Synthesis and biological activity of hydroxylated derivatives of linoleic acid and conjugated linoleic acids

Article abstract of DOI:10.1016/j.chemphyslip.2008.12.004

Allylic hydroxylated derivatives of the C18 unsaturated fatty acids were prepared from linoleic acid (LA) and conjugated linoleic acids (CLAs). The reaction of LA methyl ester with selenium dioxide (SeO₂) gave mono-hydroxylated derivatives, 13-hydroxy-9Z,11E-octadecadienoic acid, 13-hydroxy-9E,11E-octadecadienoic acid, 9-hydroxy-10E,12Z-octadecadienoic acid and 9-hydroxy-10E,12E-octadecadienoic acid methyl esters. In contrast, the reaction of CLA methyl ester with SeO₂ gave di-hydroxylated derivatives as novel products including, erythro-12,13-dihydroxy-10E-octadecenoic acid, erythro-11,12-dihydroxy-9E-octadecenoic acid, erythro-10,11-dihydroxy-12E-octadecenoic acid and erythro-9,10-dihydroxy-11E-octadecenoic acid methyl esters. These products were purified by normal-phase short column vacuum chromatography followed by high-performance liquid chromatography (HPLC). Their chemical structures were characterized by liquid chromatography-mass spectrometry (LC-MS) and nuclear magnetic resonance spectroscopy (NMR). The allylic hydroxylated derivatives of LA and CLA exhibited moderate in vitro cytotoxicity against a panel of human cancer cell lines including chronic myelogenous leukemia K562, myeloma RPMI8226, hepatocellular carcinoma HepG2 and breast adenocarcinoma MCF-7 cells (IC₅₀ 10-75 μM). The allylic hydroxylated derivatives of LA and CLA also showed toxicity to brine shrimp with LD₅₀ values in the range of 2.30-13.8 μM. However these compounds showed insignificant toxicity to honeybee at doses up to 100 μg/bee.

Full text of DOI:10.1016/j.chemphyslip.2008.12.004

Chemistry and Physics of Lipids 158 (2009) 39–45
Contents lists available at ScienceDirect
Chemistry and Physics of Lipids
journal homepage: www.elsevier.com/locate/chemphyslip
Synthesis and biological activity of hydroxylated derivatives of linoleic acid
and conjugated linoleic acids
Zhen Lia,b,c, Van H. Tran , Rujee K. Duke , Michelle C.H. Ng , Depo Yang
b
d
b
c,∗∗
, Colin C. Duke
b,∗
a
School of Biotechnology, Southern Medical University, Guangzhou, Guangdong Province 510515, PR China
Faculty of Pharmacy, University of Sydney, NSW 2006, Australia
Laboratory of Pharmacognosy, School of Pharmaceutical Sciences, Sun Yat-Sen University, Guangzhou, Guangdong Province 510080, PR China
Department of Pharmacology, Faculty of Medicine, University of Sydney, NSW 2006, Australia
b
c
d
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 17 August 2008
Received in revised form 10 December 2008
Accepted 17 December 2008
Available online 25 December 2008
Allylic hydroxylated derivatives of the C18 unsaturated fatty acids were prepared from linoleic
acid (LA) and conjugated linoleic acids (CLAs). The reaction of LA methyl ester with sele-
nium dioxide (SeO2) gave mono-hydroxylated derivatives, 13-hydroxy-9Z,11E-octadecadienoic acid,
13-hydroxy-9E,11E-octadecadienoic acid, 9-hydroxy-10E,12Z-octadecadienoic acid and 9-hydroxy-
10E,12E-octadecadienoic acid methyl esters. In contrast, the reaction of CLA methyl ester with SeO2
gave di-hydroxylated derivatives as novel products including, erythro-12,13-dihydroxy-10E-octadecenoic
acid, erythro-11,12-dihydroxy-9E-octadecenoic acid, erythro-10,11-dihydroxy-12E-octadecenoic acid and
erythro-9,10-dihydroxy-11E-octadecenoic acid methyl esters. These products were purified by normal-
phase short column vacuum chromatography followed by high-performance liquid chromatography
(HPLC). Their chemical structures were characterized by liquid chromatography–mass spectrometry
Keywords:
Allylic hydroxylation
Selenium dioxide
Linoleic acid
Conjugated linoleic acid
(LC–MS) and nuclear magnetic resonance spectroscopy (NMR). The allylic hydroxylated derivatives of
LA and CLA exhibited moderate in vitro cytotoxicity against a panel of human cancer cell lines including
chronic myelogenous leukemia K562, myeloma RPMI8226, hepatocellular carcinoma HepG2 and breast
adenocarcinoma MCF-7 cells (IC50 10–75 ␮M). The allylic hydroxylated derivatives of LA and CLA also
showed toxicity to brine shrimp with LD50 values in the range of 2.30–13.8 ␮M. However these compounds
showed insignificant toxicity to honeybee at doses up to 100 ␮g/bee.
©
2009 Elsevier Ireland Ltd. All rights reserved.
1
. Introduction
In plants, LA is one of the main substrates for lipid peroxidation
by lipoxygenases (LOX) (Fukushige et al., 2005). The metabo-
lites of LA from lipoxygenase pathways are collectively called
oxylipins, which are believed to be involved in plant pathogen
resistant strategies (Tsitsigiannis and Keller, 2007; Vellosillo et al.,
2007). Each of the C9 and C13 positions in LA are oxygenated
by LOX resulting in the metabolites 9-hydroperoxy-10E,12Z-
octadecadienoic acid and 13-hydroperoxy-9E,11Z-octadecadienoic
acid, respectively. These are further converted into more sta-
ble metabolites 9-hydroxy-10E,12Z-octadecadienoic acid (9-HPOD)
and 13-hydroxy-9E,11Z-octadecadienoic acid (13-HPOD) (Blee,
1998). Interestingly, these secondary products were reported to
have many biological activities including anti-inflammatory (Dong
et al., 2000), anti-microbial (Blee, 1998), anti-platelet (Truitt et al.,
1999), and potent cytotoxic activity on P388 cell lines (Yoshiki et
al., 1998).
9
Z,12Z-Octadecadienoic acid or linoleic acid (LA) is an omega-6
essential fatty acid. LA is a polyunsaturated fatty acid substrate in
the biosynthesis of prostaglandins and found in the lipids of cell
membranes (Lands, 2001). It is abundant in many vegetable oils,
especially safflower and sunflower oils (Knowles, 1975; Carvalho
et al., 2006). Conjugated linoleic acid (CLA) refers to the conju-
gated dienoic isomers of LA (1), 9Z,11E-CLA (11) and 10E,12Z-CLA
(
13). CLA isomers have been reported to have anti-antherogenic,
growth-promoting, body fat-reducing and anticancer properties in
several animal models (House et al., 2005). Consequently, there is
a great deal of current interest in generating CLA derivatives as
research tools to study the physiological action of CLA and as lead
compounds for drug development.
Selenium dioxide (SeO ) has been shown to be an effective
2
reagent for allylic hydroxylation introducing a hydroxyl group next
to the double bond position in a single step. It was successfully
applied in the allylic oxidation of oleic acid in the presence of tert-
butyl hydroperoxide (TBHP) (Knothe et al., 1993, 1995). There was
∗
Corresponding author. Tel.: +61 2 9351 2321; fax: +61 2 9351 4391.
Corresponding author. Tel.: +86 20 8733 3159; fax: +86 20 8733 3159.
E-mail addresses: lssydp@mail.sysu.edu.cn (D. Yang), colind@pharm.usyd.edu.au
∗
∗
(
C.C. Duke).
0
009-3084/$ – see front matter © 2009 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.chemphyslip.2008.12.004

4
0
Z. Li et al. / Chemistry and Physics of Lipids 158 (2009) 39–45
Fig. 1. (A) Linoleic acid (LA, 1) and its allylic mono-hydroxylated derivatives (3, 5, 7, and 9) and the corresponding methyl esters (2, 4, 6, 8, and 10), (B). 9Z,11E-Conjugated
linoleic acid (9Z,11E-CLA, 11) and its di-hydroxylated derivatives (15 and 17) and the corresponding methyl esters (12, 16 and 18) and (C) 10E,12Z-conjugated linoleic acid
(
10E,12Z-CLA, 13) and its di-hydroxylated derivatives (19 and 21) and the corresponding methyl esters (14, 20 and 22).
no reported synthesis of hydroxylated derivatives of CLA and, also,
the biological activities of the allylic hydroxylated derivatives LA
and CLA were not known.
This study describes allylic hydroxylation by SeO2 of LA (2) and
CLA methyl esters (12) and (14) and reports the in vivo toxicity of
the respective resulting products (4, 6, 8, 10, 16, 18, 20, and 22) as
well as the corresponding free fatty acids (3, 5, 7, 9, 15, 17, 19 and
2.2. Separation and purification of the products
The reaction mixture was purified by short-column (sil-
ica gel) vacuum chromatography using a step-wise gradient
of mobile phases consisting of hexane/dichloromethane/ethyl
acetate/ethanol. Mixtures of mono-hydroxylated and di-
hydroxylated isomeric products were respectively eluted with
dichloromethane/ethyl acetate (80/20) and dichloromethane/ethyl
acetate (50/50). The isomers were subsequently purified by semi-
preparative normal phase-HPLC (NP-HPLC) employing an Econosil
Silica (10 micron) column (10 mm × 250 mm) (Alltech, Australia)
and a refractive index detector (RID-10A, SHIMADZU). Mobile
phase systems consisting of n-hexane:2-propanol (99.4:0.6, v/v)
and n-hexane:2-propanol (98:2, v/v) were employed for mono-
hydroxylated and di-hydroxylated products, respectively. Flow rate
was operated at 8 mL/min. Six purified compounds designated as
21) (Fig. 1) on brine shrimp and honeybee and, also, cytotoxicity
towards several human cancer cell lines.
2. Experimental
2.1. General
CLA (82% purity) was provided by Zhongshan Unicare Natural
Medicine Co. Ltd. (Zhongshan, China). CLA was purified to 95%
purity (consisting of 40% 9Z,11E-CLA and 55% 10E,12Z-CLA iso-
mers by GC–MS) according to the method previously reported
4, 6, 8, 10, 20, 16 and a mixture consisting of compounds 18 and 22
were obtained as shown in Figs 1 and 2.
(
(
Li et al., 2007). Methyl linoleate (>95% purity), selenium dioxide
99.8%), DMSO and TBHP were purchased from Aldrich Chem-
2.3. Synthesis of mono-hydroxylated derivatives
ical Co. All chemicals used were of analytical reagent grade.
RPMI 1640 was purchased from Invitrogen Ltd. (Australia). Foetal
bovine serum (FBS) was purchased from Bovogen Ltd. (Australia).
To a 50-mL round-bottomed flask was added SeO2 (377.6 mg,
3.4 mmol) and dichloromethane (4 mL). The mixture was stirred
for 0.5 h at room temperature (Knothe et al., 1993, 1995). Methyl
linoleate (2) (1 g, 3.4 mmol) was then added, and the reaction
mixture was stirred under N2 for 24 h at room temperature. 10%
NaCl solution (10 mL) was added and the mixture extracted twice
with dichloromethane (10 mL). The dichloromethane portions were
combined, washed with deionised water (3× 5 mL) and concen-
trated under reduced pressure to give a pale yellow oil (0.95 g, 90%)
containing the compounds 4, 6, 8, and 10 in the approximate pro-
portions of 34, 21, 28, and 17% as determined by NP-HPLC (Fig. 2A).
Hydrolysis of the esters was carried out as described in Section 2.5
below.
®
CellTiter 96 AQueous One Solution Reagent was purchased from
Promega Ltd. (Australia). 1H NMR and C NMR spectra were
recorded on 300 MHz Varian Gemini 300 spectrometer (Palo Alto
CA, USA) in CDCl3 (Sigma–Aldrich, USA) with tetramethylsilane as
internal standard. LC–MS/MS analysis of products was performed
on a Finnigan/Mat TSQ7000 mass spectrometer with a high-
performance liquid chromatography (HPLC) flow rate of 0.3 mL/min
and ODS-C18 (Alltech) analytical column (2.1 mm × 150 mm). The
gradient of the mobile phase was from 40% acetonitrile (60% water;
13
0.5%, v/v acetic acid) to 100% acetonitrile (0.5%, v/v acetic acid)
in 40 min. For MS/MS analyses the collision energy was 30 V.
Methyl ester samples were analyzed in electro-spray ionisation
(
ESI) positive mode, whilst free fatty acid samples were analyzed
2.3.1. 13-Hydroxy-9Z,11E-octadecadienoic acid methyl ester (4)
Colorless oil of >98% purity: 1H NMR (CDCl , 300 MHz) ı 6.48
in ESI negative mode. High-resolution masses were determined
3
by ESI on a Bruker BioApex Fourier transform mass spectrome-
ter.
(dd, J = 15.2, 11.1 Hz, 1H, H-11), 5.97 (dd, J = 11.7, 11.1 Hz, 1H, H-10),
5.67 (dd, J = 15.2, 6.9 Hz, 1H, H-12), 5.43 (dt, J = 10.8, 7.8 Hz, 1H, H-9),

Z. Li et al. / Chemistry and Physics of Lipids 158 (2009) 39–45
41
Fig. 2. Normal phase-HPLC chromatograms of the hydroxylated products from LA and CLA methyl esters. HPLC elution profiles of (A) mono-hydroxylated products (4, 6, 8,
and 10) from LA methyl ester (2) and (B) di-hydroxylated products (16, 18, 20, and 22).
4
.16 (m, 1H, H-13), 3.67 (s, 3H, CH O), 2.30 (t, 2H, H-2), 2.16 (m, 2H,
2.4. Synthesis of di-hydroxylated derivatives
3
13
H-8), 1.74–1.17 (m, 18H, 9 × CH ), 0.89 (t, 3H, H-18); C NMR (CDCl3,
2
7
3
5 MHz,) ı 174.18, 135.21, 132.66, 127.90, 125.69, 72.83, 51.34, 37.42,
To a 50-mL round-bottomed flask was added SeO2 (377.6 mg,
3.4 mmol) and dichloromethane (4 mL). The mixture was stirred
for 0.5 h at room temperature (Knothe et al., 1993, 1995). Methyl
ester of CLA (12 or 14) (1 g, 3.4 mmol) was then added, and the
4.08, 31.79, 29.49, 29.06, 29.02, 28.95, 27.68, 25.09, 24.91, 22.57,
+
1
3.95. LC–MS/MS (+) m/z 293 ([(M−H O) + H] , 100%). LC–MS/MS
2
−
(
2
−) of the hydrolysis product 3 gave ions at m/z 295 ([M−H] , 100%),
−
−
◦
77 ([(M−H O) − H] , 64%), 195 ([(M−CH (CH ) CHO) − H] , 36%),
reaction mixture was stirred under N2 for 8 h at 50 C and at room
2
3
2 4
−
179 ([(M−CH (CH ) CHCH − HCOOH) − H] , 5%).
temperature for further 16 h. 10% NaCl solution (10 mL) was added
and the mixture extracted twice with dichloromethane (10 mL). The
dichloromethane portions were combined, washed with deionised
water (3× 5 mL) and concentrated under reduced pressure to give a
pale yellow oil (0.85 g, 76%) containing the compounds 22, 18, 20, 16
3
2
2
2
2.3.2. 13-Hydroxy-9E,11E-octadecadienoic acid methyl ester (6)
1
Colorless oil of >98% purity: H NMR (CDCl , 300 MHz) ı 6.18 (dd,
3
J = 15.2, 11.1 Hz, 1H, H-11), 6.03 (dd, J = 15, 10.2 Hz, 1 H, H-10), 5.70
(
order of elution) in the approximate proportions of 61% (a mixture
(
dt, J = 15, 6.9 Hz, 1H, H-9), 5.58 (dd, J = 15.2, 6.9 Hz, 1H, H-12), 4.11
of compound 22, 18), 19%, 20% as determined by NP-HPLC (Fig. 2B).
Hydrolysis of the esters was carried out as described in Section 2.5
bellow.
(
m, 1H, H-13), 3.67 (s, 3H, CH O), 2.30 (t, 2H, H-2), 2.06 (m, 2H, H-
3
13
8), 1.68-1.21 (m, 18H, 9 × CH ), 0.89 (t, 3H, H-18); C NMR (CDCl3,
7
3
2
5 MHz) ı 174.16, 135.21, 133.80, 130.83, 129.61, 72.81, 51.33, 37.41,
4.10, 31.80, 29.15, 29.10, 29.07, 29.05, 28.94, 25.09, 24.93, 22.57,
+
2.4.1. erythro-9,10-Dihydroxy-11E-octadecenoic acid methyl ester
1
3.9. LC–MS/MS (+) m/z 293 ([(M−H O) + H] , 100%). LC–MS/MS (−)
2
−
(16) (Piazza et al., 2003)
of the hydrolysis product 5 gave ions at m/z 295 ([M−H] , 36%), 277
White solid of >98% purity, m.p. 36–38 C: 1H NMR (CDCl3,
◦
−
−
(
1
[(M−H O) − H] , 100%), 195 ([(M−CH (CH ) CHO) − H] , 50%),
2
3
2 4
−
300 MHz) ı 5.76 (dt, J = 15.4, 6.0 Hz, 1H, H-12), 5.44 (ddt, J = 15.4, 7.2,
79 ([(M−CH (CH ) CHCH − HCOOH) − H] , 3%).
3
2
2
2
1
Hz, 1H, H-11), 3.86 (t, J = 6.9 Hz, 1H, H-10), 3.67 (s, 3H, CH O), 3.43
3
(
2
1
2
m, 1H, H-9), 2.30 (m, 2H, H-2), 2.05 (m, 2H, H-13), 1.72–1.17 (m,
2
.3.3. 9-Hydroxy-10E,12Z-octadecadienoic acid methyl ester (8)
13
0H), 0.88 (t, J = 6.9 Hz, 3H, H-18); C NMR (CDCl , 75 MHz) ı 174.18,
3
Colorless oil of >98% purity: 1H NMR (CDCl , 300 MHz) ı
3
34.95, 129.37, 76.31, 74.76, 51.33, 34.12, 33.02, 32.29, 31.37, 29.57,
6
1
.49 (dd, J = 15.2, 11.1 Hz, 1H, H-11), 5.97 (dd, J = 11.7, 11.1 Hz,
H, H-12), 5.66 (dd, J = 15.2, 6.9 Hz, 1H, H-10), 5.45 (dt, J = 11.1,
9.33, 29.15, 29.00, 28.75, 25.57, 24.96, 22.47, 13.99. LC–MS/MS (+)
+
analysis gave m/z 311 ([(M−H O) + H] , 100%).
2
7.5 Hz, 1H, H-13), 4.15 (m, 1H, H-9), 3.67 (s, 3H, CH O), 2.30
3
Hydrolysis of 16 gave 15 as a white solid of >97% purity, m.p.
(
0
t, 2H, H-2), 2.16 (m, 2H, H-14), 1.74–1.17 (m, 18H, 9 × CH ),
◦
7–69 C. C NMR (CDCl , 75 MHz) ı 178.80, 135.18, 129.02, 76.38,
3
13
2
6
7
2
.89 (t, 3H, H-18); 13C NMR (CDCl , 75 MHz) ı 174.17, 135.84,
3
4.71, 33.92, 32.81, 32.34, 31.66, 29.31, 29.05, 28.99, 28.88, 28.82,
5.43, 24.62, 22.60, 14.06.
132.97, 127.75, 125.85, 72.84, 51.34, 37.40, 34.09, 31.46, 29.35,
2
9.30, 29.14, 29.07, 27.75, 25.33, 24.93, 22.51, 13.97. LC–MS/MS
LC–MS/MS (−) analysis of 15 gave ion fragments at
+
(
+) m/z 293 ([(M−H O) + H] , 100%). LC–MS/MS (−) of the hydrol-
−
−
−
2
m/z 313 ([M−H] , 17%), 295 ([(M−H O) − H] , 12%), 277
2
−
ysis product
7
gave ions at m/z 295 ([M−H] , 100%), 277
−
(
2
[(M−2 × H O) − H] , 34%), 201 ([(M−CH (CH )5CHCH ) − H] ,
2
3
2
−
2
−
−
(
5
[(M−H O) − H] , 64%), 171 ([(M−CH (CH ) CHCHCHCH ) − H] ,
2
3
2
4
2
4%), 171 ([M−CH (CH )5CHCHCH OH − H] , 100%), 141
3
2
2
−
0%), 123 ([(M−HCO(CH )7COOH) − H] , 3%).
−
2
(
(
[(M−CH (CH )5CHCHCH(OH)CH OH − H) − H] , 25%). HR-MS
3
2
2
−
ESI) of 15 calcd for (C18
H
O ) [M−H] 313.2384, found 313.2381.
34
4
2.3.4. 9-Hydroxy-10E,12E-octadecadienoic acid methyl ester (10)
◦
1
White solid of >98% purity, m.p. 37 C: H NMR (CDCl3,
2.4.2. erythro-10,11-Dihydroxy-12E-octadecenoic acid methyl
3
00 MHz) ı 6.18 (dd, J = 15.2, 11.1 Hz, 1H, H-11), 6.01 (dd, J = 15,
ester (20)
White solid of >98% purity, m.p. 39–40 C: 1H NMR (CDCl3,
300 MHz) ı 5.75 (dt, J = 15.6, 6.4 Hz, 1H, H-13), 5.44 (ddt, J = 15.6,
7.2, 1 Hz, 1 H, H-12), 3.85 (t, J = 6.6 Hz, 1H, H-11), 3.67 (s, 3H,
◦
1
(
0.5 Hz, 1H, H-12), 5.71 (dt, J = 14.7, 6.9 Hz, 1H, H-13), 5.56
dd, J = 15, 4.9 Hz, 1H, H-10), 4.10 (m, 1H, H-9), 3.67 (s, 3H,
CH O), 2.30 (t, 2H, H-2), 2.07 (m, 2H, H-14), 1.69–1.18 (m, 18H,
3
13
9
× CH ), 0.89 (t, 3H, H-18); C NMR (CDCl , 75 MHz) ı 174.16,
CH O), 3.42 (m, 1H, H-10), 2.30 (t, 2H, H-2), 2.05 (m, 1H, H-14),
2
3
3
13
135.65, 133.57, 131.02, 129.41, 72.84, 51.33, 37.38, 34.10, 32.58,
1.71–1.14 (m, 20H,), 0.89 (t, J = 6.9 Hz, 3H, H-18); C NMR (CDCl3,
75 MHz) ı 174.21, 134.93, 129.36, 76.0, 74.77, 51.33, 34.10, 33.0,
32.33, 31.68, 29.14, 29.07, 29.06, 29.00, 28.81, 25.53, 24.93, 22.50,
31.41, 29.35, 29.14, 29.07, 28.92, 25.34, 24.93, 22.49, 13.95. The
+
LC–MS/MS (+) m/z 293 ([(M−H O) + H] , 100%). LC–MS/MS (−)
2
−
+
of the hydrolysis product 9 gave ions at m/z 295 ([M−H] ,
13.95. LC–MS/MS (+) analysis gave m/z 311 ([(M−H O) + 1] , 100%).
2
−
2
5%), 277 ([(M−H O) − H] , 100%), 171 ([(M−CH (CH ) CHCH–
Hydrolysis of 20 gave 19 as a white solid of > 81% purity,
2
3
2 4
−
−
◦
CHCH ) − H] , 40%), 123 ([(M−HCO(CH )7COOH) − H] , 3%).
m.p. 72–75 C. LC–MS/MS (−) analysis of 19 gave ion fragments
2
2

4
2
Z. Li et al. / Chemistry and Physics of Lipids 158 (2009) 39–45
−
−
at m/z 313 ([M−H] , 17%), 295 ([(M−H O) − H] ,12%), 277
5 ␮L acetone) was applied to each bee via a microapplicator. Each
experiments was performed on groups of 10 bees for both test and
control treatments. Mortality and sublethal effects were recorded
at 2, 4, 8, 16, 24 and 48 h after treatment. Three independent exper-
iments were performed.
2
−
−
(
[(M−2 × H O) − H] , 29%), 215 ([(M−CH (CH ) CHCH ) − H] ,
2
3
2
4
2
−
2
(
(
2%), 185 ([(M−CH (CH ) CH–CHCH OH) − H] , 100%), 127
3
2
4
−
2
[(M−CHO(CH ) COOH) − H] , 25%). HRMS (ESI) of 19 calcd for
2
8
−
C
18
H
34
O ) [M−H] 313.2384, found 313.2382.
4
2
.4.3. Methyl esters of erythro-11,12-dihydroxy-9E-octadecenoic
acid and erythro-12,13-dihydroxy-10E-octadecenoic acid (18 and
2)
Colorless oil: H NMR spectra of these compounds were similar
to those of 16 and 20. 13C NMR (CDCl , 75 MHz) ı 174.24, 134.62,
2
.8. Cytotoxicity assays
2
Two suspension cell lines including human myeloma RPMI
226 (ATCC CCL-165) and chronic myelogenous leukemia K-562
1
8
3
(
ATCC CCL-243), and two adherent cell lines including human
134.52, 129.31, 129.25, 76.27, 76.25, 74.64, 74.63, 51.35, 33.97, 33.94,
hepatocellular carcinoma HepG2 (ATCC HB-8065) and adenocarci-
noma MCF-7 (HTB-22) were used in this study. These cells were
cultured in RPMI 1640 medium supplemented with 10% (v/v)
heat-inactivated FBS without antibiotics at 37 C in a humidified
incubator with 5% of carbon dioxide.
3
2
2.86, 32.81, 32.21, 32.18, 31.79, 31.69, 29.24, 28.97, 28.93, 28.91,
5.477, 25.19, 25.47, 25.19, 24.79, 22.50, 23.79, 13.96, 13.93.
Hydrolysis of this mixture gave 17 and 21 as a mixture of
◦
colorless oil. LC–MS/MS (−) analysis of the hydrolysis product 17
−
showed ion fragments at m/z 313 ([M−H] , 17%),
−
−
2
2
95
9%),
([(M−H O) − H] ,
12%),
277
([(M−2 × H O) − H] ,
2
2
−
2.9. Cell viability assay
199
([(M−CH (CH )5CHO) − H] ,
100%),
113
169
[(M−
3
2
−
(
[(M−CH (CH )5CH–(OH)–CHO) − H] ,
70%),
3
2
−
Prior to seeding cells to the microplates, cells medium was
replaced with serum-free RPMI 1640 medium (Garcia-Pedrero et
al., 2006). Cells were then seeded onto 96-well microplates at a
density of 4 × 10 per well per 100 ␮L and incubated at 37 C in a
humidified incubator with 5% of CO2 for 4 h before treatment of
test agents which were solubilised in DMSO at a range of concen-
CH (CH )5CH(OH)CH(OH)–CHCHCH CH ) − H] , 25%). LC–MS/MS
3
2
2
3
(
−) of the hydrolysis product 21 showed ion fragments at
−
−
−
m/z 313 ([M− H] , 17%), 295 ([(M−H O) − H] , 12%), 277
2
4
◦
−
(
1
[(M−2 × H O) − H] , 29%), 213 ([(M−CH (CH ) CHO) − H] ,
2
3
2 4
−
00%), 183 ([(M−CH (CH ) CH(OH)CHO) − H] , 70%), 129
3
2
4
−
(
[(M−CH CH(CH ) COOH) − H] , 25%).
2
2 8
®
tration (0.5–200 ␮M). After 24 h incubation, CellTiter 96 AQueous
One Solution Reagent (20 ␮L) was added of to each well and incu-
2.5. Hydrolysis of the methyl esters
◦
bated for 3 h at 37 C in a humidified incubator with 5% of CO . The
2
final concentration of DMSO was 0.2% (v/v). Cell viability was mea-
sured using a microplate reader (Polarstar, BMG Labtechnologies)
at a wavelength of 490 nm. Cytotoxic effect of the compounds was
expressed as percentage of cell viability against concentrations of
fatty acids. The concentrations of fatty acids that inhibited growth
by 50% (IC50) were calculated using the Prism Software (Graphpad,
San Diego, CA) from full curves generated from three independent
experiments, each performed in triplicate.
Hydrolysis of methyl esters of mono-hydroxylated LA (4, 6, 8
and 10) and di-hydroxylated CLA (16, 18, 20, and 22) were carried
out under a N2 atmosphere. To the ester (10–50 mg) in a 50-mL
round-bottomed flask were added methanol (5 mL) and 1 M NaOH
◦
solution (5 mL). The mixture was stirred (500 rpm) at 50 C for 1 h,
followed by addition of 1 M HCl solution (5 mL), and extracted with
ethyl acetate (3× 10 mL). The ethyl acetate solvents were evaporated
under reduced pressure to give the final free fatty acids, which were
then characterized by NMR and mass spectrometry. The 1H NMR
Cytotoxic effects of the mono-hydroxylated LA (3, 5, 7, and 9)
were also examined in K569 cells in presence of FBS. Cells at a den-
spectra of free fatty acids of hydroxylated LA and CLA were similar
to that of their esters except for the absence of the signal for the
methyl group at 3.67 ppm.
4
sity of 1 × 10 per well per 100 ␮L in 96-well plate were incubated
with a range of concentrations (0.5–200 ␮M) of the compounds
for 72 h. Cell viability was measured using MTS assay as described
above.
2
.6. Brine shrimp assay
Artemia salina cysts (10 mg) (Bio-Marine, Inc. USA) were hatched
3. Results and discussion
at room temperature in artificial seawater under a continuous light
regime (Campbell et al., 1994). After 48 h incubation, the brine
shrimp (10 per well) were transferred to 24-well microplates then
treated with various concentrations of the test compounds added
as a dimethyl sulfoxide (DMSO) solution such that the final con-
centration of DMSO was 2%. Controls for the assay were brine
shrimp treated with DMSO alone. After 24 h exposure, the num-
bers of live shrimp were counted and percentages of deaths were
calculated. Brine shrimp lethalities of 50% (LC50) were calculated
using the Prism Software (Graphpad, San Diego, CA) from full curves
generated from two independent experiments, each performed in
triplicate.
3.1. Synthesis of mono-hydroxylated products
Allylic oxidation by selenium dioxide using TBHP as a re-oxidant
has been successfully applied to many olefinic compounds (Knothe
et al., 1993, 1995). Although allylic hydroxylation reactions using
selenium dioxide have been widely documented in the literature,
its use in the synthesis of hydroxylated fatty acids using polyun-
saturated fatty acids are comparatively rare as the reported allylic
hydroxylation was confined mainly to mono-unsaturated fatty
acids (Knothe et al., 1994).
In our studies, the reaction of methyl linoleate with SeO /TBHP,
2
following the reported experimental conditions (Knothe et al.,
1993), resulted in a low yield and complex mixture of products
possibly containing a small amount (∼15%) of the required hydrox-
2
.7. Contact toxicity to honeybee
1
Tests were conducted following the EPA Guideline OPPTS
ylated products. H NMR analysis of the mixture showed that there
8
50.3020 (US EPA, 1996). Worker bees (Apis mellifera var. Liguria)
was a strong signal at 4.25–4.30 ppm, and a relatively weak sig-
nal at 4.17–4.07 ppm tentatively identified as the hydrogen signals
of the –CH(OH) groups by comparison with the previous reported
similar partial structure (Kato et al., 2001; Kuklev et al., 1997).
The reaction conditions were subsequently modified to improve
were obtained directly from a healthy managed beehive (NSW
Department of Primary Industries registration number D268). Test
bees were immobilized with CO and randomly assigned to the test
2
substances and controls. A single dose (100 ␮g of test substance per

Z. Li et al. / Chemistry and Physics of Lipids 158 (2009) 39–45
43
the yield and minimise the unwanted side products. The modifi-
cations included changing the ratio of substrates, amount of TBHP,
reaction media, temperature and time. The optimised reaction con-
dition which did not involve the use of TBHP gave a cleaner reaction
product. Compared with other chemical methods (Kato et al., 2001;
Truc and Daesung, 2001) the improved method of hydroxylation of
methyl linoleate by SeO2 as described in this study was simpler,
quicker and gave higher yield (∼54%) of the required products. The
allylic mono-hydroxylated products were subjected to purification
by short column vacuum chromatography before being separated
by NP-HPLC. These mono-hydroxylated products 4, 6, 8 and 10, as
shown in Fig. 1, were consecutively eluted in four distinctive peaks
(
1–4) on NP-HPLC (Fig. 2A, Table 1) and characterized by NMR spec-
troscopy and mass spectrometry.
3.2. Synthesis of di-hydroxylated products
Our initial aim was to synthesize the mono-hydroxy derivatives
of CLA using SeO /TBHP as described above. However, using the
2
same condition described in Section 2.3, the reaction of CLA methyl
esters (11 and 13) gave mixture 1,2-dihydroxy products in 60% yield.
This effect was first reported as a novel syn di-hydroxylation of
1
,3-dienes with SeO , which involves a [4 + 2] cycloaddition fol-
2
Fig. 3. 1H NMR spectra of the products from CLA methyl ester with SeO2 (A) before
and (B) after silica gel short column vacuum chromatography.
lowed by C–Se bond oxidation, which gave unprecedented 1,2-
and 1,4-diol products (Truc and Daesung, 2001). The proposed
reaction mechanism can be explained in Scheme 1 and this was
evidenced from the formation of di-hydroxylated products (18 and
Table 1
Retention time of mono- and di-hydroxylated derivatives of LA and CLA methyl
esters.
2
0; 16 and 22) (Fig. 1). The existence of 1,2-diol cyclic selenites
ꢀ ꢀ ꢀ ꢀ
1
(
22 , 20 , 18 , and 16 ) was inferred from the H NMR spectrum of
the crude products before silica gel short column vacuum chro-
matography (Fig. 3A). These initial products are likely to be of a
similar type to the 1,2-diol cyclic selenites identified by Truc and
Daesung (2001). The H NMR spectrum shown in Fig. 3B, illustrates
a chemical shift change of two –CH(OH) protons from a down-
Compound no. Hydroxylated fatty acid methyl esters
Retention
time (min)
4
6
8
Methyl 13-hydroxy-9Z,11E-octadecadienoate
Methyl 13-hydroxy-9E,11E-octadecadienoate
Methyl 9-hydroxy-10E,12Z-octadecadienoate
Methyl 9-hydroxy-10E,12E-octadecadienoate
Methyl erythro-9,10-dihydroxy-11E-octadecenoate
Methyl erythro-11,12-dihydroxy-9E-octadecenoate
16.4
18.5
21.2
22.8
24.9
20.8
1
1
0
field region (ppm 4.81, 4.44; 4.31, and 3.95) to an upfield region
1
6
(
ppm 3.85 and 3.43) revealing the structures of the di-hydroxylated
18
products. These di-hydroxylated products gave rise to three dis-
tinctive peaks on NP-HPLC (Fig. 2B, Table 1). Peak #1 contained a
mixture of two compounds (22 and 18), whilst peaks #2 and #3
yielded compounds 20 and 16, respectively. For the four products
20
Methyl erythro-10,11-dihydroxy-12E-octadecenoate 22.9
Methyl erythro-12,13-dihydroxy-10E-octadecenoate 20.8
2
2
(
16, 18, 20 and 22), the structural moiety consisting of a 1,2-diol
0.1 ppm compared with the corresponding methyl groups of com-
pounds 16, 18 and 22 due to the influence of the double bond
at the 12 position in compound 20. As the functionality for the
four compounds was located towards the middle of a long alkyl
chain 1H and C NMR spectral analyses did not clearly distinguish
with one hydroxyl group allylic to a double bond was clearly recog-
nizable from the signals for the double bond protons at 5.8 ppm and
allylic and non-allylic –CH(OH) protons at 3.9 and 3.4 ppm, respec-
tively. The C18 methyl group of compound 20 was deshielded by
13
Scheme 1. Proposed reaction mechanism of CLA methyl ester with SeO2. The numbers shown primed denote the proposed seleno reaction intermediates.

4
4
Z. Li et al. / Chemistry and Physics of Lipids 158 (2009) 39–45
the four positional isomers however their free carboxylic acids
were clearly distinguished and identified by mass spectral analy-
ses.
The E configuration for the double bond is indicated by the
proton-coupling constant (J = 15.4 Hz) of the double bond protons
for compounds 15–22. The vicinal dihydroxy compounds (15–22)
are tentatively assigned as each having erythro stereochemistry
based on the reaction mechanism reported previously (Piazza et
al., 2003) for reaction of SeO2 with dienes. Evidence supporting
erythro stereochemistry is the formation of four products, 16, 18,
3.3. Biological evaluation
3.3.1. Effect of the hydroxylated LA and CLA on brine shrimp (A.
salina)
The hydroxylated derivatives of LA and CLA methyl esters
showed a concentration dependent toxicity to brine shrimp. Inter-
estingly these hydroxylated derivatives were more toxic to the
brine shrimp than the parent compounds LA and CLA methyl ester
(Table 2). The toxicity of the test compounds was in the order of
16 > 20 > 10 > 6 > 8 > 4 »> LA–Me > CLA–Me. This finding clearly indi-
cated that addition of a hydroxyl group, regardless of the position
on the carbon chain, adjacent to a double bond significantly
increased toxicity to brine shrimp (>30-fold) in comparison with
the parent compounds LA and CLA methyl esters (LC50 > 200 ␮M).
Interestingly, free fatty acids of hydroxylated LA and CLA showed
insignificant toxicity to the brine shrimp at the concentrations of
up to 100 ␮M.
2
0 and 22, which is the outcome predicted from the proposed
stereospecific mechanism (Scheme 1) where the initial reaction is
,4 addition to the diene followed by rearrangement and hydrol-
1
ysis to give 1,2-diols. Of the alternative mechanisms considered
regiospecific addition reaction with a Z double bond would give only
two products 16 and 22 and non-regiospecific addition and non-
stereoselective reactions would give eight or possibly more vicinal
erythro and threo diol products. The vicinal proton coupling con-
stants of the two ␣-hydroxy methine protons for the four products
3.3.2. Effect of the hydroxylated LA and CLA on honeybee (A.
mellifera)
(
16, 18, 20 and 22) of approximately 6.9 Hz determined in CDCl3
and CD OD were of intermediate value indicating free rotation of
The mono-hydroxylated compounds 4, 6, 8, and 10 showed no
contact toxicity to honey bees at the dose of 100 ␮g/bee up to 16 h
in comparison with the acetone control group. Prolonged exposure
of the honeybees to the compounds for 24 h showed only marginal
toxicity (15% death). Similarly, the compound 16, which was the
most toxic compound to brine shrimp, displayed no contact toxicity
to honeybees at the doses of 50 and 100 ␮g/bee up to 16 h, respec-
tively. Prolonged exposure time to compound 16 at same doses
to 24 h caused approximately 11% honeybee death. These findings
demonstrated that mono- and di-hydroxylated derivatives had no
significant acute contact toxicity to worker honeybees for doses up
to 100 ␮g/bee.
3
the bond between the hydroxylated carbons, consequently, erythro
or threo stereochemistry could not be determined from the vicinal
proton–proton coupling constants of the two ␣-hydroxy methine
protons.
13C NMR spectra reported (Jarvis et al., 1982) for a series of 1,2-
diols with one hydroxyl group allylic to a double bond showed
shielding of 1.0 ppm for the allylic–CH(OH)– carbon of the ery-
thro products compared with the threo products and a shielding
of 0.3 ppm for the non-allylic –CH(OH)– carbon of the erythro prod-
ucts compared with the threo products. The chemical shifts of the
allylic –CH(OH)– carbons for the four 1,2-diols (16, 18, 20 and 22)
in this study were in the range 76.00–76.31 ppm and for the non-
allylic –CH(OH)– carbons were in the range 74.63–74.77. The narrow
chemical shift range for the –CH(OH)– carbon signals for the 1,2-
diols is consistent with all 4 isomers having erythro configuration.
The erythro configuration for compound 16 is supported by simi-
larity with the previously reported 13C NMR chemical shifts for this
compound, in particular, for the allylic (76.4 ppm) and non-allylic
3.3.3. Effect of the hydroxylated LA and CLA on human cancerous
cell lines
The cytotoxicity of mono-hydroxy derivatives of LA and CLA
was examined and compared with the parent compounds CLA and
LA on four human tumor cell lines including two suspension cell
lines (K562 and RPMI8226) and two adhesion cell lines (HepG2
and MCF-7). Assays for the cytotoxic effects of the test samples
towards these cell lines were carried out in serum-free medium
for 24 h. Longer incubation time was not suitable, as under this
condition cell growth would be inhibited due to deficiency of
nutrition. Table 3 shows the cytotoxic effect of free fatty acids 3,
5, 7, 9, and CLA and LA expressed as IC50 values determined from
dose–response curves. The order of potency of these compounds
was as follows: 7 > 9 > LA > 3 > CLA > 5 on all cell lines tested. Com-
pound 7 was the most active isomer with an IC50 value of 10 ␮M for
K562 cells and 59 ␮M for HepG2 cells. This observation indicates
that the analogues with a –OH group at the C9 position exerted
higher inhibitory effect on cell proliferation than those having a
–OH group substituted at the C13 position, and conjugated double
bonds with E,Z configuration appeared to be more effective than
that those with the E,E configuration. Methyl esters (4, 6, 8, and 10)
(
75.0 ppm) –CH(OH)– carbons (Piazza et al., 2003). The correspond-
ing chemical shifts reported for the threo isomer of compound 16
were 77.3 and 75.5 for the allylic and non-allylic –CH(OH)– carbons.
The shielding of the erythro–CH(OH)– carbons relative to those of
the threo isomer is in agreement with the previously discussed 13
C
NMR spectral data reported (Jarvis et al., 1982) for a series of allylic
,2-diols.
The positions of the allylic diol moieties for compounds (16, 18,
0 and 22) were established from negative ion ESI-MS/MS mass
1
2
spectral analyses of their respective hydrolysis products (15, 17, 19
and 21). Characteristic fragments were observed for bond cleav-
ages between or adjacent to the carbons bearing the hydroxyl
groups. All characteristic fragments resulted from cleavages that
gave neutral fragments from the C18 methyl end and charged frag-
ments from the carboxyl end. For compound 15 there were three
characteristic ions at m/z 201, 171 and 141 corresponding to frag-
−
ments from [M−H] resulting from cleavages between C10–C11,
Table 2
C9–C10 and C8–C9, respectively. For compound 17 there were
two characteristic ions at m/z 199 and 169 corresponding to frag-
Lethality of the hydroxylated LA and CLA towards brine shrimp.
−
Compounds
LC50 (␮M ± S.D.)
ments from [M−H] resulting from cleavages between C11–C12,
and C10–C11, respectively. For compound 19 there were two char-
4
6
8
7.8 ± 0.6
6.9 ± 0.6
7.1 ± 0.6
5.1 ± 1.0
2.3 ± 0.15
3.1 ± 0.5
>340
acteristic ions at m/z 215 and 185 corresponding to fragments from
−
[
M−H] resulting from cleavages between C11–C12 and C10–C11,
1
0
respectively. Finally, for compound 21 there were two characteristic
1
6
−
ions at m/z 213 and 183 corresponding to fragments from [M−H]
20
CLA–Me
LA–Me
resulting from cleavages between C12–C13 and C11–C12, respec-
tively.
>340

Z. Li et al. / Chemistry and Physics of Lipids 158 (2009) 39–45
45
Table 3
di-hydroxylated derivatives had no significant acute contact toxic-
ity towards honeybee workers.
Effect of hydroxylated LA and CLA on human cancer cell proliferation.
IC50 (␮M ± S.D.)
Acknowledgements
Cell lines
K562a
K562b
RPMI8226a
HepG2a
MCF-7a
Compounds
We thank the NSFC/DEST Joint Research Scheme (2007) and
International cooperation center of Sun Yat-Sen University (2006)
for financial assistance. The assistance of Mr Bruce N. Tattam with
3
4
5
6
7
8
9
43 ± 2
118 ± 8
209 ± 15
39 ± 10
40 ± 12
33.1 ± 2.6
56.7 ± 2.4
21.2 ± 1.8
16.3 ± 3.0
110 ± 6
150 ± 7
59 ± 6
62 ± 6
153 ± 4
701 ± 3
56 ± 4
390 ± 4.0
75 ± 4
the mass spectral analyses and Dr Kenneth Mewett with the 13
NMR data are gratefully acknowledged.
C
520 ± 2.5
10 ± 2
50.8 ± 1.4
19 ± 4
115 ± 4
References
1
0
76 ± 2.9
39 ± 3
LA
CLA
133 ± 3
221 ± 5
26.3 ± 2.5
42.6 ± 4.0
206 ± 10
225 ± 7
31 ± 3
48 ± 5
Blee, E., 1998. Phytooxylipins and plant defence reactions. Prog. Lipid Res. 37, 33–72.
Campbell, D.L., Lawton, L.A., Beattie, K.A., Codd, G.A., 1994. Comparative assess-
ment of the specificity of the brine shrimp and microtox assays to hepatotoxic
57 ± 7
a
The experiments were carried out under serum-free medium for 24 h.
Serum-containing medium for 72 h.
(microcystin-LR-containing) cyanobacteria. Environ. Toxicol. Water Qual. 9 (1),
b
71–77.
Carvalho, I.S., Miranda, I., Pereira, H., 2006. Evaluation of oil composition of some
crops suitable for human nutrition. Ind. Crops Prod. 24 (1), 75–78.
Dong, M., Oda, Y., Hirota, M., 2000. (10E,12Z,15Z)-9-hydroxy-10,12,15-
octadecatrienoic acid methyl ester as an anti-inflammatory compound
from Ehretia dicksonii. Biosci. Biotechnol. Biochem. 64 (4), 882–886.
Fukushige, H., Wang, C., Simpson, T.D., Gardner, H.W., Hildebrand, D.F., 2005. Purifi-
cation and identification of linoleic acid hydroperoxides generated by soybean
seed lipoxygenases 2 and 3. J. Agric. Food Chem. 53 (14), 5691–5694.
Garcia-Pedrero, J.M., Kiskinis, E., Parker, M.G., Belandia, B., 2006. The SWI/SNF chro-
matin remodeling subunit BAF57 Is a critical regulator of estrogen receptor
function in breast cancer cells. J. Biol. Chem. 281 (32), 22656–22664.
House, R.L., Cassady, J.P., Eisen, E.J., McIntosh, M.K., Odle, 2005. Conjugated linoleic
acid evokes de-lipidation through the regulation of genes controlling lipid
metabolism in adipose and liver tissue. J. Obes. Rev. 6, 247–258.
Jarvis, B.B., Stahly, P., Pavanasasivam, G., Midiwo, J.O., DeSilva, T., Holmlund, C.E.,
Mazzola, E.P., Geoghegan, R.F., 1982. Isolation and characterisation of the tri-
choverroids and new roridins and verrucarins. J. Org. Chem. 47, 1117–1124.
Kato, T., Nakai, T., Ishikawa, R., Karasawa, A., Namai, T., 2001. Preparation of the enan-
tiomers of hydroxy-C18 fatty acids and their anti-rice blast fungus activities.
Tetrahedron: Asym. 12, 2695–2701.
Knothe, G., Weisleder, D., Bagby, M.O., 1993. Hydroxy fatty acids through hydrox-
ylation of oleic acid with selenium dioxide/tert-butyl hydroperoxide. J. Am. Oil
Chem. Soc. 70, 401–404.
Knothe, G., Bagby, M.O., Weisleder, D., Peterson, R.E., 1994. Allylic mono-
and dihydroxylation of isolated double bonds with selenium dioxide/tert-
butylhydroperoxide. NMR characterization of long-chain enols, allylic and
saturated 1,4-diols, and enones. J. Chem. Soc., Perkin Trans. 2, 1661–1669.
Knothe, G., Bagby, M.O., Weisleder, D., 1995. Fatty alcohols through hydroxylation of
symmetrical alkenes with selenium dioxide/tertbutylhydroperoxide. J. Am. Oil
Chem. Soc. 72, 1021–1026.
of hydroxylated LA exhibited relatively low cytotoxic effect against
K562 cells (Table 3), indicating that a free carboxylic acid group is
essential for significant inhibition of cell proliferation.
Cytotoxicities of di-hydroxylated derivatives of CLA against
these cancerous cell lines were also investigated. However,
they exhibited insignificant cytotoxic effect toward these cells,
even lower than the parent compound CLA. Among these di-
hydroxylated derivatives, only compound 15, at 200 ␮M, inhibited
approximately 61% and 28% cell proliferation of MCF-7 and
RPMI8226 cells, respectively. In comparison with brine shrimp
assay results, the di-hydroxylated derivatives displayed selective
toxicity towards brine shrimp but low toxicity towards a range of
human cancer cell lines.
Cytotoxic effect of compounds 3, 5, 7, 9, and CLA and LA were
also evaluated on K562 cells in serum-containing medium at 24, 48,
and 72 h of treatment, respectively. Subsequently a 72-h incubation
period was chosen for concentration dependent studies of cytotox-
icity of these hydroxylated LA as they exhibited most inhibition of
cells (K562) proliferation (results not shown). Under these condi-
tions, compound 7 was most potent with an IC50 value of 39 ␮M
(
Table 3), followed by compounds 9, 3, LA, 5 and CLA (Table 3). It is
interesting to note the IC50 values of the test compounds were sig-
nificantly lower in serum-free medium than in serum-containing
medium, even at 72 h exposure. Binding of the test compounds to
albumin or lipid in the serum thus may sequester the compounds
from coming in contact with the cells, and thereby reduce cytotoxic
effects.
Knowles, P.F., 1975. Recent research on safflower, sunflower, and cotton. J. Am. Oil
Chem. Soc. 52, 374–376.
Kuklev, D.V., Christie, W.W., Durand, T., Rossi, J.C., Vidal, J.P., Kasyanov, S.P., Akulin,
V.N., Bezuglov, V.V., 1997. Synthesis of keto- and hydroxydienoic compounds
from linoleic acid. Chem. Phys. Lipids 85, 125–134.
Lands, W.E.M., 2001. Biochemistry and physiology of eicosanoid precursors in cell
membranes. Eur. Heart J. Suppl. 3, D22–D25.
Li, Z., Yang, D., Jiang, L., Ji, J., Ji, H., Zeng, X., 2007. Lipase-catalyzed esterification of
conjugated linoleic acid with L-carnitine in solvent-free system and acetonitrile.
Bioprocess Biosyst. Eng. 30, 331–336.
4
. Conclusions
Piazza, G.J., Nunez, A., Foglia, T.A., 2003. Isolation of unsaturated diols after oxidation
of conjugated linoleic acid with peroxygenase. Lipids 38, 255–261.
Truc, M.N., Daesung, L., 2001. A novel reactivity of SeO2 with 1,3-dienes: forma-
tion of syn 1,2- and 1,4-diols via a facile C–Se bond oxidation. Org. Lett. 3 (20),
3161–3163.
Truitt, A., McNeill, G., Vanderhoek, J.Y., 1999. Antiplatelet effects of conjugated
linoleic acid isomers. Biochim. Biophys. Acta 1438, 239–246.
Tsitsigiannis, D.I., Keller, N.P., 2007. Oxylipins as developmental and host–fungal
communication signals. Trends Microbiol. 15 (3), 109–118.
US EPA, 1996. Ecological effects test guidelines OPPTS 850.3020 Honeybee acute
contact toxicity test, EPA 712–C-96–147, Washington, DC.
Vellosillo, T., Martinez, M., Lopez, M.A., Vicente, J., Cascon, T., Dolan, L., Hamberg,
M., Castresana, C., 2007. Oxylipins produced by the 9-lipoxygenase pathway in
Arabidopsis regulate lateral root development and defense responses through a
specific signaling cascade. Plant Cell. 19, 831–846.
Yoshiki, H., Yasushi, N., Hirotaka, M., Hirotoshi, T., Yoh-ichi, M., Takanao, M.J., 1998.
Antitumor activity of (10E, 12Z)-9-hydroxy-10,12-octadecadienoic acid from rice
bran. J. Ferment. Bioeng. 86 (2), 149–153.
In this study a series of mono- and di-hydroxylated derivatives
of octadecadienoic acid were prepared by reaction of homo-
conjugated and conjugated dienes with SeO . Vicinal dihydroxy
2
compounds as novel products were obtained via a proposed novel
syn di-hydroxylation of 1,3-dienes with SeO , involving a [4 + 2]
2
cycloaddition followed by C–Se bond oxidation. Our results demon-
strated that these allylic hydroxylated derivatives exhibit more
significant toxicity against brine shrimp than CLA–Me and LA–Me.
This implies that the–OH group next to the double bond con-
tributes to toxic activity towards brine shrimp. We found that the
free fatty acid of mono-hydroxy compounds have moderate cyto-
toxicity towards K562, RPMI8226 cell lines and relatively weak
cytotoxicity towards HepG2, MCF-7 cell lines. Both the mono- and