Inhibition of Lithium Sensitive Orai1/ STIM1 Expression and Store Operated Ca²⁺ Entry in Chorea-Acanthocytosis Neurons by NF-κB Inhibitor Wogonin

Chorein binds to and supports activation of phosphoinositide-3-kinase (PI3K)-p85-subunit thus participating in the regulation of actin polymerization and cell survival [1-4]. Loss-of-function-mutations of the chorein encoding gene VPS13A (vacuolar protein sorting-associated protein 13A) result in chorea-acanthocytosis (ChAc) [5, 6], a neurodegenerative disease with hyperkinetic movements, epilepsy, impaired cognitive functions, myopathy, and erythrocyte acanthocytosis [5, 7-9]. The progressive neurodegeneration eventually leads to severe disability and early death [9]. Erythrocyte shape changes [10], neuronal apoptosis [11] and altered behaviour [11] are similarly observed in gene targeted mice lacking functional chorein.

Chorein expression is observed in a wide variety of tissues and across a wide range of cell types [12-14]: Chorein-sensitive functions include dopamine release [15], platelet activation [14], cytoskeletal architecture [16], endothelial cell stiffness [13], as well as survival of tumour cells [3], neurons and skeletal muscle cells [5, 17].

Cell survival and cell death are sensitive to alterations of cytosolic Ca²⁺ activity ([Ca²⁺]_i) [18, 19]. [Ca²⁺]_i could be increased by Ca²⁺ release from intracellular stores with subsequent store-operated Ca²⁺ entry (SOCE) through the pore-forming Ca²⁺ channel subunits Orai1, Orai2 and/or Orai3 [20]. The Orai isoforms are activated following store depletion by the Ca²⁺ sensing proteins STIM1 and/or STIM2 [21-23]. Orai1 and SOCE are up-regulated by PI3K dependent signaling and thus sensitive to chorein [24]. Stimulators of Orai1 expression and SOCE include lithium [25-27], which may slow neurodegeneration [28-30].

Orai1 expression is up-regulated by serum & glucocorticoid inducible kinase SGK1 [31, 32] and the effect of lithium in ChAc neurons is reversed in the presence of SGK1 inhibitor GSK650394 [27]. SGK1 is partially effective by NFκB dependent up-regulation of Orai1 expression and by inhibition of Nedd4-2 triggered degradation of Orai1 protein [31, 32]. As chorein deficiency impairs activation of PI3K [1-3], it should interfere with PI3K/ SGK1/NFκB dependent Orai1 upregulation. The observed signaling contributes to the anti-apoptotic effect of PI3K which confers survival of a wide variety of cells including cancer cells [33-36] and neurons [37-40].

The present study explored whether the effect of lithium on neuronal Orai1 expression and SOCE in ChAc neurons requires functional NFκB. To this end, skin fibroblasts from ChAc patients were reprogrammed and differentiated to neurons and Orai1/STIM1/ STIM2 transcript levels, Orai1/STIM1/STIM2 protein abundance, SOCE and CRAC currents determined without or with prior lithium treatment in the absence or presence of NFκB inhibitor wogonin.

The study has been approved by the Ethical Commission of the University of Tübingen (598/2011). Informed consent was obtained from all participants and/or their legal guardian/s. Human dermal fibroblasts were isolated from ChAc patients (n = 3) and healthy volunteers (n = 3). Dermal fibroblasts were cultivated in fibroblast medium, consisting of DMEM (Biochrom, Berlin, Germany) supplemented with 10% fetal calf serum (FCS, Life technologies, Thermo Fisher Scientific, Waltham, Massachusetts) and 1% L-Glutamine (Biochrom). Induced pluripotent stem cells (iPSCs) were generated following a protocol published previously [41], with minor modifications. In brief, 1x10⁵ fibroblasts were electroporated (Nucleofector 2D, Lonza) with a total of 1 μg per plasmid carrying the sequences for hOCT4, hSOX2, hKLF4, hL-MYC and hLIN28. After cultivation in fibroblast medium for 1 day, 2 ng/ml FGF-2 (Peprotech) was supplemented to the medium. From day 3 on, cells were cultivated in Essential 8 (E8) medium containing 100 µM NaB (Sigma-Aldrich). The iPSC colonies were picked manually after 3 – 4 weeks and further expanded in Matrigel coated 6-well plates. At passage 7 – 10, iPSCs were characterized and frozen in E8 medium supplemented with 40% KOSR (Thermo Fisher Scientific), 10% DMSO (Sigma-Aldrich), and 1 µM Y-27632 (Selleckchem, Munich, Germany). Characterization of generated iPSCs included genomic validation via exclusion of plasmid-integration, SNParray analysis for genetic integrity, and resequencing of mutation site, as well as functional validation via confirmation of expression of pluripotency marker, and verification of the in vitro differentiation potential as described previously [42]. The opportunity to apply iPSC-derived neurons to model neurological disorders is a powerful tool to identify disease-relevant alterations in patient-specific neuronal cell types (reviewed in: [43]). A careful characterization of generated iPSCs as well as the establishment of a robust and reliable protocol to generate neurons is essential to provide consistent phenotypes. Additionally, using multiple patient and control cell lines is crucial to confirm the identified cellular phenotypes. In the present study we addressed these points by genomic and functional validation of all generated iPSCs, applying a standardized neuronal differentiation protocol yielding ß-III-tubulin/CTIP-2 positive neurons as well as verifying the observed phenotypes in three independent patient lines.

Cortical neurons were generated as described previously [44]. Briefly, neural induction of iPSCs was achieved by addition of dual SMAD inhibitors (10 µM SB431542 (Sigma-Aldrich) and 500 nM LDN-193189 (Sigma-Aldrich)) to 3N medium. Cells were collected at day 10 and further expanded by cultivation in 3N medium supplemented with 20 ng/ml FGF-2 for 2 days. From day 12 on, cells were cultivated in 3N medium with medium change every other day. Cell cultures were passaged at day 27 and replated appropriately for the specific assay (RNA/Protein isolation: 5x10⁵ cells per cm²; FACS analysis: 2.5x10⁵ per cm²; Ca²⁺ measurements: 5x10⁴ per cm2). Where indicated, 2 mM lithium (Sigma-Aldrich) and/or 50 µM wogonin (Sigma-Aldrich) was added to the medium 24 hours prior to measurements. Analysis was performed between day 37 and 41.

Transcript levels of Orai1, STIM1, STIM2 and house-keeping GAPDH were determined by RT-PCR as described previously [24, 45]. Total RNA was extracted in TriFast (Peqlab, Erlangen, Germany) according to the manufacturer’s instructions. After DNAse digestion reverse transcription of 2 µg RNA was performed using GoScript^TM Reverse Transcription System (Promega, Hilden, Germany) according to the manufacturer’s protocol. Real-time polymerase chain reaction (RT-PCR) amplification of the respective genes were set up in a total volume of 20 µl using 40 ng of cDNA, 500 nM forward and reverse primer and 2x GoTaq® qPCR Master Mix (Promega, Hilden, Germany) according to the manufacturer’s protocol. Cycling conditions were as follows: Initial denaturation at 95°C for 2 minutes, followed by 40 cycles of 95°C for 15 seconds, 55°C for 15 seconds and 68°C for 20 seconds. For amplification the following primers were used (5`-> 3` orientation):

GAPDH:

fw: TGAGTACGTCGTGGAGTCCAC;

rev: GTGCTAAGCAGTTGGTGGTG

Orai1:

fw: CGTATCTAGAATGCATCCGGAGCC;

rev: CAGCCACTATGCCTAGGTCGACTAGC

STIM1:

fw: CCTCGGTACCATCCATGTTGTAGCA

rev: GCGAAAGCTTACGCTAAAATGGTGTCT

STIM2:

fw: CAAGTTGCCCTGCGCTTTAT

rev: ATTCACTTTTGCACGCACCG

Specificity of PCR products was confirmed by analysis of a melting curve. Real-time PCR amplifications were performed on a CFX96 Real-Time System (Bio-Rad, Munich, Germany) and all experiments were done in duplicate. The house-keeping gene GAPDH (Glyceraldehyde 3-phosphate dehydrogenase) was amplified to standardize the amount of sample RNA.

Protein abundance of Orai1, STIM1/2 and GAPDH was determined by Western blotting as described previously [24, 45]. To this end, cells were centrifuged for 5 minutes at 240 g and 4°C. The pellet was washed twice with ice cold PBS and suspended in 40 μl ice-cold RIPA lysis buffer (Thermo Fisher Scientific, USA) containing Halt Protease and Halt Phosphatase Inhibitor Cocktail (Thermo Fisher Scientific, USA). Protein concentration was determined using the Bradford assay (BioRad, München, Germany). 100 µg of protein were solubilized in sample buffer at 95ºC for 10 min. The proteins were separated by 10% SDS-PAGE in Glycine-Tris buffer and electro-transferred onto nitrocellulose membranes for 90 min. After blocking with 5% milk in TBST at room temperature for 1 h, the membranes were incubated with primary anti-ORAI1 antibody (1: 1000, Proteintech), anti-STIM1 antibody (1: 1000, Cell Signaling), anti-STIM2 antibody (1: 1000, Cell Signaling) and anti-GAPDH antibody (1: 1000, Cell Signaling) at 4°C overnight. After washing (TBST), the blots were incubated with secondary anti-rabbit antibody conjugated with horseradish peroxidase (1: 2000, Cell Signaling) for 1 h at room temperature. Protein bands were detected after additional washes (TBST) with an ECL detection reagent (Amersham, Freiburg, Germany) and quantified with Quantity One Software (BioRad, München, Germany). To assign the right protein size we used Protein-Marker VI (Peqlab, Erlangen, Germany).

Fura-2 fluorescence was taken as a measure of cytosolic Ca²⁺ concentration ([Ca²⁺]_i), as described previously [24, 45]. For this purpose, cells were loaded with Fura-2/AM (2 µM, Invitrogen, Goettingen, Germany) for 20 min at 37°C. Cells were excited alternatively at 340 nm and 380 nm through an objective (Fluor 40×/1.30 oil) built in an inverted fluorescence microscope (Axiovert 100, Zeiss, Oberkochen, Germany). Emitted fluorescence intensity was recorded at 505 nm. Data were acquired using specialized computer software (Metafluor, Universal Imaging, Downingtown, USA). Cytosolic Ca²⁺ activity was estimated from the 340 nm/380 nm ratio. SOCE was determined by extracellular Ca²⁺ removal and subsequent Ca²⁺ re-addition in the presence of thapsigargin (1 µM, Invitrogen). For quantification of Ca²⁺ entry, the slope (delta ratio/s) and peak (delta ratio) were calculated following re-addition of Ca²⁺.

Experiments were performed with Ringer solution containing (in mM): 125 NaCl, 5 KCl, 1.2 MgSO₄, 2 CaCl₂, 2 Na₂HPO₄, 32 HEPES, 5 glucose, pH 7.4. To reach nominally Ca²⁺-free conditions, experiments were performed using Ca²⁺-free Ringer solution containing (in mM): 125 NaCl, 5 KCl, 1.2 MgSO₄, 2 Na₂HPO₄, 32 HEPES, 0.5 EGTA, 5 glucose, pH 7.4.

Ca²⁺ release activated channel (CRAC) currents (I_CRAC) were determined by whole cell patch clamp recording at room temperature in voltage-clamp, fast whole cell mode. Cells were continuously superfused through a flow system inserted into the dish. The bath was grounded via a bridge filled with the external solution. Borosilicate glass pipettes (2- to 4-MΩ resistance; Harvard Apparatus, UK) manufactured by a microprocessor-driven DMZ puller (Zeitz, Augsburg, Germany), were used in combination with a MS314 electrical micromanipulator (MW, Märzhäuser, Wetzlar, Germany). The currents were recorded by an EPC-9 amplifier (Heka, Lambrecht, Germany) and analyzed with Pulse software (Heka) and an ITC-16 Interface (Instrutech, Port Washington, NY). Currents were recorded at an acquisition frequency of 10 kHz and 3 kHz low-pass filtered. The pipette solution contained (in mM/l): 35 NaCl, 120 CsCl, 10 HEPES/CsOH, 10 ethylene glycol tetraacetic acid (EGTA) and 0.04 inositol 1, 4,5-triphosphate (Ins(1, 4,5)P₃ Enzo Life Sciences), pH 7.4. The external solution contained (in mM/l) 140 NaCl, 5 KCL, 10 CaCl₂, 20 glucose, 10 HEPES/NaOH, pH 7.4 [46].

Data are expressed as arithmetic means ± SEM. Statistical analysis was made by unpaired t-test or Mann-Whitney test, as appropriate. p < 0.05 was considered as statistically significant.

ChAc neurons were incubated for 24 hours in the absence or presence of 2 mM lithium. The Orai1 and STIM1/2 mRNA abundance was subsequently determined using qRT-PCR. As displayed in Fig. 1, Orai1 and STIM1/2 transcript levels were in ChAc neurons significantly increased by treatment with 2 mM LiCl. The effect of lithium on Orai1 and STIM1/2 transcript levels was abrogated by additional treatment with the NFκB inhibitor wogonin (50 µM). In the presence of lithium and wogonin, the Orai1 and STIM1/2 transcript levels were even significantly lower than the transcript levels of untreated ChAc neurons (Fig. 1). Wogonin alone tended to decrease Orai1 transcript levels and significantly decreased STIM1/2 transcript levels (Fig. 1).

Western blotting was employed for quantification of protein abundance. As illustrated in Fig. 2, the Orai1 and STIM1 protein expressions were significantly higher in ChAc neurons following 24 hours incubation in the presence than in the absence of 2 mM lithium (Fig. 2). The additional treatment with the NFκB inhibitor wogonin (50 µM) significantly decreased the Orai1 and STIM1/2 protein abundance. In the presence of both, lithium and wogonin, the Orai1 and STIM1/2 protein abundances were not significantly different from the Orai1 and STIM1/2 protein abundances in the absence of lithium and wogonin (Fig. 2).

Altered expressions of Orai1 and STIM1 were expected to be paralleled by respective changes of store operated Ca²⁺ entry (SOCE). Fura2 fluorescence was thus employed to quantify the cytosolic Ca²⁺ concentration ([Ca²⁺]_i). For determination of SOCE, the intracellular stores were emptied by exposure of the cells to the sarco-/endoplasmic reticulum Ca²⁺-ATPase (SERCA) inhibitor thapsigargin (1 µM) in the absence of extracellular Ca²⁺. Re-addition of extracellular Ca²⁺ in the continued presence of thapsigargin resulted in a sharp increase of [Ca²⁺]_i reflecting SOCE. As shown in Fig. 3B,C, emptying the intracellular Ca²⁺ stores with thapsigargin was followed by a transient increase in [Ca²⁺]_i. The increase tended to be higher in lithium treated than in untreated ChAc neurons, a difference, however, not reaching statistical significance. The increase of [Ca²⁺]i following re-addition of extracellular Ca²⁺ in the continued presence of thapsigargin was significantly higher in lithium treated than in untreated ChAc neurons (Fig. 3D,E). Treatment with the NFκB inhibitor wogonin (50 µM) significantly decreased SOCE both, in the absence and presence of lithium. In the presence of both, lithium and wogonin, SOCE was significantly lower than in untreated ChAc neurons (Fig. 3D,E).

The function of Orai1 was further quantified by measurement of Ca²⁺ release activated Ca²⁺ channel (CRAC) currents (I_CRAC) utilizing whole cell patch clamp recording. As shown in Fig. 4, I_CRAC was significantly higher in lithium treated than in untreated ChAc neurons. The additional treatment with the NFκB inhibitor wogonin (50 µM) significantly decreased I_CRAC (Fig. 4).

The present study sheds further light on the signaling contributing to the regulation of Orai1, STIM1 and STIM2 expression, Ca²⁺ release activated channel (CRAC) currents (I_CRAC) and store operated Ca²⁺ entry (SOCE) in chorea-acanthocytosis (ChAc) neurons. The results confirm the previous observations that lithium up-regulates Orai1and STIM1 expression, I_CRAC and SOCE in ChAc neurons [27]. Inhibition of NFκB inhibitor wogonin abrogates the stimulating effect of lithium on Orai1 and STIM1/2 transcript levels and protein abundance, SOCE and CRAC current. Wogonin significantly decreases Orai1 and STIM1/2 transcript levels and SOCE even in the absence of lithium indicating that NFκB is stimulating Orai1 and STIM1/2 expression as well SOCE independent of lithium. NFκB may thus play a permissive role for the effect of lithium. In any case, NFκB is apparently a stimulator of Orai1 and STIM1/2 expression as well SOCE in both, the presence and absence of lithium.

Stimulation of SOCE may result in oscillations of cytosolic Ca²⁺ activity ([Ca²⁺]_i) [47] with rapid increase of [Ca²⁺]_i due to intracellular Ca²⁺ release and activation of SOCE followed by rapid decrease of [Ca²⁺]_i due to SOCE inhibition and Ca²⁺ extrusion [48]. The repeated short increases of [Ca²⁺]_i are followed by activation of Ca²⁺ dependent transcription factors and reorganization of the actin filament network without cell injury resulting from sustained increases of [Ca²⁺]_i [49, 50].

Ca²⁺ oscillations trigger several cellular functions [51, 52] including entry into the S and the M phase of the cell cycle [53] and cell survival [54, 55]. The Orai isoforms [20] and their regulators STIM 1 or 2 [21] confer survival, proliferation, and migration of tumor cells [56-59] and neural stem/progenitor cells [60]. In contrast to Ca²⁺ oscillations, sustained increases of cytosolic Ca²⁺ activity trigger suicidal cell death [61-63].

Lithium has previously been shown to support survival of ChAc neurons [27]. Lithium has further been shown to favorably influence other neurodegenerative diseases including Huntington´s chorea, Alzheimer´s disease, Parkinson´s disease, amyotrophic lateral sclerosis as well as spinocerebellar ataxias type 1 and type 3 [28-30, 64, 65]. Mechanisms invoked in the neuroprotective effect of lithium include direct or Akt-mediated inhibition of glycogen synthase kinase GSK-3β, Akt-mediated inhibition of the proapoptotic forkhead box class O transcription factor Foxo3a and murine double minute (MDM), stimulation of production and activity of neuroprotective brain derived neurotrophic factor BDNF, up-regulation of antiapoptotic protein Bcl-2, as well as down-regulation of proapoptotic transcription factor p53, of the proapoptotic proteins Bad and Bax, of glutamate excitotoxicity, of calpain and of oxidative stress [30, 66].

The stimulating effect of NFκB on Orai1/STIM1 expression, I_CRAC and SOCE observed here could, at least in theory, impact on neuronal cell survival and contribute to the beneficial effect of lithium. The present observations, however, cannot be taken as evidence for an inhibitory effect of NFκB on neurodegeneration. NFκB expressed in microglial cells has previously been shown to participate in the orchestration of neuroinflammation and thus neurodegeneration [67-69] and NF-κB inhibitors are considered as potential treatment for neurodegenerative diseases [69]. It should be further kept in mind that neurons gained by reprogramming of skin fibroblasts and kept in cell culture do not necessarily respond to challenges in an identical way as neurons in situ. Thus, caution is warranted if the present results are extrapolated to native human physiology and pathophysiology.

In conclusion, pharmacological inhibition of NFκB markedly decreases Orai1 and STIM1 expression as well as I_CRAC and store operated Ca²⁺ entry in cultured ChAc neurons. It further abrogates the stimulating effect of lithium on Orai1 and STIM1 expression as well as I_CRAC and store operated Ca²⁺ entry in ChAc neurons. To which extent this effect impacts on ChAc and other neurodegenerative disorders, remains to be shown.

The authors acknowledge the meticulous preparation of the manuscript by Lejla Subasic and the technical support by Yvonne Schelling This work was supported by grants from Brigitte-Schlieben-Lange-Programm to L.P, by the Deutscher Akademischer Austauschdienst (DAAD) bi-nationally supervised Ph.D. Fellowship to I.S., by the Deutsche Forschungsgemeinschaft (La315-15) to F.L. and by the Open Access Publishing Fund, University of Tuebingen.

B.S., S.H., L.P., Z.H., I.S., T.M., A.M. performed experiments, C.S., L.S., F.L. designed the study and share last authorship for this paper, and F.L. drafted the manuscript. All authors corrected and approved the final version of the manuscript.

The sponsors had no role in study design, the collection, analysis and interpretation of data, in the writing of the report, and in the decision to submit the article for publication. All authors confirm that they have no competing financial interests to disclose.

B. Sukkar and S. Hauser contributed equally to this work and thus share first authorship.