HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Data Supplement All Versions of this Article: 571/1/57    most recent jphysiol.2005.100586v2 jphysiol.2005.100586v1 Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Núñez, L. Articles by Villalobos, C. Search for Related Content PubMed PubMed Citation Articles by Núñez, L. Articles by Villalobos, C. Related Collections Cellular

¹ Instituto de Biología y Genética Molecular (IBGM), Universidad de Valladolid and CSIC, C/Sanz y Forés s/n. 47003-Valladolid, Spain

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Store-operated Ca²⁺ entry (SOCE) is a ubiquitous Ca²⁺ influx pathway involved in control of multiple cellular and physiological processes including cell proliferation. Recent evidence has shown that SOCE depends critically on mitochondrial sinking of entering Ca²⁺ to avoid Ca²⁺-dependent inactivation. Thus, a role of mitochondria in control of cell proliferation could be anticipated. We show here that activation of SOCE induces cytosolic high [Ca²⁺] domains that are large enough to be sensed and avidly taken up by a pool of nearby mitochondria. Prevention of mitochondrial clearance of the entering Ca²⁺ inhibited both SOCE and cell proliferation in several cell types including Jurkat and human colon cancer cells. In addition, we find that therapeutic concentrations of salicylate, the major metabolite of aspirin, depolarize partially mitochondria and inhibit mitochondrial Ca²⁺ uptake, as revealed by mitochondrial Ca²⁺ measurements with targeted aequorins. This salicylate-induced inhibition of mitochondrial Ca²⁺ sinking prevented SOCE and impaired cell growth of Jurkat and human colon cancer cells. Finally, direct blockade of SOCE by the pyrazole derivative BTP-2 was sufficient to arrest cell growth. Taken together, our results reveal that cell proliferation depends critically on mitochondrial Ca²⁺ uptake and suggest that inhibition of tumour cell proliferation by salicylate may be due to interference with mitochondrial Ca²⁺ uptake, which is essential for sustaining SOCE. This novel mechanism may contribute to explaining the reported anti-proliferative and anti-tumoral actions of aspirin and dietary salicylates.

(Received 24 October 2005; accepted after revision 7 December 2005; first published online 8 December 2005)
Corresponding author C. Villalobos: Instituto de Biología y Genética Molecular (IBGM), c/Sanz y Forés s/n. 47003-Valladolid, Spain. Email: carlosv{at}ibgm.uva.es

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
Much evidence indicates that Ca²⁺ signals participate in the control of cell proliferation (Berridge, 1995; Kahl & Means, 2003). They activate, for example, immediate early genes responsible for inducing resting cells to enter the cell cycle, promote DNA synthesis and stimulate different events at mitosis (Berridge, 1995). Particular attention has been paid to the role of Ca²⁺ entry through store-operated channels (SOCs, Parekh & Putney, 2005). Store-operated Ca²⁺ entry (SOCE), also termed capacitative Ca²⁺ entry (Putney, 1986), is the most important Ca²⁺ influx pathway in non-excitable cells (Parekh & Penner, 1997). SOCE is activated by emptying of intracellular Ca²⁺ stores, either by inositol 1,4,5 trisphosphate produced upon activation of G-protein-coupled or tyrosine kinase receptors or by pharmacological emptying of the Ca²⁺ stores with, for instance, thapsigargin (Montero et al. 1990, 1991; Alvarez et al. 1994; Parekh & Putney, 2005). The involvement of SOCE in cell proliferation has been studied in detail in Jurkat T cells. In these cells, T cell receptor stimulation induces sustained SOCE through the Ca²⁺ release-activated current (I_CRAC) resulting in sustained activation of the nuclear factor of activated T cells (NFAT) that promotes proliferation (Lewis, 2001). Many observations in this and other cell models relate SOCE and cell proliferation: (i) entry in cell cycle is preceded by SOC activation (Sugioka & Yamashita, 2003); (ii) growth factor stimulation of cell proliferation is accompanied by increased activity and/or expression of transient receptor potential (TRP) channels that are related to SOCE (Golovina et al. 2001); (iii) SOCE inhibition by different means (Ca²⁺ removal, inorganic channel blockers and I_CRAC antagonists) abolishes tumour cell proliferation (Weiss et al. 2001; Zitt et al. 2004).

Two decades after the discovery of SOCE (Putney, 1986), the link between emptying of intracellular Ca²⁺ stores and the increased Ca²⁺ influx is still unknown (Parekh & Putney, 2005). Likewise, the nature of the SOCs responsible for SOCE remains controversial. A number of members of the TRP family of cation channels, including several canonical TRPs (TRPCs) and TRPV6, have been proposed to be involved in SOCE (Parekh & Putney, 2005). An important characteristic of SOCE, I_CRAC and the TRP channels related to SOCE is the strong Ca²⁺-dependent inactivation that limits Ca²⁺ entry (Singh et al. 2002; Parekh & Putney, 2005). In fact, I_CRAC is hardly recorded unless the Ca²⁺-dependent inactivation is prevented by a strong intracellular Ca²⁺ buffer (Gilabert & Parekh, 2000). Thus, intracellular Ca²⁺ extrusion systems are important for sustaining SOCE. Mitochondria play an important role as a Ca²⁺ sinking system able to shape the cytosolic Ca²⁺ signals (Rizzuto et al. 2000; Montero et al. 2000; Villalobos et al. 2002). With regard to SOCE, a series of reports from the laboratories of Lewis (Hoth et al. 1997, 2000) and Parekh (Gilabert & Parekh, 2000; Gilabert et al. 2001, 2002) have shown that sustained SOCE requires normal mitochondrial Ca²⁺ uptake in Jurkat and rat basophilic leukaemia (RBL-1) cells. Protonophores such as CCCP or FCCP prevent SOCE since they collapse the mitochondrial potential (), the driving force for Ca²⁺ entry into mitochondria. The inhibition of SOCE has been attributed to Ca²⁺-dependent inactivation of I_CRAC due to generation of high [Ca²⁺]_cyt domains secondary to a lack of mitochondrial Ca²⁺ uptake (Hoth et al. 1997, 2000; Glitsch et al. 2002; Malli et al. 2003; Parekh, 2003 but see also Varadi et al. 2004; Frieden et al. 2004). We have reported recently that activation of voltage-gated Ca²⁺ channels induces in excitable cells large [Ca²⁺]_cyt increases that are sensed by subplasmalemmal mitochondria (Montero et al. 2000; Villalobos et al. 2001). This pool amounting to about 50% of the mitochondria, takes up tremendous amounts of Ca²⁺ (driving [Ca²⁺]_mit to near millimolar levels) and clears efficiently the high [Ca²⁺]_cyt domains (Villalobos et al. 2002). Whether SOCE induces such high [Ca²⁺]_cyt domains leading to large [Ca²⁺]_mit increases in a mitochondrial pool has not been studied.

If SOCE is essential for cell proliferation and depends on mitochondrial Ca²⁺ uptake, then it follows that drugs interfering with mitochondrial Ca²⁺ uptake may prevent cell proliferation via SOC inactivation. Here we tested whether this rationale can explain the effects of salicylate, the major metabolite of aspirin (acetylsalicylic acid, ASA). Salicylate acts as a proton carrier and uncouples mitochondria (Pachman et al. 1971; Gutknecht, 1990). According to the above rationale, salicylate should interfere with mitochondrial Ca²⁺ uptake, SOCE and cell proliferation. In fact, aspirin and salicylate have been reported to inhibit tumour cell growth and to prevent colon and other cancers (Hanif et al. 1996; Molina et al. 1999; Perugini et al. 2000; Smith et al. 2000; Baron et al. 2003; Sandler et al. 2003), although the action mechanism is unknown. Salicylate has been recently reported to also inhibit proliferation of normal vascular smooth muscle cells (Marra & Liao, 2001). Here we have investigated (i) whether SOCE is regulated by mitochondria in colon cancer cells, (ii) whether proliferation of tumour cells depends on mitochondrial Ca²⁺ uptake and (iii) whether salicylate inhibits tumour cell proliferation by acting on mitochondrial control of SOCE. For these studies we have used HT29 human colon cancer cells and Jurkat T cells where mitochondrial control of SOCE and the role of SOCE in proliferation were first described (Hoth et al. 1997; Lewis, 2001).

	Methods

Top Abstract Introduction Methods Results Discussion References

 
Materials

Fura-2 AM, Fura-4 AM, JC-1 (a cationic dye that indicates mitochondrial polarization) and tetramethylrhodamine ethyl ester perchlorate (TMRE) were purchased from Molecular Probes. Ru360 (a calcium signalling tool) and BTP2 (an immunomodulator) were purchased from Calbiochem. Thapsigargin was obtained from Alomon Laboratories. Other reagents and chemicals were obtained either from Sigma or Merck. Mutated aequorin was kindly donated by M. Montero (Valladolid University, Valladolid). Jurkat cells were obtained from the American Type Culture Collection. HT29 cells were donated by J. C. Fernández-Checa (CSIC, Barcelona). SW480 cells were donated by A. Muñoz (CSIC, Madrid). NIH 3T3 cells expressing luciferase under control of the topoisomerase-II promoter were a kind gift of M. Sehested (Rigshospitalet, Copenhagen).

Store-operated Ca²⁺ entry

Cells were plated at about 0.5 x 10⁶ cells ml^–1 on 12 mm glass coverslips treated with poly L-lysine (HT29) or fibronectin (Jurkat) and loaded with 4 µM fura-2 /AM or fura-4 AM for 60 min at room temperature. Cells were then incubated with 1 µM thapsigargin for 10 min in Ca²⁺-free standard medium containing (mM): NaCl, 145; KCl, 5; MgCl₂, 1; EGTA, 0.5; glucose, 10; Hepes/NaOH, 10 (pH 7.42) and placed on the stage of an inverted microscope (Nikon Diaphot) maintained at 37°C. They were then perfused with prewarmed Ca²⁺-free medium, and epi-illuminated alternately at 340 and 380 nm. Light emitted above 520 nm was recorded with a Magical Image Processor (Applied Imaging). Pixel-by-pixel ratios of consecutive frames were captured and [Ca²⁺]_cyt was estimated from these ratios as previously reported (Villalobos et al. 2002). Re-addition of standard, Ca²⁺-containing (CaCl₂, 1 mM; no EGTA) medium evoked large [Ca²⁺]_cyt increases revealing store-operated Ca²⁺ entry (Montero et al. 1990; Villalobos & García-Sancho, 1995; Hoth et al. 2000; Glitsch et al. 2002). In some experiments, SOCE induced by physiological stimulation with carbachol (100 µM) in HT29 cells and TCR stimulation in Jurkat cells was tested. For TCR stimulation, Jurkat cells were suspended in external medium and loaded with Fura-4 AM (4 µM). Then, cells were plated on fibronectin-coated glass coverslips previously incubated with IgG2a for 4 h at room temperature. Cells were then placed in the fluorescence microscope and stimulated with 0.5 µg ml^–1 anti-CD3 (the T cell receptor) and 5 µg ml^–1 anti-CD28 (a co-receptor) to elicit the cross-linking activation of TCR.

Mitochondrial potential

HT29 cells were loaded with JC-1 (1 µg ml^–1) for 10 min and subjected to confocal microscopy with a Bio-Rad laser scanning system (Radiance 2100) coupled to a Nikon eclipse TE2100U inverted microscope. For TMRE measurements, HT29 cells were loaded with TMRE (100 nM) for 30 min at room temperature, placed on the perfusion chamber of a Zeiss Axiovert 100 TV inverted microscope and superfused continuously with prewarmed (37°C) standard medium. Fluorescence images were taken at 5 s intervals with a Hamamatsu VIM photon counting camera handled with an Argus-20 image processor. Traces from individual cells were expressed as the percentage value of fluorescence before addition of salicylate and averaged. In some experiments, cells were loaded with TMRE (100 nM) and fluorescence from mitochondria and nearby cytosol was analysed with the confocal system. Localization of TMRE fluorescence in mitochondria was tested by co-loading cells with Mitotracker Red. To quantify mitochondrial depolarization, the effect of salicylate and FCCP on the ratio of TMRE fluorescence in mitochondria relative to that surrounding cytosol was calculated as reported by Collins et al. (2002).

Mitochondrial calcium uptake

HT29 cells were transfected (lipofectamin) with a plasmid containing either the wild type or the mutated (Asp119Ala), low-Ca²⁺ affinity aequorin targeted to mitochondria (Montero et al. 2000). After 24 h, cells were incubated in standard medium (see above) containing 1 µM of either wild type or coelenterazine (n, acquorin cofactors) for 2 h at room temperature. The coverslips were then placed in a luminometer (Cairn Research), perfused continuously with warm (37°C) standard medium and subjected to photon counting at 1 s intervals. For some experiments, cells were permeabilized with 20 µM digitonin in ‘intracellular’ medium (130 mM KCl, 10 mM NaCl, 1 mM MgCl₂, 1 mM K₃PO₄, 0.2 mM EGTA, 1 mM ATP, 20 µM ADP, 2 mM succynate, 20 mM Hepes/KOH, pH 6.8). The cells were then incubated with the same medium containing 200 nM Ca²⁺ (buffered with EGTA) with or without salicylate for 5 min. Finally, perfusion was switched to ‘intracellular’ medium containing 6 µM Ca²⁺ (with or without salicylate) for 1 min. Photonic emissions were converted to mitochondrial-free Ca²⁺ concentration ([Ca²⁺]_mit) values as reported previously (Montero et al. 2000; Villalobos et al. 2001; Alvarez & Montero, 2001).

Cell growth, apoptosis and cell viability

Cells (HT29, SW480, Jurkat and NIH 3T3) were cultured in Dulbecco's modified Eagle's medium (DMEM) or RPMI 1640 media (Gibco) containing 10% fetal bovine serum and antibiotics. Cells were plated in wells at about 5 x 10⁴ cells ml^–1 and incubated with test solutions for 96 h. Triplicate wells were counted daily or after 96 h. Cell death was estimated in the same samples by trypan blue exclusion. Data correspond to at least three independent experiments and are expressed as cell number-fold increase, percentage growth rate relative to control or cells per hour. Differences were considered significant at P < 0.05 (Student's t test). For determination of apoptotic cells at single cell level, cells were plated at about 5 x 10⁴ cells ml^–1 and incubated with test solutions for 96 h. Apoptotic cells were revealed by the terminal deoxynucleaotidyl transferase-mediated dUTP nick-end labelling (TUNEL) method by means of fluorescence microscopy and the in situ cell death detection kit (Roche Diagnostics, Penzberg, Germany) following the protocol provided by the manufacturer.

ATP measurements

HT29 cells were plated in 6-well plates and cultured in medium containing either vehicle or different salicylate concentrations (10–2000 µM). After 24 h, cells were washed twice with phosphate-buffered saline (PBS) at 37°C and 1 ml of boiling 20 mM Tris, pH 7.75, 4 mM EDTA solution was added. After 2 min, samples were centrifuged for 4 min at 10 000 g. ATP was measured later from the supernatant by the luciferin–luciferase assay (Sánchez, 1985). Readings were taken using a scintillation counter (Wallac model 1409/11) over 20 s intervals with the windows wide open. ATP levels were determined with the aid of a standard curve prepared using pure ATP over a 10^–5–10^–9M concentration range.

Bioluminescence imaging of transcription dynamics and cell division

NIH 3T3 cells stably transfected with the luciferase reporter gene under control of the topoisomerase-II gene promoter (Falck et al. 1999) were synchronized in medium containing 0.5% serum for 3 days. Then, 10% serum and 1 mM luciferin were added and the Petri dish containing the cells was placed in an incubator (Zeiss CTI-controller 3700) controlling temperature, CO₂ and humidity, attached over the stage of an inverted microscope (Zeiss Axiovert 100 TV). Photonic emissions and bright field images were captured concurrently with a Hamamatsu VIM photon counting camera handled with an Argus-20 image processor at 15 min intervals for 72 h. Transcription activity is expressed as total photonic emissions in each image minus background photonic emissions, as previously reported (Villalobos et al. 1999).

	Results

Top Abstract Introduction Methods Results Discussion References

 
SOCE was shown by the increase in cytosolic Ca²⁺ concentration ([Ca²⁺]_cyt) that follows addition of extracellular Ca²⁺ to fura-2-loaded Jurkat cells pre-treated with thapsigargin in Ca²⁺-free medium. We found that FCCP reversibly reduced the initial rate and the size of SOCE in Jurkat cells (Fig. 1A). SOCE was also shown in HT29 human colon carcinoma cells by the same [Ca²⁺]_cyt overshoot protocol (Fig. 1B). No Ca²⁺ overshoot was observed in cells not pre-treated with thapsigargin (data not shown). A second pulse of extracellular Ca²⁺ evoked a similar [Ca²⁺]_cyt increase and FCCP reversibly prevented SOCE (Fig. 1B). We noticed that the fura-2 signals were often saturated during SOCE in many individual cells. This would result in the underestimation of both [Ca²⁺]_cyt peaks and inhibition by FCCP. To avoid this problem we repeated the experiments using the lower affinity Ca²⁺ probe fura-4 F instead of fura-2 (Villalobos et al. 2002). [Ca²⁺]_cyt increases measured by fura-4 F ranged between 1.3 and 5.8 µM (3.2 ± 1.8 µM, mean ±S.E.M., n= 9) in HT29 cells and were also inhibited by FCCP (Fig. 1C). Similar results were obtained in Jurkat cells (data not shown). The SOCE-induced increase in [Ca²⁺]_cyt as well as the FCCP-induced inhibition were not affected by the presence of F₀F₁–ATP synthase blocker oligomycin (Fig. 1D) indicating that the effects of FCCP are not due to ATP depletion. Oligomycin had no effect on subsequent control pulses either (not shown). In order to interpret the lack of effect of oligomycin, it must be remembered that mitochondrial respiration is very much inhibited in tumour cells simply by incubation in glucose-containing media (Crabtree, 1929). The inhibition is due to blocking of the F₀F₁–ATPase, but the mitochondrial potential is not lost or is even increased (Wojtczak et al. 1999). Treatment with the respiratory chain inhibitor antimycin A plus oligomycin, a cocktail that collapses and prevents SOCE and I_CRAC in Jurkat and RBL-1 cells (Glitsch et al. 2002), also inhibited SOCE in HT29 cells (Fig. 1E). Valinomycin, a K⁺ ionophore that collapses the mitochondrial but hyperpolarizes the plasma membrane, also inhibited SOCE in HT29 cells (Fig. 1F) suggesting that the effects of FCCP are due to collapse of the mitochondrial rather than to plasma membrane depolarization. In medium containing high K⁺ (50 mM), a condition that should hold the membrane potential in a quite depolarized state, FCCP also inhibited SOCE (data not shown). Carbachol activates M3 muscarinic receptors in HT29 cells provoking a biphasic [Ca²⁺]_cyt increase, a peak due to the transient Ca²⁺ release from intracellular stores followed by a plateau due to sustained Ca²⁺ entry via SOC (Kerst et al. 1995). FCCP did not affect the Ca²⁺ release induced by carbachol indicating that FCCP does not affect the ATP-dependent refilling of intracellular Ca²⁺ stores. Similar results have been reported previously in Jurkat cells (Makowska et al. 2000). However, FCCP inhibited the carbachol-induced Ca²⁺ entry (Fig. 1G and H). FCCP also inhibited SOCE in SW480 cells, a human colon cancer cell lacking cyclooxygenase-2 (COX-2) gene expression (Smith et al. 2000) (data not shown). Thus, human colon carcinoma cells display a robust SOCE that is prevented by mitochondrial depolarization and this effect is independent of ATP, plasma membrane depolarization and COX-2 gene expression.

View larger version (28K): [in this window] [in a new window]   Figure 1.  Store-operated Ca2+ entry is prevented by collapsing mitochondrial potential in Jurkat and colon cancer cells SOCE was started by addition of Ca2+ to thapsigargin-treated (1 µM, 10 min), fura-2-loaded cells incubated in Ca2+-free medium. A and B, effects of FCCP (10 µM) in Jurkat (A) and HT29 cells (B). C and D, effects of FCCP (10 µM) and oligomycin (0.12 µM) plus FCCP (10 µM) on Ca2+ entry in fura-4 F-loaded HT29 cells. E and F, effects of oligomycin (0.12 µM) plus antimycin A (0.5 µg ml–1) and valinomycin (10 µM) on SOCE in HT29 cells. G and H, carbachol (100 µM) elicited a biphasic [Ca2+]cyt increase in HT29 cells loaded with fura-4 F. In G, cells were perfused in Ca2+-containing medium with or without (control; •) FCCP (10 µM). FCCP inhibited the sustained but not the transient [Ca2+]cyt increase. In H, cells were perfused with Ca2+-free medium, except for the period indicated (Ca) in which perfusion was shifted to Ca2+-containing medium. All traces of [Ca2+]cyt are averaged values (mean ±S.E.M.) of 28–56 cells and representative of 3–9 similar experiments.

View larger version (9K): [in this window] [in a new window]   Figure 2.  Store-operated Ca2+ entry induces mitochondrial Ca2+ uptake A, HT29 cells were transfected with mitochondria targeted aequorin and subjected to photon counting measurements for estimation of [Ca2+]mit (see Methods). Cells were first treated with 1 µM thapsigargin for 10 min in Ca2+-free medium and Ca2+ entry was started by addition of external Ca2+ (1 mM; Ca). The top panel shows aequorin consumption and the bottom one the [Ca2+]mit estimate (see Methods for details). Data are representative of 3 similar experiments. B, HT29 cells were transfected with the low Ca2+ affinity mutated mitochondria-targeted aequorin, reconstituted with coelenterazine n and subjected to photon counting measurements. Other details as in A. For calibration of [Ca2+]mit, the size of the mitochondrial pool was assumed to contain 45% of the aequorin. Data are representative of 5 similar experiments.

View larger version (36K): [in this window] [in a new window]   Figure 3.  Salicylate depolarizes mitochondria A, confocal images of HT29 cells loaded with JC-1 (1 µg ml–1, 10 min) in vehicle (Control) or salicylate (+Sal, 500 µM, 10 min). Emitted red and green fluorescence corresponds to energised and depolarized mitochondria, respectively. Bar represents 10 µm. B, values of ratios (mean ±S.E.M.; n= 4) of red/green fluorescence in control, salicylate- and FCCP- (10 µM, 10 min) treated cells. *P < 0.05 versus control (Student's t test). C, decrease of TMRE fluorescence induced by FCCP (10 µM) and different salicylate concentrations (in µM) in HT29 cells. The arrow indicates addition of either vehicle, FCCP or salicylate. Fluorescence values for each individual cell were normalized relative to the value before treatment. Traces correspond to the averaged (mean ±S.E.M.), normalized recordings of 48–76 cells. Pictures show TMRE fluorescent images of the same cells before (Control) and 5 min after addition of 500 µM salicylate (+Sal) or 10 µM FCCP (see also Supplemental movie 1). Bar represents 10 µm. D, dose–response relation of TMRE fluorescence quenching by treatment with salicylate; values are mean ±S.E.M. of three independent experiments. All points were significantly different from control (P < 0.05). E, confocal images of bright field and TMRE-loaded HT29 cells in vehicle (Control), 10 min after 500 µM salicylate (+Sal) and 10 min after 10 µM FCCP (+FCCP). Bar represents 10 µm. The ratio of TMRE fluorescence in mitochondria relative to the surrounding cytosol was calculated as reported by Collins et al. (2002). Salicylate and FCCP largely decreased this ratio in both HT29 (F) and Jurkat (G) cells. Data correspond to 12 HT29 and 15 Jurkat cells studied in 3 independent experiments for each cell type.

View larger version (17K): [in this window] [in a new window]   Figure 4.  Salicylate prevents mitochondrial Ca2+ uptake Effects of salicylate (concentrations in µM) and FCCP (10 µM) on the Ca2+ uptake by mitochondria. A, HT29 cells expressing mutated mitochondrial aequorin were permeabilized and [Ca2+]mit was calculated from photonic emissions (details as in Fig. 2). Ca2+ uptake was started by perfusion with 6 µM Ca2+ (filled bar). B, average effects of different salicylate concentrations on mitochondria Ca2+ uptake in 3 experiments similar to that shown in A. Data are mean ±S.E.M. All points were significantly different from control.

View larger version (27K): [in this window] [in a new window]   Figure 5.  Salicylate prevents store-operated Ca2+ entry Effects of different salicylate concentrations (in µM) on SOCE in thapsigargin-treated, fura-4 F-loaded, HT29 (A) and Jurkat (B) cells. Each couple of traces correspond to [Ca2+]cyt recordings either before and after incubation (10 min) with different salicylate concentrations in the same cells. Each trace is the mean ±S.E.M. of 29–59 cells. C and D, dose–response curves of effects of salicylate on SOCE inhibition induced by salicylate in HT29 (C) and Jurkat (D) cells. Each value is the mean ±S.E.M. of 3 experiments. E, effects of salicylate (2 mM) on Ca2+ entry induced by carbachol (100 µM) in HT29 cells (traces are the mean ±S.E.M. of 23 and 32 cells and representative of 3 similar experiments). F, effects of salicylate (2 mM) on Ca2+ entry induced by TCR stimulation in Jurkat cells (0.5 µg ml–1 anti-CD3 and 5 µg ml–1 anti-CD28 cross-linked with IgG2a). Traces are mean ±S.E.M. of 12 and 22 cells and representative of 3 similar experiments.

View larger version (28K): [in this window] [in a new window]   Figure 6.  Salicylate inhibits tumour cell growth Effects of different concentrations of salicylate (in µM) on HT29 (A) and Jurkat (B) cell growth, expressed as cell number fold increase. All the groups differed statistically from each other except for the effects of 500 µM salicylate (not shown) that were not significantly different from the effects of 100 µM salicylate. Dose–response curve of the effects of salicylate on HT29 (C) and Jurkat (D) cell growth. E and F, correlation between growth inhibition (%) and SOCE inhibition (%) in HT29 (E) and Jurkat (F) cells. Lines are best linear fit (r= 0.94–0.96; P < 0.005–0.05). G and H, effects of salicylate on percentage of dead cells (cells stained with trypan blue) after 96 h incubation with the different salicylate concentrations (in µM) in HT29 (G) and Jurkat (H) cells. Cell death induced by 10, 100 or 500 µM salicylate were not significantly different from control (P > 0.05).

View larger version (23K): [in this window] [in a new window]   Figure 7.  Effects of salicylate on cell ATP and apoptosis A, effects of salicylate on cell ATP concentrations in HT29 cells measured by the luciferin/luciferase assay. Salicylate (10– 2000 µM) did not affect ATP concentrations (P > 0.05). B, the effect of salicylate on the per cent of apoptotic cells was tested by the tunel assay. Pictures show HT29 cell nuclei (blue) and apoptotic cells (red) in vehicle (control) and cells treated with different concentrations of salicylate. The bar represents 10 µm. B, bars show the per cent of apoptotic cells in each condition (mean ±S.E.M., n= 3). Salicylate did not induce apoptosis at concentrations of 10– 2000 µM (P > 0.05).

View larger version (72K): [in this window] [in a new window]   Figure 8.  Salicylate inhibits cell cycle-driven gene transcription Transcription activity of topoisomerase-II and cell division were monitored for 72 h at 15 min intervals by concurrent bioluminescence imaging and bright field microscopy in NIH 3T3 cells expressing luciferase under control of the topoisomerase-II gene promoter. The pictures show accumulated photonic emissions (15 min integration period) superimposed on bright field images in control (top) and salicylate-treated cells (bottom) at 0, 24, 48 and 72 h. Bar represents 10 µm. Traces show photonic emissions at 15 min intervals for 72 h in vehicle (control) and salicylate (500 µM) containing medium. Data are representative of 15 (control) and 4 (salicylate) experiments. See also Supplemental movie 2.

View larger version (21K): [in this window] [in a new window]   Figure 9.  Ruthenium compounds inhibit store-operated Ca2+ entry, cell cycle driven gene transcription and tumour cell growth A, effects of 6 µM Ru360 on SOCE in HT29 cells loaded with fura-4 F. Each trace is the mean ±S.E.M. of 54 cells (data representative of 3 experiments). B, transcription cycling of topoisomerase-II in 3T3 cells treated with vehicle (control) or 6 µM Ru360. Data representative of 15 (control) and 3 (Ru360) independent experiments. C and D, effects of ruthenium red (RR, 100 µM, 96 h) on Jurkat (C) and HT29 (D) cell growth (n= 3). *P < 0.05 versus control (Student's t test). Similar results were obtained with Ru360 (data not shown).

View larger version (25K): [in this window] [in a new window]   Figure 10.  BTP-2 inhibits store-operated Ca2+ entry (SOCE) and cell growth A, effects of different BTP-2 concentrations (in µM) on SOCE in HT29 cells loaded with fura-4 F. Each trace is the average (mean ±S.E.M.) of 29–83 cells. B, dose dependency of the effects of BTP-2 on SOCE (%). Each value, expressed as percentage of control, is the mean ±S.E.M. of 3 experiments. Each bar was significantly different from the others (P < 0.05). C, dose dependence of the effects of BTP-2 on HT29 cell growth. Each bar represents the mean ±S.E.M. of 3 experiments. Each bar was significantly different from the others (P < 0.05). D, correlation between SOCE inhibition (%) and growth inhibition (%) induced by BTP-2 in HT29 cells. The line is the best linear fit of experimental data shown in B and C (r= 0.999; P < 0.0001). E, effects of BTP-2 (BTP2, 10 µM), salicylate (500 µM, Sal) and both (BTP2 + Sal) on cell growth in HT29 cells. Salicylate did not increase inhibition of cell growth induced by BTP-2. Effects of increasing extracellular Ca2+ concentration from 1.8 mM (control) to 3.8 mM (+Ca2+). High Ca2+ essentially restored the inhibitory effects of salicylate. Each bar represents the mean ±S.E.M. of 3 experiments. Each bar was significantly different (P < 0.05) from the others except for BTP-2 alone versus BTP-2 plus salicylate.

	Discussion

Top Abstract Introduction Methods Results Discussion References

According to our results, SOCE generates high [Ca²⁺]_cyt domains that are sensed and cleared by nearby mitochondria, thus allowing sustained Ca²⁺ entry in colon cancer cells. Massive Ca²⁺ uptake by a fraction of mitochondria takes place through the Ca²⁺ uniporter, which is activated by high [Ca²⁺]_cyt domains generated by nearby SOCs. This situation is similar to the one reported previously in adrenal chromaffin (Montero et al. 2000; Villalobos et al. 2002) and anterior pituitary cells (Villalobos et al. 2001). In fact, we showed previously that most of the Ca²⁺ entering excitable cells upon activation of voltage-gated Ca²⁺ channels was cleared by nearby mitochondria rather than by other extrusion systems (Villalobos et al. 2002). Our present results suggest that Ca²⁺ entry in non-excitable cells through SOCs is similarly cleared by nearby mitochondria. As a consequence, inhibition of mitochondrial Ca²⁺ uptake by different means (FCCP, ruthenium compounds or antimycin A plus oligomycin) impairs Ca²⁺ clearance and leads to SOCE inhibition, probably by inactivation of SOC by high [Ca²⁺] domains.

If sustained SOCE requires mitochondrial Ca²⁺ uptake, then it follows that any compound that prevents mitochondrial Ca²⁺ uptake should inhibit SOCE. ASA and specially its major metabolite salicylate are mitochondrial uncouplers (Pachman et al. 1971). Salicylate enters mitochondria as salicylic acid (driven by the concentration gradient) and exits as salicylate anion (driven by voltage and concentration gradients) thus producing net proton entry (Gutknecht, 1990). Salicylate is much more potent than ASA because it contains an internal hydrogen bond that delocalizes the negative charge, thus increasing the lipid solubility and permeability of the anion. Model calculations predicted that, at therapeutic concentrations, salicylate may cause a net H⁺ influx into mitochondria enough to explain the reported ‘loose coupling’ effect (Gutknecht, 1990). Consistently, we find that salicylate induces a partial mitochondrial depolarization in intact cells. As stated above, the Nernst equilibrium predicts that even a small drop in should greatly affect equilibrium [Ca²⁺]_mit and the driving force for mitochondrial Ca²⁺ uptake. This prediction was confirmed here as salicylate concentrations as low as 10 µM induced a modest mitochondrial depolarization (< 15%) that contrasted with a much larger inhibition of mitochondrial Ca²⁺ uptake (> 70%; Fig. 3). This disparity between the effects on and on mitochondrial Ca²⁺ uptake had not been experimentally shown before because lack of reliable measurements of Ca²⁺ uptake in mitochondria of living cells. Most of the previous measurements underestimated the actual [Ca²⁺]_mit increases reached upon cell stimulation. The recent introduction of mitochondria-targeted, low affinity mutated aequorin (Montero et al. 2000) has enabled [Ca²⁺]_mit measurements in the high micromolar to millimolar range. Using this novel methodology, we have shown here that therapeutic concentrations of salicylate produce a large inhibition of mitochondrial Ca²⁺ uptake. Since prevention of mitochondrial Ca²⁺ uptake leads to SOCE inhibition (Hoth et al. 1997, 2000; Glitsch et al. 2002), it follows that therapeutic concentrations of salicylate should also prevent SOC operation. Again, this expectation was confirmed in both Jurkat and colon carcinoma cells. Moreover, we also show that salicylate prevented SOCE elicited by physiological stimulation in both HT29 and Jurkat cells. In addition, the effects of salicylate on SOCE, like those of FCCP, were not affected by oligomycin, suggesting that the salicylate-induced inhibition of SOCE is not due to ATP depletion. In support of this view, we show here that salicylate concentrations that inhibited SOCE fully had no effect on ATP levels in HT29 cells. It must be remembered that tumour cells use glycolysis more than mitochondrial respiration, which is blocked by glucose (Crabtree, 1929), for ATP synthesis. Taken together, our results indicate that salicylate inhibits SOCE by preventing mitochondrial Ca²⁺ uptake, and that this leads to Ca²⁺-dependent inactivation of SOC.

If SOCE is essential for proliferation in Jurkat and other tumour cells then salicylate should inhibit tumour cell proliferation at the same low concentrations that prevent SOCE. Our results fulfil this prediction as salicylate inhibits both SOCE and cell growth with the same concentration dependence. The lack of effects of salicylate on cell viability and apoptosis together with the effects of salicylate on cell division and transcription cycling of topoisomerase-II, a cell-cycle-related gene and marker of cell proliferation, supports the view that salicylate inhibits cell proliferation at low, therapeutic concentrations, and this effect is not due to changes in ATP concentration. Furthermore, our data support the view that SOCE inhibition is sufficient to prevent tumour cell proliferation. The pyrazole derivative BTP-2, a direct I_CRAC blocker (Zitt et al. 2004), inhibits SOCE and proliferation in Jurkat cells with the same concentration dependence. We show here similar results for colon cancer cells. It should be mentioned that BTP-2 has been reported not to affect mitochondria or other ion channels (Zitt et al. 2004). Moreover, salicylate effects on cell proliferation are not additive with the effects induced by BTP-2. In addition, the effects of salicylate on cell proliferation are restored by increasing the extracellular Ca²⁺ concentration suggesting they are not due to a possible metabolic effect but to a defect in SOCE. Finally, diazoxide and ruthenium compounds, which inhibit SOCE by interfering in other ways with mitochondrial Ca²⁺ uptake, also prevent tumour cell proliferation. Taken together, our data support the view that salicylate, at therapeutic concentrations, inhibits SOCE in a mitochondria-dependent manner and that this effect is sufficient to prevent cell proliferation.

What is the mechanism by which SOCE inhibition impairs cell proliferation? Stimulation of Jurkat cells induces sustained SOCE and activation of the Ca²⁺-dependent phosphatase calcineurin that dephosphorylates NFAT, promoting expression of interleukin-2 and proliferation (Lewis, 2001). Interestingly, inhibition of SOCE by FCCP abolishes NFAT activation (Hoth et al. 2000). Moreover, direct blockade of SOCE by BTP-2 also prevents NFAT activation and proliferation (Zitt et al. 2004). Thus, the effects of salicylate on SOCE may interfere with the NFAT signalling pathway to proliferation. Consistent with this view, salicylate has been recently reported to inhibit NFAT activation and NFAT-dependent transcription in Jurkat cells (Aceves et al. 2004). The role of SOCE in control of proliferation in colon cancer cells is scarcely known. The inducible COX-2 isozyme is up-regulated in multiple tumours including colon cancer (Molina et al. 1999) and it has been proposed that it promotes tumour cell growth. Interestingly, transcription of COX-2 is regulated by NFAT both in Jurkat (Iñiguez et al. 2000) and colon cancer cells (Duque et al. 2005). Salicylate inhibits COX-2 gene transcription at therapeutic concentrations similar to those used here (Xu et al. 1999). Thus, control of proliferation by SOCE in colon cancer cells could be transduced by NFAT-mediated regulation of COX-2 gene expression. Notwithstanding, the above possibility cannot account for the anti-proliferative effects of salicylate in colon cancer cells lacking COX-2 gene expression (Hanif et al. 1996; Smith et al. 2000) and we find that salicylate prevents SOCE and tumour cell growth also in the colon cancer cells lacking COX-2 gene expression. Therefore, SOCE must regulate proliferation by a different mechanism, perhaps involving other NFAT-induced genes. Interestingly, NFAT activation induced by sustained SOCE is considered a general proliferative signal (Lipskaia & Lompre, 2004). On the other hand, since SOCE is an early event in signal transduction, it is likely that multiple downstream transducing proteins could be recruited in the proliferative signalling pathway. One important example is NFB, another transcription factor involved in control of cell proliferation, which is inhibited by salicylate and modulated by multiple signalling pathways including Ca²⁺ signals. However, salicylate inhibits NFB activity and NFB-mediated gene expression at concentrations (IC₅₀ 5–9 mM) far larger than those reported here to inhibit SOCE and cell proliferation. Although caution is necessary in the interpretation of the results, the multiple targets proposed for aspirin (Hardwick et al. 2004) and salicylate can be consistent with our findings.

Our results may have important implications as salicylate, aspirin and other non-steroidal anti-inflammatory drugs (NSAIDs) have been reported to inhibit tumour cell growth (Hanif et al. 1996; Molina et al. 1999; Smith et al. 2000) and to protect against colon (Chan, 2003) and other cancers (Bosetti et al. 2001). Clinical trials show that ASA reduces the risk of colorectal adenomas (Baron et al. 2003; Sandler et al. 2003) but the action mechanism has remained elusive. ASA irreversibly inhibits COX activity by acetylation of constitutive COX-1 at Ser530 and inducible COX-2 at Ser516. In vivo, ASA is quickly (t_1/2= 15 min) deacetylated to salicylic acid, which remains in the plasma for much longer. Salicylic acid does not directly inhibit COX activity because it lacks the acetyl group, although, as stated above, it antagonizes COX-2 gene expression at the transcription level (Xu et al. 1999). Thus, both inhibition of COX-2 activity and COX-2 gene expression have been proposed to mediate the anti-tumoral effects of ASA. Recent results, however, do not support this view. It has been shown, for example, that ASA and COX-2 inhibitors block proliferation through a prostaglandin-independent pathway in both COX-2-expressing (HT29 and HCA-7) and non-expressing (SW480 and HTC116) colon cancer cells (Hanif et al. 1996; Smith et al. 2000). In the same line, it has been reported that the anti-proliferative effects of NSAIDs are independent of the level of COX-2 expression (Molina et al. 1999) or prostaglandin E2 production (Perugini et al. 2000), but related to cell cycle quiescence. Microarray analysis has shown that ASA represses many cell-cycle-related genes and modulates multiple signalling pathways (Hardwick et al. 2004), suggesting that an early mitotic signal may be the key target for the anti-proliferative effect. This view is consistent with the results we report here where the aspirin metabolite salicylate inhibits tumour cell proliferation by promoting inactivation of SOCE, an essential early requirement for proliferation. Thus, our results could contribute to explaining the mechanism for the anti-tumoral actions of aspirin and dietary salicylates (Paterson & Lawrence, 2001). Interestingly, ruthenium compounds have also been reported to have anti-tumoral and immunosuppresive properties both in vivo and in vitro, but the action of this mechanism has also remained elusive (Sava & Bergamo, 2000). Our results support the view that the effects of ruthenium compounds on mitochondrial Ca²⁺ uptake might contribute towards explaining also their anti-tumoral and immunosuppresive properties.

Salicylate has been also reported to inhibit proliferation of normal vascular smooth muscle cells in a prostaglandin-independent manner (Marra & Liao, 2001). In fact, salicylate has been proposed as a candidate for treatment of proliferative disorders of the vessel wall that lead to intimal hyperplasia and hypertension. The specific mechanism for this action is not clear and multiple downstream targets have been proposed (Marra & Liao, 2001). Interestingly, vascular smooth muscle cells show up-regulated TRP and enhanced SOCE during proliferation (Golovina et al. 2001) suggesting that SOCE is very relevant for proliferation of these cells. It is tempting to speculate that salicylate might inhibit normal vascular smooth muscle cell proliferation by the same mechanism reported here. Nevertheless, the effects of salicylate are achieved at larger concentrations suggesting the possibility that salicylate may inhibit transformed cells more efficiently than normal cells. Further research is required to ascertain both the mechanism and possible differential sensitivity to salicylate in normal cells relative to their transformed counterparts.

In summary, our results suggest a novel and important role of mitochondria in control of cell proliferation. Specifically, we show here that tumour cell proliferation critically depends on mitochondrial Ca²⁺ uptake. The interference of salicylate with this mechanism may ultimately provide the basis for the reported anti-proliferative and anti-tumoral effects of aspirin and dietary salicylates. In addition, as Ca²⁺ is a pleiotropic signal, our results point to a novel mechanism that may help to disentangle yet other elusive effects of aspirin such as its anti-inflammatory and immunosuppresive actions.

	Footnotes

 
L. Núñez and R.A. Valero contributed equally to this work.

	References

Top Abstract Introduction Methods Results Discussion References

 
Aceves M, Dueñas A, Gómez C, San Vicente E, Sánchez Crespo M & García-Rodríguez C (2004). A new pharmacological effect of salicylates, inhibition of NFAT-dependent transcription. J Immunol 173, 5721–5729.[Abstract/Free Full Text]

Alvarez J & Montero M (2001). Ca²⁺ measurements with luminescent proteins in the endoplasmic reticulum. In Measuring Calcium and Calmodulin Inside and Outside Cells, ed. Petersen OH, pp. 147–163. Springer Laboratory Manual, Springer, Berlin.

Alvarez J, Montero M & García-Sancho J (1994). Agonist-induced Ca²⁺ influx in human neutrophils is not mediated by production of inositol polyphosphates but by emptying of the intracellular Ca²⁺ stores. Biochem Soc Transact 22, 809–813.[Medline]

Baron JA, Cole BF, Sandler RS, Haile RW, Ahnen D, Bresalier R et al. (2003). A randomized trial of aspirin to prevent colorectal adenomas. N Engl J Med 348, 891–899.[Abstract/Free Full Text]

Bernardi P (1999). Mitochondrial transport of cations: channels, exchangers and permeability transition. Physiol Rev 79, 1127–1155.[Abstract/Free Full Text]

Berridge MJ (1995). Calcium signalling and cell proliferation. Bioessays 17, 491–500.[CrossRef][Medline]

Bosetti C, Gallus S & La Vecchia C (2001). Aspirin and cancer risk, an update to 2001. Eur J Cancer Prev 11, 535–542.

Chan AT (2003). Aspirin, non-steroidal anti-inflammatory drugs and colorectal neoplasia: future challenges in chemoprevention. Cancer Causes Control 14, 413–418.[CrossRef][Medline]

Collins TJ, Berridge MJ, Lipp P & Bootman MD (2002). Mitochondria are morphologically and functionally heterogeneous within cells. EMBO J 21, 1616–1627.[CrossRef][Medline]

Crabtree HG (1929). Observations on the carbohydrate metabolism of tumours. Biochem J 23, 536–545.[Medline]

Duque J, Fresno M & Iñiguez MA (2005). Expression and function of the nuclear factor of activated T cells in colon carcinoma cells: involvement in the regulation of cyclooxygenase-2. J Biol Chem 280, 8686–8693.[Abstract/Free Full Text]

Falck J, Jensen PB & Sehested M (1999). Evidence for repressional role of an inverted CCAAT box in cell cycle-dependent transcription of the human DNA topoisomerase II gene. J Biol Chem 274, 18753–18758.[Abstract/Free Full Text]

Frieden M, James D, Castelbou C, Danckaert A, Martinou JC & Demaurex N (2004). Ca²⁺ homeostasis during mitochondrial fragmentation and perinuclear clustering induced by hFis1. J Biol Chem 279, 22704–22714.[Abstract/Free Full Text]

Gilabert JA, Bakowski D & Parekh AB (2001). Energised mitochondria increase the dynamic range over which inositol 1,4,5-trisphosphate activates store-operated calcium influx. EMBO J 20, 2672–2679.[CrossRef][Medline]

Gilabert JA & Parekh AB (2000). Respiring mitochondria determine the pattern of activation and inactivation of the store-operated Ca²⁺ current I_crac. EMBO J 19, 6401–6407.[CrossRef][Medline]

Glitsch MD, Bakowski D & Parekh AB (2002). Store-operated Ca²⁺ entry depends on mitochondrial Ca²⁺ uptake. EMBO J 21, 6744–6754.[CrossRef][Medline]

Golovina VA, Platoshyn O, Bailey CL, Wang J, Limsuwan A, Sweeney M et al. (2001). Upregulated TRP and enhanced capacitative Ca²⁺ entry in human pulmonary artery myocytes during proliferation. Am J Physiol Heart Circ Physiol 280, H746–H755.[Abstract/Free Full Text]

Gutknecht J (1990). Salicylates and proton transport through lipid bilayer membranes: a model for salicylate-induced uncoupling and swelling in mitochondria. J Membr Biol 115, 253–260.[CrossRef][Medline]

Hanif R, Pittas A, Feng Y, Koutsos MI, Qiao L, Staiano-Coico L et al. (1996). Effects of nonsteroidal anti-inflammatory drugs on proliferation and on induction of apoptosis in colon cancer cells by a prostaglandin-independent pathway. Biochem Pharmacol 52, 237–245.[CrossRef][Medline]

Hardwick JC, Van Santen N, Van Den Brink GR, Van Deventer SJ & Peppelenbosch MP (2004). DNA array analysis of the effects of aspirin on colon cancer cells: involvement of Rac1. Carcinogenesis 25, 1293–1298.[Abstract/Free Full Text]

Holmuhamedov E, Lewis L, Bienengraeber M, Holmuhamedova M, Jahangir A & Terzic A (2002). Suppression of human tumour cell proliferation through mitochondrial targeting. FASEB J 16, 1010–1016.[Abstract/Free Full Text]

Hoth M, Button DC & Lewis RS (2000). Mitochondrial control of calc