Molecular mechanisms of tumour invasion: regulation by calcium signals

Abstract

Intracellular calcium (Ca²⁺) signals are key regulators of multiple cellular functions, both healthy and physiopathological. It is therefore unsurprising that several cancers present a strong Ca²⁺ homeostasis deregulation. Among the various hallmarks of cancer disease, a particular role is played by metastasis, which has a critical impact on cancer patients’ outcome. Importantly, Ca²⁺ signalling has been reported to control multiple aspects of the adaptive metastatic cancer cell behaviour, including epithelial–mesenchymal transition, cell migration, local invasion and induction of angiogenesis (see Abstract Figure). In this context Ca²⁺ signalling is considered to be a substantial intracellular tool that regulates the dynamicity and complexity of the metastatic cascade. In the present study we review the spatial and temporal organization of Ca²⁺ fluxes, as well as the molecular mechanisms involved in metastasis, analysing the key steps which regulate initial tumour spread.

Abbreviations

AA

arachidonic acid

EC

endothelial cell

ER

endoplasmic reticulum

EGF

epidermal growth factor

EMT

epithelial–mesenchymal transition

ECM

extracellular matrix

KCa

Ca²⁺‐activated potassium channels

MMP

matrix metalloproteinase

MT1‐MMP

membrane type 1 matrix metalloproteinase

MLCK

Ca²⁺‐dependent myosin light chain kinase

myo II

myosin II

NCX

Na⁺/Ca²⁺ exchanger

NFAT

nuclear factor of activated T‐cells

ORAI

Ca²⁺ release‐activated Ca²⁺ channel protein

PMCA

plasma membrane ATPase

SOCE

store‐operated Ca²⁺ entry

STIM

stromal interaction molecule

TEC

tumour‐derived endothelial cell

TRP

transient receptor potential

TRPA

transient receptor potential ankyrin channel

TRPC

transient receptor potential canonical channel

TRPM

transient receptor potential melastatin channel

TRPV

transient receptor potential vanilloid channel

VEGF

vascular endothelial growth factor

VGCC

voltage‐gated Ca²⁺ channel

Introduction

Calcium (Ca²⁺) is a ubiquitous second messenger which is involved in the tuning of multiple fundamental cellular functions (Berridge et al. 2000). Due to its multifaceted roles, it is not surprising that deregulated Ca²⁺ homeostasis has been observed in various disorders, including tumourigenesis (Monteith et al. 2007; Prevarskaya et al. 2011). Among the various manifestations of cancer, a particular role is played by metastasis, which has a critical impact on cancer patients’ outcome (Hanahan & Weinberg, 2011). Tumour spread is a highly regulated process that usually starts with the loss of cell–cell contact and the epithelial–mesenchymal transition (EMT) (Kalluri & Weinberg, 2009). During metastasis, cancer cells also acquire enhanced directional movement and activate molecular pathways that enable the proteolysis of an extracellular matrix (ECM) as well as local angiogenesis. As a result, cancer cells enter the body's circulation systems and disseminate to distinct sites around the organism. Importantly, Ca²⁺ signalling has been reported to control multiple aspects of the adaptive metastatic cancer cell behaviours, including EMT, migration, local angiogenesis induction and intravasation (Chen et al. 2013). In this context, it is considered to be a substantial intracellular tool that regulates the dynamicity and complexity of the metastatic cascade. Intracellular free Ca²⁺ concentration is highly controlled by the fine regulation of ‘ON’ and ‘OFF’ mechanisms that ultimately generate Ca²⁺ signals with various amplitudes and frequencies. As regarding the ON mechanisms, cytosolic Ca²⁺ can either be delivered from extracellular space due to the activity of Ca²⁺‐permeable channels and transporters in plasma membrane, or occur as a result of a release from Ca²⁺‐containing organelles (e.g. endoplasmic reticulum) (Berridge et al. 2000). In order to maintain low resting Ca²⁺ concentration, cells remove Ca²⁺ using an energy‐dependent mechanism, such as plasma membrane ATPases (PMCAs), or the Na⁺/Ca²⁺ exchanger (NCX). Moreover Ca²⁺ is sequestered intracellularly into Ca²⁺‐containing organelles, primarily endoplasmic reticulum (ER), by means of mechanisms which require either ATP hydrolysis (e.g. a sarco/endoplasmic reticulum Ca²⁺‐ATPase pump), or a favourable electrochemical gradient. In this review we will overview the spatial and temporal organization of Ca²⁺ fluxes as well as those molecular mechanisms involved in metastasis, analysing the key steps which regulate tumour spread.

Epithelial–mesenchymal transition and loss of cell–cell contact

EMT is a cellular process during which epithelial cells acquire a fibroblast‐like morphology. This process involves changes in cellular shape, a loss of epithelial polarized organization and cell–cell contacts like tight and adherens junctions. Accordingly, one of the most recognized features of cells undergoing EMT, is a suppression of multiple epithelial markers (e.g. E‐cadherin, claudins, occludins) and an overexpression of mesenchymal markers (e.g. N‐cadherin, vimentin, integrins) (Fig. 1).

Figure 1. Epithelial‐to‐mesenchymal transition (EMT), loss of cell–cell contacts and downregulation of proteins such as E‐cadherin, claudin or occludin.

EMT transition is accompanied by the changes in Ca²⁺ signals due to several factors such as growth factors, cytokines and hypoxia. The most studied Ca²⁺‐permeable channels, which are associated with EMT, are indicated.

Of note, EMT and the disruption of cell–cell contact is one of the key events in tumour progression. This can be induced by various effectors like growth factors, hypoxia and inflammation (Diepenbruck & Christofori, 2016). Interestingly, the remodelling of Ca²⁺ signals during EMT processes has been reported for a variety of cancer cells. For example, in breast cancer cells, the potency of ATP‐mediated cytosolic Ca²⁺ transients exhibits significant changes after epidermal growth factor (EGF) and hypoxia‐induced EMT (Davis et al. 2011; Azimi et al. 2016). Specifically, an attenuation of the cytosolic Ca²⁺ peak and a sustained phase of Ca²⁺ influx in the response to ATP have been attributed to the activity of G‐protein‐coupled purinergic receptors (P2Y family) and ligand‐gated Ca²⁺ channels (P2X family) (Davis et al. 2011; Azimi et al. 2016). Another study reveals that an inhibition of P2X5 reduces the expression of the EMT marker vimentin, whereas its increased expression correlates with breast cancer cells that are associated with a more mesenchymal phenotype (Davis et al. 2011). Moreover, the chelation of free cytosolic Ca²⁺ suppresses the production of mesenchymal markers like vimentin, N‐cadherin and CD44, after the exposure of breast cancer cells to EGF and hypoxia (Davis et al. 2013; Stewart et al. 2015). Similar findings have been reported for hepatic cancer cells, where chelation of intracellular Ca²⁺ reversed doxorubicin‐induced EMT (Wen et al. 2016). Furthermore, the EMT of colon cancer cells may be regulated by the small conductance calcium‐activated channel, subfamily N, such as KCNN4, through Ca²⁺‐dependent mechanisms (Lai et al. 2013). Regarding the store‐operated Ca²⁺ entry (SOCE), the data are ambiguous. On one hand, SOCE and basal Ca²⁺ influx are reduced after the EGF‐induction of EMT in the MDA‐MD‐468 breast cancer cell line (Davis et al. 2012). On the other hand, transforming growth factor β1 (TGF‐β1)‐induced EMT is associated with enhanced SOCE in the MCF‐7 breast cancer cell line (Hu et al. 2011).

It is now clear that the remodelling of Ca²⁺ signalling is a prominent feature of EMT in various cancer types. Therefore, a deregulation of Ca²⁺‐permeable channels could subserve as an important EMT regulator during carcinogenesis. Indeed, silencing and pharmacological inhibition of transient receptor potential melastatin channels (TRPM) such as TRPM7 and TRPM8 reduce the expression of a variety of mesenchymal markers in breast cancer cells (Davis et al. 2013; Liu et al. 2014). In the MCF‐7 breast cancer cell line that exhibits a more epithelial‐like phenotype, the overexpression of TRPM8 leads to EMT induction as indicated by the profile of markers expressed (Liu et al. 2014). Consistent with this data, TRPM8 has been found to be upregulated in breast cancer tumour tissues, when compared to adjacent non‐tumour tissues, thereby suggesting the role of TRPM8 as a determinant of EMT transition (Liu et al. 2014). Moreover, in breast cancer cells, EGF‐induced EMT significantly increases the mRNA level of Ca²⁺ release‐activated Ca²⁺ channel protein 1 (ORAI1) and leads to altered Ca²⁺ signalling, possibly due to the involvement of transient receptor potential canonical channel type 1 (TRPC1) (Davis et al. 2012). In hepatic cancer cells, another member of the transient receptor potential canonical channel family, TRPC6, has been shown to affect the expression of EMT markers after doxorubicin induction (Wen et al. 2016).

Overall, the studies of Ca²⁺ signalling and Ca²⁺‐permeable channels using various cancer models and EMT effectors have defined a critical role for the Ca²⁺ signal in the EMT process during tumourigenesis.

Cell migration

The principal component of cancer cell motility is directional migration, which is due to front–rear end polarity (Mayor & Etienne‐Manneville, 2016). Typically, the leading edge is represented by flat cell membrane extensions with directed actin polymerization and nascent attachment sites, whereas at the rear of the cell, adhesions are disassembled and the trailing edge is contracted (Mayor & Etienne‐Manneville, 2016). Interestingly, global cytosolic Ca²⁺ is generally higher at the rear end, whereas Ca²⁺ flickers are enriched near the front edge (Evans et al. 2007; Wei et al. 2009; Tsai & Meyer, 2012). It is suggested that such Ca²⁺ distribution is involved in controlling directed cellular locomotion (Brundage et al. 1991).

Of note, migration is a complex and multistep process that involves coordination between cytoskeleton remodelling, cell‐substrate adhesion/detachment and cellular protrusion/contraction (Gardel et al. 2010; Parsons et al. 2010). Importantly, several key molecular components and signalling events of the cellular migration machinery are Ca²⁺ sensitive (Fig. 2). For example, the myosin II (myo II)‐based actomyosin contraction is mainly mediated through the activity of Ca²⁺‐dependent myosin light chain kinase (MLCK) (Clark et al. 2007). The focal adhesion turnover is also highly dependent on Ca²⁺ signalling. On the one hand, the disassembly of cell adhesions is achieved due to the cleavage of focal adhesion proteins, such as integrins, talin, vinculin and focal adhesion kinases, by the Ca²⁺‐sensitive protease, calpain (Franco & Huttenlocher, 2005). On the other hand, Ca²⁺ is important for the modulation of nascent focal adhesion sites by activating proline‐rich tyrosine kinase 2 (Pyk2), and small GTPases like Ras and Rac (Lysechko et al. 2010; Selitrennik & Lev, 2015). S100 proteins, a subgroup of the EF‐hand Ca²⁺‐binding protein family, regulate a variety of cellular processes via an interaction with different target proteins (Bresnick et al. 2015). In particular, their influence on F‐actin polymerization and myo II–actin assembly has been suggested as governing cell migration due to cytoskeletal structural remodelling (Gross et al. 2014) (Fig. 2). Overall, it is now clear that cell migration can be considered as a Ca²⁺‐dependent process. Importantly, Ca²⁺‐permeable channels are responsible for the cytosolic Ca²⁺ delivery from external and internal cellular stores. Therefore, their activity would define the occurrence of those sustained and transient Ca²⁺ changes which are important for the orchestration of cellular migration.

Figure 2. Global cytosolic Ca²⁺ is generally higher at the rear (marked in red), whereas Ca²⁺ flickers are enriched near the front edge of migrating cell.

The key molecular components and signalling events of the cellular migration machinery are Ca²⁺ dependent. The most studied Ca²⁺‐permeable channels, which are associated with directional migration, are indicated. ERK, extracellular signal‐regulated kinase; MAPK, mitogen‐activated protein kinase; MLCK, myosin light chain kinase; Pyk2, proline‐rich tyrosine kinase 2; ROCK, Rho‐associated protein kinase; STIM, stromal interaction molecule.

Interestingly, in migrating erythrocytes and human umbilical vein endothelial cells, the low basal Ca²⁺ levels at the leading edge are maintained due to the activity of PMCA and the inhibition of PMCA leads to an abrogated front‐to‐rear Ca²⁺ gradient and decreased migration (Pérez‐Gordones et al. 2009; Tsai et al. 2014). Similar mechanisms could be utilized by the metastatic cells, since the expression of PMCA has been found to directly correlate with the tumourigenicity of breast cancer cells (Lee et al. 2005) (Fig. 2). At the same time, in the front end of ER, low local Ca²⁺ concentration provokes high sensitivity to SOCE (Tsai et al. 2014). Indeed, the ER residual Ca²⁺ sensor of SOCE, stromal interaction molecule (STIM), has been found to be distributed along the polar axis in the leading edge of the migrating cell (Tsai et al. 2014). The STIM molecule responds to the ER Ca²⁺ depletion and provokes ion influx through the plasma membrane ORAI channel (Liou et al. 2005; Roos et al. 2005). Of note, STIM–ORAI proteins have been found to be significantly upregulated in various cancer types and SOCE‐activated Ca²⁺ signalling is implemented in the mediation of actomyosin assembly and the focal adhesions required for efficient migration (Chen et al. 2011; Fiorio Pla et al. 2016; Jardin & Rosado, 2016) (Fig. 2).

Plasma membrane extensions and protrusions play the role of a mechanical stress and thus provide Ca²⁺ influx through stretch‐activated channels at the front end of a migrating cell. Indeed, TRPM7 can be activated intracellularly through phospholipase C, or by a membrane stretch (Su et al. 2006; Wei et al. 2009; Gao et al. 2011; Middelbeek et al. 2012). Interestingly, TRPM7 is located in close proximity to calpain and myo II (Clark et al. 2007). Therefore, Ca²⁺ entry provided through TRPM7 modulates actomyosin cytoskeleton contraction, as well as the dynamics of the focal adhesion turnover required for directional cell migration (Clark et al. 2007). Indeed, the pro‐migratory role of TRPM7 has been demonstrated for breast, lung, pancreatic and nasopharyngeal cancers (Visser et al. 2014). Moreover recently, the mechano‐sensitive TRPC1 activation, located at the rear of the cells, has been shown to play a role in the formation of cell polarity of U2OS bone osteosarcoma cells and their directional migration (Huang et al. 2015). Similarly, several members of the TRP channel family have been implicated in cell migration in various cancer types (Fiorio Pla & Gkika, 2013). In particular most TRP channels have been associated with an increase in migration potential. This is the case for TRPC members such as TRPC1 and TRPC6 in glioma cells (Chigurupati et al. 2010; Bomben et al. 2011). In addition, the vanilloid subfamily TRPV2 has also been associated with increased cellular migration in prostate, bladder and breast cancer (Oulidi et al. 2013; Gambade et al. 2016). In contrast, full length TRPM8, has been reported to inhibit cell migration, thus suggesting a protective role for TRPM8 in prostate metastatic cancer progression (Gkika et al. 2010, 2015), whereas the short TRPM8 isoform could have a pro‐metastatic potential (Peng et al. 2015; Bidaux et al. 2016).

Voltage‐gated Ca²⁺ channels (VGCCs) present another pathway for Ca²⁺ influx that activates a downstream mitogen‐activated protein kinase (MAPK)–extracellular signal‐regulated kinase (ERK) signalling pathway and increases migration (Mertens‐Walker et al. 2010). In particular, Cav1.3 has been found to be overexpressed in endometrial carcinoma and its knockdown has been shown to reduce migration (Hao et al. 2015). Indeed, at filopodia tips increased Ca²⁺ concentration provided through L‐type VGCCs directs cancer cell migration due to calpain activation (Jacquemet et al. 2016).

Intracellular Ca²⁺ is an important regulator of Ca²⁺‐activated potassium channels (KCa). Furthermore, ORAI and TRPC1 channels may form complexes with small conductance KCa channel SK3 (Chantome et al. 2013; Guéguinou et al. 2016). Such an SK3–ORAI complex is crucial for the migratory function of breast and prostate cancer cells and has been found in bone metastasis (Chantome et al. 2013). Similarly, colon cancer cell migration is dependent on SOCE through the SK3–TRPC1–ORAI1 channel complex (Guéguinou et al. 2016).

Invasiveness and invadopodia formation

The invasiveness of cancer cells is due to their ability to degrade ECM and migrate into neighbouring connective tissues as well as the lymph‐ and bloodstreams. There, cancer cells spread throughout the organism and give rise to secondary tumour outbursts, metastases. Consequently, the understanding and hence prevention of the process of cancer cell invasion would remarkably improve the survival rate of cancer patients. Cancer cell invasion is achieved due to special structures – invadopodia, which are dynamic actin‐enriched cell protrusions with proteolytic activity. Typically, the invadopodia formation process can be differentiated into the following steps: initiation, assembly and maturation (Fig. 3) (Jacob et al. 2015). The assembly of invadopodia is initiated in response to the focal generation of phosphatidylinositol‐3,4‐bisphosphate and the activation of the non‐receptor tyrosine kinase Src (Mader et al. 2011; Pan et al. 2011; Yamaguchi et al. 2011). The matured invadopodia recruit proteolytic enzymes, such as membrane type 1 (MT1)‐matrix metalloproteinase (MMP), MMP2 and MMP9, to facilitate the focal degradation of the extracellular matrix and allow cell invasion (Beaty et al. 2013).

Figure 3. Ca²⁺ oscillations are required for the initiation of the invadopodia formation process, whereas Ca²⁺ influx activates the focal degradation of the extracellular matrix (ECM), in particular through upregulation of the proteolytic enzymes like matrix metalloproteinases (MMPs) and cathepsins.

The most studied Ca²⁺‐permeable channels, which are associated with invasiveness, are indicated. PI3K, phosphoinositide 3‐kinase; PM, plasma membrane; PtdIns(3,4)P₂, phosphatidylinositol‐3,4‐bisphosphate. N‐WASP, neural Wiscott‐Aldrich syndrome protein; PDGF, platelet‐derived growth factor; RTK, receptor tyrosine kinase.

Intriguingly, a particular pattern of Ca²⁺ signalling, Ca²⁺ oscillations, has been revealed as a predisposing factor for invadopodia formation and activity (Fig. 3) (Sun et al. 2014). For example, Ca²⁺ oscillations mediated through STIM1 and ORAI1 channels have been reported to activate Src kinase and hence facilitate the assembly of invadopodial precursors in melanoma cells (Sun et al. 2014). The proteolytic activity of invadopodia is predetermined by the incorporation of MMP‐containing endocytic vesicles into the plasma membrane at the ECM degradation sites and can also be linked to Ca²⁺ signalling machinery (Bravo‐Cordero et al. 2007). Indeed, the inhibition of SOCE‐abrogated fusion of MMP‐containing vesicles with the plasma membrane results in a constrained ECM degradation (Sun et al. 2014). Moreover, constitutively active TRPV2 engenders an intracellular Ca²⁺ increase and has been associated with an upregulation of MMP9 and the invasive potential of prostate cancer cells (Monet et al. 2010). In oral squamous carcinoma, TRPM8 activity directly correlates with MMP9 activity and the metastatic potential of cells (Okamoto et al. 2012). The downregulation of MMP9 might also be achieved after the inhibition of VGCCs (Kato et al. 2007). Furthermore, in the highly metastatic MDA‐MB‐435 human breast cancer cell line, the activity of the ATP‐gated Ca²⁺‐permeable P2X7 receptor increases invasion by the release of gelatinolytic cysteine cathepsins (Jelassi et al. 2011). Therefore, in invadopodia, Ca²⁺ influx is required for the focal degradation of ECM, in particular through the upregulation of proteolytic enzymes like MMPs and cathepsins, whereas Ca²⁺ oscillations are required for the initiation of the invadopodia formation process (Fig. 3).

Induction of local angiogenesis

The first mechanical and functional interface between blood and tissues is blood vessels. Thus, in order to sustain metastatic dissemination, as well as providing sufficient metabolic support, tumour neovascularization is required. Indeed, tumour cells enable the ‘activation’ of the nearby endothelial cells (ECs) due to the secretion of specific molecules, such as vascular endothelial growth factor (VEGF). Hence, the complex and multistep process of angiogenesis is achieved due to the proliferation, migration, differentiation and stabilization of such tumour‐derived ECs (TECs) in a new circulatory network (Carmeliet, 2005; Folkman, 2006).

It should be noted that in tumours of various origins, Ca²⁺ signalling has been shown to regulate the release of VEGF and hence modulate angiogenesis (Fig. 4). For example, plasma membrane Cav3.2 and ER‐residing STIM1 proteins have been shown to promote angiogenesis in vivo due to the stimulation of VEGF secretion in both prostate and cervical cancers (Chen et al. 2011; Warnier et al. 2015). Moreover, the importance of TRPC6 as a modulator of angiogenic potential has been revealed in glioma cells (Chigurupati et al. 2010). According to this study, the number of EC branch points decreases after growing in a conditioned medium harvested from glioma cells, where hypoxia‐induced TRPC6 overexpression and nuclear factor of activated T‐cells (NFAT) activation are inhibited (Chigurupati et al. 2010). Of note, methyl syringate, which is suggested to be a highly specific and selective agonist of transient receptor potential ankyrin channel 1 (TRPA1) suppresses the hypoxia‐induced migration, invasion and secretion of VEGF in human lung epithelial cells (Park et al. 2016).

Figure 4. Induction of local angiogenesis by Ca²⁺ signalling remodelling.

In tumour cells Ca²⁺ signals regulate the secretion of proangiogenic stimuli, like vascular endothelial growth factor (VEGF). A newly formed vessel can be differentiated into the following structures: tip – represented by the migrating edge of the vessel; stalk – mostly proliferating part of the vessel; phalanx – tightly apposed, regularly ordered ECs that provide perfusion and oxygenation; mural – functional preformed ECs. Interestingly, VEGF‐ and ATP‐mediated Ca²⁺ signals provide proangiogenic effects specifically on tumour‐derived tissue and not on healthy ECs. The most studied Ca²⁺‐permeable channels, which are associated with local angiogenesis, are indicated. NAADP, nicotinic acid adenine dinucleotide phosphate. LP, long persistent; RS, repeated spikes; NSOCE, non‐store‐operated Ca²⁺ entry.

Importantly, during tumour neovascularization, the remodelling of Ca²⁺ machinery has been associated not only with the angiogenic potential of cancer cells, but also with the distinct functions of the ‘activated’ endothelium cells (Fig. 4). Indeed, proangiogenic Ca²⁺ signals and their related pathways are significantly altered in TECs, compared with normal ECs (Fiorio Pla & Munaron, 2014). As an example, Ca²⁺ signals mediated by specific factors like VEGF and ATP and intracellular messengers such as arachidonic acid (AA), nitric oxide, or hydrogen sulfide and cyclic AMP are involved in pro‐migratory effects in TEC, but not in normal ECs (Fiorio Pla et al. 2008, 2010, 2012b; Pupo et al. 2011; Avanzato et al. 2016).

Moreover, the role of intracellular Ca²⁺ increase has been investigated at length in endothelium (Fiorio Pla & Munaron, 2014). Both pro‐ and antiangiogenic molecules can induce an intracellular Ca²⁺ increase, often leading to different biological effects. For instance, Ca²⁺ entry triggered by VEGF, as well as by other proangiogenic factors, is often associated with an increase in vessel permeability, EC survival/proliferation, migration and in vitro tubulogenesis (Dragoni et al. 2011, 2015; Li et al. 2011). These outcomes can be achieved by the activation of a distinct intracellular mechanism, such as SOCE via ORAI and TRPC1 channels (Mehta et al. 2003; Paria et al. 2004; Jho et al. 2005; Abdullaev et al. 2008; Dragoni et al. 2011; Li et al. 2011; Fiorio Pla & Munaron, 2014), non‐SOCE mechanisms via TRPC6 channels (Cheng et al. 2006; Hamdollah Zadeh et al. 2008), the specific engagement of the two‐pore channel TPC2 subtype on acidic intracellular Ca²⁺ stores, resulting in Ca²⁺ release and angiogenic responses (Favia et al. 2014), or by reverse mode activation of NCX (Fig. 4) (Andrikopoulos et al. 2011). Of note, in a recent study VEGF‐mediated Ca²⁺ signalling in individual endothelial cells has been investigated and shown to correlate with both stochastic and deterministic response characteristics to the selection of phenotype‐associated angiogenesis. In particular, altering the amount of VEGF signalling in endothelial cells by stimulating them with different VEGF concentrations triggered distinct and mutually exclusive dynamic Ca²⁺ signalling responses, which correlated with different cellular behaviours such as cell proliferation (monitored by NFAT nuclear translocation) or cell migration (involving MLCK) (Noren et al. 2016). The in vivo role of Ca²⁺ signals has been recently studied in zebrafish, during angiogenic input by means of high‐speed, three‐dimensional time‐lapse imaging to describe intracellular Ca²⁺ dynamics in ECs at single‐cell resolution (Yokota et al. 2015; Noren et al. 2016). It may be noted that TRP Ca²⁺‐permeable channels have profound effects on the control of different steps of tumour angiogenesis. Besides their role in the VEGF‐mediated Ca²⁺ signals previously described, several data clearly show their involvement Ca²⁺‐mediated signal transduction with a prominent roles in tumour angiogenesis. In this context, TRPV4 is an emerging player in angiogenesis as on ECs it acts as a mechano‐sensor during changes in cell morphology, cell swelling and shear stress. TRPV4 plays a significant role in endothelial migration, (Fiorio Pla et al. 2012b) displaying a marked increase in ECs derived from human breast carcinomas, as compared with ‘normal’ ECs, leading to a greater Ca²⁺ entry that in turn activates migration in TECs (Fig. 4) (Fiorio Pla et al. 2012b). Moreover, TRPV4 has recently been described as an important player in tumour vasculature normalization, thereby potentially improving cancer therapies (Adapala et al. 2015; Thoppil et al. 2015, 2016). In addition, TRPM2 has been recently identified as mediating an H₂O₂‐dependent increase in macrovascular pulmonary EC permeability (Fig. 4) (Hecquet et al. 2008; Mittal et al. 2015). TRPM7 inhibits human umbilical vein endothelial cell proliferation and migration, whereas its functions in human mammary epithelial cells seem to be the opposite (Fig. 4) (Inoue & Xiong, 2009; Baldoli & Maier, 2012; Baldoli et al. 2013; Zeng et al. 2015). Recently, TRPA1 has been found to have a role in the vasodilatation of cerebral arteries, via an increase in Ca²⁺ influx generated by the detection of reactive oxygen species, a process that requires the peroxidation of membrane lipids (Sullivan et al. 2015). Similarly, TRPV2 has been shown to be expressed in aorta endothelium, but no clear functional data have been reported (Earley, 2011).

Finally, the emerging family of mechanosensitive Piezo channels has recently been described in vascular endothelial cells: Piezo2 knockdown is involved in glioma angiogenesis, both in vitro and in vivo by promoting abnormal intracellular Ca²⁺, Wnt11/β‐catenin signalling reduction, leading to altered angiogenic activity of endothelial cells (Fig. 4) (Yang et al. 2016).

In conclusion, due to its multifaceted role in the control of endothelium homeostasis, Ca²⁺ machinery is a potential molecular target for strategies against tumour neovascularization.

Conclusions

A remodelling of Ca²⁺ signalling plays an important role during tumourigenesis. Interestingly, there are some specific channel patterns through which such Ca²⁺ signals occur, at the different stages of cancer progression. This could be partially explained by the specificity of Ca²⁺ flux, its compartment localization and the proximity of downstream Ca²⁺‐dependent targets. Furthermore, some ion channels represent multimodal activity and are characterized not only as Ca²⁺‐permeable pore proteins, but also as possessing other functional domains. For example, the C‐terminal end of TRPM7 is constituted by a serine/threonine protein kinase domain and hence, due to the phosphorylation of cytoskeletal components, regulates cellular migration (Clark et al. 2008).

Importantly, plasma membrane Ca²⁺ channels are easily and directly accessible via the bloodstream. Therefore, they are potential targets for a variety of therapeutic strategies, such as their regulation on a transcriptional and translational level, their trafficking to the plasma membrane or their stabilization at the plasma membrane (Gkika & Prevarskaya, 2009; Fiorio Pla et al. 2012a; Bernardini et al. 2015; Earley & Brayden, 2015).

Additional information

Competing interests

The authors have no conflicts of interest to declare.

Author contributions

O.I. and A.F.P. collected information, conceived the concept, prepared figures and drafted the manuscript. N.P. was involved in supervising and editing the manuscript. All authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Funding

O.I. was funded by the ‘la Ligue Nationale Contre le Cancer’, France (GB/MA/CD 11813) and Laboratory of Excellence in Ion Channel Science and Therapeutics. The work of A.F.P. and N.P. was funded by Institut National du Cancer. The research in the authors' laboratory is supported by INSERM, la Ligue Nationale Contre le Cancer, Le Ministere de l'Education Nationale, the Region Nord/Pas‐de‐Calais, la Fondation de Recherche Medicale, and l'Association pour la Recherche sur le Cancer.

Biographies

Oksana Iamshanova is a PhD student at the University of Lille 1 (France) in the Laboratory of Cell Physiology headed by Natalia Prevarskaya. Investigating the role of ion channels in tumorigenesis is one of her major research interests.

Alessandra Fiorio Pla is associate professor at the University of Torino (Italy) and visiting professor at the University of Lille 1 (France). Her research activity is focused on Ca²⁺ channels characterization, in particular TRP channels, and their involvement in angiogenesis and tumour vascularization/progression by combining Ca²⁺ imaging and live cell imaging techniques with cell biology.

Natalia Prevarskaya is full‐time professor of physiology at the University of Lille 1 (France), director of the Laboratory of Cell Physiology and head of the team ‘Ca²⁺ signatures of prostate cancer’ certified by the National League Against Cancer. Her scientific interests are the function and regulation of ion channels, the role of ion channels and Ca²⁺ signalling in carcinogenesis, Ca²⁺ signalling in proliferation, apoptosis, migration and differentiation, prostate cancer and prostatic diseases.