Bis‐diphosphoinositol tetrakisphosphate ([PP]2‐InsP4 or ‘InsP8’) is a ‘high‐energy’ inositol phosphate; we report that its metabolism is receptor‐regulated in DDT1 MF‐2 smooth muscle cells. This conclusion arose by pursuing the mechanism by which F− decreased cellular levels of [PP]2‐InsP4 up to 70%. A similar effect was induced by elevating cyclic nucleotide levels, either with IBMX or by application of either Bt2cAMP (EC50 = 14.7 μM), Bt2cGMP (EC50 = 7.9 μM) or isoproterenol (EC50 = 0.4 nM). Isoproterenol (1 μM) decreased [PP]2‐InsP4 levels 25% by 5 min, and 71% by 60 min. This novel, agonist‐mediated regulation of [PP]2‐InsP4 turnover was very specific; isoproterenol did not decrease the cellular levels of either inositol pentakisphosphate, inositol hexakisphosphate or other diphosphorylated inositol polyphosphates. Bradykinin, which activated phospholipase C, did not affect [PP]2‐InsP4 levels. Regulation of [PP]2‐InsP4 turnover by both isoproterenol and cell‐permeant cyclic nucleotides was unaffected by inhibitors of protein kinases A and G. The effectiveness of the kinase inhibitors was confirmed by their ability to block phosphorylation of the cAMP response element‐binding protein. Our results indicate a new signaling action of cAMP, and furnish an important focus for future research into the roles of diphosphorylated inositol phosphates in signal transduction.
Organisms from across the phylogenetic spectrum synthesize an assortment of phosphorylated inositol derivatives. Although this family of compounds all share the cyclohexanehexol moiety, they provide the cell with a functionally diverse range of molecules. Their versatility as intracellular signals is particularly remarkable; different sub‐groups of inositol‐based compounds each have specialized roles in some quite different signal transduction processes (Berridge and Irvine, 1989; Kapeller and Cantley, 1994; Shears, 1997). In this report, we provide a new focus for studies in this field of research, by demonstrating a specific receptor‐dependent regulation of the turnover of bis‐diphosphoinositol tetrakisphosphate ([PP]2‐InsP4).
[PP]2‐InsP4 is a diphosphorylated polyphosphate, the synthesis and metabolism of which is regulated by two coupled kinase/phosphatase substrate cycles (Figure 1). [PP]2‐InsP4is formed by an ATP‐dependent PP‐InsP5 kinase (Shears et al., 1995); PP‐InsP5 is in turn synthesized from an ATP‐dependent InsP6 kinase (Menniti et al., 1993; Stephens et al., 1993; Shears et al., 1995; Voglmaier et al., 1996). A third diphosphate, PP‐InsP4, is interconverted with InsP5 in another kinase/phosphatase substrate cycle (Menniti et al., 1993). Interest in this group of compounds has in part arisen from the substantial free energy change associated with the hydrolysis of the β‐phosphate in the diphosphate groups (Stephens et al., 1993; Laussmann et al., 1997). It has been suggested that this metabolic turnover may be utilized as a regulatory molecular switch, possibly through the transphosphorylation of one or more proteins (Voglmaier et al., 1996); some exciting preliminary evidence in support of the latter possibility has been obtained (Voglmaier et al., 1994).
We previously have employed F− as a tool to measure the extent of the metabolic flux through these substrate cycles in intact cells (Menniti et al., 1993; Albert et al., 1997). By inhibiting the β‐phosphatases that attack PP‐InsP4 and PP‐InsP5 (Menniti et al., 1993; Shears et al., 1995), F− acts as a metabolic trap that causes the levels of both PP‐InsP4 and PP‐InsP5 to increase (Menniti et al., 1993). These data indicated that, every hour, 30–50% of the entire cellular pools of InsP5 and InsP6 cycle through the diphosphorylated polyphosphates (Menniti et al., 1993; Albert et al., 1997).
There was an intriguing, additional outcome of these experiments with F−. In earlier studies with the AR4‐2J pancreatoma cell line, raising the concentration of F− from 0.8 to 10 mM caused levels of [PP]2‐InsP4 to decrease (Shears et al., 1995). These results suggest that F− inhibited the rate of [PP]2‐InsP4 synthesis and/or stimulated the rate of [PP]2‐InsP4 dephosphorylation. However, work with purified enzymes has not yielded any evidence to support these two possibilities: the purified phosphatases that dephosphorylate PP‐InsP5 and [PP]2‐InsP4 are equally sensitive to inhibition by F− (Ki = 10 μM, unpublished data). Furthermore, F− did not affect the activity of the purified PP‐InsP5kinase (C.‐H.Huang, S.M.Voglmaier, M.E.Bembenek and S.H.Snyder, unpublished data). Thus, we suggested that F− has some additional and distinctive effect upon [PP]2‐InsP4 turnover in vivo (Shears et al., 1995). We now describe the mechanism by which F− affects cellular [PP]2‐InsP4 metabolism in the DDT1 MF‐2 Syrian hamster vas deferens smooth muscle cell line.
Results and discussion
Diphosphorylated inositol polyphosphates in DDT1MF‐2 cells
HPLC was used to resolve the 3H‐labeled inositol phosphates (polarity ≥InsP2) in DDT1 MF‐2 cells (Figure 2); there were three diphosphorylated compounds (PP‐InsP4, peak v; PP‐InsP5, peak vii; [PP]2‐InsP4, peak viii), all of which have been observed previously in other mammalian cells (Menniti et al., 1993; Stephens et al., 1993; Shears et al., 1995; Albert et al., 1997). Although [PP]2‐[3H]InsP4 was the least abundant of the diphosphates (4787 d.p.m., Figure 2), its levels were of sufficient magnitude to exceed those of all the InsP4 isomers combined (3906 d.p.m.). [PP]2‐InsP4 was eluted by 1.8–2 M Pi, depending upon the age of the column. No more polar 3H‐labeled material was detected in these or any further experiments described in this study, even when [Pi] in the HPLC eluate was increased to 2.6 M.
The effects of F− upon the turnover of the diphosphorylated inositol polyphosphates
Treatment of intact cells for 30 min with either 0.8 (not shown) or 10 mM NaF (Table I) approximately doubled the levels of PP‐[3H]InsP4and PP‐[3H]InsP5. This effect is consistent with the previously demonstrated ability of F− to inhibit the diphosphatase activities that attack PP‐InsP4and PP‐InsP5 (Menniti et al., 1993; Shears et al., 1995). Thus, F− acts as a metabolic trap, exposing the ongoing phosphorylation of InsP5 and InsP6.
F− also inhibits [PP]2‐InsP4 diphosphatase activity (Shears et al., 1995), but 30 min treatment of intact cells with 0.8 mM NaF had no net effect upon [PP]2‐[3H]InsP4 levels, even though 2‐fold increases in levels of both PP‐[3H]InsP4and PP‐[3H]InsP5 were observed (data not shown). More striking was the observation that 10 mM F− caused the levels of [PP]2‐[3H]InsP4 to decrease by 70% (Table I). Levels of [PP]2‐[3H]InsP4 remained at a nadir for 10–30 min (Figure 3), but then started to rise again, so that by 60 min, they were only 25% below those in time‐matched control cells. After 2–3 h, [PP]2‐[3H]InsP4 levels exceeded those of control cells (data not shown). In contrast, the levels of both PP‐[3H]InsP4 and PP‐[3H]InsP5 remained elevated throughout the duration of these experiments (Table I, and data not shown). [PP]2‐InsP4 was the only inositol diphosphate to show a decrease in cellular levels following treatment with 10 mM F− (Table I). Since F− does not directly stimulate [PP]2‐InsP4 phosphatase or inhibit PP‐InsP5 kinase (see Introduction), we considered alternative possible mechanisms that might account for this specific effect of F−.
Is the effect of F− on [PP]2‐InsP4 levels caused by a decrease in ATP levels?
At a concentration of 10 mM, F− is a metabolic poison that depletes cellular ATP levels (see, for example, Okada and Brown, 1988). Indeed, 30 min treatment of DDT1MF‐2 cells with 10 mM F− caused ATP levels to drop by 70% (Table I). It is possible that this decrease in ATP supply for the PP‐InsP5 kinase might contribute to the F−–mediated decline in [PP]2‐[3H]InsP4 levels. On the other hand, cellular [ATP] remained depleted after 60 min F− treatment (7 ± 2 nmol/mg protein), at a time when levels of [PP]2‐[3H]InsP4 began to recover (see above). This increased rate of synthesis of [PP]2‐InsP4 demonstrates the competence of the PP‐InsP5 kinase even at these reduced levels of ATP.
Is the effect of F− on [PP]2‐InsP4 levels mediated through phospholipase C?
Another consequence of adding 10 mM NaF to cells is the formation of an AlF4− complex that activates heterotrimeric GTP‐binding proteins (Sternweis and Gilman, 1982). This has a number of repercussions, among which is the activation of phospholipase C (PLC) (Blackmore et al., 1985). Indeed, 10 mM F− stimulated PLC activity in DDT1MF‐2 cells, as indicated by the increases in levels of [3H]InsP2, [3H]InsP3 and [3H]InsP4 to, respectively, 227 ± 27%, 356 ± 140% and 817 ± 84% of control levels (Table I, and data not shown). We next investigated if there was any effect upon the diphosphorylated inositol phosphates when PLC activity was activated by an appropriate ligand of the cell surface receptors in DDT1MF‐2 cells, namely, bradykinin (Sipma et al., 1995). At a concentration of 100 nM, this agonist stimulated PLC activity, since cytoplasmic [Ca2+] increased from 58 ± 2 nM (n = 11) to a peak value of 132 ± 8 nM (n = 10), and levels of [3H]InsP2, [3H]InsP3 and [3H]InsP4 increased to, respectively, 128 ± 6%, 128 ± 12% and 294 ± 32% of control levels (Table I, and data not shown). However, bradykinin did not significantly affect the levels of either [PP]2‐InsP4 or the other diphosphorylated inositol phosphates (Table I).
Is the effect of F− on [PP]2‐InsP4 levels mediated through adenylate cyclase?
Basal levels of cAMP in DDT1 MF‐2 cells (2.1 ± 0.5 pmol/mg protein, n = 12) were elevated by a 30 min treatment with 10 mM F− (to 8.1 ± 0.8 pmol/mg protein, n = 4, P<0.001 unpaired t‐test), due to AlF4−‐mediated activation of adenylate cyclase (Sternweis and Gilman, 1982). We therefore studied the effects upon [PP]2‐InsP4 levels of alternate procedures to elevate intracellular [cAMP]. The phosphodiesterase inhibitor, IBMX, caused [PP]2‐[3H]InsP4 levels to decrease by about half (Table I). Levels of this polyphosphate also decreased ∼50% after application of cell‐permeant Bt2cAMP (Table I). Thus, we conclude that the predominant mechanism by which F− modifies [PP]2‐InsP4 turnover is through activation of adenylate cyclase.
Three additional aspects of the data in Table I are important to note. First, Bt2cGMP reduced [PP]2‐[3H]InsP4 levels to almost the same degree as did Bt2cAMP (Figure 4). In fact, very similar dose–response curves were obtained for the two cyclic nucleotides (EC50 for Bt2cAMP = 14.7 μM; EC50 for Bt2cGMP = 7.9 μM). Second, there is a specificity to the effect of cyclic nucleotides upon [PP]2‐InsP4 turnover. The levels of the other inositol polyphosphates were not affected by elevations in cellular [cAMP] or [cGMP] (Table I). Finally, the mechanisms of action of IBMX, Bt2cGMP and Bt2cAMP were not targeted at cellular ATP levels, which these agents did not affect (Table I).
The effect of isoproterenol upon [PP]2‐InsP4 turnover and levels of cyclic nucleotides in intact cells
Isoproterenol was also used to elevate cAMP levels up to 16‐fold in DDT1 MF‐2 cells (Figure 5, and see Norris et al., 1983). The mean EC50 value from seven experiments was 9 nM. In agreement with previous studies using other types of smooth muscle cell, including those of non‐vascular origin (Jiang et al., 1992; Murthy and Makhlouf, 1995), levels of cGMP (0.45 ± 0.24 pmol/mg protein, n = 4) were unaffected by 2–30 min incubation with 1 μM isoproterenol. PLC was not activated (Table I).
Isoproterenol also caused a dose‐dependent decrease in [PP]2‐[3H]InsP4 levels (Figure 5; the mean EC50 value was 0.4 nM, n = 4). This agonist‐mediated effect was highly specific; incubation with isoproterenol did not lead to a drop in the levels of any other inositol polyphosphate (Table I). A time course experiment (Figure 6) shows that 1 μM isoproterenol caused the levels of [PP]2‐[3H]InsP4 to decrease 25% by 5 min. The reduction continued throughout a 60 min treatment, at which point the 71% decrease was the same degree of effect as that which was brought about by treatment with 10 mM F− (Table I). Consistent with data showing that isoproterenol acts through β2‐adrenergic receptors in DDT1 MF‐2 cells (Norris et al., 1983), the agonist's action upon [PP]2‐[3H]InsP4 turnover was unaffected when α‐adrenergic receptors were blocked with 10 μM phentolamine (data not shown). The novelty of these observations is worth emphasizing; previously, the metabolism of the diphosphorylated inositol polyphosphates was not known to be agonist‐regulated (Menniti et al., 1993; Shears et al., 1995).
cAMP regulates [PP]2‐InsP4 turnover independently of A‐kinase
We next investigated if A‐kinase participated in the response of [PP]2‐InsP4 turnover to treatment with Bt2cAMP and isoproterenol. Three cell‐permeant A‐kinase antagonists were used: H‐89 (Chijiwa et al., 1990), HA1077 (Asano et al., 1989) and RpCPT‐cAMPS (Roseboom and Klein, 1995). In these experiments, two precautions were taken to optimize the opportunity for the antagonist to effectively block A‐kinase. First, we used relatively high concentrations of the antagonists. Second, we chose slightly submaximal concentrations of agonists (3 nM isoproterenol and 300 μM Bt2cAMP) which induced 80% of the maximum possible depletion in [PP]2‐InsP4 levels. We also independently checked the efficacy of the A‐kinase antagonists by measuring the degree of phosphorylation by A‐kinase of cAMP response element‐binding protein (CREB), using an anti‐phospho‐CREB antibody (Ginty et al., 1993). Figure 7 shows the prominent increase in the phospho‐CREB signal in extracts from cells treated with 3 nM isoproterenol (lane 3) compared with control cells (lane 1). All three cell‐permeant antagonists of A‐kinase reduced the phospho‐CREB signal to levels that were either at or below basal levels (Figure 7, lanes 4–6). These data confirm the effectiveness of the antagonist protocols.
We found that the effect upon [PP]2‐InsP4 metabolism of both isoproterenol and Bt2cAMP was unaffected by any of the three A‐kinase antagonists (Figure 8). The antagonists also had no effect by themselves (Figure 8). These data indicate that cAMP regulates [PP]2‐InsP4 turnover independently of A‐kinase. We considered that one possible explanation for this mode of cAMP action might be the cross‐activation of G‐kinase (Jiang et al., 1992; Murthy and Makhlouf, 1995). However, this seems unlikely for several reasons. First, cAMP is such a weak activator of the G‐kinase (Gamm et al., 1995) that, in smooth muscle cells at least, it requires 10–100 μM isoproterenol before significant cross‐activation can be detected (Jiang et al., 1992; Murthy and Makhlouf, 1995). These concentrations are considerably higher than the EC50 value of 0.4 nM for the agonist's effects on [PP]2‐InsP4 metabolism (see above). Second, the A‐kinase antagonist, HA1077, is equally effective against G‐kinase (Asano et al., 1989), yet HA1077 also did not antagonize cyclic nucleotide‐mediated changes in [PP]2‐InsP4 metabolism (Figure 8). Third, we used another G‐kinase antagonist, Rp‐cGMPS (Butt et al., 1994), and it also failed to reverse the decrease in [PP]2‐InsP4 levels brought about by either isoproterenol or by Bt2cAMP (Figure 8). Finally, the conclusion that G‐kinase does not participate in this response is reinforced further by the observation that Rp‐cGMPS failed to reverse the effect of Bt2cGMP upon [PP]2‐InsP4 turnover (Figure 8). In other words, it seems that both cGMP and cAMP regulate the metabolism of [PP]2‐InsP4 independently of activating their respective kinases.
We further excluded the participation of protein phosphorylation processes in the regulation of [PP]2‐InsP4 turnover by blocking protein phosphatases with okadaic acid (Cohen, 1989). Treatment of intact cells with okadaic acid increased the phospho‐CREB signal (Figure 8, lane 2) over that in control cells (Figure 8, lane 1) but did not affect the cellular levels of [PP]2‐[3H]InsP4 (data not shown).
One well‐recognized kinase‐independent effect of cyclic nucleotides is to open Ca2+‐permeable cation channels in the plasma membrane (Yau, 1994). Although these ion channels were thought initially to be restricted to sensory cells, they are now known to be more widely distributed (Yau, 1994; Ding et al., 1997). We found that cyclic nucleotide‐mediated depletion in [PP]2‐InsP4 levels was not associated with any measurable increase in the rate of Ca2+ entry, as cytoplasmic [Ca2+] was unaffected by either 1 mM IBMX, 3 mM Bt2cAMP, 3 mM Bt2cGMP or 1 μM isoproterenol (data not shown). Moreover, we also incubated cells for 30 min in the absence of extracellular Ca2+ so that no Ca2+ entry could occur. In these incubations, [PP]2‐[3H]InsP4 levels were unaffected (95 ± 9% of control, n = 4), and 1 μM isoproterenol still caused [PP]2‐[3H]InsP4levels to decrease (to 51 ± 7% of control, n = 3) to the same extent as in medium with the normal extracellular [Ca2+] (to 45 ± 6% of control, see Table I).
Among several new conclusions to arise from this study, the most important is that the metabolism of [PP]2‐InsP4 is regulated by the β2‐adrenergic agonist, isoproterenol. Our data therefore provide the first evidence that there is a close relationship between [PP]2‐InsP4 levels and signal transduction events. This effect of isoproterenol was mediated through cAMP but, significantly, the latter acted independently of the activation of A‐kinase, or the cross‐activation of G‐kinase. The receptor‐regulated diminution in [PP]2‐InsP4 levels could arise as a consequence of a decrease in the rate of PP‐InsP5 phosphorylation, and/or an increase in the rate of [PP]2‐InsP4 dephosphorylation. Distinguishing between these two possibilities will be an important future direction as we unravel the molecular events that underlie this new regulatory process. It is an exciting possibility that an increased rate of utilization of [PP]2‐InsP4 could reflect up‐regulation of its proposed action as a molecular switch and/or a donor for transphosphorylation reactions (Voglmaier et al., 1994, 1996).
The model system for the current study was the non‐vascular, smooth muscle cell line, DDT1MF‐2. In these cells, F− affects [PP]2‐InsP4 metabolism through an elevation in cAMP levels. Since an F−‐mediated reduction in [PP]2‐InsP4 levels has also been observed in AR4‐2J pancreatoma cells (Shears et al., 1995), we can therefore anticipate a wider occurrence of these effects of cAMP upon [PP]2‐InsP4 turnover. Moreover, our demonstration that Bt2cGMP is as effective a stimulus as Bt2cAMP (Figure 4) suggests that physiologically relevant elevations in cellular [cGMP] [such as those induced by nitric oxide (Garthwaite and Boulton, 1995)] will also prove to regulate [PP]2‐InsP4 metabolism.
The diphosphorylated polyphosphates have already attracted attention in view of their rapid interconversion in coupled substrate cycles (Figure 1), and because of the considerable free energy change upon the hydrolysis of the β‐phosphate of the diphosphate groups. It is surely the case that the cell reaps some considerable benefit from the energy it expends in order to sustain the steady‐state levels of these metabolically active compounds. Now that we know this turnover is receptor‐regulated, the need for further research into the biological significance of this class of compounds is stronger than ever.
Materials and methods
Cell culture, [3H]inositol labeling and HPLC of [3H]inositol phosphate
DDT1 MF‐2 Syrian hamster vas deferens smooth muscle cells (provided by Dr D.Gill, University of Maryland School of Medicine, Baltimore, MD) were maintained in Dulbecco‘s modified Eagle's medium (DMEM) supplemented with ‘high‐glucose’ (i.e. 25 mM), 2 mM glutamine and 5% fetal calf serum at 37°C in 5% CO2/95% humidified air. Cells were harvested and plated at a density of ∼200 000 cells/well (16 mm diameter, 24‐well muliplates) in the DMEM‐based culture medium described above, supplemented with 50 μCi/ml [3H]inositol. Medium was replaced 3 and 5 days after plating. On the 7th day, cell monolayers were washed (2×250 μl) and then incubated (250 μl) in [3H]inositol‐free HEPES‐buffered Krebs solution (115 mM NaCl, 5 mM KCl, 1 mM NaH2PO4, 0.5 mM MgSO4, 11 mM glucose, 1.36 mM CaCl2, 25 mM HEPES, pH 7.4 with NaOH). Cells were maintained at 37°C for 3 h prior to the beginning of any experiments. All experimental protocols were time‐matched with control incubations.
Experiments were quenched by rapid aspiration of the Krebs solution, followed by addition of 250 μl of ice‐cold 0.6 M perchloric acid (PCA) containing 0.2 mM InsP6, and neutralized by the addition of 70 μl of 1 M K2CO3 containing 5 mM Na2EDTA. After being kept at 4°C for 30 min, the perchlorate precipitate was removed by centrifugation (10 000 g, 2 min). The supernatants were finally diluted with 3 vols of 1 mM Na2EDTA. Similar data were also obtained when cells were quenched with a Triton‐based buffer, followed by deproteinization using NENSORB columns (Shears et al., 1995).
Samples were stored at −20°C prior to being loaded onto a 4.6×125 mm Partisphere 5 μm SAX HPLC column. Inositol phosphates were eluted at 1 ml/min by the following gradient generated by mixing buffer A (1 mM Na2EDTA) and buffer B [buffer A plus 1.3 M (NH4)2HPO4, pH 3.85 with H3PO4; total [Pi] = 2.6 M]: 0–10 min, 0% B; 10–25 min, 0–35% B; 25–105 min, 35–100% B; 105–115 min, 100% B. Fractions of 1 ml were collected, mixed with 3 vols of Monoflow 4 scintillant, and radioactivity was determined using liquid scintillation spectrometry.
Analysis of CREB phosphorylation by Western blotting
Cells were prepared in 24‐well multiplates as described above, except that [3H]inositol was omitted from the culture medium. Following agonist stimulation, experiments were quenched in hot Laemmli sample buffer, and cell extracts were drawn through a 27 gauge needle to reduce viscosity. Samples were run on a 10% polyacrylamide gel, and electroblotted onto a nitrocellulose membrane. After washing with phosphate‐buffered saline (PBS), the membrane was incubated with an anti‐phospho‐CREB rabbit polyclonal antibody according to the manufacturer's protocol. Antibody–antigen complexes were detected using a horseradish peroxidase‐coupled anti‐rabbit IgG.
Cellular [Ca2+] measurements
DDT1 MF‐2 cells were grow on glass coverslips, washed with Krebs buffer, and equilibrated for 4 h prior to loading with 3 μM Fura‐2/AM for 30 min at room temperature in the dark. Cells were then washed with Krebs buffer and fluorescence was monitored using a PTI dual wavelength spectrofluorimeter system, with excitation at 340 and 380 nm, and emission intensity was measured at 505 nm. The average free cytoplasmic [Ca2+] was then calculated (Grynkiewicz et al., 1986).
The appropriate kits (see Materials) and methodologies recommended by the supplier were used to assay for cAMP, cGMP and ATP (in PCA‐quenched, K2CO3‐neutralized cell extracts, see above). Protein concentration was determined as described by Bradford (1976). EC50 values were derived using Prism (GraphPad, San Diego, CA).
[3H]Inositol (10–25 Ci/mmol; 10 mCi/ml; in sterile water), NENSORB columns, and standards of [3H]InsP6, PP‐[3H]InsP5 and [PP]2‐[3H]InsP4 were purchased from NEN Life Science Products (Boston, MA). Rp‐cGMPS and HA1077 were purchased from RBI (Natick, MA), RpCPT‐cAMPS was purchased from Biolog (La Jolla, CA). Bt2cAMP and Triton X‐100 were purchased from Boehringer‐Mannheim (Indianapolis, IN). Bradykinin, H‐89, Bt2cGMP, okadaic acid and the ATP assay kit were purchased from Calbiochem‐Novabiochem Int. (La Jolla, CA). cAMP and cGMP assay kits were purchased from Amersham Corp. (Arlington Heights, IL). Anti‐phospho‐CREB rabbit polyclonal antibody was supplied by Upstate Biotechnology, Waltham, MA. Fetal calf serum was purchased from Gibco (Grand Island, NY). HPLC columns were supplied by Krackler Scientific (Durham, NC). Monoflow 4 scintillant was purchased from National Diagnostics (Manville, NJ). InsP6 was purchased from Aldrich (Milwaukee, WI). Fura‐2/AM was purchased from Molecular Probes, Eugene OR. Other reagents were purchased from either Sigma or Fluka.
We thank Dr Sol Snyder and colleagues for providing us with their unpublished manuscript on the properties of the [PP]‐InsP5 kinase. We are grateful to Dr Ribeiro for the [Ca2+] measurements. We also thank Drs J.J.Caffrey, C.M.P.Ribeiro and J.W.Putney,Jr for their comments on this manuscript.
- Copyright © 1998 European Molecular Biology Organization