Biosynthesis and transport of phosphatidylserine in the cell

Historical perspective: phosphatidylserine and phosphatidylethanolamine from the 1800s to the present

This article provides a historical account of the discovery, chemistry, and biochemistry of two ubiquitous phosphoglycerolipids, phosphatidylserine (PS) and phosphatidylethanolamine (PE), including the ether lipids. In addition, the article describes the biosynthetic pathways for these phospholipids and how these pathways were elucidated. Several unique functions of PS and PE in mammalian cells in addition to their ability to define physical properties of membranes are discussed. For example, the translocation of PS from the inner to the outer leaflet of the plasma membrane of cells occurs during apoptosis and during some other specific physiological processes, and this translocation is responsible for profound life-or-death events. Moreover, mitochondrial function is severely impaired when the PE content of mitochondria is reduced below a threshold level. The discovery and implications of the existence of membrane contact sites between the endoplasmic reticulum and mitochondria and their relevance for PS and PE metabolism, as well as for mitochondrial function, are also discussed. Many of the recent advances in these fields are due to the use of isotope labeling for tracing biochemical pathways. In addition, techniques for disruption of specific genes in mice are now widely used and have provided major breakthroughs in understanding the roles and metabolism of PS and PE in vivo.

Membrane lipid interactions in intestinal ischemia/reperfusion-induced Injury

Ischemia, lack of blood flow, and reperfusion, return of blood flow, are a common phenomenon affecting millions of Americans each year. Roughly 30,000 Americans per year experience intestinal ischemia-reperfusion (IR), which is associated with a high mortality rate. Previous studies of the intestine established a role for neutrophils, eicosanoids, the complement system and naturally occurring antibodies in IR-induced pathology. Furthermore, data indicate involvement of a lipid or lipid-like moiety in mediating IR-induced damage. It has been proposed that antibodies recognize exposure of neo-antigens, triggering action of the complement cascade. While it is evident that the pathophysiology of IR-induced injury is complex and multi-factorial, we focus this review on the involvement of eicosanoids, phospholipids and neo-antigens in the early pathogenesis. Lipid changes occurring in response to IR, neo-antigens exposed and the role of a phospholipid transporter, phospholipid scramblase 1 will be discussed.

Serine base-exchange in rat liver nuclei

It has been shown that the incorporation of [(14)C]serine into phosphatidylserine (PS) in isolated rat liver nuclei is intrinsic to this organelle as attested by marker enzyme activity. Serine incorporation into PS was the highest in nuclei depleted of the outer membrane of the nuclear envelope (nucleoplasts) and negligible in the outer membrane. Trypsin treatment of nucleoplasts caused a strong inactivation of PS synthesis and only a moderate one of the NAD pyrophosphorylase activity, the marker enzyme of the inner nuclear membrane. We suggest that the serine base-exchange enzyme is located in the inner membrane of the nuclear envelope and accessible from the periplasmic surface of this membrane.

Effect of ethanol on ATP-induced phospholipases C and D and serine base exchange in glioma C6 cells

The effect of extracellular ATP, a nucleotide receptor agonist in the central nervous system, was investigated in glioma C6 cells on the intracellular Ca2+ level and the formation of phosphatidylethanol and phosphatidic acid in the presence and absence of ethanol (150 mM). In the cells prelabeled with [14C]palmitic acid, 100 microM ATP induced both the hydrolysis and the transphosphatidylation reactions leading to the formation of [14C]phosphatidic acid; addition of ethanol generated [14C]phosphatidylethanol. However, ATP-mediated increase in the level of [14C]phosphatidic acid was not inhibited by ethanol. Furthermore, ethanol augmented ATP-induced transient and sustained increase in the intracellular Ca2+ concentration, whereas ethanol alone did not produce any change in the intracellular Ca2+ level. These results indicate that in glioma C6 cells, ATP induces activation of polyphosphoinositide-specific phospholipase C and phospholipase D and that ethanol enhances this effect. In the present investigation we have also shown that long-term (2 days) ethanol treatment, at concentration relevant to chronic alcoholism (100 mM), decreased the incorporation of [14C]serine into phosphatidylserine. Since the effect of ethanol on ATP-induced activities of phospholipase C and phospholipase D and on serine base-exchange in glioma C6 cells differs significantly from that in cultured neuronal cells, these results may contribute to a better understanding of the mechanisms of ethanol action in cells of glial origin.

Serine base exchange enzyme activity is modulated by sphingosine and other amphiphilic compounds: possible role of positive charge in increasing the synthesis of phosphatidylserine

It has been found that sphingosine and sphingosylphosphorylcholine (amphiphilic cations) have a stimulatory, and cholesterol 3-sulfate (an amphiphilic anion), an inhibitory, effect on [14C]serine incorporation into phosphatidylserine in glioma C6 and rat liver microsomes. In glioma intact cells sphingosine stimulates phosphatidylserine synthesis in a process independent of protein kinase C, but suppressed by thapsigargin. We suggest that the stimulation of the enzyme occurs by the interaction of amphiphilic cations with the membrane cosubstrate phospholipids, leading to a charge redistribution on their phosphate groups, and hence facilitating the enzyme action. A new hypothesis concerning the mechanism of the serine base exchange reaction is discussed.

Protein kinase C inhibitors and PAF stimulate phosphatidylserine synthesis in human leucocytes

To study the regulation and turnover of phosphatidylserine (PtdSer) in human leucocytes, we investigated the effect of 12-O-tetradecanoylphorbol 13-acetate (TPA), I-O-octadecyl-2-O-methyl-rac-glycero-3-phosphocholine (ET-18-OCH3 or edelfosine), staurosporine and platelet activating factor (PAF) on [14C]serine incorporation into phospholipids. More than 80% of lipid radioactivity was in PtdSer. ET-18-OCH3 stimulated incorporation into PtdSer 5-fold, without increasing incorporation into other lipids. PAF stimulated PtdSer synthesis 3-fold after 1 h, while staurosporine stimulated the synthesis 2-fold after 3 h. TPA inhibited PtdSer synthesis. It abolished the ET-18-OCH3 stimulation, and reduced the staurosporine stimulation. ET-18-OCH3 and TPA did not significantly alter the incorporation of [14C]arachidonic acid into PtdSer, and did not increase PtdSer turnover judged from chase and stability experiments. The results demonstrate that PKC inhibitors and PAF induce increased incorporation of [14C]serine into PtdSer, while TPA inhibits stimulated PtdSer synthesis. This suggests that modulation of PtdSer synthesis may regulate PKC activity in PMN cells.

Immunochemical identification of the pssA gene product as phosphatidylserine synthase I of Chinese hamster ovary cells

We have previously shown that a Chinese hamster ovary (CHO) cell mutant defective in phosphatidylserine synthase I recovers the enzyme activity on transfection with a pssA cDNA clone isolated from the parental CHO-K1. The resultant transfectant, CDT-1, exhibited about 20-fold higher specific activity of the enzyme in the membrane fraction than CHO-K1 cells. Polyclonal antibodies against two peptides of the predicted pssA product cross-reacted with a membrane protein having an apparent molecular mass of 42 kDa, which was overproduced in CDT-1 cells. By immunoprecipitation with the antibody, phosphatidylserine synthase I activity as well as the 42-kDa protein was eliminated from solubilized membrane proteins of CDT-1 cells. Both the enzyme activity and the 42-kDa protein of CHO-K1 cells were enriched in the mitochondria-associated membrane fraction and the microsome fraction, but neither was enriched in the mitochondria fraction or the cytosol fraction. These results suggest that the pssA gene encodes phosphatidylserine synthase I.

Insulin inhibits changes in the phospholipid profiles in sciatic nerves from streptozocin-induced diabetic rats: a phosphorus-31 magnetic resonance study

Sciatic nerve phospholipids obtained from insulin-treated streptozocin-induced diabetic, non-treated streptozocin-induced diabetic, and healthy, control male Sprague-Dawley rats after eighteen weeks of diabetes were studied by 31P NMR spectrometry. Eleven phospholipids resonances were identified as follows: Phosphatidic acid (Chemical shift, 0.30 ppm), dihydrosphingomyelin (0.13 ppm), ethanolamine plasmalogen (0.07 ppm), phosphatidylethanolamine (0.03 ppm), phosphatidylserine (-0.05 ppm), sphingomyelin (-0.09 ppm), lysophosphatidylcholine (-0.28 ppm), phosphatidylinositol (-0.30 ppm), alkylacylglycerophosphorylcholine (-0.78 ppm), choline plasmalogen (-0.80 ppm), and phosphatidylcholine (-0.84 ppm). Diabetic rats showed that phosphatidylcholine was significantly elevated (p < 0.05), and ethanolamine plasmalogen and choline plasmalogen were significantly lower when compared with both control and insulin treated rats. The choline ratio (choline-containing phospholipids over noncholine phospholipids) was significantly elevated in the diabetic group, when compared with both control and insulin-treated groups. The ethanolamine ratio (ethanolamine-containing phospholipids over nonethanolamine phospholipids) and the ratio of the ethanolamine ratio over the choline ratio, was significantly elevated in the control and the insulin-treated groups when compared with the diabetic group. The presence of phosphatidic acid and the significance in phosphatidylcholine and ethanolamine plasmalogen, suggested that insulin had a role in the phosphatidylcholine metabolism in the rat nerve.

High-performance liquid chromatographic method for determination of the metabolism of polyunsaturated molecular species of phosphatidylserine labeled in the polar group

A reversed-phase HPLC method to monitor the incorporation of radiolabeled precursors into the polar group of individual polyunsaturated molecular species of phosphatidylserine (PS) is presented. PS labeled in the polar group was decarboxylated and subsequently converted to trinitrophenyl-phosphatidylethanolamine (Tnp-PE), which was separated into its molecular species by reversed-phase HPLC within 90 min, using a gradient of acetonitrile-methanol and ammonium acetate. A major feature of the method is the complete resolution of the stearoyl species, 18:0/20:4 and 18:0/22:6, at ambient temperature. By determining the amount of radioactivity incorporated into each fraction, the metabolism of individual molecular species of PS, and also of PE, labeled in the polar group can be investigated.

The selective use of stearoyl-polyunsaturated molecular species of phosphatidylcholine and phosphatidylethanolamine for the synthesis of phosphatidylserine

In rat liver microsomes, [3H]serine was incorporated primarily into the two most abundant molecular species of microsomal phosphatidylserine (PS), 18:0/20:4 and 18:0/22:6, by Ca(2+)-dependent base exchange. The pattern of PS molecular species synthesized was very similar to the species composition of PS and markedly different from the species composition of either microsomal precursor, phosphatidylcholine (PC) or phosphatidylethanolamine (PE). The data indicated that the enrichment of rat liver PS in mainly three fatty acids--stearic, arachidonic, and docosahexaenoic acids, occurred by (1) the preference by PS synthases for the stearoyl-polyunsaturated molecular species, 18:0/20:4 and 18:0/22:6, of PC and PE and (2) a discrimination against the use of the palmitoyl-polyunsaturated species, 16:0/20:4 and 16:0/22:6, and the stearoyl-diunsaturated species, 18:0/18:2. The preferential use of the two species of PC and PE, based on their acyl chain content and not on their relative abundance, demonstrates that an individual molecular species can be selected out of the total pool for a defined function.

Serine and ethanolamine incorporation into different plasmalogen pools: subcellular analyses of phosphoglyceride synthesis in cultured glioma cells

In cultured glioma cells, plasma membrane (PM) is enriched in phosphatidylserine (PtdSer) and plasmalogens (1-O-alk-1'-enyl-2-acyl-sn-glycero-3-phosphoethanolamine). Serine can be a precursor of headgroups of both PtdSer and ethanolamine phosphoglycerides (PE) including plasmalogens and non-plasmalogen PE (NP-PE). Synthesis of phospholipids was investigated at the subcellular level using established fractionation procedures and incorporation of [3H(G)]L-serine and [1,2-14C]ethanolamine. Specific radioactivity of PtdSer from [3H]serine was 2-fold greater in PM than in microsomes, reaching maximum by 2-4 h. Labeled plasmalogen from [3H]serine appeared in PM by 4 h and increased to 48 h, whereas almost no plasmalogen accumulated in microsomes within 12 h. In contrast, labeled plasmalogen from [1,2-14C]ethanolamine appeared in both PM and microsomes at early incubation times and became enriched in PM beyond 12 h. Thus, in glioma cells: (1) greater and faster accumulation of labeled PtdSer in PM may reflect direct synthesis from serine within PM; (2) PM is a major source of PtdSer for decarboxylation and PE synthesis; (3) NP-PE in both PM and microsome provides headgroup for synthesis of plasmalogen; and, (4) plasmalogen synthesis may involve different intracellular pools depending on headgroup origin.

Limited metabolic interaction of serine with ethanolamine and choline in the turnover of phosphatidylserine, phosphatidylethanolamine and plasmalogens in cultured glioma cells

Modulation of choline phosphoglyceride turnover has been investigated extensively but less is known about regulation of serine and ethanolamine phosphoglyceride synthesis and turnover. We investigated incorporation and interactions of [3H(G)]L-serine, [1,2-14C]ethanolamine and [methyl-3H]choline in cultured glioma cells. Exogenous serine did not compete with ethanolamine or choline incorporation and did not chase labeled headgroup from ethanolamine phosphoglycerides (PE); serine displaced headgroup of prelabeled phosphatidylserine (PtdSer) resulting in less labeled PtdSer for decarboxylation. In contrast, exogenous ethanolamine markedly chased labeled headgroup of non-plasmenylethanolamine phosphoglycerides (NP-PE) with less effect on plasmalogen (1-O-alk-1'-enyl-2-acyl-sn-glycero-3-phosphoethanolamine) whether headgroup was derived from [3H]serine or [14C]ethanolamine. Label in chase medium was mainly ethanolamine to 12 h; phosphoethanolamine was present with longer chase (12-48 h). Choline did not compete with serine incorporation and had little chase effect on PtdSer and PE. Choline and ethanolamine competitively interacted with preference for choline. These data suggest that (1) PtdSer synthesis in cultured glioma cells may involve more than headgroup exchange; (2) PE turnover with metabolite release to medium may involve both phospholipase D and phospholipase C; (3) acceleration of PE turnover by exogenous ethanolamine primarily involves NP-PE with lesser involvement of plasmalogen; and (4) in contrast to lack of interaction between serine and other headgroup precursors, choline and ethanolamine compete primarily at uptake.

Modulation of the serine base exchange enzyme activity of rat brain membranes by amphiphilic cations and amphiphilic anions

The biosynthesis of phosphatidylserine in mammalian tissues is catalyzed by the serine base exchange enzyme. The activity of this membrane-bound enzyme can be manipulated by amphiphiles. Amphiphilic cations, such as oleylamine, W-7, chlorpromazine, and didodecyldimethylamine, stimulate the serine base exchange activity. Amphiphilic anions, such as bis(2-ethylhexyl) hydrogen phosphate and cholesterol sulfate, inhibit the serine base exchange activity. These effects are more pronounced at pH 7.0 than at the pH optimum of 8.5 for this enzyme. Both the stimulators and the inhibitors alter the Vmax values without changing the Km value for serine, suggesting that their mechanism of action is related to interactions of the membrane-bound cosubstrate, phosphatidylethanolamine, with the membrane-bound enzyme. The optimal concentration of stimulator varies with the amount of membrane protein present; however, supraoptimal concentrations cause inhibitions. It is proposed that the amphiphilic cations enhance the interaction of the phosphorylethanolamine moiety of the membrane-bound cosubstrate with the enzyme and the amphiphilic anions interfere with such an interaction. Some of the pharmacological properties of these amphiphilic cations, employed clinically as antidepressants, may be mediated by modulation of the serine base exchange enzyme activity.

Inhibition of phosphatidylserine synthesis by glutamate, acetylcholine, thapsigargin and ionophore A23187 in glioma C6 cells

Phosphatidylserine synthesis was studied in glioma C6 cells with [14C]serine and in the presence or absence of agents which increase the level of [Ca2+]i. It was found that glutamate and acetylcholine inhibited this synthesis by up to 40%, whereas thapsigargin and the ionophore A23187 inhibited by up to 70%. The inhibitory effect of thapsigargin and the A23187 was observed in Ca(2+)-free medium. The data show that the inhibition of this synthesis is caused by the Ca(2+)-depletion from endoplasmic reticulum, suggesting that the synthesis of phosphatidylserine occurs on the luminal side of these structures and can be regulated by transmembrane signaling systems.

In utero ethanol exposure decreases the biosynthesis of phosphatidylserine in rat pup cerebrum

Phosphatidylserine is enriched in the brain and has been implicated to play a role in regulating neuronal membrane functions. In this study, three experimental protocols were used to examine the effects of in utero ethanol exposure on phosphatidylserine biosynthesis in rat pup brain, namely, (1) assay of the serine base-exchange enzyme activity in brain microsomes, (2) incubation of brain slices with [3H] serine, and (3) incorporation in vivo of [3H]serine into phosphatidylserine as well as serine-related phospholipids in brain. Results from all three protocols point to a decrease in phosphatidylserine biosynthesis in newborn rat pup cerebrum on exposure to ethanol in utero compared with the pair-fed controls. When in utero ethanol-exposed pups were nursed by mothers given a chow diet, the differences gradually returned to control levels by 17 days of age. The decrease in phosphatidylserine biosynthesis may be important in explaining some of the neuronal deficits associated with in utero ethanol exposure.

Sphingosine and unsaturated fatty acids modulate the base exchange enzyme activities of rat brain membranes

The base exchange enzymes catalyze the incorporation of L-serine, ethanolamine and choline into their corresponding phospholipids. The L-serine base enzyme activity was increased 120% by 0.1 mM sphingosine. There was a modest increase of the ethanolamine base exchange enzyme activity but the choline base exchange enzyme activity was unaffected. Na-arachadonate, Na-oleate and Na-linolenate at 0.2 mM concentration increased the activity of the L-serine and ethanolamine base exchange enzymes but inhibited the choline base exchange enzyme activity. A model is proposed suggesting that modulations of the L-serine base exchange enzyme may participate in the regulation of the calcium phospholipid-dependent protein kinase C.

Calcium-dependent incorporation of choline into phosphatidylcholine (PC) by base-exchange in rat brain membranes occurs preferentially with phospholipid substrates containing docosahexaenoic acid (22:6(n-3))

The fatty acid composition of phosphatidylcholine (PC) formed by base-exchange was examined in rat brain membranes in vitro. The free choline incorporated into subspecies of PC by this phospholipase-D type activity can be distinguished from that which might enter indirectly via the last enzyme of the de novo pathway for phospholipid biosynthesis, cholinephosphotransferase, by its ionic requirements. Choline base-exchange in lysed synaptosomes is optimal when assayed at extracellular (mM) calcium concentrations and is blocked by magnesium. As much as 40% of the choline incorporated by base-exchange into rat brain membranes was recovered in subspecies of PC, representing no more than 10% of the total PC pool, which contained docosahexaenoic acid (22:6(n-3)). Docosahexaenoic acid is enriched in electrically-excitable membranes and its content in phospholipids of rat and human brain change during early development and increase with age.

Mechanism of the ATP-dependent phosphatidylserine synthesis in liver subcellular fractions

It has been shown that the ATP-dependent incorporation of [14C]serine into phosphatidylserine in rat liver mitochondrial and microsomal fractions is prevented by EGTA. On the other hand, at low (microM) Ca2+ concentrations, serine incorporation is strongly stimulated by ATP and Mg2+. This stimulatory effect is reduced by calcium ionophore A23187. It is therefore suggested that the ATP-dependent process is that of serine base-exchange reaction, stimulated by endogenous Ca2+ accumulated inside the microsomal vesicles by Ca2+,Mg2+-ATPase. The mitochondrial activity can be accounted for by contamination by the endoplasmic reticulum.

A new pathway for phosphatidylserine synthesis in rat liver microsomes

It has been shown that the phosphate moiety of glycerophosphate is incorporated into phosphatidylserine of rat liver microsomes, but not of mitochondria. The reaction is dependent on CMP. This observation suggests that the new pathway of acyl-specific synthesis of phosphatidylserine proposed by J.P. Infante [(1984) FEBS Lett. 170, 1-14] can proceed in rat liver microsomes.

Phospholipid metabolism in mouse sciatic nerve in vivo

To probe the activities of various pathways of lipid metabolism in peripheral nerve, six phospholipid-directed precursors were individually injected into the exposed sciatic nerves of adult mice, and their incorporation into phospholipids and proteins was studied over a 2-week period. Tritiated choline, inositol, ethanolamine, serine, and glycerol were mainly used in phospholipid synthesis; in contrast, methyl-labeled methionine was primarily incorporated into protein. Phosphatidylcholine was the main lipid formed from tritiated choline, glycerol, and methionine precursors. Phosphatidylserine, phosphatidylethanolamine, and phosphatidylinositol were the main lipids formed from serine, ethanolamine, and inositol, respectively. With time there was a shift in label among phospholipids, with higher proportions of choline appearing in sphingomyelin, glycerol in phosphatidylserine, ethanolamine in phosphatidylethanolamine (plasmalogen), and inositol in polyphosphoinositides, especially phosphatidylinositol 4,5-bisphosphate. We suggest that the delay in formation of these phospholipids, which are concentrated in peripheral nerve myelin, may, at least in part, be due to their formation at a site(s) distant from the sites where the bulk of Schwann cell lipids are made. We propose that separating the synthesis of these myelin-destined lipids to near the Schwann cell's plasma membrane would facilitate their concentration in peripheral nerve myelin sheaths. At earlier labeling times, ethanolamine and glycerol were more actively incorporated into phosphatidylcholine and phosphatidylinositol, respectively, than later. The transient labeling of these phospholipids may reflect some unique role in peripheral nerve function.

Phospholipid substrate-specificity of the L-serine base-exchange enzyme in rat liver microsomal fraction

The specificity of the L-serine base-exchange enzyme towards the fatty acid composition of the phospholipid substrate was investigated with a rat liver microsomal fraction. The relative rates of L-serine incorporation into saturated-hexaenoic, saturated-pentaenoic, saturated-tetraenoic, saturated-trienoic, dienoic-dienoic, monoenoic-dienoic, saturated-dienoic and saturated-monoenoic + saturated-saturated phosphatidylserine molecular species were 42, 5, 23, 4, 5, 4, 5 and 11% respectively. This is similar to, but not identical with, the relative mass abundance of these molecular species in total liver cell phosphatidylserines. The results indicate that the substrate-specificity of the L-serine base-exchange enzyme can at least in part explain the observed fatty acid composition of rat liver phosphatidylserines.

Incorporation of serine into the phospholipids of phosphatidylethanolamine-depleted Tetrahymena

Phosphatidylserine formation and decarboxylation are decreased in Tetrahymena in which phosphatidylethanolamine has been replaced by its isosteric analog 3-aminopropylphosphonolipid (1,2-diacylglyceryl-3-O-(3-aminopropylphosphonate). The combined activity of the phosphatidylethanolamine: serine phosphatidyltransferase/ phosphatidylserine decarboxylase complex in isolated mitochondria from lipid-altered cells [J. D. Smith and D. A. Giegel (1981) Arch. Biochem, Biophys. 206, 420-423] is about 20% of the activity in mitochondria from control cells. The enzyme activity in the lipid-altered mitochondria is stimulated by the addition of exogenous phosphatidylethanolamine to the assay system while the enzymes of the control mitochondria are not. In vivo the lipid-altered cells are able to incorporate radioactivity from [3-14C]- or [3-3H]serine into phosphatidylserine and phosphatidylcholine in amounts comparable to normal cells. Thus, under conditions of "stress" (e.g., the depletion of phosphatidylethanolamine), the phosphatidyltransferase is apparantly capable of utilizing other phospholipids besides its normal substrate phosphatidylethanolamine.