Introduction

Ionomycin (IM) is widely used in experiments to increase intracellular Ca2+ concentration ([Ca2+]i), and is reported to increase intracellular [Ca2+]i via two Ca2+ entry pathways. One is electrogenic Ca2+/H+ exchange (1 Ca2+ for 1 H+), which is mediated by IM in phospholipid membranes and vesicles [1–4] and in PC12 cells and lymphocytes [5, 6]. The other is the store-operated Ca2+ channels (SOCs), which is activated in many cell types by depleting Ca2+ from the intracellular stores. In salivary cells [7–15], IM is used to activate SOCs, similarly to thapsigargin (TG) and acetylcholine (ACh).

However, it remains uncertain which concentration of IM activates Ca2+/H+ exchange in rat submandibular acinar cells. We examined Ca2+ entry via SOCs in rat submandibular acinar cells, using many agonists including IM. In the course of the experiments, a high concentration of IM, for example 5 μM, increased intracellular Ca2+ concentration ([Ca2+]i) to an extremely high value, which is not inhibited by 1 μM Gd3+, and induced depolarization (Yoshida, unpublished observation). However, a low concentration of IM, for example 1 μM, increased [Ca2+]i moderately, which was inhibited by 1 μM Gd3+. These findings suggest that 5 μM IM may mediate the Ca2+/H+ exchange. A previous report also showed that 6 μM IM induced electrogenic Ca2+ entry into lymphocytes, suggesting that 6 μM IM mediates the Ca2+/H+ exchange [6].

To clarify whether 5 μM IM exchanges 1 Ca2+ for 1 H+, we measured pHi in rat submandibular acinar cells stimulated with 5 μM IM. Unexpectedly, 5 μM IM induced a transient pHi decrease: a rapid pHi decrease followed by a pHi increase. Moreover, the transient pHi decrease induced by 5 μM IM was HCO3 −-dependent, because no transient pHi decrease was noted and only a gradual pHi was observed in the presence of a carbonic anhydrase inhibitor (200 μM acetazolamide). Thus, a gradual pHi increase seems to be induced by the Ca2+/H+ exchange. There are two possible means of inducing a transient pHi decrease. One is activation of ion transporters, for example H+ extrusion, HCO3 − secretion, and HCO3 − uptake. The other is a chemical characteristic of IM [5]. In this study, we examined why IM induced a transient pHi decrease in rat submandibular acinar cells.

Materials and methods

Solutions and chemicals

The control solution contained (mM): NaCl 121, KCl 4.5, MgCl2 1, CaCl2 1.5, NaHCO3 25, NaHEPES 5, HHEPES 5, and glucose 5. The control solutions were aerated with 95% O2 and 5% CO2. To prepare Ca2+-free solution, CaCl2 was excluded from the solution (nominally Ca2+-free). We did not use EGTA to chelate Ca2+, because EGTA also chelates Gd3+. There was no difference between [Ca2+]i increases after re-introduction of Ca2+ using Ca2+-free solution with or without EGTA. HCO3 −-free solution contained (mM): NaCl 146, KCl 4.5, MgCl2 1, CaCl2 1.5, NaHEPES 5, HHEPES 5, and glucose 5, and was aerated with 100% O2. The pHs of the solutions were adjusted to 6.8, 7.4, or 8.0 by the addition of 1 M HCl or 1 M NaOH, as appropriate. Ionomycin, TG, methyl isobutyl amiloride (MIA) and diisothiocyanatostilbene disulfonate (DIDS), bovine serum albumin (BSA), and collagenase were obtained from Wako Pure Chemical Industries (Osaka, Japan). The reagents were dissolved in dimethyl sulfoxide (DMSO) and stored for stock solutions. Stock solutions were diluted to their final concentrations just before the experiments. DMSO concentrations never exceeded 0.1%. DMSO (0.1%) has no effects on [Ca2+]i and pHi. All the experiments were performed at 37°C.

Cell preparation and fluorescence measurements

Male rats (Slc:Wistar/ST; Japan SLC, Hamamatsu, Japan) weighing 150–200 g were purchased and fed a standard pellet diet and water. The rats were anesthetized by intraperitoneal injection of pentobarbital sodium (60–70 mg kg−1). The submandibular glands were removed from the rats for cell isolation, and the rats were then killed by the additional injection of pentobarbital sodium (100 mg kg−1). The experiments were approved by the Animal Research Committee of Osaka Medical College, and the rats were cared for according to the guidelines of this committee. The procedures for the cell preparations have already been described in detail [13]. The submandibular glands were minced with a collagenase solution (0.1%) and then incubated (37°C) for 15 min. The digested tissue was filtered through a Nylon mesh (150 μm2). The filtrate was centrifuged at 300 rpm (20g) for 1 min, washed three times, with centrifugation, and suspended in the control solution containing 2% BSA.

The isolated cells were loaded with fura 2-AM (2.5 μM; Dojindo, Kumamoto, Japan) or BCECF-AM (2.5 μM; Dojindo, Kumamoto, Japan) for 25 min at room temperature (23°C) and washed with the control solution. They were then mounted on a coverslip precoated with neutralized Cell-Tak (Becton–Dickinson Labware, Bedford, MA, USA) to allow cells to firmly adhere to the coverslip. The coverslip with cells was set in a perfusion chamber, which was mounted on the stage of an inverted microscope (TE2000; Nikon, Tokyo, Japan) connected to an image-analysis system (AQUA COSMOS; Hamamatsu Photonics, Hamamatsu, Japan). The capacity of the chamber was approximately 100 μl and the perfusion rate was 500 μl min−1. The fura 2 (340:380 nm) and BCECF (490:450 nm) fluorescence ratios were calculated and stored in the image-analysis system [13–15]. The calibration curve for pHi was obtained from the F 490/F 450 values of the BCECF-loaded cells, which were perfused with solution II containing nigericin (10 μg/ml). The pH of solution II was set at 6.6, 7.0, 7.2, 7.4, 7.8, or 8.0. Solution II contained (mM): KCl 130, NaCl 20, MgSO4 1, and HEPES 10. One experiment was performed using 6–12 coverslips from 3 to 4 animals. A typical response obtained from 7 to 9 cells in a coverslip is shown in the figures. To compare values among experiments, mean values obtained from 3 to 6 coverslips were used.

Activation of Ca2+ entry

Before addition of IM, cells were perfused with a Ca2+-free solution for 5 min. IM was then added and cells were incubated for an additional 10 min. The Ca2+-free solution was suddenly replaced with the control solution (1.5 mM Ca2+) maintaining an IM concentration (re-introduction of Ca2+). The re-introduction of Ca2+ enables us to observe only a Ca2+ entry, because intracellular stores have already been depleted [11–13].

Calculation of time constant (τ)

To compare pHi increase among different experiments, we calculated the time constant (τ) of changes in [H+]i. The changes in [H+]i are expressed by Eq. 1:

$$ \left( {[{\text{H}}^{ + } ]_{{{\text{i}}t}} - [{\text{H}}^{ + } ]_{{{\text{i}}\infty }} } \right) = ([{\text{H}}^{ + } ]_{\text{i0}} - [{\text{H}}^{ + } ]_{{{\text{i}}\infty }} ) \cdot \exp ( - t /\tau ) $$
(1)

where “t” is the time, and the subscripts “0” and “∞” indicate [H+]i at t = 0 and t = ∞, respectively.

Equation 1 can be rewritten as Eq. 2:

$$ {\text{Ln}}\left( {\left( {[{\text{H}}^{ + } ]_{\text{it}} - [{\text{H}}^{ + } ]_{{{\text{i}}\infty }} } \right)/\left( {[{\text{H}}^{ + } ]_{\text{i0}} - [{\text{H}}^{ + } ]_{{{\text{i}}\infty }} } \right)} \right) = - 1/\tau \cdot {\text{t}} $$
(2)

When “ln (([H+]it  − [H+]i∞)/([H+]i0 − [H+]i∞))” is plotted against t (τ plot), the slope is −1/τ. In this study we used τ as an index of the rate of pHi increase. Therefore, we compared the pHi increases following the re-introduction of Ca2+ among experiments by using τ (Table 1).

Table 1 Time constants of pHi increase following the re-introduction of Ca2+
Full size table

The statistical significance of the differences between the means was assessed using paired and unpaired Student’s t tests, as appropriate. Differences were considered significant at p < 0.05.

Results

Intracellular pH changes induced by IM in the presence and the absence of HCO3 −

Changes in intracellular pH (pHi) were measured in submandibular acinar cells using 5 and 1 μM IM (Fig. 1). In the control solution, the pHis of rat submandibular acinar cells were ~7.4 (Fig. 1).

Fig. 1
figure 1

Intracellular pH (pHi) changes induced by ionomycin (IM) in the presence of HCO3 −. a 5 μM IM in HCO3 −-containing solution. IM (5 μM) induced no pHi decrease in Ca2+-free solution. The re-introduction of Ca2+ induced a large transient pHi decrease followed by a rapid pHi increase. The plateau pHi was slightly higher than that before re-introduction of Ca2+. b τ Plot of a. The pHis for 4 min after re-introduction of Ca2+ were used for the τ plot. The τ plot clearly shows that the pHi increase following the rapid pHi decrease consisted of two phases: the first phase followed by the second phase. c 1 μM IM in HCO3 −-containing solution. The re-introduction of Ca2+ induced a large transient pHi decrease followed by a rapid pHi increase. The plateau pHi was slightly lower than that before the re-introduction of Ca2+. d τ plot of c. The pHis for 4 min after re-introduction of Ca2+ were used. The τ plot revealed that the pHi increase following the rapid pHi decrease consisted of the single component during 1 μM IM stimulation

Full size image

In the presence of HCO3 −, the switch to Ca2+-free solution did not induce any pHi change; then, further addition of 5 μM IM induced a small transient pHi decrease, the duration of which was ~10 min (Fig. 1a). Re-introduction of Ca2+ induced a rapid and transient pHi decrease followed by a gradual pHi increase. The pHis 1 and 3 min after re-introduction of Ca2+ were 6.89 ± 0.04 and 7.50 ± 0.03 (n = 6), respectively. The pHi 3 min after re-introduction of Ca2+ was higher than that before re-introduction of Ca2+ (Fig. 1a). The time constant of the pHi increase was calculated from the τ plot (Fig. 1b). The τ plot showed that the pHi increase consisted of two components; a rapid increase (the first phase within 0.5 min after the rapid pHi decrease) followed by a gradual increase (the second phase within 0.5–2.5 min after the rapid pHi decrease). The time constants of the first phase (τ 1) and the second phase (τ 2) are shown in Table 1.

Similar experiments were carried out using 1 μM IM. The switch to Ca2+-free solution and addition of 1 μM IM did not change pHi. The re-introduction of Ca2+ induced a rapid transient pHi decrease followed by a plateau. The pHis 1 and 4 min after the re-introduction of Ca2+ were 7.02 ± 0.03 and 7.32 ± 0.03 (n = 7), respectively (Fig. 1c). The final pHi during 1 μM IM stimulation was low compared with that during 5 μM IM stimulation (p < 0.05). The τ plot showed that the pHi increase consisted of one component without any gradual increase (Fig. 1d; Table 1). The value of τ 1 obtained by use of 1 μM IM was not significantly different from that obtained by use of 5 μM IM (p > 0.05) (Table 1).

The experiments were also performed in the absence of HCO3 − (Fig. 2). The switch to HCO3 −-free solution transiently increased pHi, which then plateaued. The pHis 2 and 5 min after the switch were 7.60 ± 0.04 and 7.41 ± 0.03 (n = 7), respectively. Then, the switch to the Ca2+-free solution and addition of 5 μM IM did not change pHi. The re-introduction of Ca2+ induced a rapid transient pHi decrease followed by a gradual pHi increase (Fig. 2a). The pHis 1 and 5 min after the re-introduction of Ca2+ were 7.15 ± 0.03 and 7.48 ± 0.04 (n = 7), respectively. The τ plot showed that the pHi increase still consisted of two components (Table 1) (data not shown). The τ 1 in the absence of HCO3 − was significantly greater than that in the presence of HCO3 − (p < 0.05), whereas τ 2 was not affected by HCO3 −-free solution (p > 0.05) (Table 1). Thus, the gradual pHi increase stimulated by 5 μM IM is not induced by HCO3 − entry, because there is no HCO3 − outside the cells.

Fig. 2
figure 2

Intracellular pH (pHi) changes induced by ionomycin (IM) in the absence of HCO3 −. a 5 μM IM in HCO3 −-free solution. IM (5 μM) induced no change in pHi in Ca2+-free solution. The re-introduction of Ca2+ induced a rapid pHi decrease followed by a gradual pHi increase. b 1 μM IM in HCO3 −-free solution. When control solution was switched to HCO3 −-free solution, pHi transiently increased from 7.4 to 7.7 and then plateaued (pHi 7.5). The switch to Ca2+-free solution and addition of IM (1 μM) did not change pHi. The re-introduction of Ca2+ induced a rapid pHi decrease without any pHi increase

Full size image

The experiments were also carried out using 1 μM IM (Fig. 2b). The switch to HCO3 −-free solution increased pHi transiently. Then, the switch to Ca2+-free solution and further addition of 1 μM IM induced no pHi change. The re-introduction of Ca2+ induced a rapid pHi decrease followed by a small pHi increase. The pHis 1.5 and 5 min after re-introduction of Ca2+ were 7.17 ± 0.03 and 7.25 ± 0.02 (n = 7), respectively (Fig. 2b). The τ plot showed that the first phase was still observed during 1 μM IM in the absence of HCO3 and the τ 1 was significantly greater than that in the presence of HCO3 − (p < 0.05) (Table 1). Moreover, the extent of the rapid pHi decrease was less in the absence of HCO3 − than in the presence of HCO3 −. These observations suggest that the transient pHi decrease is accelerated by the presence of HCO3 −.

Experiments were carried out using thapsigargin (TG, 4 μM). In HCO3 −-containing solution, the switch to the Ca2+-free solution and the addition of TG did not change pHi. On re-introduction of Ca2+ pHi decreased gradually and plateaued within 3 min (Fig. 3a). In HCO3 −-free solution, the switch to Ca2+-free solution and addition of TG did not change pHi. On re-introduction of Ca2+ pHi decreased gradually (Fig. 3b). Thus, TG did not induce any rapid transient pHi decrease after re-introduction of Ca2+. Thus, the transient pHi decrease induced by IM is not caused by an [Ca2+]i increase.

Fig. 3
figure 3

Intracellular pH (pHi) changes induced by thapsigargin (TG). a HCO3 − containing solution. TG (4 μM) induced no pHi decrease in Ca2+-free solution. The re-introduction of Ca2+ induced a gradual pHi decrease without any rapid pHi increase. b HCO3 −-free solution. The switch to HCO3 −-free solution, increased pHi transiently, and it then plateaued (pHi 7.5). The switch to Ca2+-free solution and addition of IM (1 μM) did not change pHi. The re-introduction of Ca2+ induced a gradual pHi decrease without any pHi increase

Full size image

Effects of acetazolamide on the transient pHi decrease

We also examined the effects of acetazolamide (100 μM), an inhibitor of carbonic anhydrase. Because the absence of HCO3 − decreased the extent of rapid transient pHi decrease and increased the value of τ 1 (Figs. 1, 2; Table 1). Experiments were carried out in the presence of HCO3 −. In Ca2+-free solution, the addition of acetazolamide and the further addition of 5 μM IM induced no pHi change. Re-introduction of Ca2+ induced only a small pHi decrease followed by a gradual pHi increase (Fig. 4a). The τ plot showed that the pHi increase consisted of one component (Fig. 4b). The time constant of the gradual pHi increase was similar to τ 2 obtained by use of 5 μM IM alone (Table 1). Experiments were carried out using 1 μM IM (Fig. 4c). In a Ca2+-free solution, addition of acetazolamide and IM (1 μM) induced no pHi change. The re-introduction of Ca2+ induced no transient pHi decrease, and then pHi decreased slightly. Thus, acetazolamide abolished the transient pHi decrease after the re-introduction of Ca2+ in IM-treated cells, suggesting that the rapid transient pHi decrease induced by IM was coupled with the reaction \( ({\text{H}}^{ + } + {\text{HCO}}_{ 3}{^{ - }} \rightleftarrows {\text{CO}}_{ 2} ) \).

Fig. 4
figure 4

Effects of 100 μM acetazolamide (ACZ, an inhibitor of carbonic anhydrase) on pHi changes induced by IM. a 5 μM IM. The addition of ACZ and the further addition of 5 μM IM have no effect on pHi in Ca2+-free solution. The re-introduction of Ca2+ induced a small pHi decrease followed by a gradual pHi increase. b τ plot of a. The pHis for 4 min after the re-introduction of Ca2+ were used. The τ plot revealed that the pHi increase after re-introduction of Ca2+ consisted of the single component with ACZ. c 1 μM IM. ACZ abolished the pHi changes induced by 1 μM IM

Full size image

Effects of Na+/H+ exchange, Na+-HCO3 − cotransport, and HCO3 − on the transient pHi decrease

To examine the effects of Na+/H+ exchange on the rapid pHi decrease followed by the rapid increase in the presence of HCO3 −, an inhibitor of the Na+/H+ exchange (MIA, 10 μM) was used (Fig. 5a). After switching to Ca2+-free solution, addition of MIA and IM induced no pHi change. Re-introduction of Ca2+ induced a large rapid pHi decrease (pHi 6.53 ± 0.03, n = 3), followed by a pHi increase (pHi 7.38 ± 0.03, 5 min after re-introduction of Ca2+). The τ plot showed that the pHi increase consisted of two components (Table 1), and the value of τ 1 is significantly greater than that obtained by use of 5 μM IM alone (p < 0.05) (Table 1). Thus, MIA enhanced the extent of rapid pHi decrease (Fig. 5a) and delayed a pHi recovery in the rapid transient pHi decrease, although MIA did not significantly increase τ 2 (p > 0.05). An increase in [Ca2+]i may activate H+ extrusion via Na+/H+ exchangers in submandibular acinar cells, the inhibition of which enhances acidification.

Fig. 5
figure 5

Effects of MIA and DIDS on pHi changes induced by 5 μM IM. a MIA (10 μM, an inhibitor of Na+/H+ exchange) has no effect on pHi. Re-introduction of Ca2+ induced a rapid pHi decrease followed by a rapid pHi increase. The pHi decrease stimulated by 5 μM IM was enhanced by MIA in submandibular acinar cells. b DIDS (200 μM, an inhibitor of Cl−/HCO3 − exchange and Na+-HCO3 − cotransport). DIDS did not inhibit the rapid transient pHi decrease. The pHi increase after the pHi decrease seems to be faster with DIDS than that without DIDS

Full size image

We examined the effects of the Cl−/HCO3 − exchange on the rapid transient pHi decrease followed by a gradual pHi increase by using an inhibitor of Cl−/HCO3 − exchange (DIDS, 200 μM) (Fig. 5b). After the switch to Ca2+-free solution, addition of DIDS and IM induced no pHi change. Re-introduction of Ca2+ induced a rapid transient pHi decrease followed by pHi recovery. The pHis 1.5 and 5 min after the re-introduction of Ca2+ were 6.84 ± 0.01 and 7.54 ± 0.01 (n = 4), respectively. The τ plot showed that the pHi increase consisted of two components (Table 1). The value of τ 1 was similar to that for 5 μM IM alone (p > 0.05), but that of τ 2 was significantly reduced (p < 0.05). A decrease in τ 2 value induced by DIDS may be caused by inhibition of Cl−/HCO3 − exchange. Inhibition of the Cl−/HCO3 − exchange by DIDS increases [HCO3 −]i which accelerates pHi increase.

DIDS also inhibits Na+-HCO3 − cotransport. If Na+-HCO3 − cotransport is active in submandibular acinar cells, inhibition of the cotransport increases the extent of rapid pHi decrease and delays pHi increase. Because DIDS did not increase the extent of rapid pHi decrease and accelerated pHi increase in this study, activity of Na+-HCO3 − cotransport seems to be low in rat submandibular acinar cells.

IM-induced Ca2+ entry

[Ca2+]i was measured in rat submandibular acinar cells using 5 and 1 μM IM (Fig. 6). Experiments were carried out in the presence of HCO3 −. Cells were first perfused with control solution, and then with Ca2+-free solution. The switch to Ca2+-free solution slightly reduced [Ca2+]i, which plateaued within 3 min (data not shown). Addition of 5 μM IM induced a transient [Ca2+]i increase in the Ca2+-free solution. The re-introduction of Ca2+ immediately increased [Ca2+]i, which plateaued within 3 min. The F 340/F 380 plateau was extremely high (Fig. 6a). The large [Ca2+]i increase following re-introduction of Ca2+ was not inhibited by 1 μM Gd3+, although 1 μM Gd3+ reduced the rate of Ca2+ entry and the sustained [Ca2+]i by ~10% (Fig. 6a).

Fig. 6
figure 6

IM-induced [Ca2+]i increases after re-introduction of Ca2+. a IM (5 μM). Experiments were carried out using HCO3 −-containing solution. Addition of 5 μM IM induced a large transient increase in fura 2 fluorescence ratio (F 340/F 380) in Ca2+-free solution. Re-introduction of Ca2+ increased F 340/F 380, which plateaued within 3 min. The plateau level was extremely high. This increase was not inhibited by 1 μM Gd3+. b and c IM (1 μM, b) and TG (4 μM, c). Addition of 1 μM IM or 4 μM TG induced a small transient increase in F 340/F 380 in Ca2+-free solution. Re-introduction of Ca2+ induced a biphasic increase in F 340/F 380, which was inhibited by 1 μM Gd3+

Full size image

Similar experiments were carried out using 1 μM IM and 4 μM TG. Addition of 1 μM IM induced a small transient increase in [Ca2+]i in Ca2+-free solution. The re-introduction of Ca2+ induced a biphasic [Ca2+]i increase, that is, a rapid transient increase followed by a sustained increase in [Ca2+]i (Fig. 6b). The biphasic [Ca2+]i increase after re-introduction of Ca2+ was inhibited by 1 μM Gd3+ (Fig. 6b). The sustained F 340/F 380 stimulated with 1 μM IM was much less than that stimulated with 5 μM IM (Fig. 6a, b). Similar results were obtained by use of 4 μM TG (Fig. 6c).

Effects of pHo on the rate of [Ca2+]i increase

Experiments were carried out using the same procedure as shown in Fig. 6. The pHos used were 6.8, 7.4, and 8.0. When 5 μM IM was used for stimulation, re-introduction of Ca2+ immediately increased and plateaued [Ca2+]i (F 340/F 380). The final plateaus of [Ca2+]i were similar at the three pHos. The rate of Ca2+ entry was calculated from the rate of F 340/F 380 increase (/min). The rate of Ca2+ entry increased with changing pHo from 6.8 to 8.0 (Fig. 7a, c). In contrast, when 1 μM IM was used for stimulation, re-introduction of Ca2+ immediately increased and then gradually reduced F 340/F 380. Increases in [Ca2+]i and the rates of Ca2+ entry were almost the same at the three pHos (Fig. 7a, c). The rates of Ca2+ entry were plotted against pHos. In cells stimulated with 5 μM IM, a pHo increase enhanced Ca2+ influx whereas in cells stimulated with 1 μM IM or 4 μM TG, it did not affect Ca2+ influx (Fig. 7c). Thus, Ca2+ influxes stimulated by 5 μM IM were because of the Ca2+/H+ exchange.

Fig. 7
figure 7

Effects of extracellular pH (pHo) on the rate of F 340/F 380 increase (Ca2+ influx) stimulated by IM. Experiments were carried out using a HCO3 −-containing solution. a 5 μM IM: increases in F 340/F 380 after re-introduction of Ca2+ at pHo of 8.0 and 6.8. Increases in F 340/F 380 were enhanced at a pHo of 8.0. b 1 μM IM. Increases in F 340/F 380 after re-introduction of Ca2+ at pHo of 8.0 and 6.8. Increases in F 340/F 380 were not affected by pHo. c Rates of the F 340/F 380 increase were plotted against pHo. Upon stimulation with 5 μM IM, the rates increased with increasing pHo. However, upon stimulation with 1 μM IM, the rates remained almost constant with increasing pHo. The rates of F 340/F 380 increase obtained in HCO3 −-free solution (data calculated from Fig. 8) are also plotted in this figure

Full size image

Ca2+ entry induced by IM in the absence of HCO3 −

[Ca2+]i was also measured using 5 and 1 μM IM in HCO3 −-free solution (Fig. 8). The switch to Ca2+-free solution slightly reduced [Ca2+]i, which plateaued within 3 min (data not shown). Addition of 5 μM IM induced a transient [Ca2+]i increase in the Ca2+-free solution. Re-introduction of Ca2+ immediately increased [Ca2+]i to an extremely high value within 3 min (Fig. 8a). The large [Ca2+]i increase with 5 μM IM was not inhibited by 1 μM Gd3+, similarly to Fig. 6a (data not shown).

Fig. 8
figure 8

IM-induced [Ca2+]i increase after re-introduction of Ca2+ in HCO3 −-free solution. The re-introduction of Ca2+ increases [Ca2+]i in rat submandibular acinar cells stimulated with 5 μM or 1 μM IM. Increases in [Ca2+]i in the absence of HCO3 − were similar to those in the presence of HCO3 −. a 5 μM IM, b 1 μM IM

Full size image

Addition of 1 μM IM induced a small transient increase in [Ca2+]i in the Ca2+-free solution. The re-introduction of Ca2+ induced a biphasic [Ca2+]i increase, that is, a rapid transient increase followed by a sustained increase in [Ca2+]i (Fig. 8b). The biphasic [Ca2+]i increase following to the re-introduction of Ca2+ was inhibited by 1 μM Gd3+, similarly to Fig. 6b (data not shown). Similar results were obtained by addition of 4 μM TG (data not shown). Thus, HCO3 −-free solution did not affect [Ca2+]i increases after re-introduction of Ca2+ during stimulation with 1 or 5 μM IM. The rates of Ca2+ entry in the absence of HCO3 − at 7.4 pHo are plotted in Fig. 7. The rates of [Ca2+]i increase in the absence of HCO3 − were similar to those in the presence of HCO3 − at pHo 7.4.

Discussion

This study demonstrated that IM increases [Ca2+]i via two pathways depending on its concentration. One is the SOCs activated by depletion of intracellular stores, which are sensitive to Gd3+ and activated by a low concentration of IM (1 μM). The other is Ca2+/H+ exchange, which is Gd3+-insensitive, accelerated by an increase of pHo, and mediated by a high concentration of IM (5 μM), [5, 6, 17].

This study also demonstrated that 5 μM IM induced a large transient pHi decrease after re-introduction of Ca2+. The pHi changes induced by 5 μM IM consisted of two phases. The first phase is a rapid transient pHi decrease, and the second phase is a gradual pHi increase. The rapid transient pHi decrease was reduced by HCO3 −-free solution and abolished by ACZ, whereas the gradual pHi increase was not affected by HCO3 −-free solution or ACZ. Thus, the rapid transient pHi decrease was HCO3 −-dependent and the gradual pHi increase was HCO3 −-independent.

IM (1 μM) induced a rapid transient pHi decrease, but not a gradual pHi increase. Because 5 μM IM, not 1 μM IM, mediates Ca2+/H+ exchange, the gradual pHi increase is induced by the Ca2+/H+ exchange. Moreover, TG, which increased [Ca2+]i, induced no rapid transient pHi decrease and no gradual pHi increase, although it gradually reduced pHi. Thus, the rapid transient pHi decrease was induced by IM, not by an [Ca2+]i increase.

On the other hand, the rapid transient pHi decrease followed by the gradual pHi increase was not abolished by MIA and DIDS, although MIA enhanced the extent of rapid pHi decrease and delayed the pHi recovery (an increase in τ 1), and DIDS accelerated the pHi recovery (a decrease in τ 2). Thus, Na+/H+ exchangers (H+ extrusion) or Cl−/HCO3 − exchangers (HCO3 − extrusion) do not cause the transient pHi decrease stimulated by IM, although they modified the pHi changes induced by IM. The pHi changes in DIDS-treated cells also suggest that the activity of Na+-HCO3 − cotransporters (HCO3 − uptake) in rat submandibular acinar cells is low, because DIDS did not enhance the rapid pHi decrease and did not delay pHi recovery. Moreover, an increase in [Ca2+]i induced by TG did not induce any rapid transient pHi decrease, especially in the absence of HCO3 −. If a [Ca2+]i increase stimulates HCO3 − secretion via anion channels, TG should reduce pHi rapidly, similarly to IM. These observations indicate that the rapid pHi decrease is not induced by HCO3 − secretion. The extent of pHi decrease may affect pHi recovery. The extent of rapid pHi decrease with DIDS was less than that with MIA. However, pHi recovery with DIDS was faster than with MIA. This may suggest that the gradients of H+ between intra-cellular and extra-cellular spaces do not directly affect H+ extrusion.

As mentioned above, the gradual pHi increase was induced by the Ca2+/H+ exchange mediated by 5 μM IM. However, the rapid transient pHi decrease induced by IM is not explained by cellular ion transporters activated by an [Ca2+]i increase. An increase in [Ca2+]i, which stimulates cellular metabolisms, increases CO2 production, resulting in a pHi decrease. However, the pHi decrease induced by TG was much slower than that induced by 1 μM IM. These results suggest that a rapid transient pHi decrease is induced by a chemical characteristic of IM.

The rapid transient pHi decrease induced by IM was HCO3 −-dependent as described above. The reactions of IM were considered as follows in the absence of both Ca2+ and HCO3 − [5, 17].

$$ [{\text{H}}_{2} - {\text{IM}}] \rightleftarrows [{\text{H}} - {\text{IM}}]^{ - } + {\text{H}}^{ + } $$
(3)
$$ [{\text{H}} - {\text{IM}}]^{ - } \rightleftarrows {\text{IM}}^{2 - } \, + \,{\text{H}}^{ + } $$
(4)

Given that the highest alkaline pk a of IM is >8 [18–20], the complex [H–IM]− should be the most abundant species at a physiological pH. These IM complexes are considered to be membrane-permeable [5].

For IM interacting with Ca2+, relatively lipophilic complexes that bear a positive charge are monoprotonated forms of the 1:1 species ([H–Ca–IM]+). Therefore, addition of Ca2+ induces reaction 5 at a physiological pH.

$$ [{\text{H}} - {\text{IM}}]^{ - } + {\text{Ca}}^{{ 2 { + }}} \rightleftarrows [{\text{H}} - {\text{Ca}} - {\text{IM}}]^{ + } $$
(5)

Thus, addition of Ca2+ (re-introduction of Ca2+) does not change pHi in the absence of HCO3 −. However, in the presence of HCO3 −, reaction 5 is coupled with reaction 6.

$$ {\text{CO}}_{ 2} \rightleftarrows {\text{HCO}}_{ 3}{^{ - }} + {\text{H}}^{ + } $$
(6)

The coupling of two reactions (5 and 6) produces an electroneutral Ca2+ complex, [[H–Ca–IM]+·HCO3 −] [20] as shown in reaction 7.

$$ [ {\text{H}} - {\text{IM]}}^{ - } + {\text{Ca}}^{{ 2 { + }}} + {\text{CO}}_{ 2} \rightleftarrows \left[ {[{\text{H}} - {\text{Ca}} - {\text{IM]}}^{ + } \cdot {\text{HCO}}_{ 3}{^{ - }} } \right] + {\text{H}}^{ + } $$
(7)

Thus, in the presence of HCO3 −, a monoprotonated IM reacts with Ca2+ and CO2 to produce an electroneutral Ca2+ complex and H+, which rapidly reduces pHi after re-introduction of Ca2+. Previous reports suggest that, in IM-induced Ca2+ transport, electroneutral and electrogenic Ca2+ transports may occur, involving [[H–Ca–IM]+·X−] and [H–Ca–IM]+ species, respectively [5, 20]. Then, H+ reacts with HCO3 − to produce CO2 (reaction 6). CO2 is removed from the cell, which results in the rapid pHi increase. Our hypothesis, in which IM induces a rapid transient pHi decrease mediated via reactions 6 and 7, was supported by the experiments using ACZ: inhibition of reaction 6 abolished the rapid transient pHi decrease.

The presence of HCO3 − has been reported to enhance a transient pHi decrease after removal of NH4 + pulse in alveolar type II cells [16]. The CO2/HCO3 − reaction also enhanced the transient pHi decrease after removal of the NH4 + pulse in ATII cells. Thus, the CO2/HCO3 − reaction (reaction 6) seems to be important role in pHi regulation in cells, especially on acute acid loading.

TG or 1 μM IM caused a transient increase in [Ca2+]i after re-introduction of Ca2+, whereas 5 μM IM caused a sustained plateau increase in [Ca2+]i. During TG stimulation, an immediate increase in [Ca2+]i after re-introduction of Ca2+ is considered to be due to Ca2+ entry but the plateau level is affected by Ca2+ extrusion pathways, most notably mitochondrial Ca2+ uptake and Ca2+ extrusion via plasma membrane Ca2+-ATPase (PMCA). Because IM does not inhibit SERCA (sarcoendoplasmic reticulum Ca2+ ATPase), the transient increase in [Ca2+]i induced by 1 μM IM seems to reflect the activation of mitochondrial Ca2+ uptake and PMCA. However, during stimulation with 5 μM IM, these clearance pathways seem to be inhibited. PMCA activities are known to be inhibited by cytosolic acidification [21, 22]. In this study, 5 μM IM was demonstrated to induce marked cytosolic acidification, which may inhibit PMCA. Moreover, at pHo 6.8, the sustained [Ca2+]i responses were similar to those at pHo 7.4 and pHo 8.0, that is, not transient. This may suggest that the sustained plateau is caused by the inhibition of PMCA by 5 μM IM-induced cytosolic acidification. However, in the presence of HCO3 −, the cytosolic acidification induced by 5 μM IM was transient. After pHi recovery, the sustained plateau [Ca2+]i was not reduced. A high concentration of IM, for example 5 μM, may induce acidification inside the mitochondria by H+ uptake coupled with Ca2+ release via the IM-mediated Ca2+/H+ exchange. This may inhibit ATP synthesis resulting in inhibition of PMCA and SERCA. IM (5 μM) seems to inhibit Ca2+ extrusion pathways. Further studies will be needed to clarify this.