Gastroprotective effect and cytotoxicity of carnosic acid derivatives

Article abstract of DOI:10.1055/s-0030-1250648

Carnosic acid (CA) is the main phenolic diterpene of rosemary (Rosmarinus officinalis L., Lamiaceae) and presents gastroprotective effect in vitro and in vivo. To determine structure-activity relationships, seventeen esters and ethers of CA were prepared, comprising aliphatic, aromatic, and heterocyclic compounds. The naturally occurring 12-O-methylcarnosic acid (14) was also included in the study. The compounds were evaluated for their gastroprotective activity in the HCl/EtOH-induced gastric lesions model in mice, and for cytotoxicity in human adenocarcinoma AGS cells, Hep G2 hepatocellular carcinoma cells, and human lung fibroblasts. At 10 mg/kg, some of the CA derivatives (5, 8, 9, 12, 14, and 18) were more effective preventing gastric lesions than the reference compound lansoprazole at the same dose. The dibenzoate 9, diindoleacetate 12, and the derivative 18 showed the best gastroprotective effect combined with the lowest cytotoxicity. Georg Thieme Verlag KG Stuttgart.

Full text of DOI:10.1055/s-0030-1250648

882 Original Papers
Gastroprotective Effect and Cytotoxicity
of Carnosic Acid Derivatives
Authors
Cristina Theoduloz1, Mariano Walter Pertino2, Jaime A. Rodríguez1*, Guillermo Schmeda-Hirschmann2
1
2
Affiliations
Laboratorio de Cultivo Celular, Facultad de Ciencias de la Salud, Universidad de Talca, Talca, Chile
Laboratorio de Química de Productos Naturales, Instituto de Química de Recursos Naturales,
Universidad de Talca, Talca, Chile
Key words
Abstract
ence compound lansoprazole at the same dose.
The dibenzoate 9, diindoleacetate 12, and the de-
rivative 18 showed the best gastroprotective ef-
fect combined with the lowest cytotoxicity.
"
l Rosmarinus officinalis
!
"
l Lamiaceae
Carnosic acid (CA) is the main phenolic diterpene
of rosemary (Rosmarinus officinalis L., Lamiaceae)
and presents gastroprotective effect in vitro and
in vivo. To determine structure-activity relation-
ships, seventeen esters and ethers of CA were pre-
pared, comprising aliphatic, aromatic, and het-
erocyclic compounds. The naturally occurring
12-O-methylcarnosic acid (14) was also included
in the study. The compounds were evaluated for
their gastroprotective activity in the HCl/EtOH-
induced gastric lesions model in mice, and for cy-
totoxicity in human adenocarcinoma AGS cells,
Hep G2 hepatocellular carcinoma cells, and hu-
man lung fibroblasts. At 10 mg/kg, some of the
CA derivatives (5, 8, 9, 12, 14, and 18) were more
effective preventing gastric lesions than the refer-
"
l carnosic acid derivatives
"
l semisynthesis
"
l cytotoxicity
"
l gastroprotective effect
Abbreviations
!
AGS cells:
human gastric adenocarcinoma
cells
HEP G2 cells: human liver hepatocellular
carcinoma cells
MEM:
minimum essential Eagleʼs medium
MRC-5 cells: human lung fibroblasts
Supporting information available online at
http://www.thieme-connect.de/ejournals/toc/
plantamedica
Introduction
ed a cytotoxic effecton P388 murineleukemia cells
[13]. The effect of 20 diterpenes isolated from Sal-
via species against gram-positive and gram-nega-
tive bacteria by the agar diffusion technique was
reported [14]. The compounds investigated in-
cluded CA derivatives, products with a lactone
group between C-20 and C-6, and C-20 and C-7, as
well as other oxidation and substitution patterns
of the diterpenes. The best effect against Staphylo-
coccus aureus and Bacillus subtilis was found for a
CAderivativeandaC-20andC-6lactonewitharing
oxidized to a quinoid system [14]. The gastropro-
tective effect and cytotoxicity of carnosic acid
γ‑lactone derivatives was recently reported [15].
In the present study, we describe the preparation
of various semisynthetic derivatives of CA and ex-
amine the structure-activity relationships of
these compounds as gastroprotective agents, as
well as their cytotoxicity. The effects were com-
pared with those of the parent compound to-
wards normal lung fibroblasts MRC-5, human
gastric adenocarcinoma AGS cells, and human
received
revised
accepted
August 26, 2010
Nov. 17, 2010
Nov. 28, 2010
!
Carnosic acid (CA) is an abietane diterpene occur-
ring in the medicinal and aromatic plants rose-
mary (Rosmarinus officinalis L.) and sage (Salvia
officinalis L.) (Lamiaceae) as well as in other
plants of the same botanical family [1,2]. Crude
hydroalcoholic extracts of R. officinalis and S. offi-
cinalis display a strong gastroprotective effect,
and the activity was associated with the plant di-
terpenes, including CA [3,4]. Rosemary extract
presents also antimicrobial [5] and antioxidant ef-
fects [6,7].
Carnosic acid was also shown to prevent seric tri-
glyceride increase in olive-oil treated mice and in-
hibited the pancreatic enzyme lipase [8], modu-
lated bacterial multidrug resistance [9], inhibited
inflammation and human 5-lipoxygenase [10],
showed an antioxidant effect [11,12] and present-
Bibliography
DOI http://dx.doi.org/
10.1055/s-0030-1250648
Published online January 14,
2011
Planta Med 2011; 77: 882–887
© Georg Thieme Verlag KG
Stuttgart · New York ·
ISSN 0032‑0943
Correspondence
Guillermo
Schmeda-Hirschmann
Laboratorio de Química
de Productos Naturales
Instituto de Química
de Recursos Naturales
Universidad de Talca
2 Norte 685
Casilla 747, Talca
Chile
Phone: + 5671200288
Fax: + 5671201573
schmeda@utalca.cl
* Dedicated to the memory of Prof. Jaime A. Rodríguez.
Theoduloz C et al. Gastroprotective Effect and… Planta Med 2011; 77: 882–887

Original Papers 883
Hep G2 hepatocellular carcinoma cells, as well as on the HCl/
EtOH-induced gastric lesions model in mice. The MRC-5 cells
were used as a control of basal cytotoxicity. Epithelial gastric
AGS cells were selected because we were looking for gastropro-
tective activity. Hepatocytes Hep G2 were included because they
express the P450 cytochrome system, differing from the two pre-
vious cell lines that are unable to metabolize the compounds.
dide. The compounds 3 and 4 were obtained by acetylation of CA
and 1, respectively. The derivatives 5–13 were prepared by ester-
ification reactions between the phenolic groups at C-11 and C-12
of methylcarnosate (1) with different acids, using dicyclohexyl-
carbodiimide (DCC) and dimethylaminopyridine (DMAP) in dry
CH2Cl2 at room temperature. Esterification of 1 was performed
using chloroacetic, propionic, butyric, isobutyric, benzoic, nico-
tinic, phenylacetic, indoleacetic, and indolebutyric acid, with
DCC/DMAP in CH2Cl2 for 2 h at room temperature, affording the
corresponding chloroacetate 5, propionate 6, butyrate 7, isobu-
tyrate 8, benzoate 9, nicotinate 10, phenylacetate 11, indoleace-
tate 12, and indolebutyrate 13 in 46–84% w/w yield. 12-O-meth-
ylcarnosic acid (14) was methylated with CH2N2 to afford com-
pound 15 while compound 16 was prepared treating 15 in DMF
with NaH and then adding propyl iodide. The compound 17 was
obtained by acetylation of 15. Treatment of 14 with NaH and
propyl iodide in DMF afforded compound 18. The purity of all de-
rivatives was over 98% as assessed by 1H NMR spectroscopy. The
Materials and Methods
!
Equipment and general procedure
Melting points were determined on a Koffler hot stage apparatus
(Electrothermal 9100) and were uncorrected. Optical rotations
were measured on a Jasco DIP 370 (Jasco Analytical Instruments)
polarimeter in CHCl3 at 20°C. IR spectra were recorded on a Nico-
let Nexus 470 FT‑IR instrument (Thermo Electron Corporation).
The NMR spectra were recorded on a Bruker Avance 400 (Bruker)
spectrometer at 400 MHz for 1H and 100 MHz for 13C in CDCl3
(and expressed in δ). Low-resolution mass spectra were run on a
VG Micromass ZAB-2F and high-resolution mass spectra on a VG
Micromass ZAB-2F at 70 eV (Varian, Inc.). Merck silica gel (0.063–
0.2) was used for column chromatography; precoated silica gel
plates (Merck, Kieselgel 60 F254, 0.25 mm) were used for TLC anal-
ysis. 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 for 3 min. Diciclohexylcarbodiimide
(DCC) and dimethylaminopyridine (DMAP) were from Merck.
"
syntheses are summarized in l Figs. 1 and 2, and the detailed
preparation of the compounds is presented as Supporting Infor-
mation. All of the compounds were characterized by spectro-
scopic means and compounds 5–13, 16, and 18 are described for
the first time.
HCl/EtOH-induced ulcer model in mice
The gastroprotective effect of the compounds was determined by
the HCl/EtOH-induced gastric lesions model in mice [16]. Male
Swiss albino mice weighing 30 3 g purchased from the Instituto
de Salud Pública de Chile, Santiago de Chile, were used. Animals
were fed on certified Champion diet with free access to water
under standard conditions of a 12-h dark light period, 50% rela-
tive humidity, and 22°C. Mice were randomly distributed into
groups of 7 animals each and fasted for 24 hours with free access
to water prior to the experiment. Carnosic acid was tested at four
different doses (40, 20, 10, and 5 mg/kg) for the dose-response
study. These were selected based on previous work on gastropro-
tective diterpenes. For comparison purposes, the semisynthetic
compounds were assessed at 10 mg/kg. This dose was selected
because CA showed 49% of lesion reduction at 10 mg/kg. Com-
pounds were suspended in 12% Tween 80 (Sigma Chemical Co.)
and administered orally. Fifty min after treatment, all groups re-
ceived orally 0.2 mL of a solution containing 0.3 M HCl/60% EtOH
(HCl/EtOH) for gastric lesion induction. The antisecretory drug
lansoprazole (Sigma Chemical Co.) was used as the reference
compound. According to the manufacturer, the purity of lanso-
prazole was ≥ 98% by TLC analysis. The control group received
12% Tween 80. Animals were sacrificed by cervical dislocation
1 h after the administration of HCl/EtOH, and the stomachs were
excised and inflated by injection of 5% formalin (1 mL). The ulcer-
ated stomachs were fixed in 5% formalin for 30 min and opened
along the greater curvature. The length (mm) of each lesion was
measured, and the lesion index was expressed as the sum of the
length of all lesions [16,17]. The protocols were approved by the
Universidad de Talca Institutional Animal Care and Use Com-
mittee, which follows the recommendations of the Canadian
Council on Animal Care [18] under the number FONDECYT Grant
N° 1060841.
Plant material
Carnosic acid (CA) and 12-O-methylcarnosic acid (14) were iso-
lated from the aerial parts of R. officinalis cultivated in Curicó,
VII Región, Chile. A voucher herbarium specimen (Pertino 001/
2007) has been kept at the Herbario de la Universidad de Talca
and was identified by Dr. Patricio Peñailillo. The air-dried aerial
parts (3 kg) of rosemary were powdered and successively ex-
tracted under reflux with petrol (PE), ethyl acetate (EtOAc), and
MeOH (3 × 5 L for 30 min for each solvent). After concentration
under reduced pressure, 151.6 g (5.05%), 288.0 g (9.60%), and
320.0 g (10.67%) of solubles were obtained from the PE, EtOAc,
and MeOH extract, respectively. The PE extract was dissolved in
MeOH and defatted at 0°C. The MeOH-soluble portion of the PE
extract (126.6 g) was submitted to column chromatography on
silica gel (63–200 µm, Merck Kieselgel 60; column length: 53 cm,
internal diameter: 9 cm) using a stepwise gradient of PE (7 L),
PE:EtOAc 70:30 (5 L), 60:40 (5 L), and EtOAc (3 L). Fractions of
1 L were collected and monitored by TLC (PE/EtOAc 80:20) with
respect to carnosic acid (CA) and 12-O-methylcarnosic acid (14)
content. The fractions 16–27 (16–27 L, 41.8 g) were combined
and purified by permeation on Sephadex LH-20 (column length:
58 cm, internal diameter: 5 cm; MeOH, flow rate 2 mL/min; 3
times × 14 g). For each 14 g batch, using the same column, 23 frac-
tions of 30 mL each were collected. Fractions 14–16 (420–
480 mL, 1.5 g) were combined for compound 14. Fractions 19–23
(570–690 mL, 6.0 g) were combined for CA. Pure compound 14
(0.4 g) and carnosic acid (CA) (2.1 g) were obtained by recrystal-
lization on PE.
Preparation of derivatives
Cytotoxicity assay
Carnosic acid (CA) was methylated with CH2N2 to afford the
methyl ester 1. The methyl ether 2 was prepared treating CA in
dimethylformamide (DMF) with NaH and then adding methyl io-
The human cell lines, MRC-5 normal lung fibroblasts (CCL-171),
AGS gastric adenocarcinoma cells (CRL-1739), and Hep G2 hepa-
tocellular carcinoma cells (HB-8065) were obtained from the
Theoduloz C et al. Gastroprotective Effect and… Planta Med 2011; 77: 882–887

884 Original Papers
Fig. 1 Preparation of compounds 1–13. Reagents
and conditions: (a) CH2N2; EtO2; (b) NaH; MeI; DMF
(c) acetic anhydride; pyridine; (d) acetic anhydride;
pyridine or appropriate organic acid; DCC; DMAP;
CH2Cl2; (e) appropriate aromatic acid; DCC; DMAP;
CH2Cl2; (f) appropriate indolic acid; DCC; DMAP;
CH2Cl2. All reactions were carried out at room tem-
perature. CA: carnosic acid.
and 1.5 g/L sodium bicarbonate. All media were supplemented
with 10% heat inactivated fetal bovine serum, 100 IU/mL penicil-
lin, and 100 µg/mL streptomycin in a humidified incubator with
5% CO2 in air at 37°C. For the experiments, cells were plated at a
density of 25000 cells/mL in 96-well plates. Confluent cultures of
the different cell lines were treated with medium containing the
compounds at concentrations ranging from 0 up to 1000 µM.
Compounds were first dissolved in dimethyl sulfoxide (DMSO)
and then in medium. The final concentration of DMSO in the test
medium and controls was 1%. Final concentrations of DMSO did
not interfere with cell growth. Cells were exposed to test medium
for 24 hours, with or without the compound (control). Each con-
centration was tested in quadruplicate together with the control
and repeated three times in separate experiments. Cell viability
was determined at the end of the incubation by means of the
neutral red uptake assay [19]. Results were converted to percent-
age of controls, and the IC50 values were graphically obtained
from the dose-response curves. Culture media and fetal bovine
serum were obtained from Invitrogen Corp. Other reagents were
purchased from Sigma Chemical Co.
Fig. 2 Preparation of compounds 15–18. Reagents and conditions: (a)
CH2N2; EtO2; (b) NaH; ProI; DMF or acetic anhydride; pyridine; (c) NaH;
ProI; DMF. All reactions were carried out at room temperature.
Statistical analysis
Results were expressed as the mean SD. Statistical differences
between several treatments and their respective control were
determined by the Kruskal-Wallis test followed by the nonpara-
metric multiple comparison Nemenyiʼs test. All statistical analy-
ses were performed using the software Statistica 5.1 (StatSoft,
Inc.) and Statistical Package S-Plus 2000.
American Type Culture Collection (ATCC). The cells were grown
as monolayers in the following media: MRC-5 and Hep G2 in
MEM and AGS in Ham F-12. The MEM medium contained 2 mM
L-glutamine, 1 mM sodium pyruvate and 1.5 g/L sodium bicar-
bonate. Ham F-12 was supplemented with 2 mM L-glutamine
Theoduloz C et al. Gastroprotective Effect and… Planta Med 2011; 77: 882–887

Original Papers 885
Compounds
z = 582.3367 (calcd. for C37H42O6: 582.3281); EI‑MS: (rel. int. %)
m/z = 582 [M] (5), 447 (16), 446 (48), 347 (8), 346 (35), 287
(22), 286 (100), 118 (10), 91 (74).
11,12-O-diindoleacetylmethylcarnosate (12): Yellow oil; [α]D20
:
Carnosic acid (CA): Colorless crystals; m.p. 188–191°C; [α]2D0: + 52
+
(c0.10,CHCl3);IR(film):νmax = 3422, 2957, 1674, 1426, 1165 cm−1
HR‑MS (EI): m/z = 332.1965 (calcd. for C20H28O4: 332.1988);
EI‑MS: m/z (rel. int. %) = 332 [M] (6), 286 (100), 271 (16), 243
;
+
+ 113 (c 0.30, CHCl3); IR (film): νmax = 3398, 2964, 1760, 1712,
(17), 230 (34), 218 (18), 215 (15), 149 (11).
1457, 1230, 1118 cm−1; HR‑MS (EI): m/z = 660.3018 (calcd. for
Methylcarnosate (1): Yellow oil; [α]2D0: + 152 (c 0.10, CHCl3); IR
(film): νmax = 3406, 2948, 1668, 1425, 1158 cm−1; HR‑MS (EI): m/
z = 346.2158 (calcd. for C21H30O4: 346.2144); EI‑MS: m/z (rel. int.
C
41H44 N2O6: 660.3199); EI‑MS: (rel. int. %) m/z = 660 [M] (6),
+
487 (8), 486 (26), 485 (9), 346 (6), 286 (15), 157 (39), 130 (100),
129 (15).
+
%) = 346 [M] (12), 287 (22), 286 (100), 271(10), 243 (13), 230
12-O-indolebutirylmethylcarnosate (13): Colorless amorphous
solid; m.p. 155–157°C; [α]2D0: + 89 (c 0.46, CHCl3); IR (film):
(27), 218 (15), 217 (11), 215 (13), 204 (11).
11,12-O-dimethylmethylcarnosate (2): Yellow oil; [α]2D0: + 126 (c
ν
max = 3422, 2952, 1744, 1708, 1672, 1453, 1222, 1118 cm−1
;
0.19, CHCl3); IR (film): νmax = 2964, 1724, 1445, 1162 cm−1
;
HR‑MS (EI): m/z = 554.2883 [M + Na] (calcd. for C33H41NO5Na:
554.2852); EI‑MS: (rel. int. %) m/z = 531 [M] (6), 513 (15), 346
+
+
HR‑MS (EI): m/z = 374.2437 (calcd. for C23H34O4: 374.2457);
EI‑MS: (rel. int. %) m/z = 374 [M] (73), 316 (21), 315 (100), 314
(29), 273 (10), 245 (41), 233 (31), 229 (14), 219 (90), 206 (37).
11,12-O-diacetylcarnosic acid (3): Colorless crystals; m.p. 213–
215°C; [α]2D0: + 50 (c 0.45, CHCl3); IR (film): νma = 3374, 2928,
1780, 1688, 1366, 1206 cm−1; HR‑MS (EI): m/z = 416.2163 (calcd.
+
(14), 286 (93), 230 (22), 217 (17), 186 (32), 130 (48), 85 (71), 83
(100).
12-O-methylcarnosic acid (14): Colorless amorphous solid;
m.p. 155–157°C; [α]D20: + 173 (c 0.49, CHCl3); IR (film): νmax
=
3378, 2964, 1692, 1421, 1234, 1023 cm−1; HR‑MS (EI): m/
z = 346.2135 (calcd. for C21H30O4: 346.2144); EI‑MS: (rel. int. %)
+
for C24H32O6: 416.2199); EI‑MS: (rel. int. %) m/z = 416 [M] (2),
+
329 (10), 328 (41), 287 (21), 286 (100), 230 (19), 218 (10), 204 (8).
m/z = 346 [M] (13), 301 (23), 300 (100), 285 (15), 257 (9), 244
11,12-O-diacetylmethylcarnosate (4): Yellow oil; [α]2D0: + 139 (c
(23), 232 (10), 229 (8).
0.50, CHCl3); IR (film): νmax = 2961, 1778, 1717, 1370, 1213 cm−1
;
12-O-methylmethylcarnosate (15): Dark yellow oil; [α]2D0: + 114 (c
0.34, CHCl3); IR (film): νmax = 3422, 2956, 1724, 1417, 1226,
1162 cm−1; HR‑MS (EI): m/z = 360.2304 (calcd. for C22H32O4:
360.2301); EI‑MS: (rel. int. %) m/z = 360 [M] + (28), 314 (20), 301
(45), 300 (100), 285 (18), 257 (13), 244 (33), 232 (15), 231 (16),
229 (20).
HR‑MS (EI): m/z = 430.2362 (calcd. for C25H34O6: 430.2355);
EI‑MS: (rel. int. %) m/z = 430 [M] (5), 371 (20), 370 (69), 346
(25), 328 (26), 287 (31), 286 (100), 230 (12), 215 (9).
11,12-O-dichloroacetylmethylcarnosate (5): Yellow oil; [α]D20
+ 122 (c 0.49, CHCl3); IR (film): νmax = 2964, 1788, 1764, 1716,
1421, 1226, 1188 cm−1; HR‑MS (EI): m/z 498.2074 (calcd. for
C25H32Cl2O6: 498.2076); EI‑MS: (rel. int. %) m/z = 498 [M] (2),
+
:
12-O-methyl-11-O-propylmethylcarnosate (16): Yellow oil; [α]D20
+ 84 (c 0.31, CHCl3); IR (film): νmax = 3417, 2961, 1726, 1461,
:
+
442 (13), 441 (22), 440 (68), 439 (34), 438 (100), 369 (13), 364
(17), 363 (15) 362 (47), 357 (13), 345 (28), 309 (19), 286 (11),
267 (14), 230 (10), 215 (15).
1217, 1165 cm−1
;
HR‑MS (EI): m/z = 402.2685 (calcd. for
+
C
25H38O4: 402.2770); EI‑MS: (rel. int. %) m/z = 402 [M] (100),
371 (20), 344 (22), 343 (90), 342 (36), 328 (15), 301 (18), 300
(49), 285 (17), 273 (20), 247 (70), 234 (43), 229 (16), 55 (16).
12-O-propylmethylcarnosate (6): Pale yellow oil; [α]2D0: + 90 (c
0.31, CHCl3); IR (film): νmax = 3442, 2957, 1760, 1712, 1429,
11-O-acetyl-12-O-methylmethylcarnosate (17): Yellow oil; [α]D20
:
1230, 1166 cm−1
;
HR‑MS (EI): m/z = 402.2421 (calcd. for
+ 129 (c 0.28, CHCl3); IR (film): νmax = 2956, 1780, 1716, 1481,
+
C24H34O5: 402.2406); EI‑MS: (rel. int. %) m/z 402 [M] (5), 360
(10), 346 (15), 300 (21), 287 (22), 286 (100), 284 (21), 230 (12),
215 (16), 57 (15).
11,12-O-dibutylmethylcarnosate (7): Colorless oil; [α]2D0: + 141 (c
0.27, CHCl3); IR (film): νmax = 2968, 1764, 1720, 1457, 1230,
1146 cm−1; HR‑MS (EI): m/z = 486.300 (calcd. for C29H42O6:
1218, 1198 cm−1
;
HR‑MS (EI): m/z = 402.2398 (calcd. for
+
C
24H34O5: 402.2406); EI‑MS: (rel. int. %) m/z = 402 [M] (10),
360 (14), 342 (14), 301 (25), 300 (100), 285 (7), 244 (13), 229
(6), 205 (5).
11-O-propyl-12-O-methylpropylcarnosate (18): Colorless amor-
phous solid; mp 234–237°C; [α]2D0: + 81 (c 0.20, CHCl3); IR (film):
ν ; HR‑MS (EI): m/
max = 2960, 1724, 1461, 1162, 1071 cm−1
+
486.2981); EI‑MS: (rel. int. %) m/z = 486 [M] (2), 399 (11), 398
(36), 346 (31), 287 (22), 286 (100), 230 (7), 215 (6), 71 (14).
12-O-isobutylmethylcarnosate (8): Colorless oil; [α]2D0: + 141 (c
0.24, CHCl3); IR (film): νmax = 3438, 2964, 1756, 1712, 1465,
z = 430.3100 (calcd. for C27H42O4: 430.3083); EI‑MS: (rel. int. %)
m/z = 430 [M] + (52), 344 (25), 343 (100), 342 (28), 300 (20), 273
(14), 261 (12), 247 (59), 234 (16).
The structures of the compounds were authenticated by compar-
ing spectroscopic data with those published for carnosic acid [13,
20], methylcarnosate (1) [21,22], 11,12-O-dimethylcarnosate (2)
[23], 11,12-O-diacetylcarnosic acid (3) [14,24], 12-O-methylcar-
nosic acid (14) [11], 12-O-methylcarnosate (15) [22], and 11-O-
acetyl-12-O-methylcarnosate (17) [13].
1235, 1126 cm−1
;
HR‑MS (EI): m/z = 416.2578 (calcd. for
C
25H36O5: 416.2563); EI‑MS: (rel. int. %) m/z = 416 [M] + (5), 398
(13), 356 (9), 346 (19), 287 (28), 286 (100), 230 (12), 71 (16).
11,12-O-dibenzylmethylcarnosate (9): Colorlessresin;[α]2D0: + 73 (c
0.07, CHCl3); IR (film): νmax = 2968, 1784, 1740, 1720, 1453, 1214,
1087 cm−1; HR‑MS (EI): m/z = 554.2678 (calcd. for C35H38O6:
+
554.2668); EI‑MS: (rel. int. %) m/z = 554 [M] (2), 495 (11), 494
(27), 432 (14), 389 (9), 267 (13), 106 (8), 105 (100), 77 (12).
11,12-O-dinicotylmethylcarnosate (10): Yellow oil; [α]2D0: + 100 (c
0.20, CHCl3); IR (film): νmax = 2960, 1752, 1716, 1417, 1262,
1083 cm−1; HR‑MS (EI): m/z = 556.2550 (calcd. for C33H36N2O6:
556.2573): EI‑MS: (rel. int. %) m/z = 556 [M] (3), 497 (27), 496
(50), 433 (16), 390 (9), 267 (7), 106 (100), 78 (23).
Supporting information
1
The H NMR and 13C NMR data of compounds 5–13, 16, and 18
are presented in Table 1S and Table 2S for 1H and Table 3S and
Table 4S for 13C, respectively. Details regarding preparation of
the derivative compounds are available as Supporting Informa-
tion as well.
+
11,12-O-diphenylacetylmethylcarnosate (11): Colorless amor-
phous solid; m.p. 152–153°C; [α]2D0: + 92 (c 0.56, CHCl3); IR (film):
ν
max = 2960, 1760, 1712, 1457, 1218, 1122 cm−1; HR‑MS (EI): m/
Theoduloz C et al. Gastroprotective Effect and… Planta Med 2011; 77: 882–887

886 Original Papers
Table 1 Dose-response gastroprotective effect of carnosic acid; gastroprotective effect at 10 mg/kg and cytotoxicity (IC50, µM) of derivatives 1–18.
Compound
Lesion index (mm)a
% Lesion reduction
Cytotoxicity (IC50 SD, µM)b
Carnosic acid
5 mg/kg
35.3 9.5
22
49
62
77
–
MRC-5
AGS
Hep G2
10 mg/kg
23.2 11.5
17.1 8.2*
10.2 5.5**
45.1 8.3
21.8 3.2
89.8 5.2
41.3 2.6
20 mg/kg
40 mg/kg
Control
–
–
–
1
27.9 7.2
41
47
52
49
70
49
57
71
78
50
61
73
61
83
60
61
62
73
64
18.3 2.4
56.7 3.8
91.5 4.6
18.3 1.9
15.4 1.4
17.5 2.1
48.4 3.1
80.6 4.4
167.2 10.8
46.6 3.3
224.2 6.2
124.1 4.8
2
24.8 5.2
3
22.4 6.3*
24.0 8.2
4
> 1000
> 1000
5
14.0 7.1**
24.1 7.3
17.7 2.8
23.8 2.2
31.2 2.3
39.9 3.1
6
7
20.1 4.8*
13.6 4.4**
10.5 4.8**
23.4 9.6
> 1000
21.5 2.5
> 1000
51.5 2.8
> 1000
> 1000
77.8 3.6
> 1000
47.9 2.9
> 1000
28.3 2.6
> 1000
> 1000
203.3 12.3
> 1000
23.3 2.2
> 1000
> 1000
43.9 3.8
8
9
10
11
18.3 6.5*
12.9 4.3**
18.3 7.5*
8.2 4.0**
18.8 7.3*
18.4 8.7*
17.6 9.7*
12.8 7.2**
16.4 5.4**
12
649.1 32.8
40.1 2.9
45.2 3.1
90.3 4.4
491.3 24.9
52.7 3.2
873.4 56.7
162 11
13
14
54.9 3.9
33.4 1.9
756.2 25.5
86.1 5.4
43.5 4.2
27.7 2.6
15
16
> 1000
91.4 6.3
> 1000
221 13
17
18
> 1000
306 16
Lansoprazole
(10 mg/kg)
a Results are expressed as mean values SD. n = 7. Kruskal-Wallis test followed by the nonparametric multiple comparison Nemenyiʼs test. * P < 0.05; ** p < 0.01 compared to con-
trol group. b Each concentration was tested in quadruplicate together with the control and repeated three times in separate experiments
Results
tective action of carnosic acid methyl ester, as the effect of com-
pounds 1 and 2 are in the same range of lesion reduction (41 and
47%, respectively), neither the placement of two acetoxy func-
tions at C-11 and C-12 with a free or methylated carboxylic acid
at C-20 (compounds 3 and 4). When an electron withdrawing
substituent as chloroacetic acid (compound 5) was placed at C-11
and C-12, the gastroprotective effect increases compared with
the corresponding acetate (compound 4). The effect elicited by
compound 5 was similar to that presented by the isobutyrate at
C-12 with a free hydroxyl group at C-11 (compound 8), reducing
the gastric ulcers by 70 and 71%, respectively.
The compounds presenting better gastroprotective effect with
low-cell toxicity were the dibenzoate 9, diindoleacetate 12, and
the compound 18. The diphenylacetate 11 with some less gastro-
protective activity was also nontoxic to all three cell lines used.
The presence of nicotinate 10 increases cytotoxicity without clear
selectivity among the cell lines.
!
Starting from carnosic acid (CA) and 12-O-methylcarnosic acid
(14), 17 derivatives were prepared, including ethers, short-chain
aliphatic esters (between one and four carbons), as well as esters
with aromatic acids and heterocycles.
The compounds were then assessed for gastroprotective effect
and cytotoxicity. The results are summarized in l Table 1. A
"
dose-response study of CA in mice showed a strong, dose-depen-
dent gastroprotective effect, reducing gastric lesions in about
50% at the single oral dose of 10 mg/kg. Under the same experi-
mental conditions and at 10 mg/kg, the reference drug lansopra-
zole reduced the lesions by 64%. For comparison purposes and to
draw structure-activity relationships, the 18 derivatives of CA
were evaluated at a single oral dose of 10 mg/kg in mice.
Methylation of the carboxylic acid function decreased the gastro-
protective effect, as can be deduced comparing the activity of the
compound 14 with free carboxylic acid and the methyl ester 15,
both presenting a methyl ether at C-12, as well as comparing the
effect of CA and the methyl ester 1. The methoxy group at C-12
also increases the protective effect on gastric lesions, both with
a free acid function at C-20 or the corresponding methyl ester.
However, the presence of a propylether or acetate at C-11 does
not seem to influence the effect on the ulcers, since compounds
15, 16, and 17 presented nearly the same gastroprotective effect.
Some better effect was observed for the derivative 18 bearing a
propylester at C-20 and a propylether at C-11. Some of the new
derivatives at 10 mg/kg were as effective as or even more active
than lansoprazole at 10 mg/kg preventing gastric lesions in mice.
Methylation of both phenolic OH groups at C-11 and C-12 to the
methyl ethers does not have a significant effect on the gastropro-
Considering the short-chain esters 3–8, only the dibutyrate 7 was
nontoxic to the cells, with IC50 > 1000 µg/mL. The isobutyrate at
C-11 (compound 8), however, was cytotoxic with a stronger effect
on normal cells.
Discussion
!
Carnosic acid is one of the gastroprotective constituents of the
widely used medicinal plants R. officinalis and S. officinalis [25–
27]. When the gastroprotective effect of carnosic acid is com-
pared with that of the abietane diterpene ferruginol [17,28–30]
bearing a single phenolic OH in the aromatic ring, the effect is
higher in the more oxidized CA. The relevance of the free phe-
Theoduloz C et al. Gastroprotective Effect and… Planta Med 2011; 77: 882–887

Original Papers 887
11 Richheimer SL, Bernart MW, King GA, Kent MC, Bailey DT. Antioxidant
activity of lipid-soluble phenolic diterpenes from rosemary. J Am Oil
Chem Soc 1996; 73: 507–514
12 Erkan N, Ayranci G, Ayranci E. Antioxidant activities of rosemary (Ros-
marinus officinalis L.) extract, blackseed (Nigella sativa L.) essential
oil, carnosic acid, rosmarinic acid and sesamol. Food Chem 2008; 110:
76–82
13 Aoyagi Y, Takahashi Y, Satake Y, Takeya K, Aiyama R, Matsuzaki T,
Hashimoto S, Kurihara T. Cytotoxicity of abietane diterpenoids from
Perovskia abrotanoides and of their semisynthetic analogues. Bioorg
Med Chem 2006; 14: 5285–5291
14 Gonzalez AG, Andres LS, Herrera JR, Luis JG, Ravelo AG. Abietane diter-
penes with antibiotic activity from the flowers of Salvia canariensis.
Reaction of galdosol with diazomethane. Can J Chem 1989; 67: 208–
212
15 Pertino M, Rodríguez JA, Theoduloz C, Yáñez T, Lazo V, Schmeda-Hirsch-
mann G. Gastroprotective effect of carnosic acid γ‑lactone derivatives.
J Nat Prod 2010; 73: 639–643
16 Pertino M, Schmeda-Hirschmann G, Rodríguez JA, Theoduloz C. Gastro-
protective effect and cytotoxicity of semisynthetic jatropholone deriv-
atives. Planta Med 2007; 73: 1095–1100
17 Areche C, Rodríguez JA, Razmilic I, Yáñez T, Theoduloz C, Schmeda-Hirsch-
mann G. Gastroprotective and cytotoxic effect of semisynthetic ferrugi-
nol derivatives. J Pharm Pharmacol 2007; 59: 289–300
18 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: 1–213
nolic hydroxy groups in gastroprotection is clear when including
in the comparison carnosic acid γ‑lactone (CAL), obtained by an
intramolecular esterification of CA [15] and showing a weaker ef-
fect than CA as a gastroprotective agent. This fact suggests that
antioxidant mechanisms might be involved in the gastroprotec-
tion of this group of diterpenes. Ferruginol was found to display
a powerful effect protecting erythrocyte membranes from lipo-
peroxidation [28], and CA is a known antioxidant [10–12].
The cytotoxicity can also be related with the oxidation pattern
when comparing the IC50 values of CA towards MRC-5, AGS, and
HepG2 cells (21.8, 89.8, and 41.3 µM, respectively) with that of
carnosic acid γ‑lactone (CAL) and ferruginol. The cytotoxicity of
CAL (IC50 values) was 293, 246, and 240 µM towards MRC-5,
AGS, and HepG2 cells, respectively, while ferruginol presented
cell toxicity values closer to CA (23.0, 27.0, and 68.5 µM for MRC-
5, AGS, and HepG2 cells, respectively) [30]. In the present work,
we have shown that the gastroprotective effect and cytotoxicity
of CA can be modulated by placing different substituents either
at the phenolic hydroxyl groups or at the carboxylic acid function
at C-20. Further studies should be undertaken to disclose the ac-
tion mechanisms of the CA derivatives.
19 Rodríguez JA, Haun M. Cytotoxicity of trans-dehydrocrotonin from Cro-
ton cajucara (Euphorbiaceae) on V79 cells and rat hepatocytes. Planta
Med 1999; 65: 522–526
Acknowledgements
!
20 Rodriguez-Meizoso I, Cifuentes A, San Roman J, Ibañez E, Elvira CA. A sys-
tematic study on the interactions between carnosic acid and ethylpyr-
rolidine methacrylate methyl methacrylate copolymer in supercritical
media. J Supercrit Fluids 2007; 41: 452–460
21 Cantrell CL, Richheimer SL, Nicholas GM, Schmidt BK, Bailey DT. Seco-hi-
nokiol, a new abietane diterpenoid from Rosmarinus officinalis. J Nat
Prod 2005; 68: 98–100
22 Djarmati Z, Jankov RM, Csanádi J, Djordjevic A. The isolation of carnosic
acid 12-methyl ether from Salvia officinalis L. and NMR study of its
methyl ester. Collect Czech Chem Commun 1993; 58: 1919–1924
23 Dobrynin VN, Kolosov MN, Chernov BK, Derbentseva NA. Antimicrobial
substance of Salvia officinalis. Khim Prir Soedin 1976; 5: 686–687
24 González AG, Rodríguez CM, Luis JG. Diterpenes from the flowers of Sal-
via canariensis. Phytochemistry 1987; 26: 1471–1474
This work was supported by FONDECYT (Grant N° 1060841) and
Programa de Investigación en Productos Bioactivos, Universidad
de Talca. M.W.P. thanks the PBCT PSD50 for a postdoctoral grant.
We thank Prof. A. Palenzuela, Universidad de La Laguna, Tenerife,
Spain, for the HR‑MS. The skillful technical assistance of Tania
Yáñez is kindly acknowledged.
References
1 Troncoso N, Sierra H, Carvajal L, Delpiano P, Günther G. Fast high perfor-
mance liquid chromatography and ultraviolet-visible quantification of
principal phenolic antioxidants in fresh rosemary. J Chromatogr A
2005; 1100: 20–25
25 Akihisa T, Yasukawa K, Tokuda H. Potentially cancer chemopreventive
and anti-inflammatory terpenoids from natural sources. In: Rahman
A–U, editor. Studies in Natural Products Chemistry, Vol. 29. Cambridge:
Elsevier BV; 2003: 73–125
26 Pukalskas A, Beek TA, de Waard P. Development of a triple hyphenated
HPLC-radical scavenging detection–DAD‑SPE‑NMR system for the rap-
id identification of antioxidants in complex plant extracts. J Chroma-
togr A 2005; 1074: 81–88
2 Caruso JL, Callahan J, De Chant C, Jayasimhulu, K, Winget GD. Carnosic
acid in green callus and regenerated shoots of Rosmarinus officinalis.
Plant Cell Rep 2000; 19: 500–503
3 Dias PC, Foglio MA, Possenti A, de Carvalho JE. Antiulcerogenic activity
of crude hydroalcoholic extract of Rosmarinus officinalis L. J Ethno-
pharmacol 2000; 69: 57–62
4 Miyazaki T, Kosaka K, Ito H. Method of treating ulcers. US Patent
6638523; 2003
27 Mayer B, Baggio CH, Freitas CS, dos Santos AC, Twardowschy A, Horst H,
Pizzolatti MG, Micke GA, Heller M, Pereira dos Santos E, Otuki MF, An-
drade Marques MC. Gastroprotective constituents of Salvia officinalis
L. Fitoterapia 2009; 80: 421–426
28 Rodríguez J, Theoduloz C, Yáñez T, Becerra J, Schmeda-Hirschmann G.
Gastroprotective effect of the diterpene ferruginol in mice: protection
against membrane lipid peroxidation and involvement of prostaglan-
dins. Life Sci 2006; 78: 2503–2509
29 Areche C, Theoduloz C, Yañez T, Souza-Brito ARM, Barbastefano V, de
Paula D, Ferreira AL, Schmeda-Hirschmann G, Rodríguez JA. Gastropro-
tective activity of ferruginol in mice and rats: effects on gastric secre-
tion, endogenous prostaglandins and non-protein sulfhydryls. J Pharm
Pharmacol 2008; 60: 245–251
30 Espinoza M, Santos LS, Theoduloz C, Schmeda-Hirschmann G, Rodriguez
JA. New gastroprotective ferruginol derivatives with selective cytotox-
icity against gastric cancer cells. Planta Med 2008; 74: 802–808
5 Del Campo J, Amiot MJ, Nguyen-The C. Antimicrobial effect of rosemary
extracts. J Food Prod 2000; 63: 1359–1368
6 Almela L, Sánchez-Muñoz B, Fernández-López JA, Roca MJ, Virginia Rabe
V. Liquid chromatograpic-mass spectrometric analysis of phenolics and
free radical scavenging activity of rosemary extract from different raw
material. J Chromatogr A 2006; 1120: 221–229
7 Herrero M, Plaza M, Cifuentes A, Ibáñez E. Green processes for the ex-
traction of bioactives from rosemary: Chemical and functional charac-
terization via ultra-performance liquid chromatography-tandem mass
spectrometry and in-vitro assays. J Chromatogr A 2010; 1217: 2512–
2520
8 Ninomiya K, Matsuda H, Shimoda H, Nishida N, Kasajima N, Yoshino T,
Morikawa T, Yoshikawa M. Carnosic acid, a new class of lipid absorp-
tion inhibitor from sage. Bioorg Med Chem Lett 2004; 14: 1943–1946
9 Oluwatuyi M, Kaatz GW, Gibbons S. Antibacterial and resistance modi-
fying activity of Rosmarinus officinalis. Phytochemistry 2004; 65:
3249–3254
10 Poeckel D, Greiner C, Verhoff M, Rau O, Tausch L, Hörnig C, Steinhilber D,
Schubert Zsilavecz M, Werz O. Carnosic acid and carnosol potently in-
hibit human 5 lipoxygenase and suppress pro-inflammatory responses
of stimulated human polymorphonuclear leukocytes. Biochem Phar-
macol 2008; 76: 91–97
Theoduloz C et al. Gastroprotective Effect and… Planta Med 2011; 77: 882–887