Trp1, a candidate protein for the store-operated Ca(2+) influx mechanism in salivary gland cells.

The trp gene family has been proposed to encode the store-operated Ca(2+) influx (SOC) channel(s). This study examines the role of Trp1 in the SOC mechanism of salivary gland cells. htrp1, htrp3, and Trp1 were detected in the human submandibular gland cell line (HSG). HSG cells stably transfected with htrp1alpha cDNA displayed (i) a higher level of Trp1, (ii) a 3-5-fold increase in SOC (thapsigargin-stimulated Ca(2+) influx), determined by [Ca(2+)](i) and Ca(2+)-activated K(+) channel current measurements, and (iii) similar basal Ca(2+) permeability, and inhibition of SOC by Gd(3+) but not by Zn(2+), as compared with control cells. Importantly, (i) transfection of HSG cells with antisense trp1alpha cDNA decreased endogenous Trp1 level and significantly attenuated SOC, and (ii) transfection of HSG cells with htrp3 cDNA did not increase SOC. These data demonstrate an association between Trp1 and SOC and strongly suggest that Trp1 is involved in this mechanism in HSG cells. Consistent with this suggestion, Trp1 was detected in the plasma membrane region, the proposed site of SOC, of acinar and ductal cells in intact rat submandibular glands. Based on these aggregate data, we propose Trp1 as a candidate protein for the SOC mechanism in salivary gland cells.

[Ca 2ϩ ] i regulates the physiological function of a variety of nonexcitable cell types, including salivary gland cells (1)(2)(3). In salivary gland cells, a sustained elevation of [Ca 2ϩ ] i is required for the activation of the ion channels, such as the Ca 2ϩ -activated K ϩ channel and the Ca 2ϩ -activated Cl Ϫ channel, which critically regulate the secretion of fluid and electrolytes. [Ca 2ϩ ] i increase in these cells results from intracellular Ca 2ϩ release and from extracellular Ca 2ϩ influx. While release of Ca 2ϩ from the intracellular store(s) triggers a transient increase in [Ca 2ϩ ] i and fluid secretion, sustained fluid secretion is dependent on the influx of Ca 2ϩ from the extracellular medium (3,4). Intracellular Ca 2ϩ release is induced by inositol 1,4,5-trisphosphate (IP 3 ) 1 that is generated in reponse to agonist-dependent hydrolysis of phosphatidylinositol 4,5-bisphosphate (PIP 2 ) (1,5). Ca 2ϩ influx across the plasma membrane is mediated via an as yet unknown mechanism referred to as store-operated Ca 2ϩ influx (SOC), or capacitative calcium entry, that is activated by the depletion of Ca 2ϩ from the internal Ca 2ϩ store (3, 6 -8).
The molecular basis for the activation and regulation of SOC has not yet been determined in any nonexcitable cell type. The two critical questions that remain are regarding (i) the signal that is transmitted from the internal Ca 2ϩ store to the plasma membrane to trigger activation or inactivation of SOC and (ii) the molecule that mediates Ca 2ϩ influx into the cells. Recent studies with salivary gland and other nonexcitable cells indicate that store-operated Ca 2ϩ influx is mediated via a channellike mechanism (8,9). Electrophysiological measurements demonstrate that Ca 2ϩ influx is associated with the activation of an inward Ca 2ϩ current, which appears to be selective for Ca 2ϩ in some cell types and relatively nonspecific for cations in others (2,8). However, the mechanism involved in gating the putative Ca 2ϩ influx channel is not known. Studies toward identifying the molecule(s) mediating Ca 2ϩ influx have led to the cloning of mammalian homologues of the Drosophila transient receptor potential gene, trp (10,11). Presently, full-length or partial sequences have been reported for seven trp genes (trp1-trp7) in various mammalian species, including human, mouse, rat, rabbit, and bovine. It has been proposed that the trp gene(s) encode the store-operated Ca 2ϩ channel in nonexcitable cells. However, studies examining the functional effects induced by expression of trp genes have demonstrated that the characteristics of the expressed activity are distinct from those of the endogenous SOC activity, suggesting that the encoded Trp protein is functionally different from the endogenous SOC mechanism present in the cells used for the expression. Additionally, several studies also show an increase in receptoroperated, or basal, Ca 2ϩ influx in cells transfected with trp cDNAs. Thus, only a few trp gene products meet the functional criteria for SOC, typically defined as an increase in Ca 2ϩ influx stimulated in cells following internal Ca 2ϩ store-depletion by thapsigargin.
Expression of the Drosophila trp cDNA, dtrp (12); human trp1␤, the alternatively spliced variant, short form of htrp1, (13), but not htrp1␣ (14,15); bovine trp4, btrp4 (16); rat trp4, rtrp4 (17); and rabbit trp5 (18) but not mouse trp5, mtrp5 (19), was associated with increased Ca 2ϩ influx in response to treatment with thapsigargin. mtrp2 expression in HEK-293 cells induced a relatively small increase in SOC and also increased receptor-stimulated Ca 2ϩ influx (20). On the other hand, rtrp2 has been suggested to be associated with the sensory signaling mechanism in brain, apparently independent of store depletion (21). The reported data strongly suggest that Trp3, Trp6, and Trp7 proteins are associated with receptor-or second messenger-regulated Ca 2ϩ influx (22)(23)(24)(25). Consistent with this, it was * The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. ‡ These two authors contributed equally to this work.
¶ To whom all correspondence should be addressed: Bldg. 10 recently suggested that Trp3 might be regulated by IP 3 via interaction with the IP 3 receptor (26). However, it remains to be established whether the Trp3-IP 3 receptor interaction is involved in the mechanism regulating SOC.
More convincing data regarding a role for Trp in SOC has been obtained by expression of antisense sequences of trp cDNA. Expression of either a mixture of antisense cDNA fragments of mtrp1-mtrp6 or of mtrp4 alone attenuated carbacholstimulated Ca 2ϩ influx (10,14). However, it is difficult to assess the role of the trp gene products based on these data, since thapsigargin-stimulated influx was not tested in these studies and, as mentioned above, expression of several trp isoforms induce increases in second messenger-or receptoroperated type of Ca 2ϩ influx. In another study, expression of an antisense oligodeoxyribonucleotide against the region of the start codon of mtrp1 in Xenopus oocytes attenuated thapsigargin stimulation of a I Cl Ϫ (Ca) (17). This sequence is present in all four variants of mtrp1 genes, although only trp1␣ and -␤ genes have been suggested to encode a polypeptide (27).
The present study uses the human submandibular gland ductal cell line, HSG, to examine the role of the trp1 gene product in the SOC mechanism in salivary gland cells. Similar internal Ca 2ϩ release and store-operated Ca 2ϩ influx mechanisms are triggered upon activation of the muscarinic receptor in salivary acinar and ductal cells (28). Consistent with this, HSG cells exhibit a robust activation of Ca 2ϩ influx when the internal Ca 2ϩ store(s) is depleted by either by muscarinic receptor stimulation (via an IP 3 -dependent mechanism) or by treatment with thapsigargin or tert-butylhydroxyquinone (9,29). Furthermore, SOC appears to be the primary mechanism for Ca 2ϩ influx in these cells following muscarinic receptor stimulation or treatment with thapsigargin. Thus, HSG cells provide an excellent, well characterized model system to study the mechanism of SOC. The results presented below demonstrate that transfection of HSG cells with htrp1␣, but not htrp3, cDNA induced dramatic increases in thapsigargin-stimulated SOC, while expression of the htrp1␣ cDNA in the antisense direction significantly attenuated the endogenous SOC activity. Importantly, changes in SOC were associated with corresponding changes in the levels of Trp1 in HSG cell plasma membranes. Further, and consistent with its proposed function, we show that Trp1 is localized in the plasma membrane region of acinar and ductal cells in intact rat submandibular glands. In aggregate, these data strongly suggest that Trp1 is a candidate protein for the SOC mechanism in salivary gland cells.

EXPERIMENTAL PROCEDURES
Cell Culture-HSG cells were a kind gift from Dr. Mitsunabo Sato (Tokushima University, Japan). The conditions for cell culture were similar to those described previously (7,29). Briefly, cells were grown in Eagle's minimum essential medium under 5% CO 2 at 37°C. The culture medium was supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 units/ml penicillin, and 100 g/ml streptomycin (all from Biofluids, Rockville, MD). Cells were passaged when confluent by detaching from the tissue culture dish with 0.25% trypsin, 1.0 mM EDTA. For functional studies, a single-cell suspension was reseeded on coverslips and used after about 24 h.
RNA Isolations, Synthesis of the First Strain cDNA, and RT-PCR Analysis-Rat submandibular glands were excised from anesthetized male Wistar rats (Harlan Sprague Dawley) and frozen immediately in liquid nitrogen. Total RNA was extracted from the frozen tissues and HSG cells using TRIzol reagent (Life Technologies, Inc.) and was treated with deoxyribonuclease I (Life Technologies, Inc.) at a concentration of 1 unit of DNase I/1 g of RNA in a buffer containing 20 mM Tris-HCl (pH 8.4), 2 mM MgCl 2 , 50 mM KCl, for 15 min at room temperature. The reaction was terminated by adding EDTA at a final concentration of 2.5 mM and heated at 65°C for 10 min.
Transfection of HSG Cells-Transfection of HSG cells with htrp1␣ or htrp3 cDNA was performed using the LipofectAMINE reagent (Life Technologies, Inc.). Briefly, HSG cells were cultured to 80% confluence. 3 g of the plasmid pCDNA3, HA-tagged (5Ј-end) htrp1␣, or HA-tagged (3Ј-end) htrp3 cDNA was diluted into 300 l of serum-free Opti-MEM I medium (Life Technologies, Inc.) and was incubated at room temperature for 15 min and then mixed with 30 l LipofectAMINE reagent diluted into 300 l of serum-free Opti-MEM I medium. The mixture was incubated for 45 min at room temperature and then added to 2.4 ml of serum-free medium. The cells were rinsed serum-free medium, overlaid with the transfection mixture, and incubated at 37°C at 5% CO 2 for 5 h. Medium containing 20% FBS was added to the cells, and they were incubated overnight. The medium was then replaced with fresh growth medium containing 20% FBS, and cells were allowed to grow for 72 h. Cells were then detached using versene (Biofluids) and diluted 1:30 in growth medium containing 10% FBS and G418 (400 g/ml) to select stably transfected clones. G418-resistant single clones were selected by 3 weeks after transfection, and SOC activity was tested as described below.
Membrane Preparation and Western Blot Analysis-A crude membrane fraction was prepared from control or transfected HSG cells in buffer containing 50 mM Tris-HCl, 1 mM phenylmethylsulfonyl fluoride (Calbiochem), 19 l/ml aprotinin (Sigma). The cells were homogenized using a Polytron homogenizer, and the homogenate was centrifuged at 900 ϫ g at 4°C for 10 min. The supernatant was filtered and centrifuged at 48,000 rpm for 30 min. The pellets were suspended in the above buffer, aliquoted, frozen, and stored at Ϫ70°C until use. The plasma membrane fraction was purified as described before (31) using a Percoll gradient and stored at Ϫ80°C.
Conditions for SDS-polyacrylamide gel electrophoresis and Western blotting were as described previously (30). After transfer of the proteins, polyvinylidene difluoride membranes were blocked with blocking (TTBS) buffer containing 20 mM Tris, pH 7.5, 68.5 mM NaCl, 0.1% Tween 20, and 5% milk, washed, and then incubated in TTBS containing the primary antibody (1:1000 dilution), either anti-Trp1 or anti-HA (BABCo), for 1 h. The membrane was incubated with the secondary antibody, either anti-rabbit or anti-mouse IgG, as required, in TTBS with 0.05% milk, washed, treated with the ECL reagent (Amersham Pharmacia Biotech), and exposed to X-OMAT TM films (Kodak) as required for detection of the proteins.
Immunolocalization of Trp Proteins in HSG Cells and Submandibular Glands-HSG cells expressing hTrp1␣ were cultured on glass coverslips for 2 days. These cells were rinsed once with PBS (pH 7.5), fixed with 3% formaldehyde in PBS for 30 min, treated with 100 mM glycine-PBS for 30 min, permeabilized with methanol at Ϫ70°C on dry ice for 5 min, and then washed with PBS and blocked with 5% donkey serum for 1 h. The sample was then incubated with the primary antibody, mouse monoclonal anti-HA (Roche Molecular Biochemicals) or rabbit polyclonal anti-Trp1 diluted 1:150 or 1:100, respectively, and washed three times. The samples were then incubated with FITCconjugated anti-rabbit IgG or anti-mouse IgG at a 1:150 dilution for 1 h, washed with PBS three times, mounted on glass slides, and used for confocal microscopy.
Preparation of frozen sections was performed essentially as described previously (32). Submandibular glands were quickly removed and fixed in prechilled (Ϫ20°C) methanol/Me 2 SO (4:1, v/v) at Ϫ20°C overnight. Glands were washed for 1 h at room temperature in PBS (pH 7.5) and then transferred to 20% sucrose in PBS at 4°C overnight. After freezing at Ϫ50°C for 3 min in isopentane, glands were placed in Tissue Tek intermediate sized cryomolds (Miles Inc., Elkhart, IN), which were on dry ice and had just received Tissue Tek OCT mounting medium (Miles Inc., Elkhart, IN). Frozen samples were cryosectioned in 8-m sections, which were placed on glass slides at Ϫ20°C. Sections were stored at Ϫ70°C until examined.
Sections were rehydrated for 5 min in PBS and used either for immunocytochemistry or confocal microscopy. Immunocytochemistry was performed by using the HistoStain SP kit (Zymed Laboratories Inc.). Samples for confocal microscopy were blocked for 20 min with 5% donkey serum (Jackson ImmunoResearch) in PBS containing 0.2% bovine serum albumin (BSA; fraction V; Calbiochem) and then incubated for 1 h with the primary antibody (1:150) in PBS containing 0.2% BSA.
Sections were then washed for 5 min in PBS containing 0.2% BSA and incubated for 20 min with a FITC-conjugated goat anti-rabbit secondary antibody (Jackson ImmunoResearch) diluted 1:150 in PBS containing 0.2% BSA. The secondary antibody was then removed by washing the sections twice in PBS containing 0.2% BSA (5 min/wash) followed by two washes with PBS (5 min each). Each immunostained section received a drop of Citifluor (Ted Pella, Redding, CA), was mounted on a coverslip, and sealed. Samples were stored in the dark at 4°C until examined.
Confocal images were collected using an MRC 1024 krypton/argon laser-scanning confocal system (Bio-Rad) equipped with a Nikon Optiphot II photomicroscope (Melville, NY). Images of control and experimental samples were obtained using identical conditions (laser intensities, gain, iris aperture, black level, scan speed). Images were collected in the x-y plane using a ϫ 60 oil immersion PlanApo objective (Nikon, Japan) with a confocal aperture of 2 mm. The entire series of images was then collected into a single focused image using the Confocal Assistant software supplied by the manufacturer. A Codonics NP-1600 Photographic Network Printer (Middleton Heights, OH) was used to print the final single focused image.
Patch Clamp Measurements-A piece of coverslip (0.5 ϫ 0.5 mm) with cells was placed in the perfusion chamber (Warner Instrument Corp., Hamden, CT). Perfusion, at approximately 5 ml/min, was achieved by gravity-fed plastic tubes. A vacuum line continuously removed the bath solution. Complete solution changes were achieved within 10 -15 s. The standard extracellular solution for measuring Ca 2ϩ -activated K ϩ channel current (K Ca ) contained 145 mM NaCl, 5 mM KCl, 1 mM MgCl 2 , 1 mM CaCl 2 , 10 mM glucose, 0.2 mM EGTA, and 10 mM HEPES, pH 7.4. The pipette was filled with 150 mM KCl, 2 mM MgCl 2 , 1 mM ATP, 10 mM HEPES, pH 7.2. Any changes in the composition of the pipette and external solutions are given under "Results." Patch clamp in a whole cell configuration was performed on single HSG cells attached to coverslips using the standard patch clamp technique described previously (9,29). Membrane currents were measured with an Axopatch 200A amplifier in conjunction with pClamp 6.1 software and a Digidata 1200 A/D converter (Axon Instruments, Foster City, CA). The currents were filtered at 2 kHz (low pass bessel filter) and sampled with an interval of 10 ms. The currents were digitized and recorded directly onto the hard drive of a Dell Pentium computer.
[Ca 2ϩ ] i Measurements-The culture medium was removed, and the cells were washed and incubated in medium containing 1 M Fura-2/AM (Molecular Probes, Inc., Eugene, OR) for 45 min at 37°C. Fura-2 fluorescence in single cells was measured at excitation wavelengths of 340 and 380, with emission at 510 nm, using a SLM 8000/DMX 100 spectrofluorimeter attached to an inverted Nikon Diaphot inverted microscope with a Fluor ϫ 40 oil immersion objective. Images were acquired using an enhanced CCD camera (CCD-72, MTI) and the Image-1 software (Universal Imaging Corp., PA). Analog plots of the fluorescence in single cells are shown. The 340/380 nm ratio of Fura-2 fluorescence have been used to represent [Ca 2ϩ ] i . Other details of the experiments are given under "Results" and in the figure legends. All experiments were performed at room temperature. Fig. 1A shows RT-PCR products amplified from HSG cell RNA by using htrp1-specific primers for the conserved (400-base pair region) of trp genes (30, 33) (lane 1). RT-PCR products from rat submandibular gland (lane 2) and rat brain (lane 3) are also shown (rat-specific primer sequences used were described previously; see Ref. 30). Lane 4 shows DNA size markers. Sequencing the product from HSG cells confirmed the presence of htrp1 (data not shown). Using htrp3-specific primers and RT-PCR, a similar region in htrp3 was also amplified from HSG cell mRNA and sequenced (data not shown). Fig. 1B shows the RT-PCR products representing the 2.3-kilobase pair 3Ј-end of trp1 (lane 1, HSG cells; lane 2, rat submandibular glands; lane 3, size markers). Similar 2.3-kilobase pair products were amplified from rat parotid gland (data not shown) and brain mRNA (30).

Expression of Trp1 in HSG Cells-
Endogenous Trp1 protein expression in HSG cells was determined using the Trp1-specific antibody we have previously described (30). Trp1 protein (shown by an arrow) was detected in HSG cell plasma membranes by Western blotting (Fig. 1C,  lane 1). Since antibody to Trp3 is presently unavailable, the presence of endogenous Trp3 in HSG cells could not be determined.
Stable Transfection of HSG Cells with htrp1␣ cDNA-To examine the role of Trp1 protein in salivary gland cell storeoperated Ca 2ϩ influx, HSG cells were stably transfected with the HA-tagged trp1␣ cDNA. The transfected cells were then used for detection of the Trp1 protein and measurement of store-operated Ca 2ϩ influx. Fig. 1C, lane 2, shows that the plasma membrane fraction of transfected cells contains a higher (about 2-fold) level of Trp1 protein than nontransfected cells (lane 1; detected using anti-Trp1 antibody). The localization of the expressed HA-tagged hTrp1␣ protein was further examined by confocal microscopy. Fig. 1D shows an extended focus (stacked) image of FITC fluorescence using an anti-HA antibody. Consistent with the data obtained with the immunoblotting technique, considerable reactivity was detected in the transfected cells, and fluorescence was mainly associated with the plasma membrane and sub-plasma membrane region of the cell. A similar pattern was observed when the anti-Trp1 anti- The lighter, lower band is nonspecific, since it is also detected after incubation of the primary antibody with the Trp1 peptide. D, localization of the Trp1 protein in transfected HSG cells. Cells were fixed and treated, as described under "Experimental Procedures," and reactivity to anti-HA antibody was determined using an FITC-linked anti-mouse IgG. Fluorescence was detected in the plasma membrane region of the transfected cells, indicated by the arrows. E, fluorescence detected when the primary antibody was preincubated with the HA peptide.

Trp1 Protein Expression and SOC
body was used (data not shown). The fluorescence intensity was dramatically diminished when the anti-HA antibody was preincubated with the HA peptide (Fig. 1E).
Thapsigargin-stimulated Ca 2ϩ Influx in htrp1␣-transfected HSG Cells- Fig. 2A shows Fura-2 fluorescence measurements in control (HSG cell) and htrp1␣-transfected (hTrp1 cell) cells. The 340/380 nm fluorescence ratio has been used to represent changes in cytosolic [Ca 2ϩ ] ([Ca 2ϩ ] i ). The experimental protocol used involved stimulating cells with thapsigargin (Tg test) in a Ca 2ϩ -free medium to detect internal Ca 2ϩ release, following which 10 mM Ca 2ϩ was added to the cell medium to detect Ca 2ϩ influx. This experimental protocol was employed, since we have previously shown that SOC in HSG cells is strongly subject to feedback inhibition by [Ca 2ϩ ] i when treated with thapsigargin in a Ca 2ϩ -containing medium (9).
Thapsigargin induced a rapid release of Ca 2ϩ from intracellular stores in both control and transfected cells, demonstrated by the first transient increase in [Ca 2ϩ ] i . There was no significant difference in this initial increase in [Ca 2ϩ ] i in the two sets of cells (p Ͼ 0.05; number of cells tested was 65 and 90, respectively, for control and transfected cells). When Ca 2ϩ was readded to the cells, a second increase in [Ca 2ϩ ] i was obtained, representing the Ca 2ϩ influx component. The peak of this increase was about 3-fold higher in transfected cells compared with control cells (p Ͻ 0.01; number of cells was 65 and 90 for control and transfected cells, respectively). This increase was transient in both types of cells and decreased to a relatively sustained level that was also consistently higher in the transfected cells. Importantly, Fig. 2B shows that the addition of 10 mM Ca 2ϩ to cells incubated in a Ca 2ϩ -free medium without thapsigargin treatment induced a similar increase in fluorescence in both sets of cells (p Ͼ 0.05, n ϭ 43 for either set). These data demonstrate that the transfected HSG cells, which express higher levels of the Trp1␣ protein, display (i) a significantly higher level of SOC (see average data in Fig. 2C), (ii) similar thapsigargin-stimulated internal Ca 2ϩ release, and (iii) similar basal Ca 2ϩ influx as compared with control cells (see Fig. 2C). The SOC component (i.e. the difference between the peak [Ca 2ϩ ] i increase obtained upon readdition of 10 mM Ca 2ϩ in untreated cells and that obtained in thapsigargin-treated cells) measured in transfected cells was about 3-5-fold higher than in control HSG cells (Fig. 2C).
Thapsigargin-stimulated K Ca Current in htrp1␣-transfected HSG Cells-Since Ca 2ϩ influx is strongly affected by changes in the membrane potential, measurement of K Ca current, performed under voltage-clamped conditions, was also used to assess the increase in SOC seen in htrp1␣-transfected HSG cells. We have previously reported that (i) the K Ca current in

Trp1 Protein Expression and SOC
HSG cells represents underlying changes in [Ca 2ϩ ] i , and (ii) its sustained activation is determined by Ca 2ϩ influx (29). Thus, K Ca activity in these cells is a physiological readout for the Ca 2ϩ entering the cell. The experimental protocol used was similar to that used for [Ca 2ϩ ] i measurements, and the data are shown in Fig. 3. Following the addition of thapsigargin to the cell maintained at 0 mV in a Ca 2ϩ -free medium, a transient increase in the outward current (due to internal Ca 2ϩ release) was detected, which was similar in control (A) and transfected (B) cells. Upon the readdition of 10 mM Ca 2ϩ , a second increase in the current, due to Ca 2ϩ influx, was detected that was significantly higher in the transfected cells. Importantly, and consistent with the [Ca 2ϩ ] i measurements, the addition of Ca 2ϩ to cells incubated for the same length of time in Ca 2ϩ -free medium without thapsigargin treatment induced K Ca currents of similar magnitudes in control and transfected HSG cells (data not shown). Based on the present data and our previous studies, we suggest that the higher level of K Ca current obtained upon the readdition of Ca 2ϩ to transfected cells is due to a higher level of Ca 2ϩ influx in these cells and not due to an increase in the activity of K Ca per se in cells overexpressing hTrp1. Note that two different clones of htrp1␣ cDNA-transfected HSG cells were tested for thapsigargin-stimulated SOC and K Ca , and similar results were obtained.
Characteristics of Thapsigargin-stimulated Ca 2ϩ Influx in htrp1␣-transfected Cells-Previous reports have shown that the Ca 2ϩ influx activity associated with the expression of some trp gene products has characteristics distinct from those of the endogenous SOC in the cells used for the expression (11). To determine whether the activity due to expression of the Trp1␣ protein in HSG cells is similar to that of the endogenous SOC in these cells, we examined the sensitivity of the Ca 2ϩ influx in control and transfected cells to inhibition by divalent and trivalent cations. We have previously shown that Gd 3ϩ and Zn 2ϩ have distinct effects on the SOC in HSG cells; Gd 3ϩ , but not Zn 2ϩ , blocks Ca 2ϩ influx to the same level as La 3ϩ (9,29). Fig.  4A shows that 1 mM Zn 2ϩ was ineffective in blocking the Ca 2ϩ influx in transfected HSG cells. Increase in the Fura-2 fluorescence was similar to that in the absence of added Zn 2ϩ (see Fig.  4B) in 46 control cells and 51 transfected cells. Fig. 4, B and C, shows that Gd 3ϩ and La 3ϩ , respectively, block Ca 2ϩ influx by Ͼ90% in transfected cells. Notably, the expressed Ca 2ϩ influx activity and endogenous SOC activity were inhibited to the same level by the cations (see average data presented in Table  I). Thus, the increased SOC activity associated with Trp1␣ protein expression in HSG cells has characteristics similar to those of the endogenous SOC in these cells.
Stable Transfection of HSG Cells with htrp3 cDNA-HSG cells were stably transfected with htrp3 cDNA using a method similar to that described above for htrp1␣. SOC in these cells was assessed by determining thapsigargin-stimulated internal Ca 2ϩ release and Ca 2ϩ influx using the experimental protocol described above. Fig. 5A shows a representative trace of Fura-2 fluorescence. As seen in htrp1␣-transfected cells, thapsigarginstimulated internal Ca 2ϩ release in htrp3-transfected cells was not significantly different from that in control cells (see Fig.  2A). However, unlike in trp1␣-transfected HSG cells, the Ca 2ϩ influx component measured in thapsigargin-treated htrp3transfected HSG cells was also not different from that in control HSG cells (compare traces in Figs. 2A and 5A). Fig. 5E shows the average data from these experiments. The SOC component (thapsigargin-stimulated Ca 2ϩ influx minus basal Ca 2ϩ influx) in htrp3-transfected HSG cells is significantly lower than that in htrp1␣-transfected HSG cells but not different from that measured in control HSG cells. Fig. 5B shows a Western blot demonstrating the expression of HA-tagged hTrp3 in HSG cell extracts (lane 2). As expected, there is no reaction with the anti-HA antibody in control cells  Fig. 4B). Gd 3ϩ and La 3ϩ decreased influx to the same level in control (see Table I and Refs. 7 and 29) and transfected cells.

Trp1 Protein Expression and SOC
(lane 1). These data clearly show that increasing the levels of hTrp3 in HSG cells does not induce any changes in thapsigargin-stimulated Ca 2ϩ influx. These results are similar to those previously reported showing that htrp3 expression does not increase thapsigargin-stimulated Ca 2ϩ influx in HEK-293, COS, and Chinese hamster ovary cells (22,24).
Expression of Antisense trp1␣ cDNA in HSG Cells-The data presented above provide evidence for an increase in the SOC activity in HSG cells transfected with the htrp1␣, but not htrp3, cDNA. The data also rule out secondary effects due to changes in (i) the membrane potential or (ii) the basal Ca 2ϩ permeability of the cells. To demonstrate more directly that the Trp1 protein is involved in the endogenous SOC mechanism, HSG cells were stably transfected with htrp1␣ cDNA in the antisense direction. Two clones were tested for activity, and similar results were obtained. Fig. 5C is a representative trace showing thapsigargin-stimulated [Ca 2ϩ ] i changes in HSG cells transfected with antisense htrp1␣ cDNA. Thapsigargin-stimulated internal Ca 2ϩ release was not altered in these cells (compare traces in Figs. 5, A and B, and 2A). However, the thapsigargin-stimulated Ca 2ϩ influx (second peak of [Ca 2ϩ ] i increase) was significantly reduced compared with that in control cells (compare with trace in Fig. 2A). The magnitude of [Ca 2ϩ ] i increase upon the readdition of Ca 2ϩ to these cells is similar to that in the basal, unstimulated condition (see Fig. 2B). Average values for the SOC component measured in the HSG cells transfected with antisense htrp1␣ cDNA is shown in Fig. 5E. The SOC component in these cells is significantly lower (p Ͻ 0.05, n ϭ 80) than in the other three groups of cells (i.e. control, htrp1␣-transfected, and htrp3-transfected).
The effect of transfection of HSG cells with antisense htrp1␣ on the levels of endogenous Trp1 was determined. A dramatic decrease in the level of Trp1 protein was detected in plasma membrane fraction isolated from these cells. Fig. 5D shows a representative Western blot using anti-Trp1 antibody (similar results were obtained in three experiments using cells from different passages). Lane 1 shows endogenous hTrp1 in plasma membranes isolated from HSG cells transfected with antisense htrp1␣ cDNA, and lane 2 shows endogenous hTrp1 in control cells. In aggregate, these data strongly demonstrate that the expression of the htrp1␣ cDNA in HSG cells in the antisense direction decreases the level of Trp protein and compromises activation of SOC by thapsigargin. Thus, the data demonstrate for the first time a direct association between SOC and the level of Trp1 in the plasma membrane.
Immunolocalization of the Endogenous Trp1 Protein in Rat Submandibular Gland Cells-The data presented above strongly suggest that Trp1 is involved in the SOC mechanism of HSG cells. Consistent with this suggestion and with the detection of trp1 in rat submandibular glands, Fig. 6 shows that endogenous Trp1 protein is present in this salivary gland. Two methods were used to localize Trp1 in intact rat submandibular gland: immunocytochemistry, using a horseradish peroxidase-linked as secondary antibody (Fig. 6, A and B), and confocal microscopy, using an FITC-linked secondary antibody (Fig. 6, C-E). In both cases, anti-Trp1 antibody (30) was used as the primary antibody. A reddish color was detected in the basolateral region of the acinar cells (shown by the arrow marked a) but not in the luminal regions (A). The intensity of the reaction seen in the ductal cells (shown by the arrow marked d) was higher than that of the acinar cells, although the localization was not as distinct. Since this reactivity was blocked either by preincubating the antibody with the Trp1peptide (B) or in the absence of the secondary antibody (data not shown), the reactivity in the ductal cells does not appear to be nonspecific. Fig. 6, C-E, shows the images obtained by confocal microscopy. Consistent with the immunocytochemistry results, Trp1 protein was detected in the basolateral plasma membrane region of acinar cells (see the arrow marked a in Fig. 6C), while ductal cells (see the arrow marked d in Fig.  6D) showed a relatively stronger reactivity that did not appear to be localized to any specific region. The reactivity detected in the absence of the primary antibody is shown in Fig. 6E. Notably, the store-operated Ca 2ϩ influx activity has been suggested to occur via the basolateral membrane of salivary gland and other exocrine gland acinar cells (3,34). Thus, the localization of the Trp1 protein in intact submandibular gland cells is consistent with its proposed physiological function. DISCUSSION A number of previous studies have suggested that trp gene products are involved in SOC. However, the data presently available do not provide convincing evidence linking the localization and expression of any Trp protein with the SOC function in any nonexcitable cell type or tissue. The present study demonstrates the localization of the Trp1 protein in an intact nonexcitable tissue and functional effects of trp1␣ expression in the same cell type. Importantly, the data show for the first time an association between the level of Trp1 protein and SOC and suggest that Trp1 is a candidate protein for the SOC mechanism in salivary gland cells. HSG cells stably transfected with the htrp1␣ cDNA expressed higher levels of Trp1 and displayed a 3-5-fold increase in the thapsigargin-stimulated Ca 2ϩ influx component. This was unequivocally demonstrated by measuring (i) [Ca 2ϩ ] i changes by microfluorimetry and imaging of Fura-2 fluorescence in single transfected HSG cells and (ii) Ca 2ϩ -activated K ϩ channel current. We have shown that more Ca 2ϩ enters thapsigargin-treated htrp1␣ cDNA-transfected cells than control cells, thus accounting for the higher level of K Ca current measured in these cells. Consistent with this, the sustained level of [Ca 2ϩ ] i in these cells was about 2-fold higher than in nontransfected cells. Further, the data rule out the possibility that the increase in SOC is indirectly induced due to a change in the membrane potential or an increase in the basal Ca 2ϩ permeability of the cell. An important and novel finding of this study is that the characteristics of Ca 2ϩ influx in htrp1␣transfected HSG cells are similar to those of SOC in the nontransfected HSG cells. This is in contrast to previous reports, where the Ca 2ϩ influx activity associated with the expressed trp was shown to be distinct from the endogenous SOC in the cell used for expression (11). We have reported earlier that SOC in HSG cells can be inhibited by Gd 3ϩ but not by Zn 2ϩ (9,29). We now show that SOC activity in transfected cells displays the same differential sensitivity to these cations. Notably, the ac-  Fig. 4. The values given below are the peak increases in Fura-2 fluorescence (340/380 ratios) obtained upon the addition of 10 mM Ca 2ϩ to the external medium after internal Ca 2ϩ store depletion by treatment with thapsigargin in control (HSG) and htrp1␣-transfected cells (Trp1-HSG). The other ions were added prior to the addition of Ca 2ϩ (see Fig. 4).

Trp1 Protein Expression and SOC
tivity associated with htrp1␣ (13) was also inhibited by Gd 3ϩ .
In contrast, the store-operated I CRAC channel found in mast cells was highly sensitive to Zn 2ϩ (8). We suggest that the sensitivity of SOC to various cations might be a useful tool for fingerprinting the type of Ca 2ϩ influx channel present. Further evidence for an association between hTrp1 and SOC was provided by our demonstration that transfection of HSG cells with antisense htrp1␣ cDNA resulted in a decrease in endogenous hTrp1 and a significant attenuation of SOC in these cells. In aggregate, these data strongly suggest that Trp1 is involved in the SOC mechanism of HSG cells. Notably, transfection of HSG cells with htrp3 cDNA and increase in hTrp3 did not induce an increase in thapsigargin-stimulated Ca 2ϩ influx. This is consistent with a number of previous reports that suggest that Trp3 activity is associated with agonist-stimulation of cells and probably regulated by IP 3 (11,22,24) or diacylglycerol (35). An interaction between Trp3 and the IP 3 receptor has been proposed as a possible mechanism for Trp3 activation (26). Thus, Trp3 is more likely to be involved in mediating Ca 2ϩ influx following stimulation of cells by agonists that induce PIP 2 hydrolysis and internal Ca 2ϩ store depletion. The Trp1associated SOC activity we have described above in HSG cells is activated by internal Ca 2ϩ store depletion in the absence of PIP 2 turnover or an increase in the level of IP 3 . Although a role for IP 3 and the IP 3 receptor in Trp1 activation cannot be presently excluded, it is possible that cells might have an IP 3independent SOC mechanism that (i) serves a more "housekeeping" role to ensure maintenance of the internal Ca 2ϩ stores or (ii) achieves refill when store depletion involves an IP 3independent process. However, more detailed studies are required to determine the specific physiological function of these two Trp proteins in HSG cells.
Localization of trp transcripts has been previously shown by  a and d, respectively). Reactivity in the absence of primary antibody is shown in E, and reactivity after incubation of anti-Trp1 antibody with the peptide is shown in B. in situ hybridization in the brain (36,37). More recently, rat Trp2 protein was shown to be localized exclusively in rat brain VNO neurons and sensory villi (21). The present data describe for the first time the localization of a Trp protein in an intact nonexcitable tissue. By using a Trp1-specific antibody that we have previously characterized (30), we have shown here that endogenous Trp1 protein is localized in the basolateral plasma membrane region of submandibular gland acinar cells. In ductal cells, the localization appears to be less distinct, since reactivity against the antibody is seen on both the apical and basolateral membrane regions. The present findings are consistent with our previous findings of trp1 homologous sequences in rat submandibular gland RNA (30). Notably, the proposed route for Ca 2ϩ entry in salivary gland acinar cells is via the basolateral plasma membrane (3). Although, presently, there are no data to demonstrate the route of Ca 2ϩ entry in ductal cells, these cells appear to have a more sustained level of Ca 2ϩ influx than the acinar cells (28). Thus, the localization of Trp1 protein in submandibular gland cells described above is consistent with its proposed function (i.e. a Ca 2ϩ influx pathway in the plasma membrane). However, further studies are required to determine the role of Trp1 in salivary gland function.
As mentioned above, the function of the trp gene product(s) has not yet been clearly established. With regard to studies involving trp1 genes, the trp1␣ (long form) has been expressed in COS-M6 cells and Sf9 insect cells (14,15). In COS-M6 cells, trp1␣ expression induced a relatively small increase in the Ca 2ϩ influx activity associated with carbachol stimulation. However, this study did not assess Tg-stimulated Ca 2ϩ influx, a primary criterion used to define SOC. Transient expression of trp1␣ in Sf9 cells induced nonselective constitutively activated cation channel activity, which did not appear to be sensitive to internal Ca 2ϩ store depletion. In contrast, transient expression of htrp1␤ (the alternatively spliced, short form of the gene) in Chinese hamster ovary and Sf9 cells induced a nonselective cation channel activity and an increase in the Ca 2ϩ influx component in response to Tg treatment. However, since the Chinese hamster ovary cells used in this study did not display an endogenous SOC activity (13), it is difficult to ascribe a function to trp1␤ in this cell type. Overall, these previous studies were more consistent with a role for trp1␤, rather than trp1␣, in SOC. However, here we have demonstrated that expression of trp1␣ increased the SOC mechanism of salivary gland cells. Thus, both trp1␣ and trp1␤ appear to be capable of regulating SOC.
The exact reasons for the conflicting data and different effects seen following expression of the various trp genes are not yet understood. It is possible that distinct and tissue-specific regulatory factors are involved in regulating the function of various Trp proteins. Thus, the activity of the expressed protein will probably be determined by the endogenous SOC and associated regulatory mechanism(s) present in the cells used for the functional expression. It has been previously proposed that Trp proteins might form multimeric complexes, either homomers or heteromers with other Trp proteins, or with other as yet unidentified proteins (10). For example, interactions have been suggested between hTrp1 and hTrp3 (38). It has also been recently suggested that hTrp3 activity is regulated via an interaction with the IP 3 receptor (26). Thus, such interactions and other possible regulatory mechanisms could also determine the characteristics and regulation of expressed trp activity in various cells. Clearly, further studies are required for a more complete understanding of the function of Trp proteins. An important aspect of these studies will be to determine the presence and localization of various Trp proteins in different tissues and cell types.
Based on the rather ubiquitous distribution of the SOC mechanism in various cell types, it is reasonable to hypothesize that the molecules involved in this mechanism would also be similarly widespread in distribution. Interestingly, data with htrp1 and rtrp1 demonstrate that trp1 is more widely expressed than the other trp genes. In addition to brain and heart, both rtrp1 and htrp1 transcripts are present in a number of nonexcitable tissues, such as kidney, lung, and colon (30,33). trp1 homologous sequences have also been identified in cells such as rat and human pulmonary endothelial cells, Xenopus oocytes, and megakaryotes. On the other hand, htrp3, trp5, and rtrp2 transcripts have been predominantly found in the brain, mtrp2 was abundant in the testis, btrp4 was abundant in the adrenal gland and testis, and mtrp6 was abundant in the lung. Thus, the functional differences between various trp gene products might be related to their tissue-specific distribution. Whether these molecules fulfill a specific physiological role in these tissues remains to be established.
In summary, this study demonstrates that Trp1 protein is endogenously present in the plasma membrane region of HSG cells and of acinar and ductal cells in the intact salivary gland. Further, stable expression of trp1␣ cDNA in HSG cells induced an approximately 2-fold increase in the levels of the Trp1 protein and a 3-5-fold increase in SOC. Importantly, the increased Ca 2ϩ influx activity associated with trp1␣ expression displayed characteristics similar to the endogenous SOC in HSG cells. Thus, the data presented demonstrate that expression of htrp1␣, like trp1␤, also results in an increase in SOC activity. In addition, we have shown that (i) expression of the trp1␣ cDNA in the antisense direction significantly reduced the endogenous SOC activity in HSG cells and (ii) expression of htrp3 cDNA did not induce any change in SOC. In aggregate, these findings provide evidence for the involvement of the Trp1 protein in the SOC mechanism in HSG cells, thus suggesting a molecular basis for this process in this salivary cell line. Notably, these are the first data to directly correlate changes in the level of the Trp1 protein with changes in SOC. Further studies will be required to address the important question of whether the Trp1␣ protein forms the SOC channel and how it is regulated by internal Ca 2ϩ store depletion.
We suggest that subscribers photocopy these corrections and insert the photocopies at the appropriate places where the article to be corrected originally appeared. Authors are urged to introduce these corrections into any reprints they distribute. Secondary (abstract) services are urged to carry notice of these corrections as prominently as they carried the original abstracts.