Cyclopiazonic Acid Depletes Intracellular Ca2+ Stores and Activates an Influx Pathway for Divalent Cations in HL-60 Cells*

The filling state of intracellular Ca2+ stores has been proposed to regulate Ca2+ influx across the plasma membrane in a variety of tissues. To test this hypoth-esis, we have used three structurally unrelated inhibitors of the Ca2+-ATPase of intracellular Ca2+ stores and investigated their effect on Ca2+ homeostasis in HL-60 cells. Without increasing cellular inositol (1,4,5)trisphosphate levels, all three inhibitors (cyclo- piazonic acid, thapsigargin, and 2, 5-Di-tert-butylhy-droquinone) released Ca2+ from intracellular stores, resulting in total depletion of agonist-sensitive Ca2+ stores. The Ca2+ release was relatively slow with a lag time of 5 s and a time to peak of 60 s. After a lag time of approximately 15 s, all three Ca2+-ATPase inhibitors activated a pathway for divalent cation influx across the plasma membrane. At a given concentration of an inhibitor, the plasma membrane permeability for divalent cations closely correlated with the extent of depletion of Ca2+ stores. The influx pathway activated by Ca2+-ATPase inhibitors conducted Ca2+, Mn2+, Co2+, Zn2+, and Ba2+ and was blocked, at similar concentrations, by La3+, Ni2+, Cd2+, as well as by the imidazole derivate SK&F 96365. The divalent cation influx in response

The filling state of intracellular Ca2+ stores has been proposed to regulate Ca2+ influx across the plasma membrane in a variety of tissues. To test this hypothesis, we have used three structurally unrelated inhibitors of the Ca2+-ATPase of intracellular Ca2+ stores and investigated their effect on Ca2+ homeostasis in HL-60 cells. Without increasing cellular inositol (1,4,5)trisphosphate levels, all three inhibitors (cyclopiazonic acid, thapsigargin, and 2, 5-Di-tert-butylhydroquinone) released Ca2+ from intracellular stores, resulting in total depletion of agonist-sensitive Ca2+ stores. The Ca2+ release was relatively slow with a lag time of 5 s and a time to peak of 60 s. After a lag time of approximately 15 s, all three Ca2+-ATPase inhibitors activated a pathway for divalent cation influx across the plasma membrane. At a given concentration of an inhibitor, the plasma membrane permeability for divalent cations closely correlated with the extent of depletion of Ca2+ stores. The influx pathway activated by Ca2+-ATPase inhibitors conducted Ca2+, Mn2+, Co2+, Zn2+, and Ba2+ and was blocked, at similar concentrations, by La3+, Ni2+, Cd2+, as well as by the imidazole derivate SK&F 96365. The divalent cation influx in response to the chemotactic peptide fMLP had the same characteristics, suggesting a common pathway for Ca2+ entry. Our results support the idea that the filling state of intracellular Ca2+ stores regulates Ca2+ influx in HL-60 cells.
Activation of myeloid cells by cell surface agonists causes Ca2+ release from internal stores and Ca2+ influx across the plasma membrane (1-3). It has been clearly demonstrated that the release of Ca2+ from internal stores is mediated by inositol (1,4,5)trisphosphate (Ins(1,4,5)P3)l (4-7). In contrast, the mechanism underlying receptor-mediated Ca2+ entry is poorly understood. In myeloid cells, it has been proposed that 'This research was supported by Grant 32-30161.90 from the Swiss National Foundation for Scientific Research and by a grant from the Swiss Ligue against Rheumatism. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18  Ins(1,3,4,5)P4 might play a role in the regulation of Ca2+ entry (3) or that the initial [Ca2+Ii increase, due to Ca2+ release from intracellular stores, activates a Ca2+-conductive plasma membrane channel (8). However, subsequently it has been shown that [Ca2+], increases are neither necessary nor sufficient to induce Ca2+ influx (9). Studies in various other cell types have suggested a correlation between depletion of intracellular Ca'+ pools and Ca2+ entry across the plasma membrane (10-13). Thus, according to the so-called capacitative model of Caz+ entry, the filling state of the intracellular Ca2+ stores determines the Ca" permeability of the plasma membrane. Recently, three inhibitors of Ca2+-ATPases of intracellular Ca2+ pools have been described, thapsigargin (TG), cyclopiazonic acid (CPA), and 2,5-Di-tert-butylhydroquinone2 (DBHQ) (14)(15)(16). All three, structurally unrelated, compounds inhibit Ca2+ uptake in the Ins(l,4,5)P3-sensitive Ca2+ pool in HL-60 ho-mogenate~.~ In accordance with the capacitative model of Ca2+ entry, thapsigargin added to intact cells not only depletes Ins(l,4,5)P3-sensitive Ca2+ stores, but also induces Ca2+ influx in a variety of cellular systems (for synopsis see Table 2 of Ref. 14). In contrast, studies in hepatocytes did not observe Ca2+ influx in response to DBHQ, despite depletion of Ins( 1,4,5)PS-sensitive intracellular Ca2+ stores by the compound (17)(18)(19). The Ca2+-ATPase inhibitor CPA has so far not been tested for its effects on Caz+ release or Ca2+ influx in intact cells.
If, in HL-60 cells, the capacitative model of Ca2+ entry reflects indeed the mechanism of receptor-mediated Ca2+ influx, experimental analysis should confirm the following theoretical predictions. 1) All ca'+-ATPase inhibitors that efficiently empty intracellular Ca2+ stores should induce Ca2+ influx with dose responses similar to Ca2+ release from stores. 2 ) Emptying of the Ca2+ stores should precede Ca2+ influx. 3) Various cations should, in a similar manner, either permeate or block the Ca'+ influx pathway activated by fMLP and the three inhibitors.
In this study we demonstrate that all three Ca2+-ATPase inhibitors induced Ca2+ release from Ins( 1,4,5)P3-sensitive Ca2+ stores and promoted Ca2+ influx across the plasma membrane in HL-60 cells. Emptying of the stores preceded Ca2+ influx. The influx pathway activated by the receptor agonist fMLP and by the Ca2+-ATPase inhibitors conducted Ca2+, Mn2+, Co2+, Zn2+, and Ba2+. The sensitivity of the influx pathway to block by La3+, Ni2+, and Cd2+ was identical, whether it was activated by fMLP or by the Ca2+-ATPase inhibitors.
Culture of HL-60 Cells-Cells were cultured in RPMI 1640 medium supplemented with 10% fetal calf serum, penicillin (5 units/ml) and streptomycin (50 pg/ml). The cells were passaged twice every week and differentiated by adding Me230 (final concentration 1.3% v/v) to the cell suspension 7 days before experiments.
[Ca"], Measurements-[Ca2+], was measured with the fluorescent Ca'+ indicator fura-2 as described previously (3). Cells (2 X 107/ml) suspended in CaZ+ medium containing 0.1% bovine serum albumin were loaded for 45 min at 37 'C with 2 ~L M fura-P/AM, then diluted to 10'/ml, and kept on ice. Just before use, a sample of 5 X lo6 cells was centrifuged and resuspended in the indicated medium. Fluorescence measurements were performed on a Perkin-Elmer fluorimeter (LS3, Perkin-Elmer Cetus), thermostated at 37 "C. Excitation and emission wavelengths were 340 and 505 nm, respectively. Calibration of fura-2 fluorescence was performed as described previously (1). In additional experiments, similar results were obtained with excitation ratio measurements at 340380 nm.
Assessment of Mn" Influx-Entry of Mn2+ into cells was measured using the fura-2 fluorescence quenching technique. Cells were loaded as for [Ca"], measurements and fura-2 fluorescence monitored in a Ca'+-containing medium at the Ca2+-independent excitation wavelength 360 nm. Then, Mn2+ (0.5 mM) was added and its entry into cells measured as rate of fluorescence decrease during the first 2 min following its addition. We routinely added the heavy metal chelator DTPA (2 mM) at the end of an experiment to measure the'percentage of fluorescence decrease due to extracellular fura-2 quenching. It accounted for less than 5% of the total fluorescence quenched by Mn'+ and was subtracted before calculation of the rate of Mn2+ quenching. No increase in extracellular fura-2 was observed after exposure of cells to the CaZ+-ATPase inhibitors, demonstrating that these compounds were not cytotoxic.
Assessment of Plasma Membrane Permeability for Other Divalent Cations-Activation of the influx pathway for divalent cations was assessed using a procedure analogous to the Mn2+ quenching technique. For the measures of cations/fura-2-free acid fluorescence, fura-2-free acid (10 p~) was dissolved in a Ca2+-containing medium and the fluorescence measured at the Ca2+-independent excitation wavelength of 360 nm. Similar to Mn2+, the divalent cation Co2+ quenched the fluorescence of fura-2, while Zn2+ and Ba2+ increased fluorescence. T o study the entry of the respective ions into intact cells, fluorescence of fura-2 loaded cells was monitored at 360 nm. Co2+ (0.1 mM), Zn2+ (0.1 mM), or Ba2+ (5 mM) were added, and divalent cation entry was detected as fluorescence decrease (Coz+) or increase (Zn", Ba2+). DTPA (2-20 mM) was added at the end of the experiment to estimate the percentage of the fluorescence changes due to extracellular fura-2. Sensitivity of the Co2+, Zn2+, and Ba2+ entry pathway to inorganic Ca" channel blockers was assessed by preincubating the cells with Ni2+ (10 mM), Cd2+ (10 mM), or La3+ (100 pM).
Inositol Phosphates Measurements-Inositol phosphate concentrations were measured as described previously (20).
Data Recording and Analysis-Fluorescence values are given as percent of total Ca2+-dependent fura-2 fluorescence (for calibration, see Ref. 1). Fluorescence traces were digitalized using a 12-bit Multi-1/0 A-D converter (Acqui, SICMU, Geneva, Switzerland). Data were recorded at a rate of 50 Hz, filtered with a moving average procedure, and stored on an A T computer. Statistical analysis and fitting procedures were done with Sigmaplot software (Jandel Scientific, Corte Madera, CA). Values are given as mean (range) of 4 to 25 different experiments performed on at least 2 different days.
The fMLP-induced [Ca2++Ii increase was transient, while the [Ca"], increases caused by the three inhibitors were sustained for the time of observation (up to 20 min). The time to plateau of the [Ca2+Ii increase in response to a maximally effective concentration of CPA was 260 (120-360) s (n = 25, mean (range)) and thus markedly longer than the time to peak observed after stimulation with fMLP (19 (15-30) s, n = 15, mean (range)).
CPA induced [Ca2+Ii elevations when cells were suspended in Ca"-free medium, demonstrating that it releases Ca2+ from intracellular stores (Fig. 2 A ) . Ca2+ release due to CPA was slow compared to fMLP-induced Ca2+ release (time to peak 55 (40-80) s, n = 12 versus 13 (12-16) s, n = 15, mean (range)), but was comparable to the Ca'+ release produced by TG and DBHQ (not shown). The dose-response curves obtained in Ca'+-free medium were similar to those obtained in Ca2+ medium (Figs. 2B and lB, respectively). However the [Ca2+]i increase in Ca2+-free medium was only transient, raising the possibility that the prolonged Ca2+ increase observed in Ca'+ medium was due to Ca2+ influx across the plasma membrane.
Source of the Ca2+ Mobilized by the Inhibitors-Addition of maximally effective concentrations of CPA before exposure t o fMLP in Ca2+-free medium, entirely abolished fMLPinduced Ca2+ release (Fig. 3A, right trace), presumably by depleting the Ins(l,4,5)P3-sensitive Ca2+ store. In contrast, after addition of a maximally effective concentrations of fMLP, CPA was still able to release Ca'+ ( Fig. 3A, left trace). The dose-inhibition curves of fMLP-induced Ca2+ release by the three inhibitors ( Fig. 3B) were similar to the dose responses for induction of [Ca2+IL increase; half-maximal inhibition was obtained with 2.8 (2-6) p~ CPA, 3.8 (1-8) nM TG, and 2.6 (1-5) p~ DBHQ ( n = 4-6, mean (range)). 100% inhibition was achieved with 30 p~ CPA, 30 nM TG, and 30 PM DBHQ. To estimate which percentage of total intracellular Ca'+ stores was depleted by the Ca"-ATPase inhibitors, we added, in the absence of extracellular Ca2+, the Ca'+ ionophore ionomycin to cells preincubated with or without the Ca2+-ATPase inhibitors. After exposure to a maximal dose of CPA, TG, or DBHQ, the amount of ionomycin-releasable Ca2+ was less than 10% of the amount of ionomycin-releasable Ca2+ in the absence of inhibitors (data not shown). These results show that most of the intracellular Ca'+ stores can be depleted by any of the three Ca2+-ATPase inhibitors while the pool that is released by fMLP is only a part of the total pool that is loaded by CPA-, TG-, and DBHQ-sensitive Ca2+-ATPase.
Induction of Ca2+ Influx by CPA, TG, and DBHQ-As the sustained increase in [Ca2+], upon exposure to CPA, TG, and DBHQ was not observed in the absence of extracellular Ca2+, we next investigated whether it was due to Ca2+ influx. The divalent cation Mn2+ has been shown to permeate through the neutrophil Ca'+ influx pathway activated by chemotactic peptides (2, 3, 9). Its intracellular presence can be easily detected as it binds with a high affinity to Ca2+-sensitive fluorescent dyes and thereby quenches their fluorescence. If the Ca2+ influx pathway activated by the Ca2'-ATPase inhibitors is identical to the pathway activated by chemotactic peptides, it should also conduct Mn2+. We therefore compared the rate at which Mn2+ quenched the fura-2 fluorescence in cells preincubated with either CPA, TG, DBHQ, or the agonist fMLP. The experiments were performed at an excitation wavelength of 360 nm, a wavelength a t which the fluorescence of fura-2 is independent of [Ca"],. When added to control cells, Mn2+ caused a slow and continuous decrease in fluorescence. After exposure of cells to a maximal dose of fMLP for 1 min, this basal quenching rate was 2-fold increased (1.9 (1.8-2.1), n = 25, mean (range)) (see also Ref. 3). CPA (0.3-30 p M ) caused a concentration-dependent increase of the rate of fura-2 quenching (Fig. 4A). The dose-response curves for CPA, TG, and DBHQ were similar to the dose-response curves for [Ca2+Iz mobilization (Fig.  4B). The increase in fura-2 quenching by Mn2+ with maximal concentrations of Ca2+-ATPase inhibitors was 2.5-(2.2-2.7), 3.2-(2.8-3.5), and 2.4-(2.1-2.8) fold of basal for CPA, TG, and DBHQ, respectively ( n = 11-14, mean (range)). Thus, maximal concentrations of CPA, TG, and DBHQ caused more Mn2+ influx than 1O"j M fMLP. The fluorescence quenching by Mn2+ was not reversed by addition of the non-permeant heavy metal chelator DTPA (2 mM) and was therefore due to Mn2+ influx and not to quenching of extracellular fura-2. In a different set of experiments, Ca'+ instead of Mn2+ was added to cells after incubation with the Ca2+-ATPase inhibitors in Ca2+-free medium. Immediate increases in fura-2 fluorescence with dose-response curves similar to those for Mn2+ were observed (data not shown).
Permeation and Block by Divalent Cations of the Influx Pathway Activated by CPA and fMLP-We next investigated the sensitivity to inorganic channel blockers of the influx pathway activated by the Ca"-ATPase inhibitors and fMLP. Cells were preincubated with Ni2+, Cd", or La3+ for 1 min and then exposed to CPA (30 p M ) for 1 min prior to Mn2+ addition. The three cations inhibited the Mn2+ influx induced by CPA in a dose-dependent manner (Fig. 5A). Half-maximal inhibition was obtained with 1.1 (0.9-1.2) mM Ni2+, 2.4 (1.8-3.2) mM Cd2', and 53 (24-82) p M La3+ ( n = 4, mean (range)). Complete inhibition was obtained with 10 mM Cd2+, 10 mM Ni", and 100 p~ La3+. The Mn2+ influx induced by fMLP, TG, and DBHQ was blocked by Ni2+, Cd", and La3+ with comparable dose-inhibition curves (Fig. 5B), suggesting that agonist and Ca*+-ATPase inhibitors activate a similar pathway for Mn2+ influx. In contrast to Ni", Cd2+, and La", several other cations penetrated through the influx pathway.

Ca2+-ATPase
T o study this question, we took advantage of the fact that, similar to Mn2+, various divalent cations are able to bind to fura-2 and to alter its fluorescence. We observed that, at an excitation wavelength of 360 nm, Co2+ quenched the fura-2 free-acid fluorescence while Zn2+ and Ba2+ increased it. When unstimulated fura-2-loaded cells were exposed to either 0.1 mM co2+, 0.1 mM zn2+, or 5 mM Ba2+ only a slow increase or decrease in fluorescence was observed. However, after addition of a maximal dose of fMLP, CPA, TG, or DBHQ, a marked fluorescence increase (Zn2+ and Ba2+) or decrease (Co2+) was observed (data not shown). This suggests that the Ca2+ and Mn'+ influx pathway activated by CPA, fMLP, TG, and DBHQ also conducts Zn", Ba", and Co2+. Changes in fluorescence were not reversed by the addition of DTPA, indicating that the cations were bound to intracellular fura-2. As in the case of Mn2+, the influx of these cations was entirely inhibited by Ni" (10 mM), Cd2+ (10 mM), and La3+ (100 p~) .
Thus, the influx pathway activated by the agonist fMLP and by the Ca'+-ATPase inhibitors was permeant to Ca", Mn'+, Co2+, Zn", and Ba2+, and had the same sensitivity t o block by La", Ni", and Cd".
Time Course of Activation of the Influx Pathway for Divalent Cations-If depletion of intracellular stores is a signal for Ca"+ influx, Ca2+ release should precede influx. To determine the activation kinetics of the influx pathway for divalent cations induced by the Ca2+-ATPase inhibitors, we determined fura-2 fluorescence quenching by Mn2+ at different times after exposure of cells to a maximal dose of CPA (Fig.  6A). The time course of Mn2+ influx induced by CPA differed from that of Mn2+ influx triggered by fMLP (see also Ref. 3). In cells exposed to fMLP, the rate of fura-2 quenching increased after 10 s, was maximal after 30 s, and returned to basal levels after 6 min. In cells preincubated with CPA, the rate of fura-2 quenching started to increase after a lag period of approximately 15 s, reached its maximum at 1 min, and remained elevated for the period of observation (Fig. 6B). Thus, the Ca2+-ATPase inhibitors caused a continuous increase in plasma membrane permeability for divalent cations, while fMLP produced only a transient increase. In contrast to the delayed onset of Mn2+ influx (>15 s, n = 3), the onset of Ca2+ release from internal pools was more rapid. The Caz+ rise in response to CPA started after 4.3 (3-6) s, ( n = 7, mean (range)) when assessed in a Ca2+-free medium (see also Fig.  6B) and 3.8 (2.5-5) s ( n = 6, mean (range)) when assessed in a Ca2+-containing medium. While at 15 s no Mn2+ influx was observed, Ca2+ release, i.e. the increase of fura-2 fluorescence in Ca2+-free medium, was already 11.8 (8-15)% ( n = 7, mean (range)) i.e. more than half of total, at this time point. The increase of fura-2 fluorescence in Ca2+-containing medium was 12.6 (7-19)% ( n = 6, mean (range)) at 15 s, i.e. identical to the one observed in Ca2+-free medium. Similar results were obtained with the two other inhibitors (data not shown). Thus, by two criteria Ca2+ release preceded Ca2+ influx. (i) Lag time and initial rate of [Caz+]i rise were identical in Ca2+containing and Ca2+-free medium, suggesting that there is no early Ca2+ influx component and (ii) Mn2+ influx could be observed only with a delay of >10 s after the beginning of Ca2+ release. Thus, compatible with the concept of regulation of Ca2+ influx by the filling state of intracellular Ca2+ stores, Ca2+-ATPase inhibitors first induce Ca2+ release and then Ca2+ influx.

DISCUSSION
This study is, to the best of our knowledge, the first demonstration that the Ca2+-ATPase inhibitor CPA depletes intracellular Ca2+ stores and activates an influx pathway for divalent cations. The action of CPA is therefore, in HL-60 cells, undistinguishable from the effects of two other, struc-  1 1 10 2 0 0.1 1 10 0.1 1 1 0   6 5. Ni2+, Cd2+, and La3+ block the CPA-induced Mn2+ influx in a concentration-dependent manner. A, fura-2-loaded cells were preincubated with either Cd'+, Ni", or La'+ for 1 min and subsequently exposed to CPA (30 p~) and, 1 min later, to Mn2+ (0.5 mM). The traces are superimposed for comparison. Increasing the doses of Ni'+ or Cd2+ reduced the rate of Mn2+ quenching. B, fura-2 quenching rate of cells exposed to CPA, fMLP, TG, and DBHQ was plotted against Ni", Cd'+, or La3+ concentration. The IC50 of the different cations (in mM for Ni'+ and Cd", in p~ for La3+) for inhibition of fura-2 quenching by Mn2+ are listed below (mean (range)): turally unrelated inhibitors of the microsomal Ca"-ATPase, DBHQ, and TG. This identical cellular response to three structurally unrelated compounds provides new and strong evidence for the concept of regulation of Ca2+ influx by the filling state of the intracellular Caz+ stores. CPA, TG, and DBHQ have been shown to inhibit ATPdependent Ca'+ uptake in microsomal fractions of HL-60 cells,:' but the mechanism by which they mobilize Ca"' from Ins(l,4,5)P3-sensitive stores in intact cells is not understood. I t might be due to a relatively high Ca2+ permeability of the Ins(l,4,5)P3-sensitive pool in resting cells which would necessitate a permanent Ca2+-ATPase activity to maintain resting [Ca"],. Compatible with this hypothesis, studies in unstimulated HL-60 cells found a basal Ins(1,4,5)P3 concentration of approximately 300 nM (22). As the half-maximal concentration of Ins(1,4,5)P3 necessary to release Ca2+ from intracellular stores in permeabilized neutrophils is 750 nM (4), these concentrations would be expected to increase moderately the Ca2+ permeability of intracellular Ca2+ stores. Alternatively, one might consider that the Ca2+-ATPase inhibitors alter the Ca2+-ATPase structure in a manner that allows reversal of the Ca2+ flux through the pump. The latter event may occur a t least under certain experimental conditions (for review see Ref. 23).
Ca2+ release occurred approximately 4 s, Ca2+ influx approximately 15 s after addition of Ca*+-ATPase inhibitors. Thus, release of Ca2+ from intracellular Ca'+ stores preceded Ca2+ influx by more than 10 s. This temporal sequence of initial Ca2+ release and delayed Caz+ influx is also seen with the chemotactic peptide fMLP (3). However, due to the slower kinetics of the [Cat+]; increase, this phenomenon can be more clearly demonstrated with the Ca"-ATPase inhibitors (Fig.  6).
While in HL-60 cells all three Ca2+-ATPase inhibitors induce Ca2+ influx, one study showed that in rat hepatocytes only TG, but not DBHQ induced Ca2+ influx, although both compounds induced Ca2+ release (24). In contrast, the same study found that in a lymphocyte cell line DBHQ induced Caz+ influx. The reason for the differing effects of DBHQ in hepatocytes is not clear. It might be due to effects of the compound others than inhibition of the Ca2+-ATPase (25). It will be particularly important to test cyclopiazonic acid in these cells.
Several lines of argument suggest that the Ca2+ influx induced by CPA, TG, and DBHQ in HL-60 cells is due to specific inhibition of the Ca2+-ATPase of intracellular Ca2+ stores and not due to a nonspecific effect.
1) The plasma membrane permeability induced by the Ca2+-ATPase inhibitors could be blocked by the inorganic channel blockers La3+, Ni2+ , and Cd". 2) The three Ca2+-ATPase inhibitors are structurally unrelated, and it is therefore unlikely that they possess a common target other than the intracellular Ca'+-ATPase. 3) in particular, none of the three compounds is able t o inhibit the plasma membrane/erythrocyte type Ca2+-ATPase (14)(15)(16). 4) All three compounds act first on intracellular Ca'+ stores and induce Ca2+ influx with a delay, while a nonspecific plasma membrane perturbation would be expected to precede the effect on an intracellular organelle. 5) None of the compounds released fura-2 from the cytosol of fura-2loaded HL-60 cells (see "Experimental Procedures"), i.e. they were not cytotoxic.
Although there is now convincing experimental documentation of the induction of Ca2+ influx by TG in many cell types (14), and at least one report of such an effect of DBHQ (24), it is presently not known if the influx pathways activated by physiological agonist and by Ca2+-ATPase inhibitors is the same. In this study we have extensively characterized permeation and block by various cations, as a mean to search for putative differences between the fMLP-and Ca*+-ATPase inhibitor-induced Ca2+ influx pathways. Previous research has shown that both permeation and block by cations differ markedly among various Ca2+ entry pathways and that this type of analysis is therefore a sensitive tool to search for heterogeneity of Ca2+ influx pathways. For example, the Zn2+ impermeability of the influx pathway in a mast cell line (26) distinguishes it from the one found in HL-60 cells. In hepatocytes, a Ni*+ and Cd2+ permeable agonist-induced Ca2+ influx pathway has been described (27), whereas in HL-60 cells both cations block Ca2+ influx (Fig. 5). In the case of voltage-dependent Ca2+ channels, the sensitivity to inorganic channel blockers allows to discriminate T-type, N-type, and L-type Ca'+ channels (19,(28)(29)(30). The N-type channel, for example, has been described to be approximately 50-fold more sensitive to Cd2+ than the T-type channel (19,(28)(29)(30). The Ca2+ influx pathway activated by both fMLP and Ca2+-ATPase inhibitors is permeant for Mn2+, Co2+, Ba2+, and Zn2+ and blocked by Cd2+, La3+, and Ni2+ at the same concentrations. In addition, the influx pathway activated by fMLP and Ca2+-ATPase inhibitors is equally sensitive to the imidazole derivative SK&F 96365. The Ca2+ influx pathway activated by CPA, fMLP, TG, or DBHQ thus exhibits an undistinguishable profile of block and permeation. These results strongly suggest that the Ca2+-ATPase inhibitors and the chemotactic peptide activate an identical Ca2+ influx pathway in HL-60 cells.
How might Ca2+-ATPase inhibitors stimulate Ca2+ influx? TG, CPA, and DBHQ inhibit Ca2+ uptake into intracellular Ca2+ stores and thereby deplete them. Thus, the filling state of intracellular Ca2+ stores is an obvious candidate for the mediation of the Ca2+-ATPase inhibitor-induced Ca2+ influx. However, during the depletion of the stores by the inhibitors, a [Ca'+], rise due to Ca2+ release can be observed. For the following reasons it seems unlikely that this Ca'+ rise mediates Ca2+ influx in myeloid cells. (i) A [Ca'+], rise is not a necessary stimulus for Ca2+ influx: as demonstratedpreviously (9), Mn2+ influx may occur despite virtually complete buffering of the agonist-induced Caz+ release. (ii) The amplitude and the time course of the [Ca2+Ir rise are not correlated with the magnitude and the time course of Ca'+ influx: fMLP induces a Caz+ release with a larger amplitude than the Ca2+-ATPase inhibitors, but induces a quantitatively minor and shorter lasting Ca'+ influx. Ca2+-ATPase inhibitors cause, in Ca2+-free medium, a [Ca2+Ii rise that lasts for approximately 2 min; however, the influx pathway for divalent cations remains active for the period of observation (up to 20 min). Thus, although, at this point, we cannot exclude some modulatory role of a [Ca"], rise, it does not appear to be the crucial messenger for the mediation of Ca2+ influx in myeloid cells. In contrast, our results are compatible with a key role of the filling state of the intracellular Ca2+ pool in the regulation of Ca2+ influx.
The mechanism of regulation of Ca'+-influx by the filling state of intracellular Ca2+ stores must comprise at least three components: (i) an intravesicular Caz+ sensor; (ii) a transmembrane signaling step from the inside to the outside of the Ca'+ pool; and (iii) a signaling step from the Ca2+ pool to the plasma membrane. The molecular basis of these steps is not yet known. The Ca"-ATPase inhibitors will be powerful tools to further elucidate the mechanism of Ca2+ influx in myeloid cells and other nonexcitable tissues.