Voltage-gated ion channels in cancer cell proliferation
The pore-forming CaV-α1 subunit of high voltage-activated CaV channels is associated with auxiliary subunits, while low voltage-activated CaV channels do not co-assemble with auxiliary subunits [100]. Auxiliary CaV subunits include the groups of CaV-β subunits (β1–β4), CaV-α2δ subunits (α2δ1− α2δ4), and CaV-γ subunits (γ1-γ4) and contribute to the functional diversity of the high voltage-activated CaV channels [100].
The CACNA2D2 gene encoding the CaV-α2δ2 subunit is located on chromosome 3p21, a well-defined tumor suppressor locus [101]. Early after the genetic identification and functional analysis of the CaV-α2δ2 subunit [102], single nucleotide polymorphisms associated with small cell lung cancer were detected in the CACNA2D2 gene [103]. Similarly, there was an observed 50% loss of expression in cancer cell lines [104].
Subsequent experiments utilizing adenoviral vector-mediated gene transfer to overexpress the CaV-α2δ2 subunit in non-small cell lung cancer revealed a significant inhibition of tumor growth [105], mediated by increased apoptosis and the preceding disruption of mitochondrial membrane integrity [105].
In contrast, more recent studies on prostate cancer have shown that overexpression of the CaV-α2δ2 subunit leads to increases in the rate of cell proliferation in vitro, and that xenografts of LNCaP cells overexpressing the CaV-α2δ2 subunit into nude mice are more tumorigenic, showing evidence of increased cellular proliferation and angiogenesis [106]. In support of the direct mechanistic role of the CaV-α2δ2 subunit in their model, the authors were able to show efficacy of the selective CaV-α2δ2 subunit blocker, gabapentin, in reducing tumors in the xenografts [106].
Similarly, the related CACNA2D3 locus, encoding the CaV-α2δ3 subunit, has been identified as a frequent site of methylation in gastric cancer [107]. Furthermore, methylation of CACNA2D3, but not CACNA2D2 was associated with significantly shorter survival [107], making it a useful prognostic marker for patients with gastric cancer. The tumor repressive properties of the CaV-α2δ3 subunit were also observed in in vitro models, where overexpression resulted in reduced cell growth and adhesion, while down-regulation using small interference RNAs had the inverse effect [107]. Methylation-dependent silencing of CACNA2D3 was proposed as a biomarker for the risk of development of metastatic disease in breast cancer, where methylation correlated strongly with visceral and metastatic disease independent of α-estrogen receptor expression [108].
More recently, similar tumor suppressor function of CACNA2D3 was identified for the development and progression of nasopharyngeal carcinoma [109] and esophageal squamous cell carcinoma [110].
While there is clear evidence that supports an involvement of CaV channels and auxiliary CaV subunits in the proliferation of cancer cells, several questions remain regarding the underlying mechanism. A new hypothesis involves the formation of complexes comprised of Ca++ activated potassium (KCa) and CaV channels for review, see [86]. For example, large-conductance BKCa channels expressed in prostate cancer LNCaP cells set the resting membrane potential at approximately −40 mV, thereby promoting constitutive Ca++ entry through CaV3.2 channels [111]. Furthermore, both channels form macromolecular complexes, where CaV3 channels are located in clusters of BK channels [111]. This co-localization thus provides for an effective feed-forward loop of constitutive Ca++ entry, which in turn regulates the membrane potential by activating BKCa channels [86].
Further studies are needed to investigate the biophysical properties of CaV channels, particularly their association with other ion channels in macromolecular complexes as well as their association with auxiliary subunits.
Voltage-gated Na+ channels selectively allow the passage of Na+ ions into the cells resulting in membrane depolarization and play a pivotal role in initiation and conduction of action potential in excitable cells including neurons, heart and skeletal tissue [112]. They are distinguished into nine families identified thus far (Nav1.1–1.9) based on the alpha subunit which is the essential pore forming unit that facilitates Na+ transport across the membrane. The alpha subunits are found to associate with one or more beta subunits (ᵝ1− ᵝ4). While the alpha subunits are functional by themselves, the beta subunits can regulate the expression and gating of Nav channels. In recent years, the increase in expression of Nav channels in a broad number of cancers has been extensively reported [113,114]. Cancers seem to express variable and multiple Nav channels, with certain types expressing a predominant alpha subunit. For instance, Nav1.5 is most abundantly expressed in breast and colon cancer cell lines [115] while Nav1.6 in cervical cancer and Nav1.7 in prostate cancer [113,114]. Furthermore, some cancers express alternatively spliced more active, neonatal isoforms, such as the neonatal Nav1.5, Nav1.7 and Nav1.9 [116]. Nav expression and activity is greatly enhanced in strongly metastatic cell lines such as the MDA-MB-231 breast cancer cell line in comparison to weakly metastatic MCF 7 cell lines [117]. Similar observation is consistent in prostate cancer cell lines, wherein strongly metastatic PC3 cells demonstrate significantly higher expression of Nav1.9 alpha subunit in comparison to the weakly metastatic LNCap cell line [118,119]. These observations have been corroborated in vivo and as well as in human cancer biopsies [120].
However, few studies suggest that inhibiting Nav channel activity reduces cell proliferation while the vast majority of the data appear to associate Nav channel activity to metastasis, invasion, galvanotaxis, endocytosis and secretion, as these effects are effectively inhibited by tetrotodoxin (TTX) a selective Nav channel inhibitor and several other Nav channel blockers while having no effect on cell proliferation [121,122,123,124]. Given the direct correlation of Nav channel expression to the metastatic potential of tumors and the ability of of Nav inhibitors to block metastasis, Nav channels have emerged as novel biomarkers and as well as therapeutic targets for certain cancers.
Chloride ion channels are a diverse group of membrane proteins that facilitate the transport of chloride ions across cell membrane and intracellular organelles. They play a pivotal role in a number of physiologically important functions ranging from trans-epithelial secretion, pH regulation, cell volume regulation, proliferation and membrane potential stabilization [125]. Depending on their mode of activation, Cl− channels are classified as voltage dependent (CLC family), cAMP regulated (CFTR), or calcium dependent chloride ion channels. The transport of chloride is also regulated by volume-sensitive organic osmolyte/Anion channels (VSOAC), which regulate the flow of organic solutes and anions in response to changes in cell volume. The voltage dependent CLC family of ion channels was initially identified in torpedo membranes (CLC-0) followed by nine different CLC proteins with similar sequence homologies being identified thus far in mammalian systems including CLC 1-7, CLCka, and CLCkb. In recent years, CLC family of Cl− channels have been implicated in a number of cancers. Gliomas, in particular, seem to express higher levels and activities of CLC 2 and CLC 3 with enhanced ability to efflux Cl−, thus contributing to depolarized membranes [126]. Cl− channel blockers or siRNA mediated knockdown of CLC 3 have been shown to reduce Cl− currents and inhibit proliferation of gliomas. The inhibitory effect following CLC 3 block on glioma proliferation has been mainly attributed to its regulatory effect on cell volume control during the cell cycle [127,128].
Cell proliferation comprises an accurate control of cell volume that is achieved by the fine government of osmolarity. Therefore, changes in intra/extra-cellular ionic concentration upon activation of ion channels can directly alter cell volume and, therefore, provide a contribution to cell proliferation process. Water molecules follow the flux of ions through ion channels creating changes in cell volume, cytoplasm, and condensation/shrinkage described as pre-mitotic condensation (PMC) prior to mitosis [73,129]. Impairment in PMC can result in delay or potentiation of the cell cycle. PMC or cell shrinkage is primarily attributed to the loss of fluids associated with Cl−/K+ ion efflux. It has been observed that Cl− conductance is increased in the M phase, CLC 3 channel is enriched in the PMC phase, furthermore inhibiting Cl− channel activity or knockdown of CLC 3 results in lengthened cell cycles. Similarly, CLC 3 expression, activity and its possible role in cell volume control has been implicated in a number of cancers including nasopharyngeal carcinoma, endometrial and ovarian cancer cells [130,131,132]. While the role of voltage-gated CiLC 3 in cell volume regulation is still under debate, nevertheless, the enhanced expression of CLC family proteins in certain cancers and alleviation of the progression of these cancers by inhibiting Cl− currents, offers a potential role for Cl− ion channels in cancer and merits further investigation.
Given the enhanced expression and activity of various voltage-gated ion channels in cancer and their role in key events contributing to initiation and progression of the disease, VGICs have emerged as a novel therapeutic target for treating cancers. A number of preclinical in vitro and in vivo studies targeting VIGC have led to valuable insights on the usefulness and limitations of existing approaches, while fostering the need for further strategies to effectively target ion channels such that, its role in cancer can be harnessed to its complete potential. Several small molecules targeting VGICs that are/were clinically relevant for other non-cancer conditions incidentally, which also are VGIC channel blockers, have been investigated in preclinical cancer studies for their therapeutic potential. For example, imipramine, an antidepressant known to block Kv10.1, inhibits proliferation of cancer cells including ovarian, osteosarcoma and small cell lung carcinoma by apoptosis. However, in melanoma cell lines, it inhibits proliferation without affecting apoptosis [133,134,135,136,137,138]. Astemizole, an anti-histamine, is able to inhibit both Kv10.1 and Kv11.1 channels, contributing to inhibition of cancer cell proliferation in vitro and in vivo [139,140,141]. Several antiarrhythmic agents of the class III antiarrhythmics are known to block Kv11.1 channels. Anti-arrhythmic agents Way 123,398 and E4031 are selective blockers of Kv11.1 have been shown to impart therapeutic benefit in cancer as observed in various in vitro and pre-clinical models of cancer [135,142,143]. A major limitation that prevents the use of the Kv10.1 and Kv11.1 blockers in cancer is their adverse effects on cardiac physiology, resulting in prolonged QT intervals, which may further lead to fatal ventricular arrhythmias [144].
Thus a number of diligent studies are focused on understanding the structure and gating of Kv11.1 channels with the objective of identifying molecules that can circumvent the adverse effect of prolonged QT intervals while still retaining the Kv11.1 channel inhibition. For example, roscovitine, a cyclin dependent kinase inhibitor currently in phase II trials for treating cancers, is able to inhibit kv11.1 in clinically relevant doses [145,146,147] and yet, no adverse effects of arrhythmia have been reported thus far. The lack of arrhythmia by roscovitine has been attributed to its specific, rapid reversible binding kinetics to only open channel conformation of Kv11.1 and a lack of in use dependence commonly observed with potent Kv11.1 channel blockers. Similarly, verapamil, a voltage dependent Ca++ blocker which also inhibits kv11.1, fails to elicit a prolonged QT interval in subjects, probably due to its masking effect of Kv11.1 inhibition with that of voltage− dependent Ca++ channel inhibition and lack of action potential dependence [148,149]. Furthermore, peptide toxins, obtained from various animal venoms including, snakes, scorpions, sea anemones are potent Kv channel inhibitors, have served as a critical pharmacological tools in understanding the mechanisms leading to Kv channel inhibition and identifying toxins and designing small molecules selective for a specific Kv channel [150]. Unlike small molecules, which bind to the inner channel pore, Kv channel peptide toxins bind to the outer channel pore and occlude it or act as gating modifiers by binding to the voltage sensor domain of the Kv channel, leading to enhanced stability of the closed state and resulting in a channel that is more difficult to open.
Studies focused on peptide toxins have led to identification and designing of peptides derived from toxins with enhanced specificity for a particular Kv channel. For instance, anemone toxin APETx1 is a gating modifier with high specificity for Kv11.1 [151,152]. Similarly, a modified sea anemone toxin-ShK-186/SL5 specifically targets Kv1.3 expressed in auto-reactive memory/effector T cells implicated in autoimmune disease and ameliorates the disease symptoms in animal models [153]. While upregulation CaV1.3 has been observed in prostate cancer and its pharmacological blockade inhibits growth in prostate cancer cell lines, the CaV1 blocker, nifedipine, however, has been implicated in the proliferation of certain cancers and its effects are likely not mediated by blocking CaV channels [88,91]. One recent study demonstrated that while nifedipine promoted both the proliferation and migration of breast cancer in vitro and in vivo [91], verapamil did not affect proliferation [91]. The Cav3 inhibitors including mibefradil are able to inhibit cancer cell proliferation [94,96].
Several FDA–approved VGSC blockers that are currently used as anticonvulsants, antiarrhythmic, or tricyclic antidepressants [154,155] could be potentially repurposed in cancer therapy. For example, phenytoin, an anti-convulsing drug, effectively blocks VGSC and does not only inhibit the migration and invasion of metastatic MDA-MB-231 cells in vitro but also in clinically relevant doses is able to significantly reduce tumor growth, cancer cell proliferation in vivo, invasion and metastasis [115,156,157].
Novel hydroxyamide compounds developed using rational drug designing approaches, based on the binding site of phenytoin to VGSCs, have demonstrated significant inhibition of voltage-gated sodium channel activities and increased inhibition of androgen–independent prostate cancer cell growth in vitro in comparison to phenytoin and significant ability to reduce prostate cancer in vivo as well [124,158]. Moreover, it has been observed that the beneficial effects of EPA in inhibiting prostate cancer cell proliferation and metastasis seems to be mediated by its inhibition of voltage-gated sodium channels.
Lack of specific inhibitors has been a major limitation in targeting voltage-gated chloride channels in cancer therapy. Several known anticancer molecules including tamoxifen, 5-nitro-2-(3-phenylpropyl-amino) benzoic acid (NPPB) that are able to inhibit chloride channel currents [131,159] also indeed inhibit cancer cell proliferation.
Chlorotoxin, a 36 amino acid peptide derived from scorpion venom, is able to inhibit glioma progression [160]. Cholortoxin and its derivatives are known to bind specifically to CLC 3 channels and inhibit the channel activity, which in part explains its inhibitory effect on gliomas, as CLC 3 is known to play a critical role in glioma proliferation and progression. However, cholorotoxin, also binds to matrix metalloproteinase-2 (MMP2) [161] resulting in reduced membrane-associated MMP2 activity and therefore inhibiting migration of glioma cells. Incidentally, CLC 3 and MMP2 are both over and co-expressed specifically in gliomas rather than in the surrounding normal tissue. Thus, chlorotoxin has emerged as not only a therapeutic agent but also a diagnostic agent for glioma therapy. TM-601, an I131 synthetic derivative of chlorotoxin, has been approved by the FDA for glioma therapy and diagnostics [162,163].
A more selective targeting of a specific channel has been achieved by generating antibodies against it. Antibodies serve as an extremely specific molecule for inhibiting a peptide/protein target. Selective monoclonal antibodies have been generated that are highly specific for Kv10.1 and that are not able to cross the blood brain barrier. These antibodies are known to impart moderate anti-tumor effects and can also be used as diagnostic or drug delivery agents to deliver cytotoxic drugs selectively to the tumor expressing Kv10.1 channels [36].
Selective inhibition of ion channels have also been possible with the use of molecular tools such siRNAs. For instance selective knockdown of Kv10.1 with siRNA reduces Kv10.1 expression and cell proliferation in several cancer cell lines [39]. Similarly, plasmid mediated expression of shRNA-herg1 and shRNA-herg1/1b inhibited the cellular growth rate, viability and colony formation of SH-SY5Y neurobalstoma cells [164]. Further, in the same study, it was determined that cells intra-tumor injection of shRNA-hERG1/1b inhibited the growth of SH-SY5Y tumors inoculated subcutaneously in nude mice. Similarly, CLC 3 specific siRNAs are able to inhibit CLC 3 mediated proliferation in cancer cell lines [165,166]. The voltage gated channel blockers contribute inhibition in cell proliferation either due to cell cycle arrest or apoptosis via caspase 3 activation. The cell cycle arrest is observed either at the G0/G1 phase or the G2/M phase through the accumulation of tumor suppressors p27kip1 and p21CIP1 and down regulation of cyclins or cyclin dependents kinases critical for cell cycle progression [167]. The G2/M arrest has been attributed to the regulatory effect of VGIC in cell volume changes during cell cycle progression [168].
More recently, it has been observed that ion channel activators have also proven beneficial in inhibiting cancer cell proliferation. For example, we have observed that pharmacologic stimulation of Kv11.1 with the selective activator NS1643 in various breast cancer cells results in marked decrease in cell proliferation by activation of a senescent-like cellular program that is characterized by a cell cycle arrest at G1 phase, decreased levels of tumor markers such as cyclin E2 and increased level of tumor suppressors such as p21Waf1/Cip1 and p16INK4A [48,169]. This dramatic inhibitory effect of NS1643 is mediated by Ca++ driven signaling mechanism that contributes to selective degradation of cyclin E2 [48] and enhanced expression of p21Waf1/Cip1 [170], Moreover, Kv11.1 openers have shown to exhibit less cardiovascular adverse effects compared to Kv11.1 blockers. Similarly, it has been observed that bufalin, a bufadienolide isolated from toad venom, is a novel CLC-3 activator that has anti-tumor activity and contributes to cancer cell apoptosis through inhibition of the PI3K/Akt/mTOR pathway [171]. This suggests that stimulation of VGIC could be a valid alternative or additive therapeutic approach to treat cancer. We have summarized the potential therapeutically approaches for individual VGIC in Table 1.
Voltage-gated ion channel activity is an important event that controls several cellular functions including generation, maintenance of action potentials, and secretion. The role of VGICs in cancers is being extensively investigated and has led to valuable insight into mechanisms by which VIGCs contribute to tumorogenesis. In light of current literature and the availability of well characterized VGIC inhibitors/activators already in market, a compelling opportunity for further advancing the studies that characterize the role of VIGC in cancer and targeting them as therapeutics for cancer has been provided.
Vidhya R. Rao has contributed to the collection of literature and writing of the manuscript.
Mathew Perez-Neut has contributed to the collection of literature and writing of the manuscript.
Simon Kaja has contributed to the collection of literature and writing of the manuscript.
Saverio Gentile is the principal investigator and corresponding author who was responsible for the layout, design and organization with critical inputs.