Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Many hormones, neurotransmitters, and growth factors induce the hydrolysis of phosphatidylinositol-4,5-bisphosphate and thereby produce inositol-1,4,5-trisphosphate (IP₃) as an intracellular messenger (Berridge, 1993). IP₃ releases Ca²⁺ from intracellular stores by binding to the IP₃ receptor (IP₃R), a multigene family of Ca²⁺-release channels consisting of IP₃R1 (Furuichi et al., 1989), IP₃R2 (Südhof et al., 1991), and IP₃R3 (Blondel et al., 1993). This Ca²⁺ release results in the generation of complex cytoplasmic Ca²⁺ signals, including Ca²⁺ oscillations and propagating Ca²⁺ waves (Lechleiter et al., 1991).

Cytosolic Ca²⁺ has a bell-shaped effect on IP₃R1, with low concentrations stimulating the Ca²⁺ release and high concentrations inhibiting it (Iino, 1990; Bezprozvanny et al., 1991; Finch et al., 1991; Parys et al., 1992). The regulation of IP₃R2 and IP₃R3 by Ca²⁺ is, however, less well understood. IP₃-induced Ca²⁺ release in permeabilized rat basophilic leukemia cells, which predominantly express IP₃R2 (De Smedt et al., 1994), is not inactivated by cytosolic Ca²⁺ (Horne and Meyer, 1995), and the partially purified cardiac IP₃R2 also lacks the inhibition at high Ca²⁺ concentrations in single-channel recordings (Ramos-Franco et al., 1998). In contrast, the IP₃-induced Ca²⁺ release in permeabilized chicken B cells genetically modified to express only IP₃R2 was inhibited by 1 µM Ca²⁺ (Miyakawa et al., 1999). The effects of high Ca²⁺ concentrations on IP₃R3 have been studied using different techniques, and the reports are so far discrepant. The IP₃Rs in RIN-m5F insulinoma cells, which are between 60% (De Smedt et al., 1994) and 96% (Wojcikiewicz, 1995) of type 3, were not inhibited by up to 100 µM Ca²⁺ when incorporated in planar lipid bilayers (Hagar et al., 1998). In contrast, patch-clamp experiments on outer nuclear membranes of Xenopus oocytes overexpressing IP₃R3 revealed that micromolar Ca²⁺ did inhibit IP₃-induced channel activity (Mak et al., 1998a). Reports on the effects of high Ca²⁺ on IP₃R3 in permeabilized cells are also discrepant. IP₃-induced Ca²⁺ release in permeabilized 16HBE14o- cells, which predominantly express IP₃R3 (Sienaert et al., 1998), was inhibited by micromolar Ca²⁺ (Missiaen et al., 1998; Sienaert et al., 1998). In contrast, the release in permeabilized chicken B cells expressing only IP₃R3 was not inhibited by 1 µM Ca²⁺, but higher concentrations were not tested (Miyakawa et al., 1999). One possible explanation for these divergent results is that experimental conditions and/or regulatory mechanisms can interfere with the bell-shaped Ca²⁺ dependence of the IP₃-induced Ca²⁺ release [e.g., the effects of cytosolic Ca²⁺ on the IP₃R depend on the free Mg²⁺ concentration, pH, and the IP₃ and ATP concentrations (Tsukioka et al., 1994; Bootman et al., 1995; Mak et al., 1998b, 1999)]. In the present study, we focus on the effect of the Ca²⁺-binding protein calmodulin.

Calmodulin binds to IP₃R1 (Maeda et al., 1991; Yamada et al., 1995; Patel et al., 1997; Cardy and Taylor, 1998), and this interaction results in a decreased binding of IP₃ to IP₃R1 (Patel et al., 1997; Cardy and Taylor, 1998; Sipma et al., 1999). Exogenous calmodulin inhibits IP₃-induced Ca²⁺ release in permeabilized A7r5 cells (Missiaen et al., 1999), which express for 75% IP₃R1 and for 25% IP₃R3 (De Smedt et al., 1994). Calmodulin also inhibits the purified cerebellar IP₃R1 incorporated in planar lipid bilayers (Michikawa et al., 1999).

The aim of this work was to investigate the effects of calmodulin on IP₃-induced Ca²⁺ release in permeabilized 16HBE14o- human bronchial mucosal cells, which express for 93% IP₃R3, as judged from the relative levels of steady-state mRNA, and for 81% IP₃R3 as judged from experiments using isoform-specific antibodies (Sienaert et al., 1998).

We now report that calmodulin inhibited the IP₃-induced Ca²⁺ release if the free cytosolic Ca²⁺ concentration was 0.1 µM or higher. This inhibition occurred with an IC₅₀ value of 9 µM calmodulin. Calmodulin shifted the inhibitory part of the Ca²⁺-response curve of the IP₃-induced Ca²⁺ release toward lower Ca²⁺ concentrations. We conclude that IP₃R3 is inhibited by calmodulin and that the Ca²⁺ concentrations needed to inactivate IP₃R3 are decreased by the presence of calmodulin.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

⁴⁵Ca²⁺ fluxes were performed on saponin-permeabilized 16HBE14o- cells derived from human bronchial surface epithelium (Cozens et al., 1994) at 25°C as described previously (Missiaen et al., 1998). The nonmitochondrial Ca²⁺ stores were loaded for 45 min in 120 mM KCl, 30 mM imidazole-HCl (pH 6.8), 5 mM MgCl₂, 5 mM ATP, 0.44 mM EGTA, 10 mM NaN₃ and 150 nM free Ca²⁺ (23 µCi/ml) and then washed once in 1 ml of efflux medium containing 120 mM KCl, 30 mM imidazole-HCl (pH 6.8), 1 mM 1,2-bis(2-aminophenoxy)ethane-N,N,N',N'-tetraacetic acid (BAPTA), and 4 µM thapsigargin. Thapsigargin was added to block the endoplasmic-reticulum Ca²⁺ pumps during subsequent additions of Ca²⁺. The efflux medium was replaced every 2 min for 20 min. The additions of IP₃, Ca²⁺, and calmodulin are indicated in the figures. The free Ca²⁺ concentration was calculated with the CaBuf computer program using the following decimal logarithms of the association constants for ATP: H-ATP, 6.49; H-HATP, 4.11; Ca-ATP, 3.78; Ca-HATP, 1.98; Mg-ATP, 4.00; and Mg-HATP, 2.06 (Martell and Smith, 1982). The association constants for BAPTA were H-BAPTA, 6.36; H-HBAPTA, 5.47; and Ca-BAPTA, 6.97 (Tsien, 1980). At the end of the experiment, the ⁴⁵Ca²⁺ remaining in the stores was released by incubation with 1 ml of a 2% SDS solution for 30 min.

Calmodulin from bovine brain (purity >99%; Calbiochem, San Diego, CA) was made Ca²⁺-free by batch treatment with 50 mg/ml Chelex 100 (Bio-Rad Laboratories, Hercules, CA) for 1 h at 10°C. Calmodulin was dissolved as a 1 mM stock in 30 mM imidazole-HCl (pH 6.8). Control samples were treated with the same buffer.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

IP₃-Induced Ca²⁺ Release in Permeabilized 16HBE14o- Cells. The nonmitochondrial Ca²⁺ stores of permeabilized 16HBE14o- cells were first loaded to equilibrium with ⁴⁵Ca²⁺ and then incubated in efflux medium containing 1 mM BAPTA and no added Ca²⁺. Thapsigargin (4 µM) was added to the efflux medium to allow a unidirectional Ca²⁺ efflux. Figure 1A (filled circles) illustrates that a 2-min exposure to 1.5 µM IP₃ and 0.3 µM free Ca²⁺ accelerated the rate of Ca²⁺ loss. The traces were corrected for the passive Ca²⁺ efflux in an identical medium in the absence of IP₃. This concentration of IP₃ released 45 ± 4% of the Ca²⁺ released by a saturating dose of 100 µM IP₃ in the presence of 0.3 µM free Ca²⁺ (n = 3).

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Effect of calmodulin on the IP3-induced Ca2+ release in permeabilized 16HBE14o- cells. The nonmitochondrial Ca2+ stores were loaded to steady state with 45Ca2+ and then incubated in efflux medium containing 1 mM BAPTA and no added Ca2+. During the time period indicated by the horizontal bar, 1.5 µM IP3 and 0.3 µM free Ca2+ (A) or 1.5 µM IP3 alone (B) were added for 2 min in the absence () or presence () of 20 µM calmodulin. The traces were corrected for the passive Ca2+ efflux in an identical efflux medium in the absence of IP3. Ca2+ release is plotted as fractional loss (i.e., the amount of Ca2+ released in 2 min divided by the total store Ca2+ content at that time). Values are mean of four experiments. The S.E. was always less than 5%.

Effect of Calmodulin on IP₃-Induced Ca²⁺ Release. Figure 1 also illustrates the effect of 20 µM calmodulin (open symbols), added at the time of IP₃ addition, on the Ca²⁺ release induced by 1.5 µM IP₃ in the presence of 0.3 µM free Ca²⁺ (Fig. 1A) and in the absence of added Ca²⁺ (Fig. 1B). Exogenously added calmodulin inhibited the IP₃-induced Ca²⁺ release in the presence of 0.3 µM Ca²⁺ but was unable to inhibit the release in the absence of added Ca²⁺.

Inhibition of IP₃R by Calmodulin Is Dose-Dependent. The Ca²⁺ release induced by 1.5 µM IP₃ and a whole range of calmodulin concentrations in a medium containing 0.3 µM free Ca²⁺ (filled symbols) and in a medium with 1 mM BAPTA and no added Ca²⁺ (open symbols) is shown in Fig. 2. Calmodulin inhibited the IP₃R with an IC₅₀ value of 9 µM in the presence of 0.3 µM free Ca²⁺. No inhibition was observed in the absence of added Ca²⁺.

View larger version (15K): [in this window] [in a new window]   Fig. 2.   Inhibition of the IP3-induced Ca2+ release by calmodulin in permeabilized 16HBE14o- cells is dose-dependent. Ca2+ release induced by 1.5 µM IP3 in the absence () or presence () of 0.3 µM free Ca2+ was measured at the indicated calmodulin concentration. Values are mean ± S.E. for three experiments.

Effect of Calmodulin on Ca²⁺ Concentration Dependence of IP₃-Induced Ca²⁺ Release. Figure 3 illustrates how 20 µM calmodulin interfered with the activation of the IP₃R by Ca²⁺ in the presence of a constant IP₃ concentration (1.5 µM). The filled symbols illustrate the effects of Ca²⁺ on the IP₃-induced Ca²⁺ release in the absence of calmodulin. Low Ca²⁺ concentrations slightly activated the release, and high Ca²⁺ concentrations inhibited it. The open circles illustrate that a similar pattern also occurred in the presence of 20 µM calmodulin. Ca²⁺ inhibited the IP₃R with an IC₅₀ value of 0.92 µM Ca²⁺ in the absence of calmodulin and with an IC₅₀ value of 0.15 µM Ca²⁺ in its presence. The inactivation by Ca²⁺ therefore occurred at lower Ca²⁺ concentrations in the presence of calmodulin.

View larger version (15K): [in this window] [in a new window]   Fig. 3.   Effect of calmodulin on the Ca2+ concentration dependence of the IP3-induced Ca2+ release in permeabilized 16HBE14o- cells. The stores were challenged for 2 min with 1.5 µM IP3 and the indicated free Ca2+ concentration in the absence () or presence () of 20 µM calmodulin. Values are mean ± S.E. for four independent experiments.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

16HBE14o- cells express for 81 to 93% IP₃R3, as judged from experiments using isoform-specific antibodies and from the relative levels of steady-state mRNA as determined by quantitative ratio reverse transcription-polymerase chain reaction (Sienaert et al., 1998). Although a small fraction of the IP₃Rs are IP₃R1 and IP₃R2 isoforms, the properties of the IP₃-induced Ca²⁺ release in 16HBE14o- cells were very similar to those in genetically engineered B cells that exclusively express IP₃R3 (Miyakawa et al., 1999); that is, the release was less sensitive to IP₃ and much less affected by ATP than in cell types expressing predominantly IP₃R1 (Missiaen et al., 1998). The properties of the IP₃-induced Ca²⁺ release in 16HBE14o- cells can therefore be considered as representative of the characteristics of IP₃R3.

We observed that calmodulin inhibited the IP₃-induced Ca²⁺ release in 16HBE14o- cells in the presence of Ca²⁺ and that calmodulin shifted the inhibitory part of the Ca²⁺-response curve toward lower Ca²⁺ concentrations. IP₃-induced Ca²⁺ release in permeabilized RIN-m5F cells, which express between 60% (De Smedt et al., 1994) and 96% (Wojcikiewicz, 1995) of type 3 IP₃R, was also inhibited by calmodulin (Adkins et al., 2000). Binding studies have provided evidence for both Ca²⁺-dependent and -independent interactions between calmodulin and IP₃R1 (Maeda et al., 1991; Yamada et al., 1995; Patel et al., 1997; Cardy and Taylor, 1998; Adkins et al., 2000). Calmodulin interacts with at least two different binding sites, of which the functional significance has not yet been unequivocally demonstrated (Yamada et al., 1995; Sipma et al., 1999; Adkins et al., 2000). A Ca²⁺-dependent binding site is localized in the regulatory domain of IP₃R1 (Yamada et al., 1995) and could be involved in the Ca²⁺-dependent inhibition of IP₃R1 by calmodulin (Michikawa et al., 1999; Missiaen et al., 1999). This site was also identified in IP₃R2 but not in IP₃R3 (Yamada et al., 1995), possibly because its affinity is too low to be detected by affinity chromatography (Adkins et al., 2000).

The significance of the Ca²⁺-independent interaction of IP₃R1 with calmodulin is much less clear, but a role in the inhibition of IP₃-induced Ca²⁺ release was also proposed (Patel et al., 1997). Moreover, calmodulin was found to inhibit in a Ca²⁺-independent way IP₃ binding to the bacterially expressed ligand-binding domain of IP₃R1 (Sipma et al., 1999), and similar observations were made for the ligand-binding domains of IP₃R2 and IP₃R3 (Vanlingen et al., 2000). These effects may be mediated by a conserved low-affinity calmodulin-binding site identified in the N-terminal region of IP₃R1 (Adkins et al., 2000).

The inhibition of IP₃-induced Ca²⁺ release by calmodulin in cell types expressing predominantly IP₃R3, such as RIN-m5F insulinoma cells (Adkins et al., 2000) or 16HBE14o- bronchial epithelial cells (present work), could therefore indicate the interaction of calmodulin to IP₃R3 at a low-affinity binding site that could have been missed by calmodulin affinity chromatography. Alternatively, the effect of calmodulin may be indirect and mediated by a protein associated with IP₃R3 and in fact can even be the IP₃R1 or IP₃R2 subunits present with the predominant IP₃R3 as heterotetramers.

The Ca²⁺-induced inhibition of IP₃R1 in cerebellar microsomes in the absence of added calmodulin was prevented by 400 µM N-(6-aminohexyl)-5-chloro-1-naphthalenesulfonamide (W-7), a calmodulin inhibitor (Michikawa et al., 1999). Ca²⁺ also caused a significant inhibition of the IP₃R3 in the absence of added calmodulin in permeabilized 16HBE14o- cells. This could mean either that sufficiently high levels of endogenous calmodulin were still present after permeabilization or that calmodulin was not strictly necessary but only stimulated the Ca²⁺-induced inhibition of IP₃R3. It was technically impossible to discriminate between these two possibilities, because the calmodulin inhibitor W-7 (50 µM) induced an appreciable release of ⁴⁵Ca²⁺ on its own (data not shown), probably via nonspecific lipophilic interactions.

High levels (>10 µM) of calmodulin were found in brain, testis, and pituitary gland (Kakiuchi et al., 1982). Intermediate levels (5-10 µM) were found in lung, prostate, and adrenal gland, whereas low levels (<5 µM) occurred in liver, kidney, and spleen. In addition, calmodulin is compartmentalized, and its distribution changes during increases in intracellular Ca²⁺ concentration (Luby-Phelps et al., 1995). The concentration range over which calmodulin inhibited IP₃R3 (IC₅₀ = 9 µM in the presence of 0.3 µM free Ca²⁺) is therefore potentially physiologically relevant.

We conclude that IP₃R3 in human bronchial mucosal cells is inhibited by calmodulin and that the Ca²⁺ concentrations needed to inactivate IP₃R3 are decreased by the presence of calmodulin. The present data therefore confirm our previous finding that the type 3 IP₃R can be inhibited by Ca²⁺ (Missiaen et al., 1998). The present work extends these observations by showing that the Ca²⁺ concentration needed to inactivate IP₃R3 is largely dependent on the presence of calmodulin.