Thapsigargin-Induced Gene Expression in Nonexcitable Cells Is ...

Thapsigargin-Induced Gene Expression in Nonexcitable Cells Is Dependent on Calcium Influx

Karin D. Rodland, Robert P. Wersto, Susan Hobson, and Elise C. Kohn Department of Cell and Developmental Biology (K.D.R., S.H.) Oregon Health Sciences University Portland, Oregon 97201-3098 Laboratory of Pathology (R.P.W., E.C.K.), National Cancer Institute Bethesda, Maryland 20892

Agents such as thapsigargin and endothelin elevate intracellular calcium levels by a combination of calcium release from intracellular stores and calcium influx across the plasma membrane; however, the relative contribution of influx vs. release in modulating calcium-dependent gene expression is not as well understood in nonexcitable cells as in excitable cells. In this report we have been able to separate thapsigargin-induced elevation of intracellular calcium into release and influx components, using carboxyamido-triazole (CAI), a known inhibitor of calcium influx with antiproliferative activity against a number of human carcinomas, to selectively inhibit influx without affecting release. The results of these experiments indicate that the ability of thapsigargin to induce calcium-dependent gene expression in nonexcitable cells is dependent on the induction of calcium influx, presumably through store-operated calcium channels. CAI treatment specifically inhibited thapsigargin- or endothelin-stimulated expression from the c-fos promoter in Rat-1 cells and in epithelial cell lines derived from ovary and breast. Use of the VL30 model system confirmed the ability of CAI to inhibit calcium-dependent gene expression and further demonstrated that the ability of elevated calcium to synergize with other signaling pathways required close temporal coupling. In addition to inhibiting endothelin-induced calcium influx, CAI treatment also resulted in a partial inhibition of IP3 production and calcium release. CAI treatment also blocked the increase in ERK1 kinase activity observed in response to either endothelin or thapsigargin, suggesting a role for calcium influx in the activation of mitogen-activated protein kinase pathways. (Molecular Endocrinology 11: 281–291, 1997)

INTRODUCTION Calcium is an ubiquitous intracellular messenger and regulator of cellular activities. As both a second messenger and a modulator of selected signal transduction pathways, calcium has the potential to modulate a variety of cellular functions, including proliferation, cell-substratum interaction, cytoskeletal organization, and gene expression (1–3). Elevation of intracellular calcium has been associated with the induction of a variety of proliferation-associated immediate early genes, including c-fos, c-jun, and VL30 (4–7). In the case of neuronal cells, c-fos expression appears to be regulated at the transcriptional level by two distinct calcium-dependent pathways. Calcium influx through voltage-sensitive L-type calcium channels results in the phosphorylation of serine 133 on CREB (the cAMP response element binding protein), presumably through a Ca2⫹-calmodulin kinase-dependent pathway (8–10). In contrast, calcium influx through the N-methyl-D-aspartate receptor activates c-fos transcription through the serum response element and appears to involve phosphorylation of Elk-1 in a Ras-dependent fashion (11). Growth factor-mediated induction of c-fos gene expression appears to utilize a similar Ras-dependent phosphorylation of Elk-1, and potentially CREB (12). Elevated intracellular calcium has also been shown to promote elongation of c-fos transcripts via an attenuator sequence in the first intron (13). The importance of these pathways in regulating cfos expression in nonexcitable cells has not been well characterized. The ability of endothelin-1 to induce c-fos expression in Rat-1 fibroblasts is dependent upon the elevation of intracellular calcium, as demonstrated by the inhibitory effect of intracellular calcium chelators (6). Intracellular calcium must be elevated above a threshold of 200 nM in order for endothelin-1 to induce expression of the immediate early gene VL30 (7). Endothelin-1 is known to activate both calcium

0888-8809/97/$3.00/0 Molecular Endocrinology Copyright © 1997 by The Endocrine Society

281

MOL ENDO · 1997 282

Vol 11 No. 3

Fig. 1. Effect of CAI Treatment on Thapsigargin-Induced Changes in Intracellular Calcium Levels Rat-1 TK3R-3 cells were grown on Pronectin-coated coverslips, serum-starved, and loaded with Fura-2-AM as described in Materials and Methods. Cells were exposed to DMSO (panel A) or 10 ␮M CAI (panel B) for 4 h, then stimulated with 2 ␮M thapsigargin (TG), and changes in intracellular calcium concentration were determined as described in Materials and Methods. Results presented are mean ⫾ SD of n ⱖ 35 cells per experiment, representative of n ⫽ 3 experiments.

release and calcium influx in Rat-1 cells (14, 15). The ability of endothelin-1 to elevate intracellular calcium can be mimicked by treating cells with thapsigargin, an inhibitor of the Ca2⫹-ATPase responsible for sequestering calcium in intracellular stores (16, 17). Thapsigargin treatment leads to both increased calcium release and calcium influx across the plasma membrane (16, 18). The relative contribution of calcium release vs. calcium influx in the regulation of either c-fos or VL30 expression has not been determined. Carboxyamido-triazole (CAI) is a novel inhibitor of non-voltage-gated calcium influx in response to a variety of agonists including carbachol (19), maitotoxin [a nonionophoretic stimulator of calcium influx (20)], and the ionophore A23187 (21). In the concentration range that inhibits calcium influx (1–10 ␮M), CAI has demonstrated antiproliferative and antimetastatic properties (19, 21–24). Structure-activity investigations demonstrate that the same moieties of the CAI molecule required for modulation of calcium influx are also required for the antiproliferative actions of CAI (23). In this report we use CAI to selectively inhibit the influx component of the calcium elevation observed in response to thapsigargin treatment and further demonstrate that calcium influx is required for the transcriptional induction of the immediate early genes c-fos and VL30 by thapsigargin. We also demonstrate a previously undocumented ability of CAI to inhibit calcium release and inositol triphosphate (IP3) generation in Rat-1 fibroblasts stimulated with endothelin-1. These data suggest that the antiproliferative activity of CAI may be a consequence of CAI’s ability to inhibit calcium-dependent expression of immediate early genes required for cellular proliferation.

RESULTS Effect of CAI Treatment on Intracellular Ca2ⴙ Levels in Rat-1 Cells CAI has been previously characterized as a specific inhibitor of calcium influx (19, 21). In an attempt to selectively inhibit calcium influx in response to endothelin-1 and thapsigargin without affecting calcium release, we determined the effect of CAI pretreatment on intracellular calcium levels in Rat-1 cells treated with either thapsigargin or endothelin-1. Thapsigargin treatment produced a gradual rise in intracellular calcium concentration attributed to release from internal stores, followed by a slow return to basal levels; this calcium shoulder is a consequence of calcium influx through store-operated calcium channels (SOCC1; Refs. 16 and 18). Exposure to CAI minimally reduced the internal release component observed in response to thapsigargin treatment but markedly attenuated the component attributable to calcium influx through SOCC (Fig. 1, panels A and B). This is the first demonstration that CAI inhibits the store-operated calcium channel. CAI treatment had a pronounced effect on the magnitude of the calcium transient observed in response to endothelin-1, whether measurements were made in the presence or absence of extracellular calcium (Fig. 2, panels A and B). This would indicate that CAI is able to inhibit intracellular calcium release in response to 1 SOCC is used to denote the channel responsible for calcium influx observed after depletion of intracellular calcium stores; this channel is also referred to as a ‘refilling’ or ‘capacitance’ channel (18). ICRAC is one example of a SOCC (25).

Inhibition of Calcium-Dependent Gene Expression

283

endothelin-1. In the presence of extracellular calcium, there is a sustained elevation of intracellular calcium after endothelin-1 treatment, and this has been attributed to calcium influx (14, 15). CAI treatment eliminated this sustained increase (as did calcium-free external medium), indicating that CAI also inhibited endothelin-1-mediated calcium influx, possibly via the SOCC. To provide a potential explanation for the ability of CAI to inhibit endothelin-mediated calcium release in Rat-1 cells, we measured the effects of CAI treatment on IP3 production in response to endothelin. Rat-1 cells exposed to 10 ␮M CAI for 16 h displayed a significant inhibition of IP3 production measured 20 min after endothelin-1 treatment (45% inhibition, P ⱕ 0.03; Fig. 2C). Although the ability of CAI to inhibit thapsigargininduced calcium influx without affecting calcium release provides a useful tool for distinguishing between these two components of the intracellular calcium response to thapsigargin, the observation that CAI inhibited both calcium release and influx in response to endothelin precluded us from using CAI to separate these two components of endothelin signaling. However, when the functionally similar but structurally distinct compound SK&F 96365 was tested in a similar fashion, SK&F 96365 was observed to inhibit only the influx component of the endothelin response (Fig. 2D). This result is in accord with the published effects of SK&F 96365 on calcium influx through refilling channels (26) and provides a method for inhibiting endothelin-mediated calcium influx independent of calcium release.

Fig. 2. Effect of CAI Treatment on Endothelin-1-Mediated Signaling Panels A and B, Effects of CAI on intracellular calcium levels. Rat-1 TK3R-3 cells were grown on coated coverslips, serum-starved, and loaded with Fura-2-AM as described above. Cells were exposed to DMSO (open circles) or 10 ␮M

CAI (closed circles) for 4 h, then stimulated with endothelin-1 at 10⫺8 M (ET). Changes in intracellular calcium concentration were measured in medium containing 2 mM Ca2⫹ (panel A) or 0 mM Ca2⫹ 0.5 mM EGTA (panel B). Results presented are mean ⫾ SD of n ⱖ 30 cells; the peak Cai in response to CAI ⫹ ET is offset relative to ET alone for clarity. Panel C, Effect of CAI on IP3 production. Rat-1 cells were grown to confluence in 10-cm plates, then serum-deprived in the presence of 3 ␮Ci [3H]myo-inositol as described in Materials and Methods. After a 16-h exposure to either CAI (black bars) or DMSO (open bars), Rat-1 cells were exposed to either endothelin-1 (10⫺8 M) or vehicular control for 20 min in the presence of 100 mM LiCl. Inositol phosphates were extracted in formic acid and fractionated on Dowex-formate columns as described in Materials and Methods. Bars represent the mean ⫾ SD of pooled IP3 fractions obtained from triplicate plates; similar results were obtained in three independent experiments. Panel D, Effect of SK&F 96365 on intracellular calcium levels. Rat-1 TK3R-3 cells were grown on coated coverslips, serumstarved, and loaded with Fura-2-AM as described above. Cells were exposed to DMSO (open circles) or 50 ␮M SK&F 96365 (closed circles) for 16 h, then stimulated with endothelin-1 at 10⫺8 M (ET). Changes in intracellular calcium concentration were measured in medium containing 2 mM Ca2⫹. Results presented are mean ⫾ SE of n ⱖ 20 cells.

MOL ENDO · 1997 284

Inhibition of c-fos Expression by CAI To determine whether the dual ability of CAI to inhibit the elevation of intracellular calcium in response to endothelin-1 and to inhibit thapsigargin-stimulated calcium influx could affect calcium-sensitive gene expression, we tested the ability of CAI treatment to modulate expression of FC2CAT. These experiments used Rat-1 cells stably transfected with the FC2CAT construct containing 1.4 kb of the c-fos promoter, including the serum response element and calcium response element/cAMP response element enhancer elements, which are known to be responsive to calcium-dependent signals (27). Endothelin-1 is a potent inducer of FC2CAT expression in Rat-1 cells, as illustrated in Fig. 3 (lanes 3 and 13). As endothelin-1 is known to signal by both calcium mobilization and activation of protein kinase C, each of these component parts was tested individually, using thapsigargin to elevate intracellular calcium and 12-O-tetradecanoylphorbol-13-acetate (TPA) to activate protein kinase C. Although treatment with thapsigargin alone failed to induce FC2CAT expression over control levels (lanes 5 and 15), cotreatment with TPA plus thapsigargin produced a 2- to 8-fold increase in FC2CAT expression (Fig. 3, lanes 9 and 19). This result suggests that induction of c-fos by endothelin-1 requires concurrent activation of both protein kinase C- and calcium-dependent pathways. In the presence of 10 ␮M CAI, FC2CAT expression was inhibited to nearly basal levels whether the inducing agent was endothelin-1 or combined TPA ⫹ TG

Vol 11 No. 3

(Fig. 3A, lanes 4 and 10). Treatment with SK&F 96365 also inhibited calcium-dependent induction from the c-fos promoter (Fig. 3B). In Fig. 2D, we demonstrated that SK&F 96365 had little or no effect on calcium release in response to endothelin-1, but did reduce the sustained calcium threshold attributed to calcium influx through the SOCC. These results suggest that the initial elevation of intracellular calcium seen after either endothelin-1 or thapsigargin treatment is insufficient to induce c-fos expression, but the sustained increase dependent on activation of SOCC is essential for induction. Because elevated calcium has been implicated not only in the transcriptional induction of c-fos but also in the sustained elongation of nascent transcripts (13), it is important to determine independently the effects of CAI on the accumulation of full length c-fos mRNA. Analysis of c-fos mRNA, as opposed to chloramphenicol acetyl transferase (CAT) protein, also controls for any possible effects of thapsigargin on protein synthesis (28, 29). Endogenous c-fos mRNA accumulation was measured in the parental Rat-1 cells that had been exposed to either 10 ␮M CAI or a vehicular control before stimulation with either endothelin-1 or TPA ⫹ thapsigargin for 20 min. Induction of c-fos by either endothelin-1 or TPA ⫹ thapsigargin was completely inhibited after exposure to CAI for 20 h (Fig. 4, lanes 7 and 9 compared with lanes 2 and 4); an approximately 50% reduction in c-fos mRNA levels was seen after a 4-h exposure (data not shown). CAI treatment had a negligible effect on the accumulation of cyclophilin mRNA (Fig. 4, lower panel; see also Fig. 5A). Cyclo-

Fig. 3. Inhibition of c-fos Expression by CAI in Rat-1 Cells Panel A, Inhibition of FC2CAT expression after CAI pretreatment. Rat-1 cells stably transfected with the FC2CAT plasmid, as described in Materials and Methods, were grown to confluence in 10-cm plates as described in Fig. 1. Cells were serum-deprived for 24 h before addition of either DMSO (0.5%; open bars) or CAI (10 ␮M; hatched bars) for a 4-h pretreatment. The pretreatment medium was replaced with fresh DMEM, and agonists were added as indicated; cells were harvested for analysis of CAT activity 4 h later. Results are mean ⫾ SD, n ⫽ 3, and similar results were obtained in three experiments. Panel B, Inhibition of FC2CAT expression after SK&F 96365 pretreatment. FC2CAT Rat1 cells were serum-deprived for 24 h before addition of either DMSO (0.5%; open bars) or SK&F 96365 (50 ␮M; hatched bars) for 4 h pretreatment. The pretreatment medium was replaced with fresh DMEM, and agonists were added as indicated; cells were harvested for analysis of CAT activity 4 h later. Results are mean ⫾ SD, n ⫽ 3. Similar fold induction and percent inhibition were observed in three replicate experiments.


Fig. 4. Inhibition of c-fos mRNA Accumulation by CAI Rat-1 cells were grown to confluence in 10-cm plates, then incubated in serum-free medium containing either 10 ␮M CAI (lanes 6–10) or 0.1% DMSO (lanes 1–5) for 24 h before the addition of endothelin-1 (10⫺8 M), TPA (100 ng/ml), thapsigargin (2 ␮M), or TPA ⫹ thapsigargin where indicated. RNA was extracted and processed for Northern hybridization analysis as previously described (30). The 2.2-kb mRNA hybridizing to a c-fos riboprobe (pGem3Z-cFos, obtained from D. Pribnow) is indicated. Hybridization to the constitutively expressed gene cyclophilin indicated that equivalent amounts of RNA were present in each lane (lower panel). 32P was visualized by exposure to a PhosphorImager screen (Molecular Dynamics) for 16 h.

philin is peptidyl-prolyl cis-trans isomerase, a ubiquitously expressed gene required for collagen synthesis (31, 32). To determine whether the ability of CAI to inhibit c-fos expression in murine fibroblasts was a general phenomenon of nonexcitable cells, we tested the ability of CAI to modulate c-fos mRNA and protein expression in two human cell lines representing nonexcitable epithelial cell types. The human ovarian carcinoma cell line SKOV-3 has low endogenous levels of c-fos mRNA expression, but responds to 10⫺8 M endothelin-1 with a substantial induction of c-fos mRNA (Fig. 5A). Treatment of SKOV-3 cells with CAI at 10 ␮M for 20 h before addition of endothelin-1 completely inhibited c-fos mRNA accumulation (Fig. 5A). A similar reduction in serum-induced c-fos mRNA levels was observed after CAI treatment of MCF-10F cells, an immortalized but nonmalignant line of human mammary epithelial cells (Fig. 5B). Inhibition of VL30 Expression after CAI Exposure To extend our observations to another proliferationassociated gene with a well characterized induction in response to calcium, we tested the effects of CAI and SK&F 96365 on the VL30 calcium-sensitive enhancer. The responsiveness of the VL30 enhancer to various treatments can be quantitatively studied in the TK3RCAT cell line (described in Ref. 33), a Rat-1 derivative line containing a stably integrated VL30 enhancer-CAT reporter construct. When TK3R-CAT cells are exposed to either epidermal growth factor (EGF), TPA, or thapsigargin as single agonists, a 3-fold or less increase in CAT activity is observed; simultaneous addition of thapsigargin with either EGF or TPA produces a marked and statistically synergistic increase in TK3RCAT expression (Fig. 6 and Refs. 17 and 33). Endo-

285

Fig. 5. Inhibition of c-fos Expression by CAI in Epithelial Cell Lines Panel A, Inhibition of c-fos mRNA accumulation by CAI in SKOV-3 cells. Confluent cultures of SKOV-3 cells were serum deprived for 24 h, then exposed to 10 ␮M CAI for 20 h before addition of endothelin-1 at 10⫺8 M. RNA was harvested 20 min after endothelin-1 addition. RNA processing and Northern hybridization analysis were conducted as described in Fig. 4. Similar results were obtained in two replicate experiments. Panel B, Inhibition of c-fos mRNA accumulation by CAI in MCF-10F cells. MCF-10F immortalized but nonmalignant human mammary epithelial cells were grown to confluence in the presence of 10% bovine calf serum (HyClone). CAI was added at a final concentration of 10 ␮M 24 h before harvesting as indicated. RNA was harvested, processed, and subjected to Northern hybridization analysis as described in Fig. 4. Similar results were obtained in two experiments.

thelin-1, which is known to elevate intracellular calcium, activate protein kinase C, and activate the EGF receptor (14, 34), stimulates TK3R-CAT expression even better than the combination of TPA and thapsigargin (Fig. 6, lane 3 compared with lane 9). This increased response may be attributed to the additional ability of endothelin-1 to activate the EGF receptor in Rat-1 cells (34). To determine whether the known ability of CAI to modulate changes in intracellular calcium in response to endothelin-1 and thapsigargin (Figs. 1 and 2) would influence TK3R-CAT expression, we exposed TK3R-CAT cells to CAI for up to 24 h before the addition of agonists. Induction of TK3R-CAT expression by endothelin-1 was significantly inhibited by CAI pretreatment (P ⬍ 0.01) as shown in Fig. 6 (hatched bars). The ability of thapsigargin to stimulate TK3RCAT expression in conjunction with either TPA, EGF, or cAMP was also significantly inhibited (P ⬍ 0.01, Fig. 6). These inhibitory effects were observed even when the CAI-containing medium from the 20-h pretreatment was replaced with fresh, CAI-free medium concurrent with agonist addition, indicating that the continuous presence of CAI may not be required for

MOL ENDO · 1997 286

Fig. 6. Inhibition of Calcium-Dependent VL30 Expression by CAI The Rat-1-derived TK3R-3 cell line, stably transfected with pCBTK3R-CAT, was grown to confluency in 12-well plates in the presence of DMEM ⫹ 10% defined calf serum (HyClone). Cells were serum-deprived in DMEM for 72 h before agonist addition. Cells were exposed to either 0.5% DMSO (hatched bars) or 10 ␮M CAI (in 0.5% DMSO; open bars) for 20 h before addition of agonists. The CAI and DMSO-containing media were aspirated and replaced with fresh DMEM (without CAI) before addition of agonists as indicated. After 4 h, cells were lysed, and CAT activity in the cell extracts was assayed as described in Materials and Methods. DMEM, No agonists; ET, 10⫺8 M endothelin-1; TG, 2 ␮M thapsigargin; TPA, 100 ng/ml TPA; cAMP, 600 ␮M (Bu)2cAMP. Results presented are mean ⫾ SD, n ⫽ 3. Similar fold induction and percent inhibition were observed in four replicate experiments.

inhibition. The ability of CAI to inhibit combinations involving cAMP and either EGF or TPA was also tested to determine whether the inhibitory effects of CAI were specific for thapsigargin; CAI failed to inhibit combinations that did not include thapsigargin (Fig. 6B). Similar effects were observed when the calcium ionophores A23187 or ionomycin were substituted for thapsigargin (data not shown). These results suggest that the inhibitory effects of CAI on VL30 expression are selective for interactions involving elevated intracellular calcium. Further support for calcium influx as a major requirement for VL30 expression in response to endothelin-1 or thapsigargin was obtained in experiments in which SK&F 96365 was substituted for CAI. As was observed for c-fos expression in Fig. 3, 50 ␮M SK&F 96365 inhibited TK3R-CAT expression in response to endothelin-1 or combinations of thapsigargin and either EGF or TPA (Fig. 7, light bars), although the magnitude of the inhibition was less than had been observed after CAI treatment (compare Fig. 6, lanes 3, 4 and 9, 10 with Fig. 7, lanes 3, 4 and 7, 8). This result may reflect the generally lower potency of SK&F 96365 compared with CAI (20); in the case of endothelin-induced VL30 expression the difference may also represent the contribution of endothelin-induced calcium release, which is not inhibited by SK&F 96365.

Vol 11 No. 3

Fig. 7. Effect of SK&F 96365 Treatment on Calcium-Dependent TK3R-CAT Expression TK3R-3 cells were cultured and serum-starved as described. SKF treatment (open bars) was achieved by adding 50 ␮M SK&F 96365 to media for 20 h before agonist addition. All cells were changed into fresh serum-free DMEM immediately before addition of agonists, and cells were harvested 4 h later. Results shown are mean ⫾ SD, n ⫽ 3, and similar results were obtained in three replicate experiments.

Fig. 8. Effects of DBHQ on Calcium-Dependent TK3R-CAT Expression TK3R-3 cells grown in DMEM as described were exposed to agonists as indicated and harvested for CAT activity after 4 h. Control cells received 0.5% DMSO during stimulation in addition to other agonists as indicated, while co-DBHQ received 10 ␮M DBHQ added concurrently with other treatments, and maintained throughout the 4-h agonist treatment. In the DBHQ-removed group, cells were exposed to 10 ␮M DBHQ for 15 min, after which the DBHQ was removed and cells placed in DMEM for 15 min before addition of other agonists: DMSO, 0.05%, open bars; EGF, 10 ng/ml, crosshatched bars; TPA, 160 nM, striped bars; cAMP, 600 ␮M, black bars. Results are mean ⫾ SD, n ⫽ 3. Similar results were obtained with wash-out periods of 30, 60, and 120 min (data not shown).

Effects of a Reversible Inhibitor of Calcium Sequestration on TK3R-CAT Expression To determine whether the elevation of intracellular calcium must occur contemporaneously with the other agonists to produce the synergistic increase in VL30 expression, we used the reversible Ca2⫹-ATPase in-


hibitor 2,5-di-tert-butylhydroquinone (DBHQ) (35). TK3R-CAT cells were exposed to EGF, TPA, or cAMP either singly (Fig. 8, lanes 1–4), in the continuous presence of 10 ␮M DBHQ (lanes 5–8), or 15 min after DBHQ had been removed from the medium (lanes 9–12). Exposure to DBHQ alone for either 15 min or 4 h produced a 5-fold increase in TK3R-CAT expression, which was sustained throughout a 4-h wash-out period. Coaddition of either EGF, TPA, or cAMP throughout the DBHQ exposure produced a greater than additive increase in CAT expression similar to that observed for the combination of thapsigargin and the same agonists. However, removal of the DBHQ before addition of the other agonists resulted in levels of CAT expression indistinguishable from those induced by DBHQ alone. This result implies that elevation of intracellular calcium must be coordinated with the activation of other signaling pathways to produce a synergistic increase in gene expression. Effect of CAI on MAP Kinase Activation Activation of c-fos expression through the SRE is known to be influenced by the MAP kinase-mediated phophorylation of Elk-1 (11, 12). To determine whether this pathway of c-fos induction might be inhibited as a consequence of CAI treatment, we conducted in vitro kinase assays using immunoprecitated ERK-1 and synthetic GST-ELK as a substrate. Endothelin-1 acted as a potent activator of ERK1 kinase activity, which was increased approximately 15- to 20-fold 20 min after exposure to 10⫺8 M endothelin-1 (Fig. 9A, lane 3 compared with lane 1); in comparison, 10 ng/ml EGF for 20 min produced a 40- to 50-fold increase in ERK1 activity (lane 5 compared with lane 1). Pretreatment with CAI for 16 h completely inhibited the activation of ERK-1 by endothelin (Fig. 9, lane 4 compared with lane 3) but had no signficant effect on the EGF-induced ERK1 activation (Fig. 9, lane 6 compared with lane 5). This inhibition of endothelin-mediated, but not EGFmediated, ERK1 activation correlates well with the

287

differential effects of CAI treatment on endothelin-induced vs. EGF-induced gene expression. Neither TPA alone nor thapsigargin alone was highly effective in activating ERK1, but the combination of TPA ⫹ thapsigargin produced an increase in ERK-1 activity comparable to that observed in response to endothelin-1 (Fig. 9B, lanes 1–4). CAI treatment had little or no effect on the TPA-mediated induction (lane 7 compared with lane 3) but reduced the thapsigarginmediated response to control levels (lane 8 compared lanes 1 and 4). In the presence of CAI, the combined TPA ⫹ thapsigargin response was inhibited approximately 50% (lane 6 compared with lane 2), consistent with the relative lack of CAI effect on TPA-stimulated ERK1 activity.

DISCUSSION Calcium has been shown to be an important messenger in the regulation of several genes. In this report we show that CAI and SK&F 96365, two structurally unrelated inhibitors of calcium influx, can inhibit the calcium-dependent induction of the proliferation-associated genes c-fos and VL30. Neither compound had a significant effect on the calcium-independent expression of these two genes. This report also contains the first published demonstration that CAI can specifically inhibit store-operated calcium channels. CAI has been previously shown to inhibit non-voltage-gated calcium influx across the plasma membrane through receptoroperated and ionophoretic channels (20, 21, 23). Thapsigargin is a selective activator of SOCC activity subsequent to the depletion of intracellular calcium stores (16, 36, 37). The ability of CAI to inhibit thapsigargin-dependent FC2-CAT and TK3R-CAT expression can be attributed to the marked reduction in thapsigargin-mediated SOCC activity observed in the presence of CAI (Fig. 1). Confirmation that inhibition of calcium influx is responsible for the reduction in TK3RCAT activity is provided by studies using SK&F 96365,

Fig. 9. Effect of CAI Treatment on ERK1 Activity Rat-1 cells were grown to confluence in 10-cm plates, then serum-deprived for 24 h in either 10 ␮M CAI or 0.1% DMSO, as indicated. Cells were exposed to the indicated agonists (ET-1, 10⫺8 M; EGF, 10 ng/ml, TPA, 100 ng/ml; TG, 2 ␮M) for 20 min. Cells were lysed and cell lysates (normalized to equal amount of protein) incubated overnight with anti-ERK1 antibodies as described in Materials and Methods. The immunoprecepitated proteins were used in in vitro kinase assays and phosphorylated GST-Elk-1 visualized with a PhosphorImager (panel A, 4 h exposure; panel B, 16 h exposure). Panel A represents one complete set from triplicate lysates, while panel B represents one set from duplicate lysates. Similar results were obtained in three independent experiments.

MOL ENDO · 1997 288

an agent known to inhibit both receptor-operated calcium influx and SOCC activity (28, 38, 39). The inhibition of thapsigargin-stimulated VL30 and c-fos expression by CAI and SK&F 96365 suggests that transmembrane calcium influx, most likely due to SOCC activity in these cells, is an important component of the calcium-signaling pathways leading to induction of c-fos and VL30 gene expression. Results with the reversible inhibitor DBHQ would suggest that a sustained elevation of intracellular calcium concurrent with the stimulation of other pathways, such as protein kinase C, is required for the synergistic induction of VL30. The ability of endothelin-1 to induce both VL30 and c-fos expression has been linked to elevation of intracellular calcium above a threshold of 200 nM (6, 7). Endothelin-1 treatment induces both a rapid release of intracellular calcium, producing peak values over 700 nM, and a sustained elevation of intracellular calcium over 200 nM, attributable to calcium influx (Fig. 8 and Refs. 7, 14, 15, 40, and 41). In this report we demonstrate that the sustained threshold elevation of intracellular calcium dependent on calcium influx is required for the induction of immediate early genes in response to endothelin. CAI treatment has been previously shown to inhibit carbachol-induced IP3 production in CHO cells expressing the m5 acetylcholine receptor by about 40% (20), a reduction that is similar to that observed in endothelin-stimulated Rat-1 cells exposed to CAI (Fig. 2C). In endothelin-stimulated Rat-1 cells treated with CAI, this 45% inhibition of IP3 release was associated with an 86% decrease in peak intracellular Ca2⫹ release (Fig. 2, A and B); such a substantial inhibition of intracellular Ca2⫹ release in response to CAI has not been reported in any other system. Dose-response curves for the effect of endothelin on IP3 production and peak intracellular calcium concentration in Rat-1 cells indicate that calcium release is somewhat more sensitive to endothelin concentration than is IP3 production (15). At an endothelin concentration producing half-maximal IP3 levels (0.3 nM), intracellular calcium concentration is only 30% of the peak value observed at 10 nM (15). Thus the partial inhibition of IP3 production observed in response to CAI treatment may result in a proportionately greater reduction in Ca2⫹ release. It is also possible that the nearly complete inhibition of endothelin-stimulated Ca2⫹ release observed in the presence of CAI may reflect a previously undescribed effect of CAI independent of direct effects on calcium influx. In this report, we have demonstrated that the calcium-dependent induction of both VL30 and c-fos gene expression can be inhibited by cellular CAI exposure, despite the substantial structural differences in the regulatory elements of these two genes. The calciumresponsive sequence elements of c-fos and VL30 are functionally distinct, in that neither sequence will compete for protein binding by the other (33, 42). While the

Vol 11 No. 3

ability of elevated intracellular calcium to induce c-fos and VL30 expression has been well studied, particularly regulation of c-fos in excitable cells, little is known about the relative importance of calcium influx vs. calcium release in modulating expression in nonexcitable cells. The ability of CAI and SK&F 96365 to inhibit the disparate calcium-responsive enhancers of c-fos and VL30 suggests that calcium influx, including SOCC-mediated influx, may be effective generally as an inducer of calcium-dependent gene expression in nonexcitable cells, regardless of the specific transcription factors ultimately responsible for transactivation. The inhibition of ERK1’s ability to phosphhorylate Elk-1 observed after CAI treatment suggests that this pathway is responsive to changes in intracellular calcium concentration, particularly the sustained elevation attributed to SOCC activity and inhibited by both CAI and SK&F 96365. Although the signaling intermediary directly responsive to changes in intracellular calcium has not been identified in these studies, inhibition of Elk-1 phosphorylation does provide a a potential model for the inhibition of c-fos expression by CAI and SK&F and suggests that the activation of c-fos expression in Rat-1 cells may utilize a similar Ras-dependent pathway as that activated by calcium influx through the N-methyl-D-aspartate receptor in PC-12 cells (11). Calcium-dependent expression of c-fos has the potential to modulate expression of other genes such as c-jun (43), interleukin-2 (44, 45), proenkephalin (46), and many members of the matrix metalloproteinase family (47, 48). Kohn et al. (24) have previously demonstrated that expression of matrix metalloproteinase (MMP)-1/interstitial collagenase and MMP-2/type IV collagenase can be inhibited by CAI in the same concentration range that inhibits calcium influx. Expression of MMP-1/interstitial collagenase is known to have a c-fos-dependent component (48, 49), and the results of these studies suggest that inhibition of MMP-1/interstitial collagenase expression by CAI may be attributed to inhibition of c-fos expression. Other members of this proteinase family with AP-1 regulatory sites include the newly described matrix metalloproteinase MT-MMP (50), MMP-9 (51), matrilysin (52), and stromelysin (53). The broad effect of CAI on the expression of the MMPs is consistent with its demonstrated antiinvasive and antiangiogenic effects (19, 24, 54). The complex signaling pathways regulating cellular function are now being elucidated by dual strategies focusing, respectively, on the functional consequences of gene expression and on the propagation of signaling events from the plasma membrane to the nucleus through various cytoplasmic cascades. The use of CAI in this study combines these two concepts in demonstrating that modulation of calcium influx can be critical in driving the transcription of genes associated with cell proliferation. The ability to modulate c-fos expression


has broader implications as suggested by the ability of CAI to inhibit gene expression of MMPs with known AP-1-dependent regulation. The ability of CAI to inhibit calcium-dependent gene expression provides a unifying explanation for the biological effects of CAI on proliferation, invasion, and angiogenesis.

MATERIALS AND METHODS Cell Culture

289

nm and calibrated in situ. All measurements were acquired at room temperature. Analysis of c-fos mRNA Expression Confluent 10-cm plates of Rat-1 cells were serum-deprived for 36 h before the addition of experimental agents. Cell lysates for RNA analysis were prepared at specified times after agonist addition by the addition of a LiCl-urea-SDS lysis buffer as previously described (14). RNA was processed and subjected to Northern hybridization analysis, using a [32P]CTP-labeled c-fos riboprobe transcribed from the plasmid pGEM3Z-cFos (obtained from D. Pribnow, Oregon Health Sciences University, Portland OR, Ref. 6). Measurement of IP3 Levels

The FC2-Rat1 cell line was created by stable transfection of the FC2CAT plasmid (obtained from I. Verma, Salk Institute, La Jolla, CA) into Rat-1 cells. Clonal isolates of G418-resistant FC2-Rat1 cells were cultured and tested for responsiveness to serum; the FC2-Rat1 cell line used in these experiments displays a 30-fold increase in CAT expression after serum stimulation. The Rat-1-derived TK3R-3 cell line, stably transfected with the VL30 enhancer element linked to the chloramphenicol acetyltransferase reporter gene in the context of the herpes simplex thymidine kinase promoter, has been described previously (33). TK3R-3 and FC2-Rat1 cells were maintained in DMEM plus 10% defined calf serum (Hyclone, Logan UT) in 37 C, 95% air/5% CO2. Cultures were supplemented with 10 ␮g/ml gentamicin, 1.75 mg/ml amphotericin B (Fungizone, Sigma, St. Louis MO), and 750 ␮g/ml G418 (GIBCO BRL, Gaithersburg, MD) was used for selection, instead of neomycin.

Confluent 10-cm plates of Rat-1 cells were labeled with [3H]myo-inositol (3 ␮Ci/ml) for 36 h, followed by a 24-h incubation in serum-free DMEM lacking [3H]myo-inositol and containing either dimethylsulfoxide (DMSO) (0.1%) or 10 ␮M CAI. Cells were exposed to either 10⫺8 M endothelin-1 or vehicular control (0.05 mM HCl) for 20 min in the presence of 100 mM LiCl. Cells were extracted in 0.2 M formic acid, then diluted to 50 mM formic acid-100 mM ammonium formate by addition of NH4OH and H2O. Extracts were applied to Dowex 1 ion exchange columns and IP1 and IP2 were removed by multiple washes in 0.1 M formic acid/0.5 M NH4 formate. Elutions of IP3 in 0.1 M formic acid/0.75 M NH4 formate and IP4 in 0.1 M formic acid/1.1 M NH4 formate were collected and quantified by liquid scintillation counting, as previously described (14).

Measurement of CAT Activity

Measurement of MAP Kinase Activity

The desired cell lines were grown to confluence in 12-well plates and subjected to serum deprivation for 36 h (FC2CAT) or up to 72 h (TK3R-3) to induce quiescence before use. Agonists were added as described and cells harvested 4 h later, to allow sufficient time for synthesis of newly induced CAT protein. Cellular proteins were extracted in 10 mM TrisHCl, 0.05% Triton X-100, and the cell extracts were incubated for 10 min at 70 C to inhibit cellular acetylases as previously described (17). CAT activity was measured by the two-phase CAT assay of Neumann et al. (55), and a rate of enzyme activity was determined from the increase in [3H]acetyl-coenzyme A incorporated over time. Only data from the linear portion of the reaction curve were used.

Confluent Rat-1 cells in 10 cm plates were cultured in serumfree DMEM containing either DMSO (0.1%) or 10 ␮M CAI for 24 h before the addition of the desired agonists (endothelin-1, TPA, thapsigargin, or EGF). After 20 min of agonist stimulation, the medium was aspirated and cells were immediately lysed in 750 ␮l HEPES-KOH lysis buffer [20 mM HEPES-KOH, 2 mM EGTA, 50 mM ␤-glycerophosphate, 10% glycerol, 1% Triton X-100, 1 mM dithiothreitol (DTT), 1 mM vanadate, 0.4 mM phenylmethylsulfonyl fluoride, 0.5 ␮g/ml aprotinin, 0.5 ␮g/ml leupeptin]. Lysates were cleared by addition of 10 ␮l protein A agarose and centrifugation. Aliquots of cleared lysates normalized for protein content were subjected to immunoprecitation overnight at 4 C using anti-ERK-1 (Santa Cruz Biotechnology, Santa Cruz, CA), followed by addition of protein A agarose and an additional 2 h incubation at 4 C. Immunoprecipitates were recovered by centrifugation and washed once in HEPES-KOH lysis buffer, once in LiCl buffer [500 mM LiCl, 100 mM Tris HCl, pH 7.6, 0.1% Triton X-100, 1 mM DTT, 1 mM vanadate, 0.4 mM phenylmethylsulfonyl fluoride] and once in 3-[N-morpholino]propanesulfonic acid (MOPS) assay buffer [20 mM MOPS, pH 7.2, 20 mM MgCl2, 2 mM EGTA, 2 mM DTT, 0.2% Triton X-100]. The pellets were resuspended in 20 ␮l kinase assay buffer [10 mM MOPS, pH 7.2, 20 mM MgCl2, 1 mM EGTA, 1 mM DTT, 0.1% Triton X-100, 3 ␮g GST-Elk1, 1 ␮Ci 32P-␥-ATP] and incubated at 30 C for 20 min. Phosphorylated proteins were resolved by SDSPAGE in 12% acrylamide gels. Gels were dried and radioactivity visualized and quantitated with a Molecular Dynamics PhosphorImager (Molecular Dynamics, Sunnyvale, CA) and IP LabGel software.

Measurement of Intracellular Calcium Intracellular Ca2⫹ concentrations were measured using the Ca2⫹-sensitive fluorescent dye, Fura-2, as previously described (14, 56). Rat-1 TK3R-3 cells were grown in gelatincoated coverslip chambers (Lab-Tek, Naperville IL) and serum deprived for 48 h before loading with the acetoxyester of Fura-2 (Fura-2/AM, Molecular Probes Inc, Eugene OR; final concentration 1 ␮M) for 30 min at 37 C. An equal volume of 20% Pluronic F-127 (Molecular Probes, Inc., Eugene OR) was added to facilitate Fura-2 incorporation. After loading, cellular fluorescence was measured at 400⫻ using a using a Nikon inverted microscope coupled to a CCD camera (Videoscope Int., Herndon VA), and the images were analyzed using either the Image 1/Fluor software package (Universal Imaging, West Chester, PA) or the InCa2-double wavelength package (Intracellular Imaging, Inc., Cincinnati, OH). In each experiment, at least 30 cells per field were measured. Intracellular calcium concentrations were quantified in Fura-2-loaded cells by the fluorescence ratio method (58), using the ratio of fluorescence emission at 510 nm from excitation at 340 nm and 380

Reagents Thapsigargin (LC Services, Woburn, MA) was dissolved in dimethyl sulfoxide (DMSO) and stored at ⫺20 C before use.

MOL ENDO · 1997 290

Endothelin-1 was obtained from Peptides International (Louisville, KY) and was dissolved in H2O and stored at ⫺70 C before use. TPA, A23187, and EGF were obtained from Sigma Chemical Co. (St. Louis, MO) and were dissolved in the appropriate vehicle (DMSO for TPA and A23187; 50 mM HCl for EGF). SK&F 96365 and 2,5-di-tert-butylhydroquinone (DBHQ) were obtained from BioMol (Plymouth Meeting, PA) and dissolved in DMSO. CAI was obtained from the Developmental Therapeutics Program, National Cancer Institute. Working stocks of all reagents were prepared in 50% DMSO immediately before use, and the final DMSO concentration was held constant throughout each experiment at either 0.5% or 1%. Control experiments indicated that neither of these DMSO concentrations affected gene expression or cell viability.

Vol 11 No. 3

11. 12.

13.

14.

Acknowledgments We thank Kirsten Taylor and Thanh-Hoai Dinh for expert technical assistance, and Drs. Richard Mauer, David Levens, Lance Liotta, and Bruce Magun for thoughtful discussions.

Received April 26, 1996. Re-revision received November 26, 1996. Accepted November 27, 1996. Address requests for reprints to: Karin D. Rodland, Department of Cell and Developmental Biology, L215, Oregon Health Sciences University, Portland, Oregon 97201-3098. This work was supported by Grant CA-60738 from the NIH-National Cancer Institute (to K.D.R.).

15.

16.

17.

18.

REFERENCES 1. Cole K, Kohn EC 1994 Calcium-mediated signal transduction: biology, biochemistry, and therapy. Cancer Metastasis Rev 13:33–41 2. Schwartz MA 1993 Spreading of human endothelial cells on fibronectin or vitronectin triggers elevation of intracellular free calcium. J Cell Biol 120:1003–1010 3. Sjaastad M, Angres B, Lewis RS, Nelson WJ 1994 Feedback regulation of cell-substratum adhesion by integrinmediated intracellular calcium signaling. Proc Natl Acad Sci USA 91:8214–8218 4. Fisch TM, Prywes R, Roeder R 1987 c-Fos sequences necessary for basal expression and induction by epidermal growth factor, 12–0-tetradecanoyl phorbol 13-acetate, and the calcium ionophore. Mol Cell Biol 7:3490–3502 5. Sheng M, Dougan ST, McFadden G, Greenberg ME 1988 Calcium and growth factor pathways of c-fos transcriptional activation require distinct upstream regulatory sequences. Mol Cell Biol 8:2787–2796 6. Pribnow D, Muldoon L, Fajardo M, Theodor L, Chen L-YS, Magun BE 1992 Endothelin induces transcription of fos/jun family genes: a prominent role for calcium ion. Mol Endocrinol 6:1003–1012 7. Rodland KD, Muldoon LL, Lenormand P, Magun BE 1990 Modulation of RNA expression by intracellular calcium. J Biol Chem 265:11000–11007 8. Thompson MA, Ginty DD, Bonni A, Greenberg ME 1995 L-type voltage-sensitive Ca2⫹ channel activation regulates c-fos transcription at multiple levels. J Biol Chem 270:4224–4235 9. Sheng M, McFadden G, Greenberg ME 1990 Membrane depolarization and calcium induce c-fos transcription via phosphorylation of transcription factor CREB. Neuron 4:571–582 10. Sheng M, Thompson MA, Greenberg ME 1991 CREB:

19.

20.

21.

22. 23.

24.

25. 26.

27.

a Ca2⫹-regulated transcription factor phosphorylated by calmodulin-dependent kinases. Science 252: 1427–1430 Bading H, Ginty DD, Greenberg ME 1993 Regulation of gene expression in hippocampal neurons by distinct calcium signaling pathways. Science 260:181–186 Xing J, Ginty, DD, Greenberg ME 1996 Coupling of the RAS-MAPK pathway to gene activation by RSK2, a growth factor-regulated CREB kinase. Science 273: 959–963 Collart MA, Tourkine N, Belin D, Vassalli P, Jeanteur P, Blanchard A-M 1991 c-Fos gene transcription in murine macrophages is modulated by a calcium-dependent block to elongation in intron 1. Mol Cell Biol 11:2826–2831 Muldoon LL, Rodland KD, Forsythe ML, Magun BE 1989 Stimulation of phosphatidylinositol hydrolysis, diacylglycerol release, and gene expression in response to endothelin, a potent new agonist for fibroblasts and smooth muscle cells. J Biol Chem 264:8529–8536 Muldoon LL, Enslen H, Rodland KD, Magun BE 1991 Stimulation of Ca2⫹ influx by endothelin-1 is subject to negative feedback by elevated intracellular Ca2⫹. Am J Physiol 260:C1273–C1281 Thastrup O, Cullen PJ, Drobak BK, Hanley MR, Dawson AP 1990 Thapsigargin, a tumor promoter, discharges intracellular Ca2⫹ stores by specific inhibition of the endoplasmic reticulum Ca2⫹-ATPase. Proc Natl Acad Sci USA 87:2466–2470 Lenormand P, Muldoon LL, Enslen H, Rodland KD, Magun BE 1990 Two tumor promoters, TPA and thapsigargin, act synergistically via distinct signalling pathways to stimulate gene expression. Cell Growth Differ 1:627–635 Takemura H, Hughes AR, Thastrup O, Putney JWJ 1989 Activation of calcium entry by the tumor promoter thapsigargin in parotid acinar cells: evidence that an intracellular calcium pool, and not an inositol phosphate, regulates calcium fluxes at the plasma membrane. J Biol Chem 264:12266–12271 Kohn EC, Sandeen MA, Liotta LA 1992 In vivo efficacy of a novel inhibitor of selected signal transduction pathways including calcium, arachidonate, inositol phosphates. Cancer Res 52:3208–3212 Gusovsky F, Lueders JE, Kohn EC, Felder CC 1993 Muscarinic receptor-mediated tyrosine phosphorylation of phospholipase C-gamma. J Biol Chem 268: 7768–7772 Felder CC, Ma AL, Liotta LA, Kohn EC 1991 The antiproliferative and antimetastatic compound L651582 inhibits muscarinic acetylcholine receptor-stimulated calcium influx and arachidonic acid release. J Pharmacol Exp Ther 257:967–971 Kohn EC, Liotta LA 1990 L651582: a novel antiproliferative and antimetastasis agent. J Natl Cancer Inst 82:54–60 Kohn EC, Felder CC, Jacobs W, Holmes KA, Day A, Freer R, Liotta LA 1994 Structure-function analysis of signal and growth inhibition by carboxyamido-triazole, CAI. Cancer Res 54:935–942 Kohn EC, Jacobs W, Kim Y-S, Alessandro R, StetlerStevenson WG, Liotta LA 1994 Calcium influx regulates expression of matrix metalloproteinase-2 (72 kDa Type IV collagenase, gelatinase A). J Biol Chem 269: 21505–21511 Hoth M, Penner R 1993 Calcium release-activated calcium current in rat mast cells. J Physiol 465:359–386 Merritt JE, Armstrong WP, Benham CD, Hallam TJ, Jacob R, Jaxa-Chamiec A, Leigh B, McCarthy SA, Moores KE, Rink TJ 1990 SK&F 986365, a novel inhibitor of receptor-mediated calcium entry. Biochem J 271: 515–522 Deschamps J, Meijlink F, Verma IM 1985 Identification of a transcriptional enhancer element upstream from the


proto-oncogene fos. Science 230:1174–1177 28. Wong WL, Brostrom MA, Kuznetsov G, Gmitter-Yellen D, Brostrom CO 1993 Inhibition of protein synthesis and early protein processing by thapsigargin in cultured cells. Biochem J 289:71–79 29. Magun BE, Rodland KD 1995 Transient inhibition of protein synthesis induces the immediate early gene VL30: alternative method for thapsigargin-induced gene expression. Cell Growth Differ 6:891–897 30. Rodland KD, Jue SF, Magun BE 1986 Regulation of VL30 gene expression by activators of protein kinase C. J Biol Chem 261:5029–5033 31. Fischer G, Wittmann-Liebold B, Lang K, Kiefhaber T, Schmid F 1989 Cyclophilin and peptidyl-prolyl cis-trans isomerase are probably identical proteins. Nature 337:476–478 32. Takahashi N, Hayano T, Suzuki M 1989 Peptidyl-prolyl cis-trans isomerase is the cyclosporin A-binding protein cyclophilin. Nature 337:473–475 33. Rodland KD, Pribnow D, Lenormand P, Chen SLY, Magun BE 1993 Characterization of a unique enhancer element responsive to cyclic adenosine 3⬘, 5⬘-monophosphate and elevated calcium. Mol Endocrinol 7:787–796 34. Daub H, Weiss FU, Wallasch C, Ullrich A 1996 Role of transactivation of the EGF receptor in signalling by Gprotein-coupled receptors. Nature 379:557–560 35. Llopis J, Chow SB, Kass GE, Gahm A, Orrenius S 1991 Comparison between the effects of the microsomal Ca2⫹-translocase inhibitors thapsigargin and 2,5-di-(tbutyl)-1,4-benzohydroquinone on cellular calcium fluxes. Biochem J 277:553–556 36. Putney JJ, Takemura H, Hughes AR, Horstman DA, Thastrup O 1989 How do inositol phosphates regulate calcium signaling? FASEB J 3:1899–1905 37. Thastrup O 1990 Role of Ca2⫹-ATPases in regulation of cellular Ca2⫹ signalling, as studied with the selective microsomal Ca2⫹-ATPase inhibitor, thapsigargin. Agents Actions 29:8–15 38. Nordstrom T, Nevanlinna HA, Andersson LC 1992 Mitosis-arresting effect of the calcium channel inhibitor SK&F 96365 on human leukemia cells. Exp Cell Res 202:487–494 39. Mason MJ, Mayer B, Hymel LJ 1993 Inhibition of Ca2⫹ transport pathways in thymic lymphocytes by econazole, miconazole, and SKF 96365. Am J Physiol 264: C654–C662 40. Griendling KK, Tsuda T, Alexander RW 1989 Endothelin stimulates diacylglycerol accumulation and activates protein kinase C in cultured vascular smooth muscle cells. J Biol Chem 264:8237–8240 41. Hirata Y, Yoshimi H, Takata S, Watanabe TX, Kumagai S, Nakajima K, Sakakibara S 1989 Cellular mechanism of action by a novel vasoconstrictor endothelin in cultured rat vascular smooth muscle cells. Biochem Biophys Res Commun 154:868–875 42. Lenormand P, Pribnow D, Rodland KD, Magun BE 1992 Identification of a novel enhancer element mediating cal-

291

43. 44.

45.

46. 47.

48.

49.

50.

51. 52.

53.

54. 55. 56.

cium-dependent induction of gene expression in response to either epidermal growth factor or activation of protein kinase C. Mol Cell Biol 12:2793–2803 Angel P, Karin M 1991 The role of Jun, Fos, and the AP-1 complex in cell-proliferation and transformation. Biochim Biophys Acta 1072:129–157 Boise LH, Petryniak B, Mao X, June CH, Wang CY, Lindsten T, Bravo R, Kovary K, Leiden JM, Thompson CB 1993 The NFAT-1 DNA binding complex in activated T cells contains Fra-1 and JunB. Mol Cell Biol 13:1911–1919 Jain J, McCaffrey PG, Miner Z, Kerppola TK, Lambert JN, Verdine GL, Curran T, Rao A 1993 The T-cell transcription factor NFATp is a substrate for calcineurin and interacts with Fos and Jun. Nature 365:352–355 Sonnenberg JL, Rauscher FJ, Morgan JI, Curran T 1989 Regulation of proenkephalin by Fos and Jun. Science 246:1622–1625 McDonnell SE, Kerr LD, Matrisian LM 1990 Epidermal growth factor stimulation of stromelysin mRNA in rat fibroblasts requires the induction of the protooncogenes c-fos and c-jun, and the activation of protein kinase C. Mol Cell Biol 10:4284–4293 Angel P, Imagawa M, Chiu R, Stein B, Imbra RJ, Ramsdorf HJ, Jonat C, Herrlich P, Karin M 1987 Phorbol esterinducible genes contain a common cis element recognized by a TPA-modulated trans-acting factor. Cell 49:729–739 Angel P, Baumann I, Stein B, Delius H, Ramsdorf HJ, Herrlich P 1987 12-O-tetradecanoyl-phorbol-13-acetate induction of the human collagenase gene is mediated by an inducible enhancer element located in the 5⬘-flanking region. Mol Cell Biol 7:2256–2266 Sato H, Takino T, Okada Y, Cao J, Shinagawa A, Yamamoto E, Seiki M 1994 A matrix metalloproteinase expressed on the surface of invasive tumour cells. Nature 370:61–65 Sato H, Seiki M 1993 Regulatory mechanism of 92 kDa type IV collagenase gene expression which is associated with invasiveness of tumor cells. Oncogene 8:395–405 Gaire M, Magbanua Z, McDonnell S, McNeil L, Lovett D, Matrisian L 1994 Structure and expression of the human gene for the matrix metalloproteinase matrilysin. J Biol Chem 269:2032–2040 Kerr LD, Holt JT, Matrisian LM 1988 Growth factors regulate transin gene expression by c-fos-dependent and c-fos-independent pathways. Science 242: 1424–1427 Kohn EC, Alessandro R, Spoonster J, Wersto RP, Liotta LA 1995 Angiogenesis: role of calcium-mediated signal transduction. Proc Natl Acad Sci USA 92:1307–1311 Neumann JR, Morency CA, Russian KO 1987 A novel rapid assay for chloramphenicol acetyltransferase gene expression. BioTechniques 5:444–447 Grynkiewicz G, Poenie M, Tsien RY 1985 A new generation of Ca2⫹ indicators with greatly improved fluorescence properties. J Biol Chem 260:3440–3450