Decoding the Phosphatase Code: Regulation of Cell Proliferation by Calcineurin

Calcineurin, a calcium-dependent serine/threonine phosphatase, integrates the alterations in intracellular calcium levels into downstream signaling pathways by regulating the phosphorylation states of several targets. Intracellular Ca2+ is essential for normal cellular physiology and cell cycle progression at certain critical stages of the cell cycle. Recently, it was reported that calcineurin is activated in a variety of cancers. Given that abnormalities in calcineurin signaling can lead to malignant growth and cancer, the calcineurin signaling pathway could be a potential target for cancer treatment. For example, NFAT, a typical substrate of calcineurin, activates the genes that promote cell proliferation. Furthermore, cyclin D1 and estrogen receptors are dephosphorylated and stabilized by calcineurin, leading to cell proliferation. In this review, we focus on the cell proliferative functions and regulatory mechanisms of calcineurin and summarize the various substrates of calcineurin. We also describe recent advances regarding dysregulation of the calcineurin activity in cancer cells. We hope that this review will provide new insights into the potential role of calcineurin in cancer development.


Introduction
Intracellular calcium ions (Ca 2+ ) act as pleiotropic secondary messengers in key signaling pathways involving a variety of cellular functions. While the extracellular Ca 2+ concentration is 11.5 mM, the steady-state intracellular Ca 2+ concentration is kept very low at several tens of nanomoles. It is well known that the concentration of Ca 2+ varies greatly in different cellular compartments, for example, intracellular calcium, 100 nM; endoplasmic reticulum (ER), 0.5-1 mM; and mitochondria, 100-200 nM [1].
The binding of a hormone or growth factor to a G protein coupled receptor (GPR) or a tyrosine kinase receptor (RTK) initiates the Ca 2+ signaling cascade. The activation of such receptors transmits the signals to phospholipase C (PLC), which cleaves phosphatidylinositol 4,5-biphosphate (PIP 2 ) to produce diacylglycerol (DAG) and inositol-1,4,5-trisphosphate (InsP 3 ). InsP 3 then binds to the InsP 3 receptor (InsP 3 R) and stimulates the release of Ca 2+ from intracellular stores, such as ER [2], and allows for the entry of Ca 2+ [3,4] from the extracellular space. Spatially and temporally coordinated Ca 2+ release via InsP 3 R is regulated by complex feedback mechanisms [1].
Changes in Ca 2+ concentration trigger signal transduction, which regulates a wide range of cellular events such as gene expression, cell cycle, cell motility, autophagy, and apoptosis [1,5]. Local changes in intracellular Ca 2+ diffuse through the cell and have effects on the distal sites [6]. Long-term intracellular Ca 2+ increases in the mitochondria and causes the release of cytochrome c and subsequently triggers apoptosis [7]. Due to the diverse roles of Ca 2+ , the dysregulation of calcium homeostasis can impair cell function and trigger cell death, and by doing so, may contribute to heart disease, cancer, and mental disorders. As a result, Ca 2+ signals must be tightly regulated during the cell cycle. Indeed, previous studies have characterized Ca 2+ transients during the cell cycle [8,9]. For example, Pande CaN dephosphorylates the NFAT transcription factor, which in turn activates p21, cyclin D1, CDK4, c-myc, and cyclin A. CaN also stabilizes cyclin D1 by dephosphorylation. Furthermore, the CaN/NFAT pathway and its downstream target c-myc regulate p21. p21 is a well-known inhibitor of CDK2-cyclin E and CDK4/6cyclin D. CaMK negatively regulates the expression of p27, which is an inhibitor of CDK4-cyclin D and CDK2-cyclin E. CDK2-cyclin E and CDK4/6-cyclin D complexes phosphorylate Rb, leading to the activation of E2F1 and the subsequent G1/S progression. In G2/M, CaMK phosphorylates and NFAT transcription factor, which in turn activates p21, cyclin D1, CDK4, c-myc, and cyclin A. CaN also stabilizes cyclin D1 by dephosphorylation. Furthermore, the CaN/NFAT pathway and its downstream target c-myc regulate p21. p21 is a well-known inhibitor of CDK2-cyclin E and CDK4/6cyclin D. CaMK negatively regulates the expression of p27, which is an inhibitor of CDK4-cyclin D and CDK2-cyclin E. CDK2-cyclin E and CDK4/6-cyclin D complexes phosphorylate Rb, leading to the activation of E2F1 and the subsequent G1/S progression. In G2/M, CaMK phosphorylates and activates cdc25, leading to downstream dephosphorylation and the activation of CDK1. Solid red lines indicate phosphorylation, red dotted lines indicate dephosphorylation, and green dotted lines indicate transcriptional activation.

Characteristics of Calcineurin
CaN, also known as protein phosphatase 2 B (PP2B), is a serine/threonine protein phosphatase that is conserved in all eukaryotes [54][55][56][57]. CaN is a heterodimer composed of a catalytic subunit, calcineurin A (CnA), and a Ca 2+ -binding regulatory subunit, calcineurin B (CnB). In mammals, three independent genes, PPP3CA, PPP3CB, and PPP3CC, encode CnAα, CnAβ, and CnAγ, respectively. CnAα and CnAβ exhibit ubiquitous expression patterns, whereas the CnAγ expression is restricted to the testis and brain [58][59][60][61]. The CaN regulatory subunits CnB1 and CnB2 are encoded by two genes (PPP3R1 and PPP3R2, respectively). The CnB1 protein is expressed ubiquitously, whereas the CnB2 protein is specifically expressed in the testes. CnA contains an amino-terminal catalytic domain followed by a CnB-binding domain, a CaM-binding domain, and an autoinhibitory domain. CaN also contains a nuclear localization signal (NLS) in the catalytic domain and a nuclear export signal (NES) in the carboxyl terminus. The autoinhibitory domain of CnA blocks the catalytic site and the NLS [56,62]. CnB contains four EF-hand Ca 2+ -binding motifs and an amino-terminal myristylation site. CaN is activated by the increased intracellular Ca 2+ concentration in the cell and plays essential roles in multiple signaling processes [63]. The binding of Ca 2+ /CaM to the regulatory domain leads to the attenuation of autoinhibition, followed by dramatic enzymatic activation. Of note, although there are multiple kinases that are regulated by CaM, and CaN is the only phosphatase directly regulated by Ca 2+ /CaM.
As mentioned above, CaN is widely distributed in various mammalian tissues and is particularly abundant in neural tissues [64]. However, the abundance and sub-cellular localization of CnAα and CnAβ are different. CnAα is more abundantly expressed than CnAβ; furthermore, CnAα localizes in the nucleus while CnAβ localizes in the cytoplasm [65]. Interestingly, although CaN localizes predominantly in the cytoplasm of unstimulated cells [61,66,67] in response to elevated Ca 2+ concentrations, and a small portion of CaN can translocate to the nucleus and interact with the target substrates [68].
CHP competes with CnB to bind to CnA, and inhibits CaN activity [71,72]. While Cabin1/cain inhibits CaN by interacting with CaN in a phosphorylation-dependent manner through a binding site on CaN, which is distinct from that of the drug-immunophilin complex [73,74]. Calcipressin/RCAN/DSCR/CSP has also been identified as a CnA-binding protein that inhibits CaN activity [77,80]. A conserved peptide (FLISPPxSPP) of the calcipressin family is phosphorylated and functions as a binding site for CaN. As the expression of calcipressin is induced by CaN, it functions as a feedback inhibitor of CaN signaling.
Calcipressin/RCAN/DSCR/CSP binds CaN at the same site as NFAT and other substrates, with competition for binding between these molecules being a possible regulatory mechanism [81]. FKBP38 targets BCL-2 to the mitochondria and inhibits apoptosis. The same protein also binds to and inhibits CaN, even in the absence of FK506 [78]. Furthermore, histone H1 inhibits CaMKII and CaN by blocking CaM autophsophorylation [82]. CaN is also inactivated by the oxidation of key methionine residues [83][84][85][86]. Conversely, CaN is activated by the intramolecular cleavage by two different proteases, caspase 3 and Ca 2+dependent protease calpain [87][88][89]. Several studies have identified that CaN activity is regulated by its phosphorylation. The CaMKII-mediated phosphorylation of CaN on Ser197 inhibits the CaN activity [90][91][92]. This phosphorylation is blocked by Ca 2+ /CaM binding to CaN. In contrast, although the auto-dephosphorylation of CaN is very slow, it can be rapidly dephosphorylated by protein phosphatase IIA [91].

Functions of Calcineurin/NFAT
Cyclosporine A and FK506 are well-characterized immunosuppressive agents that prevent organ transplant rejection [93][94][95]. These compounds bind tightly to endogenous cytoplasmic cyclophilin A or FKBP12, respectively. Interestingly, cyclophilin A or FKBP12, in complex with cyclosporine A or FK506, bind to CaN and block the access of the CaN substrate to the active site of the CaN [96]. This indicates the possibility that the immunosuppressive effects of these drugs could be, in part, caused by the disruption of CaN functions. Consistent with this view, it has been show that CaN inhibition by cyclosporine A or FK506 delays G 1 /S progression in various cell types [97][98][99][100]. Mechanistically, cyclosporin A induces the expression of the cyclin inhibitor p21 and a reciprocal reduction in cyclins A and E, leading to CDK2 inactivation [101,102].
The most studied substrates of CaN are the family of nuclear factors of activated T-cell (NFATc or NFAT) transcription factors [57,103]. Four of the five members of the NFAT protein family, NFATc1 (NFAT2), NFATc2 (NFAT1), NFATc3 (NFAT4), and NFATc4 (NFAT3), are regulated by calcium signaling. In T cells, dephosphorylation of NFAT by CaN changes the structure of the NFAT protein, exposing the NLS, which is then relocated to the nucleus to regulate the transcription of immune function-associated genes [104][105][106]. In addition to T cells, NFATs are also present in a wide variety of cells and tissues, and dephosphorylation of NFAT by CaN activates transcription in the neurons and astrocytes [107,108], and also in the heart and skeletal muscle [109][110][111][112]. Of note, NFAT has been shown to regulate cell cyclerelated genes and promote cell cycle progression [103,[113][114][115]. Furthermore, both CaN and NFATc1 regulate cyclin D1 transcription [37]. In addition, Camp-responsive element binding protein 1 (CREB1) transcription factor, which binds to the cyclin D1 promoter and induces cyclin D1 mRNA expression, is also regulated by CaN [98]. The other key target of NFATc1 in cell cycle progression is c-Myc. CaN-mediated dephosphorylation of NFATc1 activates MYC transcription, allowing the cell to proceed to the S phase [116,117]. NFATc1 binds directly to the NFAT site in the MYC promoter [115]. In addition, NFATc1 increases c-myc transcription by activating the ERK1/2/p38/MAPK signaling pathway [118] or inducing histone acetylation, resulting in the binding of ELK1 to the MYC promoter [119]. Thus, NFATc1 upregulates MYC transcription via multiple mechanisms.
Given the diverse expression and function of each NFAT isoform, its dysregulation is known to be associated with tumorigenesis, Alzheimer's disease, and the development of autoimmune and inflammatory diseases. A novel therapeutic strategy for treating NFATrelated diseases is to develop new ways to selectively regulate specific NFAT isoforms [120].

Other Substrates of Calcineurin
Approximately 600 proteins with conserved sequences that bind to CaN have been identified [121]. Many of the known CaN substrates possess PxIxIT and/or LxVP motifs [56]. Such substrates include transcription factors, proteins involved in cell cycle and apoptosis, cytoskeletal proteins, scaffold proteins, membrane channels, and receptors (Table 1) [56,58,122,123]. The amino acid residues of the protein indicated by asterisks (*) are dephosphorylated when the cells are stimulated with ionomycin. It is not known whether calcineurin directly dephosphorylates them. ND means not determined.
Key proteins associated with tumor development such as myocyte enhancer factor 2 (MEF2), kinase suppressor of ras 2 (KSR2), DAXX, c-Jun, and nuclear factor I (NFI), are known CaN substrates [128,129,134,136,138]. Furthermore, CAN dephosphorylatesd the pro-apoptotic Bcl-2 family member BAD [126] and the apoptosis promoting factor ASK1 [125]. Dephosphorylation of another apoptosis related factor CaMKIIγ is promoted by CaN, leading to the nuclear translocation of CaMKIIγ. Nuclear translocation of TFEB, a master transcriptional regulator of lysosome biogenesis and autophagy, also requires dephosphorylated by CaN [127,145]. In addition, while DARPP-32 and ElK-1 are inactivated by CaN, other targets such as nitric oxide synthase, NHE1, and TRESK are activated [142,[146][147][148][149]. CaN is activated through cytoplasmic Ca 2+ increase caused by mitochondrial depolarization. Activated CaN dephosphorylates DRP1, which then relocates to the mitochondria and promotes mitochondrial fission [130]. In neurons, CaN dephosphorylates dynamin1. This promotes the endocytosis of TRKA receptors and subsequent axonal growth [131]. CaN-dependent dephosphorylation of GluA1, a component of AMPA receptors (AMPARs) at the synapses, leads to the removal of AMPARs from the synapse and endocytosis [133], while dephosphorylation of Myosin phosphatase target subunit 1 (MYPT1, a component of MP) by CaN causes MP to acquire phosphatase activity [137].
Recently, we found that CaN inhibits cyclin D1 degradation by dephosphorylation of the Thr286 residue [38] (Figure 2a). Treatment with CN585, which is a specific inhibitor of CaN phosphatase activity, or by the immunosuppressant FK506, inhibits breast cancer cell proliferation accompanied by delayed G 1 /S progression mediated by the degradation of cyclin D1. FK506 also decreases the protein level of CDK4 via a yet unknown mechanism. Taken together, CaN activates CDK4-cyclin D1 through multiple mechanisms to promote cell cycle progression [12,38]. The overexpression of cyclin D1 has been linked to the development and progression of cancer [150]. It is also known that increased levels of cyclin D1 could frequently result from deregulated proteasomal degradation [151]. This indicates that the selective inhibition of CaN could be an effective method for the treatment of cancers that are characterized by cyclin D1 overexpression. We also found that the stability and activity of ERα, a key molecule for estrogen-dependent cancer progression, is mediated by CaN [132] (Figure 2b). CaN dephosphorylates ERα on Ser294 and activates the mechanistic target of rapamycin (mTOR). CaN is also dephosphorylate MAP2, RIIα, RCAN1, and Tau, but the significance of these dephosphorylations remains unclear [124,135,143,144].
Intriguingly, CaN may possess roles that are independent of its dephosphorylation activity. CnAβ1 functionally activates the PDK1-Akt phosphorylation cascade, increasing the phosphorylation of its target, Ser253 of FOXO3a, and the inhibition of its nuclear translocation. The inhibitory effect of FOXO3a by CnAβ1 regulates myoblast proliferation and prevents myotube atrophy under starvation conditions. Of note, this function of CnAβ1 does not require a phosphatase activity, suggesting that CaN functions independently of its dephosphorylation activity [152].
indicates that the selective inhibition of CaN could be an effective method for the treatment of cancers that are characterized by cyclin D1 overexpression. We also found that the stability and activity of ERα, a key molecule for estrogen-dependent cancer progression, is mediated by CaN [132] (Figure 2b). CaN dephosphorylates ERα on Ser294 and activates the mechanistic target of rapamycin (mTOR). CaN is also dephosphorylate MAP2, RIIα, RCAN1, and Tau, but the significance of these dephosphorylations remains unclear [124,135,143,144].

Factor Alterations in Cancer Types of Cancer Reference
Calcineurin (CnA) Overexpression glioma (malignant gliomas, including grades III and IV astrocytomas) [138] breast cancer (ER-α-positive) [132] Activation lymphomas (lymphoid malignancies) [160] Overexpression, activation colon cancer [39] NFATc1 Overexpression ovarian cancer (clear-cell carcinoma) [118,163] liver cancer (hepatocellular carcinoma) [156] prostate cancer [155] lymphomas (large B-cell lymphoma) [162] Nuclear localization lymphomas (diffuse large B-cell lymphomas) [161] breast cancer (triple-negative) [153] Suppression liver cancer (hepatocellular carcinoma) [164,165] lymphomas (anaplastic large cell lymphomas and classical Hodgkin's lymphomas) [166] Overexpression, nuclear localization pancreatic cancer (pancreatic adenocarcinoma) [115] NFATc2 Overexpression melanoma [158] glioma (glioblastoma) [157] Nuclear localization lung cancer [154] NFATc1, NFATc3 Dephosphorylation leukemia (T-cell acute lymphoblastic leukemia) [159] The mechanisms that drive CaN activation in cancer are not well understood. In some cancers, CaN hyperactivation has been associated with mutations. For example, the EL4 murine T-cell lymphoma cell line expresses a mutant form of CaN, in which the aspartic acid at position 477 is mutated to asparagine and the negative regulation of the phosphatase activity by the autoinhibitory domain is impaired by this mutation [167]. In contrast, the SML B-cell lymphoma cell line expresses a truncated version of CnA, which results in constitutive activation of CaN [168]. It has been reported that the expression of Ca 2+ channels that regulate the intracellular Ca 2+ concentration is involved in the activation of CaN. TRPv6 (TRP vanilloid family member 6), a Ca 2+ -selective ion channel, is highly expressed in prostate cancer and is a prognostic marker [169,170]. Increased expression of TRPv6 induces more Ca 2+ to enter the cell, which in turn promotes NFAT activation [171]. It is also possible that the inflammatory response activates CaN [172][173][174]. Consistent with this notion, in intestinal cancer, changes in the bacterial community activate Toll-like receptor (TLR) signaling, which leads to subsequent calcium entry and CaN activation [175]. Moreover, inflammatory responses are activated in many cancers, and it is well understood that chronic inflammation is a risk factor for tumorigenesis [176]. Recently, we found that a high expression of CaN was correlated with a poor prognosis regarding the outcome of endocrine therapy in patients with ERα-positive breast cancer [132], which indicates that the selective inhibition of CaN could be effective to treat such cancers.
Highly activated CaN is supposed to activate its substrates, including NFAT, to promote proliferation. Similar to CaN, its target NFAT is also constitutively activated or overexpressed in numerous cancers and may contribute to cancer development and progression [177]. Conversely, the expression of NFATc1 is suppressed in several types of cancer, and a reduction of NFATc1 has been shown to be linked with aggressiveness and malignancy of cancer [164][165][166].

A Therapeutic Perspective for Cancer
Owing to the high frequency of CaN/NFAT activation in cancer and the contribution of these molecules in cancer progression, the CaN/NFAT pathway could be a potential therapeutic target. Indeed, the anticancer effects of CaN inhibitors have been studied extensively in the past. For instance, cyclosporine A or FK506 induced apoptosis and rapid tumor clearance, resulting in the regression of leukemia [160]. Cyclosporine A or FK506 also inhibits tumor growth in the bladder and prostate xenografts in vivo [174,178,179]. In addition, cyclosporine A itself is also directly involved in tumor growth, as it increases TGFβ production [180], activates Ras [181], suppresses PTEN expression, and increases AKT activation [182].
There is growing acceptance that targeting dysregulated Ca 2+ channels/transporters/pumps can provide promising potential for the treatment of cancer patients. [183]. Buffering nuclear Ca 2+ concentrations alters the expression levels of the genes involved in cell proliferation, resulting in an antitumor effect [184]. Moreover, buffering nuclear Ca 2+ reduces the growth rate of tumor cells without affecting the cells in normal tissues [185]. Further research is needed to clarify the differences in the sensitivity of normal and cancer cell growth to changes in nuclear Ca 2+ levels.

Conclusions
Nuclear Ca 2+ is involved in tumor growth and alters the expression of the genes involved in cell proliferation. Furthermore, previous studies have suggested that the modulation of nuclear Ca 2+ signaling may be effective in cancer therapy. Activation of CaN and its downstream dephosphorylation has been identified as a mechanism by which nuclear Ca 2+ regulates cell proliferation and cell cycle progression. As discussed above, dephosphorylation of proteins by CaN plays an important role in tumor formation and progression. Therefore, in the future, it will be necessary to identify the molecular mechanism of CaN activation and the substrates that promote cancer cell growth. Targeting the interactions of activated CaN with specific substrates in cancer cells, without affecting the normal immune function of CaN, may effectively inhibit the growth of cancer cells, leading to the establishment of new tumor-specific therapies.