Enzymatic measurement of phosphatidylserine in cultured cells.

Phosphatidylserine (PS) is a quantitatively minor membrane phospholipid involved in diverse cellular functions. In this study, we developed a new fluorometric method for measuring PS using combinations of specific enzymes and Amplex Red. The calibration curve for PS measurement was linear and hyperbolic at low (0–50 µM) and high (50–1000 µM) concentrations, respectively, and the detection limit was 5 µM (50 pmol in the reaction mixture). This assay quantified PS regardless of the chain length and the number of double bonds. We applied this new method to the determination of PS content in HEK293 cells, which was validated by a recovery study and comparison with TLC-phosphorus assay. We showed that the PS content was high in sparse cells. The overexpression of PS synthase 1 elevated not only the cellular PS content but also the phosphatidylcholine (PC) and phosphatidylethanolamine (PE) contents, suggesting the conversion of PS into PE and the enhancement of PC production. This new assay for PS measurement is simple, specific, sensitive, and high throughput, and it will be useful to clarify the metabolism and biological functions of PS.

phosphatidylethanolamine (PE), respectively ( 2,3 ). On the other hand, in bacteria and yeast, PS is synthesized from CDP-diacylglycerol and L-serine ( 4,5 ). Plants use both the base-exchange pathway and the CDP-diacylglycerol pathway to make PS ( 6 ). PS is also an important precursor of PE, which is produced by PS decarboxylase (PSD) in bacteria, yeast, plant, and mammalian cells ( 7 ).
In the plasma membrane of mammalian cells, the majority (>80%) of PS normally resides on the inner leafl et of the bilayer ( 1 ). During the early phase of apoptosis, PS becomes externalized on the outside of cells ( 8 ). The surface exposure of PS is one of the recognition signals by which apoptotic cells are removed by phagocytes. The overexpression of either PSS1 or PSS2 in Chinese hamster ovary (CHO ) cells reduces the number of cells undergoing apoptosis in response to UV irradiation ( 9 ). When blood platelets are activated, PS is exposed on the platelet surface to trigger the clotting system ( 10 ). The proteolytic activity of the factor VIIa-tissue factor complex requires very high local concentrations of PS ( 11 ). In addition, PS becomes exposed on the outside of sperm during their maturation ( 12 ). Recently, Suzuki et al. reported that transmembrane protein 16F is an essential component for the Ca 2+ -dependent exposure of PS on the cell surface ( 13 ). PS activates several signaling proteins, including protein kinase C, neutral sphingomyelinase, cRaf1 protein kinase, and dynamin-1 (14)(15)(16)(17). Hsp70 and Hsc70 display highly selective interaction with PS on membranes, followed by rapid incorporation into the lipid bilayer ( 18 ). Furthermore, PS directs proteins with moderately positive charge to the endocytic pathway ( 19 ). Histone H2B, a plasminogen receptor, tethers to the surface of cells by interacting with PS on differentiated or apoptotic monocytoid cells ( 20 ).
Abstract Phosphatidylserine (PS) is a quantitatively minor membrane phospholipid involved in diverse cellular functions. In this study, we developed a new fl uorometric method for measuring PS using combinations of specifi c enzymes and Amplex Red. The calibration curve for PS measurement was linear and hyperbolic at low (0-50 µM) and high (50-1000 µM) concentrations, respectively, and the detection limit was 5 µM (50 pmol in the reaction mixture). This assay quantifi ed PS regardless of the chain length and the number of double bonds. We applied this new method to the determination of PS content in HEK293 cells, which was validated by a recovery study and comparison with TLCphosphorus assay. We showed that the PS content was high in sparse cells. The overexpression of PS synthase 1 elevated not only the cellular PS content but also the phosphatidylcholine (PC) and phosphatidylethanolamine (PE) contents, suggesting the conversion of PS into PE and the enhancement of PC production. This new assay for PS measurement is simple, specifi c, sensitive, and high throughput, and it will be useful to clarify the metabolism and biological functions of PS. -Morita, S-y., S. Shirakawa, Y. Kobayashi, K. Nakamura, R. Teraoka, S. Kitagawa, and T. Terada. Enzymatic measurement of phosphatidylserine in cultured cells. J. Lipid Res. 2012. 53: 325-330.
Supplementary key words Amplex Red • phosphatidylcholine • phosphatidylethanolamine • phosphatidylserine synthase 1 Phosphatidylserine (PS) is a quantitatively minor component of cell membranes, accounting for ‫ف‬ 2-10% of total phospholipids, but it plays important roles in biological processes such as apoptosis, blood coagulation cascade, and cell signaling, in addition to a structural role in membranes ( 1 ). The PS synthesis in mammalian cells is catalyzed by PS synthase 1 (PSS1) and PS synthase 2 (PSS2), which are responsible for the exchange of L-serine with polar head groups of phosphatidylcholine (PC) and

Recombinant plasmid construction
The human PSS1 gene (GenBank: D14694) was obtained from Kazusa DNA Research Institute (Chiba, Japan). Using PCR, an oligonucleotide encoding a myc (EQKLISEEDL)-tagged epitope was appended to the 5 ′ end of PSS1. These PCR products were ligated into the Afl II and Bam HI sites of the pIRESneo3 mammalian expression vector (Clontech, Mountain View, CA) to generate the plasmids pIRESneo3/myc-PSS1. pIRESneo3 contains the internal ribosome entry site, which permits the translation of two open reading frames from one mRNA. This expression system facilitates the establishment of pools of stably transfected cell lines whereby nearly all cells surviving in selective media express the gene of interest, as the neomycin phosphotransferase gene is expressed under the control of the same promoter ( 26 ).

Cell culture
HEK293 cells were grown in MEM supplemented with 10% heatinactivated FBS in a humidifi ed incubator (5% CO 2 ) at 37°C.

Establishment of stable transformants of myc-PSS1
HEK293 cells were transfected with pIRESneo3/myc-PSS1 using Lipofectamine Reagent and PLUS Reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. Cells were selected with 1.2 mg/ml G418 disulfate, and a large number of G418-resistant clones were pooled in one dish.

Measurement of PS, PC, and PE contents in cells
Cells were subcultured in 10 cm dishes at various cell densities in MEM supplemented with 10% FBS. After incubation for 48 h, the cells were washed with fresh medium and incubated with MEM containing 0.02% BSA for 18 h at 37°C. After incubation, the cells were chilled on ice, washed, and scraped with cold PBS. The cells were sonicated, and the cell protein concentration was measured using a BCA protein assay kit (Thermo Scientifi c). Cellular lipids were extracted by the method of Bligh and Dyer ( 27 ) and dissolved in 1% Triton X-100. The contents of PS, PC, and PE in the lipid extract from cells were measured by enzymatic assays ( 25 ). The content of PS in the lipid extract from cells was determined using TLC and phosphorus assay, as described previously ( 25,28,29 ).

Statistical analysis
The statistical signifi cance of differences between mean values was analyzed using the nonpaired t -test. Multiple comparisons were performed using the Bonferroni test following ANOVA. Differences were considered signifi cant at P < 0.05. Unless indicated otherwise, results are given as mean ± SE. The conventional method for assaying PS is TLC followed by quantifi cation of phosphate, which is low throughput and has low sensitivity. Mass spectrometry has been used to detect and identify PS molecular species ( 21 ). However, it is diffi cult to determine the total concentration of PS using mass spectrometry. Although the evaporative light-scattering detector for HPLC has been applied to quantify lipid classes, it has a limit of sensitivity of 200 ng for various phospholipids ( 22 ). Annexin V interacts strongly and specifi cally with PS and has been used to detect apoptosis by targeting for the loss of plasma membrane asymmetry ( 23 ). However, this annexin V-binding assay cannot quantify the amount of PS. The metabolic pathways and diverse biological functions of PS have been demonstrated qualitatively, but PS has not been well characterized quantitatively because of the diffi culty in measuring the small amount of PS. Therefore, a sensitive method for PS measurement is highly desired to understand the dynamic properties of PS. We previously reported enzymatic fl uorometric assays for the quantifi cation of PC, PE, and phosphatidic acid (PA) ( 24,25 ). In this study, we developed an enzyme-based, fl uorometric method for measuring PS. This new procedure can provide simple, sensitive, specifi c, and high-throughput quantifi cation of PS. Using this novel enzymatic assay, we also examined the relationship between cell density and PS content in HEK293 cells, and we evaluated the effect of PSS1 overexpression on the cellular contents of PS, PC, and PE.
Sample (10 µl) was added to Reagent S1 (10 µl) and incubated at 25°C for 240 min. After the incubation, Reagent S2 (80 µl) was added. After 15 min of incubation at room temperature, Amplex Red Stop Reagent (20 µl) was added. The fl uorescence intensity was measured using a fl uorescence microplate reader (Fluoroskan Ascent FL, Thermo Fisher Scientifi c, Rockford, IL). The excitation and emission wavelength fi lters were set at 544 and 590 nm, respectively.
To test the accuracy of the method, we conducted a recovery test, in which known quantities of POPS were added to the cellular lipid extract ( Table 1 ). The mean percentage recovery of PS was 98.6%, indicating no interference of hydrophobic compounds extracted from the cells.
To validate our novel PS assay, we compared it with the TLC-phosphorus assay. Thirty samples of HEK293 cells were analyzed by both the present method and the TLC-phosphorus method. We found that the two approaches correlated well ( r = 0.9803, y = Ϫ 0.8007 + 0.9999×) ( Fig. 3 ).

Effect of cell density on PS content in HEK293 cells
It has been diffi cult to measure the PS content in sparse cell cultures, and the relationship between cell density and

PS measurement
We developed a new method for the enzymatic measurement of PS. The three steps for the enzymatic measurement of PS are illustrated in Fig. 1: i) PS is hydrolyzed to serine and PA by PLD; ii) serine is oxidized by LAAO, which generates H 2 O 2 , ammonia, and 2-oxo-3-hydroxypropionic acid; iii) fi nally, in the presence of peroxidase, H 2 O 2 reacts with Amplex Red to produce highly fl uorescent resorufi n, which can be measured. This method requires only a 10 µl sample volume in a 96-well format.
To validate this novel assay for PS measurement, a calibration reaction was performed using POPS standard solutions. At high concentrations of PS, the curve was less linear and fi t a hyperbolic regression equation ( r = 0.9998) ( Fig. 2A ). As shown in Fig. 2B , the standard curve for PS measurement was linear at low concentrations ( r = 0.9987), and the detection limit was as low as 5 µM (50 pmol in the reaction mixture). The fl uorescence change in response to POPS was equal to that in response to soy PS or brain PS containing mixed acyl chains ( Fig. 2C ). Therefore, the chain length or the number of double bonds does not affect this PS measurement. PS and lysophosphatidylserine (LPS) increased the fl uorescence to the same extent when normalized to moles ( Fig. 2C ), indicating that PLD also hydrolyzes LPS to release serine. Thus, this measurement cannot distinguish between PS and LPS. On the other hand, PE or PC was not detected by this assay ( Fig. 2C ).

Measurement of PS in cultured cells
Cultured cells contain considerable amounts of amino acids, which confuse the PS measurement. The method of Bligh and Dyer has been widely used for lipid extraction, followed by quantifi cation of PS using radiolabeled serine and/or TLC ( 25,29,30 ). Accordingly, to remove the contaminating amino acids that react with LAAO, lipid extraction from cells is necessary for the enzymatic measurement of PS.
The LPS content in cellular membrane is generally much lower than the PS content and not detectable ( 31,32 ). Additionally, the extraction of lysophospholipids using the method of Bligh and Dyer requires acidic conditions ( 33 ). Hence, the LPS concentration may be low enough to be negligible in the cellular lipid extract prepared by the normal method of Bligh and Dyer even though the enzymatic assay measures the total concentration of PS and LPS. Fig. 1. Strategy for PS measurement. PLD catalyzes the hydrolysis of PS to PA and serine. Oxidation of serine is catalyzed by LAAO, which produces hydrogen peroxide. In the presence of peroxidase, Amplex Red reacts with hydrogen peroxide to produce highly fl uorescent resorufi n, which can be measured. The POPS standard solution was added to Reagent S1 and incubated at 25°C for 240 min. Then, Reagent S2 was added. After 15 min of incubation at room temperature, Stop Reagent was added. The fl uorescence intensity was measured using a fl uorescence microplate reader. Background fl uorescence was 11.7 ± 0.1 (A) or 14.3 ± 0.1 (B), which was subtracted from each value. Each point represents the mean ± SD of triplicate measurements. The lines were obtained by hyperbolic regression analysis (A) and linear regression analysis (B). The correlation coeffi cients were r = 0.9998 (A) and r = 0.9987 (B). C: Fluorescence changes in response to POPS, soy PS, brain PS, LPS, PE, and PC in PS measurement. Each bar represents the mean ± SD of triplicate measurements. There were no statistically signifi cant differences among POPS, soy PS, brain PS, and LPS.
CHO-K1 cells transfected with Chinese hamster PSS1 gene show only slight increases in the PS and PE contents ( 30 ). The levels of PS, PE, and PC are not signifi cantly increased in McArdle 7777 rat hepatoma cells overexpressing mouse PSS1 ( 29 ). The PS synthesis in CHO-K1 cells is remarkably inhibited by exogenous PS, which may occur through the feedback inhibition of PSS1 and PSS2 by PS ( 30,36 ). PE is generated in mammalian cells by the CDPethanolamine pathway and the PS decarboxylation pathway ( 1 ). The medium used for the culture of animal cells usually lacks ethanolamine, and cultured cells produce PE largely through the PS decarboxylation pathway ( 1,37 ). In the present study, PSS1 overexpression led to an increase in the cellular PE content ( Fig. 6C ), which may be ascribed to the conversion of excess PS into PE by PSD. The overexpression of human PSS1 induced only a modest increase in the PS content ( Fig. 6A ), presumably due to the feedback control of PSS1 and the PS decarboxylation. Notably, the overexpression of PSS1 had no effect on the PS/PE ratio ( Fig. 6F ), suggesting the cells tightly maintain the PS/PE ratio through the synthesis and decarboxylation of PS. HEK293 cells synthesize PC via the CDP-choline pathway and do not express PE N -methyltransferase ( 25 ). CTP:phosphocholine cytidylyltransferase, a rate-limiting PS content has not yet been characterized. Using the enzymatic measurement of PS, we quantifi ed the PS content in HEK293 cells at various cell densities. As shown in Fig. 4 , the PS content was constant between the medium and highest cell densities (45.2 and 120.3 µg protein/cm 2 , respectively), whereas the PS content at the lowest cell density (22.9 µg protein/cm 2 ) was signifi cantly greater than the PS content at all other cell densities. We previously reported that an increase in the density of HEK293 cells is accompanied by a decrease in the PA content, whereas the PC and PE contents increase with increasing cell density ( 24,25 ). These fi ndings suggest that the membrane phospholipid metabolism, including the synthesis and degradation of PS, is controlled by the signaling from cell-cell adhesion and/or cellular maturation.

Effect of PSS1 overexpression on PS content in HEK293 cells
To test whether the new assay can detect the change of cellular PS content, we established a HEK293 cell line stably expressing myc-PSS1 (HEK/myc-PSS1). A myc-tag was fused to the N-terminus of PSS1 for immunodetection using antimyc antibody. It has been shown that the tagging of PSS1 does not affect the enzymatic activity or the subcellular targeting of PSS1 ( 3,34 ). Fig. 5 shows that myc-PSS1 was detected by immunoblotting with anti-myc antibody. The expression level of myc-PSS1 detected using polyclonal anti-PSS1 antibody was strikingly higher than that of endogenous PSS1 in HEK293 cells. PSS1 protein exhibited a low apparent molecular mass of 37 kDa on SDS-PAGE compared with the size predicted from the amino acid sequence (55.5 kDa), which may be attributable to the extremely high content of hydrophobic amino acids in PSS1 ( 34,35 ).
By using enzymatic assays, we determined the contents of PS, PC, and PE in HEK293 and HEK/myc-PSS1 cells at similar cell densities (73.1 ± 2.4 and 68.4 ± 2.0 µg protein/ cm 2 , respectively). As shown in Fig.6A , HEK/myc-PSS1 cells exhibited a 1.3-fold increase in the PS content compared with HEK293 cells. In addition, the PC and PE contents in HEK/myc-PSS1 cells were signifi cantly higher than those in HEK293 cells ( Fig. 6B, C ). We also assessed the effect of myc-PSS1 expression on the PS/PC, PE/PC, and PS/PE ratios. The PS/PC and PE/PC ratios were slightly but signifi cantly increased in HEK/myc-PSS1 cells ( Fig. 6D, E ). On the other hand, the PS/PE ratio was similar between the two cell lines ( Fig. 6F ). The POPS standard solution was added to the lipid extract from HEK293 cells. The concentrations of PS were measured by the enzymatic assay.  Thus, the enzymatic methods may be useful to clarify the regulation of phospholipid composition.
In conclusion, we developed and validated a novel enzymatic assay for measuring PS in cultured cells. This convenient method has better specifi city and requires a smaller quantity of samples than previously reported methods, and it allows many samples to be processed in parallel. In addition, all enzymes and substrates are available commercially. The usefulness of this assay was demonstrated using HEK293 cells overexpressing PSS1. In principle, this method will be applicable to PS measurement in subcellular organelles or in animal tissues and fl uids. This enzymatic measurement may be helpful to study the biological functions of PS and its related enzymes, including PSS1, PSS2, and PSD. In the near future, the high-throughput enzymatic methods for measuring every phospholipid will be developed and replace the conventional methods. enzyme in the CDP-choline pathway, is activated by reversible binding to cell membrane lipids. PS and PE promote the binding and activation of this enzyme ( 38,39 ). It is conceivable that the PC production was enhanced by increased PS and PE in HEK/myc-PSS1 cells, which resulted in the suppression of the changes in PS/PC and PE/PC ratios. Furthermore, it is possible that PSS1 overexpression affected de novo L-serine synthesis via phosphorylated pathway and/or exogenous L-serine uptake into the cells.   6. Effect of PSS1 overexpression on membrane phospholipid composition. HEK293 and HEK/myc-PSS1 cells on 10 cm dishes were incubated in MEM containing 0.02% BSA for 18 h at 37°C. There was no difference in the densities of HEK293 and HEK/myc-PSS1 cells (73.1 ± 2.4 and 68.4 ± 2.0 µg protein/cm 2 , respectively). PS content (A), PC content (B), PE content (C), PS/ PC ratio (D), PE/PC ratio (E), and PS/PE ratio (F) of the cells were determined by enzymatic measurements of PS, PC, and PE, and protein assay. Each bar represents the mean ± SE of three measurements. * P < 0.05, signifi cantly different from HEK293 cells.