Tumor suppressor and anti-inflammatory actions of PPARgamma agonists are mediated via upregulation of PTEN.

The PTEN tumor suppressor gene modulates several cellular functions, including cell migration, survival, and proliferation [1] by antagonizing phosphatidylinositol 3-kinase (PI 3-kinase)-mediated signaling cascades. Mechanisms by which the expression of PTEN is regulated are, however, unclear. The ligand-activated nuclear receptor peroxisome proliferator-activated receptor gamma (PPARgamma) [2] has been shown to regulate differentiation and/or cell growth in a number of cell types [3, 4, 5], which has led to the suggestion that PPARgamma, like PTEN [1, 6], could act as a tumor suppressor. PPARgamma has also been implicated in anti-inflammatory responses [7, 8], although downstream mediators of these effects are not well defined. Here, we show that the activation of PPARgamma by its selective ligand, rosiglitazone, upregulates PTEN expression in human macrophages, Caco2 colorectal cancer cells, and MCF7 breast cancer cells. This upregulation correlated with decreased PI 3-kinase activity as measured by reduced phosphorylation of protein kinase B. One consequence of this was that rosiglitazone treatment reduced the proliferation rate of Caco2 and MCF7 cells. Antisense-mediated disruption of PPARgamma expression prevented the upregulation of PTEN that normally accompanies monocyte differentiation and reduced the proportion of macrophages undergoing apoptosis, while electrophoretic mobility shift assays showed that PPARgamma is able to bind two response elements in the genomic sequence upstream of PTEN. Our results demonstrate a role for PPARgamma in regulating PI 3-kinase signaling by modulating PTEN expression in inflammatory and tumor-derived cells.

Lipoprotein-associated phospholipase A2: a potential new risk factor for coronary artery disease and a therapeutic target.

The recognition that atherosclerosis represents an inflammatory disease has begun to shift interest towards novel therapies that could specifically target the underlying inflammatory component of atherogenesis. Like low-density lipoprotein, an ideal new drug target would be a modifiable plasma risk factor that not only reflects the ongoing inflammatory process but also actively promotes it. Lipoprotein-associated phospholipase A2, also known as platelet-activating factor acetylhydrolase, is a new risk factor that may have the potential to fulfil these requirements.

Inhibition of lipoprotein-associated phospholipase A2 diminishes the death-inducing effects of oxidised LDL on human monocyte-macrophages.

The death of macrophages contributes to atheroma formation. Oxidation renders low-density lipoprotein (LDL) cytotoxic to human monocyte-macrophages. Lipoprotein-associated phospholipase A2 (Lp-PLA2), also termed platelet-activating factor acetylhydrolase, hydrolyses oxidised phospholipids. Inhibition of Lp-PLA2 by diisopropyl fluorophosphate or Pefabloc (broad-spectrum serine esterase/protease inhibitors), or SB222657 (a specific inhibitor of Lp-PLA2) did not prevent LDL oxidation, but diminished the ensuing toxicity and apoptosis induction when the LDL was oxidised, and inhibited the rise in lysophosphatidylcholine levels that occurred in the inhibitors' absence. Hydrolysis products of oxidised phospholipids thus account for over a third of the cytotoxic and apoptosis-inducing effects of oxidised LDL on macrophages.

Lipoprotein-associated phospholipase A(2), platelet-activating factor acetylhydrolase: a potential new risk factor for coronary artery disease.

A specific and robust immunoassay for the lipoprotein-associated phospholipase A(2) (Lp-PLA(2)), platelet-activating factor acetylhydrolase, is described for the first time. The immunoassay was used to evaluate possible links between plasma Lp-PLA(2) levels and atherosclerosis risk amongst susceptible individuals. Such an investigation was important because Lp-PLA(2) participates in the oxidative modification of low density lipoprotein by cleaving oxidised phosphatidylcholines, generating lysophosphatidylcholine and oxidised free fatty acids. The majority of Lp-PLA(2) was found associated with LDL (approximately 80%) and, as expected, enzyme levels were significantly positively correlated to LDL cholesterol. Plasma Lp-PLA(2) levels were significantly elevated in patients with angiographically proven coronary artery disease (CAD) when compared with age-matched controls, even though LDL cholesterol levels did not differ significantly. Indeed, when included in a general linear model with LDL cholesterol and other risk factors, Lp-PLA(2) appeared to be an independent predictor of disease status. We propose, therefore, that plasma Lp-PLA(2) mass should be viewed as a potential novel risk factor for CAD that provides information related to but additional to traditional lipoprotein measurements.

Total synthesis of the four stereoisomers of dihexadecanoyl phosphatidylinositol and the substrate stereospecificity of human erythrocyte membrane phosphatidylinositol 4-kinase.

A new and convenient method for the preparation of the four stereoisomers of dihexadecanoyl phosphatidylinositol has been developed. An enantiomeric pair of acid-labile, pentaprotected myo-inositol building blocks was synthesized in high yield and coupled with chiral phenyl dihexadecanoylglyceryl phosphates to give the fully protected phosphatidylinositols. These were subsequently deprotected by hydrogenolysis and self-hydrolysis in aqueous ethanol to give the desired pure products. Comparison of these compounds as potential substrates for a partially purified phosphatidylinositol 4-kinase (EC 2.7.1.67) derived from human erythrocyte membranes revealed that the chirality of the inositol ring is crucial for efficient phosphorylation, whereas the chirality of the glycerol moiety is relatively unimportant. Moreover, the similarity in phosphorylation rates of the naturally occurring mammalian phospholipid, I, and its synthetic stereochemical counterpart, compound 10a, suggests that the enzyme is relatively tolerant to changes in fatty acid composition.

Selective binding of the truncated form of the chemokine CKbeta8 (25-99) to CC chemokine receptor 1(CCR1).

Human CC chemokine receptor 1 (CCR1) has been proposed as a receptor for CKbeta8. To obtain conclusive evidence, binding-displacement studies of 125I-CKbeta8 (25-99) were performed on membranes of Chinese hamster ovary cells expressing human CCR1. The Ic50 for displacement of 125I-CKbeta8 (25-99) with CKbeta8 (25-99) was 0.22 nM. The longer forms of CKbeta8 (24-99 and 1-99) also displaced 125I-CKbeta8, with Ic50 values of 6.5 and 16 nM, respectively. Displacement profiles of 125I-CKbeta8 (25-99) on freshly prepared human monocytes indicated that CCR1 was the major receptor for CKbeta8. We conclude that CCR1 is a receptor for different-length CKbeta8 and that CKbeta8 (25-99) has a similar affinity for CCR1 as macrophage inflammatory protein-1alpha (MIP-1alpha). The longer variants of CKbeta8 are significantly less potent than CKbeta8 (25-99) and MIP-1a on CCR1 and monocytes (P < 0.05).

The role of fractalkine in the recruitment of monocytes to the endothelium.

Recombinant fractalkine possesses both chemoattractive and adhesive properties in vitro. Previous studies have demonstrated an upregulation of this molecule on the membranes of activated human endothelial cells and hypothesised that fractalkine plays a role in the recruitment and adherence of monocytes to the activated endothelium. Here we present data analysing both the adhesive and chemoattractive properties of this chemokine expressed by activated human umbilical vein endothelial cells. We demonstrate that both recombinant fractalkine and endogenously produced fractalkine function as adhesion molecules, tethering monocytes to the endothelium. However, our data demonstrate that although recombinant fractalkine has the potential to function as a potent monocyte chemoattractant, the endogenous fractalkine cleaved from activated human umbilical vein endothelial cells is not responsible for the observed chemotaxis in this model. Instead, we show that monocyte chemoattractant protein-1 (MCP-1), secreted from the activated human umbilical vein endothelial cells, is responsible for the chemotaxis of these monocytes.

Lipoprotein-associated PLA2 inhibition--a novel, non-lipid lowering strategy for atherosclerosis therapy.

Lipoprotein-associated phospholipase A2 (Lp-PLA2) is a serine lipase that is associated with low density lipoprotein (LDL) in human plasma. Substrates include oxidised phosphatidylcholine (PC), which is hydrolysed by Lp-PLA2 to lyso-PC and oxidised fatty acids. Both products are bioactive and proinflammatory, and implicated in monocyte infiltration into the developing plaque, deposition of foam cells, and plaque progression and instability. Lp-PLA2 has recently been shown to be a risk factor for coronary events in previously asymptomatic, hypercholesterolaemic men. A series of azetidinones was designed as potent and selective inhibitors of this enzyme; SB-222657 inhibited release of the chemotactic cleavage products from oxidised LDL, and SB-244323 reduced atherosclerotic plaque development in a 3 month rabbit study. A series of pyrimidones has been designed from a screening hit, and nanomolar inhibitors identified. Oral efficacy in inhibiting plasma Lp-PLA2 in rabbits has been demonstrated with a variety of structural classes.

Oxidatively modified phosphatidylserines on the surface of apoptotic cells are essential phagocytic 'eat-me' signals: cleavage and inhibition of phagocytosis by Lp-PLA2.

Diversified anionic phospholipids, phosphatidylserines (PS), externalized to the surface of apoptotic cells are universal phagocytic signals. However, the role of major PS metabolites, such as peroxidized species of PS (PSox) and lyso-PS, in the clearance of apoptotic cells has not been rigorously evaluated. Here, we demonstrate that H2O2 was equally effective in inducing apoptosis and externalization of PS in naive HL60 cells and in cells enriched with oxidizable polyunsaturated species of PS (supplemented with linoleic acid (LA)). Despite this, the uptake of LA-supplemented cells by RAW264.7 and THP-1 macrophages was more than an order of magnitude more effective than that of naive cells. A similar stimulation of phagocytosis was observed with LA-enriched HL60 cells and Jurkat cells triggered to apoptosis with staurosporine. This was due to the presence of PSox on the surface of apoptotic LA-supplemented cells (but not of naive cells). This enhanced phagocytosis was dependent on activation of the intrinsic apoptotic pathway, as no stimulation of phagocytosis occurred in LA-enriched cells challenged with Fas antibody. Incubation of apoptotic cells with lipoprotein-associated phospholipase A2 (Lp-PLA2), a secreted enzyme with high specificity towards PSox, hydrolyzed peroxidized PS species in LA-supplemented cells resulting in the suppression of phagocytosis to the levels observed for naive cells. This suppression of phagocytosis by Lp-PLA2 was blocked by a selective inhibitor of Lp-PLA2, SB-435495. Screening of possible receptor candidates revealed the ability of several PS receptors and bridging proteins to recognize both PS and PSox, albeit with diverse selectivity. We conclude that PSox is an effective phagocytic 'eat-me' signal that participates in the engulfment of cells undergoing intrinsic apoptosis.

Granulocyte/macrophage colony-stimulating factor affects myo-inositol metabolism in a novel manner. Implications for its priming action on human neutrophils.

Little is known about the signal transduction processes involved in the priming action of granulocyte/macrophage colony-stimulating factor (GM-CSF) on neutrophils. This study has used myo-[3H]inositol-labelled human neutrophils to determine whether preincubation with GM-CSF influences myo-inositol (Ins) metabolism in control cells, or in cells stimulated with the bacterial chemoattractant N-formyl-methionyl-leucyl-phenylalanine (fMetLeuPhe). GM-CSF pretreatment did not influence the total cellular 3H radioactivity content, demonstrating that the cytokine had no effect on Ins uptake. However, neutrophils pretreated with GM-CSF showed a dramatic 25-40% fall in the free [3H]Ins content of the cell, which was almost quantitatively recovered in a 2-4-fold increase in radioactivity within PtdIns. The remainder of the 3H radioactivity was found proportionately distributed throughout all other [3H]Ins-containing metabolites. Interestingly, in comparison with controls, the GM-CSF-stimulated increases in [3H]polyphosphoinositide (including 3-phosphorylated lipids) and [3H]Ins polyphosphate contents were consistently higher than that observed with PtdIns. This observation suggests that GM-CSF influences the hormone-sensitive pool of PtdIns, possibly through the activation of a PtdIns synthase which is rate-limiting to subsequent metabolic pathways. This is the first report of an action of GM-CSF on Ins metabolism, and highlights the conversion of Ins to PtdIns as a key regulatory metabolic step.

myo-inositol metabolites as cellular signals.

The discovery of the second-messenger functions of inositol 1,4,5-trisphosphate and diacylglycerol, the products of hormone-stimulated inositol phospholipid hydrolysis, marked a turning point in studies of hormone function. This review focuses on the myo-inositol moiety which is involved in an increasingly complex network of metabolic interconversions, myo-Inositol metabolites identified in eukaryotic cells include at least six glycerophospholipid isomers and some 25 distinct inositol phosphates which differ in the number and distribution of phosphate groups around the inositol ring. This apparent complexity can be simplified by assigning groups of myo-inositol metabolites to distinct functional compartments. For example, the phosphatidylinositol 4-kinase pathway functions to generate inositol phospholipids that are substrates for hormone-sensitive forms of inositol-phospholipid phospholipase C, whilst the newly discovered phosphatidylinositol 3-kinase pathway generates lipids that are resistant to such enzymes and may function directly as novel mitogenic signals. Inositol phosphate metabolism functions to terminate the second-messenger activity of inositol 1,4,5-trisphosphate, to recycle the latter's myo-inositol moiety and, perhaps, to generate additional signal molecules such as inositol 1,3,4,5-tetrakisphosphate, inositol pentakisphosphate and inositol hexakisphosphate. In addition to providing a more complete picture of the pathways of myo-inositol metabolism, recent studies have made rapid progress in understanding the molecular basis underlying hormonal stimulation of inositol-phospholipid-specific phospholipase C and inositol 1,4,5-trisphosphate-mediated Ca2+ mobilisation.

Thyrotropin-releasing hormone stimulates rapid breakdown of phosphatidylinositol 4,5-bisphosphate and phosphatidylinositol 4-phosphate in GH3 pituitary tumor cells.

Thyrotropin-releasing hormone (TRH) induced a rapid breakdown of phosphatidylinositol 4,5-bisphosphate (PtdIns 4,5P2) and phosphatidylinositol 4-phosphate (PtdIns 4P) in GH3 cells labeled to isotopic equilibrium with [3H]inositol. Within 10 sec of the addition of TRH (1 microM), there was a maximal 60% decrease in PtdIns 4,5P2 and 40% decrease in PtdIns 4P. Breakdown of phosphatidylinositol (PtdIns) occurred only after a lag of 30 sec. While the reduced levels of the polyphosphoinositides had almost returned to control values by 5 min, the GH3 cell PtdIns content remained at around 85% of controls for at least 2 hr. Both phosphatidic acid (PA) and 1,2-diacylglycerol levels increased in response to TRH in [32P]PO4- and [3H]glycerol-labeled GH3 cells. 1,2-Diacylglycerol accumulated in a biphasic manner with an early peak 10 sec after addition of the peptide. This early rise in 1,2-diacylglycerol levels coincided in time and was equivalent in lipid mass with the decrease in the polyphosphoinositide content, suggesting the involvement of a phospholipase C-type enzyme. 1,2-Diacylglycerol levels subsequently fell toward control values and, after 3 min of treatment with TRH, rose again to levels 50% above normal. PA levels reached a peak value approximately 2-fold above normal 1 min after the addition of TRH. At all times after TRH addition, the bulk of the inositol phospholipid lost was recovered as 1,2-diacylglycerol. These results suggest that TRH stimulates a cycle of events in which the breakdown of the polyphosphoinositides, PtdIns 4,5P2 and, perhaps, PtdIns 4P by a phospholipase C enzyme could be the initiating event.

Immunological identification of the major platelet low-Km cAMP phosphodiesterase: probable target for anti-thrombotic agents.

Immunoblot and enzyme-activity analyses, using specific immunological probes, indicated that more than 80% of the total low-Km cAMP phosphodiesterase activity present in bovine and human platelets resided in a single phosphodiesterase isozyme. In the presence of protease inhibitors, the platelet enzyme has an apparent subunit size of 110 kDa and appears immunologically and structurally indistinguishable from a recently purified bovine heart isozyme. When protease inhibitors were absent during homogenization and centrifugation, this platelet phosphodiesterase was susceptible to sequential proteolysis forming 80-kDa and 60-kDa peptides. As a previous report on the purification of the platelet low-Km cAMP phosphodiesterase described a 61-kDa protein, our data would suggest that this was a proteolytic fragment. Moreover, in our study a 40-70% increase in catalytic activity was associated with proteolysis. Further similarities between the platelet and heart phosphodiesterases were demonstrated by pharmacological studies that showed identical inhibitor profiles for both enzymes. Several known phosphodiesterase inhibitor compounds that have been found useful in inhibiting platelet aggregation also inhibited the platelet low-Km cAMP phosphodiesterase with potencies very similar to their antithrombotic effects. Cilostamide, Ro 15-2041, milrinone, papaverine, isobutylmethylxanthine, and theophylline inhibited the 110-kDa platelet enzyme with IC50 values of 0.04, 0.13, 0.46, 1.4, 2.6, and 110 microM, respectively.

Thyrotropin-releasing hormone-stimulated [3H]inositol metabolism in GH3 pituitary tumor cells. Studies with lithium.

GH3 pituitary tumor cells were labeled to isotopic equilibrium with [3H]inositol. Thyrotropin-releasing hormone (TRH), which has been shown to stimulate inositol phospholipid metabolism in these cells, enhanced the accumulation of [3H]inositol-derived radioactivity in the cell's acid-soluble fraction. Separation of the [3H]inositol metabolites by ion-exchange chromatography revealed that TRH induced a rapid rise in the cellular content of [3H]inositol mono-, bis-, and trisphosphate. The latter two metabolites accumulated in a multiphasic manner with an initial peak 5-10 sec after TRH addition. This was followed by a short-lived decline and a secondary rise which left the metabolite levels elevated for at least 50 min. The GH3 cell [3H]inositol monophosphate and [3H] inositol content also rose in response to TRH, but the latter accumulated with a considerably slower time course than the phosphorylated derivatives. None of these responses could be mimicked by the calcium ionophore A23187. Incubation of GH3 cells with TRH in the presence of lithium led to an enhanced accumulation of [3H]inositol monophosphate and, to a lesser extent, of [3H]inositol bis- and trisphosphate. This accumulation rose in a linear fashion with time for at least 20 min, by which point 50% of the [3H]inositol-containing phospholipids had been depleted. When lithium was added 30 min after TRH, [3H]inositol monophosphate accumulated at the same rate as was found when TRH and lithium were added together, indicating that the TRH-induced phospholipid response in GH3 cells does not desensitize. Under normal conditions, approximately equal amounts of the three [3H]inositol phosphates were formed within 5 sec of TRH addition. However, when TRH was added to cells grown chronically in lithium-containing medium, or to cells incubating at a subphysiological temperature (25 degrees), greater than 90% of the metabolites formed were the bis- or trisphosphates. This indicates that the primary event stimulated by TRH is the breakdown by phospholipase C of phosphatidylinositol 4,5-bisphosphate and, perhaps also, of phosphatidylinositol 4-phosphate.

Intact cell and cell-free phosphorylation and concomitant activation of a low Km, cAMP phosphodiesterase found in human platelets.

A monoclonal antibody (CGI-5) directed against the cGMP-inhibited phosphodiesterase isolated from bovine heart was used to examine the phosphorylation of this isozyme in human platelets. PGE1 promoted the phosphorylation of this isozyme, identified as a 110 kDa peptide following SDS-gel electrophoresis. Phosphorylation resulted in approximately a 40% increase in the cGMP-inhibited phosphodiesterase activity. Cell-free experiments demonstrated that cAMP-dependent protein kinase phosphorylated the cGMP-inhibited phosphodiesterase, and that this could be blocked by the heat stable inhibitor peptide (PKI). Phosphorylation of the cGMP-inhibited phosphodiesterase increases the Vmax for cAMP hydrolysis approximately 50%, but does not affect the Km for cAMP (0.12 microM).

Chemotactic peptides but not macrophage colony stimulating factor effect the synthesis of inositol lipids in macrophages.

Normal bone marrow derived macrophages display a wide variety of biological responses to a number of distinct agonists, for example, Macrophage Colony Stimulating Factor (M-CSF) and chemotactic peptides (such as FMLP). FMLP stimulates reactive oxygen intermediate production in these cells, whilst M-CSF stimulates DNA synthesis. We have compared the effects of these two agents on the production of novel inositol lipids in macrophages. Evidence is presented that FMLP, but not M-CSF elevate the levels of a lipid putatively identified as phosphatidylinositol-3,4-bisphosphate. The implications of this observation on proposed role of novel inositol lipids in macrophage proliferation are discussed.

Expression and functional analysis of chemokine receptors in human peripheral blood leukocyte populations.

Emerging evidence indicates that chemokine receptor expression patterns are critical in determining the spectrum of action of the chemokines. We have analysed the expression patterns of 17 chemokine receptors and two orphan chemokine receptor-like genes in various freshly prepared human peripheral blood leucocyte populations, including neutrophils, lymphocytes, and naïve and differentiated monocytes using real-time quantitative polymerase chain reaction (TaqMan). This is the first comprehensive study of chemokine receptor expression in such a wide variety of cell types. Human peripheral blood leukocyte populations were found to express a wide range of chemokine receptors that varies depending on cell type and differentiation state. Novel expression patterns of certain chemokine receptors were seen during our analysis. For example, the orphan chemokine receptor HCR was expressed at very high levels by both primary neutrophils and primary monocytes, and was further upregulated on neutrophil activation and during monocyte to macrophage differentiation. When neutrophil calcium transients were measured in response to a panel of 30 different chemokines the results clearly correlated with the chemokine receptor expression profile. For example strong calcium responses were seen in neutrophils following stimulation with the CXCR1 and CXCR2 ligands, interleukin (IL-)8, GCP-2 and Gro-beta. These data have implications for the study of the functional responses of leukocytes to external stimuli and will aid in our understanding of general leukocyte biology.

Prostaglandin F2 alpha stimulates phosphatidylinositol turnover and increases the cellular content of 1,2-diacylglycerol in confluent resting Swiss 3T3 cells.

Prostaglandin F2 alpha (PGF2 alpha); which stimulates DNA synthesis in resting 3T3 cells, also stimulates the incorporation of [32P]PO4 into phosphatidylinositol. The effect is selective for PGF2 alpha when compared with PGE1, PGE2, and PGF2 beta. Epidermal growth factor (EGF) also stimulates DNA synthesis but does not affect phosphatidylinositol turnover. PGE1, which acts synergistically with PGF2 alpha to enhance DNA synthesis, does not affect the ability of PGF2 alpha, to enhance the incorporation of [32P]PO4 into phosphatidylinositol. PGF2 alpha, also causes a small increase in the cellular content of 1,2-diacylglycerol. This effect is not shared by EGF or PGE1. Stimulation of phosphatidylinositol metabolism resulting in an increase in the cellular content of 1,2-diacylglycerol may thus constitute an event in the pathway leading to the initiation of DNA synthesis in which PGF2 alpha differs in its action from EGF.

Phosphorylation results in activation of a cAMP phosphodiesterase in human platelets.

Agents such as prostaglandins E1 and I2 which elevate cAMP levels in platelets also increase cAMP phosphodiesterase activity. Since much of the cAMP phosphodiesterase activity in human platelets is due to the cGMP-inhibited isozyme (Macphee, C. H., Harrison, S. A., and Beavo, J. A. (1986) Proc. Natl. Acad. Sci. U. S. A. 83, 6600-6663), we examined the regulation of this isozyme by prostaglandins E1 and I2 in intact platelets. Because this isozyme is a minor component of platelet protein, normally requiring several thousand-fold purification to achieve homogeneity, a specific monoclonal antibody (CGI-5) was utilized to identify and isolate the cGMP-inhibited phosphodiesterase activity. Treatment of intact platelets with the prostaglandins promoted an increase in the phosphorylation state of the cGMP-inhibited phosphodiesterase and a corresponding increase in phosphodiesterase activity. The effect on activity and phosphorylation of the cGMP-inhibited phosphodiesterase was observed within 2 min after intact platelets were exposed to the prostaglandins. The half-maximal effective dose for prostaglandin I2 (10 nM) was approximately 10-fold lower than that for prostaglandin E1. The phosphorylated, cGMP-inhibited isozyme migrated as a 110-kDa peptide following sodium dodecyl sulfate gel electrophoresis. Direct in vitro phosphorylation of the platelet cGMP-inhibited phosphodiesterase by the catalytic subunit of cAMP-dependent protein kinase caused a similar increase in phosphodiesterase activity. Treatment with PKI peptide, a specific inhibitor of cAMP-dependent protein kinase, blocked the phosphorylation and the effect on activity. Taken together, the data strongly suggest that the effects of prostaglandins E1 and I2 on platelet phosphodiesterase activity are mediated by a direct cAMP-dependent protein kinase-catalyzed phosphorylation of the cGMP-inhibited phosphodiesterase isozyme.

The stereoselective recognition of substrates by phosphoinositide kinases. Studies using synthetic stereoisomers of dipalmitoyl phosphatidylinositol.

Soluble phosphatidylinositol (PtdIns) 4- and 3-kinase activities were partially purified and characterized from human placental extracts. The placental PtdIns 4-kinase (type 3) has a Km for ATP of 460 microM and is kinetically different to a partially purified human erythrocyte, membrane-bound, PtdIns 4-kinase (type 2). These three inositol lipid kinases were then used to compare their substrate specificities against the four synthetic stereoisomers of dipalmitoyl PtdIns. Only the placental 4-kinase was influenced by the chirality of the glycerol moiety of PtdIns. However, neither of the 4-kinases was able to phosphorylate L-PtdIns and, therefore, these kinases have an absolute requirement for the inositol ring to be linked to the glyceryl backbone of the lipid through the D-1 position. Phosphoinositide 3-kinase, on the other hand, was found to phosphorylate both D- and L-PtdIns. While the 3-kinase phosphorylated exclusively the D-3 position of D-PtdIns, further analyses demonstrated that the same enzyme phosphorylated two sites on L-PtdIns, namely the D-6 and D-5 positions of the inositol ring. Some implications of these findings are discussed.