PI-4K IIIβ (PI-4K-β) catalyzes the synthesis of phosphatidylinositol 4-phosphate, which is a late limiting step in the synthesis of phosphatidylinositol 4,5-bisphosphate, an important lipid regulator of many cellular functions including exocytosis. A yeast homolog of NCS-1 and PI-4-K interact, and NCS-1-induced activation of PI-4-K is required for yeast survival (Hendricks et al., 1999). Structural support for interaction was obtained from a recent NMR structure of Ca²⁺-bound yeast NCS-1 (Ncs1) in complex with an N-terminal yeast PI-4-K (Pik1) fragment (Strahl et al., 2007). This interaction was also detected in neuroendocrine cells (Koizumi et al., 2002; Scalettar et al., 2002; Rajebhosale et al., 2003; de Barry et al., 2006), neurons (Taverna et al., 2002; Zheng et al., 2005) and other cell types including pancreatic beta cells (Gromada et al., 2005) and mast cells (Kapp-Barnea et al., 2003), and was shown to facilitate exocytosis and secretion in these cells (Table 1). However, previous reports have suggested no direct interaction in neurons (Bartlett et al., 2000). This contradiction may be explained by the presence of newly discovered PI-4-Kβ regulators, calneurons. While calneurons interact with PI-4K-β at low Ca²⁺ levels to inhibit its enzyme activity, NCS-1 binds to PI-4K-β at high Ca²⁺ levels to activate it (Mikhaylova et al., 2009), suggesting that calneurons and NCS-1 compete for PI-4-K-β interaction depending on intracellular Ca²⁺ levels. Thus, when the intracellular Ca²⁺ level is low, its interaction might be difficult to be detected.

The V7 Drosophila mutant that overexpresses NCS-1 has a phenotype of hyperactivity, which results in the proposal that NCS-1 facilitates neurotransmission, possibly by regulating the activities of ion channels (Pongs et al., 1993; Poulain et al., 1994). Indeed, we found that NCS-1 is a Ca²⁺-sensitive regulatory component of a native K⁺ current (Nakamura et al., 2001; Table 1). In the brain and heart, rapidly inactivating (A-type) voltage-gated K⁺ currents control cellular excitability. Although the pore-forming alpha-subunits of these channels are considered to be K_V4 channels (Serôdio et al., 1994; Fiset et al., 1997; Nakamura et al., 1997), the kinetic properties of K_V4 channels differ from native A-type currents, suggesting the presence of regulatory subunits. KChIPs, a member of the NCS protein subfamily, were initially reported as a specific K_V4 regulatory subunit (An et al., 2000). Because NCS-1 modulates K_V4 currents similar to KChIPs, by increasing current amplitude and slowing the inactivation time course, and NCS-1 physically interacts with K_V4.2 in mouse brain, it was identified as a regulator of A-type K⁺ currents in neurons (Nakamura et al., 2001; Table 1).

This interaction and activation also occurs in adult mouse cardiomyocytes (Guo et al., 2002) and in zebrafish heart (Nakamura and Coetzee, 2008), which lacks KChIPs. The differential regulation of K_V4 channels by NCS-1 and KChIPs in specific tissues and cell types was an unaddressed topic, and this was clearly demonstrated in the report by Nattel’s group (Xiao et al., 2013). Purkinje fibers (PFs) show an unusual form of transient outward K⁺ current I_to with slow recovery kinetics and TEA sensitivity compared with ventricular I_to, suggesting a distinct molecular composition. This group found that NCS-1 and DPP6, which were also reported to be auxiliary subunits of K_V4 K⁺ channels (Jerng et al., 2004), are preferentially enriched in PFs, while KChIP2, an essential subunit of ventricular K_V4.3 is weakly expressed. Moreover, NCS-1 slowed inactivation kinetics of K_V4.3, while DPP6 increased its current amplitude, thus increasing the I_to-mediated K⁺ efflux (Figure 1Ba), which would accelerate PF repolarization and shortening of action potentials (Figure 1Bb; similar computer simulation was reported by Xiao et al., 2013). Thus, overexpression of K_V4 auxiliary subunits may result in steep transmural repolarization gradients in PFs with adjacent ventricular tissues that induces coupled ectopic activity, and potentially leads to lethal arrhythmias (Figure 1Bc). NCS-1 also interacts with the anti-cancer drug taxol (Boehmerle et al., 2006), and is involved in the regulation of taxol-induced cardiac arrhythmia (Zhang et al., 2010). Thus, NCS-1 can be a potential target for anti-arrhythmic therapy.

NCS-1 regulates voltage-gated Ca²⁺ channels. Published data, however, are somewhat inconsistent with reports demonstrating both positive and negative effects (Table 1). NCS-1 was described to inhibit P/Q-type Ca²⁺ channels an regulates autocrine pathways in adrenal chromaffin cells (Weiss et al., 2000; Weiss and Burgoyne, 2001) and N-type Ca²⁺ channels in PC12 cells, which reduces neurite elongation (Gambino et al., 2007). Other studies, in contrast, have demonstrated positive regulation of N-type Ca²⁺ channels, causing glial cell line-derived neurotrophic factor (GDNF)-induced enhancement of neurotransmitter release in motoneurons (Wang et al., 2001). Activation of P/Q-type Ca²⁺ channels by NCS-1 causes activity-dependent synaptic facilitation in nerve terminals (Tsujimoto et al., 2002). In Drosophila, NCS-1 enhances neurotransmission and nerve terminal growth, by functionally interacting with the α1 subunit of the voltage-gated Ca²⁺ channel (Dason et al., 2009). The possible reason for the apparent contradictory findings is that the effects may be cell type-specific and/or mediated by accessory proteins, such as a β-subunit (Rousset et al., 2003), or dependent on other interacting proteins, such as IL1RAPL1 that cooperatively regulates the N-type Ca²⁺ channel via NCS-1 (Gambino et al., 2007). In addition, regulation of the Ca²⁺ channel by Ca²⁺ influx through the channel should be considered. Well-characterized examples are Ca²⁺-dependent inactivation (Standen and Stanfield, 1982) and facilitation (Dolphin, 1996) of Ca²⁺ channels regulated by other Ca²⁺-binding proteins, such as calmodulin (Budde et al., 2002; Christel and Lee, 2012). Future research should aim to understand the regulatory mechanisms of Ca²⁺ channels that involve NCS-1.

Dopamine plays an important role in the reward system of the brain. Disorders of the dopamine system result in several psychiatric and neurological conditions. Dopamine transmission is regulated by dopamine receptor-interacting proteins (DRIP), including NCS-1, calcyon, and DARPP-32. NCS-1 directly interacts with the D2 dopamine receptor, inhibits D2 receptor phosphorylation, and reduces the agonist-mediated internalization of the receptor (Kabbani et al., 2002), indicating that NCS-1 preserves dopamine signaling (Table 1). Indeed, modest NCS-1 overexpression in the dentate gyrus in mice promotes exploration, synaptic plasticity, and rapid acquisition of spatial memory (Saab et al., 2009). NCS-1 is upregulated in the prefrontal cortex of patients with schizophrenia and bipolar disorder (Koh et al., 2003; Bai et al., 2004). Because the levels of other DRIPs were also changed in patients with schizophrenia (Bai et al., 2004; Souza et al., 2008), DRIP signaling is possibly involved in psychiatric disorders. Furthermore, recent findings indicate the N-terminal 60 residues of NCS-1 are responsible for binding to the D2 receptor (Woll et al., 2011). Such knowledge would provide an opportunity to screen for drugs that can specifically interrupt the NCS-1-D2 dopamine receptor interaction and thus prevent psychiatric disorders.

Several studies have demonstrated that NCS-1 modulates learning and memory. For example, deletion or reduction of NCS-1 resulted in dysfunction of learning and memory in Caenorhabditis elegans (Gomez et al., 2001), as well as in mice (Mun et al., 2015), whereas mice overexpressing NCS-1 rapidly acquire spatial memory (Saab et al., 2009). Thus, NCS-1 affects neurophysiology, possibly through various interacting proteins. A mechanism underlying NCS-1-mediated learning and memory was further investigated (Nakamura et al., 2017). Ncs1^−/− mice exhibited impaired spatial learning and memory function in the Morris Water Maze test, with slight changes in their exercise activity or a structural change in the hippocampus. However, the levels of brain-derived neurotrophic factor (BDNF), a key regulator of memory function, and dopamine were decreased. Furthermore, phosphorylation of Ca²⁺/calmodulin-dependent protein kinase II-α (CaMKII-α), which regulates long-term potentiation, and BDNF levels were decreased, suggesting that CaMKII-α signaling that increases BDNF production is at least partly involved in NCS-1-mediated learning and memory function.

Ca²⁺ signaling via inositol 1,4,5-trisphosphate receptors (InsP₃Rs) regulates cellular function and is involved in pathogenesis (Table 1). NCS-1 physically interacts with InsP₃R1 and enhances InsP₃R-mediated Ca²⁺ signaling in rat brains. Indeed, physical/functional interaction of these proteins was directly demonstrated in an in vitro experiment showing that the addition of NCS-1 to InsP₃R1 in the lipid bilayer increased InsP₃R channel activity (Schlecker et al., 2006). This interaction was also detected at the growth cone region of neurites in cultured neurons, and indicated to be crucial for neurite outgrowth (Iketani et al., 2009). Metabotropic glutamate receptor-mediated cis also mediated by NCS-1/InsP₃R interaction (Jo et al., 2008). Pathologically, NCS-1/InsP₃R1 interaction is believed to be involved in bipolar disorder (Schlecker et al., 2006) because lithium, a medical drug for bipolar disorder, inhibited the NCS-1-induced enhancement of InsP₃R function. NCS-1/InsP₃R interaction is also considered to mediate neuropathy (Boehmerle et al., 2006, 2007; Blachford et al., 2009), as paclitaxel (taxol), a chemotherapeutic agent used for the treatment of solid cancers, modulates the expression/function of NCS-1, and hence InsP₃R1-mediated Ca²⁺ signaling.

NCS-1/InsP₃R interaction is also detected in the heart and is crucial for contraction at the immature stage and cardiac hypertrophy in adult (Nakamura et al., 2011; Table 1 and Figure 1C). A high expression of NCS-1 was found in the immature heart (Nakamura et al., 2003, 2011), but its function at this stage was unknown. Using Ncs1^−/− mice, Nakamura et al demonstrated that NCS-1 contributes to an increase in contraction and Ca²⁺ signaling, specifically at the immature stage (Figure 1Cb). Intracellular Ca²⁺ levels and the sarcoplasmic reticulum (SR) Ca²⁺ content were significantly lower in Ncs1^−/− myocytes at the neonatal stage than in wild-type cells. Mechanistically, the interaction of NCS-1 with InsP₃R increases InsP₃R-dependent Ca²⁺ signaling, followed by the activation of CaMKII-dependent pathways, and promotes SR Ca²⁺ pump via the phosphorylation of phospholamban (PLB), which ultimately induce increase in the SR Ca²⁺ content and global Ca²⁺ transient, thus cardiomyocyte contraction (Figures 1Ca,b). The importance of crosstalk among NCS-1, InsP₃Rs, and CaMKII in the immature hearts was evident by the high expression of all three proteins in immature hearts (Nakamura et al., 2011). In the neonatal mouse heart, the structure and function of SR are immature. Nonetheless, it is considered a primary source of Ca²⁺ necessary for muscle contractions, suggesting the existence of factors missing during development and promoting SR-dependent excitation-contraction (E-C) coupling in the postnatal stages. NCS-1 may act as one of these missing factors. Numbers of molecules which levels are high at the immature stage are often up-regulated in the disease conditions, such as cardiac hypertrophy. NCS-1 is also highly expressed during the early stages of hypertrophy in the adult heart and promotes the progression of hypertrophy, at least in part, through InsP₃R activation (Nakamura et al., 2011; Figure 1Cc). A possible molecular mechanism is suggested in the next section.

The aforementioned data indicate that NCS-1 can discretely regulate different types of Ca²⁺ signaling pathways in the heart (i.e., regulation of E-C coupling in immature heart and changes in gene expression in the adult heart). Recent evidence suggests that gene transcription is regulated by nuclear Ca²⁺ signals. However, the mechanisms underlying nuclear Ca²⁺ regulation and its relationship to cytoplasmic Ca²⁺ regulation have not been completely solved. Using a subcellular-specific, fluorescent protein-based Ca²⁺ indicator GECO (Zhao et al., 2011; Nakao et al., 2015) confirmed the following: (1) nuclear Ca²⁺ transients were elicited by both electrical and receptor stimulations (with insulin-like growth factor-1, IGF-1) in neonatal mouse ventricular myocytes; and (2) receptor stimulation-elicited nuclear Ca²⁺ transients were mainly mediated by InsP₃Rs. Furthermore, based on the evidence that IGF-1-elicited nuclear Ca²⁺ transient was significantly diminished in Ncs1^−/− cardiomyocytes, NCS-1 is involved in the receptor stimulation-induced nuclear Ca²⁺ regulation through interaction with InsP₃Rs (Nakao et al., 2015; Table 1). This may contribute to NCS-1-mediated hypertrophy, which was described above. A possible mechanism underlying a dual effect of NCS-1 on nuclear Ca²⁺ signals and E-C coupling is that NCS-1 increases the Ca²⁺ content of SR (Nakamura et al., 2011) that is interconnected to the nuclear envelope (Wu and Bers, 2006) and consequently may increase nuclear Ca²⁺ (Nakao et al., 2015; Figure 1Ca).

Physical or chemical injury and genetic abnormalities can result in neuronal degeneration, which may underlie human neurodegenerative disorders, such as Alzheimer’s disease and Parkinson’s disease. Both intrinsic and extrinsic factors, including neurotrophic factors, can activate the anti-apoptotic process to rescue neuronal cell death. The signaling pathway leading to cell survival remains unresolved. In this regard, NCS-1 was found to be a novel Ca²⁺-dependent survival-promoting factor upregulated in injured neurons (Nakamura et al., 2006), based on the following observations. (1) NCS-1 expression increases in injured neurons; (2) NCS-1 overexpression diminished various stress-induced neuronal cell death in culture; and (3) the dominant negative EF-hand NCS-1 mutant (E120Q) accelerated cell death. Mechanistically, the expression level of NCS-1 in neuron is increased by GDNF, a neurotrophic factor upregulated by neuronal injury, and NCS-1 mediates GDNF survival signaling via the activation of the Akt pathway.

Not only in neurons, NCS-1 also plays a key role in protecting cardiomyocytes against stress through the activation of mitochondrial detoxification pathways (Nakamura et al., 2016). Excessive stress induces cytosolic Ca²⁺ overload and cell death. In contrast, mild forms of stress lead to physiologically relevant changes in Ca²⁺, which activate Ca²⁺-dependent survival pathways by binding to Ca²⁺-sensor proteins. As one such protein, NCS-1 was found to play important roles in Ca²⁺-dependent survival signaling. Ncs1^−/− myocytes were more susceptible to oxidative and metabolic stress, and cellular ATP levels, mitochondrial respiration and biosynthesis were significantly reduced in these cells. In wild-type myocytes, mild oxidative stress increased the mitochondrial proton leak, which exerted a protective effect by inhibiting the production of reactive oxygen species. However, this response was diminished in Ncs1^−/− cardiomyocytes, thus resulting in cell death. Similar susceptibility was also observed in Ncs1^−/− hearts subjected to ischemia-reperfusion injury. In these hearts, molecules regulating Ca²⁺-dependent survival pathways, such as Akt and PGC-1α, which promote mitochondrial biogenesis and function, were significantly downregulated compared to wild-type hearts. These data demonstrate a novel role of NCS-1 that contributes to stress tolerance in cardiomyocytes, partly by the activation of Ca²⁺-dependent survival pathways. NCS-1 may also participate in cardioprotection by mediating receptor-signaling pathways. For example, NCS-1 associates with, and modulates, adenosine receptor activity (Navarro et al., 2012). Given the central role of adenosine in mediating the protective effects of ischemic preconditioning (Cohen and Downey, 2008), it is entirely possible that the cardioprotective effects of NCS-1 is partially mediated by this pathway.