Upload
noemi-ruiz-lopez
View
215
Download
2
Embed Size (px)
Citation preview
Sequential one-step extraction and analysis of triacylglycerolsand fatty acids in plant tissues
Noem�ıı Ruiz-L�oopez, Enrique Mart�ıınez-Force, and Rafael Garc�ees*
Instituto de la Grasa, Consejo Superior de Investigaciones Cient�ııficas, E-41012 Seville, Spain
Received 27 December 2002
Abstract
A method for plant tissue digestion and triacylglycerol (TAG) extraction followed by transmethylation of TAGs to produce the
fatty acid methyl esters (FAMEs) from small storage tissue samples is presented. The method allows the analysis of both TAGs and
FAMEs from the same sample. Several reagent mixtures and different experimental conditions were tested on sliced sunflower seeds.
The best results were obtained using a mixture that was 33.3% a solution of NaCl (0.17M) in methanol and 66.6% heptane by volume.
The TAGs in the heptane solution were transmethylated with a mixture containing methanol:toluene:dimethoxypropane:H4SO2
(39:20:5:2, by vol). The method was also tested on other oil seed storage tissue (soybean) and fruit tissues from olive and acorn. In all
cases, sunflower, soybean, olive, and acorn, the TAGs and FAMEs composition data obtained by this method were quite similar to
data from a standard analysis method. In samples with high protein content, such as soybean and sunflower seeds, the TAG
extraction was incomplete. The water content of fruit samples did not interfere with TAG extraction obtained by this method.
� 2003 Elsevier Science (USA). All rights reserved.
Keywords: Triacylglycerols; Fatty acids; Capillary gas chromatography; Extraction
The chemical, physical, and nutritional properties of
oils are affected by triacylglycerol (TAG)1 composition
and their stereospecificity (position of the fatty acids (FA)
on the glycerol backbone). Nevertheless the levels of mi-
nor constituents are also important, for example in olive
oil [1,2]. TAG composition gives more information about
a particular oil (like a fingerprint). Current methods of oilextraction and analysis are time-consuming and imprac-
tical for processing a high number of samples. Thus, it
would be very useful to develop a method that allowed a
direct extraction of lipids from a small piece of seed, in
order to analyze TAG composition and to transmethylate
FAs later. This method would allow the selection of in-
dividual oleaginous seeds from a wide collection on the
basis of their TAG composition or to characterize theTAG composition of a high number of samples.
Several methods have been used to extract oil from
seed samples. The AOAC-accepted reflux–extraction
procedures, such as Goldfisch and Soxhlet, are not
suitable for a large number of small samples. Other
methods of total lipid extraction could be used for a
relatively high number of samples, but these methods
involve several steps. Folch et al. [4] devised a methodfor total lipid extraction using a solution of chloro-
form:methanol (2:1, v/v) that was originally used for the
extraction of brain tissue lipids. Some modifications
were developed by Wren et al. [5], Dawson et al. [6], and
Privett et al. [7], making the method more suitable for
specific tissues. Hara and Radin [8] introduced a new
method with no toxic solvents using hexane:isopropanol
(3:2, v/v). These multistep methods could be used forsmall samples, but not for a large number of them.
Reske et al. [9] used a method in which approximately
100 sunflower seeds were ground with hexane in a 1:1
(w/v) ratio and stood overnight. Chaven et al. [10]
proposed a relatively simple microanalytical technique
for isolating the neutral lipid fraction of soybean lipids
and analyzing their FA composition. The microanalyt-
ical procedure was suitable for small samples but like the
Analytical Biochemistry 317 (2003) 247–254
www.elsevier.com/locate/yabio
ANALYTICAL
BIOCHEMISTRY
* Corresponding author. Fax: +34-954-616790.
E-mail address: [email protected] (R. Garc�ees).1 Abbreviations used: BHT, butylated hydroxytoluene; DMP, 2,2-
dimethoxypropane; FA, fatty acid; FAME, fatty acid methyl ester; L,
linoleic acid (18:2); MeOH, methanol; HHH, triheptadecanoylglycerol;
O, oleic acid (18:1); P, palmitic acid (16:0); S, stearic acid (18:0); TAG,
triacylglycerol.
0003-2697/03/$ - see front matter � 2003 Elsevier Science (USA). All rights reserved.
doi:10.1016/S0003-2697(03)00139-8
previous techniques still required separate steps forextraction, filtration, etc.
There are several methods for determination of
composition of isolated TAGs: thin-layer chromatogra-
phy (TLC), high-performance liquid chromatography
(HPLC), gas–liquid chromatography (GLC), and super-
critical fluid chromatography. The recommended IU-
PAC method uses isocratic nonaqueous reversed-phase
high-performance liquid chromatography, with refrac-tive index detection, rendering separations based on the
equivalent carbon number. On the other hand,GLC gives
a very high resolution and is able to separate TAGs with
saturated FAs that have poor solubility and long reten-
tion times in HPLC with refractive index detection. GLC
allows the analysis of very complex mixtures of TAGs
with many different molecular weights [11].
The TAG composition can be calculated utilizing fattyacid methyl ester (FAME) determinations and computer
programs, by applying the 1,3-random 2-random distri-
bution theory [12]. For this purpose a lipase digestion is
made and resulting lipids are separated by TLC. The
monoacylglycerol band containing the FAs in position
sn-2 in TAG molecules and the total FA composition in
TAG molecules are used with a computer program, LI-
PASE, to calculate the original TAG composition (http://niobio.grasa.csic.es/emforce/GGBLS/). Nevertheless this
method is suitable only to get a reference of the TAGs that
should be found in any oil sample, because of the asym-
metrical distribution of some FAs in the sn-1 and sn-3
positions of the TAGs [9,11].
With the aim of analyzing the FA composition of
several thousands of individual half-seed samples in
mutagenized sunflower seeds, a method for direct oilextraction and FA transmethylation was developed [13].
This method used a complex reagent mixture that di-
gested seed tissues, extracted the lipid fraction, trans-
methylated the FAs, and separated these esters from the
rest of the components in just one step.
In this work, we describe a new analytical method to
extract oil from seed samples in one step, analyze TAG
composition by GLC, transmethylate the FAs in thesample later, and analyze the FAME composition by
GLC. This allows the analysis of a large number of very
small samples like half-seeds. Because the embryo part
of the seed is not used for the analysis, it could be ger-
minated and the descendants obtained in case the de-
sired phenotype is found in a screening of mutagenized
or segregant seeds.
Materials and methods
Plant materials and growth conditions
Sunflower (Helianthus annuus L.) seeds from the high-
oleic line CAS-9 [11] and the medium stearic line CAS-4
[3] were used in this work. Seeds were collected fromplants grown in a growth chamber at 25/15 �C day/night,
16-h photoperiod, and 300lE �m�2 � s�1 light intensity.
Seeds were immersed in water for 2 h to facilitate peeling
and the achenes (seeds without seed coat) were dried and
cut into pieces. Seed pieces were placed in screw-cap test
tubes with Teflon-lined caps. For the nondestructive
half-seed method, one-third of the cotyledon (opposite
the embryo side) was analyzed to get TAG and FAMEcomposition. The rest of the seed, containing the em-
bryo, could be grown and the progeny collected. Sample
sizes are indicated in each experiment. Olive (Olea
europea) fruits, acorn (Quercus ilex) fruits, and soybean
(Glycine max) seeds were also used to test the method.
Solutions and reagents
1,2,3-Triheptadecanoylglycerol (HHH) and butylated
hydroxytoluene (BHT) were from Sigma–Aldrich Che-
mie, Gmbh (Steinheim, Germany). Hexane, heptane,
isopropanol, methanol, tethahydrofuran, and washed
sea sand were obtained from Panreac Qu�ıımica (Barce-
lona, Spain). 2,2-Dimethoxypropane (DMP) was ob-
tained from Merck (Darmstadt, Germany). Specific
solution mixtures are indicated in each experiment. So-lutions of 0.17M NaCl in methanol and 18.6mM HHH
in heptane (used as internal standard) were prepared
and added to all experiments. BHT at 0.01% (w/v) was
also added to all experiments as an antioxidant.
Oil extraction
Unless indicated, samples were incubated withoutshaking for 1 h in a water bath at 80 �C. Incubation
mixtures for TAG extraction are shown in Table 1.
After cooling, the upper phase containing the lipids was
transferred to a new test tube. For some experiments,
the lipids that remained in the seed pieces after the in-
cubation at 80 �C were extracted using the method of
Hara and Radin [8]. First, seed pieces were washed two
times in heptane to avoid any external lipid contami-nation. As control of total lipid content, seed pieces were
extracted with 50 ll of HHH in heptane as internal
control, see above. Three different replicates were made
of each experiment. TLC was carried out in 0.25-mm-
thick silica gel G-60 plates developed with hexane:di-
ethyl ether (90:10; v/v). To detect oil spots, plates were
sprayed with 50% H2SO4 in water and heated at 200 �Cuntil the spots were visible.
For time-course extraction experiments, different
amounts of seed pieces (200, 50, and 20mg) were incu-
bated with the selected mixture solution in a final vol-
ume of 7.2ml. After 15, 30, 60, 120, and 240min an
aliquot was sampled and analyzed for TAG extraction
with respect to the added internal standard. The seed
pieces remaining in the test tubes after the partial
248 N. Ruiz-L�oopez et al. / Analytical Biochemistry 317 (2003) 247–254
extraction were cleaned two times with heptane and
extracted using the method of Hara and Radin [8], see
above. These time-course experiments were repeated
four times.
TAG analysis by GLC
TAG were separated and quantified by GLC [11] with
an Agilent 6890 gas chromatograph (Palo Alto, CA,
USA), hydrogen was used as carrier gas, injector and
detector temperature was 380 �C, oven temperature was
345 �C, and a gradient of pressure from 70 to 120 kPa
was used, depending on each particular column. The
gas-chromatography capillary column was a DB-17HT
(15m length, 0.25mm i.d., and 0.15 lm film thickness)with a midpolarity liquid phase of (50% phenyl)meth-
ylpolysiloxane from J&W Scientific (Folsom, CA,
USA), the linear gas rate was 50 cm/s, and the split ratio
was 1:80. The detector used was a flame ionization de-
tector. The different TAG molecules were identified with
respect to known samples [11] and the detector response
was corrected [14]. Total lipid extracted in each experi-
ment was calculated with respect to the area of the in-ternal control HHH. For TAG analysis by GLC it is not
necessary to purify the TAG from the other neutral
lipids extracted by this method in the sample because of
their very different retention times.
FAME extraction and analysis
After the analysis of TAG, the remaining heptaneTAG solution was filled to 0.9ml of total solution with
heptane, and another 0.9ml of the transmethylation
mixture containing methanol:toluene:DMP:H2SO4 (39:
20:5:2) was added [13]. The FAMEs were quantified us-
ing a Hewlett–Packard 5890A gas chromatograph (Palo
Alto, CA, USA) with a Supelco SP-2380 (Bellefonte, PA,
USA) capillary column of fused silica (30m length;
0.25mm i.d., and 0.20 lm film thickness). Hydrogen wasused as carrier gas and the linear gas rate was 28 cm/s.
Detector and injector temperatures were 200 �C, oven
temperature was 170 �C, and the split ratio was 1:50.
FAMEs were identified by comparison with known
standards. Fatty acid composition was calculated from
TAG composition assuming that each fatty acid in a
TAG molecule represents one-third of the percentage of
that molecule in the total TAGs.
Oil extraction from other plant tissues
To obtain quantitative TAG extraction data from
olive (O. europea) and acorn (Q. ilex) fruits, 50mg of
fruit pieces was incubated at 80 �C with 7.2ml of the
new method mixture and 16 ll of HHH solution as
internal standard during 2 h. After cooling, the upper
phase containing the lipids was transferred to a newtest tube. The remaining lipids in the fruit pieces were
extracted as indicated above [8], after addition of 8 ll ofHHH solution. The lipidic phases from the extracted
samples were evaporated to dryness under N2. The
residue was dissolved in a small volume of heptane.
To compare qualitative TAG extraction data from
these fruits, 50mg of the samples was treated as indi-
cated above and 50mg of the samples was treated bythe method of Hara and Radin [8] without previous
extraction.
Soybean (G. max) seeds were immersed in water for
1 h. Fifty milligrams from one individual seed was cut
and TAGs were extracted with the new extraction mix-
ture as indicated above for olive and acorn tissues. The
seed pieces remaining in the test tubes after the partial
extraction were cleaned two times with heptane andextracted [8], see above.
Results and discussion
The goal of this work was to develop a procedure for
fast and easy TAG and FAME preparation and analysis
by GLC. In the first step TAGs were extracted andanalyzed by GLC. Then, TAGs were transmethylated.
Isolated FAMEs could also be analyzed by GLC.
Table 1
Solvent mixtures by volume assayed and total TAGs extracted expressed as percentage of the total TAG content (extracted by the Hara and Radin
method)
Mixture Heptane MeOH/NaCla Toluene DMP THF HHHb % Extracted
A 61.1 33.3 — — — 5.6 79:7� 1:1
B 56.1 33.3 5.0 — — 5.6 74:2� 3:2
C 51.1 33.3 10.0 — — 5.6 70:8� 3:5
D 55.1 33.3 5.0 1.0 — 5.6 74:0� 1:5
E 54.1 33.3 5.0 2.0 — 5.6 72:6� 2:5
F 50.1 33.3 10.0 1.0 — 5.6 76:3� 4:6
G 49.1 33.3 10.0 2.0 — 5.6 70:6� 1:2
H 51.1 33.3 — — 10.0 5.6 76:3� 8:7
Extraction time was 2 h at 80 �C. Percentage of extraction is expressed as mean� SD of three replicates.aA solution of 10 g/liter of NaCl in methanol.b Triheptadecanoylglycerol (18.6 mM) in heptane.
N. Ruiz-L�oopez et al. / Analytical Biochemistry 317 (2003) 247–254 249
This method allows the determination of TAG andFA composition of a large number of small samples
(3–5mg) in a short time.
First step—TAG extraction and analysis
In a previous work Garc�ees et al. [13] developed a
method for one-step oil extraction, FAME production,
and analysis. Following the same philosophy, our goalin this work was to perform a quick extraction of the oil
but without the transmethylation step. For this, several
solutions were designed to digest the tissue and extract
the lipids. In order to have complete extraction of lipids,
it is necessary to use organic extractants of high polarity,
usually containing some alcohol. We used solvents
containing methanol, a primary alcohol with the most
active hydroxyl group. Methanol and water stimulateddisruption of hydrogen bonds between lipid carbonyl,
hydroxyl, and amino groups and compounds of the
nonextractable residue [15]. As a nonpolar solvent we
used heptane, which has a higher boiling point than
hexane (the usual solvent). Some other chemical com-
pounds were also added to the mixture in order to test
whether the total lipid extraction was improved. Tolu-
ene or tetrahydrofuran was used, instead of benzene,because of its lower toxicity, to form a unique phase at
80 �C for lipid extraction. The addition of DMP was
also found very useful to help in the extraction of lipids
[13]. An important modification was the change of
H2SO4 to NaCl. This compound was added in order to
have two phases and a good separation of TAGs from
polar compounds at room temperature and to avoid the
production of the FAMEs because of the presence ofmethanol with acid pH. An antioxidant was added to all
mixtures in order to avoid partial oxidation of lipids.
This was especially relevant in the case of samples with a
high proportion of unsaturated FAs. We selected BHT
as antioxidant because it does not affect the polarity of
the mixture.
Table 1 shows the different mixtures we designed and
the ratio of total TAG extraction obtained after 2 h at80 �C with each mixture relative to the complete ex-
traction (100%) obtained using the method of Hara and
Radin [8]. All these mixtures have the property of
forming a single phase at 80 �C, improving extraction of
lipids, and two phases at room temperature with neutral
lipids contained in the upper phase (Fig. 1, lanes A and
B). The seeds used in this particular experiment were
from a high-oleic sunflower line because of its lownumber of TAG peaks in the chromatogram and the
peak of the internal control (HHH) being easily distin-
guishable from the rest. All the mixtures tested gave an
extraction of TAG ranging between 70 and 80%, the
best results being obtained with mixture A, which is also
the simplest. Results obtained were not better after
adding toluene, tetrahydrofuran, or DMP. Additionally,
mixtures containing tetrahydrofuran showed poor sep-
aration of phases at room temperature.
In order to verify that extraction with mixture A does
not produce any transmethylation we carried out an
experiment extracting oil from sunflower seeds. The
lipids were separated by TLC. Fig. 1 shows that the oil
extracted by our method does not produce any FAMEs(line B) but diacylglycerols are extracted (see light-gray
band); we could suggest that only neutral lipids were
extracted. Lane A shows sunflower oil lipids separated
by TLC and lane C is a partial transmethylation of
sunflower oil (see dark-gray band at the top corre-
sponding to FAMEs).
To optimize the extraction, a time-course experiment
with different sample sizes was carried out using mixtureA from 15 to 240min. Samples for these assays were
from medium-stearic sunflower line CAS-4. Total mil-
ligrams extracted from each sample is shown in Fig. 2A.
The amount of total TAG extracted increased up to
120min. In the three cases the behavior was similar,
indicating that 200mg sample did not saturate the
mixture. The amount of lipid extracted in this case was
higher than in the others with 20 and 50mg of seedsamples and was similar from 120 to 240min. Fig. 2B
shows that the extraction of lipids was very efficient,
with more than 50% of TAG extracted after 30min.
After that, the extraction slowed with no important
extraction increase up to 240min. We selected the time
of 120min, which gave near-maximum extraction under
these conditions independent of sample size.
Table 2 shows the TAG composition of the extractedoil using different sample sizes and extraction times.
Interestingly, the TAG composition is very similar in all
cases, with only a few cases exhibiting small differences
between the TAG content of two different points. The
standard deviation shows that no significant difference
could be found between the different extraction times.
Those differences might not be related to the sample size
Fig. 1. TLC plate of sunflower oil (lane A), oil extracted by mixture A
(heptane/MeOH/NaCl) for 2 h at 80 �C (lane B), and partial transme-
thylation of sunflower oil (lane C).
250 N. Ruiz-L�oopez et al. / Analytical Biochemistry 317 (2003) 247–254
or extraction time, but reflect the intrinsic differencesbetween the TAG compositions of different seeds. It is
important to note that the seed samples of 200mg, when
only 30% of TAGs were extracted, have the same TAGmolecular species content as the smallest, 20-mg seed
samples. These observations indicate that this method
gives consistent results with any extraction time and
sample size. Thus, we suggest using a sample size of
around 20 to 50mg and an extraction time of 120min,
leading to a precise TAG composition and efficient ex-
traction.
Finally, we verified the accuracy of the new extractionmethod by analyzing the TAG composition of the two
cotyledons obtained by transversally slicing an individ-
ual seed, having two identical samples and avoiding in
this case the intrinsic seed-to-seed TAG composition
differences. Each cotyledon was analyzed by extracting
the oil with mixture A (Table 1) and/or the control
method [8] and the TAG composition was compared.
Table 3 confirms that there is seed-to-seed variation,comparing the results obtained by either method be-
tween any of the seeds tested. For example, in the two
comparisons of the control method (seeds A and B) the
POP contents for each cotyledon of seed A were 0.48
and 0.51 and in seed B the values were 0.40 and 0.39. In
this case there is around 10% difference between seeds A
and B, but in the case of seeds A and C the POO con-
tents were 5.10 and 5.03 for seed A and 2.74 and 2.65 forseed C much lower than for seed A. Similar results were
obtained with the other TAG and seed method com-
parisons. In any case, it is relevant that the TAG com-
positions determined in both halves of any seed are very
similar, indicating that the new method precisely reflects
the real TAG composition of the samples.
Table 2
Molecular species of CAS-4 line TAG (mol%) found in the upper phase (heptane) using mixture A of Table 1 at 15 (A), 60 (B), and 240min (C) of
incubation and the remaining TAG found in these seeds extracted with the control method (D)
TAG type Molecular TAG speciesa (mol%)
HR 20mg 50mg 200mg
Mean�SD A B C D SD A B C D SD A B C D SD
POP 0:3� 0:0 0.3 0.4 0.3 0.3 0.1 0.4 0.3 0.3 0.3 0.1 0.3 0.3 0.3 0.3 0.0
PLP 0:5� 0:0 0.5 0.6 0.5 0.6 0.1 0.6 0.5 0.5 0.6 0.1 0.5 0.5 0.5 0.5 0.0
POS 1:5� 0:2 1.8 1.5 1.4 1.4 0.3 1.4 1.4 1.3 1.5 0.1 1.6 1.4 1.4 1.5 0.1
POO 2:8� 0:2 3.2 2.9 2.6 2.7 0.3 2.8 2.8 2.6 2.6 0.2 3.0 2.6 2.7 2.7 0.2
PLS 2:5� 0:2 2.9 2.7 2.7 2.9 0.3 2.6 2.6 2.6 2.7 0.1 2.6 2.5 2.5 2.6 0.1
POL 5:3� 0:1 6.9 6.4 5.8 5.8 0.5 5.8 5.8 5.9 5.7 0.2 5.5 5.3 5.5 5.4 0.1
PLL 4:2� 0:3 4.6 4.8 5.1 4.7 0.3 4.4 4.8 4.9 5.0 0.3 4.1 4.2 4.3 4.2 0.3
SOS 1:6� 0:1 1.5 1.5 1.3 1.4 0.3 1.4 1.3 1.2 1.3 0.1 1.6 1.4 1.5 1.6 0.2
SOO 7:2� 0:7 7.9 6.9 6.1 6.5 0.9 6.5 6.3 6.1 6.1 0.6 7.5 6.7 6.7 6.9 0.5
SLS 1:7� 0:2 2.4 2.2 2.0 2.3 0.4 1.8 1.8 2.1 1.8 0.3 1.9 2.7 2.0 1.8 0.7
OOO 7:0� 0:6 7.2 6.6 5.6 6.3 0.7 6.5 6.5 6.0 6.0 0.5 7.4 5.3 6.5 6.5 1.2
SOL 12:8� 0:4 12.0 12.0 11.7 12.4 0.6 12.1 11.9 11.8 11.8 0.4 12.9 12.8 12.4 12.8 0.6
OOL 14:1� 0:7 12.1 12.2 12.2 12.5 0.6 13.1 13.1 12.8 12.7 0.7 13.5 13.5 13.6 13.5 0.8
SLL 11:5� 0:6 10.2 11.3 12.6 13.1 1.1 11.9 11.7 12.1 12.4 0.7 11.2 12.1 11.9 11.9 0.6
OLL 17:2� 0:9 16.2 17.3 18.7 17.6 1.3 18.3 18.3 18.8 18.7 0.7 16.9 18.2 18.0 17.7 0.9
LLL 9:8� 1:0 10.2 10.9 11.4 9.4 1.2 10.5 11.0 11.0 10.9 1.1 9.4 10.6 10.3 10.1 1.3
The last column of each sample size experiment represents the standard deviation (SD) of the four TAG composition A, B, C, and D data. The
TAG molecular species extracted by the control method (HR) are also shown. The order of the fatty acids in the TAG type does not indicate the
TAG structure (POS¼PSO¼SPO).a P, palmitic acid (16:0); S, stearic acid (18:0); O, oleic acid (18:1); L, linoleic acid (18:2).
Fig. 2. Time-course experiment of TAG extraction with mixture A
(see Table 1) using seed samples of 20 (N), 50 (j), and 200mg (r).
Three replicates were made in each experiment. TAG extraction is
expressed as milligrams (A) and percentage (B) of total TAG calcu-
lated by the extraction method of Hara and Radin [8].
N. Ruiz-L�oopez et al. / Analytical Biochemistry 317 (2003) 247–254 251
Second step—FAME obtainment and analysis
As previously shown, mixture A can be used to an-
alyze the TAG composition of a half-seed sample (3 to
10mg cotyledon weight). After the analysis of the TAG
composition of an aliquot of the extracted oil, the re-
maining TAG-containing heptane solution can be
transmethylated by the method of Garc�ees et al. [13] andthe FAMEs analyzed by GLC. The heptane upper phase
is transferred to a new test tube, and filled with heptane
up to 0.9ml total volume, and 0.9ml of a solution
containing methanol:toluene:DMP:H4SO2 (39:20:5:2) is
added. The tube is closed and incubated in a water bath
at 80 �C for 1 h. At 80 �C a single phase is formed and
the FAs in the sample are transmethylated. After cool-
ing, two phases separate and the FAMEs contained inthe upper phase are analyzed by GLC.
In Table 4 the calculated FA composition according
to the TAG composition obtained in the first step is
compared to the FA composition derived from FAME
analyses. The FA compositions obtained by both meth-
ods are very similar, indicating that both of them can be
used to calculate the FA composition of simple samples.
However, this second step could be useful to identify the
TAGs in some complex samples, like those from sun-flower mutant lines CAS-5 and CAS-12, that have high
palmitic content and some special TAG-containing pal-
mitoleic, asclepic, and palmitolinoleic acids [11].
Other plant tissues
In order to validate our method we tested the oil
extraction and analysis on other types of vegetablesamples with different water, fat, and protein contents,
Table 4
Fatty acid composition calculated from the TAG composition, taking into account that each TAG molecule has three fatty acids, and from analysis
by GLC of the FAMEs obtained after transmethylation of the TAG
Half-seed
sample
FA from TAG data (mol%) FAMEs from GLC (mol%) Ratio between TAG/GLC data
P S O L P S O L P S O L
A 7.3 10.4 26.0 56.3 7.1 10.6 25.6 56.7 1.03 0.98 1.02 0.99
B 7.1 11.2 23.3 58.4 7.8 10.8 22.4 59.0 0.91 1.04 1.04 0.99
C 7.2 11.1 24.9 56.8 7.2 10.9 23.8 58.0 1.01 1.01 1.04 0.98
D 6.2 13.5 33.1 47.2 6.7 13.8 31.9 47.6 0.92 0.98 1.04 0.99
E 6.5 15.1 40.7 37.8 6.7 14.8 39.7 38.8 0.98 1.01 1.02 0.97
F 6.9 15.0 27.4 50.7 7.2 14.5 27.4 50.9 0.96 1.04 1.00 1.00
Mean 6.9 12.7 29.2 51.2 7.1 12.6 28.5 51.8 0.97 1.01 1.03 0.99
SD 0.4 1.9 6.0 7.1 0.4 1.8 5.8 7.1 0.04 0.02 0.01 0.01
Six half-seeds from mutant line CAS-4 were used in this experiment (A to F). In the last two rows the means and SD of the six seed samples are
shown. Abbreviations as in Table 2.
Table 3
TAG molecular species composition (mol%) from the analysis of twin cotyledon samples obtained by cutting one seed transversally
TAG type Seed A Seed B Seed C Seed D Seed E Seed F
HR HR HR HR HR NM HR NM NM NM NM NM
POP 0.5 0.5 0.4 0.4 0.3 0.3 0.6 0.6 0.5 0.5 0.5 0.5
PLP 0.4 0.4 0.6 0.5 0.6 0.6 0.5 0.5 0.7 0.7 0.4 0.4
POS 2.3 2.5 1.9 1.9 1.9 1.8 3.0 3.0 2.0 2.0 2.4 2.2
POO 5.1 5.0 3.8 3.8 2.7 2.7 5.1 4.9 3.8 3.5 5.0 5.1
PLS 1.8 2.0 3.0 2.7 3.5 3.5 2.5 2.5 2.9 3.1 2.0 1.8
POL 5.6 5.8 6.2 6.0 5.5 5.5 6.1 6.0 6.6 6.4 5.8 5.8
PLL 2.2 2.4 4.1 3.7 4.3 4.5 2.6 2.7 4.4 4.5 2.4 2.3
SOS 2.2 2.5 1.8 1.9 2.2 2.1 3.1 3.2 1.8 1.9 2.4 2.2
SOO 12.3 12.9 9.3 9.6 8.0 7.8 12.9 13.1 8.6 8.4 12.7 12.3
SLS 1.3 1.4 2.2 2.2 3.7 3.7 2.3 2.1 1.9 2.1 1.3 1.4
OOO 13.3 12.0 8.1 8.6 4.9 5.0 10.1 9.7 7.4 6.4 12.5 13.5
SOL 12.6 13.4 13.3 13.3 15.1 14.7 14.8 14.9 13.0 13.2 13.2 12.6
OOL 19.0 17.7 14.4 15.1 11.7 11.4 15.2 15.1 14.8 13.7 17.7 18.8
SLL 6.1 6.4 11.2 10.8 14.8 15.0 7.6 7.6 10.4 11.3 6.4 5.8
OLL 12.4 12.0 14.7 15.0 14.9 14.8 10.7 10.9 15.6 15.6 12.0 12.3
LLL 2.9 3.1 5.0 4.6 5.9 6.6 2.9 3.3 5.7 6.8 3.2 3.2
Mixture A of Table 1 was used in the new method (NM) or control method (HR). Each seed (from seed A to F) was divided into the two
cotyledons cutting across the embryo; each cotyledon was analyzed by the indicated method. The experiments were carried out twice. The order of
the fatty acids in the TAG type does not indicate the TAG structure (POS¼PSO¼ SPO). Abbreviations as in Table 2.
252 N. Ruiz-L�oopez et al. / Analytical Biochemistry 317 (2003) 247–254
such as wild olive (O. europea) fruits (high water, me-
dium fat, and low protein contents), acorn (Q. ilex)
fruits (high water and low fat and protein contents), and
soybean (G. max) seeds (very low water, low fat, and
high protein contents). Results are shown in Table 5
(olive and acorn fruits) and Table 6 (soybean seeds).
Table 5
Twin experiments comparing the TAG molecular species composition (mol%) obtained by mixture A of Table 1 (NM) and the control method (HR),
for olive and acorn fruits
TAG typea Wild olive Acorn
NM HR NM HR
Mean SD Mean SD Mean SD Mean SD
POP 4.5 0.8 4.5 0.7 4.2 0.2 4.6 0.2
PLP 2.5 0.6 2.3 0.3 4.3 0.4 5.1 0.2
PPoL/PLnP 0.4 0.3 0.3 0.1 0.3 0.4 — 0.0
POS 0.8 0.1 0.9 0.1 1.1 0.2 0.8 0.1
POO 23.3 1.4 23.7 1.4 20.0 1.3 20.7 0.5
PoOS/PLS 1.6 0.3 1.6 0.5 1.0 0.2 1.2 0.1
POL 10.6 0.8 9.9 0.7 13.6 0.7 15.1 0.2
PLL 1.7 0.3 1.5 0.1 10.1 0.7 11.2 0.5
PoOL/PLLn 0.6 0.5 0.6 0.2 1.0 0.2 0.9 0.2
SOO 2.3 0.3 2.6 0.1 2.3 0.1 1.8 0.2
OOO 31.3 2.8 32.8 1.6 18.2 1.6 17.4 0.2
SOL 4.1 0.5 4.2 0.1 2.3 0.5 1.6 0.3
OOL 12.4 2.0 11.8 2.3 9.8 0.3 8.9 0.6
SLL 1.1 0.3 0.9 0.3 1.3 0.4 0.9 0.1
OLL 2.3 0.5 2.0 0.5 7.4 0.3 7.2 0.1
OLLn 0.6 0.3 0.5 0.2 — — — —
LLL — — — — 3.2 0.6 2.8 0.2
Extracted (%) 96.8 3.2 — — 98.7 1.3 — —
In each case a sample was divided into two aliquots and each aliquot analyzed by one method. Three replicates were made. The last row shows the
means and SD of the total TAGs extracted in four experiments. The order of the fatty acids in the TAG type does not indicate the TAG structure
(POS¼PSO¼ SPO). Abbreviations as in Table 2.a Po, palmitoleic acid (16:1); Ln, linolenic acid (18:3).
Table 6
Twin experiments comparing the TAG molecular species composition (mol%) of individual soybean seeds obtained using mixture A of Table 1 (NM)
and the control method (HR)
TAG type Seed A Seed B
NM HR NM/HR NM HR NM/HR
POP 0.5 0.5 1.0 0.5 0.6 1.0
PLP 1.7 1.5 1.1 2.0 2.1 1.0
POS 0.4 0.4 1.0 0.3 0.4 0.8
POO 2.1 2.4 0.9 2.0 2.2 0.9
PLS 1.2 1.2 1.1 1.3 1.5 0.9
POL 7.8 7.9 1.0 8.6 8.8 1.0
PLL 12.9 11.8 1.1 13.6 13.3 1.0
PLLn 2.0 1.8 1.1 2.7 2.6 1.0
SOS — 0.1 — — 0.1 —
SOO 0.5 0.7 0.7 0.4 0.6 0.8
OOO 2.0 2.5 0.8 1.7 1.8 0.9
SOL 2.5 2.8 0.9 2.4 2.8 0.9
OOL 7.9 9.1 0.9 7.3 7.4 1.0
SLL 3.9 4.0 1.0 3.4 4.0 0.8
OLL 19.4 20.5 0.9 19.3 19.2 1.0
LLL 29.3 27.7 1.1 27.6 26.3 1.1
LLLn 5.8 5.2 1.1 6.7 6.4 1.1
Extracted (%) 48.6 52.8
In each case, lipids from the upper phase (heptane) were extracted by the new method described and the remaining unextracted lipids were
reextracted with the control method. The TAG composition of each lipid sample was analyzed by GLC. In the last row the total TAGs extracted by
the new method is shown. The order of the fatty acids in the TAG type does not indicate the TAG structure (POS¼PSO¼ SPO). Abbreviations as in
Tables 2 and 5.
N. Ruiz-L�oopez et al. / Analytical Biochemistry 317 (2003) 247–254 253
The TAG extraction from olive and acorn fruits wasnear 100% when the method described in this work was
applied (see last row of Table 5), so it was not feasible to
reextract and test the TAGs remaining in the fruits using
the control method [8]. Thus, with the aim of comparing
the TAG compositions obtained by both methods in the
same samples, the fruits were cut in small pieces. Each
finely chopped sample was separated into two aliquots.
These aliquots were extracted and analyzed separatelyby the new or the control method. The TAG composi-
tions obtained by both methods were very similar (Table
5). These results show that the developed method can be
used for a quick TAG extraction and analysis in these
samples.
In experiments carried out with soybean seeds, the
new method extraction ratios were around 50% (see final
row of Table 6). Thus, to compare the TAG molecularspecies compositions obtained by both methods a sam-
ple of the seed was first extracted by the method de-
scribed in this work. The TAGs remaining in this sample
were extracted by the control method [8]. In the same
sample, TAG compositions obtained by both methods
were quite similar. Data from two individual soybean
seeds are shown (Table 6). Like for the other type of
vegetable samples, this method can be useful for thiskind of determination.
Taking into account that the extraction ratio in soy-
bean was smaller, around 50%, compared to olive and
acorn ratios, it could be suggested that the percentage of
extraction applying the new method is inversely pro-
portional to the sample protein content and that the
water content is not unfavorable for the TAG extraction.
Conclusions
In this paper we propose an easy and reliable method
for TAG and FAME sequential extraction and analysis
that is suitable for a large number of small vegetable
storage tissue samples. The best mixture contains by
volume 33.3% a solution of NaCl (0.17 M) in methanoland 66.6% heptane. After 2 h at 80 �C this mixture ex-
tracts about 80% of the TAG from sunflower seeds and
about 50, 97, and 99% from soybean seeds, olive fruits,
and acorn fruits, respectively. Even under incomplete
TAG extraction the TAG composition is representative
of the total TAG found in the tissue. All these samples
could be analyzed for TAG composition by GLC. The
remaining TAG-containing heptane solution could beused to get the FAMEs, following a modification of the
method of Garc�ees et al. [13], allowing the analysis of
FAMEs from exactly the same tissue sample.
This method could be very useful to select new vari-
eties of some plants, taking into account the TAG and/
or FA composition of seeds or fruits. It allows one tocarry out the analysis with one-third of the cotyledon,
leaving the rest of the seed containing the embryo. This
fragment could be grown and the new generation col-
lected to continue the selection of any new putative
character.
Acknowledgments
Thanks are due to M.C. Ruiz and D. Cabrera for
skilful technical assistance. We are especially grateful to
A.M. Muro-Pastor for critically reading the manuscript.This work was supported by MCYT Spanish govern-
ment, Junta de Andaluc�ııa, and Advanta Seeds B.V.
References
[1] C.K. Chow, Fatty Acids in Foods and Their Health Implications,
Dekker, New York, 1992.
[2] U. Bracco, Effect of triglyceride structure on fat absorption, Am.
J. Clin. Nutr. 60 (1994) 1002–1009.
[3] J. Osorio, J. Fern�aandez-Mart�ıınez, M. Mancha, R. Garc�ees, Mutant
sunflower with high concentration of saturated fatty acids in their
oil, Crop Sci. 35 (1995) 739–742.
[4] J. Folch, M. Lees, G.H. Sloane-Stanley, A simple method for the
isolation and purification of total lipids from animal tissues, J.
Biol. Chem. 226 (1957) 497–509.
[5] J.J. Wren, H.K. Mitchell, Extraction methods and an investiga-
tion of Drosophila lipids, J. Biol. Chem. 234 (1959) 2823–2828.
[6] R.M.C. Dawson, N. Hemington, D.B. Lindsay, in: G.B. Ansell,
J.W. Hawthorne, R.M.C. Dawson (Eds.), Form and Function of
Phospholipids, Elservier, Amsterdam, London, New York, 1960,
p. 44.
[7] O.S. Privett, K.A. Dougherty, W.L. Erdahl, A. Stolyhow, Studies
on the lipid composition of developing soybeans, JAOCS 50
(1973) 516–520.
[8] A. Hara, N.S. Radin, Lipid extraction of tissues with a low-
toxicity solvent, Anal. Biochem. 90 (1978) 420–426.
[9] J. Reske, J. Siebrecht, J. Hazebroek, Triacylglycerol composition
and structure in genetically modified sunflower and soybean oils,
JAOCS 74 (8) (1997) 989–998.
[10] C. Chaven, C. Hymowitz, C.A. Newell, Chromosome number, oil,
and fatty acid content of species in the genus Glycine subgenus
Glycine, JAOCS 59 (1982) 23–25.
[11] V. Fern�aandez-Moya, E. Martinez-Force, R. Garc�ees, Identificationof triacylglycerol species from high-saturated sunflower (Helian-
thus annuus) mutants, J. Agric. Food Chem. 48 (2000) 764–769.
[12] R.J. Vander Wal, Calculation of the distribution of the saturated
and unsaturated acyl groups in fats, from pancreatic lipase
hydrolysis data, JAOCS 37 (1960) 18–20.
[13] R. Garc�ees, M. Mancha, One-step lipid extraction and fatty acid
methyl esters preparation from fresh plant tissues, Anal. Biochem.
211 (1993) 139–143.
[14] A.A. Carelli, A. Cert, Comparative study of the determination of
triacylglycerol in vegetable oils using chromatographic techniques,
J. Chromatogr. 630 (1993) 213–222.
[15] A.V. Zhukov, A.G. Vereshchagin, Quantitative content of total
polar lipid in soybean seeds, JAOCS 53 (1) (1976) 1–7.
254 N. Ruiz-L�oopez et al. / Analytical Biochemistry 317 (2003) 247–254