PREPRINT DOCUMENT
Information of the Journal in which the present paper is
published:
 Environmental Science and Pollution Research, 20:
11907-11916 (2014).
 DOI: dx.doi.org/10.1007/s11356-014-3172-5
1
Characterization of complex lipid mixtures in contaminant exposed JEG-3 cells
using liquid chromatography and high resolution mass spectrometry
Eva Gorrochategui a, Josefina Casas b, Elisabeth Pérez-Albaladalejo a, Olga Jáuregui c,
Cinta Porte a, Sílvia Lacorte a,*
a
Department of Environmental Chemistry, Institute of Environmental Assessment and
Water Research (IDAEA), Consejo Superior de Investigaciones Científicas (CSIC),
Jordi Girona 18-26, Barcelona, 08034, Catalonia, Spain.
b
Department of Biomedicinal Chemistry, Institute of Advanced Chemistry of Catalonia
(IQAC), Jordi Girona 18-26, Barcelona, 08034, Catalonia, Spain.
c
Scientific and Technological Centers, University of Barcelona, Baldiri Reixac, 4,
Barcelona, 08028, Catalonia, Spain.
*Author for correspondence: (Tel.: +34-934006133 Fax: +34-932045904 email:slbqam@cid.csic.es).
Keywords: lipidomics, JEG-3 cells, mass spectrometry, time-of-flight and orbitrap
analyzers,
perfluorinated
chemicals
2
(PFCs),
tributyltin
(TBT).
1
Abstract: The aim of this study was to develop a method based on ultra-high
2
performance liquid chromatography coupled with mass spectrometry (UHPLC-MS) for
3
lipid profiling in human placental choriocarcinoma (JEG-3) cells. Lipids were solid-
4
liquid extracted from JEG-3 cells using a solution of chloroform/methanol (2:1, v/v) in
5
a simple procedure requiring minimal sample alteration. Simultaneous separation of
6
complex lipid mixtures in their major classes was achieved with a reversed-phase (C8)
7
UHPLC column and a mobile phase containing methanol with 1 mM ammonium
8
formate and 0.2% formic acid (A)/ water with 2 mM ammonium formate and 0.2%
9
formic acid (B). Lipids were characterized using time-of-flight (TOF) and Orbitrap
10
under full scan and positive electrospray ionization mode with both analyzers. A total of
11
178 species of lipids including 37 phosphatidylcholines (PC), 32 plasmalogen PC, 9
12
lyso PC, 4 lyso plasmalogen PC, 30 triacylglycerols, 22 diacylglycerols, 7 cholesterol
13
esters, 25 phosphatidylethanolamines and 12 sphingomyelins were identified using TOF
14
and Orbitrap. The identification of all lipid classes was based on exact mass
15
characterization with an error ˂ 5 ppm. The developed methodology was applied to
16
study lipid alterations in human placental cells against the exposure to perfluorinated
17
chemicals (PFCs) and tributyltin (TBT).
3
18
Introduction
19
Lipidomics, a ramification of metabolomics, is the end point of omics cascade and
20
can be described as the system-wide study of lipids and their interaction with other
21
biochemicals. In fact, the term lipidome can be defined as the comprehensive and non-
22
exhaustive quantitative description of a group of lipid classes that may constitute a cell
23
or bio-organism (Castro-Perez et al. 2010). Lipids and their interaction with cells play a
24
crucial role in living organisms. Among the multiple biological functions of lipids, they
25
contribute in compartmentalization, energy production and storage, cell-signaling
26
processes, protein trafficking and membrane organizing tasks (Oresic et al. 2008; Van
27
Meer 2005). Moreover, several diseases including obesity (Shi and Burn 2004),
28
cardiovascular dysfunctions and diabetes, cancer and neurodegenerative alterations are
29
associated with abnormalities in lipid functions and physiological levels (Shui et al.
30
2007). Lipid alterations have been attributed to the possible involvement of
31
environmental obesogens, xenobiotics that can disrupt the normal developmental and
32
homeostatic control over adipogenesis and energy balance (Grün and Blumberg 2006).
33
Among others, tributyltin (TBT) has raised a lot of attention since it is an environmental
34
endocrine disrupter with the capacity to promote adipogenesis (Grün et al. 2006). In
35
addition to TBT, perfluorinated chemicals (PFCs) have been reported to alter lipid
36
levels in some animal species and humans (Gilliland and Mandel 1996; Nelson et al.
37
2010). Both TBT and PFCs share bioaccumulative properties, they are widely
38
distributed environmental pollutants and are the target analytes of the present study.
39
According to all that, the importance of lipidomics not only lays on its contribution
40
towards the enhanced understanding of the pathogenesis of multiple disease states
41
related to lipids but also on the study of the impact of some emerging chemicals such as
42
obesogen compounds in the environment and humans.
4
43
Lipids are exceptionally diverse in their structural, chemical, and physical properties.
44
According to the latest version of the Comprehensive Classification System for Lipids
45
firstly established on 2005 by the International Lipid Classification and Nomenclature
46
Committee (ILCNC), lipids are divided into eight main distinctive classes: 1) Fatty
47
acids (FA), 2) Glycerolipids (GL), 3) Glycerophospholipids (GP), 4) Sphingolipids
48
(SP), 5) Sterol lipids (ST), 6) Prenol lipids (PR), 7) Saccharolipids (SL) and 8)
49
Polyketides (PK) (Fahy et al. 2009). The characterized lipids in the present study
50
include some GL (di- and triacylglycerols), various GP (phosphatidylcholines and its
51
lyso forms and derivatives with plasmalogens and phosphatidylethanolamines),
52
sphingomyelins in the group of SP and cholesterol esters as ST.
53
Lipid extraction is the first step toward lipidomic analysis. Due to the water-insoluble
54
nature of lipid molecules, most of the described procedures in literature use organic
55
solvents as the preferred extractive agents. Among the organic solvents, the most
56
common for lipid extraction in biological tissues are chloroform and methanol
57
combined in a mixture (2:1, v/v) (Bligh and Dyer 1959; Folch et al. 1957). However,
58
non-traditional extractive methods have used other organic solvents such as isopropanol
59
or hexane in order to maximize the selective collection of particular lipid classes of
60
interest (Hughes and Brash 1991).
61
Separation of the extracted lipid species in their major classes is a complicate
62
procedure due to the highly complex biological matrices in which they are contained.
63
One of the earliest techniques developed was thin layer chromatography (TLC) for
64
routine analysis of lipids (Bennett and Heftmann 1962; Ruggieri 1962). However, in
65
most recent applications this technique is only used as a fast and extensive screening
66
tool prior to the exhaustive analysis with more sensitive and selective techniques such
67
as liquid chromatography (LC) (Watson 2006).
5
68
Lipidomic analysis by LC can follow either normal-phase or reverse-phase strategies.
69
While fatty acids are commonly separated on reverse-phase columns (Watson 2006),
70
separation of phospholipids (PL) can be achieved by both approaches. Normal-phase
71
LC effectively separates PL according to their different polar heads and not considering
72
their sn-1 and sn-2 fatty acid substituent. In contrast, reversed-phase strategy separates
73
PL on the basis of their fatty acid residues (Castro-Perez et al. 2010). However, for a
74
complete separation of lipids two-dimensional LC has reported to be the ideal method
75
since it allows the combination of normal- and reversed-phase approaches (Pulfer and
76
Murphy 2003, Wang et al. 2013).
77
Various classic detection methods such as spectrophotometric analysis in the UV
78
range and evaporative light scattering have reported to be adequate for detecting lipids
79
(Watson 2006). However, in recent years, mass spectrometry (MS) has evolved as a
80
superior detection method for identifying lipids in biological matrices due to its high
81
sensitivity and the additional information it provides (Sommer et al. 2006). Moreover,
82
the recent ability of high resolution mass spectrometers to obtain accurate mass
83
measurements has emplaced them at the top of MS analyzers in lipidomic research. In
84
fact, analyzers such as Orbitraps, Fourier transform ion cyclotron resonance (FTICR),
85
time-of-flight (TOF) and hybrid quadrupole orthogonal TOF (Q-TOF) have replaced the
86
conventional low resolution quadrupoles and linear ion traps, as they can resolve
87
isomeric and isobaric species and elucidate elemental composition.
88
Thus, the aim of this study was to characterize the lipidomic composition of JEG-3
89
cells using two HRMS. Moreover, the effects of PFCs and TBT on the lipidome of the
90
human placental choriocarcinoma cell line JEG-3 were investigated.
91
92
Materials and methods
6
93
Chemicals and reagents. Minimum Essential Medium, fetal bovine serum, L-
94
glutamine, sodium pyruvate, nonessential amino acids, penicillin G, streptomycin,
95
phosphate buffered saline (PBS) and trypsin-EDTA were supplied by Gibco BRL Life
96
Technologies (Paisley, Scotland, UK). Tributyltin (TBT), perfluorobutanoic acid
97
(PFBA),
98
perfluorononanoic
99
perfluorohexanesulfonate (PFHxS) were purchased from Sigma Aldrich (Steinheim,
100
Germany) and perfluorobutanesulfonate (PFBS) and perfluorooctanesulfonate (PFOS)
101
were obtained from Fluka (Austria). Stock standard solutions containing the mixture of
102
the eight PFCs were prepared in ethanol at concentrations of 0.1 mM and 1 mM and the
103
stock solution of TBT was prepared in dimethyl sulfoxide (DMSO) at a concentration of
104
0.02 mM. These solutions were stored at -20oC. HPLC grade water, methanol (> 99.8%)
105
and acetonitrile (> 99.8%) were purchased from Merck (Darmstadt, Germany).
106
Chloroform was supplied by Carlo-Erba (Peypin, France) and butylated hydroxytoluene
107
(BHT) by Sigma Aldrich (St.Louis, MO, USA).
perfluorohexanoic
acid
acid
(PFNA),
(PFHxA),
perfluorooctanoic
perfluorododecanoic
acid
acid
(PFOA),
(PFDoA)
and
108
109
Cell culture. JEG-3 cells derived from a human placental carcinoma were obtained
110
from American Type Culture Collection (ATCC HTB-36). They were grown in Eagle’s
111
Minimum Essential Medium supplemented with 5% fetal bovine serum, 2 mM L-
112
glutamine, 1 mM sodium pyruvate, 0.1 mM nonessential amino acids, 1.5 g/L sodium
113
bicarbonate and 50 U/mL penicillin and 50 g/mL streptomycin in a humidified
114
incubator with 5% CO2 at 37ºC. Cells were routinely cultured in 75 cm2 polystyrene
115
flasks (Corning; NY, USA). When 90% confluence was reached, cells were dissociated
116
with 0.25% (w/v) trypsin and 0.9 mM EDTA (trypsin-EDTA) for subculturing and
117
experiments. Experiments were carried out on confluent cell monolayers.
7
118
119
Sample preparation
120
Cell exposure to PFCs and TBT. Cells were seeded at a rate of 0.67·106 cells per
121
well (6-well plate) and allowed to attach overnight in an incubator at 37 ºC, 5% CO2.
122
Then, 6 µL of the 0.02 mM stock solution of TBT or 6 µL of the 0.1 and 1 mM stock
123
solutions containing the eight PFCs of study (PFBA, PFHxA, PFOA, PFNA, PFDoA,
124
PFBS, PFHxS and PFOS) were directly added to the wells. The final concentration of
125
DMSO and ethanol in culture wells was 0.4% (v/v), and final concentrations of PFCs
126
were 0.6 and 6.0 µM and 0.1 µM for TBT. After 24 h of exposure, the medium was
127
aspirated and cells were washed with PBS, trypsinized and centrifuged at 5,400 g for 10
128
min. The supernatant was aspirated and cells were stored at -80 ºC until analysis. In all
129
cases, addition of the tested compound was done in triplicate and controls were
130
performed by adding the corresponding solvent to cells.
131
132
Lipid extraction. Lipids were extracted from JEG-3 cells with similar extraction
133
conditions of a previous study (Christie 1985). To the cell pellets, 540 μL of a
134
methanol: chloroform (1:2, v/v) solution containing 0.01% of BHT, acting as an
135
antioxidant, were added. Samples were shaken with a vortex mixer (1 min), were settled
136
for 30 min and extracted in an ultrasonic bath for 5 min at room temperature (2 times).
137
Between each period of 5 min, samples were thoroughly mixed. Afterwards, samples
138
were centrifuged at 13,000 g for 5 min. The supernatant was transferred to a new micro
139
vial, evaporated to dryness, reconstituted with 160 µL of acetonitrile and stored at -20
140
ºC in an argon atmosphere.
141
142
Instrumental analysis
8
143
UHPLC conditions. All analyses were performed with an ultra- high performance
144
liquid chromatography (UHPLC) system using an octyl carbon chain (C8)-bonded silica
145
column. Chromatographic parameters such as column temperature, injection volume,
146
flow rate, mobile phases and gradient elution programs are summarized in Table 1.
147
148
MS conditions: Analyses were performed with an UHPLC system coupled to two
149
distinct mass analyzers. The analytical instrumentation used were an UHPLC system
150
coupled to a Waters/ LCT Premier XE TOF analyzer controlled with Waters/
151
Micromass MassLynx 4.1 software and an UHPLC system (Accela) coupled to a
152
Thermo Fischer Scientific LTQ Orbitrap Velos controlled with Thermo Fischer
153
Scientific/ Xcalibur software. The MS parameters such as the ionization mode (positive
154
electrospray), the mass acquisition range used in each mass analyzer along with other
155
parameters are summarized in Table 1.
156
157
Identification and relative quantification of lipids. Positive identification of lipids
158
was based on the accurate mass measurement with an error < 5 ppm using both high
159
resolution TOF and Orbitrap. Relative retention times in UHPLC compared to that of
160
some standards (±2) used in a previous study, analyzed under the same chromatographic
161
conditions (Garanto et al. 2013), were also considered as identification criteria. An
162
inventory of a total of 225 lipids, containing 45 phosphatidylcholines (PC), 36
163
plasmalogen PC, 18 lyso PC, 4 lyso plasmalogen PC, 39 triacylglycerols (TAG), 26
164
diacylglycerols (DAG), 7 cholesterol esters (CE), 35 phosphatidylethanolamines (PE)
165
and 15 sphingomyelins (SM), based on reported identified species (Garanto et al.,
166
2013), was first generated. Their theoretical exact masses were determined using a
167
spectrum simulation tool of Xcalibur software and the obtained list was further used as
9
168
a homemade referential database. Individual chromatographic peaks of distinct lipid
169
species were isolated from full scan MS spectra when selecting their theoretical exact
170
masses, extracted from the database. Then, a list of possible candidates fitting the
171
specific exact mass was generated using formula determination tools (Elemental
172
composition search) of both Micromass MassLynx and Thermo Fischer Scientific
173
Xcalibur softwares. The elemental number was restricted to include C, H, O, N and P.
174
The formula constraints were C, H, O ≥ 1, P ≥ 0 and N ≥ 1, following the nitrogen rule.
175
The number of double-bond equivalents (DBEs) was set between -0.5 and 15.0. The
176
search was based on single mass analysis and only considered the m/z value of the
177
monoisotopic peak.
178
Annotation of lipid species: GL, GP SP and ST are annotated as <lipid subclass>
179
<total fatty acyl chain length>:<total number of unsaturated bonds> and SM <total fatty
180
acyl chain length>:<total number of unsaturated bonds in the acyl chain>. Relative
181
quantification was done by comparison of peak areas in extracted ion chromatograms
182
between exposed cells and controls.
183
184
Data processing and statistical analysis. Significant differences among controls
185
and TBT/PFC-exposure treatment mean values (n=3) were determined by an ANOVA
186
of one factor, considering P ˂ 0.05 statistically significant. All graphics were computed
187
in Excel 2007.
188
189
Results and discussion
190
Chromatographic separation. Successful UHPLC separation of lipids in their
191
major classes was achieved using a C8-bonded silica column of 100 mm length. With
192
TOF, 10 µL of the sample extract provided good resolution and sensitivity. However,
10
193
with the Orbitrap, improved chromatographic peak shape was obtained when injecting 5
194
instead of 10 µL into the system and optimum column temperature was set to 30 ºC.
195
Concerning mobile phases, the optimum conditions were the use of methanol with 1
196
mM ammonium formate and 0.2% formic acid (A) combined with water with 2 mM
197
ammonium formate and 0.2% formic acid (B). The addition of ammonium formate to
198
the solvents has reported to be advantageous due to the formation of adduct ions [M +
199
NH4] + in positive mode, which are known to be more stable than the [M + H] + ions for
200
some lipid classes. In the present study, lipids corresponding to the families of TAG,
201
DAG and CE were identified as ammonium adducts whereas the rest of lipid groups
202
were identified in the monoprotonated form. However, the use of ammonium formate
203
requires the addition of buffering acids, such as the formic acid added in the present
204
study, to avoid the formation of double-peak chromatograms (Sommer et al. 2006).
205
Regarding elution program, the gradient conditions slightly changed when using the
206
TOF or the Orbitrap with the aim to improve the chromatographic separation. Both
207
gradients presented in conditions A and B of Table 1 allowed good separation of main
208
lipid classes. They both started at high percentage of the organic phase A, 80 and 85%,
209
respectively, and increased up to 99% to allow the desired elution of lipid species. With
210
the Orbitrap, the two gradients were tested and enhanced separation of lipid species was
211
achieved when using the latter conditions, despite having a longer chromatogram than
212
the one used with TOF. Figure 1 shows UHPLC chromatograms of TOF (Fig. 1a) and
213
Orbitrap (Fig. 1b) when working at their respective optimum conditions (Table 1).
214
215
MS conditions. As observed in Table 1, the main difference among MS parameters
216
used with TOF and Orbitrap was the acquisition range. The shorter range used in
11
217
Orbitrap respect to the one used in TOF, up to 1145 m/z values shorten, was necessary
218
to suppress high contribution of background ions in the system.
219
220
Lipid identification. Lipid species were successfully identified with both high
221
resolution TOF and Orbitrap. As shown in the UHPLC chromatograms of Figures 1a
222
and 1b, equivalent chromatographic distribution profiles were obtained when using both
223
TOF and Orbitrap. Lyso PC and lyso plasmalogen PC were the first groups of lipids to
224
elute and appeared in the early 5 min of the UHPLC-TOF chromatogram and between 4
225
and 7 min of the UHPLC-Orbitrap chromatogram. These groups of lipids were totally
226
resolved from PC, plasmalogen PC, PE, SM and DAG which eluted together in the
227
subsequent ten minutes of the chromatogram. TAG and CE appeared together in the
228
final minutes of the chromatogram. Individual lipid species unresolved in the total ion
229
chromatogram were successfully isolated when their exact masses were selected. Thus,
230
despite the incomplete chromatographic resolution of the complex lipid mixture,
231
identification of individual lipids was possible using both the high resolution TOF and
232
Orbitrap. A total of 178 species of lipids were identified by TOF and Orbitrap and are
233
shown in Table 2. In the group of glycerophospholipids, 107 species were identified
234
containing 37 PC , 32 plasmalogen PC , 9 lyso PC , 4 lyso plasmalogen PC and 25 PE.
235
In the group of glycerolipids, 52 species were identified containing 30 TAG and 22
236
DAG . As sterols 7 CE were identified and in the group of sphingolipids, 12 SM were
237
determined. The compounds identified with their accurate mass measurement, elemental
238
composition, calculated mass, error, double bound equivalents and retention times are
239
shown in the Supplementary Table 1. Although not all lipids present in the cells were
240
analyzed, the reported classes are quite representative since account for the 70% of
241
cellular lipidome. Phosphatydilserine and phosphatidylinositol were not determined
12
242
since they can only be detected in negative electrospray ionization mode and the
243
sensitivity was too low considering the amount of extracted lipids from JEG-3 cells.
244
The distinct resolution of TOF and Orbitrap, 11,500 and 30,000 FWHM at m/z
245
556and 400, respectively, resulted in a high lipid identification power. The use of
246
30,000 resolution with the Orbitrap would allow the resolution of isobaric species,
247
although in our specific case, such species were chromatographically resolved. Fig. 2
248
shows the identification ratio of the different target lipids when using both mass
249
spectrometers. Equal results were obtained in families of lyso PC, TAG, DAG, CE and
250
SM. In the rest of lipid groups, PC, plasmalogen PC lyso plasmalogen PC and PE,
251
identification was slightly superior with Orbitrap but with no significant differences. Of
252
all the 178 lipids identified in the present study 88% were found by both analyzers, 4%
253
were only seen by TOF and 8% were only found with Orbitrap. Thus, the two high
254
performance platforms, TOF and Orbitrap analyzers, showed equivalent capability for
255
lipidomic analysis in human placental choriocarcinoma JEG-3 cells.
256
257
Lipidome changes in PFC/TBT-exposed cells. The chromatographic profiles of
258
PFC/TBT-exposed cells were compared to controls. TBT effects were studied at the low
259
concentration of 0.1 µM since it was the reported level at which this compound
260
promoted adipogenesis in vertebrates (Grün et al. 2006). In contrast, PFC effects were
261
studied at two levels of concentration, 0.6 and 6 µM, the first close to the TBT exposure
262
level and the latter about 10-times higher. The selected doses of exposure were non-
263
toxic for JEG-3 cells, as shown in a previous study (Gorrochategui et al. 2014).
264
265
266
Fig. 3 represents differences in lipid amounts of exposed cells and controls caused by
the distinct treatments, expressed as increasing rates, calculated as:
Increasing rate =
∑ni=1 Ai, exposed cells
13
∑ni=1 Ai, controls
267
Where “Ai” represents peak area of lipid and “n” the number of replicates of the four
268
groups of samples (controls, cells exposed to PFCs at 0.6 µM, PFCs at 6 µM and TBT),
269
which was three in our experiment conditions. In each case, the variance associated to
270
the increasing rate was calculated as the quotient of the standard error of the mean
271
(SEM) of the corresponding cell treatment and SEM of controls.
272
Despite most lipid species were identified using both analyzers, the results shown
273
correspond to the TOF analysis with the exception of few lipid species only identified
274
with Orbitrap. Lipid species exclusively identified with TOF and Orbitrap are
275
highlighted with symbols # and ##, respectively.
276
As shown in Fig. 3a, distinct effects were observed in the four groups of GP
277
analyzed, PC, plasmalogen PC, lyso PC and lyso plasmalogen PC, when exposed to
278
PFCs and TBT. While exposure to PFCs produced significant increase in some lipid
279
species, there was no significant alteration resulting from TBT exposure in the majority
280
of cases. Even more, in groups of lyso PC and lyso plasmalogen PC, levels of lipids of
281
cells exposed to the organotin compound were lower than in control samples.
282
Surprisingly, effects of TBT were highly visible in families of TAG and DAG, whose
283
levels suffered a dramatic increase after the exposure to the organotin compound. The
284
effect of PFCs was considerably minor, even in the high concentration exposure of 6
285
µM, producing a low increase in the amount of some TAG and no significant effect on
286
DAG. In the case of CE species, only two of them, 18:0 and 18:1 suffered a significant
287
increase when exposed to TBT. Finally, effects on PE were practically undetectable and
288
were only noticed in lipid specie 38:3 whereas effects on SM resulted higher and were
289
attributed to the presence of PFCs.
290
The observed increase in the amount of lipids resulting from the presence of TBT is
291
in accordance to the findings of several studies which report the obesogenic effects of
14
292
the organotin compound. According to our findings, Janer et al. (2007) reported
293
increased accumulation of lipids and fatty acids in ramshorn snail Marisa cornuarietis.
294
Moreover, Grün et al. (2006) reported promoted adipogenesis in murine 3T3-L1 cell
295
model and elevated lipid accumulation in adipose depots, liver and testis of neonate
296
mice exposed to TBT. The observed effects of PFCs suggested a major influence on
297
membrane lipids containing phosphorylcholine such as PC, plasmalogen PC, lyso PC
298
and lyso plasmalogen PC and sphingomyelins. In contrast, alterations of TBT where
299
highly noticeable in TAG, DAG and CE species.
300
In conclusion, the presented UHPLC-TOF/Orbitrap approaches allowed successful
301
lipid profiling of JEG-3 cells. The proposed methodology was applied for the study of
302
lipid alterations in PFC/TBT-exposed cells observing significant effects of both
303
xenobiotics.
304
305
Acknowledgments
306
This study was financed by the INNPACTO project (IPT-2011-0709-060000). Dr. R.
307
Chaler, D. Fanjul and Eva Dalmau are acknowledged for TOF-MS support and Dr.
308
Alberto Adeva for Orbitrap-MS support.
309
310
References
311
Bennett RD, Heftmann E (1962) Thin-layer chromatography of sterols. J Chromatogr A
312
313
314
9: 359-362
Bligh EG, Dyer WJ (1959) A rapid method of total lipid extraction and purification.
Can J Biochem Phys 37: 911-917
315
Castro-Perez JM, Kamphorst J, Degroot J, Lafeber F, Goshawk J, Yu K, Shockcor JP,
316
Vreeken RJ, Hankemeier T (2010) Comprehensive LC-MSE lipidomic analysis
15
317
using a shotgun approach and its application to biomarker detection and
318
identification in osteoarthritis patients. J Proteome Res 9: 2377-2389
319
Christie WW (1985) Rapid separation and quantification of lipid classes by high
320
performance liquid chromatography and mass (light-scattering) detection. J
321
Lipid Res 26: 507-512
322
Fahy E, Subramaniam S, Murphy RC, Nishijima M, Raetz CRH, Shimizu T, Spener F,
323
Van Meer G, Wakelam MJO, Dennis EA (2009) Update of the LIPID MAPS
324
comprehensive classification system for lipids. J Lipid Res 50: S9-S14
325
Folch J, Lees M, Sloane Stanley GH (1957) A simple method for the isolation and
326
purification of total lipids from animal tissues. J Biol Chem 226: 497-509
327
Garanto A, Mandal NA, Egido-Gabás M, Marfany G, Fabriàs G, Anderson RE, Casas J,
328
Gonzàlez-Duarte R (2013) Specific sphingolipid content decrease in Cerkl
329
knockdown mouse retinas. Exp Eye Res 110:96-106.
330
Gilliland FD, Mandel JS (1996) Serum perfluorooctanoic acid and hepatic enzymes,
331
lipoproteins, and cholesterol: A study of occupationally exposed men. Am J Ind
332
Med 29: 560-568
333
Gorrochategui E, Pérez-Albaladejo E, Casas J, Lacorte S, Porte C (2014) Perfluorinated
334
chemicals: Differential toxicity, inhibition of aromatase activity and alteration of
335
cellular lipids in human placental cells. Toxicol Appl Pharmacol (2014),
336
http://dx.doi.org/10.1016/j.taap.2014.03.012
337
338
Grün F, Blumberg B (2006) Environmental obesogens: Organotins and endocrine
disruption via nuclear receptor signaling. Endocrinol 147: S50-S55
339
Grün F, Watanabe H, Zamanian Z, Maeda L, Arima K, Cubacha R, Gardiner DM,
340
Kanno J, Iguchi T, Blumberg B (2006) Endocrine-disrupting organotin
16
341
compounds are potent inducers of adipogenesis in vertebrates. Mol Endocrinol
342
20: 2141-2155
343
Hughes MA, Brash AR (1991) Investigation of the mechanism of biosynthesis of 8-
344
hydroxyeicosatetraenoic acid in mouse skin. BBA- Lipid Lipid Met 1081: 347-
345
354
346
Janer G, Navarro JC, Porte C (2007) Exposure to TBT increases accumulation of lipids
347
and alters fatty acid homeostasis in the ramshorn snail Marisa cornuarietis.
348
Comp Biochem Phys C 146: 368-374
349
Nelson JW, Hatch EE, Webster TF (2010) Exposure to polyfluoroalkyl chemicals and
350
cholesterol, body weight, and insulin resistance in the general U.S. population.
351
Environ Health Persp 118: 197-202
352
353
354
355
356
357
358
359
360
361
Oresic M, Hänninen VA, Vidal-Puig A (2008) Lipidomics: a new window to
biomedical frontiers. Trends Biotechnol 26: 647-652
Pulfer M, Murphy RC (2003) Electrospray mass spectrometry of phospholipids. Mass
Spectrom Rev 22: 332-364
Ruggieri S (1962) Separation of the methyl esters of fatty acids by thin layer
chromatography. Nature 193: 1282-1283
Shi Y, Burn P (2004) Lipid metabolic enzymes: Emerging drug targets for the treatment
of obesity. Nat Rev Drug Discov 3: 695-710
Shui G, Bendt AK, Pethe K, Dick T, Wenk MR (2007) Sensitive profiling of chemically
diverse bioactive lipids. J Lipid Res 48: 1976-1984
362
Sommer U, Herscovitz H, Welty FK, Costello CE (2006) LC-MS-based method for the
363
qualitative and quantitative analysis of complex lipid mixtures. J Lipid Res 47:
364
804-814
365
Van Meer G (2005) Cellular lipidomics. EMBO J 24: 3159-3165
17
366
Wang S, Li J,Shi X, Qiao L, Lu X, Xu G (2013) A novel stop-flow two-dimensional
367
liquid chromatography-mass spectrometry method for lipid analysis. J
368
Chromatogr A 1321: 65-72.
369
Watson AD (2006) Thematic review series: Systems biology approaches to metabolic
370
and cardiovascular disorders. Lipidomics: A global approach to lipid analysis in
371
biological systems. J Lipid Res 47: 2101-2111
18
372
Table 1. UHPLC and MS conditions tested for the analysis of lipids in JEG-3 cells.
UHPLC conditions
A
LC system
Column
B
Waters ACQUITY
UHPLC system
Thermo Fischer ScientificAccela
UHPLC system
ACQUITY UPLC BEH
(Waters, Ireland) C8 column (100 x 2.1 mm) 1.7 µm
Column T
30 ºC
30 ºC
Injection
volume
10 µL
5 µL
Flow rate
Mobile phase
Gradient
elution
0.3 mL/min
A: Methanol with 1 mM ammonium formiate and 0.2% formic acid
B: Water with 2 mM ammonium formiate and 0.2% formic acid
80 to 90% of A/ 3 min, held for 3 min,
increase to 99% A/ 9 min, held for 3
min, return to initial conditions in 2 min
and stabilization for 3 min.
85% of A/ 1 min, increase to 90% A/ 9
min, held for 2 min, increase to 99% A
in 6 min, held for 2 min, return to initial
conditions in 2 min and stabilization for
3 min.
MS conditions
A
MS system
Ionization
mode
Acquisition
range
B
Waters/ LCT Premier XE time-of-flight
(TOF)analyzer.
Thermo Fischer Scientific
Orbitrapanalyzer.
ESI (+)
ESI (+)
50 to 1800 m/z
395 to 1000 m/z
Capillary voltage
3.0 kV
Source voltage
3.5 kV
Desolvatation
temperature
350 ºC
Capillary
temperature
300 ºC
Desolvatation Gas
flow
600 L/hr
Sheath Gas flow
Auxiliar Gas flow
Sweep Gas flow
50 L/hr
20 L/hr
373
19
2 L/hr
374
Table 2. Lipid species identified using both TOF and Orbitrap analyzers. Symbols (#)
375
indicate lipid species only identified using TOF and (##) using Orbitrap. Lipid species
376
GL, GP SP and ST are annotated as <total fatty acyl chain length>:<total number of
377
unsaturated bonds> and SM as <total fatty acyl chain length>:<total number of
378
unsaturated bonds in the acyl chain>.
Glycerophospholipids
PC
30:0##, 32:0, 32:1, 32:2, 34:0, 34:1, 34:2, 34:3, 34:4, 34:5##, 34:6#,
36:0, 36:1, 36:2, 36:3, 36:4, 36:5, 36:6, 38:1, 38:2, 38:3, 38:4, 38:5,
38:6##, 38:8##, 40:0, 40:1, 40:2, 40:3, 40:4, 40:5, 40:6, 42:2, 42:3,
42:4, 42.5, 42:6
Plasmalogen PC
30:0, 30:1, 30:2##, 32:0, 32:1, 32:2, 34:0, 34:1, 34:2, 34:3, 34:4##,
36:0, 36:1, 36:2, 36:3, 36:4, 36:5, 38:1, 38:2, 38:3, 38:4, 38:5, 38:6,
38:7, 40:1, 40:2, 40:3, 40:4, 40:5, 40:6, 40:7, 40:8##
Lyso PC
16:0, 16:1, 18:0, 18:1, 18:2, 20:1, 20:2, 20:3, 20:4
Lyso Plasmalogen PC
16:0, 16:1##, 18:0, 18:1
PE
32:0, 32:1, 34:0, 34:1, 34:2, 34:3#, 36:0##, 36:1, 36:2, 36:3, 36:4,
36:6##, 38:1, 38:2, 38:3, 38:4, 38:5, 38:6, 38:7#, 40:1, 40:2#, 40:3,
40:4##, 40:5##, 40:6##
Glycerolipids
TAG
48:0, 48:1, 48:2, 48:3, 48:4, 48:6##, 50:0, 50:1, 50:2, 50:3, 50:4,
50:5, 50:6, 52:0, 52:1, 52:2, 52:3, 52:4, 52:5, 52:6, 52:7, 54:1#,
54:2, 54:3, 54:4, 54:5, 54:6, 54:7, 54:8, 54:9
DAG
32:0, 32:1, 32:2, 32:3, 34:0, 34:1, 34:2, 34:3, 34:4, 36:0, 36:1, 36:2,
36:3, 36:4, 36:5, 36:6, 38:0##, 38:1#, 38:2, 38:3, 38:4, 38:5
Sterol lipids
CE
18:0, 18:1, 18:2, 18:3, 20:4, 20:5, 22:6
Sphingolipids
SM
14:0, 16:0, 16:1, 18:0, 18:1, 20:0, 20:1, 22:0, 24:0, 24:1, 24:2, 24:3
(#) TOF; (##) Orbitrap
379
380
20
381
Figure captions
382
383
Fig.1. UHPLC-MS total ion chromatograms of extracted lipids from JEG-3 cells when
384
using a] the TOF analyzer and b] the Orbitrap analyzer, under conditions A and B of
385
Table 1, respectively.
386
387
Fig.2. Number of identified lipid species when using TOF and Orbitrap analyzers.
388
389
Fig.3. Increasing rates of a] PC, plasmalogen PC, lyso PC and lyso plasmalogen PC, b]
390
TAG, DAG and CE and c] PE and SM in JEG-3 cells exposed to a mixture of PFCs, at
391
0.6µ M (
392
control (set to 1) and are means  SEM (n=3), analyzed by TOF. Symbols (#) indicate
393
lipid species only identified using TOF and (##) using Orbitrap. One way-ANOVA was
394
performed to indicate statistical significant differences against the control (*P < 0.05).
395
Species of lipids are defined by their number of carbon atoms and the unsaturations of
396
their fatty acid chains.
) and 6 µM (
) and TBT at 0.1 µM (
21
). Values are relative to the cell
397
Figure 1
a] TOF
PC, Plasmalogen PC, PE, SM & DAG
Eva Gorro Pos
LipidsEva_Control_Pos1
1: TOF MS ES+
TIC
4.67e5
100
100
70
TAG & CE
60
50
%
Relative abundance (%)
90
80
Lyso PC,
Lyso Plasmalogen PC
40
30
20
10
0
0
Time
1.00
2.00
1
2
3.00
3
4.00
4
5.00
5
6.00
7.00
6
8.00
7
9.00
8
10.00
9
11.00
10
12.00
11
13.00
12
14.00
13
15.00
14
16.00
15
17.00
16
18.00
17
19.00
18
20.00
19
21.00
20
21
Retention time (min)
b] Orbitrap
PC, Plasmalogen PC, PE, SM & DAG
RT: 0,00 - 23,00
100
100
SM: 9G
NL:
4,28E7
TIC MS
Control3_2
4pouets8
95
Relative abundance (%)
90
90
85
80
80
Lyso PC,
Lyso Plasmalogen PC
75
70
70
65
TAG & CE
60
60
55
50
50
45
40
40
35
30
30
25
20
20
15
10
10
5
00
0
398
1
1
2
2
3
3
4
4
5
5
6
6
7
7
8
8
9
9
10
10
11
11
12
12
Retention time (min)
22
13
13
14
14
15
15
16
16
17
17
18
18
19
19
20
20
21
21
22
399
Figure 2
PC
33
36
31
Plasmalogen PC
29
9
9
Lyso PC
4
Lyso Plasmalogen PC
3
PE
20
22
TAG
21
21
DAG
29
29
7
7
CE
Orbitrap
12
12
SM
0
10
20
Number of identified lipids
400
23
30
TOF
40
401
Figure 3a
2.5
*
*
*
2.0
*
*
*
*
Increasing rate
*
*
*
1.5
1.0
0.5
0.0
30:0
31:0 32:0 32:1 32:2 34:0 34:1 34:2 34:3 34:4 34:5 34:6 36:0 36:1 36:2 36:3 36:4 36:5 36:6 38:1 38:2 38:3 38:4 38:5 38:6 38:8 40:0 40:1 40:2 40:3 40:4 40:5 40:6 42:2 42:3 42:4 42:5 42:6
##
##
#
## ##
PC molecular species
5.0
*
Increasing rate
4.0
*
*
*
*
3.0
*
*
*
*
**
*
*
*
*
*
2.0
1.0
0.0
30:0 30:1 30:2 32:0 32:1 32:2 34:0 34:1 34:2 34:3 34:4 36:0 36:1 36:2 36:3 36:4 36:5 38:1 38:2 38:3 38:4 38:5 38:6 38:7 40:1 40:2 40:3 40:4 40:5 40:6 40:7 40:8
##
##
##
Plasmalogen PC molecular species
2.5
*
Increasing rate
2.0
1.5
1.0
0.5
0.0
16:0
16:1
18:0
18:1
18:2
20:1
20:2
20:3
20:4
Lyso PC molecular species
3.0
*
Increasing rate
*
2.0
1.0
0.0
16:0
16:1
18:0
##
402
Lyso Plasmalogen PC molecular species
24
18:1
403
Figure 3b
10,0
*
*
*
8,0
Increasing rate
*
*
*
*
*
6,0
*
*
*
*
*
*
*
*
*
*
*
*
*
4,0
2,0
0,0
48:0 48:1 48:2 48:3 48:4 48:6 50:0 50:1 50:2 50:3 50:4 50:5 50:6 52:0 52:1 52:2 52:3 52:4 52:5 52:6 52:7 54:1 54:2 54:3 54:4 54:5 54:6 54:7 54:8 54:9
#
##
TAG molecular species
5,0
*
4,0
Increasing rate
*
*
3,0
*
*
*
*
*
2,0
1,0
0,0
32:0 32:1 32:2 32:3 34:0 34:1 34:2 34:3 34:4 36:0 36:1 36:2 36:3 36:4 36:5 36:6 38:0 38:1 328:2 38:3 38:4 38:5
##
#
DAG molecular species
5,0
*
*
Increasing rate
4,0
3,0
2,0
1,0
0,0
18:0
18:1
18:2
18:3
molecular species
CECEmolecular
species
404
25
20:4
20:5
22:6
405
Figure 3c
2,5
*
*
Increasing rate
2
1,5
1
0,5
0
32:0 32:1 34:0 34:1 34:2 34:3 36:0 36:1 36:2 36:3 36:4 36:6 38:1 38:2 38:3 38:4 38:5 38:6 38:7 40:1 40:2 40:3 40:4 40:5 40:6
#
##
# ##
## ## ##
#
PE molecular species
6
*
*
Increasing rate
5
*
4
*
*
3
*
*
*
2
1
0
14:0
406
16:0
16:1
18:0
18:1
20:0
20:1
22:0
SM molecular species
26
24:0
24:1
24:2
24:3