Ca2+ channel-forming ORAI proteins: cancer foes or cancer allies?
Summary. The ORAI family of ion channel-forming proteins in mammals includes three members, ORAI1, ORAI2 and ORAI3, encoded by homologous genes. Of these proteins the ORAI1 one received major attention as plasma membrane constituent of store-operated calcium entry (SOCE) in non-excitable cells. The functional significance of two other proteins, ORAI2 and ORAI3, is much less defined, although both of them participate to various extends in cell-specific modulation of SOCE as well as in supporting some of the store-independent calcium entry mechanisms. Calcium signaling becomes remodeled in cancer to promote cancer hallmarks — enhanced proliferation, resistance to apoptosis, motility and metastasizing. Although such remodeling commonly involves rearrangements of the whole molecular Ca2+-handling toolkit of the cell (Ca2+ pumps and transporters, Ca2+-binding and storage proteins, Ca2+ entry and release channels, Ca2+-dependent effectors), Ca2+ entry through Orai-based channels is especially important, as its dysregulation may contribute to several cancer hallmarks. The latter depend on the type of Ca2+-permeable channel formed by ORAI-proteins, spatiotemporal characteristics of Ca2+ signal that this channel contributes to, and the type Ca2+-dependent effector(s) targeted by this signal, all of which may be cancer-specific. By participating in global Ca2+ entry, ORAI-based SOCE may also contribute to cytosolic Ca2+ overload of cancer cells thereby playing pro-apoptotic, antineoplastic roles which can potentially be exploited for cancer treatment. This mini review examines various aspects of ORAI proteins in malignant transformation.
Submitted: September 3, 2019.
*Correspondence: E-mail: firstname.lastname@example.org
Abbreviations used: AA — arachidonic acid; ARC — arachidonate-regulated channel; CRAC — calcium release-activated channel; ER — endoplasmic reticulum; FAK — focal adhesion kinase; IP3 — inositol trisphosphate; LRC — LTC4-regulated channel; LTC4 — leukotriene C4; NSCLC — non-small cell lung cancer; PM — plasma membrane; SICE — store-independent calcium entry; SOAR — STIM1-ORAI1 activating region; SOC — store-operated channels; SOCE — store-operated Ca2+ entry; SPCA2 — secretory pathway Ca2+-ATPase; STIM — stromal interaction molecule.
Due to its unique physical and chemical properties, the nature has selected calcium ion as universal second messenger to regulate vast array of cellular physiological processes. Disruption of Ca2+-mediated signaling can lead to diseases. Calcium signaling events in the cell take place on the background of its very low basal cytosolic concentration (i.e. below 10–7 M), which makes even small rises of cytosolic Ca2+ concentration ([Ca2+]c) being discernible by high-affinity Ca2+-sensing effectors (kinases, proteases, phosphatases, transcription factors, Ca2+-binding proteins). In order for [Ca2+]c to rise at needed place, needed time and for needed duration to perform required action it must come from cellular compartments where its concentration is much higher and then be effectively removed back to those compartments. There are three cellular compartments which can serve both the source for cytosolic Ca2+ and the sink for its removal: extracellular space, intracellular Ca2+ stores of the endoplasmic reticulum (ER), and mitochondria. To perform Ca2+ cycling among those compartments and the cytosol the cells developed the whole molecular Ca2+-handling toolkit consisting of active energy-dependent Ca2+ transporters, Ca2+-permeable ion channels, Ca2+-binding and storage proteins. Molecular players constituting this toolkit are cell type-specific. This is especially true for Ca2+-selective plasma membrane (PM) ion channels which provide for Ca2+ entry pathways from extracellular space. Based exclusively on the primary mechanism that gates these channels (i.e. makes them open or closed) they can be roughly subdivided onto six major classes: 1) voltage-gated Ca2+ channels , 2) ligand-gated channels , 3) store-operated channels (SOC) , 4) second messenger-operated channels , 5) acid-sensing ion channels , and 6) mechano-gated channels . Voltage-gated Ca2+ channels and ligand-gated channels are mostly typical for excitable cells such as neurons and various types of muscle, whereas others have more widespread distribution.
In non-excitable cells, including those of epithelial origin from which major carcinomas develop, the main pathway for the influx of Ca2+ from extracellular space is represented by SOC-channels generating the so called store-operated Ca2+ entry (SOCE) which can be measured electrophysiologically as membrane current (ISOC) or using Ca2+-sensitive fluorescent dyes as rise in cytosolic Ca2+ concentration [3, 7, 8]. SOC-channels open only when Ca2+ content of the ER-store becomes essentially depleted. In such a way the SOCE that they provide serves the major purpose of the ER Ca2+-store refilling, whereas global and local [Ca2+]c increases associated with SOCE may participate in calcium signaling events. The common stimulus for the ER Ca2+-store depletion and concomitant activation of SOCE is agonist-mediated stimulation of the PM G-protein-coupled receptors linked to phospholipase C-catalyzed derivation of inositol trisphosphate (IP3), the known Ca2+-releasing messenger from the ER via IP3-receptors localized in the ER membrane [3, 7, 9].
Biophysical properties of SOCE in different cell types are not the same which is explained by cell-specific molecular composition of SOCs and the type of interactions that molecular components of SOCs may get involved in with other molecules . Historically first functional phenotype of SOC-channels featuring very high Ca2+ selectivity was described in the cells of immune system, and is known under the original name of calcium release-activated channel (CRAC) . Because of this in the context of our presentation we will use more general terms SOC, SOCE, SOC-mediated membrane current (ISOC) rather than CRAC and CRAC-mediated current (ICRAC) which we reserve only for the cells of immune system.
Background on ORAI and STIM proteins
On molecular level SOC, as a whole, is not just a single molecular entity, but rather multiprotein complex which includes PM and ER-membrane components. The PM component is represented by the members of four transmembrane domain ORAI family of channel-forming proteins that complex into a hexamer to produce transmembrane Ca2+-permeable pore [3, 11, 12]. The ER membrane component is represented by the single membrane-span proteins of stromal interaction molecule (STIM) family, acting as ER luminal Ca2+ sensors capable of gating ORAI-based PM channels depending on ER intraluminal Ca2+ concentration ([Ca2+]ER) [3, 12, 13].
In mammals, there are three isoforms of ORAI proteins, ORAI1 ORAI2 and ORAI3 (Figure), and two isoforms of STIM proteins, STIM1 and STIM2, encoded be respective families of homologous genes . In addition, mammalian ORAI1 may exist in two variants, longer ORAI1α and shorter ORAI1β, due to alternative mRNA translation initiation . Classically, SOC channels, including the canonical CRAC, involve ORAI1 and STIM1 interactions (with little difference between ORAI1α and ORAI1β), whereas physiological roles of ORAI2, ORAI3 and STIM2 are not so obvious and are still poorly defined. STIM1 N-terminal region facing ER lumen contains low-affinity EF-hand Ca2+-binding domain acting as Ca2+ sensor followed by a “hidden” EF‐hand motif and a sterile α‐motif important for protein stability and olimerization. Under resting conditions STIM1 is more or less evenly distributed within the ER membrane in the form of monomers and dimers. However, upon ER Ca2+-store depletion STIM1 oligomerizes and rapidly translocates into ER membrane and PM junctions where it interacts with ORAI1 through its cytosolic C-terminal STIM1-ORAI1 activating region (SOAR) domain causing ORAI1 activation [3, 12, 14].
Figure. Schematic depiction of ORAI proteins involvement in oncogenesis. Following the arrows enables functional channel types, Ca2+ entry pathways, their impact on Ca2+ homeostasis or signaling, and oncogenic effects linked to specific ORAI protein (underlined font) in indicated cancers (italic font) to be traced. Change of color shade from light to dark indicates up-regulation of respective channel or process and vise versa. Blending colors indicate multimerization. Question marks near cancer types indicate uncertain or insufficient information. Red dashed line encompasses pathway whose reversal by overactivation of ORAI1-based SOCE may be useful for cancer treatment
Heterologous co-expression of either ORAI2 or ORAI3 with STIM1 in HEK293 cells is still able to induce ISOC, but with much lesser efficiency compared to ORAI1 [11, 16]. Besides, the biophysical properties of ISOC resulting from heterologous ORAI2 or ORAI3 expression, including Ca2+ permeation, are different from ORAI1-based one [11, 16]. Finally, ORAI2 or ORAI3 knockdown in native cells does not eliminate ISOC, whilst knockdown of ORAI1 does [11, 16], suggesting that ORAI2 and ORAI3 are not essential SOC components, but rather play modulatory roles or form other types of channels, whose activation may indirectly depend on store depletion. So far an exception in this regard was demonstrated for estrogen receptor-positive (ER+) breast cancer cell lines (including the popular in cancer research MCF7 one) for which the native SOCE pathway and ISOC were shown to uniquely depend on ORAI3 . The data suggest that ORAI3 is likely to be the major SOC component also in some of the non-small cell lung cancer (NSCLC) cell lines  (see Figure). Besides, in mouse T cells  and acute myeloid leukemia cell line HL60  PM component of SOCE is likely represented by heteromeric assembly of ORAI1 and ORAI2 (see Figure). In mouse T cells ORAI2 knockdown increased, whereas ORAI1 knockdown decreased ISOC, suggesting dominant negative role of ORAI2 in heteromeric channel .
Except being involved in cell type-specific SOCE, ORAI-proteins may also form Ca2+-selective channels supporting some of the store-independent calcium entry (SICE) mechanisms. For instance, in primary breast cancer cells and MCF7 cell line ORAI1-based PM channels can be constitutively activated in a store- and STIM1/2-independent manner via protein-protein interaction with Golgi apparatus secretory pathway Ca2+-ATPase (SPCA2) . SPCA2-dependent, ORAI1-mediated SICE was implicated in supporting milk secretion during lactation  as well as in breast carcinogenesis  (see Figure). Another examples of SICE are represented by the arachidonate-regulated (ARC)  and leukotriene C4 (LTC4)-regulated (LRC)  channels which by virtue of their activation by the intracellular lipid messengers, arachidonic acid (AA) or it metabolite LTC4, can be classified as second messenger-operated ones. Detailed comparison of biophysical properties of ARC and LRC prompted to conclude that they, in fact, are represented by the same molecular entity involving heteromeric ORAI1/ORAI3 assembly (see Figure) gated by STIM1 in a way dependent on either AA or its metabolite LTC4 .
Extensive studies of ARC/LRC-mediated currents in native cells and expression systems revealed that PM component of the respective channel is represented by the heteromeric assembly of ORAI1 and ORAI3 proteins [24, 26]. Despite supporting SICE and not SOCE, activation of ARC/LRC was shown to depend on STIM1 as well [24, 26], although the type and manner of interactions involved still remains uncertain. It was initially proposed that the minor pool of PM-localized STIM1 protein which represents about 10–15% of the total cellular STIM1 is required for ARC activation . Subsequent studies, however, favored the idea that it is still ER-localized STIM1 that with its SOAR domain enters into constitutive interaction with ARC/LRC ORAI3 component to make channel functional and that out of two ORAI1 variants it is the ORAI1α one that complexes with ORAI3 [25, 26, 28].
As members of the same family, STIM1 and STIM2 share high degree of amino acid sequence homology and domain composition [16, 29]. However, a number of structural peculiarities of STIM2 in C-terminal cytoplasmic region, SOAR domain, and EF‐hand motif and a sterile α-motif domain make it less aggregable and poorer ORAI1 activator compared to STIM1 [16, 29]. It is believed that STIM2 mainly functions as homeostatic regulator of [Ca2+]c and [Ca2+]ER in response to the small decreases of [Ca2+]ER with only a minor involvement in SOCE in some cell types [16, 29, 30]. Alternative splicing gives rise to three STIM2 isoforms, the classical one, STIM2.2 (or STIM2α), a shorter STIM2.3 resulting from alternative exon 13 lacking 444 bp, and a larger STIM2.1 (or STIM2β) that includes the new exon 9 . STIM2.2 was shown to promote SOCE, while STIM2.1 to inhibit it which is why the expression ratio between them might influence SOCE amplitude .
ORAI proteins as determinants of oncogenic calcium signaling
The list of common pathophysiological features of cancer known as cancer hallmarks [31, 32] exhibiting the highest degree of Ca2+ dependence includes: 1) enhanced proliferation, 2) resistance to apoptosis (i.e. programmed cell death), 3) migration, invasion and metastasis [33–35]. Of these, the 1st and the 3rd ones manifest strengthening of the processes associated with cancer cells life cycle, whereas the 2nd one indicates that cancer cells would also not easily die.
Intracellular Ca2+ signals regulating life-related processes such as proliferation, differentiation, migration, secretion commonly exhibit complex spatiotemporal characteristics to enable targeting of specific Ca2+-dependent effectors. These signals usually take form of global cytosolic Ca2+ oscillations, Ca2+ waves propagating through the cell or short-lived spatially confined increases of [Ca2+]c known as Ca2+ sparks, spikes, flickers [36–38]. The occurrence of such complex patterns of Ca2+ signal is supported by highly coordinated in space and time Ca2+ entry, Ca2+ release and Ca2+ uptake engaging specific components of cellular molecular Ca2+-handling toolkit. On the other hand, progression to apoptosis is linked to sustained cytosolic Ca2+ overload due to excessive calcium entry and/or release promoting mitochondrial calcium uptake, mitochondrial permeability transition and release of mitochondrial apoptogenic factors (i.e. cytochrome c and apoptosis-inducing factor) into the cytoplasm with upstream activation of death-executing caspase cascade [39, 40]. Depending on cancer type and particular context, altered Ca2+ influx due to oncogenic remodeling of ORAI proteins may contribute to both life-related and death-related processes of cancer cells.
In prostate cancer, transition to the advanced, androgen-independent stage characterized by enhanced apoptosis resistance was shown to involve remodeling of the whole Ca2+-handling toolkit enabling downplaying the importance of [Ca2+]c overload and [Ca2+]ER reduction in apoptosis induction [41–43]. The key feature of such remodeling was the decrease of SOCE in androgen-independent cells that essentially limited the amount of Ca2+ that can enter from extracellular space [42, 43]. Determination of molecular nature of SOCE in prostate cancer cells showed that it relies on ORAI1 and STIM1 proteins and that reduction of SOCE upon transition to androgen-independency is associated with decreased levels of ORAI1 due to likely regulation of ORAI1 gene expression by the functional androgen receptor  (see Figure). Thus, in the event of prostate cancer ORAI1-mediated SOCE plays positive role by maintaining cancer cells’ susceptibility to apoptosis.
In contrast to ORAI1, the expression of ORAI3 was shown to increase in prostate cancer compared to the normal tissue with the tendency of increasing even further upon transition to androgen-independency . ORAI3 is capable of complexing with ORAI1 to form heteromultimeric ARC channel  which at constant or even reduced pool of ORAI1 shifted the balance in Ca2+ entry pathways towards prevalence of AA-regulated, ORAI1/ORAI3-based SICE over the classical ORAI1 only-based SOCE which is needed to maintain susceptibility to apoptosis  (see Figure). Moreover, increased ARC-mediated Ca2+ influx promoted prostate cancer cells proliferation via the activation of a Ca2+/calcineurin-dependent transcription factor, nuclear factor of activated T-cells, followed by the stimulation of the expression of the key rate-limiting controller of G1/S phase transition, cyclin D1 . Thus, overexpression of ORAI3 per se in androgen-independent prostate cancer is sufficient to promote two cancer hallmarks at once, enhanced proliferation and resistance to apoptosis , prompting to classify this channel-forming protein as oncogenic one in prostate cancer . In support of this notion, analysis of public colorectal cancer datasets from The Cancer Genome Atlas and GSE39582 revealed increased ORAI3/ORAI1 expression ratio also in colorectal cancer associated with poor prognosis .
Oncogenic roles of ORAI1-based SOCE are not limited to the involvement in determining sensitivity to apoptosis via contribution to the global Ca2+ overload. When spatial organization of SOC is such that ORAI1 is capable of providing polarized SOCE within localized signaling complexes to specific Ca2+-dependent effectors any changes in its expression and/or activity may essentially impact also other malignant behaviors of cancer cells, including motility, invasion, proliferation and susceptibility to the extrinsic apoptosis. Such localized signaling complexes often include Ca2+-dependent K+-channels (KCa) whose activation by Ca2+ hyperpolarizes membrane potential (Vm) to amplify Ca2+ influx by increasing driving force for Ca2+ entry in a positive feedback manner. For instance, in breast and colon cancer cells feedback interaction of ORAI1 with KCa2.3 (also known as SK3) channel was shown to occur within PM glycolipoprotein microdomains called lipid rafts, and the resultant amplified SOCE promoted cells migration via activation of Ca2+-dependent protease calpain [47, 48]. Interestingly, highly oncogenic voltage-gated K+-channel (Kv), Kv10.1 (also known as EAG1, human gene KCNH1), whose expression is significantly elevated in up to 80% of human tumors , is mostly known as the one promoting cancer cells proliferation [35, 50]. However, in metastatic, ER– breast cancer cells (i.e. MDA-MB-231) oncogenic Kv10.1 action was shown to promote migration and not proliferation, and this action was specifically linked to Kv10.1-driven Vm hyperpolarization and concomitant increase of driving force for SOCE through ORAI1 . STIM1- and ORAI1-dependent SOCE is in general critical determinant of breast cancer cells migration and metastasis with SOCE being important for the enhanced turnover of focal adhesions via the mechanism involving activation of small GTPases, Ras and Rac .
In ER+ MCF7 breast cancer cells which are characterized by the presence of specific type of SPCA2-gated, ORAI1-mediated SICE , the latter was implicated in supporting cancer cells survival and proliferation by collagen which is known to be a crucial component of the tumor microenvironment in breast cancer . SPCA2 and ORAI1 interaction was shown to take place at the level of lipid raft microdomains, and also involved of Kv10.1 K+-channel whose activity was likely needed to provide for Vm hyperpolarization and increasing the driving-force for Ca2+ entry.
Consistent with the importance of structured in space and time [Ca2+]c signal for orchestrating malignant behaviors, cancer cells in which elevated STIM1/ORAI1-dependent SOCE resulted in stimulation of oscillatory-type Ca2+ signaling commonly showed enhanced migration and invasion. This was demonstrated for epidermal growth factor-stimulated migration and invasion of cervical cancer cells , glioblastoma multiforme invasion , melanoma invasion and metastasis , clear cell renal cell carcinoma migration . Such signaling typically engaged Ca2+-dependent protease, calpain, and Ca2+-regulated tyrosine kinase, Pyk2, controlling focal-adhesion dynamics, as well as facilitated invadopodia formation and proteolytic activity of individual invadopodia through induction of extracellular matrix-degrading metalloproteinases.
In terms of the effects of SOCE in general and ORAI1 protein in particular on cancer cells proliferation the data seem to be cancer type-specific and less consistent. In prostate cancer cells the effect of ORAI1 on proliferation was shown to be indirect involving the Ca2+/Annexin I/S100A11-dependent translocation of highly oncogenic, pro-proliferative, Ca2+-permeable TRPV6 channel to the PM consequent to the activation of STIM1/ORAI1-dependent SOCE . Same TRPV6 and ORAI1 interplay may also take place in controlling hepatoma cells proliferation . In clear cell renal cell carcinoma , NSCLC  and esophageal squamous cell carcinoma  all of which express elevated ORAI1 mRNA and protein levels, the resultant ORAI1-dependent Ca2+ signaling was implicated in both cell proliferation and migration.
Although elevated global [Ca2+]c due to ORAI1-mediated SOCE represents important factor in the intrinsic apoptosis associated with intracellular stress response, in the event of extrinsic apoptotic signaling through cell surface death receptors of the tumor necrosis factor superfamily elevated SOCE may have anti-apoptotic significance (see Figure). As was shown in T-leukemic cell lines, polarized ORAI1/STIM1-mediated SOCE within microdomains co-localized with tumor necrosis factor death receptors family member, FasR (also known as CD95, APO-1), is crucial for preventing formation of death-inducing signaling complex, and downstream transmission of the apoptotic signal .
As was already mentioned, ORAI3 and not ORAI1 is a part of non-canonical SOC in ER+ breast cancer cell lines  and some of the NSCLC cell lines  (see Figure). Consistent with this, ORAI3-dependent SOCE supported proliferation, apoptosis resistance and migration of breast cancer cells in an estrogen receptor-dependent manner via the mechanisms involving phosphorylation of extracellular signal-regulated kinase (ERK1/2), focal adhesion kinase (FAK) as well as stimulation of nuclear factor of activated T-cells transcriptional activity , whereas in NSCLC cell lines control of proliferation and cell-cycle progression via ORAI3-dependent SOCE engaged Akt phosphorylation pathway .
The significance of ORAI2-mediated Ca2+ signaling in cancer hallmarks is still poorly understood and characterized. There is basically only one study in human myeloid leukemia HL60 cell line describing that in these cells SOCE is strongly dependent on both ORAI11 and ORAI2 and implicating both of them in cell proliferation and migration with the latter probably due to modulation of FAK tyrosine phosphorylation  (see Figure).
The bulk of available data suggests that ORAI proteins represent important determinants of cancer hallmarks through their involvement in dysregulated Ca2+ homeostasis and signaling of cancer cells. In answering the question asked in the title of this review, one can state that in the most cases up-regulated expression and/or function of ORAI proteins in cancer cells resulting in amplified SOCE or SICE promotes cancer hallmarks in a cancer type-specific manner, and in that respect they can be regarded as cancer allies. Only when ORAI proteins up-regulation can lead to the global cytosolic Ca2+ overload, as part of the intrinsic apoptotic pathway, they can be viewed as cancer foes. In this regard, potential ability of ORAI proteins to support Ca2+ overload via Ca2+ entry pathways they are involved in (i.e. SOCE or SICE) can be exploited for cancer treatment. Indeed, the function of any type of membrane ion channel can be categorized onto at least three levels: 1) moderate when it supports normal physiological processes, 2) up-regulated when its activity may contribute to the pathologic prevalence of some cellular process(es) thereby promoting cancer hallmark(s) and 3) hyper-activated when strongly augmented transmembrane ion fluxes through specific type of channel can actually kill the cell via disruption of cellular ionic and/or volume homeostasis. The 3rd level of channel function becomes especially detrimental when the channel is Ca2+-permeable, given the importance of elevated [Ca2+]c in promoting cell death. Thus, developing new molecular or pharmacological means for targeted enhancing the activity of ORAI-based channels as well as of any other Ca2+-permeable channel which becomes overexpressed in cancer can represent a viable strategy for cancer therapy.
No Comments » Add comments