Calcium Channels and Pumps in Cancer: Changes and Consequences*

Increases in intracellular free Ca2+ play a major role in many cellular processes. The deregulation of Ca2+ signaling is a feature of a variety of diseases, and modulators of Ca2+ signaling are used to treat conditions as diverse as hypertension to pain. The Ca2+ signal also plays a role in processes important in cancer, such as proliferation and migration. Many studies in cancer have identified alterations in the expression of proteins involved in the movement of Ca2+ across the plasma membrane and subcellular organelles. In some cases, these Ca2+ channels or pumps are potential therapeutic targets for specific cancer subtypes or correlate with prognosis.

Increases in intracellular free Ca 2؉ play a major role in many cellular processes. The deregulation of Ca 2؉ signaling is a feature of a variety of diseases, and modulators of Ca 2؉ signaling are used to treat conditions as diverse as hypertension to pain. The Ca 2؉ signal also plays a role in processes important in cancer, such as proliferation and migration. Many studies in cancer have identified alterations in the expression of proteins involved in the movement of Ca 2؉ across the plasma membrane and subcellular organelles. In some cases, these Ca 2؉ channels or pumps are potential therapeutic targets for specific cancer subtypes or correlate with prognosis.
Our understanding of calcium signaling and its intersection with specific processes important in tumor progression is only recent. We now appreciate that altered expression of specific Ca 2ϩ channels and pumps is a characterizing feature of some cancers. By comparison, the link between calcium signaling and other conditions, such as cardiovascular and neurological diseases, was made many years ago. The direct link between Ca 2ϩ and processes linked to a specific pathology, such as vascular tone and neurotoxicity, meant that these conditions attracted the initial focus of researchers devoted to defining the role of Ca 2ϩ in disease.
In their seminal review "The Hallmarks of Cancer," Hanahan and Weinberg (1) described six acquired characteristics of cancers: the ability to evade apoptosis, self-sufficiency in growth signaling, insensitivity to anti-growth signals, the capacity to invade and metastasize, "limitless" replication potential, and the promotion of angiogenesis. Calcium signaling is linked either directly or indirectly to each of these processes, and this has been reviewed elsewhere (2)(3)(4)(5)(6). A remodeling of calcium homeostasis can occur in cancer cells. Although alterations in Ca 2ϩ signaling may not be a requirement for the initiation of cancer, the consequences of altered calcium transport in cancer cells may be significant and contribute to tumor progression. Characterizing such changes may help to identify new therapeutic targets. In this minireview, we discuss how remodeling of Ca 2ϩ signaling is a feature of some cancers and provide examples of how this remodeling is often achieved through the differential expression of specific Ca 2ϩ pumps and channels.
Examples of this remodeling are discussed, particularly those that illustrate the complexities of expression changes and their contribution to tumor progression.

Ca 2؉ Transport in Cancer Cells
Cancer cells use the same calcium channels, pumps, and exchangers as non-malignant cells. However, there are often key alterations in calcium channels and pumps in cancer cells. Such changes in cancer cells may include the expression of calcium channels or pumps (or their specific isoforms) not normally present in non-malignant cells of the same cell type, pronounced changes in the level of expression (as outlined in Table  1), altered cellular localization, altered activity through changes in post-translational modification, gene mutations, and changes in activity or expression associated with specific cancer-relevant processes (e.g. migration). These changes are often reflected in alterations in Ca 2ϩ flux across the plasma membrane or across intracellular organelles.

Ca 2؉ Influx in Cancer
The influx of calcium across the plasma membrane into the cell is a key trigger or regulator of cellular processes relevant to tumor progression, including proliferation, migration, and apoptosis. Ca 2ϩ -permeable ion channels of almost every class have now been associated with aspects of tumor progression. This minireview will particularly focus on transient receptor potential (TRP) 2 channels and ORAI-mediated store-operated Ca 2ϩ influx as examples of Ca 2ϩ influx pathways altered in some cancers.
TRP Channels-TRP ion channels consist of six subfamilies, with most members permeable to Ca 2ϩ , many of which have a role in distinguishing sensations, including pain, temperature, taste, and pressure (7). This family is arguably the most studied ion channel class in cancer. The key early work on calcium signaling in cancer was focused on cancers of the prostate gland and more specifically the calcium-permeable ion channel TRPM8 (8). Although now studied predominately in the context of its role as a cold receptor (9, 10), TRPM8 was first identified by its overexpression in some prostate cancers (8). Early work by Zhang and Barritt (11) demonstrated that both the silencing of TRPM8 and menthol-mediated activation of TRPM8 reduced the viability of LNCaP prostate cancer cells. That both activators and inhibitors are proposed as potential therapeutic agents for prostate cancer cells that overexpress TRPM8 is reflective of the duality of the calcium signal (12), whereby Ca 2ϩ is both a key regulator of proliferation and, in the case of Ca 2ϩ overload, an initiator of cell death. The ability of TRPM8 activation by prostate-specific antigen to inhibit the migration of PC3 prostate cancer cells now extends the applicability of channel activators as therapeutics beyond just inducers of cancer cell death (13). Further detailed work on TRPM8 in prostate cancer showed androgen-mediated increases in TRPM8 in LNCaP prostate cancer cells (11,14). This finding provides one of the first examples of hormone-mediated changes in the expression of a calcium-permeable ion channel in a cancer cell line. As discussed below, this has now been seen with other calcium channels and pumps in breast cancers.
The contribution of TRPM8 to cancer progression, as we will see for other Ca 2ϩ channels and pumps, may not always involve its classic role (in this case as a plasmalemmal ion channel). As opposed to the usual plasma membrane localization, endoplasmic reticulum localization of TRPM8 is observed in some prostate cancer cells (11,15), with the consequence being reduced levels of endoplasmic reticulum Ca 2ϩ and increased resistance to apoptosis (15). Aside from prostate cancer, overexpression of TRPM8 is also associated with other cancer types, including melanoma and cancers of the pancreas, breast, colon, and lung (see Table 1). However, the utility of TRPM8 as a target for cancer therapy might be limited and require knowledge of the individual tumor expression of the channel. For example, TRPM8 expression actually appears to decrease as prostate cancer cells transition to androgen independence and increased aggressiveness (16,17).
TRPV6 is another TRP channel linked to prostate cancer. TRPV6 levels correlate with tumor progression and have been proposed as a predictor of invasiveness (18,19). TRPV6 is highly Ca 2ϩ -selective and is constitutively active (20). When TRPV6 expression is silenced in LNCaP prostate cancer cells, there is inhibition of Ca 2ϩ influx and consequently reduced activation of NFAT. Crucially, this illustrates the importance of calcium-dependent transcription pathways as a mechanism for tumor promotion (19).
Like TRPM8, alterations in TRPV6 expression are not confined to cancers of the prostate, with increased expression levels reported in thyroid, colon, ovarian, and breast cancers (see Table 1). In breast cancers, the expression of TRPV6 varies widely between tumors (21). The consequences of TRPV6 overexpression in tumors may relate to effects on cancer cell survival, as TRPV6 silencing in T47D breast cancer cells reduces cell viability (21). Further studies are needed to address the mechanisms leading to TRPV6 overexpression in cancers and the association between TRPV6 levels and breast cancer prognosis. Analogous to the androgen dependence of TRPM8 expression in LNCaP prostate cancer cells, TRPV6 levels also appear to be hormonally regulated, with estradiol increasing TRPV6 mRNA in T47D breast cancer cells (21).
Other examples of TRP channels that are overexpressed in multiple cancer types include TRPC3 and TRPC6. TRPC3 is elevated in some breast (22) and ovarian epithelial tumors, and its silencing reduces ovarian cancer cell line proliferation in vitro and tumor formation in vivo (23). TRPC6 is elevated in cancers of the breast, liver, stomach, and esophagus and in gliomas (22,24,25), and its silencing reduces the proliferation of some esophageal and breast cancer cell lines and glioma cell lines (22,24,25). For esophageal and glioma cell lines, these effects are due to G 2 /M cell cycle arrest (24,25).
The importance of some TRP channels in tumor progression appears to extend beyond the primary tumor. Fiorio Pla et al. (26) showed that migrating endothelial cells have a greater cytosolic calcium response to the TRPV4 activator 4-␣-phorbol 12,13-didecanoate than non-migrating cells. Furthermore, they showed increased expression of TRPV4 in endothelial cells derived from breast cancers compared with those derived from normal tissue, implicating TRPV4 as a possible key component in angiogenesis associated with breast cancers. Other Ca 2ϩ channels have also been associated with angiogenesis, as reviewed recently (4).
Calcium entry into the cell via some TRP channels may result in localized Ca 2ϩ signals that contribute to cancer cell migration (3). One example of such a localized event is referred to as Ca 2ϩ flickers (27), which are highly localized (ϳ5-m diameter) and transient (10 ms to 4 s) increases in Ca 2ϩ that control the direction of migration as lung fibroblasts move toward a growth factor. Ca 2ϩ flickers during migration are regulated by TRPM7 (27), which may act as a stretch or mechanical sensing channel (28). With TRPM7 inhibition, there is a reduction in migration of a number of cancer cell types, including those of the pancreas, lung, and nasopharynx (29 -31).
The examples above highlight some studies in which cancer cells have been associated with a remodeling of TRP channel expression or in which TRP channels have been linked to specific processes important in tumor progression. The interest and understanding of TRP channels in cancer are likely to expand in the coming years, and these channels may represent the first class of ion channel targeted for the treatment of a specific cancer.
Store-operated Ca 2ϩ Influx-Store-operated Ca 2ϩ entry is a critical Ca 2ϩ influx pathway and represents the major Ca 2ϩ influx mechanism in non-excitable cells (32), such as those of the epithelia, from where most cancers originate. The pathway involves the activation of Ca 2ϩ influx upon intracellular Ca 2ϩ store depletion (32)(33)(34). The canonical components of storeoperated Ca 2ϩ entry are the calcium influx channel ORAI1 and the endoplasmic Ca 2ϩ depletion sensor STIM1 (stromal interaction molecule 1) (35,36). Although this pathway has rapidly become one of the Ca 2ϩ influx pathways most studied in breast cancer, it also appears to be an important Ca 2ϩ influx route during lactation (37), suggesting an important role in normal breast function.
ORAI1 and STIM1 silencing in MDA-MB-231 breast cancer cells reduces migration, invasion through Matrigel, and the establishment of lung metastasis after tail vein injection in NOD/SCID mice (38), the latter of which can be mimicked by the pharmacological store-operated Ca 2ϩ influx inhibitor SKF96365 (38). The anti-metastasis effects of ORAI1 and STIM1 silencing appear to be due in part to alterations in focal adhesion turnover (38). The effects of ORAI1 silencing on breast cancer cells are not restricted to inhibition of processes important in migration; ORAI1 silencing has antiproliferative properties in MCF-7 breast cancer cells in culture and in vivo. These changes may be due in part to reductions in basal Ca 2ϩ influx, leading to reduced ERK1/2 phosphorylation and cyclin D 1 expression (39).
Alterations in the expression of specific components of store-operated Ca 2ϩ entry are also a feature of some breast cancer cells. ORAI1 mRNA levels are higher in some breast cancer cell lines compared with non-malignant breast cell lines (37). When breast cancer subtypes are stratified by gene expression, basal breast cancers (associated with a poor prognosis and a lack of effective therapies) are characterized by an elevated STIM1/STIM2 ratio. Correspondingly, those patients with breast cancers with a high STIM1/STIM2 ratio and high STIM1 levels have significantly reduced survival (37), placing STIM proteins as either potential key regulators or biomarkers of breast cancer progression. The significance of STIM1 may extend beyond breast cancers given its role in the migration of cervical cancer cells (40). The mechanisms responsible for enhanced ORAI1-mediated Ca 2ϩ influx in breast cancer appear to be complex and related to cancer subtypes. As discussed below, in addition to STIM1-mediated activation of ORAI1, some breast cancers that overexpress the SPCA2 isoform may be characterized by elevated ORAI1-mediated Ca 2ϩ influx.
The ORAI1 isoform is not the only ORAI protein with a cancer association. ORAI3 protein levels and ORAI3-dependent store-operated Ca 2ϩ influx are both elevated in estrogen receptor-positive breast cancer cell lines (41) compared with estrogen receptor-negative cell lines, in which store-operated Ca 2ϩ influx is mediated predominately by ORAI1. A strengthening of the causative link with cancer was provided by a study in estrogen receptor-positive MCF-7 breast cancer cells in which ORAI3 silencing inhibited proliferation through G 1 arrest (42). Although the examples above point to an up-regulation of ORAI-mediated influx, some cancer types might be associated with a down-regulation of this pathway that may in turn help in the acquisition of apoptotic resistance (6). Indeed, reduced ORAI1-mediated Ca 2ϩ influx and expression are features of androgen-independent prostate cancer cells, and silencing of ORAI1 reduces apoptosis in LNCaP cells (43).
In this minireview, we have given examples of how ORAI1 may regulate processes important for carcinogenesis, including cell proliferation, migration, and apoptosis sensitivity, and this may occur in a store-dependent or store-independent manner. Examples of how ORAI1 regulates these key cancer processes are shown schematically in Fig. 1.
Although not a focus of this minireview, voltage-gated calcium channels are increasingly studied in cancer, and in many cases, the studies have examined the reasons for changes in expression levels in cancer. This is particularly illustrated in studies assessing mechanisms of altered expression of voltagegated ion channels. For example, higher relapse in Wilms tumors is associated with higher DNA copy numbers of the ␣ 1 -subunit of the voltage-gated Ca 2ϩ channel CACNA1E (44), and reduced expression of CACNA2D3 via promoter hypermethylation is associated with poor prognosis in gastric cancer (45). These methodological approaches will be applied to other channels and pumps and other cancers in the future.

Ca 2؉ Efflux in Cancer
Ca 2ϩ efflux across the plasma membrane can be mediated by both Na ϩ /Ca 2ϩ exchangers and primary active transport via plasma membrane Ca 2ϩ -ATPases (PMCAs). However, most studies of Ca 2ϩ efflux pathways in cancer cells have focused on the latter mechanism. PMCAs are encoded by four genes (PMCA1-4), which are alternatively spliced to generate a suite of Ca 2ϩ efflux pumps responsible for maintaining resting cytosolic free Ca 2ϩ at low (ϳ100 nM) levels (46,47). PMCAs also contribute to specific cell functions, such as the transport of Ca 2ϩ into milk through PMCA2 (48).
An area in which PMCAs may be critically important in cancer is the regulation of cell death, as reflected in early work assessing the consequences of PMCA overexpression. Overexpression of some PMCA isoforms in CHO cells reduces Ca 2ϩ levels within the endoplasmic reticulum and also attenuates mitochondrial Ca 2ϩ accumulation after cell activation (49), a consequence that would be hypothesized to result in anti-apoptotic effects. Indeed, the overexpression of PMCA in HeLa cells increases their resistance to cell death induced by ceramide (50). Recent studies in T47D breast cancer cells show that the overexpression of PMCA2 reduces the degree of cell death induced by ionomycin, and this is associated with a reduction in the duration and magnitude of increases in cyto-FIGURE 1. ORAI1 regulates processes important for cancer cell proliferation, migration, and apoptosis. A, in MCF-7 human breast cancer cells, SPCA2 partially localizes to the plasma membrane and interacts with ORAI1 to mediate store-independent Ca 2ϩ influx. This is associated with phosphorylation of ERK1/2, nuclear translocation of NFAT, and increased cell proliferation (39). B, silencing of ORAI1 or STIM1 in MDA-MB-231 human breast cancer cells reduces store-operated Ca 2ϩ influx and is associated with reduced focal adhesion turnover, cell migration, and metastasis formation in vivo. Expression of constitutively active Ras or Rac in these cells partially rescues impaired focal adhesion turnover and cell migration induced by inhibition of store-operated Ca 2ϩ entry, implicating possible roles for these small GTPases in Ca 2ϩ -dependent cell migration (39). C, in LNCaP human prostate cancer cells, ORAI1 expression is regulated by the androgen receptor (AR), and ORAI1 silencing is associated with resistance to thapsigargin (TG)-, TNF␣-, cisplatin-, and oxaliplatin-induced apoptosis (43). In A and B (black), ORAI1 expression may promote carcinogenesis; in C (red), ORAI1 expression may inhibit carcinogenesis (i.e. promote apoptosis). IP 3 Rs, inositol 1,4,5-trisphosphate receptors.
solic [Ca 2ϩ ] mediated by this Ca 2ϩ ionophore (51). PMCA2 is an isoform with reported overexpression in some breast cancer cell lines (52) and in clinical human samples, in which high levels appear to be associated with a poor prognosis in some patient groups (51). Collectively, these studies suggest that the remodeling of calcium efflux associated with increases in PMCA expression contributes to the acquisition of an antiapoptotic phenotype in cancer cells.
Studies assessing the expression of PMCA isoforms during the differentiation of colon cancer cells suggest that a remodeling of PMCA isoform expression is not confined to cancers of the breast. PMCA1 expression remains fairly constant during differentiation of human colon cancer cell lines, whereas PMCA4 undergoes a pronounced increase in expression with differentiation (53,54). PMCA4 overexpression studies in HT29 colon cancer cells suggest that the down-regulation of PMCA4 in colon cancer may help to augment cytosolic Ca 2ϩ responses to proliferative stimuli without sufficiently increasing cytosolic [Ca 2ϩ ] to levels that promote apoptosis (55). The changes in PMCA4 expression seen in the differentiation models correlate well with human colon cancer clinical samples, in which PMCA4 mRNA is reduced in colon adenocarcinomas compared with normal colon (55). The up-regulation of PMCA2 expression in breast cancer and the down-regulation of PMCA4 in colon cancer may seem to conflict; however, in both cases, the changes in PMCA expression appear to bestow an advantage to the cancer cell. In the case of PMCA2, this appears to be related to the acquisition of greater resistance to cell death in breast cancer cells, and for PMCA4 augmented responses to proliferative signals in colon cancer cells.

Intracellular Organelle Ca 2؉ Channels and Pumps and Cancer
Intracellular organelles play critical roles in Ca 2ϩ -regulated processes either through the regulation of cytosolic free Ca 2ϩ or through modulation of Ca 2ϩ -regulated proteins that reside within the organelle. We will outline examples of Ca 2ϩ channels and pumps of the endoplasmic reticulum and Golgi, as these have been the most studied in cancer. However, the recent identification of proteins that play major roles in mitochondrial Ca 2ϩ influx and efflux (56 -58) and the recently identified two-pore channel proteins present in endosomes (TPC1) and lysosomes (TPC2) (59) represent new opportunities to improve our understanding of the remodeling of Ca 2ϩ signaling in some cancers and will no doubt be the focus of research in the future (60).

Regulators of Endoplasmic Reticulum Ca 2؉ Levels
One of the earliest links between the regulation of endoplasmic reticulum Ca 2ϩ and cancer comes from studies of the antiapoptotic protein Bcl-2 (B cell lymphoma-2). In addition to its early and now well established role in inhibiting the release of the pro-apoptotic factor cytochrome c (61-63), Bcl-2 decreases the Ca 2ϩ content of the endoplasmic reticulum (50,64,65). Mechanistically, this occurs at least in part through interaction with the inositol 1,4,5-trisphosphate receptor (66), likely reducing the ability to achieve the high Ca 2ϩ loads required for mitochondria to accumulate Ca 2ϩ sufficiently to trigger apoptotic cell death (67). Some examples of alterations in the expression of key calcium channels and pumps of the endoplasmic reticulum are highlighted in Table 1. Similar to increases in the expression of PMCA4 during colon cancer cell line differentiation and the down-regulation of PMCA4 expression in some colon cancers, SERCA3 pump expression increases with the differentiation of colon cell lines and is down-regulated in colon cancer (68), implicating a major remodeling of active Ca 2ϩ transport in colon cancer. The significance of the down-regulation of SERCA3 is not restricted to colon cancer given the more recent report of a significant down-regulation of SERCA3 in breast cancers, an event that is even seen in benign lesions (69). Further evidence of the potential significance of sarco/ endoplasmic reticulum Ca 2ϩ -ATPase (SERCA) down-regulation in cancer is reflected in studies of mice haplodeficient for SERCA2 (70,71). These mice are characterized by increased incidence of squamous cell tumors, the mechanism of which likely involves altered Ca 2ϩ signaling and a subsequent change in the microenvironment of skin epithelia (70).

Regulators of Golgi Ca 2؉ Levels
Although more recently identified and less widely studied in the context of contributions to cellular processes than SERCAs, secretory pathway Ca 2ϩ -ATPases (SPCAs), both the ubiquitously expressed SPCA1 isoform and the more restricted SPCA2 isoform (72,73), are beginning to be assessed in cancer cells. In MDA-MB-231 basal-like breast cancer cells (which do not express the SPCA2 isoform), SPCA1 silencing inhibits proliferation without changes in global cytosolic [Ca 2ϩ ], consistent with the minor role of SPCAs (cf. PMCAs and SERCAs) in contributing to cytosolic [Ca 2ϩ ] recovery in most cell types (74). Instead, as may be the case for other Ca 2ϩ channels and pumps located on the membranes of intracellular organelles, the mechanism by which SPCA1 silencing inhibits proliferation may involve alterations in the Ca 2ϩ levels within the Golgi lumen, where Ca 2ϩ -regulated enzymes reside. Indeed, one consequence of SPCA1 silencing in MDA-MB-231 breast cancer cells is the inhibition of cleavage of the pro-insulin-like growth factor 1 receptor likely through reduced activity of the Ca 2ϩsensitive proprotein convertase furin (74). The consequences of reduced SPCA1-mediated Ca 2ϩ sequestration may be cell typeand context-dependent as shown by the increased susceptibility of Spca1 ϩ/Ϫ mice to develop squamous skin tumors (75).
One of the proposed roles for the other SPCA isoform, SPCA2, has been the sequestration of Ca 2ϩ during lactation (76); however, this pump also appears to play a role in the pathophysiology of breast cancer. SPCA2 levels are increased in luminal-like breast cancer cell lines and clinical breast cancers belonging to the luminal B and ERBB2 molecular subtypes (39). This may be related to hormonal factors given that, in MCF-7 breast cancer cells, SPCA2 mRNA levels increase with prolactin (77). Silencing of SPCA2 in breast cancer cell lines that overexpress this Ca 2ϩ pump, such as MCF-7 cells, reduces their proliferation, anchorage-independent growth, and growth in vivo (39). However, in contrast to SPCA1 in breast cancer cells, SPCA2 does not appear to contribute to tumor progression through alterations in Ca 2ϩ levels within the Golgi. In a result that was initially counterintuitive, SPCA2 overexpression increases cytosolic basal [Ca 2ϩ ] rather than decreasing it, as might be expected for a calcium pump that sequesters Ca 2ϩ from the cytoplasm into the Golgi. Overexpression of SPCA2 leads to its localization at the plasma membrane, where it activates ORAI1 channels, the consequence of which is activation of the transcription factor NFAT (nuclear factor of activated T cells; shown in Fig. 1A) (39). SPCA2 overexpression-induced increases in Ca 2ϩ influx across the plasma membrane represent an example in which the contribution that a calcium pump makes to tumor progression is not directly related to its own Ca 2ϩ -transporting ability. The ability of SPCA2 to contribute to tumor growth independently of its own Ca 2ϩ -transporting ability suggests that pharmacological inhibitors of SPCA2 Ca 2ϩ transport function may be ineffective in breast cancers in which SPCA2 solely contributes to tumor growth through this ORAI1-dependent mechanism and demonstrates the importance of mechanistic studies assessing the contribution of Ca 2ϩ channels and pumps to tumorigenic pathways.

Calcium Signaling and Cancer: New Horizons
Major advances have occurred in the last decade in our understanding of how calcium signaling is remodeled in some cancer cells and how specific calcium channels or pumps represent potential new therapeutic targets in oncology. However, there are areas of cancer research where the link between calcium signaling is still relatively unexplored, such as the "emerging hallmarks of cancer" recently described by Hanahan and Weinberg (78). These include cellular energy metabolism reprogramming, whereby cancer cells shift their energy metabolism to glycolysis, a phenomenon first described by Otto Warburg almost a century ago (78 -80). Further studies on the possible role of Ca 2ϩ signaling in the regulation of glycolysis, the switch to glycolysis, and the use of glycolysis-generated ATP to fuel Ca 2ϩ pumps in cancer cells are required (81,82). Another aspect of cancer biology where Ca 2ϩ signaling is clearly going to be critical but has not been fully explored is the tumor microenvironment (78). Due to the depth of work in the area of tumor microenvironment, readers are encouraged to consult the numerous reviews on this topic (83)(84)(85). An aspect of the tumor microenvironment where signaling is likely to be particularly significant is cancer-associated fibroblasts, which are in an "activated" state and are in a dynamic signaling interplay with cancer cells (78,86). Ca 2ϩ may be critical to this signaling, as reflected by the importance of PDGF in the signaling between cervical cancer cells and cancer-associated fibroblasts (87) and the ability of PDGF to elevate cytosolic [Ca 2ϩ ] in other cell types (88).

Conclusions
Many processes contribute to cancer development, and Ca 2ϩ signaling seems to play a role in many of them. Numerous studies have now established that some cancers are associated with major changes in the expression of specific Ca 2ϩ channels and pumps and that inhibition of some of these proteins inhibits the proliferation and/or metastasis of cancer cells. The next decade will see the role of Ca 2ϩ in cancer further defined and may see agents that specifically target Ca 2ϩ channels or pumps used in cancer therapy.