Hypo-osmotic Shock of Tobacco Cells Stimulates Ca2+Fluxes Deriving First from External and then Internal Ca2+Stores*

Hypo-osmotic shock of aequorin-transformed tobacco cells induces a biphasic cytosolic Ca2+influx. Because both phases of Ca2+ entry are readily blocked by Ca2+ channel inhibitors, we conclude that the Ca2+ transients are mediated by Ca2+ channels. Evidence that the first but not second Ca2+ transient derives from external Ca2+ stores is that the first but not second influx is (i) eliminated by membrane-impermeable Ca2+ chelators, (ii) enlarged by supplementation of the medium with excess Ca2+, and (iii) reduced by the addition of competitive cations such as Mg2+ and Mn2+. Furthermore, entry of 45Ca during osmotic shock is prevented by inhibitors of the first but not second phase of Ca2+ entry. Evidence that the second wave of Ca2+ influx stems from release of intracellular Ca2+ is based on the above data plus observations that probable modulators of intracellular Ca2+ channels selectively block this phase of Ca2+ influx. Finally, a mechanism of communication between the two Ca2+ release pathways has become apparent, since perturbations that elevate or reduce the first Ca2+ transient lead to a compensating diminution/elevation of the second and vice versa. These data thus suggest that osmotic shock leads to the sequential opening of extracellular followed by intracellular Ca2+ stores and that these Ca2+ release pathways are internally compensated.


Hypo-osmotic Shock of Tobacco Cells Stimulates Ca 2؉ Fluxes Deriving First from External and then Internal Ca 2؉ Stores*
(Received for publication, May 4, 1997, and in revised form, August 7, 1998) Stephen G. Cessna, Sreeganga Chandra ‡, and Philip S. Low § From the Department of Chemistry, Purdue University, West Lafayette, Indiana 47907-1393 Hypo-osmotic shock of aequorin-transformed tobacco cells induces a biphasic cytosolic Ca 2؉ influx. Because both phases of Ca 2؉ entry are readily blocked by Ca 2؉ channel inhibitors, we conclude that the Ca 2؉ transients are mediated by Ca 2؉ channels. Evidence that the first but not second Ca 2؉ transient derives from external Ca 2؉ stores is that the first but not second influx is (i) eliminated by membrane-impermeable Ca 2؉ chelators, (ii) enlarged by supplementation of the medium with excess Ca 2؉ , and (iii) reduced by the addition of competitive cations such as Mg 2؉ and Mn 2؉ . Furthermore, entry of 45 Ca during osmotic shock is prevented by inhibitors of the first but not second phase of Ca 2؉ entry. Evidence that the second wave of Ca 2؉ influx stems from release of intracellular Ca 2؉ is based on the above data plus observations that probable modulators of intracellular Ca 2؉ channels selectively block this phase of Ca 2؉ influx. Finally, a mechanism of communication between the two Ca 2؉ release pathways has become apparent, since perturbations that elevate or reduce the first Ca 2؉ transient lead to a compensating diminution/elevation of the second and vice versa. These data thus suggest that osmotic shock leads to the sequential opening of extracellular followed by intracellular Ca 2؉ stores and that these Ca 2؉ release pathways are internally compensated.
It can be hypothesized that plant cells maintain high Ca 2ϩ concentrations both in their extracellular milieus and in cer-tain intracellular compartments to enable rapid gating of the cation into the low Ca 2ϩ environment of the cytosol (23)(24)(25)(26)(27)(28)(29). Indeed, cytosolic Ca 2ϩ transients in hormone/cytokine-treated animal cells have been shown to derive from both internal and external Ca 2ϩ stores (27). Because Ca 2ϩ channels have been identified in both plasma and internal membranes of plant cells (6, 9, 30 -38), both Ca 2ϩ pools should be considered as possible sources for mediating plant signaling events.
We and others have observed that elicitation of the oxidative burst in cultured plant cells generally accompanies and requires a rise in cytosolic Ca 2ϩ (7)(8)(9)(10). However, it has not been established if the induction of the oxidative burst requires Ca 2ϩ influx from internal or external Ca 2ϩ stores or from both in succession or simultaneously. In the case of the mechanically or osmotically stimulated oxidative burst, the cytoplasmic Ca 2ϩ increase is distinctly biphasic. Thus, using aequorin-transformed tobacco cells to quantitate Ca 2ϩ influx into the cytoplasm of the cell (3), Ca 2ϩ peaks have been repeatedly observed ϳ15 s and 1.5 min after hypotonic stress (8,39). In addition, other groups have established that both Ca 2ϩ entry across the plasma membrane and Ca 2ϩ release from internal stores may play roles in plant cell volume/turgor regulation (14,23). Because this unusual pattern of Ca 2ϩ flux offers the opportunity to examine the Ca 2ϩ signal required for both the induction of the oxidative burst and the regulation of cellular turgor pressure, we have decided to identify the Ca 2ϩ stores responsible for both phases of the Ca 2ϩ influx. We report here that the first Ca 2ϩ transient arises from outside the cell, whereas the second derives from intracellular stores. We further identify inhibitors that can selectively block the first, second, or both phases of Ca 2ϩ flow, and we report observations regarding possible communication between the two pathways responsible for Ca 2ϩ influx.

EXPERIMENTAL PROCEDURES
Materials-Ruthenium red, lanthanum chloride, gadolinium chloride, and niflumate were all purchased from Sigma/Aldrich. 45 Ca in the form of 45 CaCl 2 salt was obtained from ICN (Costa Mesa, CA), and coelenterazine was from Molecular Probes (Eugene, OR). Other chemicals used were of reagent grade or better and were obtained from major chemical suppliers.
Aequorin-transformed Tobacco Cell Suspension Cultures-Aequorintransformed tobacco cell (Nicotiana plumbaginifolia) suspension cultures were established from transformed seedlings and maintained as described previously (8). Two ml of packed filtered cells were transferred to 100 ml of fresh W-38 liquid medium (40) every 7 days and maintained in suspension by continuous shaking. Cells to be used ϳ12 h later for Ca 2ϩ measurements were transferred at twice their regular cell density. Functional aequorin was reconstituted from the apoenzyme by adding 5 l of coelenterazine dissolved in ethanol (1 M final) to 3 ml of suspension-cultured cells at the time of transfer. Functional aequorin capable of reproducible Ca 2ϩ measurements was found to develop 5 to 15 h after transfer of the cells to fresh medium and the addition of coelenterazine.
Aequorin Luminescence Measurements and [Ca 2ϩ ] Quantitation-Luminescence measurements were carried out in a digital luminometer (LKB Wallac model 1250, Gaithersburg, MD), as described previously * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 (8). Briefly, 0.5 ml of coelenterazine-treated cells were transferred to the luminometer cuvette and held in suspension by mild stirring during stimulation. Luminescence data was collected 10 times/s. All inhibitors, chelators, and control solvents were added at the times indicated in the figure legends. Plant cells were hypo-osmotically shocked by diluting the suspensions 1:1 with distilled water at the times indicated in the figures. At the end of each experiment, all remaining aequorin was discharged by the addition of 0.5 mM CaCl 2 in 10% Nonidet P-40, and luminescence was continually quantitated until recordings returned to basal levels and further addition of CaCl 2 /detergent solution elicited no further response. Luminescence data were then converted by computer directly to intracellular Ca 2ϩ concentration using the equation described by Allen et al. (41), [Ca 2ϩ ] ϭ ((L/L max ) 1/3 ϩ [118(L/L max ) 1/3 ] Ϫ 1)/(7 ϫ 10 6 Ϫ [7 ϫ 10 6 (L/L max ) 1/3 ]), where L is the luminescence intensity at any time point, and L max is the integrated luminescence intensity from that point to the end of the luminescence recording. 45 Ca Influx Experiments-As above, 3 ml of suspension-cultured tobacco cells transferred at twice the density of continuously cultured cells were used for determination of 45 Ca influx as described by Atkinson et al. (10). Briefly, 0.1 Ci of 45 Ca was added to 3 ml of cells just before hypo-or iso-osmotic treatment. Inhibitors were added at the times indicated in the figure legends. 45 Ca-treated cells were then incubated with gentle shaking for 7 min before suction filtration through Whatman filter paper at 20 mm of Hg and immediate rinsing with 3 ml of normo-osmotic buffer containing 10 mM CaCl 2 , 5 mM MES, 1 and 160 mM sucrose (pH 5.6) to displace any externally bound 45 Ca. Filtered and rinsed cells were then resuspended in 3 ml of rinse buffer and shaken for 20 min before suction filtering and rinsing again with 3 ml of rinse buffer. Finally, the cells were scraped into pre-weighed scintillation vials containing 3 ml of ICN Ecolite scintillation liquid. After counting the radioactivity in a Packard 1600CA liquid scintillation analyzer (Meriden, CT) and weighing the vials, the radioactivity/mg of filtered cells was calculated. Data are presented as detailed in the figure legend.

The Biphasic Ca 2ϩ Flux in Tobacco Induced by Hypo-osmotic
Shock Is the Result of the Activation of Specific Ion Channels-It was conceivable that the previously observed hypoosmotic shock generated Ca 2ϩ transients (8,39) were not due to regulated ion channels but rather to a membrane disturbance resulting in nonspecific Ca 2ϩ leaks. To evaluate this possibility, we tested the effectiveness of several broad-spectrum Ca 2ϩ channel inhibitors on the induced cytosolic Ca 2ϩ fluxes. Ruthenium red and the trivalent lanthanides gadolinium and lanthanum are known to block many types of Ca 2ϩ channels in both plasma and sarcoendoplasmic reticular membranes of animal cells (42)(43)(44), and all three have also been used for characterization of Ca 2ϩ channels and Ca 2ϩ -dependent processes in plant cells (5, 8, 30 -34, 45, 46). As can be seen in Fig.  1, each of these inhibitors is effective at blocking both phases of Ca 2ϩ flow into hypo-osmotically stimulated tobacco cells. Inhibition of Ca 2ϩ influx by ruthenium red at 50 M (Fig. 1A) is consistent with the pharmacologic data of both Allen et al. (30) and Pineros and Tester (34), who observed blockade of cyclic-ADP-ribose-activated tonoplast Ca 2ϩ channels and voltagegated plasma membrane Ca 2ϩ channels, respectively, at similar concentrations. Although this effect of ruthenium red has been previously published (8), we present these data here to allow direct comparisons of the inhibitory effects of Ca 2ϩ channel blockers on these Ca 2ϩ fluxes. Lanthanum and gadolinium ions, which required a 10-min equilibration to achieve maximal channel inhibition, also displayed concentration dependences similar to those observed by others (5,9,17). At 10 mM concentration, both lanthanides inhibited greater than 80% total Ca 2ϩ influx, whereas at 1 mM they primarily suppressed influx during the first Ca 2ϩ peak (Figs. 1, B and C).
Inhibition by these agents provides little information regarding the location of channels responsible for the cytosolic Ca 2ϩ transients. Externally provided ruthenium red has been shown to inhibit plasma membrane Ca 2ϩ channels (34) and to enter cultured plant cells and interrupt internal Ca 2ϩ release (4). In addition, lanthanum ions, classically used to identify plasma 1 The abbreviation used is: MES, 4-morpholineethanesulfonic acid. membrane Ca 2ϩ channel activities, when provided in millimolar concentrations for periods greater than 1 min have also been observed to enter cells and alter internal Ca 2ϩ channels (43,44). However, the inhibition of Ca 2ϩ influx by these channel blockers does demonstrate that the osmotically induced Ca 2ϩ uptake proceeds through specific Ca 2ϩ channels and not through nonspecific ion leaks.
The First Pulse of Cytosolic Ca 2ϩ Derives from External Ca 2ϩ Pools-To begin to identify the cellular sources of the two pulses of cytosolic Ca 2ϩ , we first performed manipulations that would specifically modify the flux of external Ca 2ϩ across the plasma membrane. EGTA, a membrane-impermeable selective chelator of Ca 2ϩ ions, was added to the extracellular medium at a concentration of 1.5 mM, approximately the concentration of free Ca 2ϩ in the medium. As shown in Fig. 2A, chelation of extracellular Ca 2ϩ with EGTA exerts an inhibitory effect selectively on the first Ca 2ϩ spike generated by hypo-osmotic shock, suggesting that this Ca 2ϩ transient, but not the second, derives from an external source. Inhibition of the first peak was also seen using 1.5 mM 1,2-bis-(o-aminophenoxy)ethane-N,N,NЈ,NЈtetraacetic acid (BAPTA) in place of EGTA, indicating that the nature of the Ca 2ϩ -chelating agent is unimportant to the inhibition (data not shown). In contrast to Ca 2ϩ chelation, the addition of extra CaCl 2 to achieve a final medium concentration approximately 2.7 and 4.3 times that normally present in the cell culture was seen to elevate the amount of Ca 2ϩ entering during the first phase of the Ca 2ϩ influx (Fig. 2B). Furthermore, supplementation of the growth medium with excess Mg 2ϩ , which can either compete with Ca 2ϩ as a channel substrate or antagonize Ca 2ϩ entry as a weak channel blocker (31,34), also reduced the magnitude of the first Ca 2ϩ peak (Fig.  2C). Mn ϩ2 and Fe 2ϩ were similarly found to inhibit the first, but not alter the second, of the two Ca 2ϩ peaks (data not shown). Inhibition of the first phase of Ca 2ϩ influx was also observed after replacement of the chloride with the sulfate salts of Mg 2ϩ and Fe 2ϩ (data not shown), indicating that this initial phase of Ca 2ϩ entry into the cytosol is not controlled by the nature of the counter ion. Based on the selective suppression of the first Ca 2ϩ transient by either EGTA or nonsubstrate cations and the selective enhancement of this peak by elevated extracellular Ca 2ϩ , we propose that the first peak of Ca 2ϩ influx during osmotic shock originates from an apoplastic pool.
To further confirm the above channel assignment, measurement of 45 Ca uptake by hypo-osmotically shocked cells was also performed. Dilution of the suspension-cultured tobacco cells with water caused an increase in the uptake of 45 Ca to ϳ5ϫ the basal level measured in iso-osmotically treated cells. As expected, 10 mM La 3ϩ was seen to largely inhibit this influx (Fig.  3), confirming that the 45 Ca uptake is mediated by a Ca 2ϩ channel. More importantly, 10 mM MgCl 2 was observed to reduce the uptake by ϳ60%, a value not inconsistent with the data shown in Fig. 2C. Since the added 45 Ca was unequivocally apoplastic and since Mg ϩ2 only reduces the first Ca 2ϩ transient measured by aequorin luminescence, we conclude that this first pulse of osmotically stimulated Ca 2ϩ entry indeed derives from extracellular Ca 2ϩ .
The Second Peak of Ca 2ϩ Derives from Intracellular Ca 2ϩ Stores-Although the results shown in Fig. 2 also argue that the second peak of cytosolic Ca 2ϩ must stem from opening an internal store, we set out to confirm this hypothesis with selective modulators of internal Ca 2ϩ release. For this purpose, caffeine, a Ca 2ϩ channel regulator believed to activate Ca 2ϩ release from cyclic-ADP-ribose and ryanodine-sensitive stores in many eukaryotic systems (4, 47, 48), was added to the tobacco suspensions, and the osmotically induced Ca 2ϩ transients were again examined. As anticipated, the addition of caffeine in the absence of any other stimulus induced substantial entry of Ca 2ϩ into the cytosol of the tobacco cell (Fig. 4A). Since these caffeine-activated Ca 2ϩ transients were found to be insensitive to membrane-inpermeant Ca 2ϩ chelators and to competition with extracellular Mn 2ϩ or Mg 2ϩ (data not shown), we reason that this Ca 2ϩ signal indeed derives from the caffeine-induced emptying of an intracellular Ca 2ϩ pool.
Based on its impact in animal cells, the ability of caffeine to autologously discharge certain internal Ca 2ϩ stores should also lead to an insensitivity of the treated cells to stimuli that discharge the same stores. Therefore, we evaluated the impact of caffeine pretreatment on the subsequent ability of hypoosmotic shock to activate the second phase of cytosolic Ca 2ϩ release. As shown in Fig. 4B, prior exposure to caffeine eliminates the second Ca 2ϩ transient without affecting or perhaps even enhancing the first. This loss of sensitivity to osmotic stimulation confirms that the latter Ca 2ϩ influx is indeed derived from caffeine-sensitive internal stores. Unfortunately, other modulators of intracellular Ca 2ϩ release commonly used in mammalian systems such as ryanodine (20 M), the inositol 1,4,5-trisphosphate receptor Ca 2ϩ channel modulator TMB-8 (200 M), or the sarcoendoplasmic reticular Ca 2ϩ ATPase inhibitor thapsigargin (up to 100 M) had little or no effect on either of the osmotically activated transients, suggesting that the similarity in pharmacology of internal Ca 2ϩ channels in animals and plants may extend no further than caffeine sensitivity. Nevertheless, the fact that moderate caffeine concentrations were able to specifically inhibit expression of the second phase of Ca 2ϩ entry suggests that this phase of influx is mechanistically distinct from the first and is likely gated through an intracellular channel.
Niflumic acid, an inhibitor of anion channels in plant and animal cells, was also found to serve as a selective inhibitor of the second peak of Ca 2ϩ entry (Fig. 4B). Other groups have observed that anion channel blockers such as niflumate and anthracene-9-carboxylate readily inhibit various anion channel activities in plant cells (14, 49 -52), and they have suggested that release of Ca 2ϩ from vacuolar stores may depend on con- FIG. 3. Effect of Ca 2؉ modulators on 45 Ca 2؉ influx. Aequorintransformed tobacco cells (3 ml) were exposed to 0.1 M Ci 45 Ca according to the protocol outlined under "Experimental Procedures" and subjected to the following treatments: control, hypo-osmotic shock induced by 1:1 dilution with distilled water; La 3ϩ , 10 mM LaCl 3 was added and followed immediately by a 1:1 dilution with distilled water; Mg 2ϩ , 10 mM MgCl 2 was added and followed immediately by a 1:1 dilution with distilled water; and niflumate, 500 M niflumate dissolved in Me 2 SO was added 10 min before 45 Ca exposure and dilution with distilled water. Data are presented as relative nuclide uptake/mg of cells and corrected for basal 45 Ca binding/uptake, measured in iso-osmotically treated control cells exposed to vehicle alone (1% water or 0.1% Me 2 SO). Error bars represent mean (S.D. for three independent experiments. Iso-osmotically treated samples gave 45 Ca uptake readings of ϳ20% of the hypo-osmotically treated samples, presumably due to 45 Ca binding in the cell wall. 45 Ca uptake was greater in control cells pretreated with 0.1% Me 2 SO than in control cells pretreated with 1% water. This difference in basal 45 Ca uptake appears to be due to a transient leak of 45 Ca due to Me 2 SO at the time of its addition. Because uptake in vehicle-treated control cells was invariably subtracted, this perturbation should not influence the results shown. current anion fluxes to depolarize the membrane (14). Consistent with these interpretations, anthracene-9-carboxylate was also observed to block the second phase of Ca 2ϩ uptake at concentrations (100 -500 M) similar to those found effective in the studies mentioned above (data not shown). Importantly and in contrast to the effect of Mg 2ϩ on 45 Ca influx, the addition of 500 M niflumate to the plant cell culture failed to inhibit 45 Ca uptake upon hypo-osmotic shock (Fig. 3). Thus, since 45 Ca influx only measures the externally derived portion of the biphasic Ca 2ϩ transients and since niflumate has no effect on this 45 Ca influx, we conclude that the second Ca 2ϩ peak is generated by release of the cation from internal stores.
The Two Channels Regulating Ca 2ϩ Influx during Hypoosmotic Shock Communicate-It did not escape our attention that inhibition of the first Ca 2ϩ transient generally resulted in enlargement of the second (Fig. 2A). Only in the case of Mg 2ϩ addition was this compensating increase in the trailing Ca 2ϩ pulse not consistently observed. Conversely, enhancement of the initial phase of Ca 2ϩ influx by supplementation of the medium with extra Ca 2ϩ invariably reduced the amplitude of the latter (Fig. 2B). Furthermore, inhibition of the second phase of Ca 2ϩ uptake by niflumic acid consistently enhanced the first (Fig. 4B). Similar but somewhat reduced effects were also seen with moderate concentrations of caffeine (Fig. 4A). Taken together, these data argue that some type of communication must occur between the two phases of osmotically stimulated Ca 2ϩ -gating and that inhibition or enhancement of one phase of influx somehow leads to a compensating Ca 2ϩ flux during the other. It will be interesting to explore the pathways along which such communication travels. DISCUSSION We have provided evidence that the two phases of Ca 2ϩ entry into the cytoplasm of osmotically shocked tobacco cells sequentially involve (i) the influx of extracellular Ca 2ϩ and (ii) the release of intracellular (compartmentalized) Ca 2ϩ . Thus, prevention of external Ca 2ϩ entry by the addition of EGTA or competing cations (i.e. Mg 2ϩ , Mn 2ϩ , Fe 2ϩ ) diminished the first phase of Ca 2ϩ uptake, whereas augmentation of the extracellular Ca 2ϩ pool specifically enhanced this same wave of Ca 2ϩ uptake. Taken together with data on the blockade of the second phase of Ca 2ϩ influx by caffeine and niflumate, i.e. probable modulators of intracellular Ca 2ϩ release, a strong case for the above Ca 2ϩ -gating assignments can now be made. This contention is also bolstered by observations that Mg 2ϩ , which blocks only the initial phase of Ca 2ϩ influx, prevents externally added 45 Ca uptake, and niflumic acid, which eliminates only the second phase of Ca 2ϩ influx, does not.
We have also observed that communication may occur between the two waves of Ca 2ϩ entry into the cytoplasm of a stimulated tobacco cell. Unfortunately, there are currently few clues in the literature regarding the mechanism through which this communication might occur. Although hypo-osmotic stress has been frequently observed in eukaryotic cells to activate release of both intracellular and extracellular Ca 2ϩ stores, little data regarding cross-talk between these two Ca 2ϩ depots has yet been presented (53)(54)(55)(56)(57)(58). Furthermore, the gating and signaling pathways activated by hypo-osmotic stimulus in nonplant systems may be very different from those in tobacco, since, opposite to the order of Ca 2ϩ -gating in tobacco cells, internal release of Ca 2ϩ appears to precede influx of externally derived Ca 2ϩ in mammalian cells (55)(56)(57)(58). This order of Ca 2ϩ channel activation in animal cells is reminiscent of the phenomenon of capacitative Ca 2ϩ entry, in which intracellular release of Ca 2ϩ leads to the activation of plasma membrane Ca 2ϩ channels and the refilling of internal stores (27, 59 -61). Although the mechanisms of communication between internal and external Ca 2ϩ stores during capacitative Ca 2ϩ entry are currently an area of intense research, no consensus mechanistic theory has been put forth (61). In osmotically stressed tobacco cells, a Ca 2ϩ -sensing mechanism that would adjust the Ca 2ϩ efflux from the second (intracellular) Ca 2ϩ store depending on the quantity of Ca 2ϩ delivered during the first (extracellular) phase of Ca 2ϩ entry can be easily envisioned. The intracellular Ca 2ϩ gate would simply need to be negatively regulated by Ca 2ϩ or a Ca 2ϩ -activated enzyme (e.g. a kinase) (62,63). It will be important in the future to look for such a signaling mediator following treatment of stimulated cells with excess Ca 2ϩ or EGTA (Figs. 2, A and B).
In contrast to the feed-forward signaling mechanism hypothesized above, identification of a communication pathway that responds to the readiness state of the second phase of Ca 2ϩ entry and pre-emptively modulates the first is more difficult to imagine. Nevertheless, pretreatment of osmotically stimulated tobacco cells with moderate concentrations of caffeine or niflumic acid invariably augments the initial Ca 2ϩ transient before eliminating the latter (Fig. 4, A and B). However, Takahashi et al. (39) have presented evidence that the general serine/threonine protein kinase inhibitors K252a and staurosporine selectively inhibit the second phase of Ca 2ϩ influx without a compensating increase in the first phase. Thus, this modification of extracellular Ca 2ϩ influx by internal Ca 2ϩ transport capability may not be a general phenomenon and may depend on the type inhibitors used. There are clearly complexities in these Ca 2ϩ signaling pathways that will require considerable research to resolve. It will be interesting to explore the area of Ca 2ϩ channel communication in plant cells and to possibly identify at both the protein and DNA level the Ca 2ϩ channels and other molecules involved.
It was encouraging to observe that several inhibitors could be identified that selectively block cytosolic influx of Ca 2ϩ from either intracellular or extracellular Ca 2ϩ stores. Although lanthanides and ruthenium red were found to inhibit both pathways of Ca 2ϩ influx, the other modulators detailed above were highly selective for just one pathway. Based on these observations, it should now be possible to resolve which downstream consequences of hypo-osmotic shock rely on which phases of Ca 2ϩ entry for signal propagation. It is conceivable, for example, that pathways designed to adjust cellular volume/turgor pressure during osmotic stress (14,23,64,65) might depend on a different Ca 2ϩ signal than pathways evolved to initiate the osmotically induced oxidative burst (8,66). Indeed, preliminary studies indicate that only the second of the two phases of Ca 2ϩ entry is required to stimulate reactive oxygen biosynthesis following hypo-osmotic stimulation. 2