Enantioselective chromatography in analysis of triacylglycerols common in edible fats and oils

Article abstract of DOI:10.1016/j.foodchem.2014.09.135

Enantiomers of racemic triacylglycerol (TAG) mixtures were separated using two chiral HPLC columns with a sample recycling system and a UV detector. A closed system without sample derivatisation enabled separation and identification by using enantiopure reference compounds of eleven racemic TAGs with C12-C22 fatty acids with 0-2 double bonds. The prolonged separation time was compensated for byfewer pretreatment steps. Presence of one saturated and one unsaturated fatty acid in the asymmetricTAG favoured the separation. Enantiomeric resolution, at the same time with stronger retention of TAGs,increased with increasing fatty acid chain length in the sn-1(3) position. Triunsaturated TAGs containingoleic, linoleic or palmitoleic acids did not separate. The elution order of enantiomers was determined bychemoenzymatically synthesised enantiopure TAGs with a co-injection method. The method is applicableto many natural fats and oils of low unsaturation level assisting advanced investigation of lipid synthesisand metabolism.

Full text of DOI:10.1016/j.foodchem.2014.09.135

Food Chemistry 172 (2015) 718–724
Contents lists available at ScienceDirect
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
Enantioselective chromatography in analysis of triacylglycerols common
in edible fats and oils
Marika Kalpio a, Matts Nylund a, Kaisa M. Linderborg a, Baoru Yang a, Björn Kristinsson b,
Gudmundur G. Haraldsson b, Heikki Kallio a,
⇑
a Food Chemistry and Food Development, Department of Biochemistry, University of Turku, FI-20014 Turku, Finland
b Science Institute, University of Iceland, Dunhaga 3, 107 Reykjavik, Iceland
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 23 June 2014
Received in revised form 12 September 2014
Accepted 23 September 2014
Available online 30 September 2014
Enantiomers of racemic triacylglycerol (TAG) mixtures were separated using two chiral HPLC columns
with a sample recycling system and a UV detector. A closed system without sample derivatisation
enabled separation and identification by using enantiopure reference compounds of eleven racemic TAGs
with C12–C22 fatty acids with 0–2 double bonds. The prolonged separation time was compensated for by
fewer pretreatment steps. Presence of one saturated and one unsaturated fatty acid in the asymmetric
TAG favoured the separation. Enantiomeric resolution, at the same time with stronger retention of TAGs,
increased with increasing fatty acid chain length in the sn-1(3) position. Triunsaturated TAGs containing
oleic, linoleic or palmitoleic acids did not separate. The elution order of enantiomers was determined by
chemoenzymatically synthesised enantiopure TAGs with a co-injection method. The method is applicable
to many natural fats and oils of low unsaturation level assisting advanced investigation of lipid synthesis
and metabolism.
Keywords:
Chiral
Enantiomer
High-performance liquid chromatography
Stereoisomer
Triacylglycerol
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
have the ability to rotate plane-polarised light in opposite direc-
tions. To measure the optical activity, a mixture must have an
When two different fatty acids are esterified in the sn-1 and
sn-3 positions of triacylglycerols (TAGs), the chiral molecule
appears in either S or R configuration due to the centre of asymme-
try in the sn-2 position. The enantiomers typically show different
biochemical behaviour and technological properties, even though
the chemical characteristics are close to each other (Stalcup,
2010). Analysis of the extremely wide selection of stereospecific
and species-specific TAG structures in edible fats and oils is a chal-
lenge (Brockerhoff, 1971; Innis, 2011; Takagi & Ando, 1991).
Today, the detailed molecular structures of TAGs are still largely
unknown (Itabashi, 2012). To understand the impact of the TAG
regioisomers and enantiomers on metabolism, bioavailability,
digestion, absorption, transport, and common health (Bracco,
1994; Kubow, 1996; Linderborg & Kallio, 2005; Small, 1991) as well
as on physicochemical properties, (Foubert, Dewettinck, Van de
Walle, Dijkstra, & Quinn, 2007; Itabashi, 2005) a fast and reliable
molecular level enantiospecific analysis is of primary importance.
The TAG enantiomers have nearly identical chromatographic
and spectroscopic behaviour, as well as physical properties,
regardless of the different absolute configurations. In theory, they
excess of one enantiomer, but in many cases the optical activity
alone does not reveal the TAG configurations (Anderson, Sutton,
& Pallansch, 1970; Brockerhoff, 1971; Schlenk, 1965). The enantio-
pure TAGs are known to possess cryptochirality (Mislow & Bickart,
1977) or being cryptoactive (Schlenk, 1965), referring to their
extremely low specific optical rotation as being very close to zero
and hardly measureable.
Several regiospecific mass spectrometric methods discriminate
between the fatty acids in the secondary position (sn-2) from those
in the primary positions (sn-1 and sn-3) (Currie & Kallio, 1993;
ˇ
Holcapek, Lísa, Jandera, & Kabátová, 2005; Kallio & Currie, 1993;
ˇ
Leskinen, Suomela, & Kallio, 2007; Lísa, Velínská, & Holcapek,
2009; Lísa et al., 2011). The loss of sn-2 fatty acids during tandem
mass spectrometry is energetically less favourable than the loss of
the fatty acids from the primary positions, whereas the positions
sn-1 and sn-3 are regarded as equivalent.
The most frequently used methods of stereospecific analysis are
enzymatic or chemical hydrolysis in combination with chromatog-
raphy (Blasi et al., 2008; Boukhchina, Gresti, Kallel, & Bézard, 2003;
Brockerhoff, 1971), via conversion to phospholipid derivatives fol-
lowed by hydrolysis with stereospecific phospholipases. Instead of
using phospholipid derivatives a chiral derivatising agent can be
applied (Christie, Nikolova-Damyanova, Laakso, & Herslof, 1991;
⇑
Corresponding author. Tel.: +358 2 333 6870; fax: +358 2 231 7666.
E-mail address: heikki.kallio@utu.fi (H. Kallio).
http://dx.doi.org/10.1016/j.foodchem.2014.09.135
0308-8146/Ó 2014 Elsevier Ltd. All rights reserved.

M. Kalpio et al. / Food Chemistry 172 (2015) 718–724
719
OC18:1
OC18:1
OC16:1
OC18:1
OC18:1
OC18:2
OC18:2
OC18:1
OC18:1
Laakso & Christie, 1990). The resolution of the diacylglycerol ure-
thanes is achieved by adsorption chromatography. Also chiral-
phase high-performance liquid chromatography (HPLC) is widely
applied as an alternative method for resolution of the isolated diac-
ylglycerol (DAG) and monoacylglycerol (MAG) derivatives in the
stereospecific analysis of TAGs (Itabashi, 2005). One significant
drawback in the laborious multistep methods is the acyl migration
of fatty acids between the positions of glycerol (Compton,
Vermillion, & Laszlo, 2007), and it is difficult to totally eliminate.
In the end, the stereochemistry of individual TAG molecules
remains unknown in these methods.
OC16:1
H
H
H
H
H
OC10:0
OC16:0
OC16:0
OC16:0
OC20:0
OC18:1
OC18:1
OC18:1
OC16:0
OC18:2
OC16:0
OC18:1
OC18:0
H
H
H
H
OC22:0
OC16:0
OC16:0
OC18:0
Fig. 1. The synthesised triacylglycerol enantiomers.
International (Radnor, PA). All solvents were either pro analysis
or HPLC grade and used without further purification.
Iwasaki and coauthors were the first to present the enantio-
meric separation of intact TAGs containing octanoic acid and eico-
sapentaenoic acid or docosahexaenoic acid, in combinations
unlikely to be found in nature (Iwasaki et al., 2001). A recent
study by Nagai et al. (2011) introduced a direct enantiomer reso-
lution of naturally occurring TAGs without derivatisation, using
HPLC in which the sample was recycled through the column sev-
eral times. Other methods based on circulation of the analytes
through one column, have also been developed (Charton, Bailly,
& Guiochon, 1994; Grill, 1998). In four applications the enantio-
meric separation of TAGs has been achieved by using polysaccha-
Fifteen racemic (Table 1) and nine enantiopure (Fig. 1) reference
compounds were used for the HPLC analyses. All racemic TAGs
were purchased from Larodan Fine Chemicals (Malmö, Sweden).
Enantiopure structured (S)-TAGs with saturated, mono- or diunsat-
urated fatty acids at predetermined positions of the glycerol back-
bone were prepared as described by Kristinsson, Linderborg, Kallio,
and Haraldsson (2014). In short, altogether 18 compounds, includ-
ing MAG and DAG intermediates, and AAB-type enantiopure struc-
tured (S)-TAGs were synthesised. (S)-TAGs may be divided into two
categories. Six of them possess one saturated fatty acyl group
located in the sn-3 position and two identical unsaturated fatty
acyl groups in the remaining sn-1 and sn-2 positions of the glycerol
backbone. The saturated fatty acids belonging to the first category
include decanoic (C10:0), palmitic (C16:0), arachidic (C20:0) and
behenic (C22:0) acids, and the unsaturated fatty acids present
include the monounsaturated palmitoleic (C16:1) and oleic
(C18:1) acids, and the diunsaturated linoleic (C18:2) acid. The
remaining three TAGs of second category possess two identical sat-
urated acyl groups in the sn-2 and sn-3 positions and one unsatu-
rated acyl group in the sn-1 position. The saturated fatty acids of
the second category are limited to palmitic and stearic (C18:0)
acids, and the unsaturated fatty acids are C18:1 and C18:2. The
synthesis of the first category TAGs is based on a five-step chemo-
ride-based chiral stationary phases (Iwasaki et al., 2001; Lísa &
ˇ
ˇ
´
Holcapek, 2013; Nagai et al., 2011; Rezanka, Lukavsky,
ˇ
& Sigler, 2012). Lísa and Holcapek
Nedbalová, Kolouchová,
(2013) separated TAG enantiomers and regioisomers with 1–8
double bonds and different chain lengths using two columns in
the normal-phase mode. However, specific co-elution problems
existed in analysis of TAGs with saturated and polyunsaturated
fatty acids in primary positions. The retention behaviour of TAG
enantiomers in chiral HPLC is highly complex and depends on
the specific molecular structures. Thus, the use of pure enantio-
meric reference compounds is essential.
The aim of the present study was to apply a sample recycling
HPLC system based on two identical chiral columns (Trone,
Vaughn, & Cole, 2006) to separate intact TAG enantiomers natu-
rally found in many food fats and oils. The other objective was to
determine the enantiomeric elution order to investigate the reten-
tion mechanisms.
enzymatic process involving
a highly regioselective Candida
antarctica lipase, and the second category TAG products were syn-
thesised by a fully chemical five-step synthetic route where no
enzyme was needed. All intermediates and final TAG products
were obtained in high chemical and regiopurity. No acyl migration
was observed to take place during these reactions. Synthesised
products were fully characterised by traditional synthetic organic
chemistry methods including 1H and 13C NMR and IR spectroscopy,
as well as high-resolution mass spectrometric analyses. Specific
rotation was determined for all chiral compounds involved, and
the melting point was determined for all compounds that were
crystalline.
2. Materials and methods
2.1. Chemicals and reference TAGs
Methanol was from J.T. Baker (Deventer, Netherlands).
n-Hexane was purchased from Sigma–Aldrich Corporation
(St. Louis, MO). Acetonitrile and 2-propanol were from VWR
Table 1
Retention times (tR, min) of all triacylglycerol racemates after the first column and separation factors (
a
).
tR
TAG
Abbreviation
a
1,2-Dioleoyl-3-caproyl-rac-glycerol
1,2-Dioleoyl-3-lauroyl-rac-glycerol
1,2-Dioleoyl-3-myristoyl-rac-glycerol
1,2-Dioleoyl-3-palmitoyl-rac-glycerol
1,2-Dioleoyl-3-stearoyl-rac-glycerol
1,2-Dioleoyl-3-arachidoyl-rac-glycerol
1,2-Dioleoyl-3-behenoyl-rac-glycerol
1,2-Dipalmitoleoyl-3-palmitoyl-rac-glycerol
1,2-Dipalmitoleoyl-3-oleoyl-rac-glycerol
1,2-Dilinoleoyl-3-palmitoyl-rac-glycerol
1,2-Dioleoyl-3-linoleoyl-rac-glycerol
1,2-Dipalmitoyl-3-linoleoyl-rac-glycerol
1,2-Dipalmitoyl-3-elaidoyl-rac-glycerol
1,2-Dipalmitoyl-3-oleoyl-rac-glycerol
1,2-Distearoyl-3-oleoyl-rac-glycerol
rac-18:1-18:1-10:0
rac-18:1-18:1-12:0
rac-18:1-18:1-14:0
rac-18:1-18:1-16:0
rac-18:1-18:1-18:0
rac-18:1-18:1-20:0
rac-18:1-18:1-22:0
rac-16:1-16:1-16:0
rac-16:1-16:1-18:1
rac-18:2-18:2-16:0
rac-18:1-18:1-18:2
rac-16:0-16:0-18:2
rac-16:0-16:0-tr18:1
rac-16:0-16:0-18:1
rac-18:0-18:0-18:1
17.9
19.8
25.0
29.4
34.9
40.0
48.6
20.4
21.6
24.6
26.0
27.3
30.9
31.0
40.0
n.d.*
1.006
1.020
1.021
1.022
1.045
1.074
1.014
n.d.
1.016
n.d.
1.030
n.d.
1.027
1.035
*
n.d.: not defined.

720
M. Kalpio et al. / Food Chemistry 172 (2015) 718–724
Fig. 2. Principle of the sample recycle HPLC equipment. The two different figures correspond to the both positions of the valve. (A) col. 1 ? UV ? col. 2, (B) col. 2 ? UV ? col. 1.
Fig. 3. UV chromatograms of eleven racemic triacylglycerols, (A) rac-18:1-18:1-12:0, (B) rac-18:1-18:1-14:0, (C) rac-18:1-18:1-16:0, (D) rac-18:1-18:1-18:0, (E) rac-18:1-
18:1-20:0, (F) rac-18:1-18:1-22:0, (G) rac-16:1-16:1-16:0, (H) rac-18:2-18:2-16:0, (I) rac-18:2-16:0-16:0, (J) rac-18:1-16:0-16:0, (K) rac-18:1-18:0-18:0. Columns:
CHIRALCEL OD-RH (150 ꢀ 4.6 mm, 5
lm), Mobile phase: methanol, Detector: UV SPD-20A at 205 nm.
2.2. Recycle HPLC instrumentation and chromatographic conditions
coated on 5
l
m silica-gel and precolumn (10 ꢀ 4.0 mm, 5
lm)
were obtained from Chiral Technologies Europe (Illkirch, France).
The Shimadzu Prominence system (Kyoto, Japan) consisted of a
SIL-20A autosampler, an LC-20AB pump, a CTO-10AC column oven,
The CHIRALCEL OD-RH (150 ꢀ 4.6 mm, 5
lm) columns with cel-
lulose tris(3,5-dimethylphenylcarbamate) stationary phases

M. Kalpio et al. / Food Chemistry 172 (2015) 718–724
721
and a DGU-20A5 degasser. The UV detector was an SPD-20A sys-
tem, and fractions were collected automatically using an FRC-
10A or manually. With the CHIRALCEL OD-RH columns, isocratic
methanol at a flow rate of 0.5 mL/min at 25 °C was used. Detection
was carried out at 205 nm with an SPD-20A UV detector. The
CS3080 Sample Peak Recycler™ (Chiralizer Services, Newtown,
PA) was installed in this system. The TAG enantiomers were
resolved by rerunning the partially separated compounds after
the first column via the UV detector to the second column. After
the first round (consisting of two columns), recycling was contin-
ued through the identical columns until the desired resolution
was attained. Either manual or automated (by using the LCsolution
program, Shimadzu, Kyoto, Japan) switching between the columns
was used.
The recycling system consisted of a controlling device and a
high-pressure 10-port valve. The valve was set to pump the solvent
through the first column via the injection valve and the pre-
column (Fig. 2A). Once the last peak in the peak pair was passed
through the UV detector, the enantiomer pair was presumed to
be eluted completely onto the second column, and the position
of the valve was changed (Fig. 2B). If the peak resolution was not
satisfactory after passing through the second column, the LC Sam-
ple Recycler™ was triggered again (col. 1 ? UV ? col. 2) by timed
switching. The analyte was circulating continuously through the
first column, the detector, back to the valve and then back to the
second column until a desired resolution was attained. The UV
detector was applied to monitor in real time the separation after
each of the columns without disturbing the run or destroying the
sample.
2.3. Chiral resolution
Altogether, 15 pure racemic TAGs, which are commonly found
in nature and in food oils, were diluted separately in n-hexane/
2-propanol (3:2, v/v) to 1 g/L, and the injection volume was
15
lL. The enantiopurity of the synthesised TAGs was confirmed
by injecting 10
l
L of a sample containing 7–15 g of enantiomer
l
diluted in n-hexane/2-propanol. The elution order of enantiomers
was determined by co-injection of enantiopure and racemic TAG.
A 15
pure TAG were diluted in n-hexane/2-propanol, and injected in
20 L.
Separation of the enantiomeric TAGs was followed by determin-
ing the progress of total retention time (tR), adjusted retention time
(t0R), separation factor (
), peak resolution (Rs), and valley-to-peak
ratio ( /p) of the analytes. The separation factor, which shows the
lg aliquot of the racemic TAG together with 5 lg of enantio-
l
a
v
ability of the column to differentiate the analytes, was defined by
the following equation:
a
¼ k2=k1; where k ¼ t0R=tM
ð1Þ
In Eq. (1), tM is the hold-up time, t0R is the adjusted retention
time of the analyte, and k is defined as the retention factor.
Peak resolution, in this context, means the separation of a race-
mate into the component enantiomers, and is calculated according
to Eq. (2):
Fig. 4. Measure of separation. (A) Resolutions of ten triacylglycerol racemates after
each round. Lower horizontal line corresponding to 50% of perfect separation (Level
1 was R = 0.75), upper line corresponding to perfect separation (Level 2 was R = 1.5).
(B) The ratio of the valley height to the peak height of eleven triacylglycerol
racemates after each round. (Round = 2 columns). (C) Elution order of fifteen
racemates according to the equivalent carbon numbers. Triacylglycerols having
C18:1 at the sn-1(3) and sn-2 positions and fatty acid C10:0 to C22:0 at the sn-3(1)
position are marked as triangles, and others are marked as crosses. Exponential
curve is marked as dashed lines. (D) Exponential curve fitted for the data points of
triacylglycerols having C18:1 at the sn-1(3) and sn-2 positions and fatty acid from
C10:0 to C22:0 at the sn-3(1) position.
2ðtR2 ꢁ tR1
wb þ wb
Þ
4
Rs ¼
; where wb2
¼
ꢂ wh
ð2Þ
2:355
1
2
In Eq. (2), wb2 is the base width of the latter peak, and wh is the
width at half height. In peak pairs, where the bottom of the valley
was higher than at the half height of the peak, peak resolution was

722
M. Kalpio et al. / Food Chemistry 172 (2015) 718–724
not determined. The ratio of the valley height to the peak height
(Christophe, 1971) may be employed as a measure of separation
even when the separation is very weak. It is calculated from Eq. (3):
3.2. Measure of separation
The chain length of fatty acids in TAGs had a significant impact
on elution rate; the shorter the carbon chain, the faster the elution.
A higher degree of unsaturation results in a shorter retention time
as shown when comparing the TAGs rac-18:1-18:1-18:0 (tR 34.9)
and rac-18:1-18:0-18:0 (tR 40.0). Table 1 lists all TAGs examined,
along with their retention times (tR) after the first column, and
m=p ratio ¼ hm=hp
ð3Þ
In this equation, hv is the valley height, and hp is the peak height
of the first peak of two.
the separation factors (a). The list starts with seven TAGs, which
3. Results and discussion
possess a saturated fatty acid C10:0–C22:0 in the sn-1(3) position
and two C18:1 fatty acids in the remaining positions. This is fol-
lowed by TAGs with two or three unsaturated fatty acids, and
finally TAGs with unsaturated fatty acid in a primary position along
with two saturated fatty acids. The separation factor defines the
distance of the peak maxima of the two enantiomers but does
not take the peak widths into account. Together with the retention
time, the separation factor is a useful value when evaluating the
separation ability of the circulating system.
Peak resolution determines the progress of separation of the
racemate into the enantiomers over time (Fig. 4A). The definition
of one round, in this context, means the full cycle including col-
umns 1 and 2 (Fig. 2). The improvement of the apparent resolution
was not in linear correlation to the number of full cycles due to the
non-ideal peak broadening. Resolution at level one (Fig. 4A) may be
regarded as a sufficient measure for estimation of the proportions
of the TAG enantiomers. During the rounds previous to those
shown in Fig. 4A, either resolution was not observed or the widths
of the peaks at the half height could not be defined. For partially
overlapping peaks, v/p ratios were determined (Fig. 4B). The v/p
ratio was determined for rac-18:1-18:1-12:0 although the resolu-
tion could not be calculated. The v/p values of TAGs examined
decreased similarly, except rac-18:1-18:1-12:0 for which the enan-
tiomeric separation was very slow.
3.1. Separation of racemic TAGs
The chiral recycling HPLC method in polar-organic phase mode
enabled successful enantioseparation of eleven out of fifteen
racemic TAGs examined. Chiral separation with the column and
conditions examined was effective when the asymmetric TAG mol-
ecule contained both saturated and unsaturated fatty acids.
The UV chromatograms of the eleven racemic TAGs separated
into enantiomers using the recycle HPLC method are presented
in Fig. 3. The progress of separation between the enantiomers is
illustrated. Molecular weights of the isolated enantiomers were
confirmed as protonated TAGs and ammonium adducts by atmo-
spheric pressure chemical ionisation mass spectrometry. Retention
of the compounds as well as the separation power increased with
the corresponding length of the fatty acid in the sn-1(3) position, as
illustrated in the chromatograms A–F of Fig. 3.
Four racemic reference mixtures analysed did not exhibit any
enantioresolution in an applicable time (rac-18:1-18:1-10:0, rac-
16:1-16:1-18:1, rac-18:1-18:1-18:2 and rac-16:0-16:0-tr18:1).
Both automated and manual column switching was applied with
rac-18:1-18:1-10:0, and after 32 elution cycles (running time
9.5 h) no enantiomeric separation was attained. After 10 columns,
no resolution was achieved with the triunsaturated TAGs consist-
ing of fatty acids with one or two double bonds. Theoretically,
the sample can be recycled as many times as needed for the
desired resolution to be obtained, but in practice peak broadening
limits the number of cycles.
3.3. Retention mechanism
The compounds were eluted in relation to the equivalent carbon
number (ECN) defined as the number of carbon atoms in the fatty
acid residues minus twice the number of double bonds (Fig. 4C).
The r2 values indicate sufficient fits of both the linear line
(r2 = 0.970) and exponential curve (r2 = 0.967). When TAGs with
C18:1 in the sn-1(3) and sn-2 positions and fatty acid from C10:0
to C22:0 in the sn-3(1) position are taken into account, the data
fit to the exponential model perfectly (r2 = 0.996), as illustrated
Trisaturated TAGs were not examined because their absorption
of UV light is very low. Routinely used evaporative light scattering
detection or mass spectrometry (Holcapek et al., 2005; Leskinen
et al., 2007) are inoperable as the sole detectors in a recycling sys-
tem, in which the detection method must be non-destructive.
However, after isolation of an enantiomer it can be further ana-
lysed with LC–MS, as the mobile phase is LC–MS compatible.
ˇ
Fig. 5. UV chromatograms of (A) rac-18:2-16:0-16:0 co-injected with 25% of sn-18:2-16:0-16:0 and (B) rac-18:2-18:2-16:0 co-injected with 25% of sn-18:2-18:2-16:0. HPLC
conditions identical to Fig. 3. U – unsaturated, S – saturated.

M. Kalpio et al. / Food Chemistry 172 (2015) 718–724
723
in Fig. 4D. The curves can be used to forecast retention behaviour of
chiral TAG racemates under the chromatographic conditions
examined. However, the curves can never fully replace enantiopure
reference compounds.
University of Turku for assistance with the recycling system.
Robert M. Badeau, Ph.D. (robert.badeau@gmx.com), of the
University of Turku Language Centre, is acknowledged for the
proofreading and scientific editing of this manuscript. Financial
support from the Academy of Finland (Grant 251623) is gratefully
acknowledged.
3.4. Identification of enantiomeric TAGs
The UV chromatograms of nine synthesised enantiopure TAGs
analysed by the chiral recycle HPLC technique showed an absence
of impurities, and that no isomerisation occurred during synthesis.
The elution order of stereoisomeric TAGs was determined by the
co-injection technique. Fig. 5 shows chromatograms of the TAG
mixtures of 75% rac-18:2-16:0-16:0 co-injected with 25% of
sn-18:2-16:0-16:0 (Fig. 5A) and 75% rac-18:2-18:2-16:0
co-injected with 25% of sn-18:2-18:2-16:0 (Fig. 5B). After the first
two column passes minimal visible resolution was noticed but from
thereon a sequenced elution of sn-18:2-16:0-16:0 and sn-16:
0-16:0-18:2 enantiomers, respectively, was recognisable. The
analogous enantiomeric separation of sn-18:2-18:2-16:0 from
sn-16:0-18:2-18:2 was visible only after the fifth column, and
improved cycle by cycle. In these two examples, as in the analyses
of all the other critical pairs of TAG enantiomers, the enantiomer
with unsaturated fatty acid in the sn-1 position eluted faster than
the enantiomer with the unsaturated fatty acid in the sn-3 position.
References
Anderson, B. A., Sutton, C. A.,
& Pallansch, M. J. (1970). Optical activity of
butterfat and vegetable oils. Journal of The American Oil Chemists’ Society, 47(1),
15–16.
Blasi, F., Montesano, D., De Angelis, M., Maurizi, A., Ventura, F., Cossignani, L., et al.
(2008). Results of stereospecific analysis of triacylglycerol fraction from donkey,
cow, ewe, goat and buffalo milk. Journal of Food Composition and Analysis, 21(1),
1–7.
Boukhchina, S., Gresti, J., Kallel, H., & Bézard, J. (2003). Stereospecific analysis of TAG
from sunflower seed oil. Journal of The American Oil Chemists’ Society, 80(1), 5–8.
Bracco, U. (1994). Effect of triglyceride structure on fat absorption. The American
Journal of Clinical Nutrition, 60(6), 1002S–1009S.
Brockerhoff, H. (1971). Stereospecific analysis of triglycerides. Lipids, 6(12),
942–956.
Charton, F., Bailly, M.,
chromatography. Journal of Chromatography A, 687(1), 13–31.
Christie, W. W., Nikolova-Damyanova, B., Laakso, P., Herslof, B. (1991).
Stereospecific analysis of triacyl-sn-glycerols via resolution of diastereomeric
diacylglycerol derivatives by high-performance liquid chromatography on
silica. Journal of The American Oil Chemists’ Society, 68(10), 695–701.
Christophe, A. B. (1971). Valley to peak ratio as a measure for the separation of two
chromatographic peaks. Chromatographia, 4(10), 455–458.
& Guiochon, G. (1994). Recycling in preparative liquid
&
4. Conclusion
Compton, D. L., Vermillion, K. E., & Laszlo, J. A. (2007). Acyl migration kinetics of 2-
monoacylglycerols from soybean oil via H-1 NMR. Journal of The American Oil
Chemists Society, 84(4), 343–348.
A straightforward sample recycling method, without enzymatic
hydrolysis or derivatisation, offers reliable information on the
enantiomeric composition of TAGs using fewer chemicals than tra-
ditional methods. The results can be used to characterise the chro-
matographic behaviour of TAG enantiomers. During one run the
sample can pass up to thirty times through the column. Amount
of stationary phase contact with the sample is multiplied contrary
to traditional column chromatography, where two or three col-
umns in series are the maximum due to the high back-pressure.
This method is applicable for chiral racemates commonly found
in nature that are only partially resolved, if the length of separation
time and the peak broadening cause no restrictions.
Only a certain number of enantiomers may be separated from
each other at a time, and the number of these compounds varies
according to TAG composition of the sample. Thus, the individual
natural oil may need to be pre-fractionated into more simple
TAG groups or even single TAGs for the enantiomeric analysis.
With the columns and conditions examined separation of triun-
saturated TAGs was practically impossible. However, the most
common natural lipids comprise mixed fatty acid composition,
and thus the limitation to TAGs with both saturated and unsatu-
rated fatty acids often does not limit the analysis substantially.
In addition to analysis of stereospecific TAG molecular compo-
sitions and structural changes from natural matrices or processed
food, the technique provide information about the enantiomeric
purity of synthesised products. Knowledge of enantiopurity and
stereospecific composition of TAGs enables more specific investi-
gation of TAG fatty acid composition and its effects on health
and lipid biosynthesis. The presented method contributes to the
area of lipid research, which has been limited by the absence of
reliable methods applicable for complex natural and biological
samples.
Currie, G. J., & Kallio, H. (1993). Triacylglycerols of human milk: Rapid analysis by
ammonia negative ion tandem mass spectrometry. Lipids, 28(3), 217–222.
Foubert, I., Dewettinck, K., Van de Walle, D., Dijkstra, A. J., & Quinn, P. J. (2007).
Physical properties: Structural and physical characteristics. In F. D. Gunstone, J.
L. Harwood, & A. J. Dijkstra (Eds.), The lipid handbook with CD-ROM (3rd,
pp. 471–490). Boca Raton, FL: CRC Press.
Grill, C. M. (1998). Closed-loop recycling with periodic intra-profile injection: A new
binary preparative chromatographic technique. Journal of Chromatography A,
796(1), 101–113.
ˇ
Holcapek, M., Lísa, M., Jandera, P.,
& Kabátová, N. (2005). Quantitation of
triacylglycerols in plant oils using HPLC with APCI-MS, evaporative light-
scattering, and UV detection. Journal of Separation Science, 28(12), 1315–1333.
Innis, S. M. (2011). Dietary triacylglycerol structure and its role in infant nutrition.
Advances in Nutrition, 2(3), 275–283.
Itabashi, Y. (2005). Chiral-phase HPLC of glycerolipids. In J.-T. K. Lin & T. A. McKeon
(Eds.), HPLC of acyl lipids (pp. 168–171). New York: HNB Publishing.
Itabashi, Y. (2012). Chiral separation of glycerolipids by high-performance liquid
chromatography. Journal of Lipid Nutrition, 21(1), 27–34.
Iwasaki, Y., Yasui, M., Ishikawa, T., Irimescu, R., Hata, K., & Yamane, T. (2001). Optical
resolution of asymmetric triacylglycerols by chiral-phase high-performance
liquid chromatography. Journal of Chromatography A, 905(1–2), 111–118.
Kallio, H., & Currie, G. (1993). Analysis of low erucic-acid turnip rapeseed oil
(Brassica campestris) by negative ion chemical ionization tandem mass
spectrometry. A method giving information on the fatty acid composition in
positions sn-2 and sn-1/3 of triacylglycerols. Lipids, 28(3), 207–215.
Kristinsson, B., Linderborg, K. M., Kallio, H., & Haraldsson, G. G. (2014). Synthesis of
enantiopure structured triacylglycerols. Tetrahedron Asymmetry, 25(2),
125–132.
Kubow, S. (1996). The influence of positional distribution of fatty acids in native,
interesterified and structure-specific lipids on lipoprotein metabolism and
atherogenesis. The Journal of Nutritional Biochemistry, 7(10), 530–541.
Laakso, P.,
& Christie, W. W. (1990). Chromatographic resolution of chiral
diacylglycerol derivatives: Potential in the stereospecific analysis of triacyl-
sn-glycerols. Lipids, 25(6), 349–353.
Leskinen, H., Suomela, J.-P., & Kallio, H. (2007). Quantification of triacylglycerol
regioisomers in oils and fat using different mass spectrometric and liquid
chromatographic methods. Rapid Communications in Mass Spectrometry, 21(14),
2361–2373.
Linderborg, K. M., & Kallio, H. P. T. (2005). Triacylglycerol fatty acid positional
distribution and postprandial lipid metabolism. Food Reviews International,
21(3), 331–355.
ˇ
Lísa, M., & Holcapek, M. (2013). Characterization of triacylglycerol enantiomers
using chiral HPLC/APCI-MS and synthesis of enantiomeric triacylglycerols.
Analytical Chemistry, 85(3), 1852–1859.
Acknowledgements
ˇ
ˇ
ˇ
Lísa, M., Netušilová, K., Franek, L., Dvoráková, H., Vrkoslav, V., & Holcapek, M.
(2011). Characterization of fatty acid and triacylglycerol composition in animal
fats using silver-ion and non-aqueous reversed-phase high-performance liquid
chromatography/mass spectrometry and gas chromatography/flame ionization
detection. Journal of Chromatography A, 1218(42), 7499–7510.
Novozymes AS in Denmark is acknowledged for the lipase, and
the authors would like to thank Dr. Sigridur Jonsdottir at University
of Iceland for accurate MS measurements and Jani Sointusalo at

724
M. Kalpio et al. / Food Chemistry 172 (2015) 718–724
ˇ
Lísa, M., Velínská, H., & Holcapek, M. (2009). Regioisomeric characterization of
triacylglycerols using silver-ion HPLC/MS and randomization synthesis of
standards. Analytical Chemistry, 81(10), 3903–3910.
Schlenk, W. Jr., (1965). Synthesis and analysis of optically active triglycerides.
Journal of The American Oil Chemists’ Society, 42(11), 945–957.
Small, D. M. (1991). The effects of glyceride structure on absorption and
metabolism. Annual Review of Nutrition, 11(1), 413–434.
Stalcup, A. M. (2010). Chiral separations. Annual Review of Analytical Chemistry, 3(1),
341–363.
Takagi, T., & Ando, Y. (1991). Stereospecific analysis of triacyl-sn-glycerols by chiral
high-performance liquid chromatography. Lipids, 26(7), 542–547.
Trone, M. D., Vaughn, M. S., & Cole, S. R. (2006). Semi-automated peak trapping
recycle chromatography instrument for peak purity investigations. Journal of
Chromatography A, 1133(1–2), 104–111.
Mislow, K., & Bickart, P. (1977). An epistemological note on chirality. Israel Journal of
Chemistry, 15(1–2), 1–6.
Nagai, T., Mizobe, H., Otake, I., Ichioka, K., Kojima, K., Matsumoto, Y., et al. (2011).
Enantiomeric separation of asymmetric triacylglycerol by recycle high-
performance liquid chromatography with chiral column. Journal of
Chromatography A, 1218(20), 2880–2886.
ˇ
Rezanka, T., Lukavsky´, J., Nedbalová, L., Kolouchová, I., & Sigler, K. (2012). Effect of
starvation on the distribution of positional isomers and enantiomers of
triacylglycerol in the diatom Phaeodactylum tricornutum. Phytochemistry, 80,
17–27.