ACTA CHROMATOGRAPHICA, NO.

18, 2007

IMPROVED SAPONIFICATION THEN MILD BASE

AND ACID-CATALYZED METHYLATION
IS A USEFUL METHOD FOR QUANTIFYING
FATTY ACIDS, WITH SPECIAL EMPHASIS
ON CONJUGATED DIENES

M. Czauderna*, J. Kowalczyk, K. Korniluk, and I. Wąsowska

The Kielanowski Institute of Animal Physiology and Nutrition, Polish Academy
of Sciences, 05-110 Jabłonna, Poland

SUMMARY
The objective of this study was to evaluate mild lipid saponification
then gentle base and acid-catalyzed methylation, in succession, at 80°C,
60°C, 40°C, or ambient temperature for quantification of fatty acids (FAs),
with special emphasis on conjugated linoleic acid (CLA) isomers. Methy-
lation at 80°C resulted in a substantial increase in the amount of trans,trans
(t,t) CLA isomers and in loss of cis,trans/trans,cis CLA isomers, as a re-
sult of intra-isomerization and formation of artefacts. Methylation at 40°C,
in contrast, seems to enable the most accurate quantification of CLA iso-
mers, because it resulted in no noticeable intra-isomerization of conjuga-
ted dienes or formation of artefacts. Under these derivatization conditions,
other FA methyl esters (FAMEs) also seem to be accurately quantified. The
proposed procedure adequately prepares FAMEs from FA standards and
from lipids in biological samples, because mild saponification and methy-
lation at 40°C using typical basic and acidic catalysts and subsequent ex-
traction with a smaller volume of heptane resulted in satisfactory accuracy
of quantification, negligible changes in the composition of FAs, especially
conjugated dienes, and a better yield of FAMEs. The proposed improved
procedure, comprising mild lipid saponification and methylation at 40°C
in solutions carefully flushed with argon (Ar) (derivatization–Ar), in the
absence of 2,6-di-tert-butyl-p-cresol, then extraction with 4 mL heptane,
seemed the most satisfactory method of preparing FAs, particularly CLA
isomers, for chromatographic quantification by argentation–liquid chro-
matography, the method of choice for analysis of fatty acids containing
conjugated double bonds.

- 59 -
INTRODUCTION
Recent efforts to increase the essential fatty acid content of food-
stuffs has led to renewed interest in analytical methods for accurate quan-
tification of mono and polyunsaturated fatty acids, particularly conjugated
linoleic acid (CLA) isomers. No single chromatographic technique is ca-
pable of fractionating the approximately 400 fatty acids (FAs) present in
biological materials, e.g. meat, fat, bovine milk, or rumen fluid lipids. The
suitability of long capillary columns (100 m) for gas–liquid chromatogra-
phy (GLC) has been improved by use of pre-fractionation by argentation–
liquid chromatography (Ag+–HPLC) [1,2]. Quantitative preparation of fatty
acid methyl esters from the complex mixtures of FAs occurring in meat,
internal organs, milk, or the fat derived from monogastric animals and ru-
minants is unfortunately, difficult. Sodium methoxide-catalyzed methyla-
tion has been used, but N-acyl lipids (i.e. glycosphingolipids or sphingoli-
pids) and free FAs are not derivatized by this reaction [3]. Acid-catalyzed
derivatization, in contrast, methylates all types of lipid. Unfortunately, nu-
merous studies have shown that FAs with conjugated double bonds (CFAs;
e.g. CLA isomers and their metabolites) are isomerized or intraisomerized
during derivatization [3–5]. The presence of oxidizing or antioxidant spe-
cies (for example tocopherols, pyrogallol, or 2,6-di-tert-butyl-p-cresol [2,
6]) results in significantly different profiles, particularly for unsaturated
fatty acids, and formation of unidentified artefacts. Polyunsaturated fatty
acids (PUFA) are especially sensitive to oxidation compared with mono-
unsaturated (MUFA) or saturated fatty acids (SFA) in processed biologi-
cal samples. Numerous studies have revealed that the profile of CFAs is
markedly changed by use of elevated temperatures and also depends on
the type of catalyst used to produce the methyl esters; for example, cataly-
sis by p-toluenesulfinic acid and iodine resulted in migration of the double
bonds and formation of eight geometric isomers of the 8,10, 9,11, 10,12,
and 11,13-octadecadienoic acid methyl derivatives [1]. Boron trifluoride
(BF3) or trimethylsilyldiazomethane methylation, in contrast, result in ex-
tensive formation of trans,trans conjugated dienes and methoxy artefacts
[3,5,7]. Liquid chromatographic analysis of CLA isomers, their metabolites,
or other FAs containing conjugated double bonds has, therefore, usually
been conducted on lipids mildly hydrolyzed to free fatty acids. Later, base
and acid-catalyzed methylation was introduced for preparation of esters of
all these FAs. The best compromise between assay accuracy and derivati-
zation yield for FA assays, particularly for conjugated dienes, would, ne-

- 60 -
vertheless, be a procedure comprising mild saponification then gentle pre-
column derivatization.
For these reasons it seemed desirable to develop a new method of
sample preparation for gas chromatography of fatty acids that would be
based on gentle saponification and esterification by procedures which did
not cause isomerization or intra-isomerization of conjugated dienes, or for-
mation of artefacts. The main objective of our study was, therefore, to in-
vestigate the effect of temperature on the profile of mono and polyunsatu-
rated fatty acids, particularly CLA isomers, subjected to methylation cata-
lyzed by commonly used basic and acidic catalysts. Our second objective
was to examine the profile of fatty acids, particularly CLA isomers, methy-
lated under optimum temperature conditions in solutions carefully flushed
with argon (Ar) (derivatization–Ar) and in solutions containing 2,6-di-tert-
butyl-p-cresol (BHT) (derivatization–BHT) [2]. The modified and original
procedures were compared using fatty acid standards and samples obtained
from laboratory rats fed a diet enriched with a 2% mixture of CLA iso-
mers.

EXPERIMENTAL
Chemicals and Materials
Heptane and acetonitrile were HPLC grade; other reagents were of
analytical grade. A mixture of free CLA isomers (95–97%) was supplied
by Larodan Fine Chemicals (Sweden). Acetonitrile (99.9%) and n-heptane
(95%) were purchased from Lab-Scan (Eire) and other FA standards and
2,6-di-tert-butyl-p-cresol were from Sigma (USA).
Kidneys and livers from rats fed a diet containing 2% of a CLA
isomer mixture were frozen, lyophilized, powdered, and the samples ob-
tained were stored at −20°C until FA analysis [8]. All FAMEs were pre-
pared by derivatization of lipids and fatty acid standards in 20-mL tubes
equipped with Teflon-lined screw-caps. All methylated FA solutions were
protected from light.
Saponification and Extraction of Fatty Acids
Lyophilized samples of kidney and liver (~50 mg) were mixed with
2 mL 2 M aqueous KOH, 2 mL 1 M methanolic KOH, and 50 µL internal
standard solution (17 mg mL−1 nonadecanoic acid in chloroform). The
resulting mixture was carefully flushed with a stream of argon (Ar) for 4–

- 61 -
5 min. Finally, the solution obtained, in a tightly closed tube, was vigoro-
usly mixed, heated at 92–95°C for 10 min, cooled for 10 min at room
temperature, then sonicated for 10 min. The solution obtained (under Ar)
was protected from light and stored overnight in a sealed vial at room
temperature.
Water (3 mL) was added to the hydrolysate, in a vial. After vigo-
rous mixing the solution obtained was acidified to pH 1–2 with 4 M HCl
and free FAs were extracted with dichloromethane (DCM; 4 × 3 mL). The
lower, DCM, layer was dried with ~0.1 g Na2SO4. To avoid any loss of
free FAs, the extraction was repeated with heptane (4 × 3 mL), the upper
heptane layers were combined with the DCM extract, and the organic
solvents were removed under a stream of Ar. The residue, I, was stored at
−20°C until base and acid-catalyzed methylation, or re-dissolved in 1 mL
DCM and 20–30 µL of this solution was injected on to an ion-exchange
column loaded with silver ions (analysis of non-methylated CFA(s) by
Ag+–HPLC).
Separation of Free CLA Isomers and their Metabolites by Ag+–HPLC
Residue I, or 7.2 mg of CLA isomer standard, was re-dissolved in
1 mL DCM and 20–30 µL of the solution was injected on to an HPLC
silver-ion column. Direct determination of non-methylated CLA isomers
and their metabolites (i.e. all fatty acids containing conjugated double bonds)
was performed by use of two analytical ion-exchange columns loaded with
silver ions (250 mm × 4.6 mm ChromSpher 5-µm Lipids column; Chrom-
pack, The Netherlands), connected in series, with photodiode-array detec-
tion (DAD) at 234 nm (Waters Model 996 photodiode-array detector). Elu-
tion was performed as described by Czauderna et al. [9].
Preparation of Fatty Acid Methyl Esters (FAMEs) in the Presence
of BHT (Derivatization–BHT)
NaOH in methanol (2 M, 2 mL) and a solution of BHT in methanol
(20 mg mL−1, 50 µL) were added to residue I, in a tube. The solution was
flushed with a stream of Ar for 3 min then reacted for 1 h at 80°C, 60°C,
or 40°C, or for 10–12 h at ambient temperature (22–24°C). After cooling
(5–10°C) 2 mL 25% BF3 in methanol was added, and the reaction mixture
was flushed with a stream of Ar for 3 min and again heated for 1 h at 80°C,
60°C, or 40°C, or for 10–12 h at ambient temperature (22–24°C). Water
(5 mL) was then added to the cooled reaction mixture and the FAMEs

- 62 -
were extracted with 5 mL heptane. The clear supernatant was transferred
to a vial.
Preparation of Fatty Acid Methyl Esters (FAMEs) in Solutions Flushed
with a Stream of Ar (Derivatization–Ar)
NaOH in methanol (2 M, 2 mL) was added to residue I, to the mi-
xed CLA isomer standard (7.2 mg), or to standards of other fatty acids (0.63
mg). The resulting solution was carefully de-aerated by use of a stream of
Ar for 4–5 min and then reacted for 1 h at 40°C. After cooling, 2 mL 25 %
BF3 in methanol was added to the reaction mixture. This was carefully flu-
shed with a stream of Ar for 4–5 min and again heated for 1 h at 40°C. Wa-
ter (5 mL) was then added to the cooled reaction mixture and the FAMEs
were extracted with 5 mL n-heptane. The clear supernatant was transferred
to a vial.
Fractionation of FAMEs by Gas Liquid-Chromatography (GLC) and
by Ag+–HPLC
Separation of all FAMEs was performed as described elsewhere
[2,8] by use of an Agilent 6890N GC equipped with a CP7489 fused silica
capillary column (100 m × 0.25 mm i.d. × 0.2 µm film thickness; Varian,
USA) and a flame-ionization detector (FID). Analysis of methylated CLA
isomers and of their metabolites with conjugated double bonds was also
performed with the method of choice for conjugated fatty acids, as descri-
bed by Czauderna et al. [2]. Briefly, direct determination of methylated
CFAs (i.e. CLA isomers and their metabolites) was performed with two
analytical ion-exchange columns loaded with silver ions (250 mm × 4.6 mm
i.d. ChromSpher 5 µm Lipids column; Chrompack, The Netherlands), con-
nected in series, in conjunction with photodiode-array detection at 234 nm
(Waters Model 996 photodiode-array detector). Elution was performed as
described elsewhere by Czauderna et al. [2]. UV spectra of the eluate (spec-
tral resolution 1.2 nm) were acquired every second and were electronically
stored on a computer hard disk [2].

RESULTS AND DISCUSSION

Analysis of Free and Methylated Conjugated Dienes by Ag+–HPLC
To investigate the tendency of conjugated fatty acid (CFA) isomers
to intra-isomerise, we assumed that fractionation of non-methylated CFAs

- 63 -
by direct Ag+–HPLC analysis [9] resulted in complete recovery of all iso-
mers and gave a true picture of the CLA isomer composition. In this stu-
dy, therefore, we first attempted to fractionate the free CLA isomers formed
during mild saponification of lipids from the kidneys and livers of rats fed
a CLA isomer-enriched diet. As expected (Table I), liquid chromatogra-
phy using two Ag+ columns, connected in series, with photodiode-array
detection at 234 nm [2,9], enabled highly discerning separation of free and
methylated trans–trans (t,t), cis,trans/trans,cis (c,t/t,c), and cis,cis (c,c)
CLA isomers in the Larodan standard CLA isomer mixture, and in extracts
of the kidneys and livers of rats fed a CLA isomer-enriched diet.
It is clear from the data summarized in Table I that the t,t and c,c
isomers were of minor abundance in the CLA isomer pool whereas
c9t10CLA and t10c12CLA (i.e. c,t/t,c isomers) were the most abundant
isomer group in all the samples assayed. The results presented in Table I
also reveal that for the CLA isomer standard and the kidneys and livers of
rats our mild saponification and base and acid-catalyzed methylation at
40°C resulted in no observable intra-isomerization of the original CLA iso-
mers and/or formation of artefacts containing conjugated double bonds.
Although methylation at 40°C resulted in a lower derivatization yield than
methylation at 60°C and 80°C, the sum of CLA isomer peak areas was
largest when derivatization was performed at 40°C. These results indicate,
therefore, that methylation at 40°C resulted in a smaller loss of CLA isomers
by conversion into non-CLA fatty acid methyl esters and/or unidentified
artefacts. When base and acid-catalysed methylation of the Larodan stan-
dard CLA isomers was conducted at 40°C the percentage distribution of
t,t, c,t/t,c, and c,c CLA isomers was, moreover, almost equal to their dis-
tribution when analyzed directly (as non-methylated CFA) by silver-ion
liquid chromatography (Ag+–HPLC) [9-11]. Larger amounts of methylated
c9t11CLA were found when samples were analysed by GLC than when
they were analysed by Ag+–HPLC, because the resolution of capillary
GLC [2] was poorer than that of Ag+–HPLC performed with two silver-
ion-loaded columns connected in series [2,8,9]. Indeed, the close proximi-
ty of numerous c,t/t,c CLA isomers [9] resulted in overlapping of the
c9t11CLA peak with some c,t/t,c isomer peaks, and, therefore, in GLC
chromatograms the c9t11CLA peak is not symmetrical. Our Ag+–HPLC
system, in contrast, resolved all the geometrical and positional isomers of
CLA (c11t13/t11c13, c10t12/t10c12, c9t11/t9c11, and c8t10/t8c10) very
efficiently [2,8,9]. The c9t11CLA peak is, therefore, symmetrical, because
this isomer did not interfere with other c,t/t,c CLA isomers. Consequently,

- 64 -
the amount of c9t11CLA measured by use of Ag+–HPLC is less than that
measured by use of GLC.

Table I
CLA isomer composition (%) of the Larodan CLA isomer standard mixturea and of
extracts of the kidneysb and livers of rats, by use of base and acid-catalyzed methylation
at different temperatures. Separations were performed by gas–liquid chromatography and
argentation–chromatography (Ag+–HPLC)
Sum of Sum
Yield, t,t c9t11 t10c12 c,t/t,c c,c of peak
Analytical Methylation
Sample from GLC CLAd CLA CLA CLA CLA areas
method temperature
assayc (%) (%) (%) (%) isomers (%) of CLA
(%) isomers
Non-methyla-
HPLCe – 5.8 44.6 33.1 91.0 3.3 –
ted FAse
Kidney 80°C 82.5 44.5 31.9 22.0 53.9 1.6 –
extract 60°C 72.2 11.0 52.2 32.7 85.6 3.4 –
GLC
40°C 62.1 7.0 56.3 31.5 89.9 3.1 –
ambient, 23°C 27.6 6.4 55.8 32.1 90.5 3.1 –
Non-methyla-
HPLCe – 5.7 43.3 32.3 92.8 1.5 –
Liver ted FAse
extract 80°C 92.3 34.3 38.7 26.2 64.7 1.0 –
GLC
40°C 50.7 5.9 50.5 33.2 92.7 1.4 –
Non-methyla-
HPLCe – 19.0 36.6 37.1 77.6 3.4 –
ted FAse
80°C 74.8 56.7 17.1 14.3 37.1 6.2 7457
Larodan
60°C 70.6 21.4 37.1 35.8 74.2 4.4 6790
CLA GLC
isomer 40°C 64.7 18.7 38.6 38.9 77.5 3.9 8605
standard (18.8)f (38.4)f (37.7)f (77.5)f (3.8)f
mixture Ag+–HPLC 80°C – 64.2 14.7 11.5 32.6 3.2 38.6 × 106
analysis of
methylated 40°C – 19.7 36.1 39.0 77.2 3.1 49.5 × 106
CLA isomers (20.1)f (35.4)f (37.3)f (76.7)f (3.1)f
a
8.39 mg free CLA isomer mixture was methylated with 2 mL 2 M NaOH in methanol
then 2 mL 25% BF3 in methanol
b
Average composition (%) of CLA isomers in kidneys obtained by saponification and
extraction; average of four results (saponification then methylation at 80, 40 or 23°C) or
three results (saponification then methylation at 60°C) for four rats fed a diet enriched
in 2% CLA isomer mixture [7]
c
Result from GLC assay based on the internal standard (C19:0) (comprising yield from
methylation and two extractions, after saponification and methylation)
d
t,t and c,c geometric forms of CLA isomers – trans,trans and cis,cis, respectively
e
Direct Ag+–HPLC analysis of non-methylated CFAs (i.e. CLA isomers and other fatty
acids containing conjugated double bonds) in kidneys (from four rats) and liver (from
one rat) and in the mixture of free CLA isomers supplied by Larodan Fine Chemicals
f
Results in parentheses: base and BF3-catalyzed methylation was performed for 75 min at
40°C

- 65 -
Analysis of Fatty Acids (FAs) Using Base and Acid-Catalyzed Methy-
lation at Different Temperatures
After several recent investigations it has been reported that poly-
unsaturated fatty acid composition, particularly conjugated linoleic acid
isomer content, was greatly affected by the catalysts used for formation of
methyl esters. Our studies (data not presented) and several others [2,3,7,8,
12,13] revealed that NaOH-catalyzed methylation for 1 h at 80°C resulted
in the formation of, predominantly, c9t11CLA-MEs with only a small
concentration of artefacts and/or isomerization products of the original
c9t11CLA in analysis of processed milk samples whereas acid-catalyzed
methylation using HCl–methanol or BF3–methanol for 1 h at 80°C resulted
in reduction in the amounts of c9t11CLA and t10c12CLA and formation
of additional positional and geometric CLA isomers, especially t,t CLA
isomers. Our results from combined methylation, summarized in Table I,
are consistent with those from studies on single methylation cited above
and reveal that combination of NaOH–methanol and BF3–methanol-cata-
lyzed methylation at 80°C also resulted in reduced amounts of c,t/t,c CLA
isomers and significantly increased amounts of t,t CLA isomers as major
intra-isomerization products of conjugated dienes. Formation of additional
t,t CLA isomers and artefacts was, fortunately, less during base and acid-
catalyzed methylation at lower temperatures. Formation of t,t CLA isomers
and artefacts during base and acid-catalyzed methylation decreased as the
temperature used for methylation was reduced. The profile of the Larodan
CLA isomer standard methylated at 40°C for 1 h and that of non-methy-
lated CLA isomers analyzed directly by Ag+–HPLC were, consequently,
similar [9]. As is apparent from the data in Table I, there were no signifi-
cant differences between the composition of the Larodan CLA isomer stan-
dard and no increase in the derivatization yield when base and acid-cata-
lyzed methylations at 40°C were also performed for longer (75 min rather
than 1 h). Under these gentle basic and acidic methylation conditions, fur-
thermore, the composition of methylated CLA isomers was equivalent to
that for non-methylated CLA isomers analysed directly by Ag+–HPLC [9].
The data presented here indicate, therefore, that combination of our mild
saponification procedure with base and acid-catalyzed methylation at 40°C
for 1 h seems to be the most suitable method for quantification of CFAs
(i.e. CLA isomers and/or their metabolites) by GLC and Ag+–HPLC [2].
The results in Table II show that mild saponification followed by
methylation at 60°C, 40°C, or ambient temperature are appropriate analy-
tical procedures for GLC analysis of saturated, monounsaturated, and non-

- 66 -
CLA polyunsaturated fatty acids in kidney and liver tissue (data for liver
are not presented), because the relative amounts of these FAs did not dif-
fer among methylations at all the temperatures examined and the control
derivatization performed at the temperature typically used (80°C). As ex-
pected, the most notable differences between results from derivatization at
different temperatures were for CLA isomers only (Table II).

Table II
Comparison of fatty acid methylation performed at different temperatures (T). The fatty
acid composition of the kidneys of rats was determined by gas–liquid chromatography
(GLC)

Ambient
T = 80°C T = 60°C T = 40°C
Fatty acida temperature
(yield, 80.3%)b (yield, 75.8%)b (yield, 59.1%)b (yield, 25.8%)b
C16:0 00.742 00.735 00.754 00.737
c9C16:1 20.74 20.05 20.90 19.94
C18:0 01.001 00.996 01.013 01.001
c9C18:1 01.502 01.426 01.501 01.445
c11C18:1 08.28 07.68 07.95 05.93
c9c12C18:2 00.984 00.966 00.984 00.970
c9c12c15C18:3 07.49 07.27 07.33 07.18
c9t11CLA 16.8 07.45 06.97 06.88
t10c12CLA 23.79 12.66 11.72 11.73
t,tCLA 10.13 39.08 45.88 33.94
c5c8c11c14C20:4 01.165 01.117 01.138 01.137
c5c8c11c14c17C20:5 09.90 11.79 12.62 12.66
c7c10c13c16c19C22:5 21.08 19.90 20.01 20.19
c4c7c10c13c16c19C22:5 10.80 09.99 09.82 10.09
a
Fatty acids (FAs) represented as the ratio of the fatty acid peak area to that of the internal
standard (C19:0) (i.e. SFA/SC19:0); c and t denote cis and trans, respectively
b
Yield of FAMEs based on the internal standard (C19:0) [2]

Mild saponification of lipids and methylation at 40°C is, therefore,

likely to lead to most accurate quantification of conjugated fatty acids
(CFAs) in the presence of other FAs. Methylation at 40°C did, however,
lead to a lower derivatization yield compared with methylation at higher
temperatures (Table II). This disadvantage of base and acid-catalyzed me-
thylation at 40°C was overcome by using a smaller volume of heptane for
extraction of the methylated fatty acids (FAMEs). Yields of methylated fat-
ty acids extracted with 4 and 5 mL heptane are compared in Table III. Higher

- 67 -
concentrations of all the FA standards assayed were obtained when extrac-
tion was performed with 4 mL heptane rather than 5 mL. As is apparent
from the results in Table III, moreover, the extraction yield was the same for
saturated and all mono and polyunsaturated fatty acids. Mild saponification
and methylation at 40°C in combination with extraction with 4 mL hepta-
ne can be regarded as the most suitable method of sample preparation for
determination of FAs as methyl esters.

Table III
Dependence of the peak areas (Sn) of the fatty acids assayed on the volume of heptane used
for extraction of FAMEs (base and acid-catalyzed methylation were performed at 40°C)

5 mL 4 mL Sn (4 mL)/Sn
Fatty acid
heptane heptane (5 mL)
c9C18:1 263 359 1.363
C19:0 683 948 1.388
c9c12C18:2 175 241 1.376
c9t11CLA 5351 7144 1.335
t10c12CLA 5563 7173 1.290
t,tCLA 2563 3465 1.349
Sum of CLA isomers 13754 18471 1.343
c5c8c11c14C20:4 863 1164 1.349
c5c8c11c14c17C20:5 76 103 1.346

The Effect of BHT on the Profile of Methylated Fatty Acid Standards

Formation of unidentified artefacts during base and acid-catalyzed
methylation resulted in inaccurate FA determination. The presence of anti-
oxidants (e.g. tocopherols, pyrogallol, or BHT [2,6]) or elimination of oxi-
dation-promoting species significantly improved the accuracy of the
chromatographic quantification of FAs, particularly for CLA isomers and
other long-chain PUFA. The CLA isomer mixture, other unsaturated fatty
acid standards, and rat liver samples were used to evaluate the effect on
the profile of the FAMEs of using Ar to remove air from solutions of the
methylated FAs (derivatization–Ar). The profiles obtained were compared
with those obtained after FA methylation in the presence of BHT (deriva-
tization–BHT) (Table IV).
To avoid compromising accuracy in the quantification of unsatura-
ted fatty acids, particularly CLA isomers, all methylation was performed
at the optimum derivatization temperature, 40°C. Detailed analysis of CLA

- 68 -
Table IV
Comparison of fatty acid methylation performed in the presence of 2,6-di-tert-butyl-p-
cresol (BHT) (derivatization–BHT) and in the absence of BHT (derivatization–Ar)

Processed fatty acid standardsa (average peak area ± SD)

Derivatization–BHT Derivatization–Ar
Base–acid Base–acid
Fully processedb Fully processedb
methylationc methylationc
Ag+–HPLC 45 × 105 ± 3 × 105 62 × 105 ± 5 × 105 46 × 105 ± 1 × 105 63 × 105 ± 1 × 105
t,tCLA
GLC 971 ± 30 1371 ± 1 946 ± 3 1369 ± 1
Ag+–HPLC 80 × 105 ± 5 × 105 113 × 105 ± 4 × 105 82 × 105 ± 2 × 105 113 × 105 ± 1 × 105
c9t11CLA
GLC 1968 ± 50 2716 ± 78 1949 ± 2 2802 ± 5
Ag+–HPLC 85 × 105 ± 5 × 105 20 × 105 ± 5 × 105 86 × 105 ± 3 × 105 19 × 105 ± 1 × 105
t10c12CLA
GLC 1894 ± 43 2723 ± 44 1871 ± 3 2721 ± 13
Sum of all Ag+–HPLC 176 × 105 ± 10 × 105 246 × 105 ± 1 × 105 179 × 105 ± 6 × 105 244 × 105 ± 1 × 105
c,t/t,c CLA GLC 3952 ± 101 5544 ± 26 3915 ± 1 5676 ± 56
Ag+–HPLC 10 × 105 ± 1 × 105 13 × 105 ± 3 × 105 10 × 105 ± 1 × 105 11 × 105 ± 1 × 105
c,c CLA
GLC 208 ± 10 290 ± 3 200 ± 1 288 ± 1
c9C18:1 GLC 617 ± 5 1019 ± 1 614 ± 1 1021 ± 1
c9c12C18:2 GLC 568 ± 4 949 ± 1 564 ± 3 952 ± 1
c9c12c15C18:3 GLC 460 ± 4 777 ± 1 458 ± 4 781 ± 1
c6c9c12C18:3 GLC 547 ± 4 918 ± 1 544 ± 5 927 ± 1
Saponification and base and acid-catalyzed methylation of fatty acids in rat liversd (µg g−1 DM)
Base-catalyzed Acid-catalyzed
C15:0 GLC 58 ± 5 57 ± 2
C16:0 GLC 4247 ± 59 4148 ± 102
c9C16:1 GLC 150 ± 6 142 ± 9
t11C18:1 GLC 18 ± 2 19 ± 2
c6C18:1 GLC 41 ± 4 40 ± 2
c9C18:1 GLC 920 ± 11 981 ± 93
c11C18:1 GLC 368 ± 6 369 ± 1
c9c12C18:2 GLC 3928 ± 53 3808 ± 44
C20:0 GLC 18 ± 1 18 ± 1
c9c12c15C18:3 GLC 610 ± 8 597 ± 6
c9t11CLA GLC 621 ± 14 653 ± 39
t10c12CLA GLC 443 ± 1 472 ± 46
c,cCLA GLC 50 ± 9 64 ± 7
t,tCLA GLC 194 ± 18 173 ± 1
c13C22:1 GLC 7±1 5±2
c11c14c17C20:3 GLC 50 ± 3 55 ± 6
c5c8c11c14C20:4 GLC 4229 ± 55 4109 ± 20
c5c8c11c14c17C20:5 GLC 1670 ± 5 1644 ± 18
c7c10c13c16c19C22:5 GLC 701 ± 20 686 ± 1
c4c7c10c13c16c19cC22:6 GLC 2889 ± 76 2845 ± 4
a
Each result is the mean from methylation of two FA standard samples; SD denotes
standard deviation and DM dry mass
b
The procedure comprised saponification and base–acid catalyzed methylation at 40°C
c
The procedure comprised solely base–acid catalyzed methylation at 40°C
d
Each result is the mean FA concentration in the liver of rats fed a diet enriched with 2%
CLA isomer mixture

- 69 -
isomers and other PUFA and MUFA (Table IV) revealed that derivatiza-
tion–Ar resulted in almost the same results for FAMEs as derivatization–
BHT. The results obtained from derivatization–Ar reveal this is a universal
procedure for accurate quantification of FAs in biological samples. The
proposed mild saponification and gentle base and acid-catalyzed methyla-
tion at 40°C, without use of BHT, (derivatization–Ar) caused no isomeri-
zation and produced no artefacts and can be recommended, particularly for
analysis of conjugated FAs (e.g. CLA isomers, their catabolites, and higher
metabolites), as FAMEs, using argentation–liquid chromatography (Ag+–
HPLC with DAD) then determination of other FAs by GLC–FID. The
absence of BHT in processed FAMEs samples enabled accurate determi-
nation of all conjugated FAs by use of Ag+–HPLC with UV detection at
232–234 nm without problems caused by overlapping of the peaks of as-
sayed conjugated FAs with the broad peak of BHT, which contains an
aromatic ring. In contrast, the peak of BHT added to processed samples
(derivatization–BHT) overlapped some of the peaks of conjugated FAs
(especially c,c and c,t/t,c CLA isomers and some CLA metabolites) when
Ag+–HPLC–DAD was used. The very large BHT peak also interfered in
the region of the short-chain fatty acids (especially C14:0 and C14:1) in
our GLC chromatograms (GLC–FID) [2,6].

CONCLUSION
Combination of mild saponification and gentle base and acid-ca-
talyzed methylation at 40°C, without addition of BHT (derivatization–Ar),
in combination with extraction with 4 mL heptane, is the best compromise
between accuracy and the yield in methylation of FAs in the quantification
of FAs, particularly conjugated dienes. The proposed procedure adequate-
ly prepared FAMEs from FA standards and lipids in biological samples,
because the mild saponification and the basic and acidic catalysts typically
used for methylation at 40°C without BHT (derivatization–Ar) resulted in
negligible alteration of the FA composition, especially CLA isomers, and
enabled simultaneous determination of the complex mixture of methylated
conjugated FAs in the presence of other FAs by use of argentation–liquid
chromatography (Ag+–HPLC) and DAD. Argentation–liquid chromatogra-
phy in conjunction with UV detection at 232–234 nm is the method of choice
for analysis of fatty acids containing conjugated double bonds (e.g. CLA
isomers and their catabolites and higher metabolites) [2,8,9]. Other methy-
lated FAs, and the methylated conjugated FAs in samples processed by de-

- 70 -
rivatization–Ar, can subsequently be accurately quantified by use of high-
resolution long-capillary column GLC–FID [2].

REFERENCES
[1] R.F. Cross, E. Ostrowska, M. Muralitharan, and F.R. Dunshea,
J. High. Resolut. Chromatogr., 23, 317 (2000)
[2] M. Czauderna, J. Kowalczyk, K. Korniluk, and I. Wąsowska,
J. Anim. Feed Sci., 14 Suppl. 1, 563 (2005)
[3] J.K.G. Kramer, V. Fellne, M.E.R. Dugan, F.D. Saue, M.M. Mossoba,
and M.P. Yurawecz, Lipids, 32, 1219 (1997)
[4] N. Aldai, B.E. Murray, A.I. Najera, D.J. Troy, and K. Osoro,
J. Sci. Food Agric., 85, 1073 (2005)
[5] L. Yu, D. Adams, and B.A. Watkins, J. Food Comp. Anal.,
16, 419 (2003)
[6] N. Aldai, K. Osoro, L.J.R. Barrón, and A.I. Nájera, J. Chromatogr. A,
1110, 133 (2006)
[7] M. Igarashi, T. Tsuzuki, T. Kambe, and T. Miyazawa, J. Nutr. Sci.
Vitaminol. (Tokyo), 50, 121 (2004)
[8] K. Korniluk, M. Czauderna, J. Kowalczyk, A. Mieczkowska,
M. Taciak, and L. Leng, J. Anim. Feed Sci., 15, 31 (2006)
[9] M. Czauderna, J. Kowalczyk, I. Wąsowska,
and K.M. Niedźwiedzka, J. Anim. Feed Sci., 12, 269 (2003)
[10] R.F. Cross and H. Zackari, J. Sep. Sci., 26, 480 (2003)
[11] E. Ostrowska, F.R. Dunshea, M. Muralitharan, and R.F. Cross,
Lipids, 35, 1147 (2000)
[12] M.Y. Yung, G.B. Kim, E.S. Jang, Y.K. Jung, S.Y. Park,
and B.H. Lee, J. Dairy Sci., 89, 90 (2006)
[13] M.Y. Jung and M.O. Jung, J. Agric. Food Chem., 50, 6188 (2002)

- 71 -