Strategies for the analysis of chlorinated lipids in biological systems

Article abstract of DOI:10.1016/j.freeradbiomed.2012.06.013

Myeloperoxidase-derived HOCl reacts with the vinyl ether bond of plasmalogens yielding α-chlorofatty aldehydes. These chlorinated aldehydes can be purified using thin-layer chromatography, which is essential for subsequent analysis of extracts from some t

Full text of DOI:10.1016/j.freeradbiomed.2012.06.013

Free Radical Biology and Medicine 59 (2013) 92–99
Contents lists available at SciVerse ScienceDirect
Free Radical Biology and Medicine
journal homepage: www.elsevier.com/locate/freeradbiomed
Methods in Free Radical Biology and Medicine
Strategies for the analysis of chlorinated lipids in biological systems
Bradley K. Wacker, Carolyn J. Albert, Benjamin A. Ford, David A. Ford n
Department of Biochemistry and Molecular Biology, School of Medicine, and Center for Cardiovascular Research, Saint Louis University, St. Louis, MO 63104,
USA
a r t i c l e i n f o
a b s t r a c t
Available online 17 June 2012
Myeloperoxidase-derived HOCl reacts with the vinyl ether bond of plasmalogens yielding
a-chlorofatty
aldehydes. These chlorinated aldehydes can be purified using thin-layer chromatography, which is essential
for subsequent analysis of extracts from some tissues such as myocardium. The -chlorofatty aldehyde
Keywords:
a
Myeloperoxidase
Hypochlorous acid
Fatty aldehydes
Fatty acids
2-chlorohexadecanal (2-ClHDA) is quantified after conversion to its pentafluorobenzyl oxime derivative
using gas chromatography–mass spectrometry and negative-ion chemical ionization detection. 2-ClHDA
accumulates in activated human neutrophils and monocytes, as well as in atherosclerotic lesions and
infarcted myocardium. Metabolites of 2-ClHDA have also been identified, including the oxidation product,
2-chlorohexadecanoic acid (2-ClHA), and the reduction product, 2-chlorohexadecanol. 2-ClHA can be
Free radicals
quantified using LC–MS with selected reaction monitoring (SRM) detection. 2-ClHA can be
o-oxidized by
hepatocytes and subsequently -oxidized from the -end, leading to the production of the dicarboxylic
b
o
acid, 2-chloroadipic acid. This dicarboxylic acid is excreted in the urine and can also be quantified using
LC–MS methods with SRM detection. Quantitative analyses of these novel chlorinated lipids are essential to
identify the role of these lipids in leukocyte-mediated injury and disease.
& 2012 Elsevier Inc. All rights reserved.
Introduction
a
-chlorofatty aldehydes (
The -ClFALD can be further metabolized to
-ClFA) and -chlorofatty alcohol ( -ClFOH). Oxidation of the
aldehyde to the -ClFA metabolite is catalyzed by a fatty aldehyde
dehydrogenase [11]. -ClFA can be further catabolized by -oxida-
tion, which is initiated by an -hydroxylation step, followed by
conversion of the intermediate to an -chlorodicarboxylic acid.
Sequential -oxidation from the -end of the dicarboxylic acids
a
-ClFALD) and lysophospholipids (Fig. 1).
a
a-chlorofatty acid
Inflammatory reactions mediated by myeloperoxidase (MPO)-
enriched phagocytes including neutrophils, monocytes, and some
macrophages contribute to the pathogenesis of atherosclerosis,
ischemic/reperfusion injury to numerous tissues, and other disor-
ders. MPO released from activated phagocytes leads to the forma-
tion of the reactive chlorinating species HOCl, via the peroxidation of
chloride ions [1,2]. The chemical reactivity of HOCl gives it potent
antimicrobial and cytotoxic properties, which are important for the
immune response, but also has the potential to mediate pathogen-
esis in many diseases and disorders [3,4]. HOCl is a two-electron
oxidant of proteins and lipids. Lipids can potentially be chlorinated
by HOCl at amine, alkene, and vinyl ether functional groups.
Plasmalogens are preferentially targeted by HOCl because of the
relatively high rate constant for reactions with the vinyl ether bond
compared to other reactive targets including alkenes [5]. Plasmalo-
gens are abundant phospholipid components in many cell types,
including endothelial cells, macrophages, neutrophils, smooth mus-
cle cells, cardiac myocytes, neurons, and glia [6–8]. Although
plasmalogens have been described as having antioxidant capabilities
[9,10], their reactivity with HOCl leads to the production of reaction
products that have been associated with cardiovascular disease.
Plasmalogens react with HOCl leading to the production of
(a
a
a
a
a
o
o
a
b
o
(through multiple dicarboxylic acid intermediates) leads to the
eventual production of 2-chloroadipic acid (2-ClAdA). The in vivo
metabolism of a-ClFA to 2-ClAdA is efficient, with the final product
2-ClAdA being excreted in the urine [12].
Chlorinated lipids have been identified under several pathophy-
siological and physiological conditions involving MPO-laden phago-
cytes. Activated neutrophils and monocytes transiently accumulate
a
-ClFALD [13,14]. Human aortic atherosclerotic plaques show a
nearly 1400-fold increase in levels of the -ClFALD, 2-chlorohex-
adecanal (2-ClHDA), compared to normal aortic tissue [15]. Addi-
tionally, 2-ClHDA acts as chemoattractant to neutrophils,
a
a
suggesting that it may play a role in the recruitment of neutrophils
to sites of inflammation [14]. Both unsaturated molecular species of
lysophosphatidylcholine (the accompanying product from the
degradation of plasmalogens by HOCl) and lysophosphatidylcho-
line-chlorohydrin, a product of HOCl attack of alkenes present in the
sn-2 aliphatic chain of unsaturated lysophosphatidylcholine mole-
cular species, induce P-selectin surface expression on human
coronary artery endothelial cells [16]. Endothelial cells show inhibi-
tion of endothelial nitric oxide synthase (eNOS) expression caused
n Corresponding author.
E-mail address: fordda@slu.edu (D.A. Ford).
0891-5849/$ - see front matter & 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.freeradbiomed.2012.06.013

B.K. Wacker et al. / Free Radical Biology and Medicine 59 (2013) 92–99
93
Fig. 1. Structures of the members of the chlorinated lipidome derived from HOCl oxidation of plasmalogens and abbreviated strategies for their quantification. PX, LPX, PFB-HA,
and PFB-Cl are the abbreviations for phospho-head group (e.g., phosphorylcholine), lysophospholipids, pentafluorobenzyl hydroxylamine, and pentafluorobenzoyl chloride.
Fig. 2. Flowchart of strategies to quantify members of the chlorinated lipidome.
by 2-ClHDA, as well as reduced eNOS expression localized to the
plasma membrane [17]. 2-ClHDA has been shown to increase in
infarcted myocardial tissue and can depress contractile function
[18]. Furthermore, both 2-ClHDA and 2-chlorohexadecanoic acid
(2-ClHA) induce COX-2 expression in human coronary artery
endothelial cells [19]. Thus, it is clear that chlorinated lipids
accumulate in an array of cardiovascular pathophysiological condi-
tions, and the accurate measurement of these novel lipids is
essential and may lead to further insights into the role of these
compounds either as putative biomarkers or mediators of proin-
flammatory mechanisms.
Several mass spectrometry-based methods have been devel-
oped to detect and accurately quantify these chlorinated lipids
(see Fig. 2 for flowchart of these quantification strategies and
Fig. 1 for the structures that are quantified). The choices of
(1) whether to derivatize the chlorinated lipids, (2) the type of
chromatography, and (3) the mass spectrometry (MS) scan mode
are dependent on the functional groups of the analytes that
dictate their chromatographic characteristics and ionization by
mass spectrometry. Accordingly, in this review of the methodo-
logical strategies employed to measure chlorinated lipids, the
quantification of (1)
(PFB) oximes using gas chromatography (GC)–MS with negative
ion chemical ionization (NICI), (2) -ClFA by reversed-phase
a-chlorofatty aldehydes as pentafluorobenzyl
a
liquid chromatography (LC) with electrospray ionization (ESI)–
MS and selected reaction monitoring (SRM) for detection, and
(3) 2-ClAdA by reversed-phase LC with ESI–MS and SRM for
detection will be detailed. For this review, we focus on the
accurate measurement of three chlorinated lipids: the
a-ClFALD
2-ClHDA, the -ClFA 2-ClHA, and the -Cl-dicarboxylic acid
a
a
2-ClAdA. These are the predominant chlorinated lipids that have
been detected in vivo. Methods to quantify the less predominant

94
B.K. Wacker et al. / Free Radical Biology and Medicine 59 (2013) 92–99
chlorinated lipids,
chlorohydrins, are available [11,16]. It should be appreciated that
other molecular species of -ClFALD and -ClFA, including 2-chloro-
a
-ClFOH (see Fig. 1) and lysophosphatidylcholine
Advantage A10 system with polishing with a Q-POD system (EMD
Millipore, Billerica, MA, USA). [d4-3,3,4,4]-Adipic acid was from
C/D/N Isotopes (Pointe-Claire, QC, Canada, Product No. D3269)
a
a
octadecanal and 2-chloro-octadecanoic acid, can also be quantified
by the methods described herein.
Instrumentation
Principles
GC–MS was performed using a DB-1 column (12.5 m, 0.2-mm
inner diameter, 0.33-mm methyl silicone film coating; Agilent, Santa
Analysis of
a-chlorofatty aldehydes
Clara, CA, USA, Product No. 128-1012) with an Agilent 6890 gas
chromatograph and Agilent 5973 mass spectrometer with chemical
ionization source (Agilent). GC–MS data analysis was performed
using enhanced Chemstation software (Agilent). LC–MS was
performed using the indicated columns (vide infra) with a Thermo
Fisher Surveyor MS LC system and Thermo Fisher TSQ Quantum
Ultra mass spectrometer (Thermo Fisher, Waltham, MA, USA).
LC–MS data analysis was performed using XCalibur software
(Thermo Fisher).
a-ClFALD are the products of chlorination of the sn-1 aliphatic
chain of plasmalogens. Chlorinated fatty aldehydes form poor
ions and are unstable under the conditions of ESI. Thus, deriva-
tives that add an electron-withdrawing group, such as the PFB
oxime derivative of the fatty aldehyde, greatly improve the sensi-
tivity for
a-ClFALD detection and quantification [20]. However, it
should be recognized that the presence of masked aldehydes in
plasmalogens, and other species that may be modified under the
acidic PFB derivatization process, can be problematic. Accordingly,
Protocols
isolation of
derivatization can remove competing lipids (that might react with
PFB hydroxylamine), while enriching -ClFALD that can then be
facilely converted to its PFB derivative. The PFB oxime of -ClFALD
a-ClFALD using thin-layer chromatography (TLC) before
Standards
a
a
For 2-ClHDA analysis, the internal standard used is 2-chloro-
[d4-7,7,8,8]-hexadecanal (20 pmol per 50 mg heart tissue or
500,000 neutrophils extracted). See Thukkani et al. [14] for
synthesis. For 2-ClHA analysis, the internal standard used is
2-chloro-[d4-7,7,8,8]-hexadecanoic acid. See Wildsmith et al.
[22] for synthesis. For 2-ClAdA analysis, the internal standard
used is [d4-3,3,4,4]-adipic acid.
can then be quantified using GC–MS by comparisons to stable
isotope-labeled internal standards using NICI, which exploits the
loss of HF from the PFB derivative to form excellent negative ions.
Analysis of a-chlorofatty acid
a-ClFALD can be further oxidized to an
a-ClFA. The carboxylic
acid functional group of -ClFA forms excellent negatively
a
charged ions that can be readily detected using ESI–MS.
Reversed-phase high-performance liquid chromatography (HPLC)
using an ammonium acetate buffer to neutralize the charge of the
acid and enhance its binding efficiency to the column can be used
to chromatographically resolve fatty acid molecular species. With
MS/MS instrumentation, quantification by SRM of the loss of HCl
Tissue lipid extraction for 2-ClHDA and 2-ClHA analyses
Homogenize or pulverize frozen tissue in a manner proper for
the tissue of interest. Homogenize soft tissue (liver, nervous
tissue, etc.) in ice-cold saline (0.154 M NaCl, 100 ml per 10 mg
tissue wet wt) using a hand-held battery-powered pestle with a
microcentrifuge tube. Tougher tissue can be frozen and pulver-
ized at the temperature of liquid N2 using a stainless steel mortar
and pestle. Extract the lipids from 10–50 mg of homogenized/
pulverized tissue using a modification of the Bligh and Dyer
method [23]. Quickly place the homogenized/pulverized tissue
into a chilled 16 Â 100-mm glass culture tube and add 3.5 ml of a
methanol/saline (2.5/1, v/v) solution followed by vortex mixing.
It is important to add methanol while the tissue is frozen to
minimize metabolism. It is critical to avoid exposure to plastic
with samples in organic solvent during any step. Add 1.25 ml of
HPLC-grade chloroform and vortex to mix. Add internal standard
from a-ClFA enables outstanding sensitivity for detection, even in
highly complex mixtures [21].
Analysis of 2-chloroadipic acid
o
-Oxidation of the
a-ClFA and subsequent sequential b-oxidation
from the -end with the production of multiple dicarboxylic acid
intermediates lead to the production of 2-ClAdA. This metabolite can
also be analyzed by LC–MS/MS using principles similar to those that
were described above for carboxylic acids (e.g.,
separated on a double-bonded column, again using an ammonium
acetate buffer to enhance binding efficiency to the column. As with
o
a-ClFA). 2-ClAdA is
in 100 ml of methanol and vortex. Incubate tissue extract at room
temperature for 5 min with occasional vortexing. Add 1.25 ml of
chloroform and 1 ml of saline to achieve a 1/1/0.8 ratio of
methanol/chloroform/saline and vortex to mix. Centrifuge the
tube at 500g for 2 min. Remove the lower chloroform layer, being
careful not to disturb the protein disc, and add to a clean glass
16 Â 125-mm screwtop tube. To the remaining upper phase, add
2.5 ml of chloroform and then vortex to mix well. Centrifuge the
tube at 500g for 2 min. Remove the lower chloroform layer again
and combine with the previous. Add 4 ml of saline to the
combined chloroform extract and vortex at least three times,
waiting 1 min between each vortexing. The lower chloroform
layer should become clearer. Centrifuge at 500g for 2 min.
Remove the lower chloroform layer into a clean glass screwtop
test tube. Evaporate the chloroform under a stream of nitrogen.
Resuspend the dried extract in 0.5 ml of chloroform, cap under
nitrogen, and store at À20 1C until further analysis.
the a-ClFA, MS/MS analysis is performed using SRM of the loss of HCl
to obtain excellent sensitivity.
Materials
Lipid extracts were prepared in borosilicate tubes from tissues
and stored under nitrogen in borosilicate glass tubes with screwtops
and polytetrafluoroethylene septa inserts. Burdick and Jackson
methanol and chloroform were of HPLC grade and were purchased
from Honeywell (Morristown, NJ, USA). Formic acid, pentafluoro-
benzyl hydroxylamine (Product No. P4190), and ammonium acetate
were purchased from Sigma Chemical (St. Louis, MO, USA). Petro-
leum ether, ethyl ether, acetic acid, cyclohexane, and ethyl acetate
were purchased from Fisher Scientific (Pittsburgh, PA, USA). All
water used in these studies was purified through a Millipore Milli-Q

B.K. Wacker et al. / Free Radical Biology and Medicine 59 (2013) 92–99
95
Lipid extraction from cultured cells for 2-ClHDA and 2-ClHA analyses
Plasma lipid extraction for esterified 2-ClHA
For a 60-mm culture dish, scrape cells into 1 ml of ice-cold
saline (0.154 M NaCl). Transfer this to a 2-ml plastic tube, and
then scrape the culture dish again with 1 ml of ice-cold saline.
Combine this with the solution from the first scrape. Keep cells on
ice, or freeze at À80 1C until ready to perform lipid extraction. For
analysis, 1.5 ml (75% of total) of cell culture homogenate or cell
culture medium is extracted by a modification of the method of
Bligh and Dyer [23] with the addition of the appropriate internal
standard. To a 16 Â 100-mm culture tube, add 2.5 ml of HPLC-
grade methanol, 1.25 ml of HPLC-grade chloroform, and 1.5 ml of
cell homogenate or culture medium. Be careful not to allow any
plastic to touch the organic solvent during any step. Add appro-
Add 25
standard. Add 155
(1 M), vortex, and cap under nitrogen. Incubate the sample in a
water bath at 60 1C for 2 h. Remove from the water bath and
m
l of plasma to a test tube containing dried internal
m
l of water and vortex. Add 200 l of NaOH
m
allow the sample to cool to room temperature. Add 120 ml of cold
HCl (2 M), vortex, and incubate the solution at room temperature
for at least 10 min. Add 2.5 ml of Dole reagent (made fresh;
isopropanol/heptane/1 M H2SO4, 40/10/1, v/v/v), then add 1.5 ml
of heptane and 1 ml of water. Vortex the solution and then
centrifuge at 500g for 2 min. Collect the upper phase into a clean
16 Â 100-mm screwtop tube. Reextract the lower phase with 2 ml
of heptane, vortex, and centrifuge at 500g for 2 min. Collect the
upper phase, and combine it in a glass tube with the previously
collected upper phase. Dry the heptane extracts under N2 and
priate standards in 100 ml of methanol. Vortex and then incubate
the cell culture extract at room temperature for 5 min with
occasional vortexing. Add 0.5 ml of saline and 1.25 ml of chloro-
form and vortex to mix. Centrifuge the tube at 500g for 2 min.
Remove the lower chloroform layer, being careful not to disturb
the protein disc, and transfer to a clean glass 16 Â 125-mm
screwtop tube. Reextract the remaining upper phase by adding
2.5 ml of chloroform and vortexing. Centrifuge the tube at 500g
for 2 min, remove the lower chloroform layer again, and combine
this fraction with the previous. Add 4 ml of saline to the combined
chloroform extract and vortex three times, waiting 2 min
between each. The lower chloroform layer should become clearer.
Centrifuge at 500g for 2 min. Remove the lower chloroform layer
into a clean glass screwtop test tube. Evaporate the chloroform
under a stream of nitrogen. For 2-ClHDA analysis, resuspend the
resuspend in 150 ml of methanol/water (85/15, v/v) containing
0.1% formic acid. Vortex vigorously, and transfer the sample to an
autosampler vial with an insert.
TLC purification of 2-ClHDA and other
a-ClFALD from tissue extracts
2-ClHDA from crude lipid extract suspended in chloroform is
purified by TLC on a silica gel 60-A plate using a mobile phase
˚
comprised of petroleum ether/ethyl ether/acetic acid (90/10/1, v/v/
v). Scrape and extract the silica region corresponding to 2-ClHDA
(relative migration 0.46). To the scraped silica, add 4 ml of metha-
nol/chloroform (1/1, v/v). Vortex and then centrifuge for 5 min at
500g. Transfer the supernatant to an 18 Â 150-mm tube. Reextract
the silica with 0.8 ml of saline and 3 ml of methanol/chloroform
(2/1, v/v). Vortex and then centrifuge for 5 min at 500g. Combine the
supernatant with the previous in the 18 Â 150-mm tube. Add 1 ml
of chloroform and 2.4 ml of saline to the combined supernatants to
bring the solution to a 1/1/0.8 ratio of methanol/chloroform/saline.
Vortex and then centrifuge for 2 min at 500g. Remove the lower
chloroform layer to a clean 16 Â 100-mm tube. Reextract the upper
phase with 4 ml of chloroform. Vortex and then centrifuge for 2 min
at 500g. Remove the lower chloroform layer and combine with the
previous fraction. Evaporate the chloroform under N2, and resus-
dried extract in 300
ml of ethanol for PFB derivatization. For free
2-ClHA analysis, resuspend the dried extract in 200
ml of metha-
nol/water (85/15, v/v) containing 0.1% formic acid. Vortex vigor-
ously and transfer to an autosampler tube with an insert for
analysis by LC–MS.
Hydrolysis of esterified 2-ClHA from tissue and cell culture
lipid extracts
pend the purified 2-ClHDA in 300
tization. Other long-chain -ClFALD, including 2-chloro-octadecanal,
copurify with 2-ClHDA using this TLC procedure.
ml of ethanol before PFB deriva-
After the modified Bligh and Dyer extraction, dry the lipid
a
extract under nitrogen and then resuspend in 180
To initiate hydrolysis, add 200 l of 1 M NaOH, cap the tube under
nitrogen, and incubate the sample for 2 h at 60 1C. After incuba-
tion, cool the reaction to room temperature and add 120 l of ice-
ml of water.
m
PFB derivatization of 2-ClHDA and other a-ClFALD
m
cold 2 M HCl. Allow the reaction to sit for 10 min at room
temperature, and then perform a modified Dole extraction [24]
(see Plasma lipid extraction for free 2-ClHA, below).
Initiate derivatization of 2-ClHDA to its PFB oxime by resuspend-
ing the extraction products in a 16 Â 100-mm glass culture tube in
300 ml of ethanol and add 300 ml of 6 mg/ml pentafluorobenzyl
hydroxylamine in water. Vortex the water/ethanol mixture for
5 min at room temperature and then incubate for an additional
25 min at room temperature. Terminate the reaction by adding
1.2 ml of Milli-Q water followed by 2 ml of cyclohexane/ethyl ether
(4/1, v/v) and subsequent vortex mixing. Then centrifuge the
samples at 400g for 2 min, and subsequently collect the upper
phase into a clean 16 Â 100-mm glass culture tube. To the remaining
lower phase, add 2 ml of cyclohexane/ethyl ether (4/1, v/v) followed
by vortex mixing. After centrifugation at 400g for 2 min collect the
upper phase and combine it with the previously collected upper
phase. Then dry these pooled organic extracts under a nitrogen
Plasma lipid extraction for free 2-ClHA
Extract plasma using a modified Dole extraction [24] in the
presence of a standard. Add 25
glass culture tube containing the dried internal standard. Add
475 l of saline and vortex. Add 2.5 ml of Dole reagent (made
ml of plasma to a 16 Â 100-mm
m
fresh; isopropanol/heptane/1 M H2SO4, 40/10/1, v/v/v), 1.5 ml of
heptane, and 1 ml of water, in order. Vortex, then centrifuge at
500g for 2 min. Collect the upper phase into a clean 16 Â 100-mm
screwtop tube. Reextract the lower phase with 2 ml of heptane,
vortex, and centrifuge at 500g for 2 min. Collect the upper phase,
and combine it in a glass tube with the previously collected upper
stream. To the dried samples, add 150
the dried sample, and then transfer it with a Pasteur pipette to an
autosampler vial containing an insert. Add another 150 l of
ml of petroleum ether, vortex
m
phase. Dry the heptane extracts under N2 and resuspend in 150
m
l
petroleum ether to the culture tube, vortex the remaining residue,
and transfer and pool it with the sample already transferred to the
autosampler vial insert. Dry the pooled petroleum ether containing
derivatized product under nitrogen and then dissolve the sample in
of methanol/water (85/15, v/v) containing 0.1% formic acid.
Vortex vigorously and transfer the sample to an autosampler vial
with an insert.

96
B.K. Wacker et al. / Free Radical Biology and Medicine 59 (2013) 92–99
50
m
l of petroleum ether and cap it until injection onto the GC
(1/1, v/v). Evaporate the organic extract under N2 and resuspend
column.
in 0.2 ml water containing 5 mM ammonium acetate and 0.25%
acetic acid and then transfer it to an autosampler vial containing
an insert. Also, for the analysis of urine it is critical to normalize to
urinary creatinine to compensate for disparate urine product
dilution due to variable hydration states in the tested subjects.
Analysis of 2-ClHDA PFB oxime derivative
PFB oximes of
a-ClFALD including 2-ClHDA are analyzed by
GC–MS on a DB-1 column utilizing an Agilent 6890 gas chroma-
tograph. Keep the injector and the transfer lines at 250 and
280 1C, respectively. Keep the GC oven at 150 1C for 3.5 min and
increase the temperature at a rate of 25 1C/min to 310 1C. Hold the
oven temperature at 310 1C for an additional 5 min. All spectra
are acquired on an Agilent 5973 mass spectrometer that is
operated in the NICI mode with methane as the reagent gas and
helium as the carrier gas. Set the source temperature at 150 1C.
The electron energy should be 170 eV and the emission current
49 mA. Use selected ion monitoring (SIM) m/z 288 to detect the
2-ClHDA (PFB oxime) and SIM m/z 292 to detect the internal
standard, 2-chloro-[7,7,8,8-d4] hexadecanal (PFB oxime) (see
Fig. 1 for the structure of this derivative and its fragmentation
leading to ions m/z 288 and 292). The product 2-chloro-octade-
canal (PFB oxime) can be detected by SIM m/z 316.
Analysis of 2-ClAdA
The urine dicarboxylic acid extract is analyzed by LC–MS/MS
using a modification of the method described by Kakola and Alen
[25]. LC–MS/MS is performed utilizing a double-bonded column
(Supelcosil LC-18 DB 150 Â 3 mm, 5
mm). The mobile phases are
as follows: (A) water containing 5 mM ammonium acetate and
0.25% acetic acid, (B) methanol containing 5 mM ammonium
acetate and 0.25% acetic acid. We use the gradient shown in
Table 2 for elution.
Use SRM in the negative-ion mode of m/z 179-143 for natural
2-ClAdA. The internal standard, d4-AdA, is detected at m/z 149-105.
The ionization voltage should be 3500 V and the ionization tempera-
ture 310 1C. For MS/MS, the collision energy should be 18 V, and use
1.5 mTorr argon as the collision gas.
Analysis of 2-ClHA
Calculations and expected results
Resuspend the organic extracts in methanol/water (85/15, v/v)
containing 0.1% formic acid and separate them on a reversed-phase
2-ClHDA
HPLC column from Supelco (Discovery HS C18, 150 Â 2.1 mm, 5
mm)
utilizing a Thermo Fisher Surveyor MS-LC system with a Thermo
Fisher Quantum Ultra electrospray ionization mass spectrometer as
a detector. For LC, use the following mobile phases: (A) 85/15
methanol/water with 5 mM ammonium acetate and 0.25% acetic
acid, (B) methanol with 5 mM ammonium acetate and 0.25% acetic
acid, (C) methanol with 5 mM ammonium acetate. We use the
gradient shown in Table 1.
The 2-ClHA is detected using SRM transition, by observing the
loss of H35Cl from the target chlorinated fatty acid. Use SRM in the
negative-ion mode of m/z 289-253 for natural 2-ClHA to
measure 2-ClHA. The internal standard, 2-Cl-[7,7,8,8-d4]–HA, is
detected at m/z 293-257. For electrospray ionization MS, the
ionization energy and temperature should be 3700 V and 310 1C,
respectively. Use collision energy of 15 V and 1.5 mTorr argon as
the collision gas. Under these conditions the Rt for 2-ClHA is
ꢀ8.5 min.
Quantification is achieved using SIM by comparing the ratio of
the integrated area of the internal standard (SIM m/z 292) for
the chromatography peak with that of the natural compound (SIM
m/z 288). Integration of the peaks can be performed within the
Agilent GC–MS Enhanced Chemstation software. Fig. 3 shows an
example application of this method in studies examining the
oxidation of human LDL plasmalogens with selected concentrations
of HOCl. It should be noted that the PFB oxime has a chiral carbon
leading to the production of two derivative peaks (anti and syn) as
shown in Fig. 3. For the quantification of 2-ClHDA in this example
the integrated areas under the anti and syn isomers are added for
each SIM.
2-ClHA
2-ClHA is detected using SRM, by observing the loss of HCl
from the target chlorinated fatty acid. As an example, 2-ClHA is
observed by SRM transition of m/z 289-253 for 2-ClHA, whereas
the stable isotope standard containing four deuteriums is
observed by SRM transition of m/z 293-257. Quantification is
achieved by comparing the ratio of the integrated area of the
internal standard for the chromatography peak with that of the
Extraction of urine for 2-ClAdA analysis—response curve generation
Add 10 ng of [3,3,4,4-d4]-adipic acid and varying amounts
(0–10 ng) of synthetic 2-ClAdA to 50 ml of urine before extraction.
To this spiked urine, add 0.75 ml of water, 0.2 ml of 6 N HCl. Then
extract the sample twice in 6 ml of ethyl acetate/diethyl ether
Table 1
Table 2
Gradient used for liquid chromatography analysis of 2-ClHA.
Gradient used for elution of 2-ClAdA.
Time (min)
% A
% B
% C
Flow rate (
ml/min)
Analysis from urine
(flow rate 300 l/min)
Analysis from cell culture
m
(flow rate 200
ml/min)
0
2
100
100
0
0
0
200
200
200
200
200
400
400
200
200
200
0
100
100
0
0
0
Time (min) % A % B
Time (min)
% A
% B
6
12
13
14
17
18
19
24
0
0
0
4
100
0
0
7.5
30
34
37
100
90
20
0
0
0
100
100
100
100
0
90 10
20 80
10
80
100
100
0
0
0
12
18
23
24
35
0
0
0
0
100
100
0
0
0
0
100
100
0
100
100
Reequilibration 42
55
100
100
0
0
0
k
0

B.K. Wacker et al. / Free Radical Biology and Medicine 59 (2013) 92–99
97
Fig. 3. 50
50 ng of 2-Cl-[d4]–HDA, and
isomers of the PFB oximes, and the identity of peak * is unknown.
m
g of human LDL was incubated with (A) 0, (B) 125, or (C) 500
mM HOCl in 50 ml of PBS for 30 min at 37 1C. Reaction products were extracted in the presence of
a-chlorofatty aldehydes were converted to their PFB oximes and subsequently subjected to GC–MS as described. a and s indicate anti and syn
Fig. 4. 25 ml of human serum was subjected to base hydrolysis in the presence of 105 fmol of 2-[d4]-ClHA and subjected to LC–MS as described under Protocols.
The concentration of 2-ClHA in this sample was 4.4 nM. This example shows results with a scan time of 0.5 s. The resolution of peaks can be further enhanced with shorter
scan times and by limiting the number of SRMs in an analysis.
natural compound. The chromatographic peaks can be integrated
using XCalibur software. Fig. 4 shows an example chromatogram
of SRM detection used to quantify 2-ClHA in human serum.
achieved by comparing the ratio of the integrated area of the
internal standard for the chromatography peak with that of the
natural compound. The chromatographic peaks can be integrated
using XCalibur software. Fig. 5 shows a dilution curve used to
measure 2-ClAdA in rat and human urine.
2-ClAdA
2-ClAdA is also detected using SRM, monitored in the nega-
tive-ion mode, to observe the loss of HCl from the chlorinated
adipic acid. This transition is seen in SRM as m/z 179-143 for
2-ClAdA. The transition and the retention times can be obtained
by comparing with synthetic 2-ClAdA. For biological samples,
a series of spiked (selected amounts of 2-ClAdA with a fixed
amount of the internal standard, [3,3,4,4-d4]-adipic acid: SRM m/z
149-105) samples is run and analyzed to assess the matrix effect
of individual urine samples, which can alter the detection sig-
nificantly from sample to sample. Urine samples should also be
normalized to urinary creatinine levels to compensate for urine
product dilution difference between animals. Quantification is
Caveats
a-Chlorofatty aldehydes are poorly ionized for ESI detection and
are not stable under ESI–MS conditions. For this reason, -chlorofatty
aldehydes are derivatized to their PFB oximes before GC–MS analyses
using NICI detection. In some instances, a complex lipid mixture
necessitates first purifying
a
a
-chlorofatty aldehydes using TLC before
-chlorofatty aldehydes can
derivatization. As an alternative method,
a
be derivatized to their dimethyl acetals, but the resulting detection
with this derivative is not as sensitive as measuring PFB derivatives
with NICI. Furthermore, under the high temperatures used for acid

98
B.K. Wacker et al. / Free Radical Biology and Medicine 59 (2013) 92–99
Our studies have shown that a-ClFALD accumulation coincides
with the activation of MPO-containing leukocytes [13,14]. Impor-
tantly, negative controls in these studies have shown that
a-ClFALD do not increase in MPO-deficient leukocytes [15] or in
either 3-aminotriazole- or azide-inhibited leukocytes [13,14].
We have not observed artificial chlorination of plasmalogens under
the conditions described herein; however, extreme care should be
directed at handling bleach in the laboratory, particularly in steriliz-
ing human-derived samples.
a-ClFALD can be generated if chloro-
form extracts are exposed to bleach vapors from bleach treatment of
specimens that is performed 10–20 ft away (unpublished data).
Thus, human specimens treated with bleach for decontamination
should be performed in hoods vented outside of the lab. We have
not observed artificial chlorination of fatty acids using our methods,
and it should be appreciated that the vinyl ether of the plasmalogens
is the key target for chlorination leading to a-ClFALD release.
Fig. 5. 100 or 50 ml of human or rat urine spiked with indicated amounts of
2-ClAdA was extracted in triplicate in the presence of 10 ng of [d4]-AdA and
subjected to LC–MS as described under Protocols. The response of 2-ClAdA to
[d4]-AdA is plotted.
Acknowledgments
This work was supported by NIH Grants (to D.A.F.) HL074214,
HL098907, and HL111906, as well as the Saint Louis University
Presidential Research Fund (to D.A.F.).
methanolysis (90 1C for 90 min with 1 M HCl in anhydrous methanol)
it is possible that chlorine can be displaced from the
aldehydes. This issue is readily observed if these conditions are used
to derivatize -bromofatty aldehydes to their dimethyl acetal, as
a-chlorofatty
a
References
bromine is a better leaving group. This artifactual loss can be reduced
by derivatizing conditions of 70 1C for 30 min [26]. Although the
sensitivity of the dimethyl acetals and the potential for loss
during derivatization are limiting, it should be appreciated that this
derivative can be measured in the absence of a mass spectrometer
because these derivatives can be readily detected by GC with a flame
ionization detector.
[1] Weiss, S. J.; Klein, R.; Slivka, A.; Wei, M. Chlorination of taurine by human
neutrophils: evidence for hypochlorous acid generation. J. Clin. Invest.
70:598–607; 1982.
[2] Klebanoff, S. J. Oxygen metabolism and the toxic properties of phagocytes.
Ann. Intern. Med. 93:480–489; 1980.
[3] Spickett, C. M.; Jerlich, A.; Panasenko, O. M.; Arnhold, J.; Pitt, A. R.;
Stelmaszynska, T.; Schaur, R. J. The reactions of hypochlorous acid, the
reactive oxygen species produced by myeloperoxidase, with lipids. Acta
Biochim. Pol. 47:889–899; 2000.
Both
a-ClFA and 2-ClAdA can also be derivatized to PFB-
containing compounds (PFB esters) that can be detected by
GC–MS with negative-ion detection. This is an excellent alter-
native method if an LC–MS triple-quadrupole instrument is not
available. In our hands the sensitivity using the LC–MS method
with SRM detection described herein is at least 1 order of
magnitude better than derivatization to the PFB esters followed
by GC–MS. Furthermore, the LC–MS method does not involve a
[4] Ford, D. A. Lipid oxidation by hypochlorous acid: chlorinated lipids in
atherosclerosis and myocardial ischemia. Clin. Lipidol 5:835–852; 2010.
[5] Skaff, O.; Pattison, D. I.; Davies, M. J. The vinyl ether linkages of plasmalogens
are favored targets for myeloperoxidase-derived oxidants: a kinetic study.
Biochemistry 47:8237–8245; 2008.
[6] Farooqui, A. A.; Horrocks, L. A. Plasmalogens: workhorse lipids of membranes
in normal and injured neurons and glia. Neuroscientist 7:232–245; 2001.
[7] Gross, R. W. High plasmalogen and arachidonic acid content of canine
myocardial sarcolemma: a fast atom bombardment mass spectroscopic and
gas chromatography–mass spectroscopic characterization. Biochemistry
23:158–165; 1984.
[8] Murphy, E. J.; Joseph, L.; Stephens, R.; Horrocks, L. A. Phospholipid composi-
tion of cultured human endothelial cells. Lipids 27:150–153; 1992.
[9] Vance, J. E. Lipoproteins secreted by cultured rat hepatocytes contain the
antioxidant 1-alk-1-enyl-2-acylglycerophosphoethanolamine. Biochim. Biophys.
Acta 1045:128–134; 1990.
derivatization step. The LC–MS method to detect
extremely sensitive, but plasma levels of -ClFA are near the
limit of detection using the methods described herein. Because of
the minute amounts of -ClFA in some samples it is necessary to
a-ClFA is
a
a
run blank samples between each two biological analyses to
ensure no sample–sample carryover. Additionally, multiple inter-
ior and exterior solvent washes of the injection needle are
recommended. In our chromatography method, the inclusion of
acetic acid facilitates fatty acid binding to the solid phase and we
have added a column solvent wash with mobile phase in the
absence of acetic acid at the end of the chromatographic protocol
to further enhance elution of nonspecifically bound 2-ClHA.
The ionization of 2-ClAdA in urine during LC–MS is variable,
probably because of the extreme differences in the matrix derived
from urine. Accordingly, dilution curves are determined for each
sample as shown in Fig. 5. Based on the calculated line para-
meters the concentration of 2-ClAdA can be determined. Further-
more, owing to the various hydration states and concentrations of
urine in laboratory animals and humans the amount of 2-ClAdA is
normalized to urine creatinine levels. Our previous studies have
shown that deuterated 2-ClFA is metabolized to deuterated
2-ClAdA that appears in the urine [12], which supports the notion
that 2-ClAdA is the catabolic clearance product of 2-ClFA.
It remains to be shown whether other mechanisms can lead to
2-ClAdA in the urine, including alternative catabolic processes as
well as acid-catalyzed chlorination in the acidic milieu of urine.
[10] Zoeller, R. A.; Grazia, T. J.; LaCamera, P.; Park, J.; Gaposchkin, D. P.; Farber, H. W.
Increasing plasmalogen levels protect human endothelial cells during hypoxia.
Am. J. Physiol. Heart Circ. Physiol 283:H671–H679; 2002.
[11] Anbukumar, D. S.; Shornick, L. P.; Albert, C. J.; Steward, M. M.; Zoeller, R. A.;
Neumann, W. L.; Ford, D. A. Chlorinated lipid species in activated human
neutrophils: lipid metabolites of 2-chlorohexadecanal. J. Lipid Res.
51:1085–1092; 2010.
[12] Brahmbhatt, V. V.; Albert, C. J.; Anbukumar, D. S.; Cunningham, B. A.;
Neumann, W. L.; Ford, D. A.
o-Oxidation of a-chlorinated fatty acids: Identifica-
tion of -chlorinated dicarboxylic acids. J. Biol. Chem. 285:41255–41269; 2010.
a
[13] Thukkani, A. K.; Albert, C. J.; Wildsmith, K. R.; Messner, M. C.; Martinson, B. D.;
Hsu, F. F.; Ford, D. A. Myeloperoxidase-derived reactive chlorinating species
from human monocytes target plasmalogens in low density lipoprotein. J. Biol.
Chem. 278:36365–36372; 2003.
[14] Thukkani, A. K.; Hsu, F. F.; Crowley, J. R.; Wysolmerski, R. B.; Albert, C. J.; Ford,
D. A. Reactive chlorinating species produced during neutrophil activation
target tissue plasmalogens: production of the chemoattractant, 2-chlorohex-
adecanal. J. Biol. Chem. 277:3842–3849; 2002.
[15] Thukkani, A. K.; McHowat, J.; Hsu, F. F.; Brennan, M. L.; Hazen, S. L.; Ford, D. A.
Identification of alpha-chloro fatty aldehydes and unsaturated lysophosphati-
dylcholine molecular species in human atherosclerotic lesions. Circulation
108:3128–3133; 2003.
[16] Messner, M. C.; Albert, C. J.; McHowat, J.; Ford, D. A. Identification of
lysophosphatidylcholine-chlorohydrin in human atherosclerotic lesions.
Lipids 43:243–249; 2008.
[17] Marsche, G.; Heller, R.; Fauler, G.; Kovacevic, A.; Nuszkowski, A.; Graier, W.;
Sattler, E.; Malle, E. 2-Chlorohexadecanal derived from hypochlorite-

B.K. Wacker et al. / Free Radical Biology and Medicine 59 (2013) 92–99
99
modified high-density lipoprotein-associated plasmalogen is a natural inhi-
bitor of endothelial nitric oxide biosynthesis. Arterioscler. Thromb. Vasc. Biol.
24:2302–2306; 2004.
[22] Wildsmith, K. R.; Albert, C. J.; Anbukumar, D. S.; Ford, D. A. Metabolism of
myeloperoxidase-derived 2-chlorohexadecanal. J. Biol. Chem. 281:16849–
16860; 2006.
[18] Thukkani, A. K.; Martinson, B. D.; Albert, C. J.; Vogler, G. A.; Ford, D. A.
Neutrophil-mediated accumulation of 2-ClHDA during myocardial infarction:
2-ClHDA-mediated myocardial injury. Am. J. Physiol. Heart Circ. Physiol
288:H2955–H2964; 2005.
[23] Bligh, E. G.; Dyer, W. J.
A rapid method of total lipid extraction and
purification. Can. J. Biochem. Physiol 37:911–917; 1959.
[24] Dole, V. P. A relation between non-esterified fatty acids in plasma and the
metabolism of glucose. J. Clin. Invest 35:150–154; 1956.
[19] Messner, M. C.; Albert, C. J.; Ford, D. A. 2-Chlorohexadecanal and 2-chlor-
ohexadecanoic acid induce COX-2 expression in human coronary artery
endothelial cells. Lipids 43:581–588; 2008.
[20] Albert, C. J.; Anbukumar, D. S.; Messner, M. C.; Ford, D. A. Chromatographic
methods for the analyses of 2-halofatty aldehydes and chlorohydrin mole-
[25] Kakola, J.; Alen, R.
A fast method for determining low-molecular-mass
aliphatic carboxylic acids by high-performance liquid chromatography–
atmospheric pressure chemical ionization mass spectrometry. J. Sep. Sci
29:1996–2003; 2006.
[26] Albert, C. J.; Crowley, J. R.; Hsu, F. F.; Thukkani, A. K.; Ford, D. A. Reactive
brominating species produced by myeloperoxidase target the vinyl ether
bond of plasmalogens: disparate utilization of sodium halides in the
production of alpha-halo fatty aldehydes. J. Biol. Chem. 277:4694–4703; 2002.
cular species of lysophosphatidylcholine. J. Chromatogr.
Biomed. Life Sci. 877:2768–2777; 2009.
B Anal. Technol.
[21] Lange, V.; Picotti, P.; Domon, B.; Aebersold, R. Selected reaction monitoring
for quantitative proteomics: a tutorial. Mol. Syst. Biol. 4:222; 2008.