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