The Role οf Ion Channels in the Development and Progression of Prostate Cancer

Ion channels have major regulatory functions in living cells. Apart from their role in ion transport, they are responsible for cellular electrogenesis and excitability, and may also regulate tissue homeostasis. Although cancer is not officially classified as a channelopathy, it has been increasingly recognized that ion channel aberrations play an important role in virtually all cancer types. Ion channels can exert pro-tumorigenic activities due to genetic or epigenetic alterations, or as a response to molecular signals, such as growth factors, hormones, etc. Increasing evidence suggests that ion channels and pumps play a critical role in the regulation of prostate cancer cell proliferation, apoptosis evasion, migration, epithelial-to-mesenchymal transition, and angiogenesis. There is also evidence suggesting that ion channels might play a role in treatment failure in patients with prostate cancer. Hence, they represent promising targets for diagnosis, staging, and treatment, and their effects may be of particular significance for specific patient populations, including those undergoing anesthesia and surgery. In this article, the role of major types of ion channels involved in the development and progression of prostate cancer are reviewed. Identifying the underlying molecular mechanisms of the pro-tumorigenic effects of ion channels may potentially inform the development of novel therapeutic strategies to counter this malignancy.


Introduction
Numerous studies have firmly established the role of ion channels in essentially all basic cellular functions [1]. Apart from their role in ion transport, they can form macromolecular complexes and interact with adhesion proteins, signaling molecules, and many other [2]. Ion channels are not only responsible for cellular electrogenesis and excitability but they also regulate the necessary conditions for tissue homeostasis, such as cellular differentiation, proliferation, and apoptosis [2].
Several molecular pathways and cellular events can be affected by changes in the ion composition inside the cells. For example, cell movement requires a sequence of cellular retractions and protrusions, in which ion channels play a crucial role [2]. The shape of the cell is largely determined by actin, microtubules and molecular motor proteins, such as myosin. They all interact and modulate the activity of ion channels [2]. Furthermore, specific ion channels are critical determinants of cell volume [3].
Research to date has shown that all cancer hallmarks require the activity of ion channels [4]. Although cancer is not officially classified as a channelopathy, it has been increasingly recognized that ion channel aberrations play an important role in virtually all cancer types, exerting protumorigenic activity due to genetic or epigenetic alterations or as a response to molecular signals [4][5][6][7][8][9]. These findings open the way to novel pharmacological strategies and to the development of personalized therapeutic approaches [6].
Prostate cancer (PCa) is the second leading cause of cancer-related death in men in the United States [10,11]. Most patients with PCa will run an indolent disease course, while a subset of patients will have an aggressive history [11]. In general, more aggressive tumors tend to have different morphological features and they upregulate markers suggestive of increased proliferation [12,13]. First-line antiandrogen therapy in patients with metastatic PCa can result in significant complete response rates. However, PCa cells that have been exposed to treatment often show distinct morphological features [14,15]. This raises the possibility that apart from their role in supporting the typical hallmarks in PCa, ion channels might also play a role in treatment resistance and subsequent failure.
In this article, the role of major types of ion channels involved in the development and progression of PCa are reviewed. Identifying the underlying molecular mechanisms of the pro-tumorigenic effects of ion channels may potentially inform the development of novel therapeutic strategies to counter this malignant ailment.
Inward rectifying potassium channels are critical for the maintenance of resting potential, by conducting inward currents easier than outward currents. Two-pore domain channels are constitutively open as 'leak channels', maintaining a negative membrane potential [2,16]. In general, cancer cells tend to be more depolarized than normal cells [17]. Potassium channels also regulate cell volume and neoangiogenesis, while some are involved in the proliferation and apoptosis of PCa cells [16,[18][19][20]. In vitro experiments by Abdul et al. showed that the potassium channel openers minoxidil, 1-ethyl-2-benzimidazolinone and diazoxide increase the growth of PC3 cells, while the potassium channel blockers amiodarone, dequalinium and glibenclamide exert dosedependent growth-inhibiting effects [21]. Comparison between androgen-dependent LNCaP cells and androgenindependent PC3 by Laniado et al. suggested that PC3 cells have a more excitable membrane due to a lower density of voltage-gated K + current, potentially reflecting their differential metastatic character [22]. This is also thought to render LNCaP cells more susceptible to apoptosis. It has been suggested that enhanced K + efflux is proapoptotic, while decreased K + efflux is antiapoptotic (Table 1) [16,23]. During apoptosis, K + channels are associated with decay of the membrane potential, calcium overload, apoptotic volume decrease, and activation of proapoptotic intracellular effectors [16,24]. Potassium channels have also been linked to senescence. Oncogenic stress increases the expression of Kv1.1, which translocates to the cellular membrane and changes the membrane potential [16,24,25]. Lallet-Daher et al. reported that Kv1.1 expression is decreased in cancer, and this reduction correlates with breast cancer disease aggressiveness [25]. Although more research is needed, relocation of Kv1.1 to the cell membrane induced a membrane potential change, which resulted in cellular senescence (Table 2) [25].
In addition, studies have shown that VGPC-induced electrical excitability in cancer cells affects metastatic invasiveness. It has been suggested that Kv2.1 channels might play a role in the migration of PCa cells [26]. Hu et al. reported evidence that increased expression enhances bone marrow mesenchymal stem cell migration under hypoxic preconditioning, while Kv2.1 also induces cell motility and focal adhesion kinase activation [27]. In vitro experiments by Park et al. showed that its expression levels and activity are increased in the highly metastatic cell line PC3 compared with other PCa cell lines of lower metastatic potential [26]. Blockade of Kv2.1 with stromatoxin-1 or small-interfering RNA (siRNA) against Kv2.1 resulted in cell migration inhibition, without affecting cell proliferation. Further experiments showed that reactive oxygen species (ROS) reduction with N-acetyl-l-cysteine or ascorbic acid resulted in Kv2.1 downregulation [26]. Although more research is needed, these data suggest that Kv2.1 might be a part of ROS-related signaling and link ROS formation with tumor metastatic potential in PCa. It is unknown if Kv2.1 overexpression is an inherent driver in PCa migration or a physiological response to high ROS [26].
On the other hand, Abdul et al. reported an inverse immunohistochemistry correlation between Kv1.3 expression in normal prostate epithelium and the grade and stage of primary prostate tumors (92 primary PCa samples) [28]. The activity of Kv1.3 is needed for the proliferation of Kv1.3-expressing normal and cancer cells as it promotes cell proliferation by setting the cell membrane potential and driving the 'force' for calcium influx, as well as by activating the MEK-ERK signaling pathway [29,30]. However, its expression was found to be higher in weaker metastatic PCa models and lower in strongly metastatic models [29]. Mitochondrial Kv1.3 channels also likely play a crucial role in the induction of the mitochondrial pathway of apoptosis.  [16,18] Increase the growth of PC3 cells [21] Induce cellular senescence [16,24,25] Enhance cell migration under hypoxic preconditioning [26,27] Promote cell proliferation by setting the cell membrane potential and driving the 'force' for calcium influx [28][29][30] Cause transient inner mitochondrial membrane hyperpolarization, increase ROS production by mitochondria, and enhance transition pore permeability (PTP) and cytochrome C release [29] Form functional complexes that set the resting potential and promote cell proliferation [33] Increase the tumor cell resistance to serum deprivation and hypoxia in poorly oxygenated areas of the tumor [39,40] Sodium Are involved in prostate cancer growth, invasion, and metastasis [16,41] Promote intracellular alkalization and extracellular acidification, disrupting the integrity of the microenvironment and enhance tumor cell metabolic adaptation, proliferation, migration, and metastasis [41,55,56,62] Activate voltage-gated calcium channels that increase intracellular calcium levels, resulting in increased formation of invadopodia/podosomes and increased invasive ability [41,46,47] Regulate gene expression regulation, galvanotaxis, endocytosis, and secretion [49] Increase adhesion, process outgrowth, invasion, and migration [53,96] Changes cell volume and enhances the development of invadopodia [57][58][59]61] Are involved in extracellular lysosome trafficking and extracellular acidification [60] Calcium Allow the intracellular influx of calcium ions, activating several downstream signaling pathways associated with prostate cancer tumorigenesis and progression [2,5,16,64,67,71] Promote intracellular calcium oscillations that can increase survival and cell proliferation [63,66,67,70,[86][87][88][89][90] Mediate a persistent transfer of calcium ions from the endoplasmic reticulum to the mitochondria [80] Are associated with enhanced tumor growth, modulation of androgen receptor, and acquisition of neuroendocrine characteristics [81][82][83][84] Promote HIF-1α degradation and decrease hypoxia-induced invasion and migration of androgen-independent cells [94,95] Mediates the progression towards the aggressive castration-resistant stage [96] Chloride Contribute to cell volume changes, regulating cancer cell migration and enhancing apoptotic resistance [16,104] Promote chemotherapy resistance by modulating intracellular pH [2,5] Enhance the endogenous expression of ClC-3 protein, strengthening the ability of cells to maintain volume constancy and avoid apoptosis [106] Promote tumor transition to androgen independence and apoptosis resistance [16,106,107] Promote cell proliferation, migration and invasion [108] Regulate cell proliferation and migration through an MAPK/ERK-dependent pathway [ Enhances prostate cancer metastatic potential, through decreased E-cadherin and increased c-Myc expression [112] Involved in endo-lysosomal pH, vesicle trafficking, migration, and invasion [116] Involved in calcium concentration in the mitochondria [136] Outer membrane-inserted Bax binds to mitochondrial Kv1.3 and leads to transient inner membrane hyperpolarization, increased ROS production, activation of permeability transition pore (PTP), and cytochrome C release [29]. Hence, downregulation of this channel might decrease the proliferation rate but render PCa cells resistant to the signals of the mitochondrial pathway of apoptosis. Increased Kv10.1 expression has also been associated with increased tumor cell proliferation [16]. Bloch et al. reported that BK channels KCa1.1 were overexpressed in 16% of late-stage PCa and PC3 cells versus 0% in normal controls, while BK inhibition with iberiotoxin and siRNA reduced PC3 cell proliferation [31]. In the androgendependent LNCaP cell line, most of the calcium-dependent K + current is carried by BK channels [32]. BK currents can be activated by very low levels of intracellular Ca 2+ . A transient Ca 2+ intracellular entry through T-type low voltageactivated Ca 2+ channels (Cav3.2 channels) can induce a persistent BK channel activation. Cav3.2 and BK channels form functional complexes that set the resting potential and promote cell proliferation [33]. The calcium-sensitive potassium channel SK3 (KCa2.3) has been recently suggested by Bery et al. to be involved in neuroendocrine differentiation of PCa [34]. The scientists also reported that ohmline, an SK3 inhibitor, was able to prevent the expression of neuroendocrine markers induced by enzalutamide. Moreover, Lallet-Daher et al. provided evidence that IK KCa3.1 activation controls cell cycle progression, through membrane potential hyperpolarization that induces calcium entry via TRPV6, a cation channel of the transient receptor potential family ( Table 2, Fig. 1) [35]. Although these data suggest that KCa1.1 (BK) and KCa3.1 (IK) are involved in PCa tumorigenesis, gene expression analysis by real-time PCR showed that these molecules are overexpressed in the group with Gleason score 5 or 6 (compared with normal prostate tissue) but decreased in Gleason score 8 or 9 tumors [36].
K2P channels are responsible for both tumorigenesis and apoptosis [37]. Zhang et al. reported that K2P2.1 expression in 100 PCa tissues by immunohistochemistry was positively correlated with Gleason score, T stage, undifferentiated state, and shorter time to castration resistance [38]. They also found that K2P2.1 knockdown inhibits cell proliferation both in vitro and in vivo, and induces a G1/S cell cycle arrest [38]. K2P9.1 amplification has also been reported in PCa. Studies suggested that K2P9.1 overexpression can increase the tumor cell resistance to serum deprivation and hypoxia in poorly oxygenated areas of the tumor [39,40]. Furthermore, KChAP is a K + -channel regulatory protein that causes 'chaperone-like' increases in K + -channel expression. Experiments with KChAP have shown the importance of augmented K + efflux in apoptosis. KChAP activity in LNCaP cells was also linked to average cell volume decrease, promotion of apoptotic death, and G0/G1 cell cycle arrest in vitro and in vivo [16].

Sodium Channels
Voltage-gated sodium channels (VGSCs) play a critical role in the generation of electrochemical action potential in excitable cells [2]. However, they are also expressed in 'non-excitable' cells, such as glial cells, fibroblasts, and some immune cells of the myeloid lineage [41], and are also expressed in cancer cells, including PCa [41]. Studies have shown that VGSCs are involved in PCa growth, invasion, and metastasis [16,41]. The underlying mechanism remains unclear and represents an area of active research.
VGSCs are transmembrane glycoproteins that are composed of one alpha subunit and one or more beta subunits. The alpha subunits are the functional centers [2,42]. There are nine alpha subunits, termed Nav1.1-Nav1.9, which are encoded by nine distinct genes (SCN1A-SCN11A). The auxiliary beta subunits have five subtypes, encoded by the genes SCN1B-SCN4B [43]. They regulate the expression and gating of VGSCs [2,41,44]. In general, the intrinsic concentration of VGSCs is higher in various cancer tissues compared with the neighboring normal tissue [45]. It has  [41,45]. Strong Na + influx results in intracellular alkalization and extracellular acidification. An extracellular acidic pH can disrupt the integrity of the microenvironment and enhance tumor cell migration and metastasis [41]. Na + influx also activates voltage-gated calcium channels. This subsequently increases intracellular calcium levels, which results in increased formation of invadopodia/podosomes and increased invasive ability [41,46,47]. Yildirim et al. provided evidence that in a Copenhagen rat model of PCa, tetrodotoxin, a VGSC inhibitor, reduced lung metastases by > 40% in vivo and extended the lifespan of experimental animals [48]. VGSCs have also been associated with gene expression regulation, galvanotaxis, endocytosis and secretion (Table 1) [49]. Whether VGSC upregulation is a primary driver of invasiveness remains debatable. Nav1.7 is predominantly expressed in PCa [16,41], but Nav1.6 and Nav1.8 were also found to be upregulated in PCa cells [41,50]. Strongly metastatic PCa PC3 cells express significantly more VGSCs compared with weakly metastatic LNCaP cells. Experiments using the patch clamp technique to record currents of whole cells showed that Nav1.6 and Nav1.7 expression had no effect in LNCaP cells but were highly functional in PC3 cells [41]. Bennett et al. suggested that the expression of VGSCs alone was sufficient and necessary to increase the invasive potential of PCa cell lines in vitro [51]. In mouse models of strongly metastatic Mat-LyLu PCa cells, nerve growth factor upregulated the functional expression of Nav1.7 [52]. The pan-tyrosine receptor kinase (pan-TRK) antagonists K252a and KT5720 (a protein kinase A inhibitor) suppressed the nerve growth factor-induced peak VGSC current density increase [52]. In strongly metastatic Mat-LyLu PCa cells, there was positive feedback in the expression of Nav1.7 [41]. It has also been suggested that naringenin (which reduces the expression of the gene that encodes for Nav1.7) can result in a VGSC-mediated reduction of invasion and proliferation in Mat-LyLu cells. Furthermore, Jansson et al. provided evidence that subunit β2 overexpression in LNCaP cells can be associated with increased adhesion, process outgrowth, invasion, and migration in vitro and in vivo [53].
Apart from voltage-gated sodium channels, there are several exchanger proteins that involve sodium ion transport. Examples include the Na + /H + exchanger (NHE1), the Na + /K + /2Cl − cotransporter (NKCC), and the Na + / HCO3 − cotransporter. They play a critical role in cellular pH maintenance by utilizing the Na + electrochemical gradient to transport other ions [54]. Notably, the tumor microenvironment is often hypoxic and tumor cells create a pH gradient, with a higher pH intracellularly and a lower pH extracellularly. This has profound effects on cellular proliferation, apoptosis evasion, and metabolic adaptation [55,56]. Moreover, in order to explore and invade the surrounding microenvironment, tumor cells develop invadopodia while retracting the rear end [57]. This requires changes in cell Fig. 1 The role of potassium channels in PCa. Potassium channels play a major role in the maintenance of resting potential. However, aberrant expression is a frequent finding in PCa. Kv1.3, Kv10.1 and BK channel overexpression has been associated with increased PCa proliferation. SK3 channel activity has been linked to neuroendocrine differentiation. K2P9.1 overexpression has been linked to increased proliferation and resistance to hypoxic conditions in PCa. Increased Kv2.1 activity has been associated with increased motility and migration in PCa. Mitochondrial Kv1.3 channels also play a critical role in the induction of the mitochondrial pathway of apoptosis. PCa prostate cancer, BK big conductance volume, a process regulated by local ion transport through the NHE1, the Na + /K + /2Cl − cotransporter (NKCC), and the Na + /HCO3 − cotransporter [58]. Li et al. showed that although NHE1 knockdown affected migration in DU145 PCa cells, this effect cannot be attributed to NHE1-dependent translocation of protons [59]. However, Steffan et al. suggested that multiple NHE1 isoforms are involved in extracellular lysosome trafficking and extracellular acidification in PCa [60]. Interestingly, the NHE1 inhibitors troglitazone and ethyl-isopropyl-amiloride prevented lysosome trafficking to the periphery of the cell [60]. On the other hand, Hiraoka et al. reported that NKCC inhibition with bumetanide and furosemide decreased cell growth of androgen-independent PC3 cells [61]. It has also been suggested that the Na + / HCO3cotransporter NBCe1 is a critical protein responsible for the acidic microenvironment in PCa. NBCe1 expression increases in hypoxic conditions and enhances tumor cell proliferation under these conditions [62]. PCa tissue microarray immunohistochemistry revealed robust NBCe1 expression in acinar and ductal adenocarcinoma. NBCe1 knockdown or pharmacological inhibition with S0859 in LNCaP and PC3 cell lines resulted in decreased cell number viability, increased cell death, and reduced cell sphere-formation in three-dimensional (3D) cultures [62].

Calcium Channels
Disruption in calcium homeostasis is a well-known phenomenon in cancer [62]. Three major classes of membrane-associated proteins are involved in calcium regulation: channels, exchangers, and pumps (ATPases). Extracellular calcium ions enter cells via different classes of channels [5]. These include (1) voltage-gated channels (activated by depolarization of the membrane); (2) second messenger-operated channels (activated by small molecules, such as cyclic nucleotides, inositol phosphate, or lipid-derived messengers); (3) receptor-operated channels (activated by the binding of a hormone or a neurotransmitter); and (4) store-operated channels (SOC) activated by the depletion of intracellular stores of calcium ions. In non-excitable cells, SOCs are among the main calcium entries [5]. They typically allow the intracellular influx of calcium ions, as a result of endoplasmic reticulum calcium depletion [5]. Mediators of the rise of intracellular calcium after intracellular store depletion include depletion sensors (STIM1, STIM2), Orai channels (highly selective calcium channels), and TRP (transient receptor potential) channels (non-selective calcium channels) [63]. This is not only necessary in order to refill the internal calcium stores but also to activate several downstream signaling pathways [2,16,64].
Resting intracellular Ca 2+ concentration is lower than the extracellular fluid (10-100 nM vs. 1.2 mM approximately). During cell stimulation, intracellular Ca 2+ levels can increase significantly, up to the micromolar level. However, large and sustained cytosolic calcium increases might potentially trigger apoptosis [65]. In general, sustained cytosolic calcium induces apoptosis, while intracellular calcium oscillations are usually associated with increased survival and cell proliferation [63,66,67]. Some calcium channels are constitutively active in resting conditions. These baseline influxes have been associated with tumorigenesis [63]. PCa is characterized by calcium signals that are different in subcellular localization, amplitude, and signal kinetics compared with normal cells. Overexpression of calcium channels leads to increased cytosolic Ca 2+ and overactivation of calciumbinding proteins. This results in overactivation of cellular processes associated with PCa tumorigenesis and progression [67]. For example, calcium/calmodulin-dependent kinase II (CAMKII) plays an important role in the resistance to antiandrogen therapy and progression of PCa to an androgen-independent state [67].
Orai1 homomultimers (calcium channel subunits) are proposed to be parts of the calcium release-activated calcium channels (CRAC) in the store-operated calcium entry (SOCE) pathway. It is believed that Orai1-Orai3 heteromultimers form at the detriment of Orai1 homomultimers and favor store-independent Ca 2+ entry [67,68]. Enhanced Orai3 expression has been associated with PCa progression [68]. Orai1-Orai3 heteromultimers are shown to promote calcium-dependent cell proliferation, while Orai1 homomultimeric channels can potentially trigger calcium-dependent apoptosis [67,68]. It has been reported that arachidonic acid (Orai1/Orai3 activator) decreases the formation of Orai1 channels and decreases SOCE in vitro and ex vivo in freshly isolated primary PCa tumor models (post-radical prostatectomy). Several studies have also linked arachidonic acid metabolism with PCa proliferation, via calcium ion entry [69]. It has also been suggested that Orai1/Orai3 multimers play an important role in proliferation stimulation following M3 muscarinic receptor activation (Table 3) [69].
Alterations in TRP melastatin 2 (TRPM2), TRPM4, TRPM8 and TRP Vanilloid 1 (TRPV1) and TRPV6 have also been reported in PCa [67]. TRPV6 translocation to the plasma membrane has been associated with constitutively increased intracellular calcium concentrations, which results in increased cell survival [70]. Although undetectable in healthy and benign prostate tissue, TRPV6 expression was shown to correlate with PCa grade, while bone metastasis models and xenografts in nude mice confirmed the link between increased tumor aggressiveness and TRPV6 overexpression [70,71]. In vitro experiments by Lehen'kyi et al. suggested that androgen receptor (AR) upregulates TRPV6 in a ligand-independent manner in LNCaP cells. Moreover, they showed that overexpressed TRPV6 channels are constitutively open and result in increased cytosolic calcium, Increase proliferation Favor store-independent calcium entry, M3 muscarinic receptor activation Fig. 2 Role of calcium channels in PCa. Calcium signals in PCa cells are different in subcellular localization, amplitude, and signal kinetics compared with normal cells. Several calcium ion channels are overexpressed in PCa. Orai3-Orai1 heteromultimer activity has been associated with increased PCa cell proliferation. Orai1 homomultimers can trigger calcium-dependent apoptosis. TRPM2 and TRPM4 overexpression has also been associated with increased PCa proliferation. TRPV2 and TRPM7 overexpression has been linked to increased cell proliferation and migration in PCa. Increased TRPV6 expression has been associated with increased PCa proliferation, survival (apoptosis evasion), and the development of osteoblastic bone metastases. Cav3.2 overexpression has been linked to increased tumor growth and neuroendocrine differentiation in PCa, whereas Cav1.3 overexpression has been linked to resistance to antiandrogen therapy. PCa prostate cancer which increases cell proliferation, via activation of NFATmediated signaling pathways (Fig. 2) [71]. TRPM8 gene expression is directly regulated by the AR, while it has been suggested that TRPM8 is essential for the survival of androgen-dependent LNCaP cells [67,72]. Localization of TRPM8 in the endoplasmic reticulum has been associated with calcium release from intracellular stores to the cytosol, and increased cell survival in LNCaP cells. On the other hand, androgen-independent PC3 cells have been reported to express low levels of TRPM8 [72]. Interestingly, recent studies by Genovesi et al. found that TRPM8 was also robustly expressed in metastatic castrationresistant PCa mouse models [73]. It has been suggested that TRPM8 overactivity is linked to anti-migratory effects and reduced motility in PCa cells [74]. Permanent transfection of TRPM8 resulted in increased susceptibility of PC3 cells to apoptosis, decreased proliferation and migration ability [74]. Alaimo et al. suggested that TRPM8 agonists might potentially sensitize treatment-refractory models of PCa to hormonal treatment, radiotherapy and chemotherapy [75]. Moreover, Genovesi et al. reported that coadministration of the highly selective TRPM8 agonist D-3263 with sublethal doses of enzalutamide and docetaxel resulted in massive apoptotic response in therapy-resistant PCa models [73]. Di Donato et al. found that while AR-negative PCa remain relatively unaffected by TRPM8 antagonists, the latter can inhibit androgen-dependent proliferation, invasiveness and migration in AR-expressing castration-resistant PCa cells. Selective antagonists interfered with non-genomic AR activity, abolished the assembly of AR/TRPM8 complex, and increased intracellular calcium levels [76]. Di Sarno et al. also reported the synthesis of a new series of TRPM8 inhibitors with remarkable efficacy against androgen-induced growth in several two-dimensional (2D) and 3D models of PCa [77].
Both TCAF1 (TRP channel-associated factor 1) and TCAF2 increase TRPM8 translocation to the plasma membrane, but they exert opposite effects on TRPM8 activity through a PI3K-dependent pathway. Evidence suggests that TCAF1 is a TRPM8 activator in PCa cells and functional interaction of TCAF1/TRPM8 can reduce both the speed and directionality of migration, while TCAF2 promotes migration [78]. Studies have shown that both the expression levels of TRPM8 and TCAF1 increase in cancerous prostate tissue [75,78].
Calcium channels mainly located in the endoplasmic reticulum (such as IP3 receptors), have been associated with apoptosis in PCa [79]. They can mediate a persistent transfer of calcium ions from the endoplasmic reticulum to the mitochondria [80]. This results in increased calcium concentration inside the mitochondria, which triggers the mitochondrial-dependent apoptotic pathway [67]. Proteasomal degradation of IP3R3 by FBXL2 overactivation (e.g., due to PTEN loss) can potentially inhibit mitochondrial apoptosis. Continuous mitochondrial calcium overload leads to apoptosis, while intermittent and low mitochondrial calcium is survival-promoting and metabolism-stimulating [67,79].
Overexpression of voltage-gated T-type calcium channels (TTCCs) such as Cav3.2 has been associated with enhanced tumor growth in PCa and acquisition of neuroendocrine characteristics [81,82]. Cav1.3 is highly expressed in PCa, especially in castration-resistant disease [64]. Studies have shown that it is upregulated during androgen deprivation therapy and promotes resistance to antiandrogen therapy [83]. Following androgen deprivation therapy with bicalutamide, PCa models were found to express a shortened 170 kDa Cav1.3 isoform, which failed to induce influx of calcium after membrane depolarization. However, it mediated an increase in SOCE and a rise in basal cytosolic calcium [83,84]. Cav1.3 was also shown to modulate AR transactivation and tumor growth [84]. Blocking L-type channels significantly decreased androgen-stimulated influx of calcium. Chen et al. also reported evidence that Cav1.3 expression might be linked to TMPRSS2-ERG gene fusion (Table 3) [84]. In LNCaP cells, the putative calcium channel α2δ2 auxiliary subunit was found to stimulate proliferation. Tumor cells overexpressing α2δ2 were shown to be more tumorigenic compared with control cells. Moreover, gabapentin (an α2δ2 ligand) reduced tumor development in the LNCaP-derived xenograft model [85]. Experiments by Wang et al. suggested that TRP canonical 6 (TRPC6) channels mediate the hepatocyte growth factor (HGF)-induced cytosolic calcium increase, and subsequently enhances cell proliferation [86]. TRPC6 inhibition in DU145 and PC3 cells resulted in G2/M phase arrest and suppressed HGFinduced proliferation [86].
TRPM4 overexpression has been associated with increased proliferation in PC3 cells, via activation of β-catenin and Akt signaling pathways [87]. TRPM4 expression has also been linked with increased biochemical recurrence after radical prostatectomy [88]. Stokłosa et al. also provided preliminary evidence suggesting that TRPM4 channels play a role in calcium-induced exocytosis in PCa cells, which depends on ion conductivity in TRPM4containing intracellular vesicles [89]. In addition, TRPM2 upregulation and nuclear localization have been observed in PCa cells but not in normal prostate tissue. Zeng et al. suggested that TRPM2 is essential for proliferation in cancer cells [90]. Scientists showed that TRPM2 knockdown inhibited the growth of PCa but had no effect in non-cancerous cells [90]. Subcellular localization was also remarkably different, with TRPM2 expression in benign prostate cells being homogeneous in the cytoplasm and near the cell membrane, while in PC3 and DU145 PCa cell lines, a significant amount was located inside the nuclei. Increased extracellular calcium concentration also contributes to increased tumor cell proliferation through the activation of calcium channels [90,91]. It has been suggested that PCa cells are able to recognize extracellular calcium by the P2X receptor or the calcium-sensing receptor [91]. Sun et al. showed that silencing of TRPM7 inhibits magnesium-nucleotide-regulated metal currents. Increased calcium/magnesium ratios facilitate a higher calcium ion influx, especially in PCa cells, which was shown to promote the proliferation of PC3 and DU145 cells. PCa patients frequently have a high serum calcium/ magnesium ratio [92].
Moreover, TRPs have been associated with tumor neoangiogenesis. TRPV2 overexpression has been linked to proliferation in PCa-derived endothelial cells [67]. It has also been suggested that TRPC3 acts as an endothelial cell attraction factor in PCa and is controlled by ER calcium filling [16,93]. The auxiliary subunit α2δ2 has been also shown to play a role in angiogenesis in LNCaP models [85]. Moreover, TRPM7 upregulation in PCa cells promotes epithelial-tomesenchymal (EMT) transition and increases tumor migration. Studies have also shown that suppression of TRPM7 promotes HIF-1α degradation and decreases hypoxiainduced invasion and migration of androgen-independent cells [94,95]; however, the involvement of calcium in the process remains debatable [95].
It has been shown that the introduction of TRPV2 into androgen-dependent LNCaP cells increases the expression of MMP-9 and cathepsin beta, and enhances cell migration [96]. Moreover, TRPV2 plays a role in the progression towards the aggressive castration-resistant stage [96]. Wang et al. reported that TRPC6 also promotes cell migration, invasion and metastasis in PCa [97]. Apart from the survival-promoting effect in prostate tumors, TRPV6 upregulation has been linked to the development of osteoblastic bone metastases. Of note, it has been reported that PC3 cells that overexpress TRPV6 can generate osteoblastic bone metastases, as opposed to the control PC3 cells, which generate osteolytic bone metastases [67,70]. Apart from being critical drivers of neuroendocrine differentiation, SK3 channels form complexes with Orai1 and likely play a (Ca 2+ -dependent) role in bone metastasis formation [98].

Chloride Channels
Chloride channels are a family of ion channels that are specific for chloride, have 10-12 transmembrane domains, and are divided into voltage-gated and ligand-gated chloride channels [99,100]. Chloride channels are also found in the membranes of various organelles and display a variety of physiological roles, such as regulation of excitable cells, volume homeostasis, pH regulation, cell cycle regulation, organic solute transport, and transepithelial transport [2,16,101]. These channels also contribute to cell volume changes, and hence play an important role in the regulation of cancer cell migration [2]. Volumeregulated anion channels (VRACs) are proteins in which leucine-rich repeat-containing protein 8A (LRRC8A) heteromers with other LRRC8 multispan membrane proteins play critical structural and functional roles [102]. Volume regulation is important during proliferation, differentiation, exocytosis and cellular movement [16,103]. Disrupted volume regulation can promote apoptosis, while strengthening the regulatory volume, through activation of chloride currents in response to cell swelling, enhances apoptotic resistance [104].
Chloride channel upregulation has been reported to have a pro-tumorigenic effect in several cancer types and promote chemotherapy resistance by modulating intracellular pH [2,5]. LNCaP cells show a remarkable ability to use chloride currents to regulate tumor cell volume under hypo-osmotic stress. This ability further increases with cell transition to androgen-independent states [16]. Neuroendocrine differentiation of LNCaP cells is associated with a wide rearrangement of the entire Ca 2+ homeostasis [105,106]. Studies have suggested that during this process, store-operated Ca 2+ channels (SOCs), one of major players in Ca 2+ homeostasis, are functionally coupled with VRACs [107]. Endogenous expression of ClC-3 protein increases, likely strengthening the ability of neuroendocrine cells to maintain volume constancy and avoid apoptosis [106]. ClC-3 likely participates in the generation of chloride currents as a response to cell swelling in LNCaP cells [16,106]. In general, intracellular calcium entering via SOCs exert inhibitory effects on VRACs [107]. The tumor transition to androgen independence and apoptosis resistance is marked by a weakened ability of SOCs to inhibit VRACs and results in a faster response to lowered tonicity. This is likely due to a decrease in the number of functional SOCs as an adaptive response to long-term decrease in endoplasmic reticulum Ca 2+ content [16].
Moreover, anoctamin 1 (ANO1), a calcium-regulated chloride channel, is highly upregulated in PCa, while its inhibition is associated with decreased cell proliferation, migration and invasion [108]. Seo et al. reported that luteolin, a potent ANO1 inhibitor, inhibited cell proliferation and migration in ANO1-expressing PC3 cells, but was less active against ANO1-deficient PC3 cells [108]. Tian et al. also suggested that chloride intracellular channel 1 (CLIC-1) regulates cell proliferation and migration in PCa, through a mitogen-activated protein kinase (MAPK)/ERKdependent pathway [109]. Knockdown of CLIC-1 negatively impacted cellular proliferation and migration in PC3 and DU145 cell lines, but showed no effect on apoptosis in vitro [109]. At the same time, the levels of p-ERK1/2, MMP-2, and MMP-9 were significantly decreased in the CLIC-1 knockout group [109].

P-Class Pumps
P-class pumps move ions against their concentration gradient. Their function is achieved by dephosphorylating ATP for each cycle of the pump [110]. Apart from ion transport, they are involved in gene transcription, cell proliferation, and cell migration in eukaryotic cells. Among P-class pumps, the sodium (Na + )/potassium (K + )-ATPases (NKA), proton (H + )/K + ATPases (HKA), and the sarco-endoplasmic reticulum calcium (Ca 2+ ) ATPases (SERCAs) have been studied extensively.
The NKA transmembrane protein is subdivided into alpha, beta and FXYD subunits [110]. The expression of alpha 1 subunit decreases in PCa and metastatic cell lines. Banerjee et al. reported evidence suggesting that this results in increased Src activity, promoting cell proliferation [111].
Scientists have also suggested that it triggers a metabolic switch from mitochondrial oxidative phosphorylation to aerobic glycolysis (Warburg effect) [111]. Subsequent experiments showed that reduced alpha 1 NKA expression and the resulting Src/FAK pathway activation enhances PCa metastatic potential through decreased E-cadherin and increased c-Myc expression [112]. On the other hand, the FXYD subunit is overexpressed in PCa compared with normal prostatic tissue [110]. Grzmil et al. showed that siRNA-mediated FXYD3 inhibition resulted in reduced cell proliferation in PC3 and LNCaP cells, while cellular apoptosis and invasive capacity remained unaffected [113].
The expression of HKA ATP12A appears to be similar in PCa and its normal counterparts. However, Streif et al. found that in normal prostate tissue, the immunostaining of ATP12A is membrane-bound with focal accumulated pattern, whereas in cancer it appears to be displaced in the luminal cells of the glandular epithelium [114]. ATP12A mRNA expression levels in LNCaP cells were 26 times higher compared with PC3 cells, while Western Blot confirmed protein expression in these PCa cell lines [114]. Whitton et al. showed that inhibition of vacuolar-ATPase, a multi-protein proton transporter, results in reduced function of AR and AR variants [115]. Vacuolar-ATPase inhibition in PCa (with bafilomycin A and concanamycin A) was also linked to impaired endo-lysosomal pH, vesicle trafficking, migration, and invasion [116].
The SERCA transports two calcium ions per ATP from the sarcoplasm to the sarco/endoplasmic reticulum lumen [110]. SERCA inhibition can result in impacted transfer of Ca 2+ in the mitochondria, which in turn might lead to the electrochemical gradient collapse, and subsequently to activation of multiple cell death-related signaling pathways [110]. In LNCaP cells, the induction of cell proliferation as a result of androgens and other physiological stimuli is controlled by SERCA [117]. Of note, thapsigargin interferes with the ability of the pump to bind calcium and produces complete growth inhibition in PSA-producing prostate tumor xenografts [117].

Discussion and Perspectives
It is evident that ion channels directly or indirectly participate in several (if not all) cellular functions. Ion transport has a central role not only in physiological processes but also in several cellular pathways that are associated with PCa tumorigenesis, apoptosis evasion, proliferation, migration, and metastasis formation. This complexity is the main reason why ion channel modulators as single agents have not replaced existing treatments in PCa. However, current therapies are not curative, and resistance mechanisms are associated with complex molecular networks in which ion channels are critically involved directly or indirectly.
Ion channel modulators can potentially complement other treatments in the future or comprise parts of therapeutic combinations. For example, the possibility that anesthetics may influence the risk or recurrence of PCa could have a wide-ranging impact on clinical practice and patient outcome. Evidence from experimental and clinical studies suggests an association between anesthetic drugs and overall survival through several mechanisms, including modulation of the innate and adaptive immune system, inflammatory system, and direct effects on ion channel signaling [118][119][120][121][122]. Interestingly, the duration of local anesthetics extends beyond the intraoperative period and they continue to inhibit the function and activity of VGSCs after anesthesia and surgery, which may potentially reduce cell proliferation and increase patient survival [51,123,124]. Moreover, in a study with SW620 metastatic colon cancer cell line and reverse transcription-polymerase chain reaction and sequencing to identify Nav1.5 variants, ropivacaine was a potent inhibitor of both Nav1.5 channel activity and metastatic colon cancer cell invasion [125]. Ropivacaine may also have a role in the inhibition of the metastatic ability of PCa cells by enhancing their secretory membrane activity and altering cellular homeostasis and intracellular ionic concentration [126][127][128].
It has also been suggested that lidocaine, an amide local anesthetic and non-specific sodium channel blocker, inhibits chemokine-induced tumor cell migration via the direct inhibition of CXCR4 activity, thus affecting PCa tumorigenesis, progression and prognosis, and blocks the expression of the Nav1.5 [9]. Lidocaine can also reduce the expression of TRPV6 and TRPM7, which are involved in Ca 2+ and Mg 2+ steady-state ion channels, and thus can potentially influence the survival rate, migration and cell division [129][130][131]. Bupivacaine, another Na + blocker, was shown by Xuan et al. to reduce viability and inhibit cellular proliferation and migration of PCa when used in clinically relevant concentrations in vitro [129].
The identification of specific benzodiazepine binding sites on the outer mitochondrial membranes of the prostatic cells could impact survival of patients with PCa. Benzodiazepines, such as midazolam, can bind to benzodiazepine receptors, which may be coupled to chloride channels, and affect the regulation of cell proliferation and tumorigenesis [132][133][134][135]. Furthermore, the γ-aminobutyric acid (GABA) receptor agonists have been reported to bind in the m-and p-fractions of prostate tumor cells. The usually high expression of these receptors in anaplastic prostate tumors suggest a potential involvement in pro-and tumorigenic activities [132]. Of note, GABA receptors have a central role in modern perioperative medicine, as many intravenous and inhalational GABA agonists are used in anesthesia and intensive care, including propofol, etomidate, thiopental, isoflurane, sevoflurane, and desflurane.
The most important aspect of current research is to implement ion channel modulators in in vitro or in vivo models, in order to further delineate the underlying tumorigenic molecular pathways and the dependencies between ion channels and other drivers in PCa development, progression, and resistance to treatment. This research is still in its infancy but has yielded useful insights to date, although more experimentation is needed before these findings translate into clinical applications. Furthermore, the association between ion channel expression/activity with disease grade and stage holds promise regarding the future use of ion channel expression levels as useful markers for diagnosis and staging of PCa, or as components of future diagnostic and staging algorithms.

Conclusion
Increasing evidence suggests that ion channels and pumps play a critical role in the regulation of PCa cell proliferation, apoptosis evasion, invasion, and migration. They represent promising targets for diagnosis, staging and treatment of PCa. However, more research is needed before these findings translate into clinical applications.

Declarations
Funding No funding was received for this work.

Conflict of interest
Minas Sakellakis and Athanasios Chalkias report no financial or non-financial competing interests that are directly or indirectly related to this manuscript.
Ethics approval Not applicable.

Consent (participate and publication) Not applicable.
Author contributions MS conceptualized the review, reviewed the literature, and drafted and critically reviewed the manuscript. AC reviewed the literature, and drafted and critically reviewed the manuscript. Both authors approved the final version of the manuscript.
Data availability statement Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.
Code availability Not applicable.