Effect of latrunculin B, jasplakinolide and brefeldin A on the store-dependent Ca2+ entry in macrophages

Using Fura-2AM microfluorimetry, effect of actin filament modifiers and vesicular trafficking inhibitor on the store-dependent Ca2+ entry induced by purinergic agonists (ATP, UTP) and endoplasmic Ca2+-ATPase inhibitors (thapsigargin, cyclopiazonic acid) in rat peritoneal macrophages was investigated. It was shown that inhibition of actin polymerization by latrunculin B had a biphasic time-dependent effect on Ca2+ entry, showing an initial potentiation followed by inhibition of Ca2+ entry. Moreover, addition of latrunculin B after induction of store-dependent Ca2+ entry inhibited the Ca2+ influx. Jasplakinolide, which reorganizes actin filaments into a tight cortical layer adjacent to the plasma membrane, prevented activation of store-dependent Ca2+ entry but did not modify this process after its activation. Vesicular transport inhibitor brefeldin A, which inactivates arfproteins, inhibited activation of store-dependent Ca2+ entry but did not alter this mechanism once being initiated. These data are compatible with the sectretion-like coupling model for store-dependent Ca2+ entry in macrophages based on a reversible trafficking and coupling of Ca2+ store with the plasma membrane.     

Phosphoinositides are required for store-mediated calcium entry in human platelets

We have recently observed that small GTP-binding proteins are important for mediation of store-mediated Ca(2+) entry in human platelets through the reorganization of the actin cytoskeleton. Because it has been shown in platelets and other cells that small GTP-binding proteins regulate the activity of phosphatidylinositol 3-kinase and phosphatidylinositol 4-kinase, whose products, phosphoinositides, play a key role in the reorganization of the actin cytoskeleton, we have investigated the role of these lipid kinases in store-mediated Ca(2+) entry. Treatment of platelets with LY294002, an inhibitor of phosphatidylinositol 3- and phosphatidylinositol 4-kinases, resulted in a concentration-dependent inhibition of Ca(2+) entry stimulated by thapsigargin or the physiological agonist, thrombin. In addition, wortmannin, another inhibitor of these kinases, which is structurally unrelated to LY294002, significantly reduced store-mediated Ca(2+) entry. The inhibitory effect of LY294002 was not mediated either by blockage of Ca(2+) channels or by modification of membrane potential. LY294002 inhibited actin polymerization stimulated by thrombin or thapsigargin. These results indicate that both phosphatidylinositol 3-kinase and phosphatidylinositol 4-kinase are required for activation of store-mediated Ca(2+) entry in human platelets and that the mechanism could involve the reorganization of the actin cytoskeleton.     

Pathways for phosphoinositide synthesis.

In eukaryotic cells, phosphatidylinositol can be phosphorylated on the inositol ring by a series of kinases to produce at least seven distinct phosphoinositides. These lipids have been implicated in a variety of cellular processes, including calcium regulation, actin rearrangement, vesicle trafficking, cell survival and mitogenesis. The phosphorylated lipids can act as precursors of second messengers or act directly to recruit specific signaling proteins to the membrane. A number of the kinases responsible for producing these lipids have been purified and their cDNA clones have been isolated. The most well characterized of these enzymes are the phosphoinositide 3-kinases. However, progress has also been made in the characterization of phosphatidylinositol 4-kinases and phosphatidylinositol-4-phosphate 5-kinases. In addition, new pathways involving phosphatidylinositol-5-phosphate 4-kinases, phosphatidylinositol-3-phosphate 5-kinases and phosphatidylinositol-3-phosphate 4-kinases have recently been described. The various enzymes and pathways involved in the synthesis of cellular phosphoinositides will be discussed.     

The role of the L3T4 molecule in mitogen and antigen-activated signal transduction

We investigated the role of the L3T4 molecule in mitogen and antigen-initiated signal transduction in the L3T4(+) murine T cell hybridoma, 3DT52.5.9 and an L3T4(-) variant, 3DT52.5.24. Both Concanavalin A (Con A) and specific antigen stimulated increases in cytosolic-free calcium ([Ca2+]i), phosphatidylinositol turnover, and interleukin-2 (IL-2) production in both cell lines. About 85% of the stimulated rise in [Ca2+]i was from an extracellular source. Anti-L3T4 monoclonal antibody (MAb) inhibited 90% of antigen- and 50% of Con A-stimulated increases in [Ca2+]i and IL-2 production but had no effect on the ability of either activation signal to stimulate phosphatidylinositol turnover in the parent L3T4(+) cells. Stimulus-response coupling in the L3T4(-) cells was unaffected by the MAb. The anti-L3T4-insensitive increase in [Ca2+]i induced by Con A was inhibited by EGTA, suggesting that this mitogen also stimulated an influx of Ca2+ via an additional transport mechanism distinct from that stimulated by antigen. The fact that anti-L3T4 antibodies inhibit antigen and Con A-stimulated Ca2+ transport and IL-2 production without affecting phosphatidylinositol turnover suggests that L3T4 may play a critical role in modulating the activation of the T cell receptor-associated Ca2+ transporter in T cell stimulus-response coupling.     

Trifluoperazine Attenuates Store-Dependent Ca<Superscript>2+</Superscript> Entry in Macrophages

Using Fura-2AM microfluorimetry, we have shown for the first time that preincubation of macrophages with the calsequestrin inhibitor neuroleptic trifluoperazine leads to a significant inhibition of the store-dependent Ca²⁺ entry induced by endoplasmic Ca²⁺-ATPase inhibitors thapsigargin or cyclopiazonic acid in rat peritoneal macrophages. The results suggest calsequestrin involvement in the regulation of the store-dependent Ca²⁺ entry in macrophages.     

The role of cytoskeleton structures in regulation of Ca2+ responses in macrophages

The role of the cytoskeleton in regulation of purinergic agonist- and endoplasmic Ca(2+)-ATPase inhibitors-induced Ca2+ signals in rat peritoneal macrophages was investigated. It has been shown that in cells pretreated with agents that disrupt microtubules (vinblastine, colchicine, colcemid) or actin microfilaments (cytochalasins, phalloidin), the ability of thapsigargin or cyclopiazonic acid to empty Ca2+ stores and activate store-dependent Ca2+ influx was significantly attenuated. On the contrary, microfilaments and microtubule disrupters did not affect ATP- or UTP-induced Ca2+ mobilization, indicating that release of Ca2+ from intracellular stores through the inositol phosphate pathway was intact. The results suggested that an intact cytoskeleton is required for capacitative Ca2+ entry but not for agonist-induced Ca2+ mobilization.     

Role of different Ca2+ sources in the superoxide production of human neutrophil granulocytes.

The role of different Ca2+ sources in the activation of the NADPH oxidase was investigated in human neutrophil granulocytes. Selective depletion of the stimulus-responsive intracellular Ca2+ -pool and the consequent opening of the store-dependent Ca2+ channel of the plasma membrane was achieved with thapsigargin, an inhibitor of microsomal Ca2+ -ATPase. Low concentration (10-100 nM) of thapsigargin did not induce any O2*- -production, indicating that elevation of [Ca2+]ic to similar level and probably via similar route as following stimulation of chemotactic receptors, by itself is not sufficient to activate the NADPH oxidase. In significantly higher concentration (1-10 microM) thapsigargin did induce O2*- -generation but this effect was not the result of elevation of [Ca2+]ic. In the absence of external Ca2+ a gradual decrease of the responsive Ca2+ pool was accompanied by a gradual decrease of the rate and duration of the respiratory response stimulated by formyl-methionyl-leucyl-phenylalanin. Maximal extent of receptor-initiated O2*- -production could only be obtained when the intracellular [Ca2+] was higher than the resting level. Under this condition Ca2+ originating from intracellular or external source was equally effective in supporting the biological response.     

Sigma-1 Receptor Antagonist Haloperidol Attenuates Store-Dependent Ca<Superscript>2+</Superscript> Entry in Macrophages

Using Fura-2AM microfluorimetry, we have shown for the first time that preincubation of macrophages with sigma-1 receptor antagonist haloperidol leads to a significant inhibition of the store-dependent Ca²⁺ entry induced by endoplasmic Ca²⁺-ATPase inhibitors thapsigargin or cyclopiazonic acid in rat peritoneal macrophages. The results suggest the involvement of the sigma-1 receptor in the regulation of storedependent Ca²⁺ entry in macrophages.     

Granulocyte-macrophage colony-stimulating factor can stimulate macrophage proliferation via persistent activation of Na+/H+ antiport. Evidence for two distinct roles for Na+/H+ antiport activation.

Thapsigargin stimulates an increase of cytosolic free Ca2+ concentration [( Ca2+]c) in, and 45Ca2+ efflux from, a clone of GH4C1 pituitary cells. This increase in [Ca2+]c was followed by a lower sustained elevation of [Ca2+]c, which required the presence of extracellular Ca2+, and was not inhibited by a Ca2(+)-channel blocker, nimodipine. Thapsigargin had no effect on inositol phosphate generation. We used thyrotropin-releasing hormone (TRH) to mobilize Ca2+ from an InsP3-sensitive store. Pretreatment with thapsigargin blocked the ability of TRH to cause a transient increase in both [Ca2+]c and 45Ca2+ efflux. The block of TRH-induced Ca2+ mobilization was not caused by a block at the receptor level, because TRH stimulation of InsP3 was not affected by thapsigargin. Rundown of the TRH-releasable store by Ca2(+)-induced Ca2+ release does not appear to account for the action of thapsigargin on the TRH-induced spike in [Ca2+]c, because BAY K 8644, which causes a sustained rise in [Ca2+]c, did not block Ca2+ release caused by TRH. In addition, caffeine, which releases Ca2+ from intracellular stores in other cell types, caused an increase in [Ca2+]c in GH4C1 cells, but had no effect on a subsequent spike in [Ca2+]c induced by TRH or thapsigargin. TRH caused a substantial decrease in the amount of intracellular Ca2+ released by thapsigargin. We conclude that in GH4C1 cells thapsigargin actively discharges an InsP3-releasable pool of Ca2+ and that this mechanism alone causes the block of the TRH-induced increase in [Ca2+]c.     

Methyl-β-cyclodextrin modulates thapsigargin-induced store-dependent Ca<Superscript>2+</Superscript> entry in macrophages

Using Fura-2AM microfluorimetry, we have shown for the first time that preincubation of macrophages with methyl-β-cyclodextrin, inducing cholesterol extraction from membranes and raft disruption, leads to significant inhibition of thapsigargin-induced store-dependent Ca²⁺ entry in rat peritoneal macrophages. In contrast, macrophage treatment with methyl-β-cyclodextrin after Ca²⁺ entry mechanisms were activated by store depletion by thapsigargin application leads to potentiation of subsequent store-dependent Ca²⁺ entry. The results suggest that intact lipid rafts are necessary for the activation but not the maintenance of store-dependent Ca²⁺ entry in macrophages.     

Role of luminal ATP in regulating electrogenic Na(+) absorption in guinea pig distal colon

Extracellular ATP regulates a variety of functions in epithelial tissues by activating the membrane P2-receptor. The purpose of this study was to investigate the autocrine/paracrine regulation by luminal ATP of electrogenic amiloride-sensitive Na(+) absorption in the distal colon from guinea pigs treated with aldosterone by measuring the amiloride-sensitive short-circuit current (I(sc)) and (22)Na(+) flux in vitro with the Ussing chamber technique. ATP added to the luminal side inhibited the amiloride-sensitive I(sc) and (22)Na(+) absorption to a similar degree. The concentration dependence of the inhibitory effect of ATP on amiloride-sensitive I(sc) had an IC(50) value of 20-30 microM, with the maximum inhibition being approximately 50%. The effects of different nucleotides and of a nucleoside were also studied, the order of potency being ATP = UTP > ADP > adenosine. The effects of ATP were slightly, but significantly, reduced in the presence of suramin in the luminal solution. The inhibitory effect of luminal ATP was more potent in the absence of both Mg2+ and Ca2+ from the luminal solution. Pretreatment of the tissue with ionomycin or thapsigargin in the absence of serosal Ca2+ did not affect the percent inhibition of amiloride-sensitive I(sc) induced by ATP. Mechanical perturbation with a hypotonic luminal solution caused a reduction in amiloride-sensitive I(sc), this effect being prevented by the presence of hexokinase, an ATP-scavenging enzyme. These results suggest that ATP released into the luminal side by hypotonic stimulation could exert an inhibitory effect on the electrogenic Na(+) absorption. This effect was probably mediated by a P2Y(2) receptor on the apical membrane of colonic epithelial cells, and a change in the intracellular Ca2+ concentration may not be necessary for this process.     

Membrane potential changes associated with calcium signals in human lymphocytes and rat mast cells

Human lymphocytes and rat mast cells, two non-excitable cellular models, were used to investigate membrane potential changes accompanying Ca2+ signals. Cells were stimulated with agents known to induce both Ca2+ release from internal stores and influx of extracellular Ca2+, namely thapsigargin, ionomycin and compound 48/80. Thapsigargin and ionomycin were used to activate lymphocytes, while compound 48/80 was used to stimulate mast cells. Membrane potential changes and Ca2+ concentration were monitored with the fluorescent dyes bis-oxonol and fura-2, respectively. In lymphocytes, thapsigargin induced a hyperpolarization temporally correlated with the increase in intracellular Ca2+ concentration. This hyperpolarization is due to activation of a K+ conductance which consists of two phases, a first phase independent on external Ca2+ and a second one blocked in a Ca2+-free medium. Ionomycin induced a Ca2+-dependent depolarization attributed to a massive influx of external Ca2+. On the other hand, stimulation of mast cells with compound 48/80 produced a fast hyperpolarization and an increase in intracellular Ca2+ levels. Besides different time-courses, this hyperpolarization differs from that induced by thapsigargin in lymphocytes in two aspects, it is mainly due to a Cl(-)-entry current and exit of K+ and it is completely inhibited in the absence of extracellular Ca2+. Compound 48/80-induced histamine release is not related to membrane potential changes.     

Role of store-dependent influx of Ca2+ and efflux of K+ in apoptosis of CHO cells

Agents mobilising Ca(2+) from the endoplasmic reticulum are known to activate apoptosis. Whatever means are used, the release of Ca(2+) is often followed by a store-dependent entry of Ca(2+). Whether apoptosis is triggered by the depletion of the stores or by the subsequent store-dependent entry of Ca(2+) is still a matter of controversy. Here we studied apoptosis in CHO cells transfected with the rat neurotensin (NT) receptor, in which the store-dependent entry of Ca(2+) is abolished by repressing the transient receptor potential channel 2 (TRPC2) by an antisense oligonucleotide strategy (TRPC2(-) cells) [Cell Calcium 30 (2001) 157]. When stimulated with thapsigargin (TG), apoptosis occurred in both TRPC2(+) and TRPC2(-) cells but 12h earlier in TRPC2(+) cells, suggesting that store-dependent entry of Ca(2+) can accelerate the process. The expression and localisation of caspase-12, an enzyme that has been involved in the apoptosis triggered by a stress on the endoplasmic reticulum, was not different in TRPC2(+) and TRPC2(-) cells. On the contrary, the expression of GADD153 (Growth Arrest and DNA Damage inducible gene 153) triggered by TG treatment depended on external Ca(2+) and occurred earlier in TRPC2(+) than in TRPC2(-) cells. In these cells, we also noted the presence of K(+) channels activated by Ca(2+) (K(Ca) channels). Stimulation of TRPC2(+) cells with TG or with NT triggered a long sustained K(+) current, parallel to [Ca(2+)](i) transients, and resulting in a sustained hyperpolarisation of the cell membrane. K(+) current and hyperpolarisation were transient and not sustained in TRPC2(-) cells. Inhibition of K(Ca) channels with charybdotoxin dramatically reduced the K(+) current and also significantly brought down the level of apoptosis, suggesting that a prolonged efflux of K(+) could be involved in the apoptosis process. We conclude that in CHO cells, store-dependent entry of Ca(2+) can accelerate apoptosis by accelerating the expression of GADD153 and by inducing a prolonged efflux of K(+) out of the cell.     

Thapsigargin decreases the Na(+)- Ca(2+) exchanger mediated Ca(2+) entry in pig coronary artery smooth muscle

Na(+)- Ca(2+) exchanger (NCX) has been proposed to play a role in refilling the sarco/endoplasmic reticulum (SER) Ca(2+) pool along with the SER Ca(2+) pump (SERCA). Here, SERCA inhibitor thapsigargin was used to determine the effects of SER Ca(2+) depletion on NCX-SERCA interactions in smooth muscle cells cultured from pig coronary artery. The cells were Na(+)-loaded and then placed in either a Na(+)-containing or in a Na(+)-substituted solution. Subsequently, the difference in Ca(2+) entry between the two groups was examined and defined as the NCX mediated Ca(2+) entry. The NCX mediated Ca(2+) entry in the smooth muscle cells was monitored using two methods: Ca(2+)sensitive fluorescence dye Fluo-4 and radioactive Ca(2+). Ca(2+)-entry was greater in the Na(+)-substituted cells than in the Na(+)-containing cells when measured by either method. This difference was established to be NCX-mediated as it was sensitive to the NCX inhibitors. Thapsigargin diminished the NCX mediated Ca(2+) entry as determined by either method. Immunofluorescence confocal microscopy was used to determine the co-localization of NCX1 and subsarcolemmal SERCA2 in the cells incubated in the Na(+)-substituted solution with or without thapsigargin. SER Ca(2+) depletion with thapsigargin increased the co-localization between NCX1 and the subsarcolemmal SERCA2. Thus, inhibition of SERCA2 leads to blockade of constant Ca(2+) entry through NCX1 and also increases proximity between NCX1 and SERCA2. This blockade of Ca(2+) entry may protect the cells against Ca(2+)-overload during ischemia-reperfusion when SERCA2 is known to be damaged.     

Retrograde regulation of store-operated calcium channels by the ryanodine receptor-associated protein triadin 95 in rat skeletal myotubes

The 95kDa triadin (or T95), the main skeletal muscle triadin isoform, negatively regulates the mechanism of excitation-contraction coupling. T95 is a ryanodine receptor (RyR)-interacting protein but it also possesses a calsequestrin-interacting domain. RyR and calsequestrin are involved in Ca2+ signalling and, for instance, influence the activity of store-dependent Ca2+ channels (SOC). This work was undertaken to determine whether T95 was able to modulate the entry of Ca2+ through SOC. The experiments were carried out on differentiated rat myotubes over-expressing T95 or DsRed (control cells) by means of an adenovirus infection. Intracellular Ca2+ signals were analyzed using the Ca2+ indicator Fluo-4. The sarco-endoplasmic reticulum Ca2+-ATPase inhibitor thapsigargin was used to deplete intracellular Ca2+ stores. When applied in the presence of a Ca2+-free medium, thapsigargin elicited transient but long-lasting Fluo-4 responses by elevating the cytoplasmic concentration of Ca2+ ([Ca2+]i). The over-expression of T95 reduced the thapsigargin-dependent [Ca2+]i increase, with respect to control myotubes. Addition of extracellular Ca2+after the depletion of this Ca2+ pool was accompanied by a [Ca2+]i increase that was sensitive to the SOC blockers 2-APB, SKF-96365 and La3+. The over-expression of T95 reduced this Ca2+ influx, without changing its pharmacological properties, showing that T95 over-expression did not alter the properties of the SOC. In conclusion, the RyR-interacting molecule T95, recently shown to inhibit the excitation-contraction coupling, has also the ability to interfere with the skeletal muscle Ca2+ signalling by depressing thapsigargin-dependent Ca2+ release and influx.     

Role of the ERK pathway in the activation of store-mediated calcium entry in human platelets

Extracellular signal-regulated kinases (ERKs), are common participants in a broad variety of signal transduction pathways. Several studies have demonstrated the presence of ERKs in human platelets and their activation by the physiological agonist thrombin. Here we report the involvement of the ERK cascade in store-mediated Ca(2+) entry in human platelets. Treatment of dimethyl-bis-(o-aminophenoxy)-ethane-N,N,N',N'-tetraacetic acid-loaded platelets with thapsigargin to deplete the intracellular Ca(2+) stores resulted in a time- and concentration-dependent activation of ERK1 and ERK2. Incubation with either U0126 or PD 184352, specific inhibitors of mitogen-activated protein kinase kinase (MEK), prevented thapsigargin-induced ERK activation. Furthermore, U0126 and PD 184352 reduced Ca(2+) entry stimulated by thapsigargin or thrombin, in a concentration-dependent manner. The role of ERK in store-mediated Ca(2+) entry was found to be independent of phosphatidylinositol 3- and 4-kinases, the tyrosine kinase pathway, and actin polymerization but sensitive to treatment with inhibitors of Ras, suggesting that the ERK pathway might be a downstream effector of Ras in mediating store-mediated Ca(2+) entry in human platelets. In addition, we have found that store depletion stimulated ERK activation does not require PKC activity. This study demonstrates for the first time a novel mechanism for regulation of store-mediated Ca(2+) entry in human platelets involving the ERK cascade.     

Fluoroaluminate activation of different components of the calcium signal in an exocrine cell.

In isolated cells from the avian supra-orbital nasal gland, used as a model for exocrine ion secretion, addition of NaF (2-15 mM) produced a slow Al3(+)-enhanced increase in intracellular Ca2+ concn. ([Ca2+]i), resulting in a more than 2-fold sustained elevation in [Ca2+]i. Simultaneously, cellular Ins(1,4,5)P3 contents became markedly elevated, suggesting an AlF4- activation of a phospholipase C-specific G-protein. Subsequent addition of the muscarinic agonist carbachol failed to produce any further sustained increase in [Ca2+]i, indicating that the AlF4(-)-induced increase in [Ca2+]i involves a Ca2(+)-entry pathway identical with that activated by carbachol. In low-Ca2+ media (extracellular [Ca2+] = 0.04 mM) no such increase in [Ca2+]i, either sustained or transient, is seen, although cellular Ins(1,4,5)P3 levels were markedly elevated. Despite the failure to observe any change in [Ca2+]i in the low-Ca2+ medium, estimation of the size of the agonist-sensitive Ca2+ stores (determined as the magnitude of the transient change in [Ca2+]i induced by carbachol) revealed that these are progressively emptied by the action of AlF4-. However, the onset of this emptying showed an initial lag period of at least 2 min (with 5 mM-NaF plus 10 microM-AlCl3). In marked contrast, determinations of the magnitude of the Ca2(+)-entry pathway under identical conditions showed that this was significantly activated after as little as 1 min of AlF4- treatment. This suggests that, under these conditions, activation of Ca2+ entry in these cells preceded the release of Ca2+ from agonist-sensitive stores, contradicting current models in which the receptor-enhanced entry of extracellular Ca2+ is entirely dependent on, and subsequent to, the prior release of Ca2+ from the intracellular stores.     

Phosphorylation of erythrocyte membrane liberates calcium

Washed and permeabilized human erythrocyte ghosts were found to discharge calcium on treatment with ATP. Concomitantly, there was a decrease in phosphatidylinositol (PI) and an increase in phosphatidylinositol-4-phosphate (PIP) and phosphatidylinositol-4,5-bisphosphate (PIP2). These results support the hypothesis that an inositide shuttle, PI in equilibrium PIP in equilibrium PIP2, operates to maintain intracellular Ca2+ levels. The cation is thought to be sequestered in a cage formed by the head groups of two acidic phospholipid molecules, e.g., phosphatidylserine and phosphatidylinositol, with participation of both PO and fatty acid ester CO groups. These cages are stabilized by inter-headgroup hydrogen bonding. When the inositol group is phosphorylated in positions 4 and 5, inter-lipid hydrogen bonding is disrupted and the cage opens to release its Ca2+.     

Lipid kinases play crucial and multiple roles in membrane trafficking and signaling

Phosphotidylinositols (PIs) are known to play an essential role in membrane trafficking and signaling transduction. PIs serve multiple functions, such as recruitment of cytosolic proteins with PI phosphate (PIP) binding domains and modification of the physical properties of the membranes in which they reside. As substrates for phosphoinositide-specific lipases they function as a switch point in phosphoinositide metabolism. Recent work with epidermal growth factor receptor (EGFR) and colony stimulating factor-1 receptor (CSFR) has identified a possible connection between endocytosis of activated receptors and type-1 phosphatidylinositol-4-phosphate-5-kinase. Furthermore, serine/tyrosine phosphorylation of phosphatidylinositol-4-phosphate-5-kinase seems to be essential for its activities. Indeed, one of the products of the phosphatidylinositol-4-phosphate-5-kinases, PIP2, has been shown to be involved in multiple steps of endocytosis, including the assembly of the clathrin coat, regulation of adaptor proteins, and production of endocytic vesicles via the regulation of dynamin. The discussion in this review focuses primarily on receptors with intrinsic enzymatic activity, specifically on receptor tyrosine kinases (RTKs). We will discuss their structure; mechanism of action and potential role in membrane trafficking and/or signaling through the regulation of phosphatidylinositol phosphate kinases.