Analyzing the Role of the Putative Inositol 1,3,4,5-Tetrakisphosphate Receptor GAP1IP4BP in Intracellular Ca2+ Homeostasis*

Inositol 1,3,4,5-tetrakisphosphate (IP4) has been linked to a potential role in the regulation of intracellular free Ca2+ concentration ([Ca2+] i ) following cellular stimulation with agonists that activate phosphoinositide-specific phospholipase C. However, despite many studies, the function of IP4 remains unclear and indeed there is still some debate over whether it has a function at all. Here we have used various molecular approaches to address whether manipulation of the potential IP4 receptor, GAP1IP4BP, affects [Ca2+] i following cellular stimulation. Using single cell imaging, we show that the overexpression of a constitutively active and a potential dominant negative form of GAP1IP4BP appear to have no effect on Ca2+ mobilization or Ca2+ entry following stimulation of HeLa cells with histamine. In addition, through the use of small interfering RNA duplexes, we have examined the effect of suppressing endogenous GAP1IP4BP production on [Ca2+] i . In HeLa cells in which the endogenous level of GAP1IP4BP has been suppressed by ∼95%, we failed to observe any effect on Ca2+ mobilization or Ca2+ entry following histamine stimulation. Thus, using various approaches to manipulate the function of endogenous GAP1IP4BP in intact HeLa cells, we have been unable to observe any detectable effect of GAP1IP4BP on [Ca2+] i .


Inositol 1,3,4,5-tetrakisphosphate (IP 4 ) has been linked to a potential role in the regulation of intracellular free Ca 2؉ concentration ([Ca
] i ) following cellular stimulation with agonists that activate phosphoinositide-specific phospholipase C. However, despite many studies, the function of IP 4 remains unclear and indeed there is still some debate over whether it has a function at all. Here we have used various molecular approaches to address whether manipulation of the potential IP 4 receptor, GAP1 IP4BP , affects [Ca 2؉ ] i following cellular stimulation. Using single cell imaging, we show that the overexpression of a constitutively active and a potential dominant negative form of GAP1 IP4BP appear to have no effect on Ca 2؉ mobilization or Ca 2؉ entry following stimulation of HeLa cells with histamine. In addition, through the use of small interfering RNA duplexes, we have examined the effect of suppressing endogenous GAP1 IP4BP production on [Ca 2؉ ] i . In HeLa cells in which the endogenous level of GAP1 IP4BP has been suppressed by ϳ95%, we failed to observe any effect on Ca 2؉ mobilization or Ca 2؉ entry following histamine stimulation. Thus, using various approaches to manipulate the function of endogenous GAP1 IP4BP in intact HeLa cells, we have been unable to observe any detectable effect of GAP1 IP4BP on [Ca 2؉ ] i . Despite many studies on the possible role of inositol 1,3,4,5tetrakisphosphate (IP 4 ) 1 in cellular physiology, its function remains unclear and indeed there is still some debate over whether it has a function at all (1,2). Formed by direct phosphorylation of inositol 1,4,5-trisphosphate (IP 3 ), a reaction catalyzed by a family of Ca 2ϩ -regulated IP 3 3-kinases (1,3), IP 4 has been linked to a potential role in the regulation of intracellular free Ca 2ϩ concentration ([Ca 2ϩ ] i ) following cellular stimulation with agonists that activate phosphoinositide-specific phospholipase C (4 -7). Evidence for this finding has come from a number of sources. For example, in endothelial cells, there is direct evidence (8) and, in neurons, there is direct (9) and indirect evidence (10) that IP 4 can activate Ca 2ϩ influx channels in the plasma membrane. Furthermore, one of the first effects of IP 4 to be reported highlighted an ability of this compound to synergize with IP 3 to mobilize Ca 2ϩ and regulate subsequent store-operated Ca 2ϩ influx (11)(12)(13). However, the marked sensitivity of this particular system to experimental protocols (11, 14 -16) has also raised a significant degree of controversy over the role of IP 4 in intracellular Ca 2ϩ homeostasis. In recent years, we have taken the view that if IP 4 does indeed constitute a novel second messenger, one criterion that must be fulfilled is the presence within cells of protein(s) that specifically bind IP 4 , i.e. an IP 4 receptor. To this end, we have described the purification (17,18) and cloning (19) of a highly specific IP 4 -binding protein termed GAP1 IP4BP . This protein, which functions as a GTPase-activating protein for members of the Ras-like family of small GTPases, at present constitutes the most promising candidate IP 4 receptor.
The Ras-like family includes H-Ras, N-Ras, and K-Ras4A and 4B, the R-Ras proteins, the Ral proteins, and the Rap proteins 1A, 1B, 2A, and 2B (20 -22). These are ubiquitously expressed, evolutionarily conserved proteins that couple extracellular signals to various cellular responses (20 -22). All of these proteins have the inherent ability to undergo conformational changes in response to the alternate binding of GDP and GTP. The GDP-bound "off" state and the GTP-bound "on" state recognize distinct effector proteins, thereby allowing these proteins to function as two-state molecular "switches." Importantly, cycling between the two forms does not occur spontaneously. Activation requires guanine nucleotide exchange factors to induce the dissociation of GDP to allow association of the more abundant GTP, and deactivation requires GTPase-activating proteins (GAPs) to bind to the GTP-bound form to enhance the rate of intrinsic GTPase activity (20 -22). GAP1 IP4BP along with the related proteins GAP1 m , RASAL, and CAPRI (23-30) is composed of tandem N-terminal C 2 domains, a C-terminal pleckstrin homology (PH) domain adjacent to a Bruton's tyrosine kinase (Btk) motif, and a central catalytic Ras GAP-related domain. Associated with the plasma membrane through a complex interaction between its PH/Btk domain and the inner plasma membrane lipid phosphatidylinositol 4,5-bisphosphate (PtdIns(4,5)P 2 ) (31, 32), GAP1 IP4BP functions as a dual Ras and Rap1 GAP (19,33). Importantly, at least in vitro, the Ras GAP activity of GAP1 IP4BP is regulated by IP 4 (19). The data that have raised the key issue of whether the effects of IP 4 on the regulation of [Ca 2ϩ ] i are mediated by GAP1 IP4BP . At present, the only direct evidence in favor of this finding has emerged from L-1210 cells (34,35). In these perme-abilized cells, the addition of exogenous GAP1 IP4BP enhances the well documented ability of IP 4 to potentiate IP 3 -stimulated Ca 2ϩ release (36). However, it is not yet clear as to the role of GAP1 IP4BP in vivo. In this study, we have addressed this issue by using various molecular approaches including overexpression of potential constitutive and dominant negative forms of GAP1 IP4BP and siRNA to examine a potential in vivo role for GAP1 IP4BP in the regulation of [Ca 2ϩ ] i in HeLa cells.

EXPERIMENTAL PROCEDURES
Expression and Purification of GST Fusion Proteins in E. coli-The pGEX plasmids containing the coding sequences for the GAP1 IP4BP mutants, isolated using the Transformer kit from Clontech as previously described (31,32), and ⌬C 2 GAP1 IP4BP were transformed individually into the E. coli strain BL21(DE3) to express and purify as GST fusion proteins using the procedure described previously (33). A single colony of the transformed strain was inoculated into 5 ml of LB containing ampicillin (100 g/ml) and grown overnight at 37°C with shaking at 250 rpm. The overnight culture was diluted 1:100 with fresh LB containing ampicillin and grown at 37°C with shaking until the cell density reached an A 600 of 0.5. Protein expression was induced with 0.2 mM isopropyl-1-thio-␤-D-galactopyranoside for 3 h at 37°C after which cells were collected by centrifugation and resuspended in 25 ml of ice-cold buffer A (PBS containing 1 mM EDTA, 1 mM EGTA, 0.2 mM phenylmethylsulfonyl fluoride, and 1 mM ␤-mercaptoethanol). Cells were lysed by sonication (3 ϫ 20-s pulses with 1 min at 4°C between each sonication) and incubated at 4°C with 1% (v/v) Triton X-100 for 1 h with gentle mixing prior to the removal of cell debris by centrifugation. 1 ml of a 50% slurry of glutathione-Sepharose 4B resin (Amersham Biosciences) prewashed with buffer A was added to the resultant supernatant and incubated overnight at 4°C with constant shaking. The resin was washed with 3 ϫ 10 ml of ice-cold buffer A prior to elution of bound GST fusion protein with 3 ϫ 1 ml of 50 mM Tris-HCl containing 10 mM glutathione, pH 8.0, by incubating with constant mixing for 10 min at room temperature. The eluates were pooled, and protein content was estimated by the Bradford method using bovine serum albumin as a standard. SDS-polyacrylamide gel electrophoresis was used to check the purity of the eluted protein.
[ 32 P]IP 4 -binding Assays-These were performed as described previously. An individual binding assay contained 100 mM KCl, 20 mM NaCl, 10 mM Hepes/NaOH, pH 7.0, 1 mM EDTA, 30,000 dpm of [ 32 P]IP 4 (see Ref. 17), 0.5-1 g of GST fusion protein, and various concentrations of competing unlabeled inositol phosphates in a final volume of 0.5 ml. Equilibrium binding was reached by a 15-min incubation at 4°C after which the receptor-ligand complex was precipitated by the addition of 100 l of 5 mg/ml ␥-globulin and 1 ml of 25% (w/v) polyethylene glycol. The samples were spun for 10 min prior to the removal of the supernatant with the resultant pellet being briefly washed prior to counting.
In Vitro Ras GAP Assays-These were performed under first order kinetics as described previously (33). The particular GTP-binding protein was loaded with [␥-32 P]GTP (3,000 Ci mmol Ϫ1 , Amersham Biosciences) for 5 min at 25°C. GTPase activity was assayed at 25°C by the addition of the various GAPs to the loaded GTP-binding protein. At the required time points, activity was stopped by the addition of 5 mM silicotungstate, 1 mM H 2 SO 4 with the liberated [ 32 P]P i being extracted with isobutanol/toluene (1/1 v/v), 5% (w/v) ammonium molybdate, and 2 M H 2 SO 4 . The upper phase was removed for scintillation counting.
Generation of a GAP1 IP4BP Monoclonal Antibody-A monoclonal antibody (GP-3) was raised against a GST fusion of a GAP1 IP4BP mutant lacking the N-terminal tandem C 2 domains as follows. 250 g of the isolated GST-⌬C 2 GAP1 IP4BP was emulsified in Freund's complete adjuvant prior to intramuscular injection into four male mice. At 22 days, an additional 250 g emulsified in Freund's complete adjuvant was injected subcutaneously. Ten days later, a tail vein bleed showed a high antibody titer (as detected by immunoblotting of ⌬C 2 GAP1 IP4BP ) and thus 1 day later, the pre-kill boost (250 g in 150 mM NaCl, 10 mM K 2 PO 4 , pH 7.4) was administered intravenously. Three days later, test serum and the spleen were removed. Fusion was undertaken by mixing spleen cells with myeloma cells in a ratio of 5:1. Mixed cells were plated in 4 ϫ 96-well plates containing MRC-5 feeder cells and 100 l of 20% fetal calf serum in Dulbecco's modified Eagle's medium. These cells were incubated in a CO 2 incubator at 37°C. Subsequent culture of hybridomas and cloning by limited dilution were followed as standard. Antibodies were used in experiments in the form of untreated culture supernatants from cells grown to stationary phase.
Immunoprecipitation of Endogenous GAP1 IP4BP Using the GP-3 Monoclonal Antibody-Immunoprecipitations were carried out using antibodies immobilized on protein G-Sepharose beads (Amersham Biosciences). Protein G beads were washed in PBS and incubated with the GP-3 hybridoma supernatant for 4 h at 4°C with constant mixing. The beads were washed in PBS and then twice in 0.2 M sodium borate. Dimethyl primelimidate was added to a final concentration of 20 mM, and the slurry was mixed at room temperature for 30 min. After one wash in 0.2 M ethanolamine, the beads were incubated in 0.2 M ethanolamine for an additional 2 h to complete the covalent coupling of the antibody to the beads. The beads were washed thoroughly in PBS and then stored at 4°C in PBS containing 0.01% azide. For immunoprecipitations, cell pellets were resuspended in ice-cold lysis buffer (PBS, 1% Triton X-100, 1 mM EGTA, 0.01% azide), and then cell debris was pelleted by high speed centrifugation. The GP-3-coupled Sepharose beads were washed thoroughly in lysis buffer, and then the lysates were added to the beads and incubated, mixing end-over-end at 4°C for between 2 and 4 h. Beads were washed several times with lysis buffer, and the final wash aspirated tight to the beads. Samples were separated by SDS-PAGE prior to visualization.
Isolation of Stable, Inducible HeLa Cells-pTet-Off-transfected HeLa cells (Clontech) were cultured in Dulbecco's modified Eagle's medium containing 10% (v/v) fetal-calf serum (Invitrogen), 100 units/ml penicillin, 100 g/ml streptomycin, 100 g/ml G418, and 2 g/ml doxycycline. Cells were co-transfected with GAP1 IP4BP pTRE and pTK-Hyg (Clontech) using LipofectAMINE (Invitrogen) according to the manufacturer's instructions. Note that pTRE does not contain a mammalian-selectable marker, but pTK-Hyg confers hygromycin resistance. Stable cell lines were selected using 500 g/ml hygromycin B (Roche Molecular Biochemicals) and maintained in 250 g/ml hygromycin B. When the expression of GAP1 IP4BP was desired, the medium was replaced with doxycycline-free medium 24 -48 h prior to using the cells.
RNA Interference of Endogenous GAP1 IP4BP Expression-GAP1 IP4BP (o/e 3 ) cells were seeded into the wells of a six-well plate, and 24 h later at a confluency of 30%, they were transiently transfected using OligoFECTAMINE (Invitrogen) with 200 pmol of GAP1 IP4BPspecific siRNA according to the manufacturer's instructions (siRNA sequence 5Ј-AAGUCACUCUGCCCGUUUUAC-3Ј). Cells were cultured in the presence of doxycycline for the required period prior to Ca 2ϩ imaging and the analysis of GAP1 IP4BP level by immunoblotting using the GP-3 monoclonal antibody.
Single-cell Ca 2ϩ Measurements-Cells were grown on 22-mm coverslips and then washed with 2 ml of extracellular medium (121 mM NaCl, 5.4 mM KCl, 1.6 mM MgCl 2 , 6 mM NaHCO 3 , 9 mM glucose, 25 mM Hepes, 1.8 mM CaCl 2 , pH 7.4) and incubated in 1 ml of extracellular medium containing 2 M Fura-2/AM for 30 min at room temperature. Cells were then washed twice with 2 ml of extracellular medium. Ca 2ϩ imaging was performed using the Merlin ratiometric Ca 2ϩ -imaging system (PerkinElmer Life Sciences), an UltraPix FKL300 digital camera, and an Olympus IX50 inverted microscope fitted with a ϫ40 oil-immersion objective lens. The data were processed using Merlin software. All imaging was performed at 37°C with constant perfusion. [Ca 2ϩ ] calibrations were calculated with Merlin software using 340/380 ratio maxima and minima obtained by the treatment of cells with 2 M ionomycin followed by 20 mM EGTA.

RESULTS
Generation of a Constitutively Active and a Potential Dominant Negative Form of GAP1 IP4BP -Our initial approach to examining the role of GAP1 IP4BP in the regulation of [Ca 2ϩ ] i was to design a series of site-directed mutants with the aim of isolating constitutively active and dominant negative forms that could potentially be used to interfere with the function of endogenous GAP1 IP4BP . As shown in Fig. 1, the conversion of arginine into cysteine at position 601 within the PH/Btk domain of GAP1 IP4BP resulted in a mutant protein that, although displaying a greatly reduced IP 4 binding activity, ϳ10% of that seen with wild-type protein (33) (Fig. 1A) nevertheless retained full Ras GAP activity when assayed in vitro (Fig. 1B). In contrast, the conversion of leucine into alanine at position 484, a residue located within the predicted Ras-binding site present in the Ras GAP-related domain (37), resulted in a GAP1 IP4BP mutant that, although displaying no detectable in vitro Ras GAP activity (Fig. 1B), retained the ability to bind IP 4 with an affinity and isomeric specificity identical to that of wild type ( Fig. 1A and data not shown).
To assay the Ras GAP activity of these mutants in vivo, we transiently transfected HeLa cells with H-Ras and the relevant GAP1 IP4BP and assayed the level of Ras-GTP using a GST fusion of the Ras-GTP-binding domain from Raf. In serumstarved cells, a substantial quantity of overexpressed H-Ras was in the GTP-bound state (Fig. 2). Under these conditions, wild-type GAP1 IP4BP caused a significant decrease in the amount of Ras-GTP that was recovered compared with controls. This basal Ras GAP activity was dependent on the Ras GAP-related domain because the expression of GAP1 IP4BP (L484A) had no effect on Ras-GTP levels, even when this mutant was expressed at significantly higher levels than wild-type GAP1 IP4BP (Fig. 2). The expression of the cytosolic GAP1 IP4BP (R601C) resulted in a significant decrease in the amount of recoverable Ras-GTP when compared with the expression of the same level of wild-type GAP1 IP4BP (Fig. 2). Similar data were also obtained using green fluorescent protein-tagged versions of wild type and the various GAP1 IP4BP mutants (data not shown).

Expression of the GAP1 IP4BP Mutants Has No Apparent Effect on Intracellular Ca 2ϩ Homeostasis following Histamine
Stimulation of HeLa Cells-To examine the effect of overexpressing wild-type GAP1 IP4BP and the various GAP1 IP4BP mutants on intracellular Ca 2ϩ homeostasis, we transiently transfected HeLa cells with the GAP1 IP4BP constructs tagged at their N termini with yellow fluorescent protein (YFP). This spectral variant of green fluorescent protein allowed visualization of those cells expressing GAP1 IP4BP and, when coupled with Fura-2 imaging, allowed the analysis of Ca 2ϩ transients in single cells following stimulation with histamine (Fig. 3). We performed experiments in the absence of extracellular Ca 2ϩ to allow discrimination between the Ca 2ϩ released from internal stores and that from Ca 2ϩ entry. In these assays, the initial increase in [Ca 2ϩ ] i that arose from IP 3 -stimulated release of Ca 2ϩ from internal stores (38) was not significantly altered in those cells expressing YFP-GAP1 IP4BP (R601C), YFP-GAP1 IP4BP (L484A), or YFP-GAP1 IP4BP (Fig. 3). Such data suggest that the overexpression of wild-type GAP1 IP4BP or the various GAP1 IP4BP mutants has no detectable effect on IP 3stimulated Ca 2ϩ release when assayed under these conditions.
In addition, after allowing the [Ca 2ϩ ] i to return to the basal state, we analyzed store-operated Ca 2ϩ entry by adding extracellular Ca 2ϩ . This induced an elevation of [Ca 2ϩ ] i that was not significantly altered by the expression of various GAP1 IP4BP constructs (Fig. 3). As a positive control, we also analyzed the histamine-stimulated Ca 2ϩ transient that was present in cells overexpressing the store-operated Ca 2ϩ influx channel Trp3. Here we observed a clear enhancement of Ca 2ϩ influx, consistent with that reported previously (39). Thus, under these particular assay conditions, GAP1 IP4BP has no detectable effect on store-operated Ca 2ϩ influx. Fig. 3 suggest that when assayed under these particular conditions GAP1 IP4BP appears to have no detectable effect on [Ca 2ϩ ] i following histamine stimulation of HeLa cells, we extended our experimental approach by examining the effect of manipulating the level of endogenous GAP1 IP4BP . To undertake this approach, we first generated a monoclonal antibody against GAP1 IP4BP that was capable of detecting endogenous protein (our previously characterized GAP1 IP4BP polyclonal antibody, H113 (19), was unable to detect endogenous levels of GAP1 IP4BP in HeLa cells (data not shown)). As described under "Experimental Procedures," a mouse monoclonal antibody (GP-3) was raised against a GST fusion of a GAP1 IP4BP mutant lacking the N-terminal tandem C 2 domains (33).

Isolation of a Monoclonal Antibody Capable of Detecting Endogenous GAP1 IP4BP -Although the data presented in
In HeLa cells overexpressing each of the mammalian GAP1 family members, GP-3 was only capable of detecting a protein in those cells transiently transfected with an expression vector for GAP1 IP4BP (Fig. 4A). In non-transfected HeLa cells, GP-3 detected an endogenous protein of ϳ96 kDa on Western blots (Fig. 4B). Furthermore, GP-3 was capable of immunoprecipitating a 96-kDa protein that was detected by the GAP1 IP4BPspecific polyclonal antibody H113 (Fig. 4C). The GP-3 immunoprecipitate also displayed Ras GAP activity and specific IP 4 binding activity (data not shown). Taken together, these data suggest that the 96-kDa protein detected by GP-3 in HeLa cell lysates is endogenous GAP1 IP4BP . Therefore, this antibody was employed in the subsequent assays designed to silence the expression of endogenous GAP1 IP4BP .
Using Double-stranded RNA Interference to Deplete Endogenous GAP1 IP4BP -To deplete endogenous GAP1 IP4BP , we used RNA interference. RNA interference is the process of sequencespecific post-transcriptional gene silencing in animals and plants initiated by double-stranded RNA that is homologous in sequence to the silenced gene (40). Recently, 21-nucleotide siRNA duplexes have been shown to specifically suppress the expression of endogenous genes in mammalian cell lines (41)(42)(43). Therefore, we designed and used a specific siRNA to sup-  1000). B, cell lysates (50 g) from HeLa cells either non-transfected or transiently transfected with pCIneo-GAP1 IP4BP were separated by SDS-PAGE and visualized by immunoblotting with 1:1000 diluted GP-3. C, non-transfected HeLa cell lysates were incubated with Sepharose beads coupled to either GP-7 (a control mouse monoclonal antibody that does not detect GAP1 IP4BP ) or GP-3 as described under "Experimental Procedures." After washing, the resultant beads were boiled in SDS-PAGE loading buffer prior to separation. Immunoprecipitated GAP1 IP4BP was visualized by immunoblotting with H113, a polyclonal anti-peptide antibody raised against the C-terminal region of GAP1 IP4BP . press the expression of endogenous GAP1 IP4BP .
To achieve an experimental system amenable to manipulation, we initially isolated a stably transfected HeLa cells line expressing full-length human GAP1 IP4BP under the control of a tetracycline-regulated promoter (see "Experimental Procedures"). To examine the leakiness and inducibility of GAP1 IP4BP , the expression in this cell line named GAP1 IP4BP (o/e 3 ), we compared whole cell lysates derived from the parental pTet-Off HeLa cell line with those from GAP1 IP4BP (o/e 3 ) following 48-h culture in the presence or absence of 2 g/ml doxycycline (a tetracycline analogue). When cultured in the presence of doxycycline, the endogenous level of GAP1 IP4BP in GAP1 IP4BP (o/e 3 ) and the parental cell line were almost identical (Fig. 5A). However, in the absence of doxycycline, immunoblotting with GP-3 revealed a high level of inducible overexpression of GAP1 IP4BP (Fig. 5A). Furthermore, the degree of overexpression could be tightly regulated by titration of the doxycycline concentration within the culture medium (Fig. 5B).
To deplete endogenous GAP1 IP4BP , we transiently transfected GAP1 IP4BP (o/e 3 ) cells with siRNA designed against GAP1 IP4BP . After 24 -48-h culture in the continued presence of 2 g/ml doxycycline, immunoblotting with GP-3 revealed that the level of endogenous GAP1 IP4BP was reduced by 91.3 Ϯ 6.1% (Fig. 5C). This depletion was specific for the GAP1 IP4BP siRNA because no detectable loss of endogenous GAP1 IP4BP was observed in cells treated with a control siRNA (Fig. 5C).
Thus, for the GAP1 IP4BP (o/e 3 ) cell line, we could manipulate the level of GAP1 IP4BP in a number of ways. First, culturing in the continued presence of 2 g/ml doxycycline resulted in a level of GAP1 IP4BP that was similar to the endogenous level present in the parental cell line. Second, culturing in the absence of doxycycline generated cells in which GAP1 IP4BP was significantly overexpressed. Third, culturing in the presence of 2 g/ml doxycycline following transient transfection with the GAP1 IP4BP -specific siRNA resulted in cells in which the endogenous GAP1 IP4BP was dramatically depleted. Finally, culturing cells following transient transfection with the GAP1 IP4BP -specific siRNA in the presence of a carefully determined concentration of doxycycline allowed the regulation of promoter activity such that the overall level of GAP1 IP4BP returned to near endogenous levels (data not shown). This latter condition was of particular importance because in the event that suppression of endogenous GAP1 IP4BP resulted in a detectable change in [Ca 2ϩ ] i , revertion of the change by the manipulation of GAP1 IP4BP back to the endogenous levels would strongly suggest that the phenotype was the result of depletion of GAP1 IP4BP and not a nonspecific effect of the siRNA.
In the initial experiments, we analyzed the level of Ras-GTP present within GAP1 IP4BP (o/e 3 ) cells cultured under the various conditions. As shown in Fig. 6, under conditions in which GAP1 IP4BP was significantly overexpressed, the level of Ras-GTP was significantly reduced compared with cells expressing endogenous levels of GAP1 IP4BP . This finding contrasted with conditions in which endogenous GAP1 IP4BP had been depleted. Here, Ras-GTP levels were similar to control cells. Thus, manipulating GAP1 IP4BP levels within this cell line results in the alteration of the levels of Ras-GTP.
To analyze the intracellular free Ca 2ϩ homeostasis within GAP1 IP4BP (o/e 3 ) cells cultured under these conditions, we performed Ca 2ϩ -imaging experiments in the absence of extracellular Ca 2ϩ . However, using Fura-2-loaded cells, we failed to observe any effect of overexpressing GAP1 IP4BP or depleting endogenous GAP1 IP4BP on the initial increase in [Ca 2ϩ ] i following histamine stimulation (Fig. 7). In addition, after allowing the [Ca 2ϩ ] i to return to the basal state, we analyzed storeoperated Ca 2ϩ entry by adding extracellular Ca 2ϩ . This induced an elevation of [Ca 2ϩ ] i that was again not significantly altered by the overexpression of GAP1 IP4BP or the depletion of endogenous GAP1 IP4BP (Fig. 7). DISCUSSION In this report, we have described results from a series of experiments aimed at addressing whether the putative IP 4 receptor GAP1 IP4BP may function to allow IP 4 to regulate [Ca 2ϩ ] i . Our initial approach was to design a series of site-directed mutants with the aim of isolating constitutively active and interfering forms of GAP1 IP4BP that could be used to potentially block the function of endogenous GAP1 IP4BP . First, the conversion of arginine into cysteine at position 601 within the PH/Btk domain of GAP1 IP4BP resulted in a mutant protein that, although displaying a greatly reduced IP 4 binding activity (33), retained full Ras GAP activity when assayed in vitro. Importantly, when assayed in vivo, this mutant, GAP1 IP4BP (R601C), displayed a significantly elevated level of basal Ras GAP activity when compared with wild-type GAP1 IP4BP . In other words, GAP1 IP4BP (R601C) may function as a constitutively active mutant. Such a conclusion is consistent with the in vitro evidence that when associated with plasma membrane-like liposomes through PtdIns(4,5)P 2 binding to its PH/Btk domain, GAP1 IP4BP has a suppressed Ras GAP activity, a suppression that can be alleviated following the binding of IP 4 to the PH/Btk domain (19). In the case of the GAP1 IP4BP (R601C) mutant, the inability of the PH/Btk domain to bind IP 4 is mirrored by a loss in binding to PtdIns(4,5)P 2 (32). This results in GAP1 IP4BP (R601C) residing in the cytosol rather than at the plasma membrane (31). Importantly, the induced activity of the GAP1 IP4BP (R601C) mutant does not appear to simply result from a loss in plasma membrane association. Targeting this mutant back to the plasma membrane by the addition of a CAAX motif does not result in a suppression of the Ras GAP activity (data not shown). This finding suggests that the constitutive activity of the GAP1 IP4BP (R601C) mutant may result from the removal of the negative influence of PtdIns(4,5)P 2 binding to the PH/Btk on the Ras GAP activity. Such a scenario may help explain how IP 4 can activate the Ras GAP activity of this protein in vitro (19). IP 4 may remove the negative control exerted by PtdIns(4,5)P 2 by competing for binding to the PH/Btk domain.
In contrast, the conversion of leucine into alanine at position 484 resulted in a GAP1 IP4BP mutant that, although displaying no detectable in vitro or in vivo Ras GAP activity, retained the ability to bind IP 4 with an affinity and isomeric specificity identical to wild-type GAP1 IP4BP . This mutant also retained the ability to associate with the plasma membrane (data not shown). Therefore, it is possible that upon overexpression, this particular mutant, GAP1 IP4BP (L484A), may interfere with the function of endogenous GAP1 IP4BP by binding to IP 4 , but because of the loss of GAP activity, this mutant will not induce the subsequent regulation of Ras function. Indeed, the overexpression of a comparable mutant of the GAP1 IP4BP -related protein CAPRI appears to interfere with endogenous CAPRI function (30).
Having generated and characterized the various GAP1 IP4BP mutants, we analyzed their effect on the changes in [Ca 2ϩ ] i that occur following histamine stimulation of HeLa cells. Through single cell analysis of HeLa cells overexpressing each of these mutants, we failed to observe any alteration in the release of internally stored Ca 2ϩ or store-operated Ca 2ϩ entry following histamine stimulation. Thus, using these particular mutants, we were unable to detect any apparent role for GAP1 IP4BP in the regulation of [Ca 2ϩ ] i observed following histamine stimulation of HeLa cells.
To complement these studies and negate potential problems especially with interpreting the ability of the GAP1 IP4BP (L484A) mutant to function as a dominant negative, we observed the affect of depleting endogenous GAP1 IP4BP on [Ca 2ϩ ] i . Using siRNA technology, we were able to significantly suppress the expression of endogenous GAP1 IP4BP in a HeLa cell line stably expressing GAP1 IP4BP under a tetracycline-regulated promoter. However, under conditions in which we manipulated the level of endogenous GAP1 IP4BP either by overexpression or siRNA suppression, we again failed to observe any affect on the release of internally store-operated Ca 2ϩ or store-operated Ca 2ϩ influx following stimulation with histamine.
As far as we are aware, this is the first study in which agonist-dependent changes in [Ca 2ϩ ] i have been analyzed in cells in which the endogenous level of GAP1 IP4BP has been so dramatically varied within the same cell line. The only other study that has reported the effect of reducing the expression of GAP1 IP4BP on [Ca 2ϩ ] i within mammalian cells is that of Lu et al. (44). In this study (44), the authors examined the role of GAP1 IP4BP in store-operated Ca 2ϩ entry by comparing two clonally selected human erythroleukemia (HEL) cell lines, one A, the data shown are from one series of experiments on GAP1 IP4BP (o/e 3 ) cells that had been cultured under the various conditions required to manipulate GAP1 IP4BP (see "text"). After culturing for the required period, cells were loaded with Fura-2/AM and washed in Ca 2ϩ -free medium. Cells were challenged by the addition of 100 M histamine, and after the initial Ca 2ϩ elevation had returned to basal, 10 mM extracellular Ca 2ϩ was added. The inset shows quantification of the Ca 2ϩ entry as defined as the percent of the peak value of the initial Ca 2ϩ elevation following histamine addition. B, visualization of GAP1 IP4BP levels using the GP-3 monoclonal antibody from samples obtained from the same series of experiment from which the data in A were obtained. expressing a constitutively active GAP1 IP4BP antisense construct (AS-HEL cells) and the other a vector lacking the antisense construct (V-HEL cells) (44). In the AS-HEL cells in which endogenous GAP1 IP4BP was reduced by between 85 and 96%, a substantial increase in Ca 2ϩ entry was observed following thrombin addition (44). This enhanced entry was shown to result from the activation of intermediate conductance Ca 2ϩactivated K ϩ channels that induced membrane hyperpolarization, thereby stimulating Ca 2ϩ entry (44). However, partly because the expression of the GAP1 IP4BP antisense was not inducible, the authors (44) were unable to firmly establish that the difference in [Ca 2ϩ ] i observed between the two cell lines was not simply down to clonal variation. Furthermore, no attempt was made to revert the observed Ca 2ϩ phenotype by expression of exogenous GAP1 IP4BP in the AS-HEL cells. Therefore, it remains unclear whether the observed affect of depleting endogenous GAP1 IP4BP on intracellular Ca 2ϩ homeostasis within HEL cells stems from the direct loss of GAP1 IP4BP .
Currently, the only direct evidence for a role of GAP1 IP4BP in Ca 2ϩ homeostasis has emerged from L-1210 cells (34,35). In these permeabilized cells, the addition of exogenous GAP1 IP4BP enhances the well documented ability of IP 4 to potentiate IP 3stimulated Ca 2ϩ release (36). This is a reproducible observation that clearly suggests that in these cells under these particular conditions the binding of IP 4 to GAP1 IP4BP does appear to play a role in the regulation of IP 3 -stimulated Ca 2ϩ release (36). However, our approach of manipulating endogenous GAP1 IP4BP within intact HeLa cells suggests that a global role for GAP1 IP4BP in the regulation of [Ca 2ϩ ] i may not be a detectable phenomenon.