Ceramide inhibition of mammalian phospholipase D1 and D2 activities is antagonized by phosphatidylinositol 4,5-bisphosphate

Ceramides inhibit phospholipase D (PLD) activity in several mammalian cell types. These effects have been related to preventing activation by ARF1, RhoA, and protein kinase C-alpha and -beta and therefore indicate that PLD1 is inhibited. In the present work, we investigated the effects of ceramides in inhibiting both PLD1 and PLD2 and the interaction with another activator, phosphatidylinositol 4,5-bisphosphate (PIP2). PLD1 and PLD2 were overexpressed separately in Sf9 insect cells using baculovirus vectors. In our cell-free system, PLD1 activity was inhibited completely by C2-ceramide at sub-optimum concentrations of PIP2 (3 and 6 microM), whereas at supra-optimum PIP2 concentrations (18 and 24 microM) C2-ceramide did not inhibit PLD1 activity. Partially purified PLD2 exhibited an absolute requirement for PIP2 when the activity was measured using Triton X-100 micelles. Ceramides inhibited PLD2 activity, and this inhibition was decreased as PIP2 concentrations increased. However, C2-ceramide also reversibly inhibited the activity of PLD1 and PLD2 mutants in which binding of PIP2 was decreased, indicating that ceramides are interacting with the catalytic core of the mammalian PLDs. By contrast, C2-ceramide failed to produce a significant inhibition of PLDs from bacteria and plants. Our results provide a novel demonstration that ceramides reversibly inhibit mammalian PLD2 as well as PLD1 activities and that both of these actions are more pronounced when PIP2 concentrations are rate-limiting.     

Potent inhibition of mammalian progesterone synthesis by 2 alpha-cyanoprogesterone.

2 alpha-Cyanoprogesterone potently inhibits the conversion of [3H]pregnenolone into progesterone catalysed by bovine corpora lutea, bovine adrenal cortex and human term placenta microsomes (microsomal fractions), yielding IC50 (concentration causing 50% inhibition) values of 66 nM, 120 nM and 700 nM respectively. By contrast, it is an exceedingly poor inhibitor of the isomerization of pregn-5-ene-3,20-dione, yielding IC50 values between 50 and 70 microM. On this basis, 2 alpha-cyanoprogesterone would appear to be an extraordinarily selective inhibitor of the 3 beta-hydroxysteroid dehydrogenase. Dixon plots indicate that it is a very-tight-binding competitive inhibitor of the corpus-luteum enzyme, yielding a Ki of 15 nM. In the bovine adrenal cortex and human placenta the steroid is less potent and inhibits the dehydrogenase non-competitively with Ki values of 150 nM and 1.0 microM respectively. Thus 2 alpha-cyanoprogesterone inhibits the corpus-luteum dehydrogenase with substantial selectivity. Because of its high affinity for the ovarian enzyme, the presence of low-micromolar concentrations of 2 alpha-cyanoprogesterone can promote a complete cessation of progesterone synthesis in corpora-lutea microsomes for several hours. Since this effect is observed in the presence of saturating concentrations of pregnenolone (50 microM), it is predicted that this inhibitor may be even more potent in vivo. 2 alpha-Cyanoprogesterone displays very low affinity for the human progesterone receptor, yielding a Kd of 600 nM as against a Kd of 1.6 nM for progesterone. It is suggested that 2 alpha-cyanoprogesterone may be a selective inhibitor of ovarian progesterone synthesis and may act as an effective anti-gestational agent in vivo.     

Substrate specificity in phospholipid transformations by plant phospholipase D isoenzymes

Phospholipase D (PLD) catalyzes the hydrolysis and transesterification of glycerophospholipids at the terminal phosphodiester bond. In many plants, several isoforms of PLD have been identified without knowing their functional differences. In this paper, the specificities of two PLD isoenzymes from white cabbage (Brassica oleracea var. capitata) and two ones from opium poppy (Papaver somniferum L.), which were recombinantly produced in Escherichia coli, were compared in the hydrolysis of phospholipids with different head groups and in the transphosphatidylation of phosphatiylcholine with several acceptor alcohols. In a biphasic reaction system, consisting of buffer and diethyl ether, the highly homologous isoenzymes are able to hydrolyze phosphatidylcholine, -glycerol, -ethanolamine, -inositol and - with one exception - also phosphatidylserine but with different individual reaction rates. In transphosphatidylation of phosphatidylcholine, they show significant differences in the rates of head group exchange but with the same trend in the preference of acceptor alcohols (ethanolamine > glycerol ? l-serine). For l- and d-serine a stereoselectivity of PLD was observed. The results suggest a physiological relevance of the different hydrolytic and transphosphatidylation activities in plant PLD isoenzymes.     

Identification of a phosphoinositide binding motif that mediates activation of mammalian and yeast phospholipase D isoenzymes.

Phosphoinositides are both substrates for second messenger-generating enzymes and spatially localized membrane signals that mediate vital steps in signal transduction, cytoskeletal regulation and membrane trafficking. Phosphatidylcholine-specific phospholipase D (PLD) activity is stimulated by phosphoinositides, but the mechanism and physiological requirement for such stimulation to promote PLD-dependent cellular processes is not known. To address these issues, we have identified a site at which phosphoinositides interact with PLD and have assessed the role of this region in PLD function. This interacting motif contains critical basic amino acid residues that are required for stimulation of PLD activity by phosphoinositides. Although PLD alleles mutated at this site fail to bind to phosphoinositides in vitro, they are membrane-associated and properly localized within the cell but are inactive against cellular lipid substrates. Analogous mutations of this site in yeast PLD, Spo14p, result in enzymes that localize normally, but with catalytic activity that has dramatically reduced responsiveness to phosphoinositides. The level of responsiveness to phosphoinositides in vitro correlated with the ability of PLD to function in vivo. Taken together, these results provide the first evidence that phosphoinositide regulation of PLD activity observed in vitro is physiologically important in cellular processes in vivo including membrane trafficking and secretion.     

Specific inhibition of rat brain phospholipase D by lysophospholipids

Although the importance of phospholipase D (PLD) in signal transduction in mammalian cells is well documented, the negative regulation of PLD is poorly understood. This is primarily due to a lack of known specific inhibitors of PLD. We herein report that the activity of partially purified rat brain PLD is inhibited by certain lysophospholipids, such as lysophosphatidylinositol, lysophosphatidylglycerol, and lysophosphatidylserine in a highly specific manner. Inhibition of PLD by lysophospholipids was dose-dependent: the concentration of lysophosphatidylinositol required for half-maximal inhibition was about 3 micrometer. An analysis of the enzyme-kinetics suggested that lysophospholipids act as non-competitive inhibitors of PLD activity. As expected, PLD activity was stimulated by ADP-ribosylation factor (Arf) and phosphatidylinositol 4,5-bisphosphate (PIP(2)). The inhibition of PLD by lysophospholipids, however, was not affected by the presence or absence of Arf or by an increase in PIP(2) concentration. A protein-binding assay suggested that lysophospholipids bind directly to PLD. These results indicate that the observed inhibition of PLD by lysophospholipids is due to their direct interaction rather than to an interaction between lysophospholipids and either Arf or PIP(2). The present study suggests that certain lysophospholipids are specific inhibitors of rat brain PLD in a cell-free system and may provide the new opportunities to investigate mechanisms by which PLD is regulated by lysophospholipids, presumably liberated by phospholipase A(2) activation, in mammalian cells.     

Potent inhibition of mammalian ribonucleases by 3', 5'-pyrophosphate-linked nucleotides.

Molecular modeling based on the crystal structure of the complex of bovine pancreatic RNase A with the inhibitor 5'-diphosphoadenosine 3'-phosphate (ppAp) (Leonidas, D. D., Shapiro, R., Irons, L. I., Russo, N., and Acharya, K. R. (1997) Biochemistry 36, 5578-5588) was used to design new inhibitors that extend into unoccupied regions of the enzyme active site. These compounds are dinucleotides that contain an unusual 3',5'-pyrophosphate linkage and were synthesized in solution by a combined chemical and enzymatic procedure. The most potent of them, 5'-phospho-2'-deoxyuridine 3'-pyrophosphate, P' --> 5'-ester with adenosine 3'-phosphate (pdUppAp), binds to RNase A with Ki values of 27 and 220 nM at pH 5.9 and 7, respectively. These values are 6-9-fold lower than those for ppAp and 50-fold lower than that for the transition state analogue, uridine vanadate. pdUppAp has broad specificity; it is an effective inhibitor of at least two other members of the pancreatic RNase superfamily, human RNase-2 (eosinophil-derived neurotoxin) and RNase-4, which share only 36-44% sequence identity with the pancreatic enzyme. The potency of pdUppAp and the other inhibitors described here depends critically on the extended internucleotide linkage; the pyrophosphate group enhances dinucleotide binding to the three RNases by 2.1-2.9 orders of magnitude, as compared with a monophosphate. These data give further insight into the organization of the catalytic centers of the various RNases. Moreover, the new class of inhibitors provides a useful means by which to probe the biological actions of these and other related enzymes.     

Potent inhibition of dendritic cell differentiation and maturation by vitamin D analogs

We show that the immunosuppressive effects of 1alpha, 25-dihydroxyvitamin D(3) (1alpha,25(OH)(2)D(3)) are due, in part, to inhibition of the T cell stimulatory functions of dendritic cells (DCs). Addition of 10(-12) and 10(-8) M 1alpha,25(OH)(2)D(3) to murine DC cultures resulted in a concentration-dependent reduction in levels of class II MHC and the co-stimulatory ligands B7-1, B7-2, and CD40 without affecting the number of DCs generated. Higher concentrations of 1alpha,25(OH)(2)D(3) reduced DC yield. The capacity of DCs to induce proliferation of purified allogeneic T cells was reduced by 1alpha,25(OH)(2)D(3). The vitamin D(3) analog, 1alpha,25(OH)(2)-16-ene-23-yne-26,27-hexafluoro-19-nor -D(3), exerted identical effects at 100-fold lower concentrations. Inhibition of DC maturation and stimulatory function was absent in cultures from mice genetically lacking vitamin D receptors (VDR). Vitamin D analogs effectively reduce DC function via VDR-dependent pathways.     

Aluminum fluoride inhibition of cabbage phospholipase D by a phosphate-mimicking mechanism

The intermediate CuI-semiquinone radical species in the catalytic mechanism of copper-amine oxidase from Lens esculenta and Pisum sativum seedlings has been studied by optical, Raman resonance and ESR spectroscopies and by stopped-flow and temperature-jump measurements. Treatment of highly purified enzyme preparations with good, poor or suicide substrates, under anaerobic and aerobic conditions, at different pH values and temperatures, makes it possible to generate, detect and characterize this free radical intermediate.     

Intracellular localization of phospholipase D1 in mammalian cells

Phospholipase D (PLD) hydrolyzes phosphatidylcholine to generate phosphatidic acid. In mammalian cells this reaction has been implicated in the recruitment of coatomer to Golgi membranes and release of nascent secretory vesicles from the trans-Golgi network. These observations suggest that PLD is associated with the Golgi complex; however, to date, because of its low abundance, the intracellular localization of PLD has been characterized only indirectly through overexpression of chimeric proteins. We have used highly sensitive antibodies to PLD1 together with immunofluorescence and immunogold electron microscopy as well as cell fractionation to identify the intracellular localization of endogenous PLD1 in several cell types. Although PLD1 had a diffuse staining pattern, it was enriched significantly in the Golgi apparatus and was also present in cell nuclei. On fragmentation of the Golgi apparatus by treatment with nocodazole, PLD1 closely associated with membrane fragments, whereas after inhibition of PA synthesis, PLD1 dissociated from the membranes. Overexpression of an hemagglutinin-tagged form of PLD1 resulted in displacement of the endogenous enzyme from its perinuclear localization to large vesicular structures. Surprisingly, when the Golgi apparatus collapsed in response to brefeldin A, the nuclear localization of PLD1 was enhanced significantly. Our data show that the intracellular localization of PLD1 is consistent with a role in vesicle trafficking from the Golgi apparatus and suggest that it also functions in the cell nucleus.     

Optimization of halopemide for phospholipase D2 inhibition

Halopemide, which was identified by HTS to inhibit phospholipase D2 (PLD2), provided the basis for an exploratory effort to identify potent inhibitors of PLD2 for use as inflammatory mediators. Parallel synthesis and purification were utilized to rapidly identify orally available amide analogs derived from indole 2-carboxylic acids with superior potency versus PLD2.     

PKN regulates phospholipase D1 through direct interaction

The association of phospholipase (PLD)-1 with protein kinase C-related protein kinases, PKNalpha and PKNbeta, was analyzed. PLD1 interacted with PKNalpha and PKNbeta in COS-7 cells transiently transfected with PLD1 and PKNalpha or PKNbeta expression constructs. The interactions between endogenous PLD1 and PKNalpha or PKNbeta were confirmed by co-immunoprecipitation from mammalian cells. In vitro binding studies using the deletion mutants of PLD1 indicated that PKNalpha directly bound to residues 228-598 of PLD1 and that PKNbeta interacted with residues 1-228 and 228-598 of PLD1. PKNalpha stimulated the activity of PLD1 in the presence of phosphatidylinositol 4,5-bisphosphate in vitro, whereas PKNbeta had a modest effect on the stimulation of PLD1 activity. The stimulation of PLD1 activity by PKNalpha was slightly enhanced by the addition of arachidonic acid. These results suggest that the PKN family functions as a novel intracellular player of PLD1 signaling pathway.     

Mammalian phosphoinositide-specific phospholipase C isoenzymes

Procaryotic and eucaryotic cells have evolved multiple pathways for communication with their external environment. The inositol 1,4,5-trisphosphate/diacylglycerol second messenger system is an example of such a signal transduction pathway which is present in multicellular eucaryotic organisms. Binding of an agonist to a specific cell surface receptor promotes rapid hydrolysis of phosphatidylinositol 4,5-bisphosphate. The pivotal enzyme for this second messenger system is phosphoinositide-specific phospholipase C which hydrolyzes phosphatidylinositol 4,5-bisphosphate to generate the two second messengers, inositol 1,4,5-trisphosphate and diacylglycerol. Recently, much progress has been made in the purification, characterization and cDNA cloning of multiple PI-PLC isoenzymes. The results of the recent studies on phosphoinositide-specific phospholipase C are reviewed.     

Potent inhibition of sperm motility by palytoxin

Palytoxin, produced by a stationary marine animal and one of the most toxic substances known, was used as a spear poison in ancient Hawaii to cause death by cardiovascular contracture. We report here that the motility of hamster caudal epididymal (HCE) and other sperm can be inhibited by as little as 10⁻¹³ M palytoxin in a time-dependent manner. This inhibition manifested itself as a loss in flagellar amplitude, often accompanied by an increase in beat frequency, resulting in a loss of forward progression and ultimately cessation of movement. Similar effects were observed in sperm from guinea pigs, rabbit, cattle, sea urchins and man. Preincubation with palytoxin did not prevent the induction of motility from quiescence in HCE sperm when free calcium ion was added. However, regardless of the timing of palytoxin addition this very vigorous motility disappeared shortly after it appeared. These, plus earlier observations showing palytoxin did not cause lysis under similar conditions or inhibit the progressive motility of demembranated sperm axoneme preparations, suggest both that this large molecule acts via the plasma membrane to cause its exceedingly toxic effects and that spermatozoa may be useful for the investigation of the mechanism of action of palytoxin.     

Two uncommon phospholipase D isoenzymes from poppy seedlings (Papaver somniferum L.)

Phospholipase D (PLD) has been detected in seedlings of Papaver somniferum L. cv. Lazúr (Papaveraceae). Purification of the enzyme revealed the existence of two forms of PLD (named as PLD-A and PLD-B). The two enzymes strongly differ in their catalytic properties. The pH optima were found at pH 8.0 for PLD-A and at pH 5.5 for PLD-B. While both enzymes show hydrolytic activity toward phosphatidylcholine (PC) and phosphatidyl-p-nitrophenol (PpNP), PLD-B only was able to catalyze the exchange of choline in PC by glycerol. Both enzymes were activated by Ca(2+) ions with an optimum concentration of 10 mM. In contrast to PLDs from other plants, PLD-B was still more activated by Zn(2+) ions with an optimum concentration of 5 mM. The apparent molecular masses of PLD-A and PLD-B, derived from sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE), were estimated to be 116.4 and 114.1 kDa. N-terminal protein sequencing indicated N-terminal blockage in both cases. The isoelectric points were found to be 8.7 for PLD-A and 6.7 for PLD-B. Both enzymes were shown to be N-linked glycoproteins. This paper is the first report on PLD in poppy and indicates some important differences of the two enzyme forms to other PLDs known so far.     

Cloning and direct G-protein regulation of phospholipase D from tobacco

Phospholipase D (PLD) and heterotrimeric G-proteins are involved in plant signal transduction pathways at the plasma membrane. There is evidence suggesting that PLD acts downstream from G-proteins, but a direct interaction of specific members has not been shown. In the present paper, a PLD cDNA clone was isolated from tobacco, expressed as a GST fusion in bacteria, and the recombinant protein was purified by glutathione affinity. Its enzymatic properties identified it as an alpha-type PLD. The alpha-subunit of a G-protein from tobacco was isolated in a similar way. Both proteins were functional in biochemical assays. When the G-protein was included in the PLD assay, a strong dosage-dependent inhibition of the PLD activity was observed. Different control proteins did not exhibit this inhibitory effect. When GST-NtGPalpha1 was activated by incubation with GTPgammaS the inhibitory activity was greatly reduced. These results provide a first indication for a direct regulation of PLDalpha by a heterotrimeric G-protein alpha-subunit in plants.     

Potent and specific inhibition of mammalian histone deacetylase both in vivo and in vitro by trichostatin A

(R)-Trichostatin A (TSA) is a Streptomyces product which causes the induction of Friend cell differentiation and specific inhibition of the cell cycle of normal rat fibroblasts in the G1 and G2 phases at the very low concentrations. We found that TSA caused an accumulation of acetylated histone species in a variety of mammalian cell lines. Pulse-labeling experiments indicated that TSA markedly prolonged the in vivo half-life of the labile acetyl groups on histones in mouse mammary gland tumor cells, FM3A. The partially purified histone deacetylase from wild-type FM3A cells was effectively inhibited by TSA in a noncompetitive manner with Ki = 3.4 nM. A newly isolated mutant cell line of FM3A resistant to TSA did not show the accumulation of the acetylated histones in the presence of a higher concentration of TSA. The histone deacetylase preparation from the mutant showed decreased sensitivity to TSA (Ki = 31 nM, noncompetitive). These results clearly indicate that TSA is a potent and specific inhibitor of the histone deacetylase and that the in vivo effect of TSA on cell proliferation and differentiation can be attributed to the inhibition of the enzyme.     

Expression and regulation of phospholipase D isoforms in mammalian cell lines

Phospholipase D (PLD) is activated in mammalian cells in response to diverse stimuli that include growth factors, activators of protein kinase C, and agonists binding to G-protein-coupled receptors. Two forms of mammalian PLD, PLD1 and PLD2, have been identified. Expression of mRNA and protein for PLD1 and PLD2 was analyzed in the following cell lines: A7r5 (rat vascular smooth muscle); EL4 (mouse thymoma); HL-60 (human myeloid leukemia); Jurkat (human leukemia); PC-3 (human prostate adenocarcinoma); PC-12K (rat phaeochromocytoma); and Rat-1 HIR (rat fibroblast). All, with the exception of EL4, express agonist-activated PLD activity. PLD1 is expressed in A7r5, HL-60, PC-3, and Rat-1, while PLD2 is expressed in A7r5, Jurkat, PC12K, PC-3, and Rat-1. Neither isoform is expressed in EL4. Guanine nucleotide-independent PLD activity is present in membranes from all cells expressing PLD2. In PC12K cells, which express only PLD2, treatment with nerve growth factor causes neurite outgrowth and increases expression of PLD2 mRNA and protein within 6-12 h. A corresponding increase is observed in membrane PLD activity and in phorbol-12-myristate-13-acetate (PMA)-stimulated PLD activity in intact cells. These results show that PLD2 can be regulated both pretranslationally and posttranslationally by agonists.