Production of conjugated linoleic acids through KOH-catalyzed dehydration of ricinoleic acid

Article abstract of DOI:10.1016/S0009-3084(02)00052-X

Production of conjugated linoleic acids (CLA) using castor oil as starting material involves conversion of ricinoleic acid to methyl 12-mesyloxy-octadec-9-enoate (MMOE) followed by dehydration. This process usually uses 1,8-diazabicyclo-(5.4.0)-undec-7-ene (DBU) as an expensive dehydrating reagent. The present study reports that potassium hydroxide (KOH) can serve as a dehydrating reagent in replacement of DBU. The results showed that conversion of MMOE to CLA catalyzed by KOH was an efficient reaction, with a 77% conversion efficiency at 80°C. The CLA isomeric profile produced in KOH-catalyzed dehydration reaction was similar to that catalyzed by DBU. The CLA mixture produced in KOH-catalyzed dehydration of MMOE at 80°C contained 72% 9c,11t-18:2 and 26% 9c,11c-18:2 while in that catalyzed by DBU, 9c,11t-18:2 and 9c,11c-18:2 accounted for 78 and 16%, respectively. It was found that the temperature of dehydration was an important factor in the determination of CLA isomer composition and yield of conversion. Elevating the temperature from 78 to 180°C decreased not only the conversion efficiency but also production of total c,t-18:2 and c,c-18:2 isomers regardless of dehydration catalyzed by either DBU or KOH. It is concluded that KOH may replace DBU as a dehydrating reagent in conversion of MMOE to CLA when the reaction conditions are optimized.

Full text of DOI:10.1016/S0009-3084(02)00052-X

Chemistry and Physics of Lipids 119 (2002) 23Á31
/
www.elsevier.com/locate/chemphyslip
Production of conjugated linoleic acids through KOH-
catalyzed dehydration of ricinoleic acid
a,b
Lin Yang , Yu Huang , Han Qing Wang , Zhen-Yu Chen *
c
a
b,
a
State Key Laboratory of OXO Synthesis & Selective Oxidation, Lanzhou Institute of Chemical Physics, The Chinese Academy of
Sciences of Lanzhou, and Department of Chemistry, Henan Normal University, Henan, People’s Republic of China
b
Department of Biochemistry, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, People’s Republic of China
Department of Physiology, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong, People’s Republic of China
c
Received 3 December 2001; received in revised form 14 May 2002; accepted 14 May 2002
Abstract
Production of conjugated linoleic acids (CLA) using castor oil as starting material involves conversion of ricinoleic
acid to methyl 12-mesyloxy-octadec-9-enoate (MMOE) followed by dehydration. This process usually uses 1,8-
diazabicyclo-(5.4.0)-undec-7-ene (DBU) as an expensive dehydrating reagent. The present study reports that potassium
hydroxide (KOH) can serve as a dehydrating reagent in replacement of DBU. The results showed that conversion of
MMOE to CLA catalyzed by KOH was an efficient reaction, with a 77% conversion efficiency at 80 8C. The CLA
isomeric profile produced in KOH-catalyzed dehydration reaction was similar to that catalyzed by DBU. The CLA
mixture produced in KOH-catalyzed dehydration of MMOE at 80 8C contained 72% 9c,11t-18:2 and 26% 9c,11c-18:2
while in that catalyzed by DBU, 9c,11t-18:2 and 9c,11c-18:2 accounted for 78 and 16%, respectively. It was found that
the temperature of dehydration was an important factor in the determination of CLA isomer composition and yield of
conversion. Elevating the temperature from 78 to 180 8C decreased not only the conversion efficiency but also
production of total c,t-18:2 and c,c-18:2 isomers regardless of dehydration catalyzed by either DBU or KOH. It is
concluded that KOH may replace DBU as a dehydrating reagent in conversion of MMOE to CLA when the reaction
conditions are optimized. # 2002 Elsevier Science Ireland Ltd. All rights reserved.
Keywords: Alkali isomerization; Conjugated linoleic acids; Linoleic acid; Ricinoleic acid
1
. Introduction
health benefits associated with its consumption.
CLA has been shown to inhibit tumor growth in
experimental animals (Scimeca, 1999), to reduce
atherosclerotic plaque (Lee et al., 1994; Nicolosi et
al., 1997) and to lower serum cholesterol (Yeung et
al., 2000). CLA may boost immune functions
There is increasing interest in producing con-
jugated linoleic acids (CLA) as a food ingredient
and health supplement because of its possible
(Cook et al., 1993; Miller et al., 1994) and reduce
*
Corresponding author. Tel.: ꢀ
603-5123
E-mail address: zhenyuchen@cuhk.edu.hk (Z.-Y. Chen).
/
85-2-2609-6382; fax: ꢀ
/
85-2-
fat accumulation while it increases muscle and
bone mass (Dugan and Aalhus, 1999; Li and
2
0
009-3084/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0 0 0 9 - 3 0 8 4 ( 0 2 ) 0 0 0 5 2 - X

24
L. Yang et al. / Chemistry and Physics of Lipids 119 (2002) 23Á31
/
Watkins, 1998). However, CLA as an antioxidant
is found inconclusive (Chen et al., 1997; Ha et al.,
2.2. Conversion of ricinoleic acid from castor oil to
CLA
1
990; Ip et al., 1991; van den Berg et al., 1995).
High-performance liquid chromatographic ana-
2
(
.2.1. Preparation of ricinoleic acid methyl ester
RAME)
As shown in Fig. 1, ricinoleic acid was converted
lysis has demonstrated that CLA is a mixture of
several isomers (Sehat et al., 1998). Dietary CLA
predominately originates from dairy products via
biohydrogenation of polyunsaturated fatty acids
by rumen bacteria (Chin et al., 1992; Ha et al.,
to RAME by mixing 6 g castor oil, 50 ml of
methanol and 12 ml of 14% boron trifluoride in
methanol. The mixture was refluxed for 1 h at
1989; Kepler and Tove, 1967). CLA is also
biosynthesized from desaturation of trans-11 oc-
65 8C and then cooled to room temperature
followed by adding 100 ml of distilled water. The
reaction mixture was then extracted three times
with 50 ml of diethyl ether, followed by removal of
the solvent using a rotary evaporator under
vacuum. RAME (4.7 g) was purified on a silica
tadecenoic acid catalyzed by delta-9 desaturase
(
Mahfouz et al., 1980) in mammary tissue (Grii-
nari and Bauman, 1999; Griinari et al., 2000),
rodents and humans (Palmquist and Santora,
1
(
999). Commercial production of CLA includes:
1) alkali isomerization of linoleic acid; (2) dehy-
column (60ꢁ3.4 cm, I.D.) using a mixture of
/
petroleum ether and diethyl ether (6:4, vol./vol.) as
an eluting solvent, and then dissolved in CDCl for
dration of ricinoleic acid methyl ester (Body and
Shorland, 1965); and (3) microbial synthesis of
3
1
H-NMR analysis (Bruker Avance DPX 300
Spectrometer, 300 MHZ, F a¨ llanden, Switzerland).
9
c,11t-18:2 from linoleic acid using cultures of
different microorganisms (Pariza and Yang, 1999;
Jiang et al., 1998). Each of these methods produces
a different mixture of CLA isomers. Among these
methods, alkali isomerization of linoleic acid is the
most commonly method for production of CLA
because it is economically viable. In contrast, the
production of CLA via dehydration of ricinoleic
acid is an efficient reaction but it involves many
steps (Berdeaux et al., 1997; Lie Ken Jie et al.,
1
H-NMR spectrum showed that RAME had the
following characteristics (CDCl , d): 0.76 (t, 3H,
3
CH ); 1.20Á1.50 (m, 20H, CH ); 1.93 (m, 2H, 8-
/
3
2
H); 2.09 (t, 2H, 11-H); 2.18 (t, 2H, 2-H); 3.48 (s,
H, CH-OH); 3.54 (s, 3H, COOCH ); 5.26Á5.42
1
/
3
(m, 2H, 9H and 10H).
2.2.2. Preparation of methyl 12-mesyloxy-octadec-
9-enoate (MMOE)
1
997), and uses 1,8-diazabicyclo-(5,4,0)-undec-7-
ene (DBU), an expensive dehydration reagent. The
objectives of the present study were to simplify the
dehydration process of ricinioleic acid, and to
compare the isomeric distribution of CLA pro-
duced by the different dehydrating agents.
As shown in Fig. 1, MMOE was prepared by
mixing RAME (2 g), methanesulfonyl chloride
(1.6 g, 1.4 mmol), triethylamine (2 ml) and
dichloromethane (20 ml), and stirring it for 45
min on an ice water bath. The reaction mixture
was diluted with 50 ml of dichloromethane and
successively washed with 10 ml of HCl solution (2
N) and twice with 20 ml of distilled water. The
mixture was then dried over anhydrous sodium
sulfate, filtered, and the solvents were removed in
2
. Materials and methods
2
.1. Reagents
1
a rotary evaporator to obtain 2.2 g MMOE. H-
NMR spectrum showed that MMOE had the
Castor oil (89% ricinoleic acid), boron trifluor-
ide in methanol (14%, w/w), and linoleic acid were
purchased from Sigma (St. Louis, MO). Methan-
sulfonylchloride was obtained from Merck-Schu-
chardt (Hohenbrunn, Germany) while DBU was
purchased from Fluka Chemie AG (Buchs, Swit-
zerland). CLA standard was obtained from Sigma.
following characteristics (CDCl , d): 0.86 (t, 3H,
3
CH ); 1.25Á1.63 (m, 20H, CH ), 1.98 (m, 2H, 8-
/
3
2
H), 2.26 (t, 2H, 2-H), 2.56 (m, 2H, 11-H), 2.95 (s,
3H, CH of OMS), 3.65 (s, 3H, COOCH ), 4.66
3
3
(quintet, 1H, CH-OMs), 5.33 (m, 1H, 9-H), 5.51
(m, 1H, 10-H).

L. Yang et al. / Chemistry and Physics of Lipids 119 (2002) 23Á
/
31
25
Fig. 1. Production of conjugated linoleic acids. (A) Castor oil was converted to fatty acid methyl esters followed by isolation of
ricinoleic acid methyl ester (RAME, 1) which was then used for preparation of methyl 12-mesyloxy-octadec-9-enoate (MMOE, 2);
conversion of MMOE to conjugated linoleic acids was catalyzed by 1,8-diazabicyclo-(5.4.0)-undec-7-ene (DBU) or potassium
hydroxide (KOH). (B) Alkali isomerization of linoleic acids.
2.2.3. Production of CLA via dehydration of
MMOE catalyzed by KOH
of HCl solution (6 N) and extracted three times
with 40 ml of diethyl ether. The combined ether
extract was then washed with 15 ml of NaCl
solution (0.9%) and dried over anhydrous sodium
sulfate. Diethyl ether was removed in a rotary
evaporator and the product was purified on a silica
KOH-dependent dehydration of MMOE was
accomplished in either absolute ethanol or ethyl-
ene glycol (50, 80, 120, 160, 180 8C). In brief, 20
ml of absolute ethanol (or ethylene glycol) and 1.4
g KOH were placed in a two-neck round-bottom
flask (100 ml) equipped with a reflux condenser
and a stream of nitrogen gas was bubbled through
the solution for 5 min followed by adding 0.5 g
MMOE. Then reaction mixture was refluxed with
constant stirring for 8 h under a gentle stream of
nitrogen gas. The flask was removed from the oil
bath and cooled to room temperature. To the
flask, 10 ml of distilled water was added and the
solution was then transferred to a 250 ml separa-
tion funnel. The mixture was acidified with 10 ml
column (30ꢁ2.5 cm, I.D.) and eluted using the
/
solvent mixture of petroleum ether and diethyl
ether (6:4, vol./vol.). The fractions containing
CLA were pooled. Individual CLA isomers were
identified according to the eluting pattern of silver-
ion high performance liquid chromatography (Ag-
HPLC) previously described (Sehat et al., 1998;
Ostrowska et al., 2000). The structure of the CLA
was further confirmed by comparing its NMR
spectra with those published previously (Lie Ken
1
Jie et al., 1997). H-NMR spectrum showed that

26
L. Yang et al. / Chemistry and Physics of Lipids 119 (2002) 23Á31
/
CLA prepared at 78 8C had the following char-
acteristics (CDCl , d): 0.89 (t, 3H, CH ), 1.20Á
water, three times with 15 ml of distilled water,
and then dried over anhydrous sodium sulfate.
Hexane was removed in a rotary evaporator under
/
3
3
1
1
1
.40 (m, 18H, CH ), 2.00Á2.30 (m, 4H, 8-H and
/
2
2
3-H ), 2.28 (t, 2H, 2-H), 5.27Á
/6.28 (m, 4H, 9-,
vacuum. The crude CLA produced the following
1
H- NMR characteristics (CDCl , d): 0.89 (t, 3H,
3
2
0-, 11-, 12-H). The CLA mixture was also
subjected to Ag-HPLC analysis.
CH ); 1.30Á
/
1.40 (m, 18H, CH ); 2.08Á
/2.17 (m,
3
2
4H, 8-H and 13-H ), 2.35 (t, 2H, 2-H ); 5.28-6.30
(m, 4H, 9/10-H, 10/11-H, 11/12-H and 12/13-H).
2
2
2
2.2.4. Production of CLA via dehydration of
MMOE catalyzed by DBU
The method previously described by Lie Ken Jie
et al. (1997) was adopted for dehydration of
MMOE. In brief, 0.5 g MMOE was mixed with
2.4. Ag-HPLC analysis
Individual CLA isomers were quantified accord-
ing to Ostrowska et al. (2000). In brief, the above
mentioned CLA preparations (8 mg/ml) were
separated using an Alltech Model 525 HPLC
equipped with a ternary pump delivery system, a
stainless steel silver-ion impregnated column
1
.3 g DBU in 4 ml of dimethylsulfoxide in a two-
neck round-bottom flask. The reaction mixture
was refluxed at different temperatures under
nitrogen gas for 12 h. The reaction mixture was
purified on a silica column (30ꢁ2.5 cm, I.D.)
/
using the solvent mixture of n-hexane and diethyl
ether (9:1, vol./vol.). The CLA methyl ester frac-
tions were hydrolyzed with 10 ml of ethanolic
KOH solution (5%) at room temperature for 1 h.
After the reaction, 12 ml of distilled water was
added and the solution was acidified with 10 ml of
(250ꢁ4.6 mm, I.D., 5 mm, Chrompack, Bridge-
/
water, NJ), a rheodyne valve injector. Hexane
containing 1.4% acetic acid and 0.014% acetoni-
trile was chosen as a mobile phase at a flow rate of
1.0 ml/min. The separated individual CLA isomers
were quantified at 233 nm using a UV detector
(UVIS-205, Alltech, Deerfield, IL). The Ag-HPLC
is not capable of separating a cis,trans-18:2 species
with a trans,cis species with same conjugated
double bond positions. For example, 9c,11t-18:2
and 9t,11c-18:2 co-elute as a single peak. To
simplify the presentation, the two isomers were
hereafter designated as 9c,11t-18:2 only. This will
apply for 8c,10t-18:2, and 10c,12t-18:2.
1
N HCl solution followed by three extractions
with 20 ml of diethyl ether. The ether extract was
washed twice with 15 ml of distilled water and
dried over anhydrous sodium sulfate. After re-
moval of ether, the crude CLA mixture was
crystallized three times in 10 ml of ethanol. The
1
resulting CLA had the following H-NMR char-
acteristics (CDCl , d): 0.88 (t, 3H, CH );1.20Á
/1.62
3
3
(
m, 18H, CH ); 2.00Á
/
2.09 (m, 4H, 8-H and 13-
6.28 (m, 4H, 9-H, 10-
2
2
H ); 2.33 (t, 2H, 2-H ); 5.36Á
/
2
2
H, 11-H and 12-H).
3. Results
2.3. Production of CLA by alkali isomerization of
linoleic acid
Castor oil was used as a source of ricinoleic acid.
Transesterification process led to produce a mix-
ture of fatty acid methyl esters in which 89% was
RAME. A silica column was used to isolate
RAME with purity of 99%. RAME was subse-
quently used to prepare MMOE. Potassium hy-
Linoleic acid was isomerized to CLA as pre-
viously described by Chin et al. (1992). In brief,
linoleic acid (1 g) was mixed with 2.6 g KOH in 10
ml of ethylene glycol and refluxed at different
temperatures under nitrogen for 4 h. To the
reaction mixture, 20 ml of methanol was added
followed by acidification with 40 ml of 3 N HCl.
The reaction mixture was then extracted three
times with 20 ml hexane. The hexane extract was
washed thrice with 15 ml of 30% methanol in
droxide was evaluated in this study as
dehydrating reagent to produce CLA. As shown
in Fig. 2, only two CLA isomers, namely 9c,11t Á
a
/
18:2 and 9c,11c-18:2, were obtained at 50 8C.
When the temperature was increased to 120 8C,
two additional isomers namely 10t,12t-18:2 and
9t,11t-18:2 were also produced. At 160 8C, t,t-

L. Yang et al. / Chemistry and Physics of Lipids 119 (2002) 23Á
/
31
27
Fig. 2. Typical silver-ion high-performance liquid chromatograms of conjugated linoleic acids as free fatty acids produced by KOH-
catalyzed dehydration of methyl 12-mesyloxy-octadec-9-enoate (MMOE). Separation was performed on a silver-ion impregnated
Chrompack analytical column (250ꢁ4.6 mm. I.D.) using hexane/1.4% acetic acid/0.014% acetonitrile as mobile phase at a flow rate of
/
1
ml/min.
1
8:2 including 9t,11t-18:2, 10t,12t-18:2 and
t,10t-18:2 became the major products followed
production of trace amounts of 9c,11c-18:2,
10c,12c-18:2, and 10t,12t-18:2. The temperature
had less influence on the profile of CLA isomers
formed. In the absence of oxygen, 94% of total
CLA formed was c,t-18:2 isomers.
The conversion efficiency was different among
the three methods. As shown in Fig. 6, the yield of
alkali isomerization could reach 90% when the
temperature was set at 160 8C or above, but below
8
by total c,t-18:2 isomers, while only minor
amounts of total c,c-18:2 isomers were formed
during the dehydration process. The change in
CLA profile with increasing temperature is best
illustrated in Fig. 3. Elevation of temperature
increased production of total t,t-18:2 isomers
while it decreased both total c,t-18:2 and c,c-
120 8C, the conversion was 20% or less. In
1
8:2 products.
When DBU was used as a dehydrating reagent,
contrast, when KOH was used to dehydrate
MMOE, the conversion was most efficient at
the CLA profile was similar to that when MMOE
was dehydrated by KOH (Fig. 4). At less than
80 8C (77%). When the temperature of dehydra-
tion was increased to 160 8C, the conversion
efficiency decreased gradually to 34%. The yield
of CLA production by dehydration of MMOE
using DBU reached 88% at 80 8C similar to
KOH. At higher reaction temperatures, the con-
8
0 8C, the major product was 9c,11t-18:2 fol-
lowed by 9c,11c-18:2 with production of a trace
amount of 9t,11t-18:2. When the temperature was
elevated to 120 8C, the amount of 9t,11t-18:2 was
increased to a level similar to that of 9c,11c-CLA,
and at 180 8C, there were total 11 CLA isomers
formed with t,t-18:2 being the major isomers.
Among these t,t-18:2 isomers, 9t,11t-18:2 was
most abundant. Similar to KOH as a dehydrating
reagent, the profile of CLA isomers was a function
of temperature. At a lower temperature, c,t-18:2
isomers were predominant while total t,t-18:2
isomers became the major CLA isomers at a
higher temperature (Fig. 3).
version efficiency was less (60Á63%) but relatively
/
consistent. Among the three methods, the conver-
sion efficiency of alkali isomerization was most
temperature-dependent, while the change in the
CLA profile was most significant with increasing
temperature when MMOE was dehydrated by
either DBU or KOH.
The profile in alkali isomerization of linoleic
acid was different from that of the CLA mixture
produced by dehydration of MMOE with DBU or
KOH. As shown in Fig. 5, 10c,12t-CLA and
4
. Discussion
Dehydration of MMOE is the most attractive
alternative to produce CLA by alkali isomeriza-
tion of linoleic acid. The starting material, castor
9
c,11t-CLA were the two major isomers with

28
L. Yang et al. / Chemistry and Physics of Lipids 119 (2002) 23Á31
/
oil, is relatively inexpensive. However, DBU, the
previously reported (Lie Ken Jie et al., 1997), is six
times as expensive as KOH. The present study was
to seek an alternative dehydrating reagent to
replace DBU without comprising the conversion
efficiency. This is the first to show that KOH
served as an efficient dehydrating reagent of
MMOE. This method has the potential of being
scaled up to produce an inexpensive and efficient
CLA product.
The result clearly demonstrated that conversion
of MMOE to CLA by KOH was an efficient
reaction with a 77% conversion efficiency com-
pared to 88% conversion efficiency using DBU at
80 8C (Fig. 3). The present results using DBU are
in agreement with those reported by Berdeaux et
al. (1997) to produce pure 9c,11t-18:2 at a
conversion efficiency of 87% (sum of 9c,11t-18:2,
9
c,11c-18:2 and 9t,11t-18:2). Lie Ken Jie et al.
1997) showed a similar conversion efficiency of
1% (sum of 9c,11c-18:2 and 9c,11t-18:2) for
(
9
dehydration of MMOE using DBU. The present
study found that replacement of DBU with KOH
was chemically viable without significantly com-
prising the yield of CLA produced.
The CLA composition produced by KOH
dehydration was similar to that produced by
DBU dehydration. In KOH dehydration at
Fig. 3. Effect of temperature on isomeric distribution of total
t,t-18:2, c,t-18:2 and c,c-18:2. (A) KOH-catalyzed dehydra-
tion of methyl 12-mesyloxy-octadec-9-enoate (MMOE). (B) 1,8-
diazabicyclo-(5.4.0)-undec-7-ene (DBU)-catalyzed dehydration
of MMOE. (C) Alkali-isomerization of linoleic acid.
8
0 8C, 9c,11t-18:2 was predominant, accounting
for 72% of total CLA, followed by 9c,11c-18:2
26%). Similarly, DBU dehydration led to produc-
tion of a CLA mixture in which 9c,11t-18:2
(
Fig. 4. Typical silver-ion high-performance liquid chromatograms of conjugated linoleic acids as free fatty acids produced by1,8-
diazabicyclo-(5.4.0)-undec-7-ene (DBU)-catalyzed dehydration of methyl 12-mesyloxy-octadec-9-enoate (MMOE). Separation was
performed on a silver-ion impregnated Chrompack analytical column (250ꢁ/4.6 mm. I.D.) using hexane/1.4% acetic acid/0.014%
acetonitrile as mobile phase at a flow rate of 1 ml/min.

L. Yang et al. / Chemistry and Physics of Lipids 119 (2002) 23Á
/
31
29
Fig. 5. Typical silver-ion high-performance liquid chromatograms of conjugated linoleic acids as free fatty acids produced by alkali-
isomerization of linoleic acids. Separation was performed on a silver-ion impregnated Chrompack analytical column (250ꢁ4.6 mm.
I.D.) using hexane/1.4% acetic acid/0.014% acetonitrile as mobile phase at a flow rate of 1 ml/min.
/
dration, because the trans configuration is chemi-
cally more stable than cis configuration. In
contrast, the isomer distribution during alkali
isomerization of linoleic acid was less influenced
by temperature. The conversion of MMOE to
CLA was most efficient at 80 8C using either
DBU or KOH as dehydrating agent. The conver-
sion rate declined at 180 8C probably due to
thermal oxidation of CLA (Eulitz et al., 1999;
Yang et al., 2000). The lower temperature (B
/
80 8C) for conversion of MMOE to CLA using
KOH dehydration has an advantage over CLA
produced by alkali isomerization of linoleic acid
Fig. 6. Effect of temperature on production of conjugated
linoleic acids. j, KOH-catalyzed dehydration of methyl 12-
mesyloxy-octadec-9-enoate (MMOE). m, 1,8-diazabicyclo
because of possible double bond migrating (ꢀ
60 8C).
We are unaware of any studies that have
/
1
(5.4.0)undec-7-ene (DBU)-catalyzed dehydration of MMOE.
'
, alkali-isomerization of linoleic acid (LA).
compared the isomeric distribution of CLA pro-
duced either by DBU dehydration of MMOE or
alkali isomerization of linoleic acid. In the present
study, the dehydration reaction of MMOE cata-
lyzed by both KOH and DBU produces mainly
one positional isomer (9c,11t-18:2) at temperature
below 120 8C (Figs. 2 and 4). In contrast, alkali
isomerization reaction of linoleic acid produces
mainly two CLA isomers (9c,11t-18:2 and
10c,12t-18:2) at the three temperatures studied
(Fig. 5). This can be explained by the different
natures of the two reaction mechanisms. The
dehydration of MMOE tends to occur between
positions 12 and 11, leading to production of
9c,11t-18:2 with the lowest free energy due to
resonance conjugated structure. However, the
isomerization of linoleic acid at high temperature
involves migration of the two double bonds. Thus,
accounted for 78% followed by 9c,11c-18:2, which
accounted for 16% of total CLA at 80 8C. The
data were in agreement with that of Berdeaux et al.
(
1997), who showed that DBU dehydration of
MMOE produced 66% 9c,11t-18:2, 20.7% 9c,11c-
8:2 and 0.6% 9t,11t-18:2. However, the profile of
1
CLA isomers produced by KOH dehydration was
markedly different from that of alkali isomeriza-
tion of linoleic acid, in which two CLA isomers
were produced (c,t-10,12-18:2 and 9c,11t Á
about equal abundance (Fig. 5).
/18:2) in
The temperature of the dehydration reaction
was an important factor in the CLA isomer
composition and yield of conversion. When the
temperature was elevated to 180 8C, 9t,11t-18:2
was predominant for both DBU and KOH dehy-

30
L. Yang et al. / Chemistry and Physics of Lipids 119 (2002) 23Á31
/
leic Acid Research, vol. 1. AOCS Press, Champaign, IL,
USA, pp. 354Á368.
Eulitz, K., Yurawecz, M.P., Ku, Y., 1999. The oxidation of
conjugated linoleic acid. In: Yurawecz, M.P., Mossoba,
M.M., Kramer, J.K.G., Pariza, M.W., Nelson, G.J. (Eds.),
Advances in Conjugated Linoleic Acid Research, vol. 1.
the double bond at position 9 migrates towards
position 10 to produce 9c,11t-18:2, and the double
bond at position 12 migrates towards position 11
to form 9c,11t-18:2.
/
The present study examined the effects of
temperature and dehydration reagents on the
CLA production and when ricinoleic acid and
linoleic acid were used as starting materials. The
results clearly demonstrated that dehydration of
MMOE and alkali isomerization of linoleic acid
produced different CLA profiles. It is premature
to assess merits of one method over another at the
present time because the information on biological
activities of each CLA isomer is still limited. Some
evidences suggested that 9c,11t-18:2 was the active
isomer (Ip et al., 1991; Knekt et al., 1996) while
several recent reports showed that 10c,12t-18:2
was biologically more potent than 9c,11t-18:2
AOCS Press, Champaign, IL, USA, pp. 55Á63.
/
Griinari, J.M., Bauman, D.E., 1999. Biosynthesis of conjugated
linoleic acid and its incorporation into meat and milk in
ruminants. In: Yurawecz, M.P., Mossoba, M.M., Kramer,
J.K.G., Pariza, M.W., Nelson, G.J. (Eds.), Advances in
Conjugated Linoleic Acid Research, vol. 1. AOCS Press,
Champaign, IL, USA, pp. 180Á200.
/
Griinari, J.E., Corl, B.A., Lacy, S.H., Chouinard, P.Y.,
Nurmela, K.V., Bauman, D.E., 2000. Conjugated linoleic
acid is synthesized endogenously in lactating dairy cows by
delta(9)-desturase. J. Nutr. 130, 2285Á2291.
/
Ha, Y.L., Grimm, N.K., Pariza, M.W., 1989. Newly recognized
anticarcinogenic fatty acids: identification and quantifica-
tion in natural and processed cheeses. J. Agric. Food Chem.
3
7, 75Á81.
/
(
Leung and Liu, 2000; Lin et al., 2001; Park et
Ha, Y.L., Storkson, J., Pariza, M.W., 1990. Inhibition of
benz(a)pyrene-induced mouse forstomach neoplasia by
conjugated dienoic derivatives of linoleic acid. Cancer Res.
al., 1999). If 9c,11t-18:2 were the only active
isomer, dehydration of MMOE catalyzed by
KOH would be a preferred method to produce
CLA. If 9c,11t-18:2 and 9c,11t-18:2 were both
biologically active, alkali isomerization would be
preferred to synthesize the CLA because it pro-
duced equal amounts of these two isomers.
5
Ip, C., Chin, S.F., Scimeca, J.A., Pariza, M.W., 1991. Mam-
0, 1097Á1101.
/
mary cancer prevention by conjugated dienoic derivative of
linoleic acid. Cancer Res. 51, 6118Á6124.
/
Jiang, J., Bjorck, L., Fonden, R., 1998. Production of con-
jugated linoleic acid by dairy starter cultures. J. Appl.
Microbiol. 85, 95Á
Kepler, C.R., Tove, S.B., 1967. Biohydrogenation of unsatu-
rated fatty acids. J. Biol. Chem. 242, 5606Á5612.
/
102.
/
Knekt, P., Jarvinen, R., Seppanen, R., Pukkala, E., Aromaa,
References
A., 1996. Intake of dairy products and the risk of breast
cancer. Br. J. Cancer 73, 687Á691.
Lee, K.N., Kritchevsky, D., Pariza, M.W., 1994. Conjugated
/
Berdeaux, O., Christie, W.W., Gunstone, F.D., Sebedio, J.L.,
1
997. Large-scale synthesis of methyl cis-9, trans-11-octa-
decadienoate from methyl ricinoleate. J. Am. Oil Chem.
Soc. 74, 1011Á1015.
linoleic acid and atheroclerosis in rabbits. Atheroclerosis
1
08, 19Á25.
/
/
Leung, Y.H., Liu, R.H., 2000. Trans-10, cis-12-conjugated
linoleic acid isomer exhibits stronger oxyradical scavenging
capacity than cis-9, trans-11 conjugated linoleic acid
Body, D.R., Shorland, F.B., 1965. The geometric isomers of
conjugated octadecadienonates from dehydration of methyl
ricinoleate. J. Am. Oil Chem. Soc. 42, 5Á8.
/
Chen, Z.Y., Chan, P.T., Kwan, K.Y., Zhang, A., 1997.
Reassessment of the antioxidant activity of conjugated
isomer. J. Agric. Food Chem. 48, 5469Á5475.
/
Li, Y., Watkins, B.A., 1998. Conjugated linoleic acids alter
bone fatty acid composition and reduce ex vivo bone
prostaglandin E2 biosynthesis in rats fed n-6 or n-3 fatty
linoleic acids. J. Am. Oil Chem. Soc. 74, 749Á
Chin, S.F., Liu, W., Storkson, J.M., Ha, Y.L., Pariza, M.W.,
992. Dietary sources of conjugated dienoic isomers of
linoleic acid, a newly recognized class of anticarcinogens. J.
Food Comp. Anal. 5, 185Á197.
/753.
1
acids. Lipids 33, 417Á425.
/
Lin, Y., Kreeft, A., Schuurbiers, J.A., Draijer, R., 2001.
Different effects of conjugated linoleic acid isomers on
lipoprotein lipase activity in 3T3 adipocytes. J. Nutr.
/
Cook, M.E., Miller, C.C., Park, Y., Pariza, M.W., 1993.
Immune modulation by altered nutrient metabolism: nutri-
tional control of immune-induced growth depression. Poul-
Biochem. 12, 183Á189.
/
Lie Ken Jie, M.S.F., Pasha, M.K.&, Alam, M.S., 1997.
Synthesis and nuclear magnetic resonance properties of all
geometrical isomers of conjugated linoleic acids. Lipids 32,
try Sci. 72, 1301Á1305.
/
Dugan, M.E.R., Aalhus, J.L., 1999. Feeding CLA to pigs. In:
Yurawecz, M.P., Mossoba, M.M., Kramer, J.K.G., Pariza,
M.W., Nelson, G.J. (Eds.), Advances in Conjugated Lino-
1
041Á1044.
/

L. Yang et al. / Chemistry and Physics of Lipids 119 (2002) 23Á
/
31
31
Mahfouz, M.M., Valicenti, A.J., Holman, R.T., 1980. Desa-
turation of isomeric trans-octadecenoic acids by rat liver
Park, Y., Albright, K.L., Storkson, J.M., Liu, W., Pariza,
M.W., 1999. Evidence that the trans-10, cis-12 isomers of
conjugated linoleic acid induces body composition in mice.
microsomes. Biochim. Biophys. Acta 618, 1Á12.
/
Miller, C.C., Park, Y., Pariza, M.W., Cook, M.E., 1994.
Feeding conjugated linoleic acid to animals partially over-
comes catabolic responses due to endotoxin injection.
Lipids 34, 235Á241.
/
Scimeca, J.A., 1999. Cancer inhibition in animals. In: Yur-
awecz, M.P., Mossoba, M.M., Kramer, J.K.G., Pariza,
M.W., Nelson, G.J. (Eds.), Advances in Conjugated Lino-
leic Acid Research, vol. 1. AOCS Press, Champaign, IL,
Biochem. Biophys. Res. Commun. 198, 1107Á1112.
/
Nicolosi, R.J., Rogers, E.J., Kritchevsky, D., Scimeca, J.A.,
Huth, P.J., 1997. Dietary conjugated linoleic acid reduces
plasma lipoproteins and early aortic atherosclerosis in
USA, pp. 420Á443.
/
Sehat, N., Yurawecz, M.P., Roach, J.A.G., Mossoba, M.M.,
Kramer, J.K.G., Ku, Y., 1998. Silver-ion high-performance
liquid chromatographic separation and identification of
hypercholesterolemic hamsters. Artery 22, 266Á
Ostrowska, E., Dunshea, F.R., Muralitharan, M., Cross, R.E.,
000. Comparison of silver-ion high-performance liquid
chromatographic quantification of free and methylated
conjugated linoleic acids. Lipids 35, 1147Á1153.
/277.
2
conjugated linoleic acid isomers. Lipids 33, 217Á221.
van den Berg, J.J.M., Cook, N.E., Tribble, D.L., 1995.
/
/
Reinvestigation of the antioxidant properties of conjugated
Palmquist, D.L., Santora, J.E., 1999. Endogenous synthesis of
rumenic acid in rodents and humans. In: Yurawecz, M.P.,
Mossoba, M.M., Kramer, J.K.G., Pariza, M.W., Nelson,
G.J. (Eds.), Advances in Conjugated Linoleic Acid Re-
linoleic acid. Lipids 30, 599Á605.
/
Yeung, C.H., Yang, L., Huang, Y., Wang, J., Chen, Z.Y., 2000.
Dietary conjugated linoleic acid mixture affects. The activity
of intestinal acyl coenzyme A:cholesterol acyltransferase in
search, vol. 1. AOCS Press, Champaign, IL, USA, pp. 201Á
/
hamsters. Br. J. Nutr. 84, 935Á941.
/
2
08.
Yang, L., Leung, L.K., Huang, Y., Chen, Z.Y., 2000. Oxidative
stability of conjugated linoleic acid isomers. J. Agric. Food
Pariza, M.W., Yang, X.Y., 1999. Method of producing con-
jugated linoleic acids, US Patent 5,856,149.
Chem. 48, 3072Á3076.
/