Chiral discrimination of branched-chain fatty acids by reversed-phase HPLC after labeling with a chiral fluorescent conversion reagent

Article abstract of DOI:10.1271/bbb.68.153

Anteiso fatty acids having 16 to 29 carbon atoms were labeled with the chiral fluorescent conversion reagents, (1R,2R)- and (1S,2S)-2-(2,3- anthracenedicarboximido)-cyclohexanol. The diastereomeric esters of anteiso acids having up to 20 carbon atoms were

Full text of DOI:10.1271/bbb.68.153

This article was downloaded by: [McGill University Library]
On: 24 November 2014, At: 12:38
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Bioscience, Biotechnology, and Biochemistry
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/tbbb20
Chiral Discrimination of Branched-chain Fatty Acids
by Reversed-phase HPLC after Labeling with a Chiral
Fluorescent Conversion Reagent
Kazuaki AKASAKAa & Hiroshi OHRUIa
a Graduate School of Life Sciences, Tohoku University
Published online: 22 May 2014.
To cite this article: Kazuaki AKASAKA & Hiroshi OHRUI (2004) Chiral Discrimination of Branched-chain Fatty Acids by
Reversed-phase HPLC after Labeling with a Chiral Fluorescent Conversion Reagent, Bioscience, Biotechnology, and
Biochemistry, 68:1, 153-158, DOI: 10.1271/bbb.68.153
To link to this article: http://dx.doi.org/10.1271/bbb.68.153
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of
the Content. Any opinions and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied
upon and should be independently verified with primary sources of information. Taylor and Francis shall
not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other
liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or
arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions

Biosci. Biotechnol. Biochem., 68 (1), 153–158, 2004
Chiral Discrimination of Branched-chain Fatty Acids by Reversed-phase
HPLC after Labeling with a Chiral Fluorescent Conversion Reagent
y
Kazuaki AKASAKA and Hiroshi OHRUI
Graduate School of Life Sciences, Tohoku University,
Tsutsumidori-Amamiyamachi 1-1, Aoba-ku, Sendai 981-8555, Japan
Received July 23, 2003; Accepted September 6, 2003
Anteiso fatty acids having 16 to 29 carbon atoms were
labeled with the chiral fluorescent conversion reagents,
(1R,2R)- and (1S,2S)-2-(2,3-anthracenedicarboximido)-
cyclohexanol. The diastereomeric esters of anteiso acids
having up to 20 carbon atoms were separated into two
peaks in an ODS column under low column-temper-
ature conditions, while those having more than 21
carbon atoms were not separated. A C30 column made
it possible to separate diastereomeric esters up to C29
anteiso acid. It was possible to predict the absolute
configuration of each acid by the elution order of the
derivatives.
Fig. 1. Structures of Reagent 1 and (S)-Anteiso Acids 2–16.
Key words: chiral discrimination; enantiomeric separa-
tion; fluorometric conversion; chiral
branched-chain fatty acid
having a 100% chiral gauche conformation. Chiral
branched-chain fatty acids having a methyl group up to
the C12 position could be separated by reversed-phase
HPLC after their conversion with these reagents,9) and
the method was successfully applied to determine the
absolute configuration of a naturally occurring prod-
uct.10) However, there are many branched-chain fatty
acids which have more than 11 methylene spacers
between the chiral center and the carboxyl group.11–15) In
this paper, to show the limitation of this method, we
report the improved separation of anteiso fatty acids
having 15–29 carbon atoms (2–16, Fig. 1) in a C30
column after labeling with these reagents.
The chiral discrimination of both naturally occurring
and synthetic materials has received much attention
because the enantiomers would have different biological
properties. Determination of the absolute stereochemis-
try is therefore very important for assessing chiral
compounds.
Although methyl branched-chain fatty acids and
alcohols have been found in many natural products,1–5)
it has been very difficult or practically impossible to
discriminate their enantiomers with very small amounts
of samples. We have reported (S)-(þ)-2-(2,3-anthra-
cenedicarboximido)-1-propyl trifluoromethanesulfonate
and (S)-(þ)-1-(2,3-anthracenedicarboximido)-2-propa-
nol as chiral fluorescent conversion reagents which have
made it possible to separate and detect the enantiomers
of chiral fatty acids having a hydroxy or a methyl group
at the C2 to C6 position by HPLC at the fmol level.6–8)
In respect of both reagents, the gauche trans conformer
in the ethanol amine part of their derivatives played a
very important role in the chiral discrimination. To
improve the chiral discriminating ability, we have
recently reported (1S,2S)- and (1R,2R)-2-(2,3-anthra-
cenedicarboximido)cyclohexanol (1, Fig. 1) as reagents
Experimental
Chemicals. (1R,2R)- and (1S,2S)-2-(2,3-Anthracene-
dicarboximido)cyclohexanol were prepared by the meth-
od previously reported.9) The solvents used for the
HPLC analysis were of HPLC grade purchased from
Kanto Chemical Co. (Tokyo, Japan). The other reagents
and solvents were purchased from Wako Pure Chemical
Co. (Osaka, Japan) and Tokyo Kasei Organic Chemicals
(Tokyo, Japan). (S)-12-Methyltetradecanoic acid (2) was
prepared by the method previously describ.9)
y
To whom correspondence should be addressed. Fax: +81-22-717-8806; E-mail: ohrui@biochem.tohoku.ac.jp
Abbreviations: DMF, dimethylformamide; THF, tetrahydrofuran; EDC, 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide hydrochloride; DMAP,
4-dimethylaminopyridine; ꢀ, separation factor; Rs value, resolution factor; k0, retention factor

154
K. AKASAKA and H. OHRUI
All synthesized products were purified by silica gel
been prepared by the coupling reaction of (S)-2-
methylbutyl tosylate and 8-benzyloxyoctyl magnesium
bromide with subsequent catalytic debenzylation and
then tosylation. Optically pure (S)-11, 12, 15 and 16
were respectively prepared in a similar manner via (S)-
12-methyltetradecyl tosylate (for 11 and 12) and (S)-16-
methyloctadecyl tosylate (for 15 and 16).
column chromatography. High-resolution mass spectra
were measured in the EI or FAB mode by using 4-
nitrobenzaldehyde as a matrix with a JMS HX-105
1
instrument (Jeol, Tokyo, Japan). H-NMR spectra were
measured in CDCl3 by a Varian Unity INOVA 500
instrument, and IR spectra were measured by an Impact
410 FT-IR (Nicolet, U.S.A.). Optical rotation was
measured by a DIP-370 polarimeter (Jasco, Tokyo,
Japan).
(S)-13-methylpentadecanoic acid (3): mp29.0ꢀC;
[ꢀ]20 +3.3ꢀ (c 1.73, CHCl3); HRMS (EI)): calcd. for
D
C16H32O2([M]þ), 256.2402; found, 256.2402.
(S)-14-methylhexadecanoic acid (4): mp34.5ꢀC;
Synthesis of optically pure anteiso acids (S)-3–16.
Enantiometrically pure (S)-3 was prepared by adding
sodium hydride (3.7 g, 0.16 mol) to 1,9-nonandiol (25 g,
0.16 mol) in DMF at 0ꢀC. The mixture was stirred at 0ꢀC
for 1 h and then at room temperature for 14 h. Benzyl-
bromide (26.7 g, 0.16 mol) was then added to the
reaction mixture to give 9-benzyloxy-1-nonanol (15.7
g, 40.2%). This monobenzyloxy alcohol (0.8 g, 83
mmol) and CBr4 (33.2 g, 100 mmol) were dissolved in
70 ml of CH2Cl2, and the mixture was stirred for 30 min
at 0ꢀC, before triphenylphosphine (32.8 g, 0.13 mol) was
added to it. The mixture was stirred for 1 h to give 1-
benzyloxy-9-bromononane (22.2 g, 85.3%). This bro-
mide (3.3 g, 10 mmol) was reacted with magnesium
(0.30 g, 13 mmol) in dry THF under an Ar atmosphere to
give the corresponding Grignard reagent. (S)-4-methyl-
hexyl tosylate (0.7 g, 2.6 mmol), which had been
prepared from optically pure (S)-4-methylhexanol, and
catalytic amount of CuI were added to the Grignard
reagent in THF at 0ꢀC, and a the mixture was stirred
overnight at room temperature to give (S)-1-benzyloxy-
13-methylpentadecane (0.61 g, 18.9%). Catalytic deben-
zylation of this compound over palladium-carbon in
ethanol gave (S)-13-methylpentadecan-1-ol (0.57 g,
93.6%). This alcohol (0.30 g, 1.2 mmol) was oxidized
with KMnO4 (0.6 g) in the presence of tetrabutylammo-
nium bromide (0.2 g) in 50 ml of a 1:1 mixture of
CH2Cl2 and H2O while vigorously stirring for 16 h at
room temperature to give (S)-3 (0.21 g, 68.4%).
Enantiometrically pure (S)-4 was also prepared via
(S)-14-methylhexadecanol in the same manner as that
just described above by using 1,12-dodecanediol and
optically pure (S)-2-methylbutyl tosylate. (S)-14-Meth-
ylhexadecanol was also used as an intermediate for the
synthesis of (S)-5, 6, 7, 8, 12, 13, and 14. (S)-14-
Methylhexadecanol was tosylated with tosylchloride,
and the tosylate was then coupled with vinyl, allyl, 3-
butenyl, 4-pentenyl, 9-decenyl or 10-undecenyl magne-
sium bromide in the presence of a catalytic amount of
CuI in THF. Their terminal double bonds were oxidized
with KMnO4 in CH2Cl2/ H2O in the presence of
tetrabutylammoium bromide as a phase-transfer catalyst
while vigorously stirring at room temperature for 16 h to
respectively give (S)-5, 6, 7, 8, 13 and 14.
[ꢀ]20 +5.0ꢀ (c 2.10, CHCl3); HRMS (FAB): calcd.
D
for C17H33O2([M-H]ꢁ), 269.2490; found, 269.2491.
(S)-15-methylheptadecanoic acid (5): mp37.0ꢀC;
[ꢀ]20 +4.4ꢀ (c 2.66, CHCl3); HRMS (FAB): calcd.
D
for C18H35O2 ([M-H]ꢁ), 283.2637; found, 283.2646.
(S)-16-methyloctadecanoic acid (6): mp43.0ꢀC;
[ꢀ]20 +2.5ꢀ (c 1.69, CHCl3); HRMS (EI): calcd. for
D
C19H38O2 ([M]þ), 298.2872; found, 298.2873.
(S)-17-methylnonadecanoic acid (7): mp46.0ꢀC;
[ꢀ]20 +2.4ꢀ (c 2.07, CHCl3); HRMS (EI): calcd. for
D
C20H40O2 ([M]þ), 312.3028; found, 312.3026.
(S)-18-methylicosanoic acid (8): mp50.0ꢀC; [ꢀ]20
D
+3.9ꢀ (c 1.34, CHCl3); HRMS (FAB): calcd. for
C21H41O2 ([M-H]ꢁ), 325.3107; found, 325.3105.
(S)-19-methylhenicosanoic acid (9): mp58.0ꢀC;
[ꢀ]20 +4.1ꢀ (c 2.25, CHCl3); HRMS (EI): calcd. for
D
C22H44O2 ([M]þ), 340.3341; found, 340.3341.
(S)-20-methyldocosanoic acid (10): mp60.5ꢀC; [ꢀ]20
D
+4.1ꢀ (c 1.06, CHCl3); HRMS (FAB): calcd. for
C23H46O2 ([M]þ), 354.3498; found, 354.3499.
(S)-21-methyltricosanoic acid (11): mp66.0ꢀC; [ꢀ]20
D
+6.2ꢀ (c 0.459, CHCl3); HRMS (FAB): calcd. for
C24H48O2 ([M]þ), 368.3654; found, 358.3657.
(S)-22-methyltetracosanoic acid (12): mp69.0ꢀC;
[ꢀ]20 +4.3ꢀ (c 0.672, CHCl3); HRMS (FAB): calcd.
D
for C25H50O2 ([M]þ), 382.3811; found, 382.3816.
(S)-23-methylpentacosanoic acid (13): mp70.5ꢀC;
[ꢀ]20 +4.7ꢀ (c 0.473, CHCl3); HRMS (FAB): calcd.
D
for C26H52O2 ([M]þ), 396.3967; found, 396.3969.
(S)-24-methyhexacosanoic acid (14): mp72.0ꢀC;
[ꢀ]20 +4.5ꢀ (c 0.411, CHCl3); HRMS (FAB): calcd.
D
for C27H54O2 ([M]þ), 410.4124; found, 410.4123.
(S)-25-methylheptacosanoic acid (15): mp76.0ꢀC;
[ꢀ]20 +3.2ꢀ (c 0.563, CHCl3); HRMS (FAB): calcd.
D
for C28H56O2 ([M]þ), 424.4280; found, 424.4281.
(S)-26-methyloctacosanoic acid (16): mp77.5ꢀC;
[ꢀ]20 +4.3ꢀ (c 0.768, CHCl3); HRMS (FAB): calcd.
D
for C29H58O2 ([M]þ), 438.4437; found, 438.4442.
Equipment. The HPLC pump used was a Jasco PU-
980 equipped with a Rheodyne 7125 sample injector
with a 20-ꢁl sample loop. The fluorescence detector was
a Jasco FP-920 with a 5-ꢁl flow cell, and the integrator
was a Chromatocorder 12 (System Instrument, Tokyo,
Japan). Cryocool CC100-II (Neslab Instruments) was
used to control the column temperature.
Enantiometrically pure (S)-9 and 10 were respectively
prepared from 9-decenyl and 10-undecenyl magnesium
bromide and (S)-10-methyldodecyl tosylate which had

Chiral Discrimination of Branched-chain Fatty Acids
155
Table 1. Effects of Flow Rate and Column Temeperature on the ꢀ,
Rs, and Theoretical Plate Values of the (1R,2R)-1 Esters of (S) and
(R)-3
Sample preparation procedure. The S enantiomer
of each branched-chain fatty acid was respectively
esterified with (R,R) and (S,S)-1 by using 1-ethyl-3-
(3-dimethylaminopropyl)carbodiimide
hydrochloride
Flow rate Column Retention factor
(ml minꢁ1) temp. (ꢀC)
Theoretical
plate
ꢀ
Rs
k0R
k0S
(EDC) in the presence of 4-(dimethylamino)pyridine
(DMAP) in a mixture of toluene and acetonitrile (1:1,
v/v). To a solution of the fatty acid were added an
excess amount (more than 5 equimolar equiv) of either
enantiomer of reagent 1 and DMAP (more than 5
equimolar equir). After adding an excess amount of
EDC to the mixture, it was stirred at room temperature
for more than 10 h. An aliquot was loaded onto a silica
gel TLC plate (10 cm length, Silicagel 60 F254, Art-
5744, Merck, Darmstadt, Germany) and developed with
n-hexane/ethyl acetate (4:1, v/v). The target spot was
collected and packed in a Pasteur pipette which was
eluted with ethyl acetate/ethanol (4:1, v/v). After
evaporating the solvent, the residue was dissolved in
methanol and subjected to an HPLC analysis.
0.2
0.2
0.2
0.2
0.1
0.2
0.4
0.6
0.8
1.2
ꢁ30
ꢁ40
ꢁ50
ꢁ60
ꢁ50
ꢁ50
ꢁ50
ꢁ50
ꢁ50
ꢁ50
2.15
5.97
2.27 1.06 1.13
6.65 1.11 2.45
12100
9900
5600
3100
8100
5600
4100
3200
2300
1000
11.3
13.2
1.17 2.76
1.27 3.33
1.17 3.35
2.76
34.5
11.4
43.7
13.3
2.39
2.05
1.72
1.17
Column: Develosil ODS-3 (single).
Mobile phase: MeCN/THF/n-hexane (100:100:5, v/v/v).
acceptable resolution by the system. The addition of
THF to the mobile phase solvents greatly improved the
separation with significantly increasing ꢀ value. The use
of those solvents, in addition to the dramatic improve-
ment in the separation of the (1R,2R) and (1S,2S)-1
esters of (S)-2, (S)-3 and (S)-4, enabled the (1R,2R) and
(1S,2S)-1 esters of (S)-5, (S)-6 and (S)-7 to also be
respectively separated into two peaks. The addition of
THF also improved the reproducibility of the retention
time.
Throughout this report, the chromatographic data for
the (1S,2S)-1 esters of the (S)-acids are regarded as those
for the (1R,2R)-1 esters of the corresponding (R)-acids.
HPLC separation. The esters of the anteiso acids were
separated in an ODS column (Develosil ODS-3, 3 ꢁm,
4.6 mm I.D. ꢂ 150 mm) or C30 column (Develosil C30
UG-3, 3 ꢁm, 4.6 mm I.D. ꢂ 150 mm) (Nomura Chemi-
cal Co., Aichi, Japan). The fluorescence intensity of
each derivative was monitored (emission at 462 nm,
excitation at 298 nm). Mixtures of methanol/acetoni-
trile/THF/n-hexane were used as mobile phase solvents,
and the column temperature was kept in the rouge from
ꢁ50 to ꢁ10ꢀC. The optimized conditions for each
derivative are shown in Tables 2 and 3.
Table 1 shows the effects of the flow rate and column
temperature on the separation factor (ꢀ), resolution
factor (Rs) and theoretical plate value of the diastereo-
meric (1R,2R)-1 esters of (S) and (R)-3. The ꢀ values
increased as the column temperature was reduced.
However, the theoretical plate with values decreased
with decreasing column temperature, and significantly
increased with decreasing flow rate at low column
temperatures. This might be attributable to insufficient
diffusion of the solutes in the column under low column-
temperature conditions. Since the column efficiency
could be improved by reducing the flow rate, the mobile
Results and Discussion
HPLC separation of the anteiso acids in ODS
columns
Since the (1S,2S)-1 ester of an (S)-acid is an
enantiomer of the (1R,2R)-1 ester of the corresponding
(R)-acid, they must have the same chromatographic
characteristics in an achiral HPLC system. Therefore,
using the (S) enantiomer of an acid and (1S,2S)-1 made
it possible to obtain chromatographic data for the
(1R,2R)-1 ester of the (R)-acid.
In our previous report, HPLC separation of the
(1R,2R) and (1S,2S)-1 esters of (S)-anteiso acids having
chiral centers at the C6 to C12 positions was accom-
plished by using an ODS column eluted with methanol/
acetonitrile/n-hexane at a low column temperature.9) In
the case of (S)-13-methylpentadcanoic (3) and (S)-14-
methylhexadecanoic (4) acids, their diastereomeric
esters of (1R,2R) and (1S,2S)-1 could be separated into
two peaks by using this solvent system with acceptable
resolution (Rs) (2.40 and 1.42, respectively). However,
it was difficult to separate the (1R,2R) and (1S,2S)-1
esters of (S)-15-methylheptadecanic acid (5) with
phase was pumped at 0.1–0.4 ml minꢁ1
.
Table 2 shows the retention factor (k0), and ꢀ and Rs
values of the (1R,2R)-1 esters of (S) and (R)-2–8 under
the optimized conditions. The farther away the chiral
center was from the carboxyl group, the poorer the
obtained resolution. Separation of the (1R,2R)-1 esters
of (S) and (R)-8 was unsatisfactory (Rs value = 0.58)
even though four ODS columns were used in series and
a long time (18 h) was allowed. Therefore, the 17
position was the limit to practically discriminate the
chirality of the anteiso fatty acids by using the ODS
column by this method.
HPLC separation of the anteiso acids in the C30
column(s)
We suspected that a longer alkyl chain stationary
phase than that in the fatty acid would be required to
provide sufficient interaction between the stationary

156
K. AKASAKA and H. OHRUI
Table 2. Chromatographic Data for the Diastereomeric Separation of the (1R,2R)-1 Esters of (S) and (R)-2–8 in an ODS Column
Mobile phase of
MeCN/THF/n-hexane
(v/v/v)
Combination
Develosil
ODS-3
Anteiso
acid
Flow rate
(ml minꢁ1
Column
temp. (ꢀC)
Retention factor
k0S
ꢀ
Rs
)
k0R
2
3
4
5
6
7
8
100:100: 5
100:100: 5
100:100:15
100:100:10
100:100:10
100:100:10
150:200:28
0.4
0.1
single
single
single
single
single
single
quartet
ꢁ50
ꢁ50
ꢁ45
ꢁ40
ꢁ40
ꢁ38
ꢁ40
18.7
11.4
17.0
13.3
1.10
1.17
1.09
1.05
1.04
1.04
1.02
1.48
3.35
1.54
0.94
0.80
0.70
0.58
0.15
0.15
0.1
8.00
9.87
18.7
7.36
10.2
17.9
24.6
15.0
0.1
0.1
23.7
15.3
Table 3. Chromatographic Data for the Diastereomeric Separation of the (1R,2R)-1 Esters of (S) and (R)-4–16 in a C30 Column
Mobile phase of
MeCN/THF/n-hexane
(v/v/v)
Combination
Develosil
C30 UG-3
Anteiso
acid
Column
temp; (ꢀC)
Retention factor
ꢀ
Rs
k0R
k0S
4
5
150:200: 30
150:200: 30
150:200: 45
150:200: 80
150:200: 80
150:200: 80
150:200:100
150:250: 90
150:250:120
150:250: 90
150:250: 60
150:250: 25
150:250: 60
150:250: 25
150:250: 60
150:250: 90
150:250: 90
single
single
single
single
single
single
single
single
tandem
single
tandem
single
tandem
single
tandem
tandem
triple
ꢁ48
ꢁ48
ꢁ50
ꢁ50
ꢁ50
ꢁ45
ꢁ45
ꢁ30
ꢁ40
ꢁ25
ꢁ30
ꢁ15
ꢁ20
ꢁ15
ꢁ20
ꢁ20
ꢁ15
8.75
12.3
14.1
5.67
8.05
13.0
13.0
6.10
9.61
13.5
13.9
7.85
10.29
7.06
6.69
22.0
12.4
34.0
19.6
21.1
18.3
1.09
1.06
1.09
1.08
1.08
1.08
1.10
1.08
1.11
1.06
1.07
1.05
1.05
1.05
1.07
1.07
1.05
1.81
1.26
1.95
1.42
1.43
1.16
1.22
1.45
1.68
1.12
1.30
1.00
1.22
0.882
1.06
0.771
0.56
6
7
8
10.3
12.5
15.2
9
10
11
7.27
9.31
7.50
7.17
12
13
14
20.9
11.8
36.8
20.9
19.7
19.2
15
16
Flow rate: 0.1 ml minꢁ1
.
phase and the esters. We therefore tested a C-30 column,
which has a triacontylsilane group as a stationary phase
modifier, for the longer anteiso fatty acids. As shown in
Table 3, separation of the (1R,2R)-1 esters of (S) and
(R)-4–8 was much better with this column than by the
ODS column.
Table 4 shows the effects of the flow rate and column
temperature on the ꢀ, Rs and theoretical plate values of
the diastereomeric 1 esters from 11. The lower column
temperature increased the ꢀ value, but the theoretical
plate values were significantly decreased by lowering
the column temperature. Moreover, the theoretical plate
values were also decreased significantly with increasing
alkyl chain length of the anteiso acid (Fig. 2). These
effects might be partly attributable to a reduction in the
solubility and mobility of the derivatives in both the
mobile phase and the stationary phase. It was therefore
necessary to control the column temperature as well as
the flow rate to get optimum resolution.
Table 4. Effects of Flow Rate and Column Temperature on the ꢀ, Rs,
and Theoretical Plate Values of the (1R,2R)-1 Esters of (S) and (R)-11
Flow rate Column Ratention factor
(ml minꢁ1) temp. (ꢀC)
Theoretical
plate
ꢀ
Rs
k0R
k0S
0.1
0.1
0.1
0.1
0.1
0.2
0.4
0.6
0.8
1.0
ꢁ20
ꢁ30
ꢁ40
ꢁ50
ꢁ30
ꢁ30
ꢁ30
ꢁ30
ꢁ30
ꢁ30
1.49
3.65
9.54
1.55 1.04 0.650
3.91 1.07 1.26
9100
6000
2900
1300
6000
4200
2700
1800
1600
1400
10.5
25.5
1.11 1.30
1.15 1.28
22.2
3.65
3.91 1.07 1.26
0.966
0.787
0.620
0.505
0.527
Column: Develosil C30 UG-3(single).
Mobile phase: MeCN/THF/n-Hexane (150:250:120, v/v/v).
resolution with a single column. By using tandem
columns, the esters of (S) and (R)-13 and 14 were also
separated with Rs values of more than 1.0. Moreover,
the esters of (S) and (R)-15 and 16 were also separated
by using tandem and triple columns in series, respec-
tively, although their resolution was not sufficient.
Table 3 also shows the k0, ꢀ and Rs values for the
(1R,2R)-1 esters of (S) and (R)-8–16 under the optimized
conditions for each case. Each diastereomeric 1 ester of
(S) and (R)-8–12 were separated with acceptable

Chiral Discrimination of Branched-chain Fatty Acids
157
group from the 2 to 11 positions.9) The results of the
NMR study suggest that the turning position of the rule
almost agrees with the border position where the
anisotropic effects of an anthracenedicarboxyimide
group worked or not. This means that the branched
methyl groups up to C11 position were located over the
imide and those of acids 2–16 were not. These results
might suggest that different mechanisms work for
discriminating the chirality of a methyl branch nearer
or farther away than the C11 position from the carbonyl
group.
It will be possible by using the present rule to predict
the absolute configuration of methyl-branched acids
without authentic samples. Since the (1S,2S)-1 ester of
one enantiometrically pure acid can be assumed to be
the equivalent of the (1R,2R)-1-ester of the enantiomer
of the acid from achiral HPLC, it is possible to estimate
the elution order of the sample acid by converting with
both enantiomers of the reagents.
Although it has been thought impossible to discrim-
inate the chirality far from where the functional group
could be tagged, reagent 1 has made it possible to
separate even those anteiso acids having 24 methylene
spacers between the stereogenic center and the carboxyl
group. We initially considered that direct diastereomeric
interaction between the stereogenic center of the acids
and the anthraceneimide ring of the reagent would make
it possible to discriminate the remote chirality of the
acid. However, the C26 position of the (S)-16 esters of
(1R,2R)- and (1S,2S)-1 is too far from the anthracene-
imide ring for direct interaction with each other. Since
the longer alkyl chain of the stationary phase of the
column proved very effective for the chiral separation of
longer anteiso acids, the stationary phase played a very
important role in remote chiral discrimination.
Fig. 2. Effect of Column Temperature on the Theoretical Plate
Values for the (1R,2R)-1 Esters of(S)-9–15.
Column: Develosil C30-UG-3; flow rate: 0.1 ml minꢁ1; mobile
phase: MeCN/THF/n-hexane (150:250:120).
acid 11, acid 12, acid 13, acid 14,
acid 9,
acid 15.
acid 10,
Figure 3 shows HPLC chromatograms of the (1R,2R)-1
esters of (S) and (R)-10–16 as representatives.
Elution order of the diastereomers
We found an empirical rule in the elution order of the
(1R,2R)-1 esters of (S)- and (R)-2–16 regardless of the
column used that is, all the (1R,2R)-1 esters of the (R)-
branched fatty acids having a methyl group at an odd-
number position were eluted faster than the correspond-
ing (S)-branched fatty acid esters, while the (S)-
branched fatty acid esters having a methyl group at an
even-number position were eluted faster than the
corresponding (R)-branched fatty acid esters, and this
rule is the reverse of that previously reported for
branched-chain fatty acids having the branched methyl
In addition to this surprising chiral discrimination
ability, this reagent made it also possible to detect acid
derivatives with very high sensitivity (10ꢁ15 mol level).
We have recently reported the absolute configurations of
Fig. 3. Chromatograms of the (1R,2R)-1 Esters of Acids (S) and (R)-11, 12, 13, 14, 15 and 16.
HPLC conditions are shown in Table 3. The diastereomeric esters of (S) And (R)-11–14 were separated in a single column, and those of (S)
and (R)-15 and 16 were separated in tandem and triple columns, respectively.

158
K. AKASAKA and H. OHRUI
6853–6856 (1997).
two anteiso fatty acid moieties of glycoglycerolipid
S365A13) produced by Corynebacterium aquaticum by
applying the present method.16) Moreover, this reagent
also made it possible to separate fatty acid enantiomers
having a chiral cyclopropane structure, and the method
was successfully applied to determine the absolute
configuration of plakoside A.17) In conclusion, this
reagent should be a powerful tool for determining the
absolute configuration of many natural products having
branched alkyl chains.
7) Akasaka, K., Ohrui, H., Meguro, H., and Umetsu, T.,
Highly sensitive isomeric determination of beraprost
sodium in plasma using a fluorescent chiral derivatiza-
tion reagent. Anal. Sci., 13 (Supplement), 461–466
(1997).
8) Akasaka, K., Imaizumi, K., and Ohrui, H., Enantiomeric
separation of branched fatty acids having chiral centers
remote from the carboxy group by labeling with chiral
fluorescent derivatization reagents. Enantiomer, 3, 169–
174 (1998).
9) Akasaka, K., and Ohrui, H., Enantiomeric separation of
branchecd fatty acids after conversion with trans-2-(2,3-
anthracenedicarboximide)cyclohexanol, a highly sensi-
tive chiral fluorescent conversion reagent. Biosci. Bio-
technol. Biochem., 63, 1209–1215 (1999).
10) Akasaka, K., Shichijyukari, S., Matsuoka, S., Murata,
M., Meguro,H., and Ohrui,H., Absolute configuration of
a ceramide with a novel branched-chain fatty acid
isolated from the epiphytic dinoflagellate, Coolia mono-
tis. Biosci. Biotechnol. Biochem., 64, 1842–1846 (2000).
11) Ratnayake, W. M. N., Olsson, B., and Ackman, R. G.,
Novel branched fatty acids in certain fish oils. Lipids, 24,
630–637 (1989).
12) Spitzer, V., Screening analysis of unknown seed oils.
Fett/Lipid, 101, 2–19 (1999).
13) Yanagi, H., Matsufuji, M., Nakata, K., Nagamatsu, Y.,
Ohta, S., and Yoshimoto, A., A new type of glycogly-
cerolipid from Corynebacterium aquaticum. Biosci.
Biotechnol. Biochem., 64, 424–427 (2000).
Acknowledgment
This work was supported in part by grant-in-aid for
scientific research from the Ministry of Education,
Culture, Sports, Science and Technology of Japan.
References
1) Buckner, J. S., Kolattukedy, P. E., and Rogers, E.,
Synthesis of multimethyl-branched fatty acids by avian
and mammalian fatty-acids synthetase and its regulation
by malonyl-CoA decarboxylase in uropygial gland.
Arch. Biochem. Biophys., 186, 152–163 (1978).
2) Costantino, V., Fattorisso, E., and Mangoni, A., Isolation
of 5-membered cyclitol glycolipids, crasserides-unique
glyceride from the sponge Pesudoceratina crassa. J.
Org. Chem., 58, 186–191 (1993).
3) Mori, K., Harada, H., Zagatti, P., Cork, A., and Hall, D.
R., Pheromone synthesis. 126. Synthesis and biological-
activity of 4 stereoisomer of 6,10,14-trimethyl-2-penta-
decanol, the female-produced sex pheromone of rice
moth (Corcyra cephalonica). Leibigs Ann. Chem., 259–
267 (1991).
4) Yamazaki, T., Seyama, Y., Otsuka, H., Ogawa, H., and
Yamakawa, T., Identification of alkyldiacylglycerols
containing saturated methyl branched chains in the
harderian gland of guinea pig. J. Biochem., 89, 683–691
(1981).
5) Hall, D. R., Cork, A., Laster, R., Nesbitt, B. F., and
Zagatti, P., Sex pheromones of rice moth, Corcyra
cephalonica stainton. 2. Identification and role of female
pheromone. J. Chem. Ecol., 13, 1575–1589 (1987).
6) Akasaka, K., Meguro, H., and Ohrui, H., Enantiomeric
separation of carboxylic acids having chiral centers
remote from the carboxyl group by labeling with a chiral
fluorescent derivatization reagent. Tetrahedron Lett., 38,
14) Tanaka, I., Matsuoka, S., Murata, M., and Tachibana, K.,
A new ceramide with a novel branched-chain fatty acid
isolated from the epiphytic dinoflagellate, Coolia mono-
tis. J. Natural Products, 61, 685–688 (1998).
15) Tofern, B., Mann, P., Kaloga, M., Jenett-Siems, K.,
Witte, L., and Eich, E., Aliphatic pyrrolidine amides
from two tropical convolvulaceous species. Phytochem-
istry, 52, 1437–1441 (1999).
16) Akasaka, K., Shichijyukari, S., Meguro, H., and Ohrui,
H., Determination of the absolute configuration of the
anteiso acid moieties of glycoglycerolipid S365A iso-
lated from Corynebacterium aquaticum. Biosci. Biotech-
nol. Biochem., 66, 1719–1722 (2002).
17) Mori, K., Tashiro, T., Akasaka, K., Ohrui, H., and
Fattorusso, E., Determination of the absolute configu-
ration at the two cyclopropane moieties of plakoside A,
an immunosuppressive marine galactosphingolipid. Tet-
rahedron Lett., 43, 3719–3722 (2002).