Dystonia caused by ANO3 variants is due to attenuated Ca2+ influx by ORAI1

Abstract

Background: Dystonia is a common neurological hyperkinetic movement disorder that can be caused by mutations in anoctamin 3 (ANO3, TMEM16C), a phospholipid scramblase and ion channel. We previously reported patients that were heterozygous for the ANO3 variants S651N, V561L, A599D and S651N, which cause dystonia by unknown mechanisms.

Methods: We applied electrophysiology, Ca²⁺ measurements and cell biological methods to analyze the molecular mechanisms that lead to aberrant intracellular Ca²⁺ signals and defective activation of K⁺ channels in patients heterozygous for the ANO3 variants.

Results: Upon expression, emptying of the endoplasmic reticulum Ca²⁺ store (store release) and particularly store-operated Ca²⁺ entry (SOCE) were strongly inhibited, leading to impaired activation of K_Ca3.1 (KCNN) K⁺ channels, but not of Na⁺-activated K⁺ channels (K_Na; SLO2). The data provide evidence for a strongly impaired expression of store-operated ORAI1 Ca²⁺ influx channels in the plasma membrane of cells expressing ANO3 variants.

Conclusions: Dysregulated Ca²⁺ signaling by ANO3 variants may impair the activation of K⁺ channels in striatal neurons of the brain, thereby causing dystonia. Furthermore, the data provide a first indication of a possible regulation of protein expression in the plasma membrane by ANO3, as has been described for other anoctamins.

Keywords: ANO3; Anoctamin 3; Ca2+ signaling; Dystonia; K+ channels; TMEM16C.

Fig. 1

ANO3 variants reduce ER store filling and store-operated Ca²⁺ influx. A Measurement of [Ca²⁺]_i by Fura2 in mock-transfected HEK293 cells and HEK293 cells expressing wtANO3 (wt), or the ANO3 variants V561L, S651N, A599D, or S116L. Removal of extracellular Ca²⁺ by a Ca²⁺-free Ringer-like solution (Ca²⁺-free) and application of the SERCA-inhibitor cyclopiazonic acid (CPA, 10 µM). Subsequent replacement of Ca²⁺ free solution by Ringer solution. B Summary of changes in [Ca²⁺]_i induced by CPA (store emptying; Store) and removal of Ca²⁺ free solution (Ca²⁺ influx; SOCE). Mean ± SEM (number of experiments). ^#significant difference when compared to mock (p < 0.05; ANOVA)

Fig. 2

The S651N variant inhibits purinergic activation of SK4 K⁺ channels. A Original whole cell current overlays before and after stimulation of HEK293 cells expressing Ca²⁺-sensitive SK4 K⁺ channels with ATP (10 µM), and additional inhibition of SK4 channels by TEA/Ba (10 mM/5 mM). Activation of SK4 currents was not affected by coexpression of wt-ANO3 but was attenuated in the presence of S651N-ANO3. B Current–voltage (I/V) relationships corresponding to the whole cell currents shown in A. C Summary of the current densities corresponding to the data shown in A and B. Mean ± SEM (number of experiments). *significant increase by ATP (p < 0.05, paired t-test). ^#reduced activation when compared to SK4 only (p < 0.05; ANOVA)

Fig. 3

Currents produced by the K⁺ channel SLO2 are not activated by ATP and are not potentiated by wtANO3 or the ANO3 variants S651N and V561L. A Whole cell current overlays, B corresponding I/V curves, and C current densities in HEK293 cells overexpressing SLO2 together with wtANO3, S651N-ANO3 or V561L-ANO3. Cells were stimulated with ATP (10 µM). Mean ± SEM (number of experiments)

Fig. 4

Activation of SLO2 by bithionol is not affected by the ANO3 variants S651N or V561L. A Whole cell current overlays, B corresponding I/V curves, and C current densities in HEK293 cells overexpressing SLO2 alone or together with S651N-ANO3 or V561L-ANO3. Mock-transfected cells served as controls. Cells were stimulated with the SLO2-activator bithionol (10 µM). ATP (10 µM). Mean ± SEM (number of experiments). *significant increase due to bithionol (p < 0.05, paired t-test)

Fig. 5

Basal and activated phospholipid scrambling is upregulated in ANO3 variants. A 4-quadrant dot blots of flow cytometry in mock transfected HEK293 cells or in cells expressing wtANO3. Cells were analyzed under control conditions (basal scrambling) and after incubation with ionomycin (Iono; 10 µM, 20 min), which strongly enhanced exposure of annexin V (AnxV). B Summary of % annexin V (AnxV) positive cells before and after stimulation with ionomycin. Increase of AnxV positive cells by ionomycin (^#unpaired t-test (p < 0.05), which was augmented in wtANO3 expressing cells (^§unpaired t-test, p < 0.05). C Phospholipid scrambling in nonstimulated (basal AnxV) cells expressing wtANO3 or ANO3 variants. ^#significant difference when compared to wtANO3 (p < 0.05; ANOVA). D Summary of the knockdown of ANO6-expression by siRNA-ANO6. E Additive effects on ionomycin-induced scrambling (AnxV positivity) by endogenous ANO6 and overexpressed anoctamins (wtANO3, A599N-ANO3, and S1116L-ANO3). The pronounced scrambling activity of V561L-ANO3 and S651N-ANO3 fully compensated for the lack of ANO6 in siANO6 treated cells. Mean ± SEM (number of experiments). ^#significant inhibition by siRNA-knockdown of ANO6-expression. (p < 0.05; unpaired t-test)

Fig. 6

Plasma membrane expression of ORAI1 is inhibited in the presence of S651N-ANO3. A Overexpression of ORAI1 alone or coexpression of ORAI1 with wtANO3 (WT) or the ANO3 variant S651N. Bar = 20 µm. B Summary of ORAI1 membrane expression as detected by fluorescence intensity (arbitrary units; au) suggests no difference for the different conditions. Stainings were performed as triplicates. C Biotinylation of membrane proteins indicates strongly reduced expression of ORAI1 in the plasma membrane of ORAI1/S651N-ANO3 coexpressing cells (unbnd, unbound ORAI1 protein; biotin, biotinylated ORAI1 protein. Experiments were performed as triplicates. Mean ± SEM (number of experiments). D Ratios for ORAI1 biotinylation in the presence of wtANO3 or ANO3 variants (as shown in C and Additional file 3)

Fig. 7

Proposed disease mechanism for dystonia caused by ANO3 variants. Wild type ANO3 expressed in the plasma membrane (PM) and in intracellular membranous compartments such as intracellular vesicles of striatal cells, allows for trafficking of ORAI1 Ca²⁺ channels to the PM, possibly by means of its phospholipid scramblase function. Thus, store-operated Ca²⁺ entry (SOCE) through ORAI1 activated through STIM1 causes proper filling of the endoplasmic reticulum (ER) Ca²⁺ store via the Ca²⁺ pump SERCA. Ca²⁺ release from the ER during activation of striatal cells causes activation of Ca²⁺ dependent KCNN K⁺ channels and repolarization of the membrane voltage to largely negative values (e.g. −80 mV). In striatal cells expressing mutant ANO3 variants, ORAI1 function, Ca²⁺ influx and ER store filling and activation of K⁺ channels are compromised causing reduced hyperpolarization (e.g. – 60 mV). This may lead to hyperexcitability and dystonia