Store-operated Ca2+ Entry and Coupling to Ca2+ Pool Depletion in Thapsigargin-resistant Cells*

(Received for publication, July 9, 1996, and in revised form, November 18, 1996)

Richard T. Waldron Dagger , Alison D. Short § and Donald L. Gill

From the Department of Biochemistry and Molecular Biology, University of Maryland School of Medicine, Baltimore, Maryland 21201

ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS AND DISCUSSION
FOOTNOTES
Acknowledgments
REFERENCES


ABSTRACT

The release of Ca2+ from intracellular Ca2+ pumping pools and the entry of extracellular Ca2+ are tightly coupled events. The potent and specific intracellular Ca2+ pump inhibitor, thapsigargin, blocks Ca2+ accumulation and allows Ca2+ release from pools within mammalian cells, inducing major changes in endoplasmic reticulum function and cell growth. Recent studies characterized the pools of Ca2+ within permeabilized DC-3F/TG2 cells (a thapsigargin-resistant variant form of the DC-3F Chinese hamster lung fibroblast line, able to grow in 2 µM thapsigargin), revealing highly thapsigargin-resistant intracellular Ca2+ pumping activity capable of accumulating Ca2+ within an inositol 1,4,5-trisphosphate-releasable Ca2+ pool (Waldron, R. T., Short, A. D., and Gill, D. L. (1995) J. Biol. Chem. 270, 11955-11961). Using intact fura-2-loaded thapsigargin-resistant DC-3F/TG2 cells, the present study investigated the role of this unusual Ca2+ pumping activity in maintaining cytosolic Ca2+, generating Ca2+ signals, and mediating Ca2+ entry. The thapsigargin-resistant Ca2+ pumping pool was capable of generating rapid cytosolic Ca2+ signals in response to the phospholipase C-coupled agonist, oleoyl lysophosphatidic acid. The resting level of cytosolic Ca2+ in DC-3F/TG2 cells was 2-fold elevated compared with control cells (the parent DC-3F line), and transient extracellular Ca2+ removal induced a large "overshoot" in cytosolic Ca2+. The overshoot response was blocked by the Ca2+ influx inhibitor, SKF96365, and was kinetically identical to that induced in parent DC-3F cells after thapsigargin-induced Ca2+ pool emptying, indicating that the thapsigargin-resistant DC-3F/TG2 cells had "constitutively" opened Ca2+ entry channels coupled to an emptied or partially emptied thapsigargin-sensitive Ca2+ pumping pool. Even though oleoyl lysophosphatidic acid-mediated Ca2+ release induced little Ca2+ entry, complete ionomycin-activated emptying of the thapsigargin-resistant Ca2+ pool in DC-3F/TG2 cells induced a large, sustained entry of Ca2+ that was also completely blocked by SKF96365. The results revealed that the thapsigargin-resistant Ca2+ pump does maintain physiological Ca2+ levels, is able to fill an agonist-responsive Ca2+ pool in DC-3F/TG2 cells, and is likely responsible for the ability of these cells to function and grow in the presence of thapsigargin. In addition, Ca2+ influx in the resistant DC-3F/TG2 cells reflects emptying of pools that accumulate Ca2+ by both thapsigargin-sensitive and -resistant Ca2+ pumps; since these pumps accumulate Ca2+ in distinct pools in parent DC-3F cells, it is possible that more than one pool is coupled to Ca2+ influx in the resistant DC-3F/TG2 cells.


INTRODUCTION

Ca2+ signals in many cells comprise both release of Ca2+ from intracellular pools and entry of Ca2+ across the plasma membrane. These two events are closely coupled, entry of Ca2+ being triggered by emptying of Ca2+ from pools in response to InsP31 (1, 2). Whereas the precise localization of Ca2+ pools involved in activating Ca2+ entry is uncertain (3, 4), in general it appears that the ER or subcompartments thereof are the site of InsP3-mediated Ca2+ release (5-7). Quite substantial changes in intraluminal Ca2+ levels occur in response to InsP3-induced Ca2+ release from the ER (8). Such changes in ER Ca2+ are responsible for a number of different cellular responses. Thus, the decrease in Ca2+ stored within the ER appears to be the primary determinant for activating the opening of store-operated channels (SOCs)2 in the plasma membrane, allowing the entry of Ca2+, which serves to enhance cytosolic Ca2+ signals as well as allow refilling of intracellular pools (1-3). The level of Ca2+ in pools also appears to control the function of InsP3-sensitive Ca2+ release channels (5, 6). In addition to controlling Ca2+ signal generation within the cytosol, the level of Ca2+ within pools mediates important control over intraluminal events including the essential ER functions of translation, folding, processing, and assembly of proteins (9-13); such effects may be mediated by the large array of intraluminal Ca2+-binding proteins, several of which function as molecular chaperones (14, 15). Last, it has become clear that the Ca2+ content of intracellular Ca2+ pools exerts profound control over cell proliferation and progression of cells through the cell cycle (16-19).

The ER accumulates Ca2+ via the function of intracellular sarcoplasmic/endoplasmic reticulum Ca2+-ATPase (SERCA) Ca2+ pump proteins (20-22); these SERCA pumps have been shown to be highly sensitive to the Ca2+ pump blocker, thapsigargin (23, 24). Thapsigargin binds with extremely high affinity to intracellular Ca2+ pumps resulting in a virtually irreversible inhibition of Ca2+ accumulation within the ER (16, 24-26). Emptying of ER Ca2+ with either thapsigargin or other Ca2+ pump blockers, including 2,5-di-tert-butylhydroquinone and cyclopiazonic acid, causes cells to undergo one of two different growth responses. DDT1MF-2 smooth muscle cells progress through S-phase and become arrested in a quiescent G0-like state in which they remain viable and stable for as long as 7 days (16, 17). Other cells such as prostatic cancer cells enter an irreversible apoptotic state in which endonucleases are activated, DNA becomes fragmented, and cells undergo morphological degeneration and death (27). One criterion that may determine these growth responses to Ca2+ pump blockade is the extent of Ca2+ influx activated by pool emptying. Prolonged Ca2+ influx via SOCs and hence sustained elevated cytosolic Ca2+ appears to trigger apoptosis (4, 7, 27), whereas survival in a growth-arrested state is observed in cells in which Ca2+ influx is rapidly deactivated, such as DDT1MF-2 smooth muscle cells (7, 17, 28). In the latter line, application of high serum or arachidonic acid induces the pool-depleted quiescent cells to synthesize new pump protein, to develop new functional Ca2+ pools, and to reenter the cell cycle (17-19).

Even though thapsigargin-induced Ca2+ pool emptying has such profound effects on cell function and growth, a variant of the DC-3F Chinese hamster lung fibroblast cell line was recently developed that is resistant to thapsigargin (29). The resistant variant line, DC-3F/TG2, was generated by exposure of DC-3F cells to gradually increasing levels of thapsigargin over a 10-month period. Unlike the parent DC-3F cells that have a high sensitivity, cytotoxic response to thapsigargin, DC-3F/TG2 cells retain normal appearance and grow and divide in culture medium containing 2 µM thapsigargin (29). This remarkable resistance to thapsigargin did not appear to stem from expression of high levels of the multidrug resistance factor, P-glycoprotein, which can confer resistance to toxic hydrophobic molecules; nor did it appear to reflect substantially increased levels of SERCA pump expression (29). Instead, the resistance is attributable to expression of a novel intracellular Ca2+ pump activity with 20,000-fold lower sensitivity to thapsigargin (30). In spite of this vast difference in thapsigargin sensitivity, the resistant pump has similar high affinity for Ca2+ (Km 0.1 µM), ATP dependence, and sensitivity to vanadate as normal SERCA pumping activity (30). A small amount of what appears to be the same thapsigargin-resistant pump activity is detectable in parent DC-3F cells; however, it does not mediate any Ca2+ accumulation within InsP3-sensitive Ca2+ pools; for this reason, it was considered that in parent DC-3F cells the resistant Ca2+ pump functions to accumulate Ca2+ within a pool that is distinct from the InsP3-sensitive Ca2+ pool (30). In the resistant DC-3F/TG2 cells, InsP3-sensitive Ca2+ pools clearly exist, but in contrast with parent DC-3F cells, their uptake of Ca2+ occurs exclusively via the thapsigargin-resistant Ca2+ pump (30). In addition to this important Ca2+ pump distinction, other changes in Ca2+ pool function within resistant cells were noted, including differences in pool heterogeneity, anion permeability, and the translocation of Ca2+ between pools. From these results it was concluded that the cells contain two different Ca2+ pools distinguished by sensitivity to thapsigargin and that in the resistant DC-3F/TG2 cells, InsP3 receptors are expressed within the pool accumulating Ca2+ via the thapsigargin-insensitive Ca2+ pump (30).

Whereas these analyses using non-intact cells revealed that the resistant DC-3F/TG2 cells contained thapsigargin-resistant Ca2+ pumps capable of accumulating Ca2+ within InsP3-sensitive pools, no information was available on the physiological function of these pools in intact cells. Two important questions remained to be addressed. The first was whether the Ca2+ pumping activity operating within the thapsigargin-resistant cells was able to fill agonist-releasable Ca2+ pools. The second was to determine whether release of Ca2+ from such pools was coupled to SOC-mediated Ca2+ influx. The results indicate that Ca2+ pools inside the resistant DC-3F/TG2 cells are indeed functional, that the resistant pump accumulates Ca2+ within a pool that can be mobilized by agonists, and that emptying of this pool within the intact cells is coupled to activation of Ca2+ influx through authentic store-operated channels. The results also reveal that the activation of Ca2+ influx in the resistant DC-3F/TG2 cells reflects operation of both normal and resistant Ca2+ pumping activities possibly residing in distinct pools.


EXPERIMENTAL PROCEDURES

Culture of Parent DC-3F and Thapsigargin-resistant DC-3F/TG2 Cells

DC-3F Chinese hamster lung fibroblasts were cultured in alpha -modified Eagle's medium (Life Technologies, Inc.) supplemented with 5% heat-inactivated fetal bovine serum (Life Technologies, Inc.) as described previously (29, 30). Cells received a change of medium after 2 days and were passaged the following day. Selection of the thapsigargin-resistant DC-3F/TG2 cell line by successive culturing in the presence of increasing thapsigargin concentrations was described previously (29). DC-3F/TG2 cells were cultured in alpha -modified Eagle's medium with 5% heat-inactivated fetal bovine serum together with 2 µM thapsigargin; as for the parent line, the cells also received a change of medium (still containing 2 µM thapsigargin) on the second day after passaging and were passaged the following day. Both cell types were also cultured on glass coverslips for use in intracellular Ca2+ measurements.

Intracellular Free Ca2+ Measurements

The methodology for measurement of intracellular free Ca2+ within DC-3F and DC-3F/TG2 cells using fura-2 was as described previously (8, 17, 28, 30). Cells were grown on glass coverslips as described above and fura-2-loaded by incubation with 2 µM fura-2/AM ester for 10 min at 20 °C in Hepes-buffered Krebs medium (107 mM NaCl, 6 mM KCl, 1.2 mM MgSO4, 1 mM CaCl2, 1.2 mM KH2PO4, 11.5 mM glucose, 20 mM Hepes-KOH, pH 7.4, 0.1% bovine serum albumin). Under these conditions approximately 95% of dye was restricted to the cytosol as judged by the signal remaining after permeabilization with saponin. Fluorescence emission at 505 nm was monitored at 25 °C using a PTI dual wavelength spectrofluorometer system with excitation at 340 and 380 nm. Calculations of free intracellular Ca2+ concentrations were as described by Grynkiewicz et al. (31) using a Kd of 135 nM. Dye was considered saturated upon addition of 40 µM ionomycin, whereas minimum fluorescence ratio was determined in the presence of 10 mM EGTA together with ionomycin. For each of the traces shown, groups of 10-15 cells were analyzed. Unless indicated otherwise, all results are representative of a minimum of three experiments.

Materials and Miscellaneous

Parent DC-3F and the thapsigargin-resistant DC-3F/TG2 cell lines were kindly provided by Dr. Arif Hussain (University of Maryland Cancer Center). Thapsigargin was purchased from LC Services, Worcester, MA. Fura-2 was from Molecular Probes, Inc., Eugene OR. Miscellaneous procedures and materials were as described previously (16-18).


RESULTS AND DISCUSSION

Cytosolic Ca2+ Responses to Thapsigargin in Intact Parent DC-3F and Thapsigargin-resistant DC-3F/TG2 Cells

Whereas almost all mammalian cells accumulate Ca2+ within pools via intracellular Ca2+ pumps with extreme sensitivity to thapsigargin (24-26), the DC-3F/TG2 thapsigargin-resistant cell line contains highly resistant Ca2+ pump activity (30). Unknown was whether this distinct Ca2+ pumping could function to sustain physiologically operational Ca2+ pools within intact thapsigargin-resistant cells. Initial experiments examined Ca2+ levels in parent DC-3F and resistant DC-3F/TG2 cells and cytosolic Ca2+ changes in response to exogenously added thapsigargin. Using fura-2-loaded parent DC-3F cells grown on coverslips, application of a high dose of thapsigargin (2 µM) induced a rapid and large increase in cytosolic Ca2+ (Fig. 1A). Although the peak Ca2+ level partially subsided with time, the increase in Ca2+ was persistent with time as a result of continued influx of Ca2+ through store-operated Ca2+ channels (see below). As shown in Fig. 1A, 30 min after thapsigargin addition, the level of Ca2+ was still above 120 nM. From many experiments, the mean resting level of Ca2+ in these cells (i.e. without thapsigargin) was measured as 27.7 ± 7.0 nM (n = 94), and the mean level of Ca2+ after 30 min of thapsigargin treatment was 148.4 ± 33.1 nM (n = 52). Even several hours following thapsigargin treatment this increased level of Ca2+ was maintained. We noted that in another cell type, the DDT1MF-2 smooth muscle line, although thapsigargin induced a similarly rapid and large Ca2+ increase due to pool release, the ensuing SOC-mediated influx of Ca2+ was short-lived due to rapid deactivation (7, 17-19). Indeed, it was considered that the efficient deactivation of SOCs may explain why DDT1MF-2 cells remain viable (albeit in a growth-arrested state) and able to undergo growth recovery with high serum treatment for up to 7 days following thapsigargin treatment (7). In contrast, once treated with thapsigargin, DC-3F cells cannot be induced to reenter the growth cycle and undergo necrosis (30). This may result from the cytotoxic action of the prolonged increase in cytosolic Ca2+ that follows thapsigargin-induced pool release.


Fig. 1. Resting Ca2+ and cytosolic Ca2+ responses to thapsigargin in parent DC-3F cells and thapsigargin-resistant DC-3F/TG2 cells. Cytosolic Ca2+ responses were measured in fura-2-loaded cells attached to coverslips as described under "Experimental Procedures." Thapsigargin (100 nM or 2 µM) was added in the presence of normal external medium to either parent DC-3F cells (A) or DC-3F/TG2 (B) at the times indicated.
[View Larger Version of this Image (22K GIF file)]


In marked distinction from the parent DC-3F cells, the thapsigargin-resistant DC-3F/TG2 cells showed no response to exogenously added thapsigargin. Thus, as shown in Fig. 1B, no change in cytosolic Ca2+ was observed after addition of either 100 nM or 2 µM thapsigargin. Although there was no effect of thapsigargin, it was clear from the data in Fig. 1B that the resting Ca2+ level in these cells was higher than in the parent DC-3F line. Indeed, from a large number of experiments, the mean resting level of cytosolic Ca2+ in resistant DC-3F/TG2 cells was 70.4 ± 15.5 nM (n = 154), that is approximately double that in the parent line. This suggested that either there was less Ca2+ pumping activity in these cells or that persistent entry of Ca2+ was occurring perhaps as a result of some degree of pool emptying. Although levels of plasma membrane Ca2+ pumps are unknown, the level of SERCA pump protein in resistant DC-3F/TG2 cells is slightly increased relative to parent DC-3F cells (29). Experiments described below indicate that there is an endogenously activated increased basal level of Ca2+ influx in resistant DC-3F/TG2 cells.

Agonist-mediated Ca2+ Pool Release in Parent DC-3F and Resistant DC-3F/TG2 Cells

The absence of a thapsigargin-sensitive Ca2+ pool in intact DC-3F/TG2 cells was consistent with observations that these cells lack thapsigargin-sensitive Ca2+ pumping activity. It was also clear from previous experiments with permeabilized DC-3F/TG2 cells that they express a highly thapsigargin-insensitive Ca2+ pump (30). This activity has an IC50 for thapsigargin of approximately 4 µM, about 20,000-fold less sensitive than most of the Ca2+ pumping activity inside parent DC-3F cells, which has an IC50 for thapsigargin of approximately 200 pM (30). Interesting was the finding that a small fraction (approximately 20%) of the total pump activity in parent DC-3F cells has an IC50 for thapsigargin (4 µM) identical to that of pumping in resistant DC-3F/TG2 cells. From this it was inferred that both pump types are coexpressed in normal cells. However, each pump appears to function in a distinct Ca2+ pool; thus, from experiments with permeabilized DC-3F cells, the resistant pump did not function to accumulate Ca2+ within an InsP3-sensitive pool. In contrast, in permeabilized resistant DC-3F/TG2 cells, the resistant pump clearly did accumulate Ca2+ within an InsP3-releasable pool (30). The important question was therefore whether in intact DC-3F/TG2 cells this pool was coupled to agonist-induced Ca2+ signal generation.

As shown in Fig. 2A, release of agonist-sensitive Ca2+ pools in parent DC-3F cells is activated by LPA receptors coupled to phospholipase C stimulation and InsP3 production, as in many other cell types (32). Upon addition of a maximally effective dose of LPA (100 µM), there was a rapid increase in cytosolic Ca2+, reaching a peak within 30 s and thereafter returning to resting levels between 100 and 150 s later. The peak Ca2+ release from pools was not as high as with thapsigargin and did not include a significant influx component. Reapplication of LPA a second time after return to basal Ca2+ levels caused no further Ca2+ response (data not shown), likely reflecting desensitization of the LPA receptor. Importantly, the resistant DC-3F/TG2 cells also responded to LPA (Fig. 2B). The response in these cells was very similar to that in the parent line. The rate of onset was rapid and there was a fast termination of the response; the decrease in Ca2+ back to basal levels was consistently faster and often dipped below the starting Ca2+ level. Although the peak level of Ca2+ attained in resistant DC-3F/TG2 cells was a little higher, the peak height above resting Ca2+ was very similar with both cell types, approximately 100 nM. Thus, the size of the response did not appear to be affected by the increased level of basal Ca2+ within DC-3F/TG2 cells. Whereas the LPA response in parent DC-3F cells was completely abolished by prior treatment with 3 µM thapsigargin, the Ca2+ pump blocker had no effect on LPA-induced Ca2+ release in resistant DC-3F/TG2 cells (data not shown). These results are significant since they provide direct proof that the thapsigargin-resistant Ca2+ pumping within DC-3F/TG2 cells causes Ca2+ accumulation within an agonist-sensitive Ca2+ pool.


Fig. 2. Cytosolic Ca2+ responses to the mitogenic Ca2+ mobilizing agonist, LPA, in parent DC-3F and resistant DC-3F/TG2 cells. Fura-2-loaded DC-3F cells (A) or DC-3F/TG2 cells (B) were treated with 100 µM LPA at the times indicated. Measurements were as described under "Experimental Procedures."
[View Larger Version of this Image (21K GIF file)]


Coupling of Ca2+ Pools to Ca2+ Influx in Parent DC-3F and Resistant DC-3F/TG2 Cells

An important further question to address was the relationship between Ca2+ pool release and the activation of Ca2+ entry in both cell types. In the parent DC-3F cells, emptying of pools with thapsigargin appeared to induce quite substantial and long lasting Ca2+ influx (Fig. 1A). The smaller degree of Ca2+ influx observed following agonist-induced Ca2+ release (Fig. 2A) could reflect less than complete pool emptying by LPA, resulting from desensitization of the LPA receptor, as mentioned above. In apparent contrast, neither LPA nor thapsigargin induced Ca2+ entry in the resistant DC-3F/TG2 cells. To investigate activation of Ca2+ influx further, we took advantage of the fact that SOCs can be induced to open by transient removal of extracellular Ca2+. As described above, in many cells, SOC-mediated Ca2+ influx becomes partially or completely deactivated with time (7, 28), and brief exposure to low extracellular Ca2+ causes temporary reactivation of SOCs (28, 33). As shown in Fig. 3A, treatment of parent DC-3F cells with nominally Ca2+-free medium for 3 min had virtually no effect upon cytosolic Ca2+ levels. However, if the cells were first treated with thapsigargin for 30 min, the removal of Ca2+ caused a fast and very substantial decrease in cytosolic Ca2+ (Fig. 3B). Under this condition, pools have been emptied and SOCs remain at least partially activated, as shown in Fig. 1A. The removal of Ca2+ from the external medium has two effects; first, the driving force for Ca2+ entry through channels is greatly diminished and second, the lowered external Ca2+ level may increase the coupling efficiency of the plasma membrane Ca2+ pump as has been observed for operation of intracellular Ca2+ pumps with decreased luminal Ca2+ levels (34). The net result is a rapid decrease in cytosolic Ca2+, reaching a new equilibrium within 30 s. As discussed below, the lowered cytosolic Ca2+ level and possibly also the lowered extracellular Ca2+ level enhance opening of Ca2+ influx channels. When Ca2+ was added back, the open Ca2+ channels allowed Ca2+ to rapidly enter the cells and a transient overshoot of Ca2+ was observed (Fig. 3A). As channels returned to a partially deactivated state, the cytosolic Ca2+ level returned to almost the same level as before Ca2+ removal. The decrease in Ca2+ entry likely reflects an inhibitory effect of cytosolic Ca2+ on Ca2+ entry through SOCs, Ca2+ interacting with a site possibly close to the channel itself (2, 28, 33, 35). There is also evidence that Ca2+ may associate with an extracellular site (or possibly a site within the channel) and exert an inhibitory effect on SOC opening (36).


Fig. 3. Measurement of Ca2+ influx overshoots in response to external Ca2+ removal in parent DC-3F cells and resistant DC-3F/TG2 cells. Cytosolic Ca2+ was measured in normal DC-3F cells (A), in DC-3F cells already treated with 2 µM thapsigargin for 30 min (B), or resistant DC-3F/TG2 cells as described under "Experimental Procedures." Cell bathing medium was changed to nominally Ca2+-free medium for the periods shown by bars (200 s); at other times the medium contained 1 mM Ca2+.
[View Larger Version of this Image (18K GIF file)]


Significantly, as shown in Fig. 3C, the same brief extracellular Ca2+ removal applied to DC-3F/TG2 cells caused a cytosolic Ca2+ decrease followed by a substantial overshoot, both responses being similar to the responses seen with thapsigargin-treated parent DC-3F cells. This is important because it indicates that in the DC-3F/TG2 cells, Ca2+ influx channels are at least partially activated. The interpretation of this observation is that even though these cells have agonist-releasable pools that are not emptied, their state of SOC activity closely resembles that of the parent cells after thapsigargin-induced emptying of pools. Therefore, it appears that a thapsigargin-sensitive pool that operates to activate Ca2+ entry exists in the resistant DC-3F/TG2 cells and remains in an empty or partially empty state. Since the DC-3F/TG2 cells are continually exposed to thapsigargin during growth and since the action of thapsigargin is irreversible (16) it is reasonable that such a pool would be empty.

Even though the signal for SOC-mediated Ca2+ entry appears to be turned on in DC-3F/TG2 cells, the channels remain in a mostly deactivated state, as judged by the relatively low resting Ca2+ levels in these cells. A question arising is how reactivation of channels by transient external Ca2+ removal occurs and whether this process is the same in both DC-3F and DC-3F/TG2 cells. As shown in Fig. 4, experiments assessed the time dependence of Ca2+ removal on subsequent reactivation of Ca2+ influx in the two cell types. Using parent DC-3F cells treated for 30 min with 2 µM thapsigargin, extracellular Ca2+ was removed for 30 s to allow a new steady state to be attained. Normal extracellular Ca2+ was returned either immediately following this time or at 30-s intervals up to 2 min later (Fig. 4A). The immediate return of Ca2+ following the 30-s removal resulted in a rapid return of cytosolic Ca2+ to the original equilibrium level; under this condition there was no overshoot. Overshoots were observed following removal of Ca2+ for 1 min or longer; maximal overshoots were observed after removal of Ca2+ for 1.5 min or longer. Results from a series of experiments that analyzed the kinetics of the activation of overshoots showed that the peak level of Ca2+ influx resulted in a doubling of the basal Ca2+ level and that half-maximal peak height was induced between 0.5 and 1 min after the Ca2+ removal period (Fig. 4A, inset). This period of time is very similar to that shown by Zweifach and Lewis (35) for the current mediated by SOCs to recover in Jurkat T lymphocytes following Ca2+ removal. Most likely, the reactivation of Ca2+ influx reflects dissociation of Ca2+ from a Ca2+ binding site that mediates inhibitory control over SOC activity, either within cells (2, 35) or possibly on the outer surface (36). Clearly, the reactivation process that occurs in the DC-3F/TG2 cells following Ca2+ removal is virtually identical (Fig. 4B). In this case, even though the peak reactivation following Ca2+ readdition was slightly less than with parent DC-3F cells, it represented a 3-fold enhancement over the basal level of Ca2+ influx occurring; the time dependence for reactivation was very similar to parent cells. Therefore, it appears that the state of activation of SOCs and the mechanisms for deactivating and reactivating Ca2+ influx remain the same for both resistant DC-3F/TG2 cells and the thapsigargin-treated parent DC-3F cells.


Fig. 4. Kinetics of activation of Ca2+ influx overshoots in DC-3F cells and DC-3F/TG2 cells. Ca2+ responses were measured either in parent DC-3F cells pretreated for 30 min with 2 µM thapsigargin (A) or in thapsigargin-resistant DC-3F/TG2 cells (B). Medium was changed to nominally Ca2+-free medium for 30 s, followed by continuation in this Ca2+-free condition for either 0, 0.5, 1.0, 1.5, or 2.0 min, as shown. The times shown represent times from the moment of return to normal (1 mM) external Ca2+. Insets show analyses of the kinetics of overshoot activation (expressed as -fold increase above starting Ca2+ levels) over a range of Ca2+ removal times from 0 to 5 min; these are means ± S.D. of measurements taken from three separate experiments. Measurements of Ca2+ were as described under "Experimental Procedures."
[View Larger Version of this Image (39K GIF file)]


Steady State Ca2+ Influx and Ca2+ Overshoots Are Both Inhibited by the Ca2+ Influx Blocker, SKF96365

From the above studies, we considered that the increased resting levels of Ca2+ within the resistant DC-3F/TG2 cells were due to continual influx through Ca2+ entry channels. We sought to investigate this further by analyzing pharmacological modification of Ca2+ entry channels using the imidazole derivative, SKF96365. This compound was originally shown to inhibit SOC-mediated Ca2+ entry by Merritt et al. (37). Even though the selectivity of SKF96365 is not confined to SOC activity (for example, it has effects on voltage-sensitive Ca2+ influx as well as other channel activities), its effects on blocking SOC-mediated influx have been widely described (2, 37, 38). Clearly, SKF96365 does terminate the influx of Ca2+ in DC-3F cells induced by thapsigargin-mediated Ca2+ pool depletion. As shown in Fig. 5A, the application of 100 µM SKF96365 to parent DC-3F cells treated with 2 µM thapsigargin for 30 min caused a large decrease in cytosolic Ca2+ levels, reducing the level from more than 150 µM down toward 50 µM within 100 s. Subsequent removal of external Ca2+ made little further difference, and readdition of Ca2+, even after a maximally effective 3-min delay (see Fig. 4), caused only a slight change in Ca2+. Thus, under this emptied pool condition, clearly, channels are open, and addition of SKF96365 causes a rundown of cytosolic Ca2+ to a new steady state as a result of channel closure. Once blocked, the overshoot response almost completely disappeared (Fig. 5A), indicating that very little reopening of channels occurred as a result of transient Ca2+ removal. SKF96365 applied to normal DC-3F cells with filled pools was without any effect on cytosolic Ca2+ levels (data not shown), indicating that there is no measurable effect on other channel or pump activities in these cells. In Fig. 5B, thapsigargin-treated parent DC-3F cells were treated with SKF96365 but, in this case, after Ca2+ removal. In this experiment the rate at which the new steady state is reached is significantly faster, presumably, as described above, due to more efficient pumping across the plasma membrane as a result of the decreased Ca2+ gradient. The new equilibrium level of Ca2+ (approximately 25 µM) was unaffected by addition of SKF96365, and the overshoot response following readdition of Ca2+ was completely abolished. From this, it is clear that the blocking action of SKF96365 is not restricted to channels actively conducting Ca2+ and is not dependent on the presence of external Ca2+. These experiments show that the action of SKF96365 is to effectively block SOC activity. Moreover, they suggest that both the steady state increases in Ca2+ as well as the overshoot responses are due to Ca2+ entry through the same channel. When applied to resistant DC-3F/TG2 cells (Fig. 5C), although the resting level of Ca2+ was not as high, the effect of SKF96365 in reducing Ca2+ is unmistakable and closely resembles its action on thapsigargin-treated parent cells. Again, the Ca2+ overshoot after Ca2+ readdition was completely prevented. These experiments provide strong evidence that the increased levels of Ca2+ in resistant DC-3F/TG2 cells result from continued entry of Ca2+ through SOCs.


Fig. 5. SKF96365-induced inhibition of Ca2+ influx in parent DC-3F cells and in thapsigargin-resistant DC-3F/TG2 cells. Changes in cytosolic Ca2+ measured as described under "Experimental Procedures" were followed in DC-3F cells pretreated for 30 min with 2 µM thapsigargin (A and B) or in DC-3F/TG2 cells that were not treated with thapsigargin after removal from dishes (C). At the times shown, cells were treated with 100 µM SKF96365 or nominally Ca2+-free medium.
[View Larger Version of this Image (19K GIF file)]


Ionomycin-induced Emptying of the Thapsigargin-resistant Ca2+ Pool Activates Additional Ca2+ Entry in Resistant DC-3F/TG2 Cells

An intriguing question to address was the relationship between possible different Ca2+ pools and the activation of Ca2+ entry in the resistant DC-3F/TG2 cells. Thus, whereas a thapsigargin-resistant Ca2+ pumping pool exists in these cells that is agonist-responsive, these cells also have some constitutively activated Ca2+ influx that appears to result from the continuously empty (or partly empty) state of a thapsigargin-sensitive Ca2+ pumping pool. Although agonist-mediated release of the thapsigargin-insensitive Ca2+ pool did not appear to induce significant Ca2+ influx (Fig. 2B), it was possible that agonist-induced Ca2+ release was incomplete in emptying this pool as a result of receptor desensitization, as described above; certainly, the finding that agonist-sensitive Ca2+ pools are a distinct entity from Ca2+ pools coupled to SOC activation would be curious. We therefore sought to determine whether more complete emptying of pools could be effected using the Ca2+ ionophore, ionomycin, and whether such pool emptying was coupled to Ca2+ entry. Initial experiments shown in Fig. 6 assessed the relative extent of ionomycin-induced Ca2+ pool release in parent DC-3F cells and resistant DC-3F/TG2 cells. These experiments were undertaken in the absence of external Ca2+ so that release could be measured without any Ca2+ entry. Using parent DC-3F cells, removal of Ca2+ had virtually no effect on resting Ca2+ levels (approximately 25 µM), and thapsigargin caused a large release of Ca2+, reaching a peak within 50 s and then declining back to the resting level within a further 5 min (Fig. 6A). After this emptying of thapsigargin-sensitive Ca2+ pools (and the subsequent pumping of released Ca2+ out of the cells), the application of 10 µM ionomycin caused a further release of Ca2+, accounting for approximately 20% of the Ca2+ released by thapsigargin. Although we cannot be sure all thapsigargin-sensitive Ca2+ pools were emptied, the ionomycin-induced Ca2+ release is believed to be derived predominantly from pools filled via thapsigargin-insensitive Ca2+ pumps. Indeed, the size of this pool relative to the thapsigargin-sensitive pool corresponds very well with the relative sizes of thapsigargin-sensitive and -insensitive Ca2+ pumping pools determined from 45Ca2+ flux experiments using permeabilized parent DC-3F cells (30). Thapsigargin treatment periods longer than that in Fig. 6A resulted in similar ionomycin-induced Ca2+ release from DC-3F cells; however, prolonged Ca2+-free treatment does eventually result in some rundown of all pools. In the resistant DC-3F/TG2 cells, as shown above, removal of Ca2+ caused the higher resting level of Ca2+ (approximately 70 µM) to decrease as the endogenously active Ca2+ entry was prevented. Thapsigargin addition had no effect, whereas addition of ionomycin released the whole thapsigargin-insensitive Ca2+ pumping pool (Fig. 6B). Generally the amount of Ca2+ released from this pool was less than the total Ca2+ released from pools in the parent cells, and this again is in agreement with total size of pools measured in flux experiments (30).


Fig. 6. Relative release of Ca2+ from pools in response to thapsigargin or ionomycin in parent DC-3F cells and resistant DC-3F/TG2 cells in the absence of extracellular Ca2+. Changes to nominally Ca2+-free medium and additions of either 2 µM thapsigargin or 10 µM ionomycin were made at the times shown. Measurements of cytosolic Ca2+ using either parent DC-3F cells (A) or thapsigargin-resistant DC-3F/TG2 cells (B) were undertaken as described under "Experimental Procedures."
[View Larger Version of this Image (24K GIF file)]


Pool release experiments were then conducted in the presence of external Ca2+ to assess whether ionophore-induced Ca2+ release from thapsigargin-resistant pools was linked to Ca2+ entry. In these experiments, a limiting concentration of ionomycin (2 µM) was used which, although still effectively releasing Ca2+ from intracellular stores, did not increase the permeability of the plasma membrane to Ca2+. Parent DC-3F cells were treated with thapsigargin for 30 min to completely release thapsigargin-sensitive Ca2+ pools. As shown in Fig. 7A, after steady state Ca2+ influx had been achieved as a result of opening of SOCs (see Fig. 1), the addition of ionomycin caused only a slight change in cytosolic Ca2+, suggesting that emptying of a residual Ca2+ pool was not inducing any significant Ca2+ influx. This result was in contrast to the actions of ionomycin on resistant DC-3F/TG2 cells shown in Fig. 7B. In this case, ionomycin induced a rapid peak of cytosolic Ca2+ due to pool release, followed by a sustained increase in Ca2+ lasting for many minutes (trace a). This sustained increase was likely due to activation of SOCs since the application of SKF96365 at any time (for example, 200 s in trace b) caused an immediate and rapid decrease in cytosolic Ca2+ as a result of inhibiting entry through SOCs. If SKF96365 was added together with ionomycin (trace c) then only the ionophore-induced peak of Ca2+ release was observed. As shown in the latter two traces, the level of Ca2+ reached after addition of SKF96365 was below the starting level, which was consistent with the action of SKF96365 in blocking both the constitutively activated Ca2+ entry (see Fig. 5C) and the additional Ca2+ entry resulting from ionophore-induced emptying of the thapsigargin-insensitive Ca2+ pool.


Fig. 7. Ca2+ entry in response to ionophore-induced Ca2+ pool depletion and blockade by SKF96365 parent DC-3F cells and in resistant DC-3F/TG2 cells. Measurements of cytosolic Ca2+ were undertaken using either DC-3F cells pretreated with 2 µM thapsigargin for 30 min before use (A) or DC-3F/TG2 cells (B), as described under "Experimental Procedures." In A, 2 µM ionomycin was added after 25 s as shown by the arrow. In B, additions were: 2 µM ionomycin alone at 25 s (trace a), 2 µM ionomycin at 25 s followed by 100 µM SKF96365 at 200 s (trace b), or 2 µM ionomycin together with 100 µM SKF96365 at 25 s (trace c).
[View Larger Version of this Image (26K GIF file)]


Concluding Remarks

Previous 45Ca2+ flux studies (30) using permeabilized cells of the parent DC-3F fibroblast line revealed that InsP3-sensitive Ca2+ pools are exclusively filled via an intracellular Ca2+ pump with high sensitivity to thapsigargin (IC50 200 pM). A small amount of highly thapsigargin-resistant pumping activity (IC50 4 µM) is detectable in these cells but is not involved in filling InsP3-releasable pools. However, similar flux studies with the thapsigargin-resistant DC-3F/TG2 cell line revealed that an identical thapsigargin-resistant Ca2+ pump is the only apparent operational pump in these cells and pumps Ca2+ into InsP3-sensitive Ca2+ pools (30). Thus, it appears that the adaptive change that permits these cells to continuously grow in the presence of 2 µM thapsigargin (29) is the coexpression of InsP3 receptors and thapsigargin-resistant pumps within a single pool (30). The present studies reveal that in intact resistant DC-3F/TG2 cells, this same pool is capable of generating agonist-responsive cytosolic Ca2+ signals. Moreover, it is most likely that the same pumping activity serves to sustain the normal intraluminal Ca2+ levels within the ER that appear necessary to maintain the many essential functions of ER (4, 7, 9-15) and normal cell growth (7, 16-19).

The resting level of Ca2+ in the resistant DC-3F/TG2 cells is higher and appears to reflect a significant amount of constitutively activated SOC-mediated Ca2+ influx. This view is confirmed by the effects of brief removal of external Ca2+, which causes a characteristic transient reactivation of the Ca2+ influx channels; in contrast, in parent DC-3F cells, transient removal of Ca2+ has no effect and SOCs are clearly closed under resting conditions. From these experiments, SOCs appear to exist in three different states: either closed, maximally open, or in a partially deactivated state. Only closed SOCs are present in the parent DC-3F cells under resting conditions. The maximally open state of SOCs appears to be a transient event and occurs only after complete emptying of at least one Ca2+ pool; the maximally open state can be transiently reactivated in cells after brief removal and readdition of external Ca2+. Apart from this transient maximally open state, SOCs return to a partially deactivated state in which a small amount of Ca2+ enters and the level of cytosolic Ca2+ is modestly raised. Most likely, the deactivation of Ca2+ channels occurs as a result of the interaction of Ca2+ with a site that negatively controls SOCs and is responsive to levels of Ca2+ significantly above resting levels. When Ca2+ is removed from the outside, decreased entry of Ca2+ and efficient clearing of cytosolic Ca2+ by the plasma membrane Ca2+ pump result in Ca2+ being rapidly lost from the vicinity of the influx channel and hence dissociating from a site that inhibits Ca2+ influx through SOCs. Clearly, there is ample precedent for such a negative controlling action of Ca2+ on SOC activity (2, 28, 33); indeed, evidence from Zweifach and Lewis (35) indicates that a Ca2+-dependent deactivation site is located close, perhaps within 3-4 nm of the cytosolic mouth of the entry channel. In addition, Putney and co-workers (36) revealed that in NIH 3T3 cells an additional slower mechanism of control over SOCs occurs as a result of Ca2+ acting at an extracellular site; at present, our results do not really distinguish between these possibilities.

Whereas the resting level of cytosolic Ca2+ in resistant DC-3F/TG2 cells is higher than in parent DC-3F cells, it is less than half of the level that persists within parent DC-3F cells following treatment with thapsigargin. As described earlier, thapsigargin treatment of parent DC-3F cells induces an irreversible cytotoxic state that may be a consequence of the persistently elevated cytosolic Ca2+ level (30). In contrast, other cells such as DDT1MF-2 smooth muscle cells have more complete deactivation of influx and, following thapsigargin-induced pool emptying, enter a growth-arrested state from which recovery can be induced (17-19). Whereas the resistant DC-3F/TG2 cells do have some persistent Ca2+ entry likely due to an empty thapsigargin-sensitive pool, the level of resting Ca2+ (70 nM) may be below the cytotoxic threshold; thus, in these cells, either there are less channels opened by emptying of the pool or the deactivation of the open channels is more efficient. The results in Figs. 3 and 4 do not really distinguish between these possibilities.

The additional Ca2+ influx resulting from ionophore-induced emptying of the thapsigargin-resistant pool in resistant DC-3F/TG2 cells is an intriguing observation. Our previous evidence indicated that the two distinct intracellular Ca2+ pump activities expressed in parent DC-3F cells are clearly located in distinct pools, with only the thapsigargin-sensitive Ca2+ pump functioning to accumulate Ca2+ within InsP3-sensitive pools (30). Although only thapsigargin-resistant pumping can be observed in DC-3F/TG2 cells, the constitutive influx of Ca2+ suggests that a thapsigargin-sensitive pool does exist in these cells but, obviously, in a persistently and irreversibly empty state as a result of the 2 µM thapsigargin present in the culture medium. Even though we do not know the molecular identity of the thapsigargin-resistant pump, Western analysis reveals the presence of at least the same or even increased levels of SERCA pump protein in the DC-3F/TG2 cells (29). We suggest that in the resistant DC-3F/TG2 cells both types of Ca2+ pumping pool exist and that Ca2+ influx is activated by the empty or partially empty state of either pool. Although the two components of Ca2+ influx are additive and both are blocked by SKF96365, at this stage we are unable to definitively conclude that the same single entry channel is activated by emptying of the different pools. Indeed, whereas the evidence is strong for two distinct pools within the parent DC-3F cells, the evidence for two separate pools in the resistant DC-3F/TG2 cells is less compelling; hence we cannot rule out the possibility that the two components of influx could represent different degrees of emptying of a single pool. In the parent DC-3F cells, emptying of the separate thapsigargin-resistant Ca2+ pool with ionophore does not result in any significant influx of Ca2+. Our previous studies indicated that in resistant DC-3F/TG2 cells, the InsP3 receptor becomes redirected to be expressed within the thapsigargin-insensitive Ca2+ pumping pool (30). If this is the case and assuming the equivalent separate Ca2+ pools exist in both DC-3F and DC-3F/TG2 cells, then it appears that the ability of the pool to signal Ca2+ entry is conferred upon the pool by the presence of InsP3 receptors and not merely by emptying of Ca2+. This presents an interesting scenario wherein both pool emptying and the presence of InsP3 receptors are required for SOC activation. Such a scheme may be in keeping with earlier (39) and more current hypotheses (2) that suggest InsP3 receptors are closely associated with the Ca2+ influx machinery.


FOOTNOTES

*   This work was supported by National Institutes of Health Grant HL55426, by National Science Foundation Grant MCB 9307746, and by a grant-in-aid from the American Heart Association, Maryland Affiliate.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Dagger    Present address: Growth Regulation Laboratory, Imperial Cancer Research Fund, Lincoln's Inn Fields, London WC2A 3PX, UK.
§   Present address: Dept. of Pharmacology, Cambridge University, Tennis Court Rd., Cambridge CB2 1QJ, UK.
   To whom correspondence should be addressed: Dept. of Biological Chemistry, University of Maryland School of Medicine, 108 North Greene St., Baltimore, MD 21201. Tel.: 410-706-2593; Fax: 410-706-6676; E-mail: dgill{at}umabnet.ab.umd.edu.
1   The abbreviations used are: InsP3, inositol 1,4,5-trisphosphate; ER, endoplasmic reticulum; fura-2/AM, fura-2 acetoxymethylester; SERCA, sarcoplasmic/endoplasmic reticulum Ca2+-ATPase; SOC, store-operated channel; LPA, oleoyl lysophosphatidic acid.
2   Whereas discrete channel proteins have yet to be identified, the convention SOC was adopted in May 1995 at the International Conference on Receptor-regulated Calcium Influx, Pacific Grove, CA, to provide a general description of Ca2+ store-activated influx channels variously described as capacitative Ca2+ entry, depletion-activated channels, and ICRAC (Ca2+ release-activated current).

Acknowledgments

We wish to thank Michele Vitolo for expert technical assistance and Dr. Amparo Alfonso and Carmen Ufret-Vincenty for technical advice and discussions.


REFERENCES

  1. Putney, J. W., Jr., Bird, G., and St, J. (1994) Trends Endocrinol. Metab. 5, 256-260
  2. Berridge, M. J. (1995) Biochem. J. 312, 1-11 [Medline] [Order article via Infotrieve]
  3. Rossier, M. F., and Putney, J. W., Jr. (1991) Trends Neurosci. 14, 310-314 [Medline] [Order article via Infotrieve]
  4. Gill, D. L., Ghosh, T. K., Bian, J., Short, A. D., Waldron, R. T., and Rybak, S. L. (1992) Adv. Second Messenger Phosphoprotein Res. 26, 265-308 [Medline] [Order article via Infotrieve]
  5. Berridge, M. J. (1993) Nature 361, 315-325 [CrossRef][Medline] [Order article via Infotrieve]
  6. Clapham, D. E. (1995) Cell 80, 259-268 [Medline] [Order article via Infotrieve]
  7. Gill, D. L., Waldron, R. T., Rys-Sikora, K. E., Ufret-Vincenty, C. A., Graber, M. N., Favre, C. J., and Alfonso, A. (1996) Biosci. Rep. 26, 1-19
  8. Short, A. D., Klein, M. G., Schneider, M. F., and Gill, D. L. (1993) J. Biol. Chem. 268, 25887-25893 [Abstract/Free Full Text]
  9. Sambrook, J. F. (1990) Cell 61, 197-199 [Medline] [Order article via Infotrieve]
  10. Kuznetsov, G., Brostrom, M. A., and Brostrom, C. O. (1992) J. Biol. Chem. 267, 3932-3939 [Abstract/Free Full Text]
  11. Lodish, H. F., Kong, N., and Wikström, L. (1992) J. Biol. Chem. 267, 12753-12760 [Abstract/Free Full Text]
  12. Prostko, C. R., Dholakia, J. N., Brostrom, M. A., and Brostrom, C. O. (1995) J. Biol. Chem. 270, 6211-6215 [Abstract/Free Full Text]
  13. Srivastava, S. P., Davies, M. V., and Kaufman, R. J. (1995) J. Biol. Chem. 270, 16619-16624 [Abstract/Free Full Text]
  14. Milner, R. E., Famulski, K. S., and Michalak, M. (1992) Mol. Cell. Biochem. 112, 1-13 [Medline] [Order article via Infotrieve]
  15. Nigam, S. J., Goldberg, A. L., Ho, S., Rohde, M. F., Bush, K. T., and Sherman, M. Yu. (1994) J. Biol. Chem. 269, 1744-1749 [Abstract/Free Full Text]
  16. Ghosh, T. K., Bian, J., Short, A. D., Rybak, S. L., and Gill, D. L. (1991) J. Biol. Chem. 266, 24690-24697 [Abstract/Free Full Text]
  17. Short, A. D., Bian, J., Ghosh, T. K., Waldron, R. T., Rybak, S. L., and Gill, D. L. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 4986-4990 [Abstract]
  18. Waldron, R. T., Short, A. D., Meadows, J. J., Ghosh, T. K., and Gill, D. L. (1994) J. Biol. Chem. 269, 11927-11933 [Abstract/Free Full Text]
  19. Graber, M. N., Alfonso, A., and Gill, D. L. (1996) J. Biol. Chem. 271, 883-888 [Abstract/Free Full Text]
  20. Burk, S. E., Lytton, J., MacLennan, D. H., and Schull, G. E. (1989) J. Biol. Chem. 264, 18561-18568 [Abstract/Free Full Text]
  21. Lytton, J., Westlin, M., and Hanley, M. R. (1991) J. Biol. Chem. 266, 17067-17071 [Abstract/Free Full Text]
  22. Maclennan, D. H., Toyofuku, T., and Lytton, J. (1992) Ann. N. Y. Acad. Sci. 671, 1-10 [Medline] [Order article via Infotrieve]
  23. Thastrup, O., Cullen, P. J., Drøbak, B. K., Hanley, M. R., and Dawson, A. P. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 2466-2470 [Abstract]
  24. Bian, J., Ghosh, T. K., Wang, J. C., and Gill, D. L. (1991) J. Biol. Chem. 266, 8801-8806 [Abstract/Free Full Text]
  25. Sagara, Y., and Inesi, G. (1991) J. Biol. Chem. 266, 13503-13506 [Abstract/Free Full Text]
  26. Davidson, G. A., and Varhol, R. J. (1995) J. Biol. Chem. 270, 11731-11734 [Abstract/Free Full Text]
  27. Furuya, Y., Lundmo, P., Short, A. D., Gill, D. L., and Isaacs, J. T. (1994) Cancer Res. 54, 6167-6175 [Abstract]
  28. Ufret-Vincenty, C. A., Short, A. D., Alfonso, A., and Gill, D. L. (1995) J. Biol. Chem. 270, 26790-26793 [Abstract/Free Full Text]
  29. Gutheil, J. C., Hart, S. R., Belani, C. P., Melera, P. W., and Hussain, A. (1994) J. Biol. Chem. 269, 7976-7981 [Abstract/Free Full Text]
  30. Waldron, R. T., Short, A. D., and Gill, D. L. (1995) J. Biol. Chem. 270, 11955-11961 [Abstract/Free Full Text]
  31. Grynkiewicz, G., Poenie, M., and Tsien, R. Y. (1985) J. Biol. Chem. 260, 3440-3450 [Abstract]
  32. Moolenaar, W. (1995) J. Biol. Chem. 270, 12949-12952 [Free Full Text]
  33. Missiaen, L., De Smedt, H., Pary, J. B., Oike, M., and Casteels, R. (1994) J. Biol. Chem. 269, 5817-5823 [Abstract/Free Full Text]
  34. Yu, X., and Inesi, G. (1995) J. Biol. Chem. 270, 4361-4367 [Abstract/Free Full Text]
  35. Zweifach, A., and Lewis, R. S. (1995) J. Biol. Chem. 270, 14445-14451 [Abstract/Free Full Text]
  36. Louzau, C. A., Ribeiro, C. M. P., Bird, G., St, J., and Putney, J. W., Jr. (1996) J. Biol. Chem. 271, 14807-14813 [Abstract/Free Full Text]
  37. Merritt, J. E., Armstrong, W. P., Benham, C. D., Hallam, T. J., Jacob, R., Jaxa-Chamiec, A., Leigh, B. K., McCarthy, S. A., Moores, K. E., and Rink, T. J. (1990) Biochem. J. 271, 515-522 [Medline] [Order article via Infotrieve]
  38. Demaurex, N., Monod, A., Lew, D. P., and Krause, K.-H. (1994) Biochem. J. 297, 595-601 [Medline] [Order article via Infotrieve]
  39. Irvine, R. F. (1990) FEBS Lett. 263, 5-9 [CrossRef][Medline] [Order article via Infotrieve]

©1997 by The American Society for Biochemistry and Molecular Biology, Inc.