Lipidomic data uncover extensive heterogeneity in phosphatidylcholine structural variants in HepG2 cells

The data contain information related to the research article entitled “Profiling of promoter occupancy by the SND1 transcriptional coactivator identifies downstream glycerolipid metabolic genes involved in TNFα response in human hepatoma cells” (DOI: 10.1093/nar/gkv858). In the article alluded to, we reported that tumor necrosis factor alpha (TNFα) increases notably the cellular content of the major glycerolipid phosphatidylcholine (PC). Here, accompanying lipidomic data determine the PC structural variants that have been identified in human hepatoma HepG2 cells and those whose relative abundance is modified by TNFα. We used ultrahigh performance liquid chromatography (UHPLC) coupled to electrospray ionization (ESI) tandem mass spectrometry (MS/MS)-based lipidomic profiling to analyze lipid extracts of control and TNFα-treated HepG2 cells. The identity of PC individual species was elucidated using the values of the retention time and molecular weight in addition to the fragmentation patterns. MS data were then processed and analyzed for the characterization of statistically significant differences in detected structural variants. We have annotated the dataset of PC species that characterize HepG2 cells' phenotype, both under normal and pro-inflammatory conditions.


a b s t r a c t
The data contain information related to the research article entitled "Profiling of promoter occupancy by the SND1 transcriptional coactivator identifies downstream glycerolipid metabolic genes involved in TNFa response in human hepatoma cells" (DOI: 10. 1093/nar/gkv858). In the article alluded to, we reported that tumor necrosis factor alpha (TNFa) increases notably the cellular content of the major glycerolipid phosphatidylcholine (PC). Here, accompanying lipidomic data determine the PC structural variants that have been identified in human hepatoma HepG2 cells and those whose relative abundance is modified by TNFa. We used ultrahigh performance liquid chromatography (UHPLC) coupled to electrospray ionization (ESI) tandem mass spectrometry (MS/MS)-based lipidomic profiling to analyze lipid extracts of control and TNFa-treated HepG2 cells. The identity of PC individual species was elucidated using the values of the retention time and molecular weight in addition to the fragmentation patterns. MS data were then processed and analyzed for the characterization of statistically significant differences in detected structural variants. We have annotated

Data
The Tables and Figures provided in this Data in Brief article gather the raw data of individual phosphatidylcholine (PC) species that have been identified in total lipid extracts of HepG2 cells and highlight those whose relative abundance is modulated by TNFa. Acute TNFa treatment of HepG2 cells is known to increase total cellular PC [1]. The methodology here adopted, tandem mass spectrometry (MS/MS) preceded by a UHPLC separation step, uncovers the heterogeneity in PC species enabling the identification of PC molecular variants at the level of the fatty acids that are bound through an ester bond to the sn-1 and sn-2 position of the glycerol backbone. The most intense ions detected for PC are in ESI (electrospray ionization) in the positive-ion mode (ESIþ, Fig. 1A

Experimental factors
Cancer cell culture, lipid extraction and PC variants' analysis by UHPLC-MS E -based lipidomics.

Experimental features
Human hepatoma HepG2 cells were cultured in the presence or the absence of TNFa, then lysed and protein determined. Lipids were extracted from cell lysates using organic solvents following standard procedures and the lipid extracts were submitted to the analysis core facility of the University for PC molecular species identification and quantification.

Data source location
Lipids & Liver Research Group, Department of Physiology, Faculty of Medicine and Nursing, University of the Basque Country UPV/EHU, Leioa, Spain.

Data accessibility
The data are available in this article.

Related research article
"Profiling of promoter occupancy by the SND1 transcriptional coactivator identifies downstream glycerolipid metabolic genes involved in TNFa response in human hepatoma cells", (DOI: 10.1093/nar/gkv858).

Value of the Data
These data facilitate a guidance for further insights into the environment dependency of the human cancer cells lipid code. This analysis is an example of how lipidomics helps to uncover perturbations of individual lipid species that, while blinded by a global quantitative approach, may be relevant to intracellular signaling and cell function. The phosphatidylcholine structural variability found in hepatocarcinoma cells might serve as potential biomarker for cancer-associated inflammation.
from both ESIþ and ESI-analysis have been used for a proper assignment. The set of individual PC species detected in HepG2 cells together with their peak markers (retention time and m/z value) and structure are annotated in  very low and that it is agreed they introduce overfitting and statistical noise [2], the sodium adducts were not taken into account for the compositional estimation. Hence, individual PC species are described using relative abundances (%) of the protonated molecules [M þ H]þ in the positive-ion mass spectra obtained by the peak integration of all PCs (Fig. 1B). We annotate in Table 2 the identity and relative abundance of the protonated variants detected both in control and in TNFa-treated HepG2 cells. Fig. 2A shows clearly that the composition of PC in terms of the number of carbons and unsaturations of fatty acyls is highly heterogeneous. Interestingly, two species, PC(32:1) or PC(16:0/16:1) and PC(34:1) or PC(16:0/18:1), account for nearly 50% of the total intensity while more than ten species exhibit a representation lower than 1%. Statistical analysis of the differences between groups revealed that PC synthesis was not globally up-regulated by TNFa as there was a significant change in the relative abundance of six out of the 23 individual species examined according to the t-test (numerical P values are listed in Table 2 and included in Fig. 2B). Three species were up-regulated, namely PC(36:2), PC(36:4) and PC(36:5), and an opposite trend was seen in PC(30:1), PC(32:1) and PC(32:2), (data are shown as Box-Whisker plots in Fig. 2B). The absolute quantitative data (in nmol/mg protein) of Table 1 Phosphatidylcholine (PC) individual species and their corresponding peak markers (mass to charge ratio (m/z) and retention time (RT) pairs) detected in HepG2 cells. E, experiment. The PC specific structure at the level of the fatty acid bound to the sn-1 and sn-2 position of the glycerol backbone was assigned as described in the Experimental Design, Materials, and Methods section.   individual lipid species may be calculated as their relative abundances multiplied by the total PC concentration in the corresponding sample. In parallel experiments aimed at replicating those reported previously [1], we measured PC mass in aliquots of the lipid extracts [3]. We found that it averaged 27.20 ± 1.23 nmol/mg protein in the control and 32.69 ± 3.27 nmol/mg protein in the treated group (mean ± SD), indicating that TNFa increased the absolute PC mass in HepG2 cells by an extent (20.18%) similar to that earlier found [1]. For those readers that are not familiar with this type of lipidomic approach, it is important to outline that present data provide the identification of PC structural variants whose TNFa-promoted increase is above or below the average 20.18% rise. In other words, present dataset enables the identification of prevalent combinations of fatty acids in the PC class of phospholipids. Remarkably, pro-inflammatory TNFa seems to favor the contribution of PC species containing di-and polyunsaturated fatty acids (like eicosanoids) in detriment of less unsaturated fatty acid-containing species.
2. Experimental design, materials, and methods
Cultures were either left untreated or treated with TNFa (50 ng/ml) for 24 h before harvesting as described [1]. Cells were then lysed and lipids were exhaustively extracted from cell lysates with CHCl 3 / MeOH [4], dried in a Savant SpeedVac concentrator, and stored at À80 C under N 2 atmosphere prior to their injection into the UHPLC-MS/MS system. Aliquots of lipid extracts were analyzed for the content of total PC using our developed thinelayer-chromatography/optical densitometry method [3].

UHPLC-MS/MS analysis
Ultrahigh performance liquid chromatography was carried out using an ACQUITY UHPLC system from Waters (Milford, MA, USA) equipped with a binary solvent delivery pump, an autosampler and a column oven. Separation of the PC structural variants was performed with Acquity UHPLC HSS T3 column (100 Â 2.1 mm, 1.8 mm) and precolumn (VanGuard, 1.8 mm) as thoroughly described elsewhere [5]. All UHPLC-MS E data were acquired on a SYNAPT G2 HDMS, with a quadrupole time of flight (Q-TOF) configuration (Waters, Milford, MA, USA) equipped with an ESI source operated in positive or negative ion mode. The mass spectrometer was operated in the continuum MS E acquisition mode for both polarities. During this acquisition method, the first quadrupole is operated in a wide band rf mode only, allowing all ions to enter the T-wave collision cell. Two discrete and independent interleaved acquisition functions were automatically created: the first function, typically set at 6 eV, collected low energy or unfragmented data, whereas the second function collected high energy or fragmented data typically obtained by using a collision energy ramp from 15e40 eV. The other MS parameters were described previously in Ref. [6]. Representative chromatograms in the positive-and in the negative-ion mode are shown in Fig. 1A and an example of PC mass spectra in Fig. 1B.

Lipid identification and MS data processing
For lipid identification, the data generated by the UHPLC-MS E analysis in positive and negative ion mode were extracted using MS E data viewer (Waters MS Technologies, Manchester, UK) generating an exportable text file, which was used for lipid identification using SimLipid software (Premier Biosoft, USA). This software matches the exact masses of the precursor and product ions of unknown lipids with those on an in-silico database containing over 40,298 lipid species and 1,509,305 structure specific MS/MS characteristic ions. SimLipid assigns a probability score to the unknown lipid structure according to the best fit of experimental m/z values with theoretical m/z values of both precursor and product ions of the SimLipid database. By matching the exact masses of the characteristic fragment ions, in addition to precursor ions, SimLipid was able to identify isomers with similar m/z. Tolerance for MS and MS/MS used for identification was 5 mDa. Then, diagnostic fragments of the polar head group or the fatty acyls chains were investigated to confirm the annotation proposed by the database and discriminate isomers. Fig. 3   18:1) during our MS E analyses. In addition to this, elution order was used to minimize misidentification. Within the PC class, retention time decreases with the number of double bonds and increases with the length of the acyl chains. Lipids that did not meet these two conditions were considered false positives.
For MS data processing and, in order to convert the three-dimensional LCeMS raw data (RT, m/z, intensity) into a table of time-aligned detected features, with their RT, m/z, and intensity in each sample, data were acquired with the MassLynx V4.1 software and processed using the Databridge 3.5 converter (Waters, Milford, USA), XCMS 1.42.0 (Metlin, La Jolla, CA, USA), as reported previously [6]. Individual PC species were quantitated using relative abundances (%) of the protonated molecules [M þ H]þ in the positive-ion mass spectra obtained by the peak integration of all PCs. The final dataset were exported into SPSS V20 for statistical analysis.

Statistical analyses
Data are reported as the mean ± SD and an unpaired Student's t-test was carried out to compare control and TNFa-treated groups (SPSS Statistics version 24, IBM, Armonk, NY, USA). Statistical significance is defined as P 0.05.

Funding
This work was supported by the Basque Government Departments of Education (grant IT971-16) and Economic Promotion (grant KK2018/00090), Basque Country, Spain.