Biovalorization of friedelane triterpenes derived from cork processing industry byproducts

Article abstract of DOI:10.1021/jf0531151

Here, we describe the synthesis, bioactivity screening, and structure-activity relationships of various synthetic triterpenoids prepared from the cork processing byproducts friedelin (1) and 3-hydroxyfriedel-3-en-2- one (2) via oxidative procedures. The synthesis of compounds 2α-trimethylsiloxyfriedelan-3-one (17), friedelin-2,3-lactone (18), friedelin-3-oxime (19), and friedelin-3,4-lactam (20) is also described. We have studied the insecticidal and phytotoxic potential of these compounds, their selective cytotoxic effects on insect and mammalian cells, and their antiparasitic effects. Structural modifications of the A-ring of friedelin (1) improved its insecticidal activity with derivatives 5, 2,3-secofriedelan-2-al-3- oic acid (6), its acetylated derivative 6a, 3β- and 3α- hydroxyfriedelane (9 and 10), 3α-hydroxyfriedel-2-one (11), 4β-hydroxyfriedel-3-one (16), the acetylated 10a, 3,4-secofriedelan-4-oxo- 3-oic-acid (14), lactone 18, and the oxime 19 being stronger insecticides than the parent compound. Methyl-3-nor-2,4-secofriedelan-4-oxo-2-oic acid (12) and its acetylated derivative 12a also showed insecticidal activity in contrast to their inactive parent compound 2. The postingestive effects and cytotoxicity of these compounds suggest a multifaceted insecticidal mode of action. These structural modifications did not result in better phytotoxic agents than the parent compounds except for lactam 20 and yielded several moderately active antiparasite derivatives (seco acids 6, 12, 14, and 4β -hydroxyfriedel-3- one 16) with cytotoxic effects on mammalian cells.

Full text of DOI:10.1021/jf0531151

3566 J. Agric. Food Chem. 2006, 54, 3566−3571
Biovalorization of Friedelane Triterpenes Derived from Cork
Processing Industry Byproducts
†
†
‡
CRISTINA MOITEIRO, MARIA JO A˜ O MARCELO CURTO, NAGLA MOHAMED,
MAR ´ı A BAIL EÄ N, RAFAEL MART ´ı NEZ-D ´ı AZ,§ AND AZUCENA GONZ AÄ LEZ-COLOMA*,‡
‡
INETI, Instituto Nacional de Engenharia, Tecnologia e Inova c¸ a˜ o I.P., Estrada do Pa c¸ o do Lumiar,
1
649-038 Lisboa, Portugal, Instituto de Ciencias Agrarias, CCMA, CSIC, Serrano 115-dpdo,
§
2
8006 Madrid, Spain, and Departamento de Medicina Preventiva, Salud P u´ blica y Microbiolog ´ı a,
Facultad de Medicina, Universidad Aut o´ noma de Madrid, Spain
Here, we describe the synthesis, bioactivity screening, and structure-activity relationships of various
synthetic triterpenoids prepared from the cork processing byproducts friedelin (1) and 3-hydroxyfriedel-
3-en-2-one (2) via oxidative procedures. The synthesis of compounds 2R-trimethylsiloxyfriedelan-3-
one (17), friedelin-2,3-lactone (18), friedelin-3-oxime (19), and friedelin-3,4-lactam (20) is also
described. We have studied the insecticidal and phytotoxic potential of these compounds, their
selective cytotoxic effects on insect and mammalian cells, and their antiparasitic effects. Structural
modifications of the A-ring of friedelin (1) improved its insecticidal activity with derivatives 5, 2,3-
secofriedelan-2-al-3-oic acid (6), its acetylated derivative 6a, 3â- and 3R-hydroxyfriedelane (9 and
10), 3R-hydroxyfriedel-2-one (11), 4â-hydroxyfriedel-3-one (16), the acetylated 10a, 3,4-secofriedelan-
4-oxo-3-oic-acid (14), lactone 18, and the oxime 19 being stronger insecticides than the parent
compound. Methyl-3-nor-2,4-secofriedelan-4-oxo-2-oic acid (12) and its acetylated derivative 12a also
showed insecticidal activity in contrast to their inactive parent compound 2. The postingestive effects
and cytotoxicity of these compounds suggest a multifaceted insecticidal mode of action. These
structural modifications did not result in better phytotoxic agents than the parent compounds except
for lactam 20 and yielded several moderately active antiparasite derivatives (seco acids 6, 12, 14,
and 4â -hydroxyfriedel-3-one 16) with cytotoxic effects on mammalian cells.
KEYWORDS: Quercus suber; cork smoker wash solids; triterpene friedelanes; Spodoptera littoralis;
Lactuca sativa; Trypanosoma cruzi; Leishmania infantum; antifeedant; phytotoxic; antiparasitic
INTRODUCTION
process of making corkboard from ground cork under pressure
by treatment with superheated steam. This black wax has no
practical application and represents an environmental problem
Biobased products derived from nature contribute to sustain-
able industrial growth in sectors such as health care, agrifood,
forest products, chemicals, plastics, energy, and the environment.
The cork industry comprises one of the world’s most important
biobased, nontimber forest products. Natural cork is a renewable
resource harvested from the living bark of the cork oak, Quercus
suber L. (Fagaceae), with minimal environmental impact. This
tree is native to the Mediterranean forest with Portugal and Spain
being among the main producers worldwide. The Quercus forest
provides immeasurable ecological value to the fragile ecosys-
tems of its arid native habitat (1, 2).
(Comiss a˜ o de Coordena c¸ a˜ o e Desenvolvimento Regional de
Lisboa e Vale do Tejo” Portuguese Environmental Department).
This residue is a convenient source of friedelane triterpenes
friedelin (1) (8%, w/w) and 3-hydroxyfriedel-3-en-2-one (2)
(
(
5.4%, w/w). Small amounts of â-sitosterol (1%), campesterol
0.3%), â-amyrin and sitost-4-en-3-one (0.3%), and waxes have
also been isolated from this residue (3). Triterpenoids are
abundant in the plant kingdom, and pharmacological (antiin-
flammatory, antibacterial, antifungal, antiviral, cytotoxic, and
antiparasitic) and pesticidal (insecticidal, phytotoxic) effects have
been reported (4).
As part of our ongoing assessment of cork-derived biobased
products, we previously prepared compounds 3-16 by structur-
ally modifying the A-ring of triterpenes 1 and 2 (Scheme 1),
including seco acid derivatives with antitumoral activity (3, 5).
In this work, we have obtained a mixture of compounds 5,
6, 11, and the new derivative 17 from compound 3 via oxidative
The cork processing industry generates large amounts of cork
powder and cork smoker wash solids (ca. 200 tons/year of cork
smoker wash solids in Portugal), a black wax obtained in the
*
To whom correspondence should be addressed. Tel: 34-91-7452500.
Fax: 34-91-5640800. E-mail: azu@ccma.csic.es.
†
INETI.
‡
Instituto de Ciencias Agrarias.
§
Universidad Aut o´ noma de Madrid.
1
0.1021/jf0531151 CCC: $33.50
© 2006 American Chemical Society
Published on Web 04/25/2006

Biovalorization of Friedelane Triterpenes
J. Agric. Food Chem., Vol. 54, No. 10, 2006 3567
Scheme 1. Friedelane Derivatives
mass spectra (EIMS) were recorded on a Kratos model MS-25 RF mass
spectrometer at 70 eV. Thin-layer chromatography was carried out using
precoated silica gel plates 60 F254, and detection was achieved by
spraying with methanolic DNP and heating to 150 °C. Column
chromatography was performed using silica gel 60 (0.040-0.063
Merck, 9385). MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazo-
lium bromide] was from Sigma-Aldrich. Cell and parasite viability were
measured in a microplate reader SLT Lab Instruments (Austria).
Material. Cork smoker wash solids were a gift from Corticeira
Amorim Algarve (Portugal).
Extraction and Isolation. Friedelin (1) and 3-hydroxyfriedel-3-en-
2
-one (2) were extracted from powdered cork smoker wash solids (3).
Synthesis. Details on the synthesis of compounds 3-16 and their
physical and spectral data have been previously reported (3, 5). Here,
we scaled-up the synthetic method to prepare compound 5. The
synthesis of compounds 17-20 is also described along with their
spectroscopic data.
Synthesis of 2r-Hydroxyfriedel-3-one (5). A solution of NaHCO
0.5 M, 70 mL) at 0 °C was added to 3-trimethylsiloxyfriedel-2-ene
3) (932 mg) dissolved in CH Cl (117 mL), and the mixture was
3
(
(
2
2
vigorously stirred at room temperature. Then, m-CPBA (776,7 mg) was
added at 0 °C and the mixture was stirred at room temperature for 1 h.
The organic phase was extracted and washed with a saturated solution
of sodium sulfite, NaHCO
was purified by column chromatography (silica gel with a CH
MeOH gradient) to obtain a mixture of 2R-hydroxyfriedel-3-one (5)
28.3 mg; 2.7%), 2,3-secofriedelan-2-al-3-oic acid (6) (133.4 mg;
2.4%), 3R-hydroxy-friedel-2-one (11) (548 mg; 53.2%), 2R-trimeth-
3
(0.5 M), and H
2
O, and the residue (1.20 g)
2
Cl
2
:
(
1
ylsiloxyfriedelan-3-one (17) (41.7 mg; 3.5%), and friedelin (1) (55.8
mg; 5.6%). Compound 5 isomerized in the presence of silica gel to
1
1. Compound 17 was obtained for the first time in trace amounts.
r-Trimethylsiloxyfriedelan-3-one (17). Amorphous brown solid.
H NMR (CDCl ): 0.10 (9H, s, OSiMe ), 0.66 (3H, s, Me-24), 0.83
3H, s, Me-25), 0.84 (3H, d, J ) 6,4 Hz, Me-23), 0.95 (3H, s, Me-30),
2
1
3
3
(
1
2
.00 (6H, s, Me-26 and Me-29), 1.07 (3H, s, Me-27), 1.17 (3H, s, Me-
8), 1.90 (1H, m, H-10), 2.87 (1H, q, J ) 6.4 Hz, H-4), 3.96 (1H, t, J
)
3.0 Hz, H-2). 13C NMR (CDCl
3
): 0 (O SiMe
3
), 32.16 (C-1), 74.89
(
C-2), 212.73 (C-3), 52.45 (C-4), 43.03 (C-5), 40.93 (C-6), 18.21 (C-
7
1
), 52.90 (C-8), 36.72 (C-9), 51.91 (C-10), 35.28 (C-11), 30.35 (C-
2), 39.66 (C-13), 38.28 (C-14), 32.74 (C-15), 35.98 (C-16), 29.93
(
(
(
(
[
(
C-17), 42.76 (C-18), 35.28 (C-19), 28.10 (C-20), 32.29 (C-21), 39.17
C-22), 6.28 (C-23), 14.00 (C-24), 18.05 (C-25), 18.54 (C-26), 20.07
C-27), 32.05 (C-28), 31.79 (C-29), 34.92 (C-30). EIMS m/z (%): 514
+
+
+
52) [M] , 499 (43) [M - CH
3
] , 424 (4) [M - SiMe
3
OH] , 409 (5)
+
+
+
M - (SiMe
3
O + O)] , 341 (5) [C25
H
41] , 273 (22) [C19H29O] , 205
+
+
40) [C15 ] , 159 (98), 143 (100), 137 (31),
H
25] , 191 (20) [C15
H
25-CH
2
+
1
25 (38), 123 (58) [C H15] , 109 (73), 95 (91), 81 (71), 69 (69), 73
9
procedure. We have also synthesized derivatives 18-20. We
have studied the biopesticide potential (insecticidal and phyto-
toxic) of these compounds on the insect pest Spodoptera
littoralis and Lactuca satiVa seeds. We have also tested their
selective cytotoxic effects on insect Sf9 cells derived from S.
frugiperda pupal ovarian tissue and mammalian chinese hamster
ovary (CHO) cells. Additionally, we studied their antiparasitic
effects against Trypanososma cruzi and Leishmania infantum,
which are associated with diseases responsible for high morbid-
ity and mortality in developing countries. In both cases, the
etiologic agent is a flagellated protozoa and the chemotherapies
used are not effective in all phases of the infection.
+
+
(
89) [Me
Synthesis of Friedelin-2,3-lactone (18). Friedelin (1, 65.8 mg) was
dissolved in CHCl (1.3 mL), and then, m-CPBA (98 mg) was added.
3 3 3
Si] , 55 (77) [C H O] , 43 (45).
3
The mixture was stirred continuously at 60 °C for 17 h. A NaHCO
saturated solution was added to the ice-cooled mixture, the pH was
adjusted to 7, and the organic phase was extracted with CHCl . The
residue was purified by preparative liquid chromatography [silica gel,
CH Cl :(CH CO, 98:2] to obtain friedelin-2,3-lactone (48 mg, 71%
yield).
Friedelin-2,3-lactone (18). White crystals; mp 288-290 °C [lit. (8)
3
3
2
2
3 2
)
-
1
2
47-248 °C]. IV υmax (KBr) (cm ): 2944, 1734 (CdO), 1073 (C-
1
O), 752. H NMR (CDCl
0
2
3
): 0.83 (3H, s, Me-25), 0.89 (3H, s, Me-24),
.95 (3H, s, Me-26), 0.99 (6H, s, Me-27 and Me-30), 1.00 (3H, s, Me-
9), 1.17 (3H, s, Me-28), 1.20 (3H, d, J ) 6,3 Hz, Me-23), 1.94 (1H,
m, H-1ax), 2.52 (1H, td, J ) 1.5, 13.0, 13.0 Hz, H-2ax), 2.63 (1H, ddd,
MATERIAL AND METHODS
1
3
J ) 1.5, 7.0, 13.0 Hz, H-2eq), 4.22 (1H, q, J ) 6,3 Hz, H-4). C NMR
(CDCl ): 18.55 (C-1), 34.35 (C-2), 175.64 (CdO), 84.91(C-4), 40.76
General Experimental Procedures. Melting points were determined
using a Reichert microscopic Thermovar model and are uncorrected.
IR spectra were recorded on a Perkin-Elmer FT-IR 1725X spectro-
3
(C-5), 38.44 (C-6), 18.03 (C-7), 52.72 (C-8), 38.18 (C-9), 63.94 (C-
10), 35.29 (C-11), 30.59 (C-12), 39.33 (C-13), 38.37 (C-14), 32.35
(C-15), 35.40 (C-16), 29.98 (C-17), 42.71 (C-18), 35.95 (C-19), 28.16
(C-20), 32.73 (C-21), 39.22 (C-22), 13.45 (C-23), 16.22 (C-24), 17.90
(C-25), 18.59 (C-26), 20.20 (C-27), 32.05 (C-28), 31.75 (C-29), 35.03
1
13
photometer. H and C NMR spectra were recorded on a Bruker AMX
00 MHz spectrometer using CDCl solutions with SiMe as an internal
3
3
4
13
standard. Multiplicities of C signals were determined by distortionless
enhancement of polarization transfer experiments. All NMR experi-
ments were carried out at constant temperature (298 K). Electron impact
+
+
(C-30). EIMS m/z: 442 (9) [M] , 427 (3) [M - CH
3
] , 398 (17) [M
+
+
- CO
2
] , 383 (6) [M - (CO
2
+ CH
3
)] , 274 (10), 245 (15), 218 (24),

3568 J. Agric. Food Chem., Vol. 54, No. 10, 2006
Moiteiro et al.
2
1
05 (44), 191 (23), 179 (16), 163 (20), 149 (23), 135 (27), 123 (53)
09 (67), 69 (100), 95 (95).
Synthesis of Friedelin-3-oxime (19). Friedelin (1, 80 mg) and
hydroxylamine hydrochloride (59 mg) were dissolved in C
5 5
H N (1 mL)
and absolute C OH (1 mL), and the mixture was refluxed for 3 h
2
H
5
and 30 m. The solvents were evaporated, a solution of HCl (10%) was
added, and the organic phase was extracted with EtOAc and washed
with a solution of NaHCO3. The residue was purified by column
chromatography (silica gel, CH
mg, 85% yield).
2 2
Cl ) to obtain friedelin-3-oxime (70
Figure 1. Natural fridelane triterpenes isolated from cork smoker wash
Friedelin-3-oxime (19). White crystals; mp 280-282 °C [lit. (9)
solids.
-
1
2
1
0
89-292 °C]. IV υmax (KBr) (cm ): 3293 (OH), 2928, 1686 (CdN),
1
458 (C-O), 945 (N-O); H NMR (CDCl
3
): 0.74 (3H, s, Me-24),
.84 (3H, s, Me-25), 0.94 (3H, d, J ) 6.9 Hz, Me-23), 1.00 (3H, s,
Phytotoxicity Tests. These experiments were conducted with L.
satiVa var. Carrascoy seeds placed on paper disks (Whatman no. 1,
Me-30), 1.03 (6H, s, Me-27 and Me-29), 1.17 (3H, s, Me-28), 1.21
2
2
(3H, s, Me-26), 1.77 (1H, m, H-2eq), 2.02 (1H, q, J ) 6.9 Hz, H-4),
2.0 cm ) treated with 50 µg/cm of the test compound or solvent for
the control. The disks were placed in lidded clear plastic boxes (2 ×
13
3
3.43 (1H, ddd, J ) 2.4, 4.5, 13.0 Hz, H-2ax). C NMR (CDCl ): 20.58
2
(C-1), 36.03 (C-2), 162.39 (CdN), 50.96 (C-4), 41.15 (C-5) 39.67 (C-
2 cm ) lined with 4 g of sterilized sand soil humidified with 200 µL of
6
3
3
3
2
4
), 18.35 (C-7), 53.10 (C-8), 37.29 (C-9), 60.00 (C-10), 35.34 (C-11),
0.54 (C-12), 40.30 (C-13), 38.30 (C-14), 32.40 (C-15), 35.57 (C-16),
0.02 (C-17), 42.79 (C-18), 35.90 (C-19), 28.18 (C-20), 32.79 (C-21),
9.27 (C-22), 8.37 (C-23), 14.21 (C-24), 17.97 (C-25), 18.67 (C-26),
0.24 (C-27), 32.09 (C-28), 31.79 (C-29), 35.04 (C-30). EIMS m/z:
deionized water and then placed in a plant growth chamber (25 ( 1
°C, >70% relative humidity, with a photoperiod of 16:8 h L:D) for 6
days. A total of 100 seeds were used (20 seeds/box, five boxes). The
germination was monitored daily, and the radicle length was measured
at the end of the experiment (20 digitalized radicles randomly selected
for each experiment) with the application Scion Image for Windows
release Alpha 4.0.3.2 (www.scioncorp.com). An analysis of variance
+
+
+
41 (11) [M] , 426 (37) [M - CH ] , 410 (48) [M - NOH] , 272
3
(11), 220 (10), 218 (18), 205 (19), 179 (16), 149 (20), 137 (28), 125
21), 109 (58), 95 (81), 81 (77), 69 (97), 55 (27).
(ANOVA) was performed on germination and radicle length data.
(
Cytotoxicity. Sf9 cells derived from S. frugiperda pupal ovarian
Synthesis of Friedelin-3,4-lactam (20). Friedelin-3-oxime (19, 60
tissue (European Collection of Cell Cultures, ECCC) and mammalian
CHO cells (a gift from Dr. Pajares, I. C. Biom e´ dicas, CSIC) were grown
as previously described (7). Cell viability was analyzed by an adaptation
of the MTT colorimetric assay method (7). The active compounds were
tested in a dose-response experiment to calculate their relative potency
5 5
mg) and p-toluenesulfonyl chloride (19.6 mg) were dissolved in C H N
(1.14 mL), and the mixture was refluxed for 5 h. The reaction was
2 3
cooled and diluted with H O, and the product was extracted with CHCl .
The CHCl phase was washed with HCl (10%), NaCl (saturated
3
solution), and H
yield).
2
O to afford pure friedelin-3,4-lactam (59.9 mg, 99.8%
(EC50 values, the effective dose to give 50% cell viability), which was
determined from linear regression analysis (% cell viability on log dose).
Friedelin-3,4-lactame (20). Amorphous yellowish solid; mp 266-
-
1
Trypanocidal Activity. This activity was assayed on epimastigote
forms of T. cruzi, Y strain, as described (7). The active compounds
were tested in a dose-response experiment to calculate their relative
potency (EC50 values, the effective dose to give 50% parasite growth
reduction), which was determined from linear regression analysis (%
cell viability on log dose).
2
68 °C. IV υmax (KBr) (cm ): 2928, 2867, 1672 (CdO), 1458 (N-
1
H), 1362 (C-N), 734 (N-H). H NMR (CDCl
3
): 0.81 (3H, s, Me-
4), 0.82 (3H, s, Me-25), 0.95 (3H, s,Me-30), 0.98 (3H, s, Me-27),
.99 (3H, s, Me-29), 1.00 (3H, s, Me-26), 1.04 (3H, d, J ) 6.9 Hz,
2
0
Me-23), 1.17 (3H, s, Me-28), 1.88 (1H, m, H-1ax), 2.40 (2H, m, H-2ax
a nd H-2eq), 3.27 (1H, quint., J ) 6.4 Hz, H-4), 5.30 (1H, brs, N-H).
13
Leishmanicidal Activity. The leishmanicidal activity was assayed
on promastigote forms of L. infantum, PB75 strain, cultured at 28 °C
in RPMI medium supplemented with fetal calf serum. Parasites in
logarithmic growth phase were distributed in 96-well flat bottom plates.
The compounds to be tested were dissolved in DMSO and added to
the cultures at several concentrations (100, 50, 25, 10, 5, and 1 µg/
mL; except for 1, assayed at 100, 10, and 1 µg/mL) for 72 h.
Amphotericin B was used as a reference drug, and parasite viability
was analyzed by a modified MTT colorimetric assay (7). The activity
was calculated as described for T. cruzi.
3
C NMR (CDCl ): 18.24 (C-1), 36.39 (C-2), 177.01 (C-3), 58.91 (C-
4
6
3
2
1
3
), 39.14 (C-5), 39.14 (C-6), 18.33 (C-7), 52.63 (C-8), 38.30 (C-9),
5.09 (C-10), 35.16 (C-11), 30.57 (C-12), 39.72 (C-13), 38.37 (C-14),
2.27 (C-15), 35.37 (C-16), 29.87 (C-17), 42.60 (C-18), 35.89 (C-19),
8.03 (C-20), 32.64 (C-21), 39.12 (C-22), 15.82 (C-23), 13.75 (C-24),
7.81 (C-25), 18.45 (C-26), 20.07 (C-27), 31.91 (C-28), 31.63 (C-29),
+
+
4.89 (C-30). EIMS m/z (%) 441 (9) [M] , 426 (3) [M - CH
3
] , 398
+
+
+
(2) [M - (CH
3
+ CO)] , 273 (2) [C19
] , 179 (2), 163 (4), 137 (7), 125 (3), 123 (11) [C
H
31N] , 205 (8) [C15
H
25] , 191
+
+
(5) [205 - CH
2
9
H
15] ,
+
1
09 (18), 95 (22), 81 (23), 69 (21), 55 [C
3
3
H O] (27), 42 (100).
Insect Bioassays. S. littoralis and Leptinotarsa decemlineata were
reared on artificial diet (6) and potato foliage, respectively, and
maintained at 22 ( 1 °C, >70% relative humidity, with a photoperiod
of 16:8 h (L:D) in a growth chamber.
RESULTS AND DISCUSSION
The triterpene derivatives studied here were obtained through
simple chemical modifications of natural triterpenes 1 and 2
(Figure 1). These reactions are suitable for possible large-scale
applications (Scheme 1). The conversions focused mainly on
the modification of the carbonyl function at C-3 of friedelin 1,
via controlled silyl enol ethers 3 and 4, which were modified
to hydroxyl ketones (5 and 16) to obtain seco acids 6 and 14,
respectively. Seco acid 12 was prepared by oxidation of 2 in
the presence of a phase transfer catalyst and potassium per-
manganate (Scheme 1), a convenient method for the preparation
of the seco acids.
Choice Feeding Assays. These experiments were conducted with
newly emerged sixth-instar S. littoralis larvae and adult L. decemlineata.
The percent feeding inhibition (%FI) was calculated as previously
described (6). Compounds with an FI > 70% were tested in a dose-
response experiment to calculate their relative potency (EC50 values,
the effective dose for 50% feeding reduction), which was determined
from linear regression analysis (% FI on log dose).
Oral Cannulation. This experiment was performed with preweighed
newly emerged S. littoralis L6 larvae. Each experiment consisted of
20 larvae orally dosed with 40 µg of the test compound (6). An analysis
of covariance (ANCOVA1) on biomass gains with initial biomass as
the covariate (covariate p > 0.05) showed that initial insect weights
were similar among all treatments. A second analysis (ANCOVA2)
was performed on biomass gains with food consumption as the covariate
to test for postingestive effects (6).
Oxidation of 3 with m-chloroperbenzoic acid under buffered
pH conditions produced quantitatively 5 (3). However, the scale-
up of this reaction based on the procedure described above and
the subsequent purification with silica gel gave a mixture of 5

Biovalorization of Friedelane Triterpenes
Table 1. Consumption ( I) and Biomass Gain (∆B) of Orally Injected
S. littoralis L6 Larvae, Expressed as Percent of the Control and
Cytotoxic Effects on Spodoptera frugiperda Sf9 and Mammalian CHO
Cellsa
J. Agric. Food Chem., Vol. 54, No. 10, 2006 3569
ranking as follows: 6a, 10 > 18, 19 > 9, 10a, 11, 16 > 12, 14
> 1, 12a, 17, and 20 (Table 1). Treatment effects on ∆B
disappeared with covariance adjustment except for 6a. There-
fore, these compounds acted as postingestive antifeedants with
varying potencies, while 6a had additional toxic action. Com-
pounds 5 and 6 reduced ∆B but not ∆I. Treatment effects of
these compounds on ∆B did not disappear with covariance
adjustment, indicating that they acted as strong postingestive
toxins without antifeedant effects.
∆
EC50
(µ
g/mL)
compd
∆
B
∆
I
pANCOVA2b
>0.05
Sf9
CHO
1
2
73*
97
102
35**
51*
34**
77*
105
115
80
89
73*
>100
100
>100
>100
>100
∼
2
5
6
6
7
7
8
9
9
1
1
1
1
1
1
1
1
1
1
1
2
a
23.10 (17.77, 30.02)
47.6 (24.31, 93.25)
7.26 (12.63, 23.60)
>100
6.99 (4.62, 10.58)
73.46 (56.36, 95.76)
>100
0.0001
0.007
0.008
41.92 (26.52, 66.27)
9.32 (7.08, 12.27)
>100
27.70 (20.89, 36.72)
52.35 (38.63, 70.94)
>100
A lower number of derivatives had cytotoxic effects to insect-
derived Sf9 and mammalian CHO cells (45 and 40%, respec-
tively). Compound 6 had the strongest cytotoxic effect followed
by 7, 11, 12, 14, 14a, 5, and 7a, while 2a, 9a, and 19 showed
selective and moderate-low cytotoxic effects to Sf9 (Table 1).
Similarly, compound 6 has been reported as a potent inhibitor
of human lymphocyte and cancer cell proliferation, followed
by 5, 7, 12, and 14 (3, 5). Other friedelane, oleanane, lupane,
and ursane triterpenes have also been reported as tumoral cell
growth inhibitors (10-12), but this is the first report on this
class of compounds acting on insect-derived cells.
a
a
79
53*
93
73
>0.05
>100
>100
a
0
0a
1
2
2a
4
4a
6
7
8
79.56 (41.87, 151.21) >100
36*
58*
57*
65*
75*
61*
62*
74*
71*
77*
79*
73*
>0.05
>0.05
>0.05
>0.05
>0.05
>0.05
>100
>100
>100
>100
8.32 (6.25, 11.07)
20.23 (14.28, 28.65)
>100
23.84 (18.78, 30.27)
33.40 (28.17, 39.61)
>100
50> EC50 >25
34.33 (29.41, 40.07)
>100
20.11 (17.98, 22.50)
19.88 (8.30, 47.57)
>100
The lack of short-term feeding behavior modulating the
effects of these terpenes on S. littoralis indicates that the
noncytotoxic insect growth regulators (6a, 10, 18, 19, 9, 16, 1,
12a, and 17) could be digestive toxins, while the cytotoxic ones
(5, 6, 11, 12, and 14) could act unspecifically on cell membranes
in addition to being digestive toxins. Therefore, the growth
regulation effects observed here could be the result of a
multifaceted biological action. Triterpenes such as oleanolic and
ursolic acid have fluidity-modulating effects on liposomal
membranes (13). Asiatic acid reduced the consumption and
growth of the rice grasshopper Oxya fuscoVita and inhibited
the digestive enzymes amylase and invertase (14). Angiotensin
blockers induced ecdisteroid secretion in the lepidopteran
Lymantria dispar (15), and triterpene acids including oleanolic
and â-glycyrrhetinic blocked agiotensin II receptors (16),
suggesting that these triterpenes could interfere with the insect’s
ecdysteriod levels.
53*
77
44*
47*
73*
65*
65*
46*
49*
77*
>0.05
>100
>100
>100
>0.05
>0.05
>0.05
∼
100
100
9
0
62.55 (27.05, 144.62)
∼
∼
100
>100
a
p < 0.05 and **p < 0.0005, ANCOVA1 (initial larval weight as covariate). b
I as covariate).
*
p
Level, ANCOVA2 (
∆
(2.7%), 11 (53.2%), seco acid 6 (12.4%), and the new reaction
product 17 (3.5%). The purification of the mixture on silica
gel gave the hydroxy ketone 11 by epimerization of 5, which
was formed via an epoxide intermediate.
The selective reduction of the carbonyl function of hydroxyl
ketones 5 and 1 by a mild treatment with NaBH4 produced
alcohols 7 and 9. Additionally, the reduction of 1 with sodium
produced alcohol 10. Alcohols 7, 9, and 10 were subjected to
acetylation (with acetic anhydride and pyridine). The R,â-
unsaturated carbonyl 8 derivative was prepared directly from 1
by phenylselenenyl chloride (PhSeCl) oxidation in 93% yield,
by contrast DDQ oxidation via sylil enol ether 3 procedure (76%
yield) (3).
The preparation of the 3,4-lactone (18) and 3-oxime of
friedelin (19) was carried out as a modification of a previously
reported protocol (8, 9). Subsequent reaction of 19 with
p-toluenesulfonyl chloride yielded the 3,4-lactam 20 (99.8%).
These compounds were identified by their spectroscopic data
Compounds 1, 6 > 18 > 17, 19, 20 > 9, 10a, 11, and 16
inhibited L. satiVa germination during the first 24-48 h of the
experiment (Table 2). All of the compounds tested inhibited L.
satiVa radicule elongation, N-derivative 19 being the most active,
followed by 1, 10, 18, 12, 16, and 17. Therefore, modification
of the A-ring of parent terpenes 1 and 2 did not result in
improved phytotoxic effects except for the oxime 19. The lack
of molecular selectivity observed suggests a nonspecific phy-
totoxic effect for these compounds on L. satiVa. Previous reports
have shown that 1 affected the radicule growth of Echinochloa
crusgalli (17). The phytotoxic effects of these type of molecules
have been attributed to their membrane-modifying properties
(
see Experimental Procedures). The NMR data of oxime 20
showed signals associated with a single product.
Assessment of these derivatives involved their screening for
insecticidal, phytotoxic, and antiparasitic effects. None of the
triterpene derivatives had antifeedant effects on S. littoralis or
L. decemlineata (data not shown). However, most of the
compounds tested (84%) had negative postingestive effects on
S. littoralis larvae (Table 1). A covariance analysis (AN-
COVA1) of food consumption (∆I) and biomass gains (∆B)
with initial larval weight (BI) as the covariate (covariate p >
(
13, 18).
The triterpene derivatives tested did not show relevant
antiparasite effects. Seco acids 6 and 12 were moderately active
against both parasite species (EC50 values of 37.6 and 44.3 µg/
mL, respectively, against L. infantum and 46.3 and 50.4 µg/
mL against T. cruzi). Compound 16 and its seco acid derivative
1
4 had selective antiparasite affects against L. infantum (EC50
of 19.2 µg/mL) and T. cruzi (EC50 of 49.3 µg/mL). There are
previous reports on the leishmanicidal and trypanocidal effects
of some pentacyclic triterpenes with 1 being inactive (19).
Furthermore, the presence of a carboxyl moiety in these types
of compounds was found to be important for their antileish-
manial activity (20).
0
.05) was performed to test for significant effects of the test
compounds on these variables. An additional ANOVA analysis
and covariate adjustment on ∆B with ∆I as the covariate
(ANCOVA2) was performed for those compounds that signifi-
cantly reduced ∆B to understand their postingestive mode of
action (antifeedant and/or toxic) (6).
The following compounds reduced biomass gains (∆B) and
food consumption (∆I) of orally injected S. littoralis larvae,
The relatively large number of bioactive derivatives allowed
us to draw some structure-activity relationships for these

3570 J. Agric. Food Chem., Vol. 54, No. 10, 2006
Moiteiro et al.
cytotoxic effects on mammalian CHO cells. Therefore, the cork
processing byproduct friedelin is a convenient model for the
design of insect growth regulators with phytotoxic effects.
Table 2. Effects of the Test Compounds on Germination and Radicle
Length of L. sativa
germination (% control)
radicle length
(% control)
ACKNOWLEDGMENT
a
compd
24 h
48 h
72 h
96 h
120 h
144 h
We gratefully acknowledge S. Carlin for language revision.
1
2
31*
97
94
81
34*
97
89
89
94
78*
100
96
70*
77*
81
97
99
94
76*
96
91
97
93
100
100
98
100
94
100
100
100
100
94
100
100
100
100
94
100
100
100
100
94
37.5
49.3
65.0
57.8
51.2
79.6
56.5
68.0
52.1
52.2
67.7
35.6
51.5
55.6
42.9
51.0
72.0
-
2
5
6
6
7
7
8
9
9
1
1
1
1
1
1
1
1
1
1
1
2
×
bb
LITERATURE CITED
(1) FAO. Programme d’action forestier m e´ diterran e´ en: Cadre de
r e´ f e´ rence des plans d’action forestiers nationaux des pays
m e´ diterran e´ ens. Committee on Mediterranean Forestry Questions,
Sylva mediterranea, Rome, Italy, 1993.
a
a
93
99
99
99
99
100
100
100
95
100
100
94
100
98
100
100
100
99
100
100
100
99
100
100
100
100
100
100
100
100
99
100
100
100
100
100
100
100
100
100
100
100
100
99
100
100
100
100
100
100
100
100
100
100
100
100
99
100
100
100
100
100
100
100
100
100
100
100
100
99
(2) FAO. Evaluation des ressources foresti e` res 1990. Pays non
tropicaux en d e´ veloppement. R e´ gion m e´ diterran e´ enne. FO: Misc/
a
0
0a
1
2
2a
4
4a
6
7
8
9
9
4/3, 1994.
(
3) Moiteiro, C.; Justino, F.; Tavares, R.; Marcelo Curto, M. J.;
Flor eˆ ncio, M. H.; Nascimento, M. S. J.; Pedro, M.; Cerqueira,
F.; Pinto, M. M. M. Synthetic seco-friedelane and friedelane
derivatives as inhibitors of human lymphocyte proliferation and
growth of human cancer cell lines in vitro. J. Nat. Prod. 2001,
6
4, 1273-1277.
4) Connolly, J. D.; Hill, R. A. Triterpenoids. Nat. Prod. Rep. 2005,
2, 487-503.
72*
67*
59*
66*
68*
40.07
44.26
38.43
25.82
56.5
(
(
100
87*
86*
100
100
98
96
100
100
97
100
100
97
100
100
97
2
5) Moiteiro, C.; Manta, C.; Justino, F.; Tavares, R.;. Marcelo Curto,
M. J.; S a˜ o Jos e´ Nascimento, M. M.; Pedro, M.; Pinto, M. M.
Hemisynthetic secofriedelane triterpenes with inhibitory activity
against the growth of human tumor cell lines in vitro. J. Nat.
Prod. 2004, 67, 1193-1196.
6) Reina, M.; Gonz a´ lez-Coloma, A.; Guti e´ rrez, C.; Cabrera, R.;
Rodr ´ı guez, M. L.; Fajardo, V.; Villarroel, L. Defensive chemistry
of Senecio miser Hook. J. Nat. Prod. 2001, 64, 6-11.
7) Gonz a´ lez-Coloma, A.; Guada n˜ o, A.; de In e´ s, C.; Mart ´ı nez-D ´ı az,
R.; Cortes, D. Selective action of acetogenin mitochondrial
complex I inhibitors. Z. Naturforsch. 2002, 57c, 1028-1034.
0
100
100
100
100
a
p < 0.05 in all cases. *p < 0.005, LSD test.
(
triterpenes. The presence of a hydroxyl group at C-2 (5) or a
lactone/hydroxyl at C-4 (18, 16) gave rise to a strong insecticidal
effect with respect to 1. The presence of a bulky substituent at
C-2 (17) and the substitution of the C-4 lactone of 18 by a lactam
(
(
20) lowered the insecticidal activity and increased its phytotoxic
effect. Substitution of the ketone at C-3 by hydroxyl (10, 11,
and 9), acetate (10a), or oxime (19) group also resulted in
elevated activity. The stereochemistry of the C-3 substituent
also played a role, with the R-hydroxyl (10) resulting in stronger
insecticidal and lower phytotoxic effects than the â-hydroxyl
(8) Talapatra, B.; Lahiri, B.; Basak, A.; Pradhan, D. K.; Talapatra,
S. K. Terpenoids & related compounds: Part XXIIIsMechanism
of some transformation reactions of friedelin. Indian J. Chem.
1
983, 22B, 741-745.
(9) Stevenson, R. Friedelin and related compounds. IV. A convenient
isolation of friedelin. J. Org. Chem. 1961, 2142-2143.
10) Setzer, W. N.; Setzer, M. C. Plant-derived triterpenoids as
(9). Similarly, recent studies have shown that lower polarity of
(
the C-3 functional group of oleanane triterpene acids reduced
their postingestive insect toxicity (21).
potential antineoplastic agents. Mini ReV. Med. Chem. 2003, 3,
5
40-556.
The A-ring resulted in inactive compounds (2, 2a, and 8).
Among the open A-ring terpenes, seco acid 6 showed the
strongest insecticidal effect, followed by 12 and 14. The
methylated derivatives 6a and 12a were stronger insecticides
than their parent compounds lacking cytotoxic or antiparasite
effects. Similarly, diacetate 7a was less cytotoxic than the parent
alcohol 7, suggesting a polarity-dependent mode of action of
these compounds on the cell/parasite. The phytotoxic (radicule
elongation) and insecticidal effects had a similar structure-
activity pattern with the acetylated derivatives being less active.
In summary, structural modifications of the A-ring of the cork
processing byproduct 1 improved its insecticidal activity with
the C-2 hydroxyl derivative 5, the aldehyde seco acid 6, its
acetylated derivative 6a, alcohols 9-11 and 16, the acetylated
(
11) Chang, C. I.; Kuo, C. C.; Chang, J. Y.; Kuo, Y. H. Three new
oleanane-type triterpenes from Ludwigia octovalvis with cyto-
toxic activity against two human cancer cell lines. J. Nat. Prod.
2004, 67, 91-93.
(12) Fu, L.; Zhang, S.; Li, N.; Wang, J.; Zhao, M.; Sakai, J.;
Hasegawa, T.; Mitsui, T.; Kataoka, T.; Oka, S.; Kiuchi, M.;
Hirose, K.; Ando, M. Three new triterpenes from Nerium
oleander and biological activity of the isolated compounds. J.
Nat. Prod. 2005, 68, 198-206.
13) Han, S. K.; Ko, Y. I.; Park, S. J.; Jin, I. J.; Kim, Y. M. Oleanolic
acid stabilize liposomal membranes Lipids 1997, 32, 769-773.
14) Sanjan, K. P.; Partho, P. D. Influence of asiatic acid, a triterpenoid
from Shorea robusta, on the feeding and enzyme activity of Oxya
fuscoVittata (Insecta, Orthoptera). J. Appl. Entomol. 1993, 115,
506-510.
(
(
1
0a, seco acid 14, lactone 18, and the oxime 19 being stronger
insecticides than the parent compound. Additionally, seco acid
2 and its acetylated derivative 12a were also insecticides in
(15) Loeb, M. J.; De Loof, A.; Schoofs, L.; Isaac, E. Angiotensin II
and agiotensin-converting enzyme as candidate compounds
modulating the effects of testis ecdysiotropin in testes of the
gypsy moth Lymantria dispar L. Gen. Comput. Endocrinol. 1998,
1
contrast to their inactive parent compound 2, another cork
processing byproduct. The larval postingestive effects and insect
cell cytotoxicity of these compounds suggest a multifaceted
mode of action. These structural modifications did not result in
better phytotoxic agents than the parent compounds 1 and 2,
except for lactam 20, and yielded several moderately active
antiparasite compounds (seco acids 6, 12, 14, and 16) with
1
12, 232-239.
(
16) Caballero-George, C.; Vanderheyden, P. M. L.; Okamoto, Y.;
Masaki, T.; Mbwambo, Z.; Apers, S.; Gupta, M. P.; Pieters, L.;
Vauquelin, G.; Vlietinck, A. Evaluation of bioactive saponins
and triterpenoidal aglycons for their binding properties on human
endothelin ETA and agiotensin AT receptors. Phytother. Res.
1
2004, 18, 729-736.

Biovalorization of Friedelane Triterpenes
J. Agric. Food Chem., Vol. 54, No. 10, 2006 3571
(
17) Casta n˜ eda, P.; Garcia, M. R.; Hernandez, E.; Torres, B. A.;
Anaya, A. L.; Mata, R. Effects of some compounds isolated from
Celaeodendron mexicanum Standl (Euphorbiaceae) on seeds and
phytopathogenic fungi. J. Chem. Ecol. 1992, 18, 1025-1037.
18) Duke, S. O.; Oliva, A. Mode of action of phytotoxic terpenoids.
In Allelopathy. Chemistry and Mode of Action of Allelochemicals;
Mac ´ı as, F., Galindo, J. C. G., Molinillo, J. M. G., Cutler, H. G.,
Eds.; CRC Press: Boca Raton, Florida, 2004; pp 201-216.
19) Camacho, M. R.; Mata, R.; Casta n˜ eda, P.; Kirby, G.; Warhurst,
D. C.; Croft, S. L.; Phillipson, J. D. Bioactive compounds from
Celaeodendron mexicanum. Planta Med. 2000, 66, 463-468.
20) Takahashi, M.; Fuchino, H.; Sekita, S.; Sakate, M. In vitro
leishmanicidal activity of some scarce natural products. Phy-
tother. Res. 2004, 18, 573-578.
(21) Pungitore, C. R.; Garcia, M.; Gianello, J. C.; Sosa, M.; Tonn,
C. E. Insecticidal and antifeedant effects of Junellia aspera
(Verbenaceae) triterpenes and derivatives on Sitophilus oryzae
(Coleoptera: Curculionidae). J. Stored Prod. Res. 2005, 41,
(
4
33-443.
Received for review December 13, 2005. Revised manuscript received
March 6, 2006. Accepted March 8, 2006. This work was supported by
Grant BQU2001-1505, and M.B. was supported by a predoctoral I3P-
CSIC fellowship.
(
(
JF0531151