-MS: m/z (rel.
int.%) = 480 (38), 465 (78), 395 (33), 383 (75), 381 (13), 369 (40),
286 (19), 285 (18), 271 (13), 195 (100), 149 (43), 75 (31), 69 (99),
55 (23).
pounds 2–17 were not found in the literature and are described
for the first time. The structures are presented in ● Fig. 1.
"
An investigation on the structure-gastroprotective effect of the
diterpene dehydroabietic acid (DHA) and its semisynthetic de-
rivatives was undertaken by Wada et al. [22]. In this study, 70
DHA derivatives at the position 12 and/or 18 were prepared and
tested for a possible gastroprotective effect using the antisecre-
tory and antipepsin assays in rats. Recently, Sepúlveda et al. [23]
reported the gastroprotective effect of 19 DHA derivatives, as-
sessed in the HCl/EtOH-induced gastric lesions model in mice
at a single oral dose of 100 mg/kg. Several compounds presented
a strong gastroprotective effect and some structure-activity
trends were observed.
The derivatives 2–8, with C2 to C18 alkyl chains, included satura-
ted and unsaturated fatty acids and two halogenated C2 esters. In
the C2 esters with a halogen atom (compounds 2 and 3), the
chlorine derivative 2 reduced the lesion index by 51% while the
iodine derivative 3 was inactive. In the aliphatic esters 4–8, the
isobutyrate (4) was devoid of activity and the gastroprotective
effect increased with the ester chain length as can be observed
by comparing the activity of the laurate (5) with those of the pal-
mitate (6) and oleate (7) derivatives. Regarding the fatty acid es-
ters 5–8, including both saturated C-12 and C-16 compounds (5
and 6) as well as unsaturated C-18:1 and C-18:2 products (7
and 8), the gastroprotective effects of the palmitate and oleate
derivatives were similar to that of ferruginol but were decreased
in the C-18:2 derivative 8.
Ferruginyl 3-methylbenzofuran-2-carbonyl ester (12): Colorless
crystals; m.p.133–135°C; [α]2D0: + 54.3 (c 0.04, CHCl3); IR (film):
ν
max = 2932, 2860, 1078, 1593, 1290, 1134, 1087, 752 cm–1; HR-MS
(EI): m/z = 444.26628 (calcd. for C30H36O3 :444.26645); EI-MS:
m/z (rel. int.%) = 444 (11), 429 (3), 285 (31), 159 (100), 149 (6),
103 (7), 85 (18), 83 (28), 69 (8), 55 (10).
Ferruginyl nicotinate (13): Colorless powder; m.p.109–110°C;
[α]2D0: + 56.6 (c 0.05, CHCl3); IR (film): νmax = 2964, 2860, 1740,
1585,1266,1162, 1087, 756 cm–1; HR-MS: m/z = 391.25168 (calcd.
for C26H33 NO2 :391.25116); EI-MS: m/z (rel. int.%) = 391 (17), 376
(21), 334 (4), 308 (9), 294 (15), 285 (14), 280 (8), 147 (9), 106
(100.00), 78 (46), 69 (11), 51 (13).
Ferruginyl indoelacetate (14): Colorless crystals; m.p. 67–69°C;
[α]2D0: + 39.7 (c 0.06, CHCl3); IR (film): νmax = 3408, 2926, 2868,
1740, 1242, 1119 cm–1; HR-MS (EI): m/z = 443.28300 (calcd. for
C
30H37NO2 :443.28243); EI-MS: m/z (rel. int.%) = 443 (14), 286
(14), 271 (8), 157 (100), 130 (80).
Ferruginyl indolebutyrate (15): Colorless crystals; m.p. 56–58°C,
[α]2D0: + 40.4 (c 0.05, CHCl3); IR (film): νmax = 3414, 2972, 2928,
1744, 1453, 1138, 740 cm–1; HR-MS (EI): m/z = 471.31136 (calcd.
for C32H41NO2:471.31375); EI-MS: m/z (rel. int.%) = 471 (6), 286
(100), 271 (32), 186 (82), 130 (34).
Ferruginyl pthalate (16): Colorless crystals; m.p.120–122°C;
[α]2D0: + 36.0 (c 0.05, CHCl3); IR (film): νmax = 2956, 2920, 1736,
1493, 1266, 1246, 1162, 1111, 756 cm–1; HR-MS (EI): m/z =
702.46031 (calcd. for C48H62O4 :702.46484); EI-MS: m/z (rel.
int.%) = 702 (1.6), 687 (0.03), 605 (0.11), 579 (0.18), 551 (0.11),
417 (100), 375 3 (10), 286 (7), 271 (6), 147 (22), 133 (6), 131 (6),
105 (4), 69 (15).
When comparing the effect of the newly prepared aliphatic es-
ters with those described by Areche et al. [17], the isobutyrate
(C4:4) and the propionate (C3) derivative tested by them were in-
active but they observed that the presence of a chlorine atom in
the C3 ester chain or a double bond (propenoate) led to active de-
rivatives. For C2 and C3 esters, we also observed that the presence
of a chlorine atom is relevant for the gastroprotective effect. For
fatty acid esters, our results showed that the effect appears to in-
crease with longer side chains but unsaturation decreases the
activity.
Ferruginyl succinate (17): Colorless crystals; m.p. 85–87°C;
[α]2D0: + 59.1 (c 0.04, CHCl3); IR (film): νmax = 2966, 2848, 1752,
1133, 754 cm–1
; HR-MS (EI): m/z = 654.46440 (calcd. for
C44H62O4 :654.46484); EI-MS: m/z (rel. int.%) = 654 (2.7), 570
(1), 444 (4.3), 429 (1.2), 417 (0.4), 387 (0.3), 369 (18), 286 (100),
271 (45), 201 (20), 189 (21), 187 (13), 175 (21), 159 (49), 149 (13),
147 (16), 83 (14), 69 (39), 55 (25).
Considering the aromatic derivatives 9–12, the best effect was
elicited by the 3,5-dinitrobenzoate (11) and compound 12,
which showed a gastroprotective effects similar to that of ferru-
ginol. The dinitro derivative 11 proved to be more active than the
mononitro compounds and the m-substituted product was
much more active than the p-substituted aromatic ester [17].
The salicylate 9 presented a similar gastroprotective effect as
that of the benzoate described by Areche et al. [17], with lower
activity for the phenylacetate 10. The heterocyclic derivative 13
presented a significant gastroprotective effect, higher than that
of ferruginol, being worthy of additional dose-response studies
to find the corresponding EC50. In the indoleacetate 14 and indo-
lebutyrate 15 derivatives, differing in the chain length, the gas-
troprotective effects were similar, reducing lesions by around
50%. A similar observation was made for the diesters 16 and 17.
Both compounds differ in the nature of the acid linkers (phthalic
and succinic acid), suggesting that the gastroprotective effect
does not depend on the identity of the selected diacids. The ac-
tivity of compounds 1–17 in the gastric lesion model in mice is
Supporting information
1H-NMR data and 13C-NMR data of compounds 2–17 are avail-
able as Supporting Information.
Results and Discussion
!
Ferruginol is an aromatic diterpene containing an abietane moi-
ety. It was first obtained from Podocarpus ferruginea (Podocarpa-
ceae) and proved to be a powerful gastroprotective agent [16]. In
a previous report, a dose-response study of ferruginol showed
that, at 25 mg/kg, the gastroprotective effect of the diterpene
was similar to that of lansoprazole at 20 mg/kg in HCl/EtOH-in-
duced gastric lesions in mice. Therefore, we chose the dose of
20 mg/kg to compare the effect of ferruginol with the new semi-
synthetic derivatives.
"
summarized in ● Table 1.
Starting from ferruginol (1),16 derivatives (2–17) were obtained
by semisynthesis. The 1H- and 13C-NMR spectroscopic data of
these compounds are presented in Tables 1S–5S (Supporting In-
formation). All compounds prepared in this work present spec-
troscopic data in agreement with the proposed structures. Com-
Regarding cytotoxicity, esterification with chloroacetic and io-
doacetic acids (2 and 3) led to products with higher or similar cy-
totoxicity than the starting compound. The 2-chloroacetate de-
rivative was more cytotoxic than ferruginol, presenting no selec-
tivity towards both AGS cells and fibroblasts. Esterification of
Espinoza Met al. New Gastroprotective Ferruginol… Planta Med 2008; 74: 802–808

806 Original Paper
Fig. 1 Structures of ferruginol 1 and ferruginol
derivatives 2–17.
ferruginol with carboxylic acids (5–8) afforded less toxic deriva-
μM against AGS cells, compared to >1000 μM for fibroblasts.
The heterocycles 13–15 proved to be less cytotoxic on fibro-
blasts but had a strong and selective effect on AGS cells, particu-
larly compounds 13 and 14 with IC50 values of 22 and 12 μM, re-
spectively. Comparing the cytotoxicity on AGS cells of com-
pounds 14 and 15, differing only in one methylene (CH2) be-
tives with almost no effect on fibroblasts (IC50 >1000 μM). The
isobutyrate, however, was highly selective with an IC50
>1000 μM and 60 μM towards fibroblasts and AGS cells, respec-
tively. Among the aromatic derivatives 9–11, the 3,5-dinitroben-
zoate (11) presented remarkable selectivity with an IC50 of 29
Espinoza Met al. New Gastroprotective Ferruginol… Planta Med 2008; 74: 802–808

Original Paper 807
Table 1 Gastroprotective effect of ferruginol (1) and the semisynthetic derivatives 2–17 at the single oral dose of 20 mg/kg, and cytotoxicity (IC50, μM) to-
wards MRC-5 fibroblasts, AGS and Hep G2 cells
Compound
Lesion Index
(mm)
% Lesion
Cytotoxicity IC50 (μM)
inhibition
MRC-5
AGS
Hep G2
1
15.5 7.7**
19.4 6.9**
30.4 5.7
61
51
24
23
42
53
54
40
38
31
66
58
71
55
50
49
47
23.0
27.0
7.3
68.5
2
14.2
> 1 000
> 1 000
39.2
3
44.9
23.8
59.7
> 1 000
567
838
554
903
710
28.8
695
22.4
11.5
465
774
455
122
162
4
30.7 7.2
> 1 000
> 1 000
> 1 000
> 1 000
> 1 000
> 1 000
> 1 000
> 1 000
> 1 000
> 1 000
> 1 000
> 1 000
> 1 000
773
5
23.3 6.2*
18.9 5.1**
18.3 5.3**
23.8 5.5*
24.7 5.7*
27.8 6.4*
13.4 5.2**
16.6 5.5 **
11.7 4.5**
18.1 5.1**
20.1 6.2**
20.5 4.8**
21.3 5.9*
> 1 000
519
6
7
> 1 000
> 1 000
> 1 000
> 1 000
22.1
8
9
10
11
12
> 1 000
161
13
14
315
15
> 1 000
> 1 000
637
16
17
Vinblastine
Lansoprazole
Control
78
254
9.2 4.8**
39.9 6.5
77
306
221
* P < 0.05; ** P < 0.01; n = 7 – 8; mean S.D.
tween the aromatic ring and the carbonyl group, a higher selec-
tivity can be seen for compound 14 (12 μM) than for compound
15 (465 μM). The diesters 16 and 17 were less toxic than ferrugi-
nol towards both fibroblasts and AGS cells, however, they were
lacking selectivity.
aromatic ring enhances the gastroprotective activity. Additional
studies including computational design should be undertaken to
prepare more active derivatives and disclose the structure-rela-
tionship trends in this group of compounds.
Hep G2 cells can biotransform the compounds added to culture
media because they express cytochrome P450, which is able to
metabolize xenobiotics. For compounds 2, 3, 7–10, 12 –16,
lower cytotoxicity was found towards hepatocytes than to-
wards AGS cells suggesting that the derivatives are being me-
tabolized in the liver cells. The cytotoxicity of compounds 1–
17 towards AGS cells, fibroblasts and hepatoma cells is sum-
Acknowledgements
!
Financial support by FONDECYT, Grant N° 1060841 and the Pro-
grama de Productos Bioactivos, Universidad de Talca is gratefully
acknowledged. L.S.Santos also thanks the International Founda-
tion for Science (F/4195–1) and the Organization for Prohibition
of Chemical Weapons for support of research activity.
"
marized in ● Table 1.
When comparing the gastroprotective effect and cytotoxicity
data, the most promising compounds are the derivatives 11, 13
and 14. The three compounds have a gastroprotective effect at
least comparable to that of ferruginol, but with higher selectivi-
ties against the tumoral AGS cells with an IC50 of 12 μM for the
indoleacetate 14, followed by the nicotinate 13 (IC50 :22 μM)
and the 3,5-dinitrobenzoate 11 (IC50 :29 μM). The IC50 values of
these compounds on fibroblasts were >1000 μM. The three com-
pounds bear an N atom either as an amino (indoleacetate and
nicotinate) or nitro (3,5-dinitrobenzoate) function. The isobuty-
rate 4, lacking a gastroprotective effect at 20 mg/kg is of interest
because of its selective cytotoxicity against AGS cells
(IC50 :60 μM). Several derivatives showed low cytotoxicity values
and good gastroprotective activity, suggesting that the two ef-
fects are not related. For aliphatic ester derivatives of ferruginol,
the gastroprotective effect seems to be related to the side chain
length. Better activity was found in the C16 and C18 derivatives
but tended to decrease with increasing aliphatic chain unsatura-
tion. For short-chain esters, the gastroprotective effect can be
observed when the chain contains a chlorine atom. For aromatic
esters, the presence of dinitro groups or a nitrogen atom in the
References
1 WHO. The World Health Report 2003: 154–9
2 Luo J, Nordenvall C, Nyren O, Adami HO, Permert J, Ye W. The risk of
pancreatic cancer in patients with gastric or duodenal ulcer disease.
Int J Cancer 2007; 120: 368–72
3 Bahmanyar S, Ye W, Dickman PW, Nyren O. Long-term risk of gastric
cancer by subsite in operated and unoperated patients hospitalized
for peptic ulcer. Am J Gastroenterol 2007; 102: 1185–91
4 San Feliciano A, Gordaliza M, Salinero M, Del Corral J. Abietane acids:
sources, biological activities, and therapeutic uses. Planta Med 1993;
59: 485–90
5 Cambie RC, Cox RE, Sidwell D. Phenolic diterpenoids of Podocarpus
ferrugineus and other podocarpus. Phytochemistry 1984; 23: 333–6
6 Sharp H, Latif Z, Bartholomew B, Bright C, Jones CD, Sarker SD et al. To-
tarol, totaradiol and ferruginol: three diterpenes from Thuja plicata
(Cupressaceae). Biochem Syst Ecol 2001; 9: 215–7
7 Ulubelen A, Topcu G. Abietane diterpenoids from Salvia microstegia.
Phytochemistry 1991; 30: 2085–6
8 Ulubelen A, Topcu G. New abietane diterpenoids from Salvia
montbretti. J Nat Prod 1992; 55: 441–4
9 Ono M, Yamamoto M, Masuoka C, Ito Y, Yamashita M, Nohara T. Diter-
penes from the fruits of Vides rotundifolia. J Nat Prod 1999; 62: 1532–
7
Espinoza Met al. New Gastroprotective Ferruginol… Planta Med 2008; 74: 802–808

808 Original Paper
10 Dictionary of Natural Products on CD-ROM, Version 2007. Chapman
17 Areche C, Rodríguez JA, Razmilic I, Yánez T, Theoduloz C, Schmeda-
Hirschmann G. Gastroprotective and cytotoxic effect of semisynthetic
ferruginol derivatives. J Pharm Pharmacol 2007; 59: 289–300
18 Santos LS, Pilli RA, Rawal VH. Enantioselective total synthesis of (+)-ar-
borescidine A, (–)-arborescidine B, and (–)-arborescidine C. J Org
Chem 2004; 69: 1283–9
19 Olfert ED, Cross BM, McWilliam AA. Guide to the care and use of exper-
imental animals, Vol.1. Ottawa: Canadian Council on Animal Care;
1993: 213–4
and Hall/CRC;
11 Chang S, Chen P, Wang S, Wu H. Antimite activity of essential oils and
their constituents from Taiwania cryptomerioides. J Med Entomol
2001; 38: 455–7
12 Becerra J, Flores C, Mena J, Aqueveque P, Alarcón J, Bittner M et al. Anti-
fungal and antibacterial activity of diterpenes isolated from wood ex-
tractables of Chilean Podocarpaceae. Bol Soc Chil Quím 2002; 47:
151 – 7
13 Ulubelen A, Birman H, Oksuz S, Topcu G, Kolak U, Barla A et al. Cardioac-
tive diterpenes from the roots of Salvia eriophora. Planta Med 2002;
68: 818–21
14 Clarkson C, Musonda C, Chibale K, Campbell WE, Smith P. Synthesis of
totarol amino alcohol derivatives and their antiplasmodial activity
and cytotoxicity. Bioorg Med Chem 2003; 11: 4417–22
15 Iwamoto M, Minami T, Tokuda H, Ohtsu H, Tanaka R. Potencial antitu-
mor promoting diterpenoids from the item bark of Thuja standishii.
Planta Med 2003; 69: 69–72
16 Rodríguez JA, Theoduloz C, Yánez T, Becerra J, Schmeda-Hirschmann G.
Gastroprotective and ulcer healing effect of ferruginol in mice and
rats: assessment of its mechanism of action using in vitro models.
Life Sci 2006; 78: 2503–9
20 Rodríguez JA, Haun M. Cytotoxicity of trans-dehydrocrotonin from
Croton cajucara (Euphorbiaceae) on V79 cells and hepatocytes. Planta
Med 1999; 65: 522–6
21 Ying B, Kubo I. Complete 1H and 13C NMR assignments of totarol and
its derivatives. Phytochemistry 1991; 6: 1951–5
22 Wada H, Kodato S, Kawamori M, Morikawa T, Nakai H, Takeda M et al.
Antiulcer activity of dehydroabietic acid derivatives. Chem Pharm
Bull 1985; 33: 1472–87
23 Sepúlveda B, Astudillo L, Rodríguez JA, Yáñez T, Theoduloz C, Schmeda-
Hirschmann G. Gastroprotective and cytotoxic effect of dehydroabietic
acid derivatives. Pharmacol Res 2005; 52: 429–37
Espinoza Met al. New Gastroprotective Ferruginol… Planta Med 2008; 74: 802–808

Copyright of Planta Medica is the property of Georg Thieme Verlag Stuttgart and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright
holder's express written permission. However, users may print, download, or email articles for
individual use.