Are polyphosphoinositides involved in platelet activation?

Targeted Phosphoinositides Analysis Using High-Performance Ion Chromatography-Coupled Selected Reaction Monitoring Mass Spectrometry

Phosphoinositides are minor components of cell membranes, but play crucial roles in numerous signal transduction pathways. To obtain quantitative measures of phosphoinositides, sensitive, accurate, and comprehensive methods are needed. Here, we present a quantitative targeted ion chromatography-mass spectrometry-based workflow that separates phosphoinositide isomers and increases the quantitative accuracy of measured phosphoinositides. Besides testing different analytical characteristics such as extraction and separation efficiency, the reproducibility of the developed workflow was also investigated. The workflow was verified in resting and stimulated human platelets, fat cells, and rat hippocampal brain tissue, where the LOD and LOQ for phosphoinositides were at 312.5 and 625 fmol, respectively. The robustness of the workflow is shown with different applications that confirms its suitability to analyze multiple less-abundant phosphoinositides.

Profiling of phosphoinositide molecular species in human and mouse platelets identifies new species increasing following stimulation

Phosphoinositides are bioactive lipids essential in the regulation of cell signaling as well as cytoskeleton and membrane dynamics. Their metabolism is highly active in blood platelets where they play a critical role during activation, at least through two well identified pathways involving phospholipase C and phosphoinositide 3-kinases (PI3K). Here, using a sensitive high-performance liquid chromatography-mass spectrometry method recently developed, we monitored for the first time the profiling of phosphatidylinositol (PI), PIP, PIP₂ and PIP₃ molecular species (fatty-acyl profiles) in human and mouse platelets during the course of stimulation by thrombin and collagen-related peptide. Furthermore, using class IA PI3K p110α or p110β deficient mouse platelets and a pharmacological inhibitor, we show the crucial role of p110β and the more subtle role of p110α in the production of PIP₃ molecular species following stimulation. This comprehensive platelet phosphoinositides profiling provides important resources for future studies and reveals new information on phosphoinositides biology, similarities and differences in mouse and human platelets and unexpected dramatic increase in low-abundance molecular species of PIP₂ during stimulation, opening new perspectives in phosphoinositide signaling in platelets.

Forty five years with membrane phospholipids, phospholipases and lipid mediators: A historical perspective

Phospholipases play a key role in the metabolism of phospholipids and in cell signaling. They are also a very useful tool to explore phospholipid structure and metabolism as well as membrane organization. They are at the center of this review, covering a period starting in 1971 and focused on a number of subjects in which my colleagues and I have been involved. Those include determination of phospholipid asymmetry in the blood platelet membrane, biosynthesis of lysophosphatidic acid, biochemistry of platelet-activating factor, first attempts to define the role of phosphoinositides in cell signaling, and identification of novel digestive (phospho)lipases such as pancreatic lipase-related protein 2 (PLRP2) or phospholipase B. Besides recalling some of our contributions to those various fields, this review makes an appraisal of the impressive and often unexpected evolution of those various aspects of membrane phospholipids and lipid mediators. It is also the occasion to propose some new working hypotheses.

Translocation of an SH2-containing protein tyrosine phosphatase (SH-PTP1) to the cytoskeleton of thrombin-activated platelets

A significant protein tyrosine phosphatase (PTP) activity was found to be associated with the cytoskeleton of thrombin-stimulated platelets. Translocation of the enzyme became maximal within 1-2 min of thrombin stimulation and was suppressed by cytochalasin D or upon inhibition of aggregation. Immunoblotting as well as immunoprecipitation revealed that a PTP with two SH2 domains (SH-PTP1) displayed the same behaviour, translocation to the cytoskeleton showing the same time course as that observed for pp60c-src. We conclude that SH-PTP1 might represent a critical enzyme in the complex interplay between the various proteins regulating protein tyrosine phosphorylation in the cytoskeletal matrix.

Rapid and transient thrombin stimulation of phosphatidylinositol 4,5-bisphosphate synthesis but not of phosphatidylinositol 3,4-bisphosphate independent of phospholipase C activation in platelets

When platelets are stimulated by thrombin they immediately undergo inositol lipid hydrolysis via phospholipase C activation. However, subsequently an increased production of phosphatidylinositol 4,5-bisphosphate is observed. Phospholipases C were inhibited by lowering the cytoplasmic free calcium concentration by preincubation with Quin-2-tetra(acetoxymethyl) ester. Aggregation and secretion were also totally suppressed. Under these conditions we observed an increased labeling of phosphatidylinositol 4,5-bisphosphate, indicating a stimulation of inositol lipid kinases, independent of lipid hydrolysis by phospholipase C. Conversely the production of phosphatidylinositol 3,4-bisphosphate was totally abolished. These results suggest a different regulation of the kinases/phosphatases responsible for the production of phosphatidylinositol 4,5-bisphosphate and phosphatidylinositol 3,4-bisphosphate.

Signal transducing mechanisms in human platelets stimulated by cotton bract tannin

Cotton bract tannin is a potent stimulus for platelet aggregation and secretion. Tannin has been shown to stimulate the phosphorylation of two 19-kD and 47-kD cytosolic proteins in platelets, but earlier steps in signal transducing mechanisms of platelet activation are unknown. In this study, measurements of 32P-labeled phospholipids, 14C-labeled arachidonic acid, and levels of intracellular free calcium (Ca2+) were performed before and after the addition of thrombin (1 U/ml) or various concentrations of tannin to human platelets. The results showed that tannin induced a dose-dependent synthesis of phosphatidic acid, an early and transient hydrolysis of phosphatidylinositol monophosphate and bisphosphate, a transient synthesis of diacylglycerol, and a release of arachidonic acid metabolites. The kinetics of phosphatidic acid, diacylglycerol, and arachidonic acid metabolite synthesis were similar after platelet stimulation by tannin (75 micrograms/ml) or thrombin. Tannin also induced a reversible rise of intracellular Ca2+ due to a mobilization of the internal stores and an influx of extracellular Ca2+. These results suggest that cotton bract tannin, as thrombin, activates human platelets by phospholipase C and A2 activations, release of diacylglycerol, and mobilization of intracellular free Ca2+.

Effect of albumin-bound DHA on phosphoinositide phosphorylation in collagen stimulated human platelets

The effect of exogenous albumin-bound docosahexaenoic acid (22:6n-3) (DHA), arachidonic acid (20:4n-6) (AA), and eicosapendaenoic acid (20:5n-3) (EPA) on phosphoinositide metabolism following collagen stimulation was studied using [3H]inositol prelabelled platelets. Collagen stimulation (3 min, 1.8 micrograms/ml) increased the labelling of both phosphatidylinositol 4-monophosphate (PIP), and phosphatidylinositol 4,5-biphosphate (PIP2). Of the fatty acids tested, only pre-incubation (2 min) with DHA (20 microM) significantly attenuated the collagen-induced increased PIP and PIP2 labelling; EPA was without effect, while AA enhanced PIP labelling. Forty microM DHA was less effective at attenuating the increased PIP and PIP2 labelling even though this concentration of DHA resulted in greater inhibition of platelet aggregation. Neither concentration of DHA attenuated the increased polyphosphoinositide labelling resulting from stimulation by the endoperoxide analogue U46619, or the phorbol ester, PMA. These data suggest that the effect of DHA on attenuating the increased PIP and PIP2 labelling following collagen stimulation likely occurs before thromboxane receptor occupancy, may not occur at the level of protein kinase C activation, and could be mediated in part via a lessened synthesis of thromboxane A2.

Association of phosphatidylinositol kinase and phosphatidylinositol 4-phosphate kinase activities with the cytoskeleton in human platelets

The inositol lipid kinases were investigated in the cytoskeletons of human platelets. In the absence of added lipids the kinases were only barely detectable in the Triton-soluble fractions and undetectable in cytoskeletons of resting cells. However at least 30% of the total phosphatidylinositol kinase was present in the cytoskeleton as revealed by saturation of the enzyme. Phosphatidylinositol 4-phosphate kinase was also found in significant amounts in the cytoskeletons. On the other hand, both enzymes being only recovered in the particulate fraction of the cells, we suggest that inositol lipid kinases may be present near the anchoring points of the cytoskeletons at the membranes.

Phosphatidylinositol 4,5-bisphosphate is selectively retained by platelet-fibrin clots formed by thrombin

Stimulation of human or rabbit platelets with thrombin in the presence of fibrinogen caused a large decrease, compared with unstimulated controls, in the amount of phosphatidylinositol 4,5-bisphosphate (PIP2) that could be extracted with acidified chloroform/methanol (60% at 60 s). In contrast, stimulation in the absence of added fibrinogen increased the amount of PIP2. The decrease was specific for PIP2, because similar decreases could not be demonstrated for other phosphoinositides or phospholipids. The interaction of polymerizing fibrin with stimulated platelets was required for the decrease in PIP2, since polymerized fibrin formed by reptilase did not cause the decrease in the amount of extractable PIP2, and inhibition by glycyl-L-prolyl-L-arginyl-L-proline of polymerization of fibrin formed by the action of thrombin prevented the large decrease in extractable PIP2. The decrease in extractable PIP2 could not be explained by increased degradation of PIP2, since sufficient degradation products were not formed. Thus, when platelets are stimulated with thrombin in the presence of fibrinogen, an association of polymerizing fibrin with the stimulated platelets occurs that leads to decreased extractability of PIP2. This may mean that PIP2 forms a specific association with platelet proteins that are involved in clot retraction.

Turnover of the phosphomonoester groups of polyphosphoinositol lipids in unstimulated human platelets

The metabolic activity of the polyphosphoinositol lipids in unstimulated human platelets was studied by short-term labelling with [32P]Pi, by replacement of [32P]Pi from pre-labelled platelets with unlabelled phosphate, and by depriving the cells of metabolic ATP. Under short-term labelling conditions, the 4- and 5-phosphate groups of phosphatidylinositol 4-phosphate (PtdIns4P) and phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P2] had the same specific 32P radioactivity as the gamma-phosphate of metabolic ATP. The specific 32P radioactivity of the 1-phosphates of phosphatidylinositol, PtdIns4P and PtdIns(4,5)P2 was similar, but only 4-13% compared to that of the ATP-gamma-phosphate. When [32P]Pi pre-labelled platelets were incubated with up to 25 mM of unlabelled phosphate, the displacement of the 32P label from PtdIns4P, PtdIns(4,5)P2 and metabolic ATP followed similar kinetics. Inhibition of ATP regeneration in platelets pre-labelled with [32P]Pi resulted in a rapid fall in metabolic ATP with a much slower fall in [32P]PtdIns(4,5)P2, whereas [32P]PtdIns4P increased initially. However, ATP turnover was not abolished, as indicated by the marked (25% of the control) incorporation of extracellular [32P]Pi into PtdIns4P and PtdIns(4,5)P2 in metabolically inhibited platelets. This low phosphate turnover may explain the relative resistance of PtdIns4P and PtdIns(4,5)P2 to metabolic inhibition. We conclude that PtdIns4P and PtdIns(4,5)P2 are present as a single metabolic pool in human platelets. Turnover of the 4- and 5-phosphates of PtdIns4P and PtdIns(4,5)P2 in unstimulated platelets is as rapid as that of the gamma-phosphate of metabolic ATP, and accounts for about 7% of basal ATP consumption.

Characterization of phosphoinositide-specific phospholipase C from human platelets

Phosphoinositide-specific phospholipase C (PI-PLC) from human platelet cytosol was purified 190-fold to a specific activity of 0.68 mumol of phosphatidylinositol (PI) cleaved/min per mg of protein. It hydrolyses PI and phosphatidylinositol 4,5-bisphosphate (PIP2), but not phosphatidylcholine, phosphatidylserine or phosphatidylethanolamine. The enzyme exhibits an acid pH optimum of 5.5 and has a molecular mass of 98 kDa as determined by Sephacryl S-200 gel filtration. It required millimolar concentrations of Ca2+ for PI hydrolysis, whereas micromolar concentrations are optimal for PIP2 hydrolysis. Mg2+ could substitute for Ca2+ when PIP2, but not PI, was used as the substrate. EDTA was more effective than EGTA in inhibiting the basal PI-PLC activity towards PIP2. Sodium deoxycholate strongly inhibits the purified PI-PLC activity with either PI or PIP2 as substrate. Ras proteins, either alone or in the form of liposomes, have no effect on PI-PLC activity.

Activation of phospholipase C associated with isolated rabbit platelet membranes by guanosine 5'-[gamma-thio]triphosphate and by thrombin in the presence of GTP

Rabbit platelets were labelled with [3H]inositol and a membrane fraction was isolated in the presence of ATP, MgCl2 and EGTA. Incubation of samples for 10 min with 0.1 microM-Ca2+free released [3H]inositol phosphates equivalent to about 2.0% of the membrane [3H]phosphoinositides. Addition of 10 microM-guanosine 5'-[gamma-thio]triphosphate (GTP[S]) caused an additional formation of [3H]inositol phosphates equivalent to 6.6% of the [3H]phosphoinositides. A half-maximal effect was observed with 0.4 microM-GTP[S]. The [3H]inositol phosphates that accumulated consisted of 10% [3H]inositol monophosphate, 88% [3H]inositol bisphosphate ([3H]IP2) and 2% [3H]inositol trisphosphate ([3H]IP3). Omission of ATP and MgCl2 led to depletion of membrane [3H]polyphosphoinositides and marked decreases in the formation of [3H]inositol phosphates. Thrombin (2 units/ml) or GTP (4-100 microM) alone weakly stimulated [3H]IP2 formation, but together they acted synergistically to exert an effect comparable with that of 10 microM-GTP[S]. The action of thrombin was also potentiated by 0.1 microM-GTP[S]. Guanosine 5'-[beta-thio]diphosphate not only inhibited the effects of GTP[S], GTP and GTP with thrombin, but also blocked the action of thrombin alone, suggesting that this depended on residual GTP. Incubation with either GTP[S] or thrombin and GTP decreased membrane [3H]phosphatidylinositol 4-phosphate ([H]PIP) and prevented an increase in [3H]phosphatidylinositol 4,5-bisphosphate ([3H]PIP2) observed in controls. Addition of unlabelled IP3 to trap [3H]IP3 before it was degraded to [3H]IP2 showed that only about 20% of the additional [3H]inositol phosphates that accumulated with GTP[S] or thrombin and GTP were derived from the action of phospholipase C on [3H]PIP2. The results provide further evidence that guanine-nucleotide-binding protein mediates signal transduction between the thrombin receptor and phospholipase C, and suggest that PIP may be a major substrate of this enzyme in the platelet.

Thin-layer chromatography of polyphosphoinositides from platelet extracts: interference by an unknown phospholipid

Different ratios of radioactive polyphosphoinositides in platelets pulse-labelled with 32p-orthophosphate have been reported by various laboratories. We studied whether these differences originate from differences in methodology. Extracts of 32p-Pi labelled human platelets were prepared at various times after gel-filtration and phosphatidylinositol (PI)-, mono (PIP)- and bisphosphate (PIP2) were separated by thin-layer chromatography using four different solvent systems. The 32p-levels in PIP and PIP2 remained constant during one hour after gel-filtration, whereas 32p-PI increased continuously and more than doubled within the first h. In two of the systems PIP co-chromatographed with a radioactive compound which separated well from PIP in the two other systems. This unknown compound was also labelled with 3H-glycerol, 3H-inositol and 3H-arachidonic acid, but it was metabolically and functionally different from the polyphosphoinositides. Both the co-chromatography of this unknown phospholipid and the increase in 32p-PI in gel-filtered platelets can explain the difference in 32p-labelling in phosphoinositides reported in the literature.

Hydrolysis of polyphosphoinositides in human platelets

Phospholipase C was purified 110 fold from human platelets. The activity of the enzyme was totally dependent upon Ca2+. The activity of the enzyme was markedly enhanced in the presence of arachidonic acid and was strongly inhibited by aminoglycoside antibiotics. The enzyme hydrolyzed endogenous polyphosphoinositides in addition to PI in Ca2+ dependent manner, suggesting the involvement of this enzyme in stimulus-linked rapid hydrolysis of polyphosphoinositides in platelets. The stimulation by thrombin of 32P-labelled human platelets induced about 30% decrease in 32P-TPI and about 220% increase in 32P-PA at the first 10 sec. The degree of hydrolysis of TPI was dependent upon the amount of agonist and it was not affected by the extracellular concentration of Ca2+. The changes in 32P-phospholipids in thrombin-stimulated platelets in the absence of Ca2+ were inhibited in a dose dependent manner by preincubation with relatively higher amount of quin 2 AM. The inhibition was completely overcome by an addition of CaCl2 to the suspending buffer. By such treatment in the absence of extracellular Ca2+, the intracellular Ca2+ concentration was significantly lowered below the basal level (less than 100 nM). Those observations suggest that TPI breakdown in thrombin-stimulated platelets is primary mediated by the agonist receptor coupling and requires at least the basal level of intracellular Ca2+.

Inositol lipids, phosphatidate and diacylglycerol share stearoylarachidonoylglycerol as a common backbone in thrombin-stimulated human platelets

Gel-filtered human platelets were stimulated with 5i.u. of thrombin/ml for times up to 1 min. The fatty acid composition of inositol-containing phospholipids, phosphatidic acid and diacylglycerol was determined by g.l.c. in control and thrombin-stimulated platelet suspensions. Inositol phospholipids were found to have similar proportions of stearic and arachidonic acids, the sum of these representing 86.6% of the total fatty acids in phosphatidylinositol (PtdIns), 76.9% in phosphatidylinositol 4-phosphate (PtdIns4P) and 85.4% in phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P2]. However, arachidonic and stearic acids were less abundant in phosphatidic acid (PtdA) and diacylglycerols in non-stimulated platelets. A transient decrease in the mass of PtdIns(4,5)P2 was observed after 5-10s of thrombin stimulation, followed by an increase after 30s. The amounts of PtdIns4P and PtdIns decreased throughout the experiment. A transient accumulation of stearoylarachidonoylglycerol was observed at 5s, whereas stearoylarachidonoylglycerol 3-phosphate (PtdA) was produced in increasing amounts throughout the experiment. The decrease in inositol-containing phospholipids was not fully compensated for by the production of diacylglycerol or PtdA [or PtdIns(4,5)P2] at 1 min. All the changes in inositol phospholipids, as well as those observed in diacylglycerols and PtdA, were due to a parallel reduction or increase in the contents of stearic and arachidonic acids, with a stoichiometry equal to 1. Taken together, this suggests an interconversion of all these lipids with the utilization of a common backbone, stearoylarachidonoylglycerol. The deacylation of this diacylglycerol could account for up to 4-5nmol of arachidonate/10(9) platelets after 1 min stimulation by thrombin.

Tight coupling of thrombin-induced acid hydrolase secretion and phosphatidate synthesis to receptor occupancy in human platelets

Human platelets incubated with [32P]Pi and [3H]arachidonate were transferred to a Pi-free Tyrode's solution by gel filtration. The labile phosphoryl groups of ATP and ADP as well as Pi in the metabolic pool of these platelets had equal specific radioactivity which was identical to that of[32P]phosphatidate formed during treatment of the cells with thrombin for 5 min. Therefore, the 32P radioactivity of phosphatidate was a true, relative measure for its mass. The thrombin-induced formation of[32P]-phosphatidate had the same time course and dose-response relationships as the concurrent secretion of acid hydrolases. 125I-alpha-Thrombin bound maximally to the platelets within 13s and was rapidly dissociated from the cells by hirudin; readdition of excess 125I-alpha-thrombin caused rapid rebinding of radioligand. This binding-dissociation-rebinding sequence was paralleled by a concerted start-stop-restart of phosphatidate formation and acid hydrolase secretion. [3H]Phosphatidylinositol disappearance was initiated upon binding but little affected by thrombin dissociation and rebinding. ATP deprivation caused similar changes in the time courses for [32P]-phosphatidate formation and acid hydrolase secretion which were different from those of [3H]phosphatidylinositol disappearance. The metabolic stress did not alter the magnitude (15%) of the initial decrease in phosphatidylinositol-4,5-bis[32P]phosphate, but did abolish the subsequent increase of phosphatidylinositol-4,5-bis[32P]-phosphate in the thrombin-treated platelets. It is concluded that in thrombin-treated platelets (1) phosphatidate synthesis, but not phosphatidylinositol disappearance, is tightly coupled to receptor occupancy and acid hydrolase secretion in platelets, (2) successive phosphorylations to phosphatidylinositol-4,5-bisphosphate is unlikely to be the main mechanism for phosphatidylinositol disappearance, and (3) only a small fraction (15%) of phosphatidylinositol-4,5-bisphosphate is susceptible to hydrolysis.

Fertilization increases the polyphosphoinositide content of sea urchin eggs

Fertilization of the sea urchin egg stimulates a wave of exocytosis of cortical vesicles, but the mechanism by which fertilization regulates this secretion is not fully understood. We describe here experiments which suggest that polyphosphoinositide metabolism could be a factor in this regulation. We find that the cortical vesicle exocytosis in the egg of Strongylocentrotus purpuratus is preceded by a 40% increase in its content of triphosphoinositide (TPI) and a 22% increase of diphosphoinositide (DPI).

Activation of phospholipase C in thrombin-stimulated platelets does not depend on cytoplasmic free calcium concentration

Human platelets loaded with the fluorescent Ca2+ indicator quin2 and with different radioactive compounds including [3H]serotonin, [14C]arachidonic acid (AA) and [32P]orthophosphate were stimulated by thrombin under conditions producing secretion. In the absence of external Ca2+ (Ca2+e), cytoplasmic free [Ca2+], [Ca2+]i, increased to 340 nM, against 1685 nM at 1 mM [Ca2+]e. In both cases, diglyceride and phosphatidic acid production proceeded at the same rate, whereas AA release was inhibited at low [Ca2+]i. It is concluded that, at variance with phospholipase A2, phospholipase C activation does not depend on [Ca2+]i. These results give further support to the hypothesis of a Ca2+-independent pathway of cell activation involving phospholipase C and protein kinase C.

Triphosphoinositide breakdown and dense body release as the earliest events in thrombin-induced activation of human platelets

The activation by thrombin of human platelets prelabelled with 32P induced a 30-40% decrease in 32P-triphosphoinositides (TPI) in the first 10 sec; the decrease in the other 32P-labelled phosphoinositides occurred by 20-30 sec. At 10 sec., the intensity of these effects was maximum with 0.2-0.4 U/ml thrombin. Under these conditions, 53, 20 and 15% of the dense granule, alpha-granule and lysosome constituents, respectively were released and thromboxane B2 synthesis reached only 10% of its maximum. Together with experiments carried out with chlorpromazine - or PGE1 - treated platelets, our results suggest the existence of a close relationship between TPI-breakdown and dense body release which appear to be the earliest events resulting from the activation of human platelets by thrombin.