Functional analysis of the green fluorescent protein-tagged inositol 1,4,5-trisphosphate receptor type 3 in Ca²⁺ release and entry in DT40 B lymphocytes

Abstract

We examined the function of GFP-IP₃R3 (green fluorescent protein-tagged inositol 1,4,5-trisphosphate receptor type 3) in Ca²⁺ release and entry using a mutant DT40 cell line (IP₃R-KO) in which all three IP₃R genes had been disrupted. GFP-IP₃R3 fluorescence largely overlapped with the distribution of endoplasmic reticulum, whereas a portion of GFP-IP₃R3 apparently co-localized with the plasma membrane. The application of IP₃ to permeabilized WT (wild-type) DT40 cells induced Ca²⁺ release from internal stores. Although this did not occur in IP₃R-KO cells it was restored by expression of GFP-IP₃R3. In intact cells, application of anti-IgM, an activator of the BCR (B-cell receptor), or trypsin, a protease-activated receptor 2 agonist, did not cause any Ca²⁺ response in IP₃R-KO cells, whereas these treatments induced oscillatory or transient Ca²⁺ responses in GFP-IP₃R3-expressing IP₃R-KO cells, as well as in WT cells. In addition, BCR activation elicited Ca²⁺ entry in WT and GFP-IP₃R3-expressing IP₃R-KO cells but not in IP₃R-KO cells. This BCR-mediated Ca²⁺ entry was observed in the presence of La³⁺, which blocks capacitative Ca²⁺ entry. Thapsigargin depleted Ca²⁺ stores and led to Ca²⁺ entry in IP₃R-KO cells irrespective of GFP-IP₃R3 expression. In contrast with BCR stimulation, thapsigargin-induced Ca²⁺ entry was completely blocked by La³⁺, suggesting that the BCR-mediated Ca²⁺ entry pathway is distinct from the capacitative Ca²⁺ entry pathway. The present study demonstrates that GFP-IP₃R3 could compensate for native IP₃R in both IP₃-induced Ca²⁺ release and BCR-mediated Ca²⁺ entry.

INTRODUCTION

The IP₃R (inositol 1,4,5-trisphosphate receptor) is an intracellular IP₃-induced Ca²⁺ release channel composed of an N-terminal cytoplasmic IP₃-binding domain and a C-terminal channel domain. In mammalian cells, there are at least three IP₃R subtypes, type 1 (IP₃R1), type 2 (IP₃R2) and type 3 (IP₃R3), which exhibit different tissue distributions [1,2].

Immunocytochemical and subcellular fractionation studies have demonstrated that IP₃Rs are distributed to Ca²⁺ storage organelles such as the ER (endoplasmic reticulum) and nuclear envelope [2]. It has also been reported that IP₃Rs localize to certain areas in some cell types. For example, in polarized epithelial cells, such as pancreatic acinar cells and salivary cells, the IP₃R is mainly localized to the apical pole, which corresponds to the trigger zone from which apical-to-basal Ca²⁺ waves originate [3–5]. Thus the subcellular localization of the IP₃R is considered to play an important part in determining the spatio-temporal characteristics of Ca²⁺ signals.

In addition to its distribution in intracellular organelles, there is also evidence indicating that a portion of the IP₃R is localized to, or is in close proximity to, the plasma membrane in a variety of cell types [6–10]. These IP₃Rs are considered to participate in Ca²⁺ entry across the plasma membrane [6,8,9,11]. Moreover, the results of several recent studies suggest that IP₃Rs migrate during cell growth and maturation [12,13].

Recently, GFP (green fluorescent protein) has been used as an expression tag to study targeting and dynamics of intracellular molecules. In a previous study, we constructed an expression vector for a fusion protein of GFP and the full-length rat IP₃R3 (GFP-IP₃R3) [14]. This enabled us to visualize the distribution of IP₃Rs in living cells, although we could not determine the function of GFP-IP₃R3 since this was hampered by the existence of native IP₃Rs. In the present study, we expressed GFP-IP₃R3 in a mutant DT40 cell line (IP₃R-KO), in which all three IP₃R genes had been disrupted by a homologous recombination [15]. Our results demonstrate that GFP-IP₃R3 acts as a functional IP₃-induced Ca²⁺ release channel. In addition, we provide strong evidence that IP₃Rs are required for BCR (B-cell receptor)-mediated Ca²⁺ entry in DT40 cells. Moreover, GFP-IP₃R3 is also effective at compensating for native IP₃R in BCR-mediated Ca²⁺ entry.

MATERIALS AND METHODS

Materials and media

Anti-chicken IgM (supernatant, M-4 clone) was from Southern Biotechnology Associates (Birmingham, AL, U.S.A.). Polyclonal anti-GFP was from Medical & Biological Laboratories (Nagoya, Japan). Peroxidase-conjugated goat anti-rabbit antibody was purchased from Pierce (Rockford, IL, U.S.A.). Rabbit polyclonal antibodies, termed ABsI, ABsII and ABsIII, were raised against the C-termini of human IP₃R1, IP₃R2 and IP₃R3 respectively and were affinity-purified and shown to be type-specific [10].

Fura 2/AM (fura 2 acetoxymethyl ester) and mag-fura 2/AM were from Molecular Probes (Eugene, OR, U.S.A.). Poly(L-lysine) was from Sigma (St. Louis, MO, U.S.A.). Cellmatrix I-C was from Nitta Gelatin (Osaka, Japan). Tosyl-L-phenylalanylchloromethylketone (‘TPCK’) was obtained from Roche (Basel, Switzerland). Soya bean trypsin inhibitor was from Sigma. PMSF, pepstatin A, leupeptin and dithiothreitol were from Wako Pure Chemicals (Osaka, Japan). Tris/glycine SDS sample buffer and 3–8% NuPAGE Tris/acetate gels were from Novel Experimental Technology (San Diego, CA, U.S.A.). pEGFP-C3 plasmid (where EGFP stands for enhanced GFP) was purchased from Clontech Laboratories (Palo Alto, CA, U.S.A.).

Modified Hanks' balanced salt solution buffered with Hepes (HBSS-H) contained 137 mM NaCl, 5.4 mM KCl, 1.3 mM CaCl₂, 0.41 mM MgSO₄, 0.49 mM MgCl₂, 0.34 mM Na₂HPO₄, 0.44 mM NaH₂PO₄, 5.5 mM glucose and 20 mM Hepes/NaOH (pH 7.4). Nominally, Ca²⁺-free HBSS-H was identical in composition except for the omission of added CaCl₂. Na₂HPO₄ and NaH₂PO₄ were omitted in phosphate anion-free HBSS-H.

ICM (intracellular-like medium) contained 125 mM KCl, 19 mM NaCl, 10 mM Hepes (pH 7.3 with KOH), 330 μM CaCl₂ and 1 mM O,O′-bis (2-aminoethyl) ethyleneglycol-N,N,N′-N′-tetraacetic acid (free Ca²⁺ concentration was 50 nM).

The homogenization buffer contained 93.5 mM trisodium citrate, 6.2 mM citrate, 200 μM PMSF, 10 μM pepstatin A, 100 μM tosyl-L-phenylalanylchloromethylketone, 1 mM dithiothreitol, 10 μM leupeptin and 0.1 mg/ml soya bean trypsin inhibitor.

Cell culture and transfections

The DT40 chicken B-cell line and triple IP₃R-deficient cell line were cultured in RPMI 1640 (Sigma), supplemented with 10% (v/v) fetal bovine serum (Trace, Noble Park, Vic., Australia), 1% chicken serum (Gibco BRL, Rockville, MD, U.S.A.), 4 mM glutamine (Gibco BRL), 50 μM 2-mercaptoethanol (Nakarai Chemicals, Kyoto, Japan), 100 units/ml penicillin and 100 μg/ml streptomycin (Gibco BRL).

Transient transfections were performed using LIPOFECTAMINE™ 2000 reagent (Invitrogen, Carlsbad, CA, U.S.A.) with 1.6 μg/ml of plasmid according to the manufacturer's instructions.

Stable transfections of HSY-EA1 cells with the GFP-IP₃R3 plasmid vector (pCB-EGFP:IP₃R3) [14] were established using LIPOFECTAMINE™ 2000 reagent. G418 (Gibco BRL) selection (0.4 mg/ml) was initiated 24 h after the transfection. After 1 week, cells were expanded for an additional week without G418. Colonies of GFP-IP₃R3-expressing cells were screened by fluorescence microscopy. Stably transfected HSY-EA1 cells of GFP-IP₃R3 were cultured in Dulbecco's modified Eagle's medium Nutrient Mixture F-12 HAM (Sigma) supplemented with 10% (v/v) newborn calf serum (Gibco BRL), 100 units/ml penicillin and 100 μg/ml streptomycin. The A431 human carcinoma cell line was cultured in Dulbecco's modified Eagle's medium (Sigma) with high glucose (25 mM), supplemented with 10% fetal bovine serum, 4 mM glutamine, 1 mM sodium pyruvate, 100 units/ml penicillin and 100 μg/ml streptomycin.

Imaging of [Ca²⁺]_i (cytosolic Ca²⁺ concentration) and [Ca²⁺]_L (Ca²⁺ concentrations within Ca²⁺ stores)

For monitoring of [Ca²⁺]_i, cells were attached to a small recording chamber consisting of a 10 mm ring and 22 mm glass coverslips coated with 0.5 mg/ml poly(L-lysine) and Cellmatrix (diluted 1/10). Attached cells were incubated with culture medium for 1 h at 37 °C and loaded with 2 μM fura 2/AM in HBSS-H containing 1% BSA for 30 min at room temperature (22±3 °C). The fura 2-loaded cells were washed with HBSS-H and rested for at least 15 min before Ca²⁺ measurements. Cells were alternately excited at 340 and 380 nm, with emission signals being recorded at 500–530 nm. Fluorescence images of GFP were captured at an excitation wavelength of 490 nm and emission wavelength of 500–530 nm using the Argus-HiSCA imaging system (Hamamatsu photonics, Shizuoka, Japan) attached to an inverted fluorescence microscope, equipped with a Nikon Fluor 40× objective.

For [Ca²⁺]_L monitoring, cells were loaded with mag-fura 2 and permeabilized, essentially as described previously [16]. Briefly, cells were incubated with 8 μM mag-fura 2/AM in HBSS-H containing 1% BSA for 45 min at 37 °C in the dark. The mag-fura 2-loaded cells were washed with BSA-free HBSS-H and attached to Cell-Tak (BD Biosciences, Bedford, MA, U.S.A.) coated glass coverslips at the bottom of the recording chambers. Attached cells were washed with ICM, followed by exposure to ICM containing 100 μg/ml (w/v) saponin (ICN, Cleveland, OH, U.S.A.) for 1–2 min. The permeabilized cells were washed with ICM and then incubated in ICM containing 3 mM ATP and 1.4 mM MgCl₂ for at least 5 min. Fluorescent images of mag-fura 2 were acquired with dual excitation at 343 and 380 nm and an emission wavelength of 500–530 nm using the Argus-HiSCA imaging system equipped with a Nikon UV-Fluor ×100 objective. All these experiments were performed at room temperature.

Evaluation of IP₃R activity

Mag-fura 2-loaded cells were permeabilized and incubated in ICM containing 3 mM ATP and 1.4 mM MgCl₂. The cells were then washed with Ca²⁺-releasing medium, consisting of ICM with 497 μM CaCl₂ (free Ca²⁺ concentration was 100 nM), 5 mM ATP and no MgCl₂. After exposure to various concentrations of IP₃, 10 μM IP₃ was added to deplete Ca²⁺ stores at the end of the experiment. The time course of Ca²⁺ release was monitored by following the change in the fluorescence ratio of mag-fura 2, and the time taken for the fluorescence ratio to decrease by half the value at the depleted level was noted. The average rate of fluorescence change was used as an index of IP₃R activity.

Confocal microscopy

The employed multi-photon microscope system included a femtosecond, pulsed Ti:sapphire laser (Mai Tai from Spectra-Physics, Mountain View, CA, U.S.A.) and a confocal microscope (Radiance 2100; BioRad, Hemel Hempstead, Herts., U.K.) with a four-line argon-ion laser and a Green HeNe laser. The pulsed laser is optically pumped by a 5 W green diode laser and has an output wavelength tuning range from 780 to 920 nm, which provides the source for a two-photon excitation. A Nikon Plan Apo×60 H objective (NA=1.4) was used for imaging.

The GFP-IP₃R3-expressing cells were attached to poly(L-lysine)/Cellmatrix-coated chambers and were incubated with culture medium for 1 h at 37 °C. Cells were then stained with 100 nM ER-Tracker Blue-White DPX (Molecular Probes) or 50 μM TR-DHPE (Texas Red-1,2-dihexadecanoyl-sn-glycero-3-phosphoethanolamine; Molecular Probes) in HBSS-H containing 1% BSA for 10–50 min at 37 °C.

GFP-IP₃R3 and TR-DHPE fluorescence were visualized using dual-excitation beams in a line sequential mode with the argon-ion laser at 488 nm (GFP-IP₃R3) and the Green HeNe laser at 542 nm (TR-DHPE) in combination with a 560DCLP dichroic mirror and two emission filters, 500LB and 570LP. ER-Tracker was excited by an 800 nm pulse laser and the emission signal was obtained using a 625SP blocking filter.

Preparation of protein extracts

DT40 cells grown in 100 mm culture dishes (Becton Dickinson, Franklin Lakes, NJ, U.S.A.) were washed with HBSS-H and were resuspended in 5 ml ice-cold homogenization buffer. These cells were disrupted with a Polytron homogenizer (2×20 s) and were centrifuged at 1500 g for 5 min. The supernatant was centrifuged (40000 g for 20 min at 4 °C), and the resulting microsomal pellet was resuspended in homogenization buffer and stored at −80 °C until use. Preparation of the membrane fraction from stably transfected HSY-EA1 GFP-IP₃R3 cells was performed as described previously [14].

Western blotting

Samples were subjected to electrophoresis on 3–8% NuPAGE Tris/acetate gels and transferred on to nitrocellulose membranes. For detection of IP₃Rs, transferred membranes were blocked with Block Ace (Dainippon Pharmaceuticals, Osaka, Japan) containing 10% (v/v) goat serum, and were then incubated for 2 h with either ABsI, ABsII or ABsIII in 10% Block Ace. For detection of GFP, transferred membranes were blocked with PBS containing 5% (w/v) skimmed milk, 10% goat serum and 0.05% Tween 20, and then incubated for 1 h with PBS containing 5% skimmed milk, 0.05% Tween 20 and anti-GFP polyclonal antibody. These membranes were incubated for 1 h with peroxidase-conjugated goat anti-rabbit antibody in the 10% Brock Ace containing 0.04% Tween 20. Immunoreactive bands were visualized using the SuperSignal West Dura substrate (Pierce) and the intensity of chemiluminescence was measured using a Bio/Chemi-Luminescence imaging system (Light Capture, Atto, Tokyo, Japan).

RESULTS

GFP-IP₃R3 expression in IP₃R-KO DT40 cells

Figure 1 shows the expression of native type3 IP₃R and GFP-IP₃R3 in DT40 cells. In HSY-EA1 cells stably transfected with GFP-IP₃R3 (lane 4), anti-IP₃R3 recognized native IP₃R3 at approx. 260 kDa and GFP-IP₃R3 at approx. 275 kDa. Anti-GFP recognized only the approx. 275 kDa band, confirming that this upper band represents GFP-IP₃R3 (lane 4). WT (wild-type) DT40 cells (lane 1) expressed IP₃R3, whereas no signal was detected in IP₃R-KO cells (lane 2). After transient transfection of IP₃R-KO cells with the GFP-IP₃R3 plasmid, GFP-IP₃R3 expression was demonstrated with both anti-IP₃R3 and anti-GFP (lane 3). Expression of IP₃R1 and IP₃R2 was also observed in WT DT40 cells, but not in IP₃R-KO cells, confirming that no IP₃R subtypes were expressed in IP₃R-KO cells (results not shown).

Figure 1. Expression of native IP₃R3 and GFP-IP₃R3.

Microsomal fractions from WT DT40 cells (lane 1), IP₃R-KO cells (lane 2) and pCB-EGFP:IP₃R3-transfected IP₃R-KO cells (lane 3) (15 μg of protein/lane) and the membrane fraction from pCB-EGFP:IP₃R3 stably transfected HSY-EA1 cells (lane 4) (5 μg of protein/lane) were analysed by immunoblotting with ABsIII (upper panel) or anti-GFP (lower panel).

Subcellular localization of GFP-IP₃R3 in DT40 cells

The subcellular distribution of GFP-IP₃R3 was analysed by confocal microscopy (Figure 2). When GFP-IP₃R3 was expressed in IP₃R-KO cells, its fluorescence largely overlapped with the distribution of the ER (Figure 2C). However, a portion of GFP-IP₃R3 appeared to be distinct from the ER distribution and apparently overlapped with the plasma membrane (Figures 2F and 2G). Thus we anticipate that GFP-IP₃R3 serves as a Ca²⁺ channel not only on the ER but also on the plasma membrane.

Figure 2. Subcellular localization of the GFP-IP₃R3 in IP₃R-KO DT40 cells.

pCB-EGFP:IP₃R3-transfected IP₃R-KO cells were stained with ER-Tracker or TR-DHPE, and the fluorescence of GFP-IP₃R3 (A, D), ER-Tracker (B) and TR-DHPE (E) was visualized with a multi-photon microscope. (C) A merged view of (A and B). (F) A merged view of (D and E). (G) A higher magnification of the area indicated in (F). Scale bar, 10 μm.

Luminal Ca²⁺ monitoring

To determine whether GFP-IP₃R3 serves as a functional Ca²⁺ channel on internal Ca²⁺ stores, we monitored IP₃-induced changes in [Ca²⁺]_L in permeabilized cells. As we have described previously [16], mag-fura 2 was compartmentalized within Ca²⁺ stores in saponin-permeabilized cells, and its fluorescence ratio (343/380 nm) rapidly decreased following the application of 1 μM IP₃, reflecting the IP₃-induced decrease in [Ca²⁺]_L.

This decreased ratio recovered within 5 min of IP₃ removal, reflecting the re-uptake of Ca²⁺ by sarcoplasmic/ER Ca²⁺-ATPase. IP₃-dependent changes in fluorescence ratio were observed in most WT DT40 cells (21 of 27 cells), but not in IP₃R-KO cells (n=49), confirming the lack of an IP₃-mediated Ca²⁺ release pathway in IP₃R-KO cells (Figures 3A and 3B). The expression of GFP-IP₃R3 restored the IP₃-induced Ca²⁺ release in IP₃R-KO cells (10 of 15 cells, Figure 3B). These results clearly indicate that GFP-IP₃R3 acts as an IP₃-sensitive Ca²⁺ channel.

Figure 3. IP₃-induced Ca²⁺ release in saponin-permeabilized cells.

WT DT40 (A) and GFP-IP₃R3-expressing IP₃R-KO DT40 (B) cells were loaded with mag-fura 2 and were permeabilized with saponin. Cells were then stimulated with IP₃ (1 μM) in ICM containing ATP and Mg²⁺. (A) (a) Fluorescence image of mag-fura 2; (b–d) pseudocolour images of the mag-fura 2 ratio obtained at time points indicated in (e); (e) time-dependent changes as relative (%) fluorescence ratio in the cells depicted in the fluorescence image (a). (B) (a) Merged fluorescence image of GFP-IP₃R3 and mag-fura 2; (b–d) pseudocolour images of the mag-fura 2 ratio obtained at time points indicated in (e); (e) time-dependent changes as relative (%) fluorescence ratio in the cells depicted in the fluorescence image (a). The GFP-IP₃R3-expressing cell is indicated with an arrow. The horizontal bars at the top of the Figure indicate the presence of IP₃ in ICM. The ratio is presented as relative fluorescence intensity (F₃₄₃/F₃₈₀). Scale bar, 10 μm. (C) IP₃-concentration dependence of Ca²⁺ release for GFP-IP₃R3 and WT IP₃R3. Stable GFP-IP₃R3-expressing cells (a) and A431 cells (b) were permeabilized and exposed to various concentrations of IP₃ in ICM containing 100 nM free Ca²⁺, 5 mM ATP and no MgCl₂. (a, b) Time taken for the fluorescence ratio to decrease by half the value at the depleted store level were plotted as a function of IP₃ concentration. (c) Normalized Ca²⁺ release rates at various IP₃ concentrations were compared between GFP-IP₃R3-expressing cells and A431 cells. Results are means±S.E.M.

We then examined the concentration–response relationship for IP₃ on Ca²⁺ release in permeabilized GFP-IP₃R3-expressing cells, and compared it with that in A431 cells, which predominantly expressed IP₃R3 [10]. The time taken for a 50% decrease in the fluorescence ratio became shorter as the IP₃ concentration increased [Figures 3C(a) and 3C(b)]. The concentration–response curve for the estimated IP₃R activity of GFP-IP₃R3 was similar to that of WT IP₃R3 [Figure 3C(c)].

GFP-IP₃R3 function in intact cells

We next examined the function of GFP-IP₃R3 in intact cells. The DT40 cell has an αIgM isotype BCR, which, through activation of PLC (phospholipase C)-γ, mediates an increase in [Ca²⁺]_i [17,18]. Figure 4(A) shows typical responses of WT DT40 cells to activation of BCR by anti-chicken IgM (anti-IgM, 5 μg/ml). Activation of BCR induced Ca²⁺ responses in 75% of WT cells tested (n=95). The anti-IgM-induced Ca²⁺ responses observed included Ca²⁺ oscillations (48 cells) and monophasic Ca²⁺ transient (20 cells). We also found that trypsin (20 units/ml), an agonist of PAR2 (protease-activated receptor 2), induced transient or oscillatory Ca²⁺ increases in most WT DT40 cells (93 of 96 cells, Figure 4A).

Figure 4. Agonist-induced Ca²⁺ responses in DT40 cells.

WT (A) and GFP-IP₃R3-transfected IP₃R-KO (B, C) cells were stimulated with anti-IgM antibody (5 μg/ml) and trypsin (20 units/ml) in HBSS-H (A, B) or in nominally Ca²⁺-free HBSS-H (C). (A) (a) Fluorescence image of fura 2; (b–d) pseudocolour images of the fura 2 ratio obtained at time points indicated in (e); (e) time-dependent changes in the fluorescence ratio in the cells depicted in (a). (B) (a) Merged fluorescence image of GFP-IP₃R3 and fura 2; (b–d) pseudocolour images of the fura 2 ratio obtained at time points indicated in (e); (e) time-dependent changes in the fluorescence ratio in the cells depicted in (a). GFP-IP₃R3-expressing cell is indicated by an arrow. (C) (a) Merged fluorescence image of GFP-IP₃R3 and fura 2; (b–c) pseudocolour images of the fura 2 ratio obtained at time points indicated in (d); (d) time-dependent changes in the fluorescence ratio in the cells depicted in (a). A GFP-IP₃R3-expressing cell is indicated by arrow. The horizontal bars at the top of the Figure indicate the presence of anti-IgM and trypsin. The ratio is presented as relative fluorescence intensity (F₃₄₀/F₃₈₀). Scale bar, 10 μm.

In 17 of the GFP-IP₃R3-expressing cells examined, anti-IgM induced oscillatory (6 cells), monophasic and transient (5 cells) or gradual (1 cell) increases in [Ca²⁺]_i (Figure 4B). Trypsin also induced Ca²⁺ responses in most of these cells (16 of 17 cells, Figure 4B). When GFP-IP₃R3 was not expressed in IP₃R-KO cells, anti-IgM and trypsin did not induce Ca²⁺ responses (Figure 4B). Anti-IgM-induced Ca²⁺ responses in GFP-IP₃R3-expressing DT40 cells were also observed in the absence of extracellular Ca²⁺, indicating that GFP-IP₃R3 mediates Ca²⁺ release from internal stores (Figure 4C).

BCR-stimulated Ca²⁺ entry requires IP₃R

We next examined the involvement of IP₃R and/or GFP-IP₃R3 in Ca²⁺ entry. BCR-mediated Ca²⁺ responses in WT DT40 cells ceased within 10 min in the absence of extracellular Ca²⁺ (48 of 64 cells). Subsequent increases in external Ca²⁺ led to Ca²⁺ entry, and thus increased [Ca²⁺]_i in approx. 50% of cells examined (Figures 5A and 5C). Note that this increase in Ca²⁺ concentration occurred in the presence of 0.3 μM La³⁺, the store-operated channel blocker. BCR-mediated Ca²⁺ entry was also observed in the presence of 1 μM La³⁺ or 1 μM Gd³⁺ (results not shown). In the absence of anti-IgM, La³⁺-insensitive Ca²⁺ entry was observed in <10% of cells examined (24 of 258 cells, Figure 5C). Thus it appears that the activation of BCR mediates Ca²⁺ entry in a manner that is distinct from the capacitative pathway, which is activated in response to the depletion of intracellular stores. BCR-mediated La³⁺-insensitive Ca²⁺ entry was not observed in IP₃R-KO cells, and was restored in approx. 50% of GFP-IP₃R3-expressing IP₃R-KO cells (Figures 5B and 5C). Similarly, activation of BCR stimulated La³⁺-insensitive Ba²⁺ entry in GFP-IP₃R3-expressing cells (Figure 6). These results indicate that IP₃R or GFP-IP₃R3 is required for BCR-mediated cation entry into DT40 cells.

Figure 5. BCR-mediated Ca²⁺ entry in DT40 cells.

WT (A) and GFP-IP₃R3-transfected IP₃R-KO (B) cells were stimulated with anti-IgM in nominally Ca²⁺-free medium, and La³⁺ (0.3 μM) and Ca²⁺ (1.3 mM) were subsequently added. (A) (a) Fluorescence image of fura 2; (b–d) pseudocolour images of fura 2 ratio obtained at time points indicated in (e); (e) time-dependent changes in the fluorescence ratio in the cells depicted in the fluorescence image (a). (B) (a) Merged fluorescence image of GFP-IP₃R3 and fura 2; (b–d) pseudocolour images of fura 2 ratio obtained at time points indicated in (e); (e) time-dependent changes in the fluorescence ratio in the cells depicted in the fluorescence image (a). GFP-IP₃R3-expressing cell is indicated by an arrow. The horizontal bars at the top of the Figure indicate the presence of anti-IgM, La³⁺ and Ca²⁺. The ratio is presented as relative fluorescence intensity (F₃₄₀/F₃₈₀). Scale bar, 10 μm. (C) La³⁺-insensitive Ca²⁺ entry was monitored in WT-DT40 cells and GFP-IP₃R3-transfected IP₃R-KO cells. The number of cells showing La³⁺-insensitive Ca²⁺ entry in the absence (−) and presence (+) of anti-IgM was quantified and plotted as a percentage of the total number of cells. Total number of cells counted is shown above each column.

Figure 6. BCR-mediated Ba²⁺ entry.

GFP-IP₃R3-expressed IP₃R-KO (A) and IP₃R-KO (B) cells were stimulated with anti-IgM in nominally Ca²⁺-free medium and La³⁺ (0.3 μM) and Ba²⁺ (10 mM) were subsequently added. The horizontal bars at the top of the Figure indicate the presence of anti-IgM, La³⁺ and Ba²⁺. The ratio is presented as relative fluorescence intensity (F₃₄₀/F₃₈₀).

In contrast, ThG (thapsigargin) caused a transient increase in [Ca²⁺]_i and depleted Ca²⁺ stores in the absence of extracellular Ca²⁺. Subsequent addition of extracellular Ca²⁺ increased [Ca²⁺]_i through capacitative Ca²⁺ entry in WT-DT40 cells (Figure 7A). This capacitative Ca²⁺ entry was also observed in IP₃R-KO cells, irrespective of their GFP-IP₃R3 expression state, indicating that IP₃Rs or GFP-IP₃R3 are not essential for capacitative Ca²⁺ entry (Figure 7B). In contrast with BCR stimulation, ThG-induced Ca²⁺ entry was completely blocked by 0.3 μM La³⁺. These results support the idea that BCR-induced Ca²⁺ entry is not mediated by the depletion of Ca²⁺ stores.

Figure 7. Capacitative Ca²⁺ entry in DT40 cells.

WT (A) and GFP-IP₃R3-transfected IP₃R-KO (B) cells were stimulated with ThG (1 μM) in nominally Ca²⁺-free medium, and Ca²⁺ (1.3 mM) was subsequently added to the medium in the presence (d–f, h) or absence (a–c, g) of La³⁺. (A) (a, d) Fluorescence image of fura 2; (b, c, e, f) pseudocolour images of fura 2 ratio obtained at time points indicated in (g, h); (g, h) time-dependent changes in the fluorescence ratio in the cells depicted in the fluorescence image (a, d). (B) (a, d) Merged fluorescence image of GFP-IP₃R3 and fura 2; (b, c, e, f) pseudocolour images of fura 2 ratio obtained at time points indicated in (g, h); (g, h) time-dependent changes in the fluorescence ratio in the cells depicted in the fluorescence image (a, d). A GFP-IP₃R3-expressing cell is indicated by an arrow. The horizontal bars at the top of the Figure indicate the presence of ThG, La³⁺ and Ca²⁺. The ratio is presented in terms of relative fluorescence intensity (F₃₄₀/F₃₈₀). Scale bar, 10 μm.

DISCUSSION

We have demonstrated in the present study that the direct application of IP₃ to saponin-permeabilized DT40 cells caused a rapid release of Ca²⁺. This IP₃-induced Ca²⁺ release was not detected in IP₃R-KO cells, but was restored by the expression of GFP-IP₃R3. In addition, GFP-IP₃R3 showed similar IP₃ sensitivity to that of WT IP₃R3.

The function of GFP-IP₃R3 was also confirmed in intact cells. The M-4 clone of anti-chicken IgM (anti-IgM) is known to stimulate BCR and induce the production of IP₃ via phosphorylation of PLC-γ [15,19]. In addition, trypsin is known to activate PAR2, which is coupled with Gq/11 and induces the production of IP₃ via activation of PLC-β. Both these stimulatory treatments increased [Ca²⁺]_i in WT, but not in IP₃R-KO, cells. The expression of GFP-IP₃R3 restored the Ca²⁺ response in IP₃R-KO cells. Taken together, these results established that GFP-IP₃R3 serves as a functional IP₃-sensitive Ca²⁺ release channel.

Anti-IgM induced oscillatory Ca²⁺ responses in GFP-IP₃R3-expressing IP₃R-KO cells. A similar oscillatory response has been reported in IP₃R-KO DT40 cells expressing native rat IP₃R3 [20]. It is generally accepted that the oscillatory Ca²⁺ response requires biphasic feedback control by cytoplasmic Ca²⁺. The activity of IP₃R1 is potentiated at low concentrations of Ca²⁺ (up to 300 nM) and is attenuated at higher concentrations [21]. Similar biphasic control has also been reported for IP₃R3 [22], whereas previously it has been reported that IP₃R3 is not inhibited by high concentrations of cytoplasmic Ca²⁺ [23,24]. Thus the mechanisms of Ca²⁺ oscillation and the role of Ca²⁺ in regulating the function of IP₃R3 are still controversial. GFP-IP₃R3 could become a useful tool for elucidating these issues.

In addition to the well-established role of IP₃Rs as a Ca²⁺ release channel on intracellular organelles, there is also evidence suggesting the involvement of IP₃Rs in Ca²⁺ entry into the cell. This hypothesis, termed the conformational-coupling model, proposes that a conformational change in IP₃R, caused by store depletion, mediates store-operated channel opening [25]. However, the present study, as well as several other recent studies, demonstrated that capacitative Ca²⁺ entry occurs in IP₃R-KO DT40 cells [15,19,26]. These results strongly indicate that IP₃R is not required for capacitative Ca²⁺ entry, at least in DT40 cells.

In the present study, La³⁺, a blocker of capacitative Ca²⁺ entry, inhibited ThG-induced Ca²⁺ entry but did not affect BCR-mediated Ca²⁺ entry. This indicated that BCR mediates Ca²⁺ entry via mechanisms that are distinct from capacitative Ca²⁺ entry. In addition, this result is consistent with the observation that BCR mediates the activation of Ba²⁺-permeable channels [20]. It is considered that the Ba²⁺ entry pathway differs from capacitative Ca²⁺ entry in that Ba²⁺ entry is not caused by the depletion of stores. Both BCR-mediated Ca²⁺ entry and Ba²⁺ entry require IP₃Rs, since they occur in WT DT40 cells but not in IP₃R-KO cells. Moreover, GFP-IP₃R3 was capable of compensating for native IP₃R in BCR-mediated Ca²⁺ entry.

The role(s) of IP₃Rs in controlling Ca²⁺ entry is not known. It is possible that IP₃R and GFP-IP₃R3 themselves are channels for cation entry or that they regulate other channels located on the plasma membrane. In the present study, a portion of GFP-IP₃R3 fluorescence apparently overlapped with that of the plasma membrane, implying that some GFP-IP₃R3 is localized to the plasma membrane. There is also evidence that IP₃R is integrated to the plasma membrane [10]. In addition, several reports have demonstrated that functional IP₃Rs are located on the plasma membrane [11,27]. These lines of evidence suggest that IP₃Rs act as Ca²⁺ entry channels on the plasma membrane.

Several recent studies have suggested that IP₃Rs may migrate. It is suggested in Madin–Darby canine kidney cells that IP₃R1 and IP₃R3 are translocated to the membrane from the cytoplasm during cell polarization [12,13]. Furthermore, in T-lymphocyte Jurkat cells, capping of the T-cell receptor–CD3 complex induces the accumulation of IP₃R to the cap region, which leads to Ca²⁺ entry from that region [6]. Since the distribution of IP₃Rs significantly affects Ca²⁺ signalling, it is predicted that GFP-IP₃R3 will be a useful tool for elucidating the translocation of IP₃R and associated Ca²⁺ responses in living cells.