Gene expression profiles in HEK-293 cells with low or high store-operated calcium entry: can regulatory as well as regulated genes be identified?
Tatiana K. Zagranichnaya,
Xiaoyan Wu,
Arpad M. Danos and
Mitchel L. Villereal
Department of Neurobiology, Pharmacology and Physiology, The University of Chicago, Chicago, Illinois
 |
ABSTRACT
|
---|
Gene expression profiles were generated using cDNA microarray technology for clones of human embryonic kidney (HEK)-293 cells selected to have either high or low levels of store-operated Ca2+ entry (SOCE). For five high clones, three low clones, and control HEK-293 cells, duplicate Affymetrix U133A human gene arrays were run after extraction of total RNA from cells growing in the presence of serum. Of the
22,000 genes represented on the microarray, 58 genes had readings at least twofold higher, while 32 genes had readings at least twofold lower, in all five high SOCE clones compared with control HEK-293 cells. In the low SOCE clones, 92 genes had readings at least twofold higher, while 58 genes had readings at least twofold lower, than in HEK-293 cells. Microarray results were confirmed for 18 selected genes by real-time RT-PCR analysis; for six of those genes, predicted changes in the low SOCE clone were confirmed by an alternative method, monitoring mRNA levels in HEK-293 with SOCE decreased by expression of small interfering (si)RNA to canonical transient receptor potential protein-1. Genes regulated by SOCE are involved in signal transduction, transcription, apoptosis, metabolism, and membrane transport. These data provide insight into the physiological role of SOCE. In addition, a potential regulator of SOCE, insulin receptor substrate (IRS)-2, has been identified. A reduction of IRS-2 levels by siRNA methods in two high clones dramatically reduced SOCE, whereas overexpression of IRS-2 in a low SOCE clone elevated SOCE.
cDNA microarray; fluorescence-activated cell sorting analysis; thapsigargin; insulin receptor substrate-2; small interfering RNA; real-time PCR; canonical transient receptor potential protein-1
 |
INTRODUCTION
|
---|
MANY PLASMA MEMBRANE RECEPTORS utilize Ca2+ as a second messenger to initiate downstream physiological processes (4, 5, 8, 54). Activation of these receptors generally results in a biphasic Ca2+ response involving an initial release of internal Ca2+ stores, followed by Ca2+ entry through receptor-operated or capacitative Ca2+ entry channels (43), also called store-operated channels (SOCs). Although progress has been made in identifying proteins that assemble to form SOCs, little is known about the downstream physiological consequences of Ca2+ entry via these channels. Recent studies have indicated that activation of store-operated Ca2+ entry (SOCE), via depletion of stores with inhibitors of sarco(endo)plasmic reticulum Ca2+-ATPase pumps, can lead to the regulation of a handful of specifically monitored genes such as Nur77 (33), c-fos and grp78 (23), and pip92 (10); however, these studies do not provide an insight into how widespread the involvement of SOCE is in regulating mRNA levels. Several recent investigations have utilized cDNA microarrays to broaden the range of mRNA expression that can be monitored. One such study in T lymphocytes utilized cDNA microarrays (representing 7,396 cDNA clones) to investigate the gene regulation in response to ionomycin and phorbol ester treatment of control cells, or cells thought to have a defect in SOCE (18). Although this study greatly advanced our understanding of the role of SOCE in regulating gene expression, there are several reasons that additional studies are required. First, a significantly larger number of genes are represented on more recent cDNA microarrays, allowing a fuller description of the expression profile. Second, the lymphocyte is a specialized cell in which induced gene expression is heavily directed toward producing cytokines to regulate the population of immune cells, and therefore the expression profile is possibly quite different from the gene expression profile elicited in other types of cells. Third, while ionomycin and phorbol ester approximate a physiological stimulus in lymphocytes, an ionomycin stimulus in most other cell types fully depletes the endoplasmic reticulum (ER) stores to the point of blocking ER protein processing, resulting in the activation of the ER stress response and ultimately activation of apoptosis. The peripheral effects of fully depleting internal Ca2+ stores can be seen in a gene expression profiling study of RBL-2H3 mast cells stimulated with 2,5-di(tert-butyl)-1,4-hydroquinone (DTBHQ), another inhibitor of the ER Ca2+-ATPase, where DTBHQ upregulated many more stress-inducible genes than cross-linking of IgE receptors, a more physiological stimulus (38). Therefore, the gene expression profiles produced by thapsigargin and ionomycin may vary significantly from those produced by activation of SOCE by receptor agonists. Another recent study utilized cDNA arrays (Atlas 1.2 mouse array from Clontech, which represents 1,200 genes) to demonstrate that inhibition of serum-stimulated 3T3 cells with SKF-96365, an inhibitor of SOCs, leads to the upregulation, or downregulation, of 29 genes (27). However, the interpretation of these SKF-96365 experiments is complicated by the lack of specificity of this reagent. Recent papers report SKF-96365 inhibition of Na+ channels (26), K+ channels (49), maitotoxin-induced Ca2+ entry (14, 51), and facilitation of nicotinic receptor desensitization (25) in addition to the more widely recognized inhibition of SOCs. One recent study used SKF-96365 to distinguish between receptor-operated channels (ROCs) and SOCs, but as an inhibitor of ROCs, not SOCs (36). Given the limitations of the previous investigations into the role of SOCE in regulating mRNA levels, we decided to take an alternative approach to investigate this question.
Previously, we reported that a clonal variation of SOCE exists in the human embryonic kidney (HEK)-293 cell population (2). In the current study, we took advantage of that clonal variation and selected HEK-293 cell clones that varied in their levels of SOCE. We used these clonal populations to investigate the effect of low or high levels of SOCE on the gene expression profile for cells maintained in their normal growth environment. The mRNA expression profiles for the various clones were evaluated by utilizing Affymetrix cDNA microarrays. Comparisons of mRNA profiles of five clones high in SOCE, the parent HEK-293 cell population, and three clones low in SOCE provide valuable information about which genes are regulated by changes in SOCE for cells in their normal growth environment. In addition, this study provides important information on potential upstream regulators of SOCE, as evidenced by our results demonstrating that reduction of the elevated insulin receptor substrate (IRS)-2 levels in two high SOCE clones by small interfering (si)RNA methods, or overexpression of IRS-2 in a low SOCE clone, alters their levels of SOCE.
 |
MATERIALS AND METHODS
|
---|
Materials
Fura-2 free acid, fura-2 AM, indo-1, and Pluronic F-127 were purchased from Molecular Probes; thapsigargin was purchased from LC Laboratory. Cyclopiazonic acid (CPA) was purchased from Calbiochem. Hanks balanced salt solution (HBSS); Ca2+-free, Mg2+-free, and HCO3-free HBSS; DMEM; penicillin-streptomycin, L-glutamine, and trypsin-EDTA were purchased from Life Technologies (GIBCO BRL). Chelex-100 came from Bio-Rad.
Cell Culture
HEK-293 cells were cultured in DMEM supplemented with 10% FBS, 50 U/ml penicillin, 50 µg/ml streptomycin, and 2 mM glutamine. Cells were grown in an incubator at 37°C with humidified 5% CO2 and 95% air.
Cell Sorting and Clone Selection
HEK-293 cells were grown to confluency on 15-cm plates, and then the cells were loaded with Fluo-3 and Fura Red. In the absence of Ca2+, cells were stimulated with 10 µM CPA for 10 min to deplete intracellular Ca2+ stores. Cells were then removed from the dish by rinsing in EDTA medium, transferred to a 50-ml sterile centrifuge tube, and washed with nominally Ca2+-free HBSS. After sufficient time for the return of cytosolic Ca2+ to basal levels, intracellular Ca2+ concentration ([Ca2+]i) was measured by fluorescence-activated cell sorting (FACS) before and after the addition of 1.8 mM Ca2+. The sorted population of cells with high SOCE (or low SOCE) was plated on six-well plates and grown until individual cell clones were visible. Cell clones were then selected and expanded into clonal cell lines. Thirty-one high SOCE clonal cell lines and twenty-four low SOCE clonal cell lines were generated and tested to confirm that they were high or low in SOCE by monitoring thapsigargin-stimulated Ba2+ entry. Five high SOCE clones and three low SOCE clones were selected for use on the basis of their level of SOCE and low Ba2+ leak influx before thapsigargin stimulation.
Ca2+ Imaging
[Ca2+]i was measured in cells loaded with the fluorescent indicator fura-2. Cells were plated onto 25-mm coverslips 1 day before the experiment. On the next morning, cells were washed twice with HEPES-buffered HBSS, loaded for 30 min with 5 µM fura-2 AM in HBSS supplemented with 1 mg/ml BSA + 0.025% Pluronic F-127, and then unloaded in HBSS for another 30 min. The coverslips were mounted as the bottom of a chamber, and cells in the chamber were perfused via an eight-channel syringe system. A suction pipette maintained a constant volume of solution (
0.5 ml) in the chamber. An InCyt IM2 dual-wavelength fluorescence imaging system (Intracellular Imaging, Cincinnati, OH) was used to measure [Ca2+]i during the experiment, as previously described (57). In short, Ba2+ influx was measured before (to determine leak flux) and after (to determine total flux) store depletion by thapsigargin. SOCE was defined as the difference between total and leak Ba2+ influxes. Nominally Ca2+-free HBSS was prepared by stirring Ca2+-free, Mg2+-free, and HCO3-free HBSS with Chelex-100 beads. After the Chelex-100 beads were filtered out, MgCl2 was added to a final concentration of 1 mM.
Total RNA Isolation
Total RNA was isolated from HEK-293 cells and clones by use of the RNeasy Mini Kit (Qiagen) and treated with DNase I (Invitrogen). The RNA sample was additionally purified by ethanol precipitation, and its concentration was determined by measuring absorbance at 260 nm.
cDNA Microarray Analysis
The five high SOCE clones, three low SOCE clones, and HEK-293 cells (to serve as the general control population) were plated onto 15-cm plates 1 day before RNA purification. Two separate experiments were run. In the first experiment, microarrays were run on duplicate samples from five high clones (H1, H15, H24, H36, and H39), one low clone (L3), and the HEK control cells. In the second experiment, microarrays were run on duplicate samples from two low clones (L28 and L29), one high clone (H36), and the HEK control cells. Cells were taken from their growth medium, and total RNA was immediately purified. The quality of the RNA was evaluated by agarose gel electrophoresis. A quantity in excess of 20 µg of RNA from each clone (duplicate sample) was submitted to our Functional Genomics Core Facility for microarray analysis. To confirm the integrity of the RNA, samples were applied on an Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA), and the purity and the concentration were determined with a GeneSpec III (Miraibio). The target preparation protocol followed the Affymetrix GeneChip Expression Analysis Manual (Santa Clara, CA). Briefly, 10 µg of total RNA were used to synthesize double-stranded cDNA using the Superscript Choice System (Life Technologies). First-strand cDNA synthesis was primed with a T7-(dT24) oligonucleotide. From the phase-log gel-purified cDNA, biotin-labeled antisense cRNA was synthesized with BioArray High Yield RNA Transcript Labeling Kit (Enzo Diagnostics, Farmingdale, NY). After precipitation with 4 M lithium chloride, 12 µg of fragmented cRNA were hybridized to human 133A arrays for 16 h at 45°C and 60 rpm in an Affymetrix Hybridization Oven 640. The arrays were washed and stained with streptavidin phycoerythrin in Affymetrix Fluidics Station 400, using the Affymetrix GeneChip protocol, and then scanned using the Affymetrix Agilent GeneArray Scanner. The acquisition and initial quantification of array images were performed using the Affymetrix Microarray Suite Version 5.0 (MAS 5.0) with the default analytic parameters. Complete microarray expression data are available at the National Center for Biotechnology Information (NCBI) Gene Expression Omnibus (GEO) database (GEO submissions GSE1309, GSM2123421235, GSM2125821264, GSM2137821382, GSM21385, GSM21391, GSM2139321396, GSM2140021401) at http://www.ncbi.nlm.nih.gov/geo. Subsequent data analyses were performed by our laboratory.
The Functional Genomics Core provided us with two types of data output, each contained in a separate Excel file. The first file contained normalized fluorescence values and an absolute call for presence or absence of a transcript (see Supplemental Table S1; available at the Physiological Genomics web site)1 The calls were present (P), absent (A), or marginal (M) to reflect whether the particular gene was expressed on the basis of a complex algorithm that weighs the number of matched vs. mismatched probes that are positive for a particular gene. Supplemental Table S2 contained comparative data; for each gene, a ratio of the normalized fluorescence value for each experimental case (each high or low clone) vs. the normalized fluorescence value for its HEK cell control was calculated. In the absolute data set, all genes with fluorescence values less than 200, both in the control HEK samples and in the high or low clones, were identified; this list of genes was matched against the Excel file containing the comparative data, and those genes were eliminated from further consideration. From this modified comparative file, a new Excel file was generated that contained the comparative data for only the five high SOCE clones. With duplicate values for each clone being individually compared with the duplicate values for the control (HEK cells), four comparative ratios for each clone were generated, each accompanied by an identifier for increase (I), decrease (D), or no significant change (NC). A Countif routine was run within Excel to total the number of I or D identifiers present within a given row (for a given gene). Thus, if the Countif routine gave a value of 20 Is for a given row, this would mean that this gene increased in all four comparisons for each of the five high clones. The file was then sorted based on the value within the "Countif column," and all genes containing a "Countif value" above 16 were selected and pasted into a new file. A Countif value of 16 almost always meant that the gene had increased for all four comparisons in four of the five clones. This data set was then sorted based on the ratio values in one of the high clone columns. Any gene with a value above 1 or below 1 (values were based on a log2 scale, so a ratio >1 meant the level had increased at least 2-fold) for at least 16 of the 20 comparisons was selected and pasted into a new file. Each of these genes was then checked against the data for the low clone (L3) run in this experiment, and genes were maintained on the list only if their low clone levels either did not change dramatically or changed in the opposite direction seen in the high clones.
For the low clones, we report the genes from the second experiment (see Supplemental Tables S3 and S4) that change in both the L28 and L29 clones compared with the HEK control cells, but either do not dramatically change or change in the opposite direction in the high clone (H36) run in this experiment.
PCR Primers
PCR primers for the genes chosen for real-time RT-PCR were designed based on published sequences in GenBank.
RT-PCR
First-strand cDNA was prepared from 1 µg of total RNA, using SuperScript III RNase H RT (Invitrogen) and 1 µg of oligo(dT). The mRNA samples were denatured at 65°C for 5 min. Reverse transcription was performed at 50°C for 55 min and was stopped by heating samples at 75°C for 10 min.
Quantitative Real-Time RT-PCR
Real-time PCR was performed on the ABI Prism 7700 Sequence Detection System using SYBR7 Green PCR Core Reagents (Applied Biosystems) and cDNA synthesized as described above. PCR was performed using the kit protocol in a 25-µl reaction volume. The integrity of the RT-PCR products was confirmed by melting-curve analysis. Melting curves for each primer pair showed one specific signal. The amount of PCR products in parental HEK-293 cells or in clones H36 and L28 was calculated in reference to the individual calibration curves based on cDNA obtained from parental HEK-293 cells.
Western Blotting
Cells were grown on 10-cm dishes under the conditions described above. Cells were lysed in modified radioimmunoprecipitation (RIPA) buffer (10 mM Tris·HCl, pH 7.5, 500 mM NaCl, 0.1% SDS, 1% NP-40, 1% Na-deoxycholate, 2 mM EDTA, 2 mM Na2VO4, 2 mM Na4P2O7, 2 mM NaF). The lysates were clarified by centrifugation, and protein concentration was measured by a bicinchoninic acid (BCA) kit (Pierce). Total protein extract (50 µg) was applied on 8% SDS-PAGE (16 cm x 16 cm gels) and run overnight. The proteins were transferred onto an Immobilon-P membrane (Millipore), and the uniformity of protein transfer for all the lanes was evaluated by reversibly staining with BLOT-Fast-Stain (Geno Technology). After 1 h of blocking, membranes were treated with monoclonal anti-IRS-2 antibodies (Upstate) at a dilution of 1:1,000. Membranes were washed 4x 10 min with Tris-buffered saline containing 0.1% Tween 20 (TBS-T), incubated for 30 min at room temperature with secondary anti-mouse antibody (1:10,000 in TBS-T), washed under the same conditions, and developed with SuperSignal West Pico Chemiluminescent Substrate (Pierce) for a suitable time so as not to saturate the film. The films were digitized on a flatbed scanner, and the relative spot intensities were determined in Photoshop 6.0. The images were inverted, the bands were outlined, and the average gray level and number of pixels in the spot were obtained from the histogram function. The product of the average gray level value and the number of pixels was used to represent the integrated signal in the band. Each Western blot was repeated at least three times using different cell lysates.
siRNA Constructs
For human IRS-2 (GenBank no. AF073310), potential siRNA target sites (19 nucleotides in length) were chosen. The location of the selected IRS-2 gene sequence is 573591. The potential target sites were compared with the human genome database by using BLAST (http://www.ncbi.nlm.nih.gov/BLAST), and any target sequences with homology to other coding sequences were eliminated from consideration. Hairpin siRNA template oligonucleotide design was done by entering siRNA target sequences into the web-based insert design tool at the following address: http://www.ambion.com/techlib/misc/psilencer_converter.html. Then, two complementary oligonucleotides (forward 5'-GATCCCGCCTCAACAACAACAACAACTTCAAGAGAGTTGTTGTTGTTGTTGAGGTTTTTTGGAAA-3' and reverse 5'-AGCTTTTCCAAAAAACCTCAACAACAACAACAACTCTCTTGAAGTTGTTGTTGTTGTTGAGGCGG-3') were synthesized, annealed, and ligated into the linearized pSilencer 3.1 H1 neo vector (Ambion). All procedures were performed as directed by the manufacturers instruction manual (Ambion). The inserts were sequenced to confirm that there were no unwanted mutations.
Transfection
Cells were grown in 75-cm2 flasks to 60% confluency and transfected by use of PerFectin Transfection reagent (Gene Therapy Systems). For transient transfection experiments (IRS-2 overexpression), cells were used 48 h after transfection. For stable transfection experiments (siRNA expression), cells were transfected and later treated with 400 µg/ml G418. Cells that survived after 2 wk were collected, and this population of cells was used for future experiments.
 |
RESULTS
|
---|
Selection of HEK-293 Cell Clones That Have Low or High Levels of SOCE
In a previous study, we reported that HEK-293 cells plated at clonal density demonstrate considerable clone-to-clone variation of SOCE (2). This clonal variation was utilized to select cell populations that had either high or low levels of SOCE. To screen for clones on a high throughput basis, HEK-293 cells were grown on 15-cm plates, loaded with Fluo-3 and Fura Red, and removed from their plates in EDTA-containing medium. They were washed by centrifugation and resuspended in Ca2+-free HBSS. An aliquot of these cells was aspirated into the FACS machine, and a baseline for intracellular Ca2+ was established. The aspiration of the sample was interrupted to add Ca2+, and then the aspiration was resumed. The data in Fig. 1A show that there is no significant change in [Ca2+]i when Ca2+ is added back to control cells. Another aliquot of cells was incubated for 10 min in 10 µM CPA in Ca2+-free HBSS, and the cells were removed from the dish. The data in Fig. 1B illustrate that, after depletion of internal Ca2+ stores, the addition of external Ca2+ resulted in a dramatic elevation of [Ca2+]i. A series of similar experiments was run, and the cells were sorted on the basis of the sorting profile shown in Fig. 1C. Cells in the low end of the profile, and in other runs the cells in the high end of the profile, were collected in sterile centrifuge tubes. This provided us with two populations of cells, one enriched for cells with low levels of CPA-stimulated Ca2+ entry and one enriched for cells with high levels of CPA-stimulated Ca2+ entry.

View larger version (26K):
[in this window]
[in a new window]
|
Fig. 1. Fluorescence-activated cell sorting (FACS) of human embryonic kidney (HEK)-293 cells based on Ca2+ entry levels. HEK-293 cells were grown to confluency on 15-cm plates. Cells were loaded with Fluo-3 and Fura Red, removed from the dish by rinsing in EDTA medium, transferred to a 50-ml tube, and washed in Ca2+-free HBSS. Cells were aspirated by FACS and excited at 488 nM, and the emission was monitored at 530 and 585 nM. The ratio of 530-to-585 nM emission intensity (FL1/FL2) was recorded as a measure of the Ca2+ level, the aspiration was interrupted to add 1.8 mM Ca2+, and the aspiration was resumed (A). In a parallel experiment, cells were stimulated with 10 µM cyclopiazonic acid (CPA) in Ca2+-free HBSS for 10 min before their removal from the dish. The cells were then treated as described above (B). Ca2+ entry into CPA-treated cells was initiated, and then the cells were sorted by FACS. A typical sorting profile is shown to illustrate the fractions collected for either low store-operated Ca2+ entry (SOCE) cells or high SOCE cells (C).
|
|
These populations of cells were plated onto six-well plates at clonal densities, and the cells were fed on a regular basis until individual macroscopic cell clones could be observed. Individual cell clones were selected and grown up for analysis of SOCE on a Ca2+ imaging rig. Thirty-one potentially high SOCE clones were harvested and rescreened to confirm high levels of SOCE. Because the initial high throughput screen was on the basis of changes in cytosolic Ca2+ levels, it seemed possible that some clones were selected because they had low Ca2+ pump rates rather than high levels of SOCE. Thus, for the second screen, thapsigargin-stimulated Ba2+ entry was monitored, as Ba2+ is not pumped by the Ca2+-ATPases (29, 48), and therefore observed changes would result from changes in entry rates. An initial incubation in HBSS allowed a baseline to be established, and then cells were incubated in Ca2+-free HBSS + 2 mM Ba2+ to measure the Ba2+ leak before store depletion. The cells were then incubated in Ca2+-free HBSS, and 1 µM thapsigargin was added to deplete intracellular stores. After the return of the cytosolic Ca2+ to basal levels, Ba2+ was added, and the slope of Ba2+ entry was measured. The SOCE is defined as the difference between the Ba2+ entry after thapsigargin stimulation and the Ba2+ leak entry before thapsigargin stimulation. This approach has been used in several of our previous papers, and time courses for Ca2+ release and Ba2+ influx can be viewed there (3, 50, 57). Because we were interested in clones with high SOCE, we discarded clones with high leak rates, regardless of whether they also had high thapsigargin-stimulated Ba2+ entry.
Of the 31 potentially high clones rescreened by image analysis, 13 clones appeared to be authentic high SOCE clones. Of the 24 potentially low clones rescreened by image analysis, 5 clones appeared to be authentic low SOCE clones. On the basis of their low basal Ba2+ leaks, we selected five high clones and three low clones to utilize in cDNA microarray experiments (Fig. 2, top). Ba2+ uptake rates were as follows: HEK-293 = 0.00061 ± 0.00005 min1 (n = 20); H1 = 0.00098 ± 0.00012 min1 (n = 18); H24 = 0.00106 ± 0.00009 min1 (n = 19); H39 = 0.00097 ± 0.00011 min1 (n = 10); H15 = 0.00095 ± 0.00010 min1 (n = 11); H36 = 0.00107 ± 0.00011 min1 (n = 16); L3 = 0.00039 ± 0.00002 min1 (n = 7); L28 = 0.00034 ± 0.00003 min1 (n = 10); L29 = 0.00034 ± 0.00004 min1 (n = 9). To assure that the variation in SOCE between clones was not simply due to variations in membrane potential, similar experiments were performed in a high-potassium medium to depolarize the membrane potential. As seen in Fig. 2, bottom, although the magnitude of the SOCE was reduced in all clones, membrane depolarization did not normalize the differences in SOCE between the various clones and the parent HEK-293 cell population. An analysis of the Ca2+ transients (area under the curve) in response to thapsigargin and the Ba2+ leak fluxes indicated that there were no statistically significant differences in these parameters between the clones and the parent HEK-293 cells (data not shown). This indicates that the clonal variations in thapsigargin-stimulated Ba2+ entry were not the result of clonal variations in cell volumes or cytoplasmic Ca2+ buffering capacities. Changes in cell volume or buffering capacity would be expected to alter Ba2+ leak fluxes as well as thapsigargin-stimulated Ba2+ flux. Likewise, clonal variations in cytoplasmic Ca2+ buffering capacity would be expected to produce clonal variations in the Ca2+ transients in response to thapsigargin.

View larger version (24K):
[in this window]
[in a new window]
|
Fig. 2. SOCE in selected HEK cell clones. Subpopulations of HEK-293 cells selected by FACS were plated at clonal density, and macroscopic clones were selected and expanded to produce monoclonal lines. Top: high or low levels of SOCE were confirmed in each clone by monitoring thapsigargin-stimulated Ba2+ entry. In the absence of Ca2+, 2 mM Ba2+ was added to Ca2+-free HBSS to measure the rate of Ba2+ leak. After switching to Ca2+-free medium, Ca2+ stores were released by adding 1 µM thapsigargin. After Ca2+ returned to basal levels, 2 mM Ba2+ was added. SOCE was determined by subtracting initial Ba2+ leak influx from the thapsigargin-stimulated Ba2+ influx. Compared with HEK control cells, 5 clones (H1, H15, H24, H36, H39) have significantly (P < 0.01) higher SOCE levels and 3 clones (L3, L28, L29) have significantly (P < 0.0005) lower SOCE levels. B: SOCE was measured in a high K+ medium (133 mM K+, with K+ above 5.4 mM being added as a replacement for Na+). Compared with the HEK control, all 5 high SOCE clones had significantly higher SOCE levels (P < 0.05) and all 3 low SOCE clones had significantly lower SOCE levels (P < 0.05). The no. of coverslips tested for SOCE is shown in parenthesis, with each determination being the average response of 800 cells.
|
|
These data extend our earlier report on clonal variation of SOCE (2) by demonstrating that, when individual cell clones are selected and grown into populations of cells, the various cell populations maintain their elevated or decreased level of SOCE. The successful generation of multiple clonal lines with high or low SOCE promises to be useful in evaluating the physiological roles of SOCE. Experiments can be performed to investigate the physiological consequences of having elevated or decreased levels of SOCE in the absence of pharmacological interventions. Our initial investigations in this direction were to assess the effect of elevated or decreased SOCE on gene expression profiles in HEK cells. One should note at this point that more of the clones from the high SOCE selection than from the low SOCE selection were confirmed by Ba2+ imaging experiments. This is likely the result of the steeper shape of the distribution curve at the low end vs. the high end of the sorting profile (Fig. 1C).
Because we wanted to investigate the effect of low or high SOCE levels on gene expression under normal physiological conditions, we harvested logarithmically growing cells directly from their growth medium (DMEM + 10% FBS) for the subsequent RNA purification. Previous studies have reported that various growth factors stimulate SOCE (32, 34, 59), suggesting that an increase or decrease in SOCE should be reflected in a change in Ca2+ entry for cells growing in serum. The data in Fig. 3 confirm that serum-stimulated Ca2+ entry is higher in the five high SOCE clones than in HEK-293 cells and is lower in the low SOCE clones (all statistically significant, P < 0.03).

View larger version (18K):
[in this window]
[in a new window]
|
Fig. 3. Serum-stimulated Ca2+ entry in the high and low SOCE clones. Cells from different clones were plated on coverslips 24 h before the experiment, and serum-stimulated Ba2+ entry was monitored in each clone. In the absence of Ca2+, 2 mM Ba2+ was added to Ca2+-free HBSS to measure the rate of Ba2+ leak. After switching to Ca2+-free medium, Ca2+ stores were released by adding 0.5% FBS. After Ca2+ returned to basal levels, 2 mM Ba2+ was added in Ca2+-free medium containing 0.5% FBS. The serum-stimulated Ba2+ entry was determined by subtracting initial Ba2+ leak influx from the Ba2+ influx measured after serum treatment. Compared with the HEK control, all 5 high SOCE clones had significantly higher serum-stimulated Ba2+ entry levels (P < 0.03) and all 3 low SOCE clones had significantly lower serum-stimulated Ba2+ entry levels (P < 0.03). The no. of coverslips tested is shown in parenthesis, with each determination being the average response of 800 cells.
|
|
Gene Expression Profiles in High and Low SOCE Clones
The quantity of total RNA purified was determined by absorbance measurements, and the quality of the RNA was checked by running several micrograms of RNA on gel electrophoresis. A second check on the RNA quality was performed at our Functional Genomics Core Facility. For each sample, at least two cDNA microarrays were run, but for the control HEK-293 cells and the H36 clone, four cDNA microarrays were run. In Fig. 4A, an example of the quality of the technical replication of the microarray assay is shown; the fluorescent signals from one duplicate sample of the H1 clone are plotted on the x-axis vs. the fluorescence signals from the other duplicate sample of the H1 clone on the y-axis. This plot demonstrates that there is good replication between the duplicate microarrays run in these experiments. This is especially true for those genes with a signal in excess of 100 on the fluorescence scale; there are few points with signals >100 that fall above the line indicating a greater than twofold change in gene expression level.

View larger version (47K):
[in this window]
[in a new window]
|
Fig. 4. Microarray data for HEK clones with high or low SOCE. Representative scatter plots are shown for the microarray experiments comparing gene expression profiles of HEK cells, high SOCE clones, and low SOCE clones. Green data points represent those genes for which the absolute fluorescence measurements fell below the threshold of 200. Red points represent the genes for which the fluorescence was above 200 for both measurements. A: gene-by-gene comparison of the fluorescence values obtained from the 2 duplicate chips run for the H1 clone. This comparison is shown to indicate the quality of the technical replication of the data. B: gene-by-gene comparison of the fluorescence values obtained from 1 chip run for the H1 clone (high SOCE) and 1 chip run for the HEK control. C: gene-by-gene comparison of the fluorescence values obtained from 1 chip run for the L3 clone (low SOCE) and 1 chip run for the HEK control. D: gene-by-gene comparison of the fluorescence values obtained from 1 chip run for the H24 clone (high SOCE) and 1 chip run for the L3 clone (low SOCE).
|
|
In Fig. 4B, the data for one replicate of the high clone H1 is plotted against the data for one replicate of the control HEK-293 cells. This plot stands in dramatic contrast to the plot illustrating the excellent technical replication. There are many more genes with signals above 100 that fall in the region indicating a greater than twofold change in expression level. There also are several genes with expression levels above 100 that fall above the line indicating a >10-fold change in expression level. These results suggest that there will be reproducible changes in expression levels in the clones with elevated levels of SOCE.
In Fig. 4C, we have plotted the data for one replicate of the L3 clone vs. the data for one replicate of the parent HEK-293 cells. This plot also looks quite different from the plot demonstrating the quality of technical replication. There are a larger number of genes with a signal above 100 that fall in the region of the plot that indicates a greater than twofold change in gene expression. These results suggest that there may be a significant number of genes that increase in cells that have low levels of SOCE.
Finally, in Fig. 4D, we have plotted one replicate of the L3 clone vs. one replicate of the H24 clone. Compared with Fig. 4B, there are many more genes within the region representing changes in level of gene expression between 2-fold and 10-fold. This suggests that there may be a number of genes that increase in cells having high SOCE levels and also decrease in cells having low levels of SOCE.
Changes in Gene Expression in Clones with High Levels of SOCE
We report all genes that change their expression level by at least twofold in at least four of the five high clones and that, in the low clone, do not undergo a change in the same direction. We chose to report the four-of-five category of genes because it was felt that two classes of interesting genes might be present in this category. Some genes would be in this category because changes were above threshold in four of five genes and slightly below threshold in the fifth clone. For example, if, for a particular gene, the ratios (on the log2 scale) were 1.1, 1.2, 1.2, 1.1, and 0.9 for the five high clones, we would consider it to be an interesting gene. We were also interested in the much smaller subclass of genes where the change was dramatic in four of the five clones but did not change, or changed in the opposite direction, in the fifth clone. Our prediction was that genes regulated downstream of SOCE would change comparably in five of five high clones. We also theorized that some genes that are elevated might be responsible for the elevated SOCE. Three obvious theories to explain the elevated SOCE would be an increase in channel proteins, an increase in a positive channel regulator, or a decrease in a negative channel regulator; it is not necessary for all high clones to have the same underlying mechanism for the elevated SOCE. We hypothesized that we possibly could identify an elevated positive regulator as being the underlying mechanism responsible for the elevated SOCE in some of the high clones. Thus the lack of response of a gene in one of five clones would suggest that it is not Ca2+ regulated and is not the underlying cause for elevated SOCE in that particular clone, but could be the cause for the elevated SOCE in the other four clones. This is only one of a number of potential scenarios one can use to explain elevated SOCE levels in the high clones, but it serves to explain our rationale for wanting to examine genes that respond in four of five clones. Thus we have reported genes that changed above the twofold threshold in at least four of five high clones.
Identifying genes that changed expression levels by at least twofold in at least four of the five high clones involved two different sorts of the comparative data file. We initially ran a Countif routine to determine the number of I (significant increase) parameters on each row (each row representing all of the 20 comparisons for a single gene). For each gene with 16 or more Is, we determined that they had all presents (P) in the absolute file and that they achieved a value of 200. Thus genes that started with a value below 200 in HEK cells but increased above 200 in the high clones were counted. Genes that increased dramatically in the low clone for this experiment were excluded. The genes that fit these criteria are listed in Table 1.
After determining which genes increased in the high SOCE clones, we then ran a Countif routine to determine the number of D (significant decrease) parameters on each row. For each gene with 16 or more Ds, we determined that they achieved a value of 200. Thus genes that started with a value above 200 in HEK cells but dropped below 200 in the high clones were counted. Genes that decreased dramatically in the low clone for this experiment were excluded. The genes that fit these criteria are listed in Table 2.
Changes in Gene Expression in Clones with Low Levels of SOCE
As discussed in MATERIALS AND METHODS, we wanted to pull out those genes that changed expression levels by at least twofold in the low SOCE clones compared with the parental HEK cells. This involved two different sorts of the comparative data file for the second set of microarrays. We initially ran a Countif routine to determine the number of I (significant increase) parameters present on each row (each row representing all of the 8 comparisons for a single gene). For each gene with eight Is, we determined that they had all presents (P) in the absolute file and that they achieved a value of 200. Thus genes that started with a value below 200 in HEK cells but increased above 200 in the low clones were counted. Genes that increased dramatically in the high clone for this experiment were excluded. The genes that fit these criteria are listed in Table 3.
After determining which genes increased in the low SOCE clones, we then ran a Countif routine to determine the number of D (significant decrease) parameters present on each row. For each gene with eight Ds, we determined that they achieved a value of 200. Thus genes that started with a value above 200 in HEK cells but dropped below 200 in the low clones were counted. Genes that decreased dramatically in the high clone for this experiment were excluded. The genes that fit these criteria are listed in Table 4.
Confirmation of Gene Changes by Real-Time PCR
High SOCE clones.
To evaluate the ability of the gene microarray to reveal changes in gene expression levels, we ran real-time PCR assays on a select number of genes that changed in either the low or high SOCE clones. The data in Fig. 5 show a comparison between the microarray results and the quantitative PCR (Q-PCR) results for several genes that increased their level of expression in the high clones. The responses for dihydrolipoamide branched-chain transacylase (DBT) match very well between the microarray and Q-PCR results; both methods show no change in the low clones but a dramatic elevation in the high clones. The data for ATPase, class 1, type 8B, member 1 (ATP1B8) show an example of a gene that responds in both the low and high clones, by being dramatically higher in the high clones and significantly lower in the low clones. Fibroblast growth factor (FGF)-13 represents a case where the microarray data reveal a trend in directions (i.e., low in low clones and high in high clones), but the Q-PCR shows a more dramatic elevation in the high clones than was predicted by the microarray.

View larger version (16K):
[in this window]
[in a new window]
|
Fig. 5. Quantitative (Q)-PCR confirmation for selected genes whose expression levels increase in high SOCE clones. Expression levels measured in the microarray assay for dihydrolipoamide branched-chain transacylase (DBT), ATP1B8, and fibroblast growth factor (FGF)-13 in control, low SOCE clones, and high SOCE clones are shown at left. Microarray results for these 3 genes were confirmed by Q-PCR, and the data are shown at right. As indicated, the increases in gene expression in the high SOCE clones predicted by the microarray assay for DBT, ATP1B8, and FGF-13 were confirmed by the real-time PCR assay. In some cases, the Q-PCR assay demonstrated more dramatic changes than were predicted by the microarray assay (e.g., the decrease in ATP1B8 in the low SOCE clones).
|
|
The data in Fig. 6 are a comparison between microarray results and Q-PCR results for several genes the expression of which decreases in the high clones and increases in the low clones. The Q-PCR results for glutaminyl peptide cyclotransferase (QPCT) agree quantitatively with the microarray results showing dramatic decreases in expression in the high clone and dramatic increases in expression in the low clone. For G protein-coupled receptor (GPR)50 and 14-3-3 epsilon, there is good qualitative agreement between the microarray data and the Q-PCR data, with the observed decrease of GPR50 expression measured in high clones being less dramatic than predicted by the microarray assay, but the increase in the low clone being more dramatic than predicted. For 14-3-3 epsilon, the Q-PCR data showed a more dramatic decrease in the high clone and a more dramatic increase in the low clone.

View larger version (19K):
[in this window]
[in a new window]
|
Fig. 6. Q-PCR confirmation for selected genes whose expression levels decreased in the high SOCE clones and increased in the low SOCE clones. Expression levels measured in the microarray assay for glutaminyl peptide cyclotransferase (QPCT), G protein-coupled receptor (GPR)50, and 14-3-3 epsilon in control, low SOCE clones, and high SOCE clones are shown at left. Microarray results for these 3 genes were confirmed by Q-PCR, and the data are shown at right. As indicated, the microarray and Q-PCR showed qualitative agreement, but in some instances the Q-PCR assay demonstrated more dramatic changes than the microarray assay predicted.
|
|
Low SOCE clones.
The data in Fig. 7 show a comparison between the microarray results and the Q-PCR results for several genes that have a decreased expression level in the low clones. The responses for bone morphogenetic protein (BMP)-2 and lymphotoxin ß-receptor (LTBR) match very well between the microarray and the Q-PCR results, with BMP-2 showing no significant change in the high clones but a dramatic decrease in the low clones, while LTBR shows a dramatic decrease in the low clones and a dramatic increase in the high clones. The data for ras-related C3 botulinum toxin substrate-3 (RAC3) show an example of a gene that responds in the low clones as predicted by the microarray data, but the Q-PCR picks up a significant elevation (note break in y-axis scale) in the high clones that was missed in the microarray assay.

View larger version (14K):
[in this window]
[in a new window]
|
Fig. 7. Q-PCR confirmation for selected genes whose expression levels decreased in low SOCE clones. Expression levels measured in the microarray assay for bone morphogenetic protein (BMP)-2, lymphotoxin ß-receptor (LTBR), and ras-related C3 botulinum toxin substrate 3 (RAC3) in control, low SOCE clones, and high SOCE clones are shown at left. Microarray results for these 3 genes were confirmed by Q-PCR, and the data are shown at right. As indicated, the decreases in gene expression in the low SOCE clones predicted by the microarray assay for BMP-2, LTBR, and RAC3 were confirmed by the real-time PCR assay, as was the increased LTBR expression in the high SOCE clones.
|
|
The data in Fig. 8 are a comparison between microarray results and Q-PCR results for several genes the expression of which increases dramatically in the low clones. Diaphanous protein homolog 2 (DIAPH2) and C-type lectin show little change in the high clones in both the microarray and the Q-PCR assay but show dramatic increases in gene expression in the low clones. Ribosomal protein, large, P1 (RPLP1)-like shows good agreement between the microarray and the Q-PCR results in the low clone and qualitative agreement in the high clone (i.e., there is a small increase in the high clones relative to the large change observed for the low clone).

View larger version (18K):
[in this window]
[in a new window]
|
Fig. 8. Q-PCR confirmation for selected genes whose expression levels increased in low SOCE clones. Expression levels measured in the microarray assay for diaphanous protein homolog 2 (DIAPH2), C-type lectin, and ribosomal protein, large, P1 (RPLP1) in control, low SOCE clones, and high SOCE clones are shown at left. Microarray results for these 3 genes were confirmed by Q-PCR, and the data are shown at right. As indicated, the increases in gene expression in the low SOCE clones predicted by the microarray assay for DIAPH2, C-type lectin, and RPLP1 were confirmed by real-time PCR assay.
|
|
Because fewer clones were analyzed for the low SOCE condition than for the high SOCE condition, the confidence level for gene changes associated with low SOCE was somewhat less than that for those associated with high SOCE. To confirm that a decrease in SOCE is causative for the changes in gene expression in the low SOCE clones, SOCE was reduced by an alternate method and select genes were assayed by real-time RT-PCR methods. Recently, our laboratory has demonstrated that a reduction of canonical transient receptor potential (TRPC)1 protein levels by siRNA techniques leads to a 65% reduction in SOCE in HEK-293 cells (unpublished observation), a result consistent with that observed after reduction of TRPC1 levels in cultured hippocampal cells (58). The data in Fig. 9 show that changes in gene expression, predicted by the microarray results for the low SOCE clones, can be confirmed in HEK-293 cells in which SOCE is reduced by 65% due to the expression of siRNA specific for TRPC1. Thus, for six test genes [cyclin T2 (CCNT2), melanoma antigen A1 (MAGEA1), rab3-GTPase-activating protein (rab3-GAP), Rho exchange factor (proto-LBC), meningioma (MN1), and zinc finger protein 22 (KOX15)], the changes in mRNA expression levels associated with decreased SOCE in the low clones were duplicated when SOCE was reduced by expressing siRNA specific for TRPC1.

View larger version (16K):
[in this window]
[in a new window]
|
Fig. 9. Comparison of expression levels for select genes in low SOCE clones vs. cells in which SOCE is suppressed by expression of small interfering (si)RNA to TRPC1. Expression levels for MN1, KOX15, CCNT2, Rab3-GAP, Proto-LBC, and MAGEA1 genes were compared by the real-time RT-PCR method in 3 cell lines: HEK-293 cells, L29 cells, and siTRPC1 cells (HEK-293 cells stably expressing siRNA to TRPC1). A: expression levels of 2 genes (MN1 and KOX15) are confirmed to be downregulated in L29 cells, as predicted by the microarray data (Table 4). They are also downregulated when SOCE is suppressed by siRNA to TRPC1. B: expression levels of 4 genes (CCNT2, Rab3-GAP, proto-LBC, and MAGEA1) are confirmed to be upregulated in L29 cells, as predicted by the microarray results (Table 3). They are also upregulated when SOCE is suppressed by siRNA to TRPC1.
|
|
Could Some of These Genes Be Regulators of SOCE?
It is possible that some of the signaling molecules that are present on the list of genes in Tables 14 could be responsible for the difference in SOCE between the high or low clones. The task is to distinguish those genes that are downstream from, and regulated by, an elevated Ca2+ signal from those that are upstream of, and perhaps causative for, an elevated SOCE. As discussed earlier, it seemed that some insight might be gained by looking at genes that were elevated in four of five of the clones. One gene that fell into this category, and is a well-known signaling molecule, is IRS-2. Western blot data indicate that IRS-2 protein levels are high in four of the five clones compared with the levels in HEK cells (data not shown). The data in Fig. 10A illustrate the difference in protein levels among control HEK cells, L3 cells, and H36 cells. This compares favorably with the differences in mRNA levels predicted by the microarray assay (Fig. 10B). We wanted to investigate whether reducing the high level of IRS-2 in the H36 clone would effect the level of SOCE. Thus we designed constructs that, upon stable expression in cells, would express hairpin siRNA specific for IRS-2. We stably transfected H36 cells with this siRNA construct and mixed all of the surviving cells (in excess of 200 surviving clones) to eliminate possible effects from clonal selection. We monitored the level of IRS-2 protein by Western blotting and found IRS-2 levels in the H36 clone to be reduced to approximately the level observed in control HEK-293 cells (Fig. 11A). Protein expression was reduced by
80%, averaged over three experiments (Fig. 11B).

View larger version (21K):
[in this window]
[in a new window]
|
Fig. 10. Elevated endogenous insulin receptor substrate (IRS)-2 protein expression in high SOCE clone. Cells (HEK control, L3 clone, and H36 clone) were grown on dishes as described in MATERIALS AND METHODS. A: Western blot analysis. Cells were lysed in modified radioimmunoprecipitation (RIPA) buffer, and Western blots were performed using monoclonal anti-IRS-2 antibody (1:1,000 dilution) and anti-mouse antibody (1:10,000 dilution) as the secondary antibody. Western blots were repeated at least 3 times, using different cell lysates. B: microarray analysis. Data represent average fold changes in IRS-2 gene expression calculated for HEK control, L3, and H36 cells.
|
|
Next, we investigated whether a reduction in IRS-2 levels in H36 cells could alter the level of SOCE. We observed that reducing IRS-2 levels in the H36 cells led to a 60% reduction in leak-corrected, thapsigargin-stimulated Ba2+ entry (Fig. 11C). Parallel experiments in the H24 clone gave similar results (data not shown). Complementary studies were performed in which IRS-2 was overexpressed in a low SOCE clone. The results in Fig. 12 show that overexpression of IRS-2 in L29 cells resulted in a dramatic increase in SOCE. These data suggest that IRS-2 does play a role in determining the level of SOCE in these cells. Although much work is required to determine where IRS-2 might be positioned in the signaling pathway that regulates SOCE, the observed results do indicate that it will be important to investigate some of the other signaling molecules that appear in Tables 14 for their potential role as regulators of SOCE.

View larger version (24K):
[in this window]
[in a new window]
|
Fig. 12. Overexpression of IRS-2 on SOCE. L29 cells (low SOCE) were transiently cotransfected with pcDNA3.1EYFP vector and pSMVhis-IRS2 as described in MATERIALS AND METHODS. L29 cells transiently cotransfected with pcDNA3.1EYFP and pSMVhis were used as a control (Mock). Cells were circled based on their EYFP fluorescence when excited at 500 nM. SOCE was determined as described in Fig. 2. Cells overexpressing IRS-2 had significantly higher SOCE levels than did their controls (*P < 0.0001). The no. of cells assayed is shown in parenthesis.
|
|
 |
DISCUSSION
|
---|
Implications of the Microarray Results
The results from the microarray experiments using cell clones with differing levels of SOCE indicate that the level of SOCE has a profound effect on the gene expression profile in HEK-293 cells. The investigation of five high SOCE clones vs. three low SOCE clones provides a higher level of confidence in the cDNA microarray results associated with an increase in SOCE than in the results associated with a reduction of SOCE. However, the case for the association between low SOCE and changes in gene profile is greatly strengthened by a supporting set of experiments using an alternative method to decrease SOCE. Real-time RT-PCR was used to confirm changes in mRNA levels for a select number of genes predicted to respond to a decrease in SOCE in HEK-293 cells expressing TRPC1 siRNA to reduce SOCE by 65% (unpublished observation). This led to changes in gene expression for these six selected genes in line with predictions from the cDNA microarray results in the low SOCE clones.
The types of genes affected by changes in SOCE cover a wide range on the functional spectrum, including signal transduction molecules, transcription factors, regulators of apoptosis, metabolic enzymes, cytoskeletal elements, and membrane transporters. Although it is obviously impossible to discuss all of the implications that can be drawn from a complete review of Tables 14, it is worthwhile to discuss several important points.
Changes in the level of SOCE have a significant impact on several genes that are related to cell cycle or cell proliferation.
There is extensive literature describing changes in Ca2+ levels during proliferation and at specific points within the cell cycle, as well as reports that imposed changes in intracellular Ca2+ levels can block the cell cycle at specific points. (4, 54) Thus the genes listed below could serve as an important focus for future investigations to clarify the role of Ca2+ in regulating cell cycle and proliferation.
1) In HEK cell clones with high levels of SOCE, the G2 and S phase expressed-1 (GTSE-1) gene expression is upregulated. Murine GTSE-1 was cloned as a p53-inducible gene (11), and it, as well as its human homolog, have been demonstrated to be cell cycle regulated (37). The product of GTSE-1, which localizes to microtubules, has been reported to delay G2/M progression and to negatively regulate p53 function and p53-dependent apoptosis (11). Given the previously described changes in Ca2+ within the cell cycle and the reported effects of Ca2+ on G2/M progression, it will be important to determine whether GTSE-1 plays any role in cell cycle regulation attributed to changes in cytosolic Ca2+.
2) Cyclin T2 is another example of a cell cycle- or proliferation-related gene that is regulated by SOCE (upregulated in the low clones). This protein is found complexed to cyclin-dependent kinase (CDK)9, which is known to phosphorylate the protein product of the retinoblastoma gene (15). Although the kinase activity of CDK9 does not appear to be cell cycle dependent, there is recent evidence that it can be involved in controlling cell growth and/or cell viability (13). Other studies point to a role for CDK9 in differentiation, and, given the role of Ca2+ in differentiation, it would be important to investigate possible Ca2+-mediated changes in cyclin T2 in differentiating cells (15).
3) MAD1L1, the human homolog of the yeast mitotic checkpoint gene MAD1 (16, 28), is also regulated by SOCE. MAD1L1 mRNA was observed to decrease in cells with high SOCE. Although the precise function of MAD1L1 in the mitotic spindle checkpoint remains unknown, this again would be an interesting protein to explore in terms of the role of Ca2+ in regulating cell cycle.
4) In the high clones, a decrease in the expression of a cdk inhibitor p21 binding protein (TOK-1) was observed. A p21(Cip1/Waf1/Sdi1) protein is thought to negatively regulate the cell cycle by inhibiting kinase activity of a variety of cyclin-dependent kinases. TOK-1 is expressed at the G1/S boundary of the cell cycle and is thus thought to be a new type of CDK2 modulator (42). An investigation of whether TOK-1 plays a role in the Ca2+-mediated regulation of cell cycle is warranted.
Changes in the level of SOCE have a significant impact on several genes that are related to various disease states.
1) The most striking observation in this regard is the large number of melanoma-associated genes the expression levels of which are increased when SOCE is decreased. In the low SOCE clones, an increase in gene expression was observed for melanoma antigen family members MAGEA1, MAGEA2A, MAGEC1, and MAGE6. The melanoma antigen genes were initially identified in melanomas and were subsequently found to have an expression pattern almost exclusively confined to tumors (39). Given that distribution pattern and our finding that four family members are upregulated in cells with low SOCE, it would be important to investigate whether changes, especially decreases, in SOCE levels occur in melanomas or other types of tumors.
2) The tumor protein D52 is another cancer-associated gene the expression of which is changed in cell clones with modified SOCE. Its expression level is decreased in the high SOCE clones. This gene was discovered in a differential screening of a breast carcinoma cDNA library; subsequent studies showed that this gene is overexpressed in
40% of breast carcinomas (9).
3) The dyskeratosis congenita 1 gene is downregulated in clones with high SOCE. The X-linked form of this disease, which results in skin and bone marrow failure, is due to mutations in the dyskerin gene. The dyskerin protein is a component of small nucleolar ribonuclear protein particles as well as the telomerase complex (35). The gene or genes involved in the recessive forms of the disease still remain unknown, so the regulation of dyskerin gene expression by SOCE should be of interest to those who study this particular disease.
4) Finally, the gene DSHP [Src homology (SH)2 domain protein-1A; Duncans disease] is upregulated in clones with low SOCE. DSHP encodes a single SH2 domain protein that is mutated in some patients with X-linked lymphoproliferative syndrome. Because DSHP is upregulated late in the immune response, this form of immunodeficiency differs from most others, where the mutations occur in a signaling molecule that is hard wired into the signaling complex (47). Thus it will be of great interest to determine whether changes in levels of SOCE can play a role in Duncans disease.
Changes in the level of SOCE alter the expression levels of several hormones, cytokines, and growth factors in HEK-293 cells.
1) In the cells with low levels of SOCE, a decreased expression of BMP-2, a member of the transforming growth factor (TGF)-ß superfamily of polypeptide signaling molecules, was observed. Although BMPs were first discovered for their osteogenic effects (56), they were later found to be expressed in a wide range of vertebrate embryonic structures (24) and to be involved in dorsal-ventral axis specifications (21). Mice having null mutations in BMP-2 die early in embryogenesis (62). In early reports, BMP-2 expression was found to be controlled by both retinoic acid and cAMP (44), but in recent studies in limb bud mesechymal cells, BMP-2 gene expression was increased by ionomycin and suppressed by the calcineurin inhibitor cyclosporine A (53). These results, coupled to our observation of a reduction of BMP-2 expression in low SOCE clones, suggest that further study into the role of store-operated channels in regulating BMP-2 would be important.
2) Another member of the TGF-ß superfamily found to be decreased in the clones with low SOCE is TGF-ß1. The TGF-ß compounds have three well-characterized biological activities. They inhibit growth in most cells except for chondrocytes and osteoblasts, in which they stimulate growth. They have an immunosuppressive effect by inhibiting T and B lymphocytes. They also stimulate the deposition of collagens, fibronectin, and proteoglycans (7). Additional studies will be required to determine the importance of the downregulation of TGF-ß1 levels in HEK cells that have low levels of SOCE.
3) The level of gene expression for FGF-13 was also found to be decreased in clones with low levels of SOCE. FGF-13 is a member of the large family of FGFs that were originally found to stimulate growth (20). They are now known to regulate differentiation and a number of other physiological functions in a wide variety of cells. They play an important physiological role in development, maintenance of tissues, and wound repair and have been postulated to play a pathophysiological role in arthritis, tumor proliferation, and arteriosclerosis. Because little is known about the regulation of gene expression for this recently discovered FGF family member (22), our observation that SOCE plays a role in regulating FGF-13 mRNA levels is an important contribution to this area.
Changes in the level of SOCE in HEK cells had an effect on a number of genes related to apoptosis.
Given the rich, but confusing, literature on the role of Ca2+ in apoptosis, there should be considerable interest in apoptotic genes regulated by SOCE.
1) The expression of Fas-associated death domain (FADD)-like apoptosis regulator (FLAME-1) was observed to increase in clones with high SOCE. FLAME-1, which contains FADD death effector domain homology regions, can be recruited to the Fas receptor complex, where it inhibits Fas/TNF receptor (TNFR)-induced apoptosis, possibly by acting as a dominant-negative inhibitor (52).
2) The programmed cell death 4 (PDCD4) gene was observed to decrease in clones having high levels of SOCE. This gene was first discovered as one that is upregulated after initiation of apoptosis in a number of different cell types, and recent evidence suggests that PDCD4 may function as a tumor suppressor gene (30). A recent paper (19) described the upregulation of PDCD4 in HEK-293 cells that were transfected with the fas ligand gene. However, with other apoptotic signals, PDCD4 can be unaffected or even downregulated (40, 41), indicating that we do not fully understand the role of this molecule in apoptosis.
3) The apoptosis-inducing factor (AIF, PDCD8) gene was also observed to decrease in clones with high levels of SOCE. AIF is expressed in both normal cells and a variety of cancer cells. The mature protein is confined to the mitochondrial intermembrane space, but in response to apoptosis-inducing conditions, it is released to the cytosol to act by a caspase-independent process to promote nuclear chromatin condensation and DNA fragmentation (12).
4) The Fas apoptotic inhibitory molecule (FAIM) is also downregulated in cells with high levels of SOCE. This gene was cloned by differential display comparing Fas-resistant and Fas-sensitive primary murine B lymphocytes. FAIM is evolutionarily conserved and expressed in a wide range of tissues, suggesting that its gene product plays a key physiological role. It will be interesting to investigate why this apoptotic inhibitory molecule is downregulated along with the cell death genes (listed above) when SOCE is increased.
A number of enzymes involved in metabolism are represented in the list of genes in Tables 14.
These include enzymes involved in carbohydrate metabolism such as galactokinase 1, glycerol kinase, phosphoglycerate kinase, solute carrier family 2, phosphofructokinase, dehydrogenase/reductase SDR family member, and HEP27; enzymes involved in amino acid metabolism such as serine hydroxymethyltransferase, phosphoserine aminotransferase, pyrroline-5-carboxylate synthetase, phosphoribosyl pyrophosphate amidotransferase, and glutamate decarboxylase 1; and enzymes involved in lipid metabolism such as very-long-chain acyl-CoA synthetase homolog 2, hydroxyacyl-CoA dehydrogenase, 3,2-trans-enoyl-CoA isomerase, dehydrocholesterol reductase, steroid sulfatase, and androgen-regulated short-chain dehydrogenase/reductase.
Changes in the level of SOCE had a dramatic effect on a number of signaling molecules.
These include protein kinases, protein phosphatases, and transcription factors. We will only discuss a subset of these, some of which have been demonstrated in other studies to be regulated by SOCE and some that are of interest to investigate as potential regulators of SOCE.
A number of signaling genes whose levels of expression were altered in our studies were also found to be Ca2+ regulated in other studies linking gene expression to SOCE (18, 27). These include calcineurin, cAMP-dependent protein kinase catalytic-ß, FLAME-1, c-myc, frizzled homolog 7, Sine oculis homeobox homolog 2 (SIX2), and TGF-ß1. The number of genes the expression of which is modified in the low SOCE clones (150 genes either increasing or decreasing) is not out of line with the number of genes proposed to respond to a decrease in SOCE in the T lymphocyte study (18) mentioned earlier (111 genes either increasing or decreasing), especially when one considers that the cDNA microarray used for our study had significantly more genes represented. Although there are a few genes in common between our study and the lymphocyte study, for the most part, SOCE appears to regulate different populations of genes in the two cell types, a possibility that we considered at the outset of the investigation (see INTRODUCTION). Compared with the data from the fibroblast study (27), the 150 genes regulated by a decrease in SOCE are much higher than the 29 genes seen to change. However, the microarray used in that study only represented 1,200 cDNA clones compared with the 22,000 cDNA clones represented on the Affymetrix microarray.
1) Our initial hypothesis was that some of the signaling genes that are high or low in the clones with high or low SOCE might be setting the levels of SOCE rather than responding to alterations in SOCE. This hypothesis was further supported by the observation that there were not significant changes in levels of TRPC genes in the high or low clones. Our results showing that expression of IRS-2 is high in four of five high clones and that expression of siRNA specific for IRS-2 inhibits SOCE in H36 cells, together with the observation that overexpression of IRS-2 in L29 cells elevates SOCE, support the hypothesis that IRS-2 may be involved in the SOCE signaling pathway. Because IRS-2 is clearly not causative for the high SOCE levels in the H1 clone, it is worthwhile to discuss what other signaling molecules listed in Tables 14 should be investigated as potential regulators of SOCE. Previous studies based on microinjection of GTP
S (6, 17), microinjection of Clostridium C3 transferase, or overexpression of wild-type rho (60) suggested that small-molecular-weight G proteins might be involved in regulating SOCE. Thus it will be important to further investigate the following genes on our lists: Rab3 GTPase-activating protein, ras-related C3 botulinum toxin substrate 3 (RAC3), rho GDP dissociation inhibitor (GDI)-
, and Rab1.
2) Previous data from our laboratory (1, 31) and from several other laboratories (45, 46, 55) suggest that there is a tyrosine phosphorylation step involved in regulating SOCE. Thus it will be important to investigate those molecules on our list that mediate changes in protein tyrosine phosphorylation levels or that interact with tyrosine phosphorylated proteins. These include the following: protein tyrosine kinase 9, H-Ryk receptor tyrosine kinase, protein tyrosine phosphatase receptor type F, CAK tyrosine protein kinase, protein tyrosine phosphatase nonreceptor type 4 (PTPN4), and SH2 domain protein 1A (DSHP).
It is possible that some of the genes upregulated in high SOCE may serve as negative regulators of SOCE, thereby preventing cells from achieving even higher levels of SOCE. Alternatively, it is possible that some of the genes upregulated in the low clones may serve to prevent the SOCE from falling to lower levels.
In summary, the selection of HEK-293 cell clones with high or low levels of SOCE, and the suppression of SOCE levels by the expression of siRNA specific for TRPC1, has enabled us to begin asking important questions concerning the physiological role of SOCE. The gene expression profiles for these cells have allowed us to gain important information about genes regulated by SOCE. The fact that these profiles were determined for cells under normal growth conditions (presence of serum), and in the absence of pharmacological interventions, allows a more accurate determination of the physiological importance of store-operated channels. These results will be particularly useful in evaluating the role of SOCE in regulating genes involved in cell proliferation and cell cycle. In addition to providing a more physiological measure of the genes regulated downstream of SOCE, these studies provide several candidate genes to investigate for their potential role in regulating store-operated channels. One of these, IRS-2, shows real promise as a regulator of SOCE based on the inhibition of SOCE in H36 cells observed after suppression of IRS-2 levels by siRNA techniques, and the increase in SOCE observed when IRS-2 is overexpressed in a clone low in SOCE. Future studies can investigate how IRS-2 might fit in with other proposed mechanisms for regulation of SOCE and explore the role of some of the other potential regulators of SOCE.
 |
GRANTS
|
---|
This work was supported by National Institute of General Medical Sciences Grant GM-54500 and a supplemental grant for microarray analysis (GM-54500-05S1) (to M. L. Villereal).
 |
ACKNOWLEDGMENTS
|
---|
We acknowledge the generous assistance that Dr. Xinmin Li from our Functional Genomics Core Facility provided in the analysis of the gene microarray data. The IRS-2 expression construct was generously provided by Dr. Xiao Jian Sun, Medicine Department, University of Chicago.
 |
FOOTNOTES
|
---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: M. L. Villereal, Dept. of Neurobiology, Pharmacology and Physiology, The Univ. of Chicago, 947 E. 58th St., Chicago, IL 60637 (E-mail: mitch{at}bsd.uchicago.edu).
10.1152/physiolgenomics. 00099.2004.
1 The Supplemental Material for this article (Supplemental Tables S1S4) is available online at http://physiolgenomics.physiology.org/cgi/content/full/0099.2004/DC1. 
 |
REFERENCES
|
---|
- Babnigg G, Bowersox SR, and Villereal ML. The role of pp60c-src in the regulation of calcium entry via store-operated calcium channels. J Biol Chem 272: 2943429437, 1997.[Abstract/Free Full Text]
- Babnigg G, Heller B, and Villereal ML. Cell-to-cell variation in store-operated calcium entry in HEK-293 cells and its impact on the interpretation of data from stable clones expressing exogenous calcium channels. Cell Calcium 27: 6173, 2000.[CrossRef][ISI][Medline]
- Babnigg G, Zagranichnaya T, Wu X, and Villereal ML. Differential tyrosine phosphorylation of plasma membrane Ca2+-ATPase and regulation of calcium pump activity by carbachol and bradykinin. J Biol Chem 278: 1487214882, 2003.[Abstract/Free Full Text]
- Berridge MJ. Calcium signalling and cell proliferation. Bioessays 17: 491500, 1995.[CrossRef][ISI][Medline]
- Berridge MJ. The versatility and complexity of calcium signalling. Novartis Found Symp 239: 5264, 2001.[ISI][Medline]
- Bird GS and Putney JW Jr. Inhibition of thapsigargin-induced calcium entry by microinjected guanine nucleotide analogues. Evidence for the involvement of a small G-protein in capacitative calcium entry. J Biol Chem 268: 2148621488, 1993.[Abstract/Free Full Text]
- Blumenfeld I and Livne E. The role of transforming growth factor (TGF)-ß, insulin-like growth factor (IGF)-1, and interleukin (IL)-1 in osteoarthritis and aging of joints. Exp Gerontol 34: 821829, 1999.[CrossRef][ISI][Medline]
- Bootman MD, Berridge MJ, and Roderick HL. Calcium signalling: more messengers, more channels, more complexity. Curr Biol 12: R563R565, 2002.[CrossRef][ISI][Medline]
- Byrne JA, Tomasetto C, Garnier JM, Rouyer N, Mattei MG, Bellocq JP, Rio MC, and Basset P. A screening method to identify genes commonly overexpressed in carcinomas and the identification of a novel complementary DNA sequence. Cancer Res 55: 28962903, 1995.[Abstract]
- Chung KC, Sung JY, Ahn W, Rhim H, Oh TH, Lee MG, and Ahn YS. Intracellular calcium mobilization induces immediate early gene pip92 via Src and mitogen-activated protein kinase in immortalized hippocampal cells. J Biol Chem 276: 21322138, 2001.[Abstract/Free Full Text]
- Collavin L, Monte M, Verardo R, Pfleger C, and Schneider C. Cell-cycle regulation of the p53-inducible gene B99. FEBS Lett 481: 5762, 2000.[CrossRef][ISI][Medline]
- Daugas E, Nochy D, Ravagnan L, Loeffler M, Susin SA, Zamzami N, and Kroemer G. Apoptosis-inducing factor (AIF): a ubiquitous mitochondrial oxidoreductase involved in apoptosis. FEBS Lett 476: 118123, 2000.[CrossRef][ISI][Medline]
- de Falco G and Giordano A. CDK9 (PITALRE): a multifunctional cdc2-related kinase. J Cell Physiol 177: 501506, 1998.[CrossRef][ISI][Medline]
- de la Rosa LA, Alfonso A, Vilarino N, Vieytes MR, Yasumoto T, and Botana LM. Maitotoxin-induced calcium entry in human lymphocytes: modulation by yessotoxin, Ca(2+) channel blockers and kinases. Cell Signal 13: 711716, 2001.[CrossRef][ISI][Medline]
- De Luca A, De Falco M, Baldi A, and Paggi MG. Cyclin T: three forms for different roles in physiological and pathological functions. J Cell Physiol 194: 101107, 2003.[CrossRef][ISI][Medline]
- Elledge SJ. Cell cycle checkpoints: preventing an identity crisis. Science 274: 16641672, 1996.[Abstract/Free Full Text]
- Fasolato C, Hoth M, and Penner R. A GTP-dependent step in the activation mechanism of capacitative calcium influx. J Biol Chem 268: 2073720740, 1993.[Abstract/Free Full Text]
- Feske S, Giltnane J, Dolmetsch R, Staudt LM, and Rao A. Gene regulation mediated by calcium signals in T lymphocytes. Nat Immun 2: 316324, 2001.[CrossRef][ISI]
- Goke A, Goke R, Knolle A, Trusheim H, Schmidt H, Wilmen A, Carmody R, Goke B, and Chen YH. DUG is a novel homologue of translation initiation factor 4G that binds eIF4A. Biochem Biophys Res Commun 297: 7882, 2002.[CrossRef][ISI][Medline]
- Gospodarowicz D, Jones KL, and Sato G. Purification of a growth factor for ovarian cells from bovine pituitary glands. Proc Natl Acad Sci USA 71: 22952299, 1974.[Abstract]
- Graff JM. Embryonic patterning: to BMP or not to BMP, that is the question. Cell 89: 171174, 1997.[CrossRef][ISI][Medline]
- Greene JM, Li YL, Yourey PA, Gruber J, Carter KC, Shell BK, Dillon PA, Florence C, Duan DR, Blunt A, Ornitz DM, Ruben SM, and Alderson RF. Identification and characterization of a novel member of the fibroblast growth factor family. Eur J Neurosci 10: 19111925, 1998.[CrossRef][ISI][Medline]
- He H, McColl K, and Distelhorst CW. Involvement of c-Fos in signaling grp78 induction following ER calcium release. Oncogene 19: 59365943, 2000.[CrossRef][ISI][Medline]
- Hogan BL. Bone morphogenetic proteins: multifunctional regulators of vertebrate development. Genes Dev 10: 15801594, 1996.[CrossRef][ISI][Medline]
- Hong SJ and Chang CC. Facilitation of nicotinic receptor desensitization at mouse motor endplate by a receptor-operated Ca2+ channel blocker, SK&F 96365. Eur J Pharmacol 265: 3542, 1994.[CrossRef][ISI][Medline]
- Hong SJ, Lin WW, and Chang CC. Inhibition of the sodium channel by SK&F 96365, an inhibitor of the receptor-operated calcium channel, in mouse diaphragm. J Biomed Sci 1: 172178, 1994.[Medline]
- Jenkins RE, Hawley SR, Promwikorn W, Brown J, Hamlett J, and Pennington SR. Regulation of growth factor-induced gene expression by calcium signalling: integrated mRNA and protein expression analysis. Proteomics 1: 10921104, 2001.[CrossRef][ISI][Medline]
- Jin DY, Spencer F, and Jeang KT. Human T cell leukemia virus type 1 oncoprotein Tax targets the human mitotic checkpoint protein MAD1. Cell 93: 8191, 1998.[CrossRef][ISI][Medline]
- Kwan CY and Putney JW Jr. Uptake and intracellular sequestration of divalent cations in resting and methacholine-stimulated mouse lacrimal acinar cells. Dissociation by Sr2+ and Ba2+ of agonist-stimulated divalent cation entry from the refilling of the agonist-sensitive intracellular pool. J Biol Chem 265: 678684, 1990.[Abstract/Free Full Text]
- Lankat-Buttgereit B and Goke R. Programmed cell death protein 4 (pdcd4): a novel target for antineoplastic therapy? Biol Cell 95: 515519, 2003.[CrossRef][ISI][Medline]
- Lee KM, Toscas K, and Villereal ML. Inhibition of bradykinin- and thapsigargin-induced Ca2+ entry by tyrosine kinase inhibitors. J Biol Chem 268: 99459948, 1993.[Abstract/Free Full Text]
- Li WP, Tsiokas L, Sansom SC, and Ma R. Epidermal growth factor activates store-operated Ca2+ channels through an inositol 1,4,5-trisphosphate-independent pathway in human glomerular mesangial cells. J Biol Chem 279: 45704577, 2004.[Abstract/Free Full Text]
- Liu W, Youn HD, and Liu JO. Thapsigargin-induced apoptosis involves Cabin1-MEF2-mediated induction of Nur77. Eur J Immunol 31: 17571764, 2001.[CrossRef][ISI][Medline]
- Ma R and Sansom SC. Epidermal growth factor activates store-operated calcium channels in human glomerular mesangial cells. J Am Soc Nephrol 12: 4753, 2001.[Abstract/Free Full Text]
- Marrone A and Mason PJ. Dyskeratosis congenita. Cell Mol Life Sci 60: 507517, 2003.[CrossRef][ISI][Medline]
- Moneer Z and Taylor CW. Reciprocal regulation of capacitative and non-capacitative Ca2+ entry in A7r5 vascular smooth muscle cells: only the latter operates during receptor activation. Biochem J 362: 1321, 2002.[CrossRef][ISI][Medline]
- Monte M, Collavin L, Lazarevic D, Utrera R, Dragani TA, and Schneider C. Cloning, chromosome mapping and functional characterization of a human homologue of murine gtse-1 (B99) gene. Gene 254: 229236, 2000.[CrossRef][ISI][Medline]
- Nakamura R, Ishida S, Ozawa S, Saito Y, Okunuki H, Teshima R, and Sawada J. Gene expression profiling of Ca2+-ATPase inhibitor DTBHQ and antigen-stimulated RBL-2H3 mast cells. Inflamm Res 51: 611618, 2002.[ISI][Medline]
- Ohman Forslund K and Nordqvist K. The melanoma antigen genesany clues to their functions in normal tissues? Exp Cell Res 265: 185194, 2001.[CrossRef][ISI][Medline]
- Onishi Y, Hashimoto S, and Kizaki H. Cloning of the TIS gene suppressed by topoisomerase inhibitors. Gene 215: 453459, 1998.[CrossRef][ISI][Medline]
- Onishi Y and Kizaki H. Molecular cloning of the genes suppressed in RVC lymphoma cells by topoisomerase inhibitors. Biochem Biophys Res Commun 228: 713, 1996.[CrossRef][ISI][Medline]
- Ono T, Kitaura H, Ugai H, Murata T, Yokoyama KK, Iguchi-Ariga SM, and Ariga H. TOK-1, a novel p21Cip1-binding protein that cooperatively enhances p21-dependent inhibitory activity toward CDK2 kinase. J Biol Chem 275: 3114531154, 2000.[Abstract/Free Full Text]
- Putney JW Jr. A model for receptor-regulated calcium entry. Cell Calcium 7: 112, 1986.[CrossRef][ISI][Medline]
- Rogers MB, Rosen V, Wozney JM, and Gudas LJ. Bone morphogenetic proteins-2 and -4 are involved in the retinoic acid-induced differentiation of embryonal carcinoma cells. Mol Biol Cell 3: 189196, 1992.[Abstract]
- Rosado JA, Graves D, and Sage SO. Tyrosine kinases activate store-mediated Ca2+ entry in human platelets through the reorganization of the actin cytoskeleton. Biochem J 351: 429437, 2000.[CrossRef][ISI][Medline]
- Sargeant P, Farndale RW, and Sage SO. ADP- and thapsigargin-evoked Ca2+ entry and protein-tyrosine phosphorylation are inhibited by the tyrosine kinase inhibitors genistein and methyl-2,5-dihydroxycinnamate in fura-2-loaded human platelets. J Biol Chem 268: 1815118156, 1993.[Abstract/Free Full Text]
- Satterthwaite AB, Rawlings DJ, and Witte ON. DSHP: a "power bar" for sustained immune responses? Proc Natl Acad Sci USA 95: 1335513357, 1998.[Free Full Text]
- Schilling WP, Rajan L, and Strobl-Jager E. Characterization of the bradykinin-stimulated calcium influx pathway of cultured vascular endothelial cells. Saturability, selectivity, and kinetics. J Biol Chem 264: 1283812848, 1989.[Abstract/Free Full Text]
- Schwarz G, Droogmans G, and Nilius B. Multiple effects of SK&F 96365 on ionic currents and intracellular calcium in human endothelial cells. Cell Calcium 15: 4554, 1994.[CrossRef][ISI][Medline]
- Shalabi A, Zamudio F, Wu X, Scaloni A, Possani LD, and Villereal ML. Tetrapandins, a new class of scorpion toxins that specifically inhibit store-operated calcium entry in human embryonic kidney-293 cells. J Biol Chem 279: 10401049, 2004.[Abstract/Free Full Text]
- Soergel DG, Yasumoto T, Daly JW, and Gusovsky F. Maitotoxin effects are blocked by SK&F 96365, an inhibitor of receptor-mediated calcium entry. Mol Pharmacol 41: 487493, 1992.[Abstract]
- Srinivasula SM, Ahmad M, Ottilie S, Bullrich F, Banks S, Wang Y, Fernandes-Alnemri T, Croce CM, Litwack G, Tomaselli KJ, Armstrong RC, and Alnemri ES. FLAME-1, a novel FADD-like anti-apoptotic molecule that regulates Fas/TNFR1-induced apoptosis. J Biol Chem 272: 1854218545, 1997.[Abstract/Free Full Text]
- Tomita M, Reinhold MI, Molkentin JD, and Naski MC. Calcineurin and NFAT4 induce chondrogenesis. J Biol Chem 277: 4221442218, 2002.[Abstract/Free Full Text]
- Villereal ML and Byron KL. Calcium signals in growth factor signal transduction. Rev Physiol Biochem Pharmacol 119: 67121, 1992.[ISI][Medline]
- Vostal JG, Jackson WL, and Shulman NR. Cytosolic and stored calcium antagonistically control tyrosine phosphorylation of specific platelet proteins. J Biol Chem 266: 1691116916, 1991.[Abstract/Free Full Text]
- Wozney JM, Rosen V, Celeste AJ, Mitsock LM, Whitters MJ, Kriz RW, Hewick RM, and Wang EA. Novel regulators of bone formation: molecular clones and activities. Science 242: 15281534, 1988.[ISI][Medline]
- Wu X, Babnigg G, Zagranichnaya T, and Villereal ML. The role of endogenous human Trp4 in regulating carbachol-induced calcium oscillations in HEK-293 cells. J Biol Chem 277: 1359713608, 2002.[Abstract/Free Full Text]
- Wu X, Zagranichnaya TK, Gurda GT, Eves EM, and Villereal ML. A TRPC1/TRPC3-mediated increase in store-operated calcium entry is required for differentiation of H19-7 hippocampal neuronal cells. J Biol Chem 279: 4339243402, 2004.[Abstract/Free Full Text]
- Yang H, Sun X, Wang Z, Ning G, Zhang F, Kong J, Lu L, and Reinach PS. EGF stimulates growth by enhancing capacitative calcium entry in corneal epithelial cells. J Membr Biol 194: 4758, 2003.[CrossRef][ISI][Medline]
- Yao Y, Ferrer-Montiel AV, Montal M, and Tsien RY. Activation of store-operated Ca2+ current in Xenopus oocytes requires SNAP-25 but not a diffusible messenger. Cell 98: 475485, 1999.[CrossRef][ISI][Medline]
- Zhang H and Bradley A. Mice deficient for BMP2 are nonviable and have defects in amnion/chorion and cardiac development. Development 122: 29772986, 1996.[Abstract/Free Full Text]
Copyright © 2005 by the American Physiological Society.