Ⅰ. INTRODUCTION

Calcium (Ca²⁺) has been established as a major second messenger within many cells, and the relationship with inositol triphosphate (IP₃) is well documented (Berridge and Irvine, 1989;Rink and Merritt, 1990). Indeed, there are very few cellular functions that are not affected directly or indirectly by Ca²⁺. A rise in the concentration of cytosolic free intracellular calcium ([Ca²⁺]_i) can trigger rapid specialized events such as secretion and muscle contraction. At the same time Ca²⁺ plays a central role in activating slower metabolic processes like glycogenolysis, steroidogenesis, and exocytosis.

Ion-selective electrodes and ion-sensitive dyes indicate that the [Ca²⁺]_i in resting (unstimulated) neurons, as in most cells, is ~10^-7 M (Blaustein, 1988). Also, as in most other types of cells, the total intracellular calcium concentration in neurons is ~10^-3 M (Blaustein et al., 1978a, b). This ratio between total cell calcium and free Ca²⁺ (the calcium buffer ratio) is ~10,000:1; it indicates that ~99.99% of the intracellular calcium is buffered and sequestered. The implication is that very small changes in free Ca²⁺ may be associated with large changes in total cell calcium; for examples, a 1nM increase in steady-state [Ca²⁺]_i may be expected to increase total cell calcium by ~1-10 μmol/l of cell water.

The Ca²⁺ buffer curve (the relation between free Ca²⁺ and total cell calcium) plays an important role in cell physiology. In many (perhaps most) types of cells, including neurons (Llano et al., 1991), the Ca²⁺ needed to activate numerous physiological processes is derived, at least in part, from intracellular stores in the endoplasmic reticulum (ER). Stored Ca²⁺ is released by IP₃ (Nahorski, 1988) and at least in some types of cells, by Ca-induced Ca²⁺ release (Beukelmana and Wier, 1988). For any given stimulus, the amount of release may be directly related to the amount of Ca²⁺ in the ER stores (Hashimoto et al., 1986). Furthermore, temporal and spatial changes in [Ca²⁺]_i during cell activation are likely to be governed, in part, by the relative saturation of the intracellular Ca²⁺ buffers and sequestration sites.

The use of video-imaging techniques has revealed that stimulus-induced Ca²⁺ signals can be highly organized within cells. Some cells respond to electrical stimuli that directly promote Ca²⁺ influx (e.g., the squid presynaptic neuron), whereas other, nonexcitable, cells respond only to IP₃-mobilizing stimuli. Video-imaging of Fura-2 has revealed localized initiation sites of Ca²⁺ signals induced by IP₃-mobilizing stimuli in many smaller cells (Cheek et al., 1989;Neylon et al., 1990;Cheek, 1991).

Regulatory volume decreases after cell swelling involves the increase in K⁺ and/or Cl^- fluxes through the cell’s membrane in a variety of cell types. Possible mechanisms by which the increase in cell volume is sensed and transduced involve the dilution of intracellular messengers or macromolecules or a direct or indirect mechanical activation of ion channels by changes in tension of the cell membrane or the cytoskeleton (Lang et al., 1998). For example, lowering the osmolarity of the bath solution will cause water to enter the cell, consequently leading to cell swelling. These swelling produces membrane stretch and activates a volume-sensitive Cl^- conductance in a variety of cell types (Nilius et al., 1996;Strange et al., 1996) including human osteoblasts and a human osteoblast-like cell line (Steinert and Grissmer, 1997).

It was the aim of this study to provide a basis for the relationship of the G-protein and Ca2+ signaling on the influx of Ca²⁺ in osteoblast-like cells in response to hypotonic (50% Na⁺ solution with 4 mM CaCl₂) external solution. In this paper, we investigated human osteoblast-like cells in order to find a cation influx. We investigated first the [Ca²⁺]_i. Simultaneously, we measured the time course in [Ca²⁺]_i of osteoblast-like cells in response to a hypotonic (50% Na⁺ solution with 4 mM CaCl₂) external solution. Second, we examined the effects of Cd²⁺ (300 μM) on the influx of Ca²⁺ in human osteoblast-like cells in response to a hypotonic (50% Na⁺ solution with 4 mM CaCl₂) external solution. Third, we investigated the effects of incubation with cholera toxin (CTX) and pertussis toxin (PTX) for 6h (250 ng/ml) on the influx of Ca²⁺ in osteoblast-like cells in response to a hypotonic (50% Na⁺ solution with 4 mM CaCl₂) external solution.

Ⅱ. MATERIAL and METHODS

1. Cells

Osteoblast-like hFOB 1.19 cells were obtained from American Type Culture Collection (ATCC). The cell permeant acetoxymethyl (AM) esters and 0.02% pluronic acid purchased from Molecular Probes (Eugene, OR, USA). CTX and PTX were purchased from Sigma (St. Louis, MO), dissolved in mammalian Na⁺ solution as stock solution (10 μM).

2. Cell culture

The cells were grown in 75-cm² culture flasks and then subcultured onto glass cover slips (No. 1.5, Chance Propper, UK) 2 days prior to investigation using 1% trypsin for 5 min at 37°C. The cells were grown in minimum essential medium (2 mM Ca, 4 mM K) at 37°C in 4% CO₂.

3. Solutions

All experiments were done at room temperature (21-25°C). The cells under investigation were normally bathed in mammalian Na⁺ solution containing (mM): 160 NaCl, 4.5 KCl, 2 CaCl₂, 1 MgCl₂, 10 Hepes; adjusted to pH 7.4 with NaOH. The osmolarity was adjusted with either sucrose or mannose to 310 mOsm. Cell swelling was induced by perfusing the bath with dilutions of the Na⁺ solutions to 50% while setting CaCl₂ to 4 mM. The rise in [Ca²⁺]_i induced by these two different solutions, however, was not different. The resulting osmorality for bath solutions was 190 mOsm. Hypotonic Ca²⁺ -free Na⁺ solution was prepared by diluting the Ca²⁺ -free Na⁺ solution to 50%.

4. [Ca²⁺]_i – measurements

The [Ca²⁺]_i–measurements were similarly carried out as described previously (Lewis and Cahalan, 1989;Verheugen et al., 1997). Briefly, the cells were incubated in Dulbecco-MEM containing 4 μM fura-2 AM and 0.02% pluronic acid for 20 min at 37°C. Then the cells were washed three times with mammalian Na⁺ solution. The AM ester served to protect the polar groups of the dye, thus making the substance electrically neutral and so lipophilic (Fontana and Blaustein, 1993). [Ca²⁺]_i was estimated according to the following equation [Ca²⁺]_i = K_D·S_f·(R-R_min)/(R_max)-R (Grynkiewicz et al., 1985), where KD is the dissociation constant of fura-2 (assumed to be 350 nM) (Negulescu and Machen, 1990), S_f is a device dependent scale factor and R is the ratio of the two measured fluorescence intensities 350 nm/380 nm. R_min and R_max were determined in vitro using solutions containing zero Ca²⁺ and 39.8 μM Ca²⁺, respectively (Fura-2 calcium imaging calibration kit, Molecular Probes, Eugene, OR, USA).

5. G protein linkage of calcium flux

Toxins were used to assess the involvement of G proteins. A series of cells were examined which had been preincubated with PTX and CTX at a level of 250 ng/ml for 6h at 37°C prior to observation.

Ⅲ. RESULTS

[Ca²⁺]_i of a human osteoblast-like cell line was determined following loading 4 mM fura-2 AM and 0.02% pluronic acid for 20 min at 37°C. The intact human osteoblast-like cells’ [Ca²⁺]_i in normally bathed in mammalian Na⁺ solution was 260 nM (21-25°C, pH 7.4). The basic observation is illustrated in Fig. 1 which shows the time course of averaged changes in [Ca²⁺]_i of human osteoblast-like cells before and during application of a hypotonic Na⁺ external solution (50% Na⁺ solution with 4 mM CaCl₂). Decreasing the tonicity of the extracellular to 50% leads to an increase in [Ca²⁺]_i from 26 nM up 569 nM. This increase in [Ca²⁺]_i was mainly due to an influx of extracellular Ca²⁺. To find out whether the [Ca²⁺]_i transient was due to Ca²⁺ entering the cell from the outside or Ca²⁺ release from internal stores we preincubated the cells for 10 min in Ca²⁺-free Na⁺ solution and applied then hypotonic Ca²⁺-free Na⁺ solution. After changing to hypotonic Ca²⁺-free Na⁺ solution, [Ca²⁺]_i increased to about 15 nM. Then [Ca²⁺]_i decreased again within 150 sec to 200 sec to the initial level (Fig. 1). From the comparison of the changes in the [Ca²⁺]_i signal in the presence and absence of external Ca²⁺ we conclude that most of the change in [Ca²⁺]_i induced by the hypotonic solution comes from the outside, implying a Ca²⁺ influx pathway in this osteoblast- like cell line.

The influx of Ca²⁺ of the cells in the hypotonic Na⁺ external solution was blocked by 300 μM Cd²⁺, a specific blocker of Ca²⁺ channels, demonstrating an extracellular source of Ca²⁺ (Fig. 2). Preincubation with CTX (250 ng/ml for 6 h) prolonged the elevation of Ca²⁺ induced by a hypotonic Na⁺ external solution (Fig. 3). In contrast, PTX (250 ng/ml for 6 h) completely eliminated the Ca²⁺ response to a hypotonic Na⁺ external solution (Fig. 4). Cells maintained in solutions free of Ca²⁺ demonstrated no change in [Ca²⁺]_i.

Ⅳ. DISCUSSION

We have characterized an increase in [Ca²⁺]_i in response to cell swelling by application of hypotonic solutions. Such an increase in [Ca²⁺]_i in response to cell swelling has also been reported in various cell types including isolated nerve terminals (Mongin et al., 1997), lacrimal acinar cells (Speake et al., 1998), neuronal cell lines (Altamirano et al., 1998), cardiomyocytes (Taouil et al., 1998), multicellular prostate cancer spheroids (Sauer et al., 1998) and thymocytes (Ross and Cahalan, 1995). In human osteoblast [Ca²⁺]_i reached its maximum after 120 sec and was regulated back within 510 sec. Removal of external Ca²⁺ diminished the Ca²⁺ transient about 90% indicating that the main source of the Ca²⁺ seems to come from the extracellular solution and only a minority comes from intracellular stores, as it has been shown for GH3 cells (Chen et al., 1996).

Previously it has been shown that 200-300 nM [Ca²⁺]_i was sufficient to activate K(Ca) channels of the IK type in T lymphocytes (Grissmer et al., 1993). Therefore, we would have concluded that the loss of osmolytes under [Ca²⁺]_i free conditions would be mediated by K+ current flowing through IK channels, however, our experiments with PTX and regulatory volume decrease excluded such a mechanism.

Regulatory volume decrease was reduced when MaxiK channels were blocked by PTX. This indicates an important role of these channels in regulatory volume decrease of human osteoblasts although apparently [Ca²⁺]_i did not reach levels to activated those channels under isotonic conditions. One possibility to account for the activation of MaxiK channels in the absence of a sufficiently high enough [Ca²⁺]_i would be to assume that those channels might be either activated through stretch or that stretch influenced the sensitivity to activation by [Ca²⁺]_i. Application of a hypotonic solution will lead to an influx of H₂O and therefore to cell swelling. This will lead to the activation of a volume- sensitive Cl^- conductance described earlier (Steinert and Grissmer, 1997), to and increase in [Ca²⁺]_i and to the activation of other volume regulatory mechanisms. The increase in [Ca²⁺]_i or cell swelling or a combination there of could activate the MaxiK channel. This activation will lead to an efflux of K⁺. Therefore, the cell will lose these solutes and this in turn will cause H₂O efflux leading to a reduction in cell volume.

One main possibility is a mechanosensitive Ca²⁺ channel, a Ca²⁺ coupled to mechanosensor via an intermediary such as a G protein, a Ca²⁺ channel activated by an intracellular messenger generated by signal transduction pathway coupled to the mechanosensor. Our data clearly show involvement of G proteins by the disruption of signaling with PTX and by the prolongation of the signal by CTX. Action of PTX argues strongly that cell swelling induced Ca²⁺ signals require activation of heterotrimeric G proteins of i subtype. Several G proteins of this family have been implicated in channel activation and regulation in other cell types. The data from CTX and PTX preincubation suggest another pathway involving G proteins. No mechanosensitive protein has as yet been identified although the integrins are close to what may be considered as a specific load sensor. This work, however, demonstrates the first linkage of a manner similar to seven transmembrane spanning receptor (G protein) activation as by ligands. Data with CTX potentiation are interesting in that they show the mechanoresponse is regulated by cAMP/protein kinase A pathway. Therefore, stretch sensitivity may be further regulated by hormone- stimulated adenyl cyclase.

The precise mechanism of the cellular activation in our report is unclear but in view of the speed of activation of the influx of Ca²⁺ it is quite feasible that a three or four-element membrane protein activation sequence exists inducing a coupled ion channel, i.e., a mechanism integral to the cell membrane (Brown, 1991;Breitwieser, 1991). Additional information is required to be precise as to the mechanism of this activation but a possible means of activation is by deformation of the extracellular matrix followed by integrin deformation, the latter being closely related physically with the relevant ion channels. Further activation of the signal appears to occur as a consequence of G protein involvement and obviously this mechanism requires further examination.