Materials and Methods
Contents
Cell culture conditions and maintenance
To investigate the eicosanoid response of human cells, two immortalised cell lines were selected (HaCaT keratinocytes and 46BR.1N fibroblasts). Immortalised cell lines were used as they offer unlimited lifespan, reproducible results and unrestricted quantities, due to the fact they indefinitely proliferate (Segrelles et al., 2011). However, the modification to the genetic code of the cells can result in aberrant functions and behaviour (Lewis et al., 2006), therefore all results need to be confirmed with primary cells. Here, immortalised skin cell lines were treated with various inflammatory stimuli and used to validate the corresponding model predictions.
Cell lines
Human adult low-calcium high-temperature (HaCaT) keratinocytes were purchased from CLS Cell Lines Service GmbH. This is a non-tumorigenic, spontaneously transformed, immortal human epidermal keratinocyte cell line which displays no major functional defects (Boukamp et al., 1988). The recommended cell culture medium for HaCaT keratinocytes was DMEM containing 10% FBS.
The 46BR.1N human fibroblast cell line was purchased from The European Collection of Cell Culture (ECACC) (Salisbury, UK). This cell line represents immortalised dermal fibroblasts, showing normal cell morphology and function (Arlett et al., 1988). The recommended cell culture medium for 46BR.1N fibroblasts was MEME containing 15% FBS, L-glutamine (200nM), non-essential amino acids (200 nM) and sodium pyruvate (200 nM).
Cells were seeded into plastic dishes, typically 75 cm2 culture flasks or 100 mm Petri dishes for UV experiments, and were grown in the recommended cell culture medium at 37°C, 5% CO2 and 95% humidity. The media was changed every 2-3 days depending on the cell growth, and to subculture the cells, they were washed with PBS (without Ca2+/Mg2+) and detached using trypsin/EDTA. The cell suspension was neutralised using an equal volume of serum-containing medium. Cells were centrifuged at 200xg for 3 min, and the pellet was reconstituted in the appropriate culture media. For maintenance, both cell lines were normally passaged at a dilution ratio of 1:2 three times a week.
Cell counting
Cell counting was performed automatically by using a TC20 cell counter. The reconstituted cell pellet (10 µL) was mixed with 0.4% trypan blue solution (1:1 dilution). The automated counter reading was expressed as a number of cells/mL, and the total amount of cells were calculated using the following equation.
Total number of cells (cells⁄mL)=Number of cells × Original volume (mL)
Cell treatments
UVR
Both cell lines were irradiated using a Herbert Waldmann 236 B (UV6) UV lamp (Germany). The lamp was switched on for at least 10 min prior to the experiment to allow it to warm-up. Directly before any experiment, the UV irradiance (mW/cm2) was measured using Variocontrol UV meter (Waldmann, Germany). The time of exposure was re-adjusted to the required dose using the UV irradiance and was calculated using the following equation.
UV Dose (mJ⁄cm^2 )=UV irradiance (mW⁄cm^2) × Time (s)
All cell types (80% confluent, grown in 100 mm Petri dishes) were washed with PBS and then fresh PBS (5 mL) was added. The lid of the Petri dish was opened and cells were irradiated with 15 mJ/cm2 UV dose. The PBS was then removed and cell-appropriate serum-free medium (10 mL) was added. Non-irradiated cells were subjected to the same procedure, but the UV lamp was switched off. The cells were incubated for the required amount of time (0.5, 1, 3 and 6 h).
A23187
A23187 treatments were prepared by dissolving commercially available A23187 (5 mg) in DMSO (955 µL) to a final 10 mM stock solution. The stock solution was stored at -20°C in 100 uL aliquots. Each aliquot was defrosted once and then disposed of. To fresh cell-appropriate serum free media, calcium chloride (1.8 mM) was added. Treatment media was prepared by directly pipetting A23187 (10 mM) into the culture medium at no more than 0.1% v/v of DMSO. For example, the 5 μM dose was prepared by adding 9 μL of A23187 (10 mM) to 18 mL of cell-appropriate serum free media. The control treatment was made by dilution of 9 μL of DMSO in to 18 mL of cell-appropriate serum free media.
Prior to treatment, both cell types were washed with PBS (5 mL/dish), and replaced with the cell-appropriate treatment media (18 mL/dish) and incubated for the required amount of time (0.5, 1, 3 and 6 h).
ATP
ATP treatments were prepared by dissolving commercially available ATP (551.1 mg) in water (10 mL) to a final 100 mM stock solution. The stock solution was stored at -20°C in 500 uL aliquots. Each aliquot was defrosted once and then disposed of. Treatment media was prepared by directly pipetting the ATP stock solution (100 mM) into the culture medium. For example, the 2 mM dose was prepared by adding 360 μL of ATP (100 mM) to 18 mL of cell-appropriate serum free media. The control treatment was made by dilution of 360 μL of water in to 18 mL of cell-appropriate serum free media.
Prior to treatment, both cell types were washed with PBS (5 mL/dish), and replaced with the cell-appropriate treatment media (18 mL/dish) and incubated for the required amount of time (0.5, 1, 3 and 6 h).
Calcium Ionophore and COX inhibited
Indomethacin treatments were prepared by dissolving commercially available indomethacin (17.9 mg) in DMSO (5 mL) to a final 10 mM stock solution. The stock solution was stored at -20°C in 100 uL aliquots. Each aliquot was defrosted once and then disposed of. Treatment media was prepared by directly pipetting the indomethacin stock solution (10 mM) into the cell-appropriate serum free culture medium at no more than 0.1% v/v of DMSO. For example, the 10 μM dose of indomethacin was prepared by adding 18 μL of indomethacin (10 mM) to 18 mL of cell-appropriate serum free media. The control treatment was made by dilution of 18 μL of DMSO in to 18 mL of cell-appropriate serum free media.
Prior to treatment, both cell types were washed with PBS (5 mL/dish), and replaced with the cell-appropriate treatment media (18 mL/dish) and incubated for 1 hour. The consequential A23187 treatments were prepared by dissolving commercially available A23187 (5 mg) in DMSO (955 µL) to a final 10 mM stock solution. The stock solution was stored at -20°C in 100 uL aliquots. Each aliquot was defrosted once and then disposed of. To fresh cell-appropriate serum free media, calcium chloride (1.8 mM) was added. Treatment media was prepared by directly pipetting A23187 (10 mM) into the culture medium at no more than 0.1% v/v of DMSO. For example, the 5 μM dose was prepared by adding 9 μL of A23187 (10 mM) to 18 mL of cell-appropriate serum free media. The control treatment was made by dilution of 9 μL of DMSO in to 18 mL of cell-appropriate serum free media.
After incubation, both cell types were washed with PBS (5 mL/dish), and replaced with the cell-appropriate serum free treatment media (18 mL/dish) and incubated for the required amount of time (0.5, 1, 3 and 6 h).
MTT Assay Method
Seeding density optimisation
In order to identify the exponential growth phase of HaCaT keratinocytes and 46BR.1N fibroblasts, eight seeding densities (0 – 2 x106 cells/well) were seeded into 6 well-plates in complete media. The cells were then incubated for two days in order to reach 80% confluency.
Dose optimisation of UVR
In order to assess the toxicity of UVR, cell lines were seeded into 6 well-plates in complete media until fully 80% confluent. Monolayers were then irradiated with 0-100 mJ/cm2 (for more details see section *) and cultured for 6h in serum-free culture media.
Dose optimisation of A23187
In order to assess the toxicity of A23187, cell lines were seeded into 6 well-plates in complete media until fully 80% confluent. Monolayers were then treated with 0-10 µM of A23187 (for more details see section *) and cultured for 6h in serum-free culture media.
Dose optimisation of ATP
In order to assess the toxicity of ATP, cell lines were seeded into 6 well-plates in complete media until fully 80% confluent. Monolayers were then treated with 0-32 mM of ATP (for more details see section *) and cultured for 6h in serum-free culture media.
Cell viability assessment
After seeding, treatment or irradiation, cells were washed with PBS. The MTT reagent (0.5 mg/mL) was added to each well (2.5 mL/well in a 6 well plate, 200 µL/well in a 96 well plate). The plate was returned to the incubator for 4 hours at 37ºC, 5% CO2, 95% humidity. The reagent was then aspirated and DMSO (2.5 mL/well in a 6 well plate, 200 µL/well in a 96 well plate) was added to each well to release formazan crystals. The plate was gently agitated for 10 min and 200 μL of each well content was transferred to a 96-well plate in triplicate. The blank sample was represented by DMSO to subtract the background reading. Absorbance (A) was measured at 540 nm. The mean absorbance of three technical repeats (triplicate wells) was regarded as one independent experiment. The percent of viability was assessed using the formula:
Viability (%)= ("Mean A of the sample - Mean A of the blank" )/("Mean A of the control - Mean A of the blank" )×100%
Proteomic Analysis Method
Proteomic analysis was performed at the Stoller Biomarker Discovery Centre, University of Manchester, under the guidance of Professor A. Whetton. Samples were prepared by Dr I. Baricevic-Jones and J. Kelsall. Data handling and processing was performed by Dr D. Lee.
Cell Lysis
Phosphatase inhibitor reagent was formulated to a 10x concentration by adding 1 PhoSTOP tablet to dH2O (1 mL) and mixed by vortexing. Lysis buffer was made up using 1M TEAB (2500 µL, final 0.5M), 5% SDS (50 µL, final 0.05%), PhoSTOP 10X (500 µL, final 1X), 25U/µL Benzonase (10 µL, Final 50 U/mL) and dH2O (1940 µL). The cell pellets were retrieved from storage at -80°C. Cells were lysed by the addition of lysis buffer (200 µL) and incubation on ice for 30 min. The cell lysate was mixed by vortexing every 10 mins. Samples were checked for clarity to ensure DNA had degraded. After incubation, the lysate was centrifuged (10,000 xg, 4°C, 10 min) and the supernatant transferred to a fresh microfuge tube. A 10 µL aliquot was analysed in duplicate for protein content.
Protein Quantitation
BCA assay working reagent was formulated by adding Reagent A (1mL) to Reagent B (50mL). 10 µL of standards and samples were transferred to a clear flat bottomed 96-well plate. Cell lysates were diluted 1 in 5 or 1 in 15 with lysis buffer. 200 µL of working reagent was added to each well. The plate was sealed and vortexed for 1 min. The plate was incubated at 37°C for 20 min. The optical density was read on the Spectramax i3 at 562 nm wavelength.
Digestion
Samples were diluted to a concentration of 50 µg protein per 120uL in ammonium bicarbonate (25mM). Samples were denatured and reduced in a final volume of 150µL containing DTT (5mM) and 1% sodium deoxycholate and incubated at 60ºC for 30 min. Proteins were then alkylated in iodoacetamine (50mM) and incubated at room temperature in the dark for 30 min. Digestion was performed by incubation at 37ºC overnight in the presence of trypsin at a 1:20, enzyme: protein ratio. Following trypsin incubation, sodium deoxycholate was precipitated from solution in 0.5% formic acid and centrifuged (12000 xg, 10ºC, 10 min). The supernatant was transferred to a fresh microfuge tube and the peptides vacuum dried.
Reconstitution
The dried peptides were reconstituted in 10 µL sample mix per injection containing 8.8 µL 2% Acetonitrile/0.1% Formic acid, 1 µL 100 fmol/µL PepCalMix (Final conc. 10 fmol/µL) and 0.2 µL 10x iRT Standard (Final 2 IE/µL). Samples were analysed by SWATH-MS with a micro-flow LC-MS system comprising an Eksigent nanoLC 400 autosampler and an Eksigent nanoLC 425 pump coupled to a 6600 Triple-TOF mass spectrometer with a DuoSpray Ion Source (Sciex). The analytical column was a YMC-Triart C18 12nm, 3µm, 0.3mm I.D. x 150mm, 1/32” column (YMC Europe GmbH TA12S03-15HORU) and the trap Column was YMC-Triart C18 12nm, S-3µm, 5 x 0.5mm I.D., 1/32” column (YMC Europe GmbH TA12S03-E5JORU). Samples were eluted on the following LC gradient at a flowrate of 5 µL /min.
Data analysis
Data-dependent acquisition (DDA) files were processed through Stoller computational pipeline by Dr D. Lee. The DDA files were run through X!Tandem database search engine against in-house canonical human sequence. The False discovery rates for each peptide-spectral match (PSM) were computed using Qvality (5% q-value threshold). Peptides and proteins mapped onto these PSMs were extracted.
Fatty acid analysis Method
Preparation of internal standard and FAME cocktail
To correct for variability during the extraction, a fixed amount of IS was added to each sample. To create the IS solution, heneicosanoic acid (C21:0) (0.025g) was dissolved in BHT:CHCl3:CH3OH (25 mL). The solution (1 mg/m; v/w) was then stored at 4°C for up to three months.
Vaccenic acid methyl ester (10 mg/mL) was dried under N2. The residue was reconstituted in DCM (1 mL) to create a stock solution. The working stock solution (200 ng/µL) was created by diluting the stock solution (200 µL) with DCM (800 µL). Both the stock and the working stock solutions were stored in an amber vial and kept at -20°C for up to a year.
Similarly, docosapentaenoic acid methyl ester (10 mg/mL) was dried under N2. The residue was reconstituted in DCM (1 mL) to create a stock solution. The working stock solution (200 ng/µL) was created by diluting the stock solution (200 µL) with DCM (800 µL). Both the stock and the working stock solutions were stored in an amber vial and kept at -20°C for up to a year.
To make the split FAME cocktail, commercially available FAME standards cocktail (1 mg/mL, 20μL), vaccenic acid methyl ester (200 ng/μL, 20μL) and docosapentaenoic acid methyl ester (200 ng/μL, 20μL) and DCM (140 μL) were mixed an amber vial. This cocktail was used for the split injection method.
To make the splitless FAME cocktail, the split FAME cocktail (20 μL) was diluted in DCM (40 μL) in an amber vial and stored at -20°C. To accurately determine retention times, the split/splitless FAME cocktail was prepared fresh before any fatty acid analysis.
Lipid extraction
BHT (0.03g) was dissolved in chloroform (200 mL) and methanol (100 mL) (abbreviated as BHT:CHCl3:CH3OH). The solution was stored at room temperature and used for up to a week. All samples were kept on ice, unless stated otherwise. N2 was used to dry the samples to avoid oxidation and remove solvent.
To each cell pellet, ice-cold water (1 mL) was added and transferred to a flat-bottomed tube. To each sample, BHT:CHCl3:CH3OH (4 mL) was added and incubated for 30 min. To separate the organic and aqueous layer, samples were vortexed, centrifuged (1,500xg, 5 min, 4°C) and using a glass pasteur pipette, the organic layer was collected and transferred into a glass round-bottomed tube. Additional BHT:CHCl3:CH3OH (4 mL) was added to the flat-bottomed glass tube containing the water layer and cell pellet. The vortex, centrifuge, separation and BHT:CHCl3:CH3OH replenishment steps were then repeated three times. Pellets containing the aqueous layer were stored at -20°C until analysed for protein content.
Dry sodium sulphate (approximately 9g or until the satisfactory snow storm effect was achieved) was added to the organic extract. The extract was then filtered through pasteur pipettes packed with glass wool filters (previously conditioned with BHT:CHCl3:CH3OH (3 mL)) into a new round bottomed tube. Additional BHT:CHCl3:CH3OH (1 mL) was used to rinse the glass pasteur pipette. A gentle stream of N2 was used to dry the samples. BHT:CHCl3:CH3OH (1 mL) was added to each sample and sealed using parafilm, and stored at -20°C for up to seven days.
Fatty acid derivatization
Prior to use, Hamilton syringes were washed with DCM. After drying the samples under a gentle stream of N2, the IS solution (C21:0; 1 mg/mL; v/w) (15 µL) was added using a Hamilton syringe. Between samples, syringes were washed with DCM. To remove the solvent, samples were dried using N2. BF3CH3OH (14%, 250 μL) and toluene: methanol (1:1 v/v; 250 μL) was added to each sample. Each tube was closed tightly, sealed with parafilm, vortexed and placed on a heating block (100˚C, 90 min). Glass tubes were then placed on ice (10 min). K2CO3 (10g) was dissolved in deionised water (100 mL) (stored for up to a week at room temperature). To each tube, 10% K2CO3 solution (1.5 mL) and isooctane (2 mL) were added, vortexed and centrifuged (5 min, 1,500xg, 4°C) to separate the organic and aqueous layer. Using a pasteur pipettes, the organic layer containing FAME was transferred to a new round-bottomed glass tube and kept on ice. The replenishment of 10% K2CO3 solution, isooctane, centrifugation and recovery of organic layer steps were then repeated twice. The samples were dried under N2 and reconstituted in DCM (60 μL). The samples were transferred to clean conical inserts in injection glass vials, using a Hamilton syringe and sealed with parafilm. Vials were stored for up to a week at -20°C until GC analysis.
GC-FID analysis
Each sample (1 μL) was injected into the GC heated port via an auto-sampler (room temperature) using the splitless injection method. Helium was used as a carrier gas to transfer the vaporised sample to the head of the column. The sample was run on the column (65 min) followed by a post-run (10 min) to re-initialise. The flame in the FID was a hydrogen and air mixture (1:10) and was fixed at 250ºC. The column temperature was initially set at 70ºC for 2 min.
By employing a temperature gradient with four temperature ramps, a clear resolution and separation of peaks were ensured. The first ramp included changing the temperature at a rate of 20ºC/min until it reached 150ºC, this condition was then held for 5 min. The second ramp included changing the temperature at a rate of 2.5ºC/min until it reached 218ºC with no holding time. The third ramp included changing the temperature at a rate of 30.6ºC/min until it reached 225ºC with 10 min holding time. The final ramp included changing the temperature at a rate of 42.5/min until it reached 230ºC, this condition was held for 3 min.
At the beginning of each run, several blank samples (DCM) were tested to check that the system is clean, followed by multiple injections of the splitless FAME cocktail, to check for instrument sensitivity, and determine retention times. All peaks were integrated using OpenLab software (Version 2.1.0433). The relative amount of each FAME was calculated by normalising the peak area against the IS peak area. DCM was used as the blank injection to minimise carryover and were applied between the FAME cocktail and the duplicate sample injections.
Eicosanoid Analysis Method
Internal standard and calibration stock solutions
The COX/LOX assay included a wide range of lipid mediators that were divided into two main categories: COX - prostanoids and related lipid mediators and LOX - hydroxy fatty acids and related compounds. All lipid mediators were diluted to form 10 ng/μL stock solutions with ethanol, which were stored in amber vials at -80 ºC for up to a year. The IS cocktail (1 ng/μL) for the COX/LOX assay was prepared by mixing PGB2-d4, 12-HETE-d8, 8(9) EET-d11 and 8,9-DHET-d11 (all at 10 ng/μL in ethanol, 100 μL each) and ethanol (600 μL) using Hamilton syringes in an amber vial. The vial was mixed thoroughly, sealed with parafilm and stored at -20 ºC for up to three months. Both calibration lines were prepared at 100pg/μL. The calibration line solution for the COX assay was prepared by diluting PGE1, PGD1, 6-keto PGF1α, 13,14-dihydro-15-keto PGE1, 13,14-dihydro-15-keto PGF1α, PGF1α, PGD2, PGE2, 15-keto-PGE2, 13,14-dihydro-PGE1, 13,14-dihydro-PGF2α, 13,14-dihydro-PGF1α, 13,14-dihydro-15-keto-PGF2α, PGF2α, 8-iso-PGF2α, PGF3α, Δ12-PGJ2, PGJ2, TXB2, TXB3, 15-deoxy-Δ12,14-PGJ2, 13,14-dihydro-15-keto-PGE2, PGD3 and PGE3 (10ng/μL, 10μL each) and ethanol (760μL) in an amber vial. The calibration line solution for the LOX assay was prepared by diluting 9-HODE 13-HODE, RvD1, RvD2, MaR1, PDX, 11-HDHA, 4-HDHA, 7-HDHA, 10-HDHA, 13-HDHA, 14-HDHA, 17-HDHA, 20-HDHA, RvE1, LTB4, 14,15-DHET, 11,12-DHET, 8,9-DHET, 5,6-DHET, 5(6)-EET, 11(12)-EET, 14(15)-EET, 8(9)-EET, 5-oxoETE, 15-HETrE, 5-HETE, 8-HETE, ± 9-HETE, 11-HETE, 12-HETE, 15-HETE, 20-HETE, 5-HEPE, 8-HEPE, 9-HEPE, 11-HEPE, 15-HEPE, 18-HEPE, 12-HEPE, 9-HOTrE, 13-HOTrE, 19(20)-DiHDPA, 9(10) EpOME, 12(13) EpOME, 9-oxoODE, 13-oxoODE, 5(15)- DiHETE, 8(15)-DiHETE, 19(20) EpDPE, 16(17) EpDPE, trans-EKODE 9,10-DiHOME, 12,13-DiHOME standards (all at 10 ng/μL, 10 μL each) and ethanol (460 μL) in an amber vial. Solutions for both calibration lines were sealed with parafilm and stored at -80 ºC for up to three months.
For quantification, a working stock solution (10 pg/μL) of lipid standards was prepared. This was achieved by mixing lipid standards (100 pg/μL, 20 μL) with ethanol (180 μL) using Hamilton syringes. To create the calibration line a serial dilution was performed, where the working stock solution (10 pg/μL, 100 μL) was first diluted in ethanol (100 μL) and labelled the 5 pg/μL standard, this was then repeated to create the 2.5, 1.25 and 0.625 pg/μL standards. The final 100 μL of 0.625 pg/μL standard was discarded.
The COX/LOX internal standard cocktail (20 μL of 1 ng/μL PGB2-d4, 12-HETE-d8, 8(9) EET-d11 and 8,9-DHET-d11) was added to all vials. Lipid standards were consequently dried under N2 and reconstituted in ethanol (100 μL). Amber vials were stored at -20ºC until the UPLC-ESI-MS/MS analysis for up to a week.
Solid phase extraction
To defrost cell culture media, samples were placed on ice. Once defrosted, methanol (1.77 mL) and the internal standard cocktail of COX/LOX species (20 µl) was added to each sample. The samples were then incubated for 15 min on ice, and then centrifuged (300 xg, 10 min). To each sample HCl (0.6 mL, 1M) was added to adjust the pH to 3.0 (verified with narrow range pH paper strips). SPE cartridges were prepared by washing with 100% methanol (6 mL) and water (6 mL). After acidify the samples, they were immediately transferred to the SPE. The SPE cartridges were then washed with 15% methanol (6 mL), water (6 mL) and hexane (6 mL). To collect the lipid extract methyl formate (6 mL) was added. Samples were then dried under N2. The remaining lipid residue was dissolved in ethanol (100%, 100 µL), vortexed and centrifuged (300 xg, 1 min). The samples were transferred to 100 µL insert vials, placed in amber glass vials and stored at -20 °C for up to one week for analysis.
UPLC-MS/MS analysis
To analyse samples, the sample chamber of the UPLC/ESI-MS/MS was set up at 8°C and the column temperature was set to 25 °C. The source temperature was 150°C, the desolvation temperature was 500°C and the capillary voltage was set at 3100V.
Collision energy settings, cone voltage and multiple reaction monitoring (MRM) transitions for each detectable compound are described in tables * and *. The MRM transitions, cone voltage and collision energy values were optimised manually or using an IntelliStart protocol. Liquid chromatographic analysis was performed on a C18 column (Acuity UPLC BEH, 1.7 µm, 2.1 x 50 mm; Waters) and argon was used as the collision gas. The injection volume was 3 µL (each sample was injected twice) and the flow rate was 0.6 mL/min. The COX/LOX assay was performed by two separate chromatographic runs, one for COX and one for LOX. The running time was 5 min for the COX assay and 5.8 min for the LOX assay.
The COX analysis was performed using an acetonitrile-based gradient system of two solvents, A and B. Solvent A was: ultrapure water: acetic acid, 100:0.02 (v/v); solvent B was: acetonitrile: acetic acid, 100:0.02 (v/v). The LOX analysis was performed using an isocratic system of two solvents, C and D. Solvent C was: ultrapure water: acetic acid, 100:0.02 (v/v); solvent D was: acetonitrile: acetic acid, 100:0.02 (v/v). The gradient between mobile phases varied depending on the assay and is shown in Table * and Table * . The instrument was operated in a negative ionisation mode and MS/MS settings were different for COX and LOX injections as shown in the Table *.
At the beginning of each run, several blank samples (ethanol) were tested to check that the system is clean, and several injections of a cocktail of standards are tested, to check for instrument sensitivity, and correct resolution of compounds. For quality control purposes, each sample was run twice and ethanol blanks were run after each sample to minimise carryover.
Detection and quantification
To detect peaks at the appropriate retention times, MassLynx (Target Lynx extension) methods were used. A calibration line was created in the software of the respective COX and LOX lipid standards. All samples were analysed within the linear range of the calibration curve (Figure * and *). The calibration curve was used to normalise sample areas using the slope and the intercept. To quantify the sample, the sample area was normalised against the appropriate internal standard peak area. The appropriate internal standard equates to the deuterated internal standard of the closest family of lipids e.g. PGB2-d4 only for PGs. The final concentration of each metabolite was expressed in pg/μL of injected extract. The average of duplicate injection was taken and multiplied by the reconstitution volume. This total concentration was normalised against the number of cells per sample (pg/106 cells for culture media). The limit of detection was set to signal to noise ratio of 3. The limit of quantitation was set to signal to noise ratio of 10.
