XB-ART-55461Brain December 1, 2018; 141 (12): 3308-3318.
Hypokalaemic periodic paralysis is a rare genetic neuromuscular disease characterized by episodes of skeletal muscle paralysis associated with low serum potassium. Muscle fibre inexcitability during attacks of paralysis is due to an aberrant depolarizing leak current through mutant voltage sensing domains of either the sarcolemmal voltage-gated calcium or sodium channel. We report a child with hypokalaemic periodic paralysis and CNS involvement, including seizures, but without mutations in the known periodic paralysis genes. We identified a novel heterozygous de novo missense mutation in the ATP1A2 gene encoding the α2 subunit of the Na+/K+-ATPase that is abundantly expressed in skeletal muscle and in brain astrocytes. Pump activity is crucial for Na+ and K+ homeostasis following sustained muscle or neuronal activity and its dysfunction is linked to the CNS disorders hemiplegic migraine and alternating hemiplegia of childhood, but muscle dysfunction has not been reported. Electrophysiological measurements of mutant pump activity in Xenopus oocytes revealed lower turnover rates in physiological extracellular K+ and an anomalous inward leak current in hypokalaemic conditions, predicted to lead to muscle depolarization. Our data provide important evidence supporting a leak current as the major pathomechanism underlying hypokalaemic periodic paralysis and indicate ATP1A2 as a new hypokalaemic periodic paralysis gene.
PubMed ID: 30423015
PMC ID: PMC6262219
Article link: Brain
Genes referenced: atp1a2
GO Terms: potassium channel activity
Disease Ontology terms: hypokalemic periodic paralysis
OMIMs: HYPOKALEMIC PERIODIC PARALYSIS, TYPE 1; HOKPP1
Article Images: [+] show captions
|Figure 1. Muscle pathology and genetic analysis of the hypoPP patient. (A) Histological examination of a biceps brachii muscle biopsy performed at age 2 years and 3 months. (i) Haematoxylin and eosin staining showed variation in fibre diameter, increased internal nuclei (black arrows), a necrotic fibre (white arrow) and subtle endomysial fibrosis (also seen in iii) (ii) Myosin ATPase histochemistry at pH 9.4 indicated type I fibre predominance (pale stained fibres). (iii) Gomori trichrome staining. (B) Sequence chromatogram demonstrating the heterozygous mutation c.2336 G>A; p.S779N in the proband (above) and sequence alignment of human Na+/K+-ATPase alpha subunits. ATP1A2-S779 and analogous serine residues are underlined. (C) Structural context of S779. Left: An overview of the Na+/K+-ATPase structure with the alpha subunit in grey, the beta subunit in blue and the gamma subunit in red. The two potassium ions are yellow spheres, S779 is in orange stick, and the ion-coordinating transmembrane helices are light cyan (M4), wheat (M5) and light blue (M6). The boundaries of the membrane are indicated by horizontal lines. Right: Close-ups of the ion binding sites viewed from the extracellular side, top with two potassium ions (yellow), bottom with three sodium ions (violet). S779 is close to ion binding sites I and III. The figure was made using PDB structures 2ZXE (potassium bound) and 3WGU (sodium bound).|
|Figure 2. Steady state and transient Na+ currents of WT and p.S779N pumps in the absence of extracellular K+. (A) Representative raw current traces in absence (left) and presence (middle) of 10 mM ouabain and the ouabain-sensitive currents (right) in wild-type (black) and p.S779N (red) pumps in response to voltage steps from −160 to +60 mV, in 20-mV increments. Steady state currents at each voltage were measured in the last 50 ms of the 200 ms stimulus, indicated by dotted lines. (B) Average ouabain-sensitive steady state leak currents. Slope conductance between −140 mV and 0 mV was 0.75 ± 0.08 nA/mV for the p.S779N pump (n = 27) while it was close to 0 for the wild-type pump (n = 29). (C) Charge-voltage relationships. Na+ charge transfer was determined from the integral of the first 50 ms of the current trace at −30 mV following the steps to test voltages. Individual QV curves were fit by a standard Boltzmann function and normalized to their respective fits. V1/2: p.S779N −95.4 ± 7.1 mV, n = 25; wild-type 2.6 ± 1.7 mV, n = 29; slope: p.S779N 66.8 ± 4.8 mV, wild-type 28.9 ± 0.7 mV, P < 10−5, unpaired t-test. (D) Rate constants of transient currents at the onset of the stimulus. Solid line represents a Boltzmann fit and yielded the overall forward/backward rate constants (p.S779N, 773.9 ± 27.7/542.6 ± 12.7 s−1, n = 22 wild-type 102.3 ± 3.6/882.2 ± 32.9 s−1, n = 26; P < 0.0001, unpaired t-test).|
|Figure 3. Forward pumping currents of WT and p.S779N α 2 at various extracellular [K+]o. (A) Example current traces of wild-type (black) and mutant (red) pumps in response to 200 ms steps from −160 to +40 mV, from a holding voltage of −30 mV in absence (left) and presence (middle) of 15 mM K+, and in presence of 15 mM K+ and 10 mM ouabain (right). K+-induced pump currents were isolated off-line by subtraction of currents in the presence of ouabain 10 mM. (B) Current-voltage relationship of steady state ouabain-sensitive currents measured in last 50 ms of each step (p.S779N n = 17; wild-type n = 20; P < 0.0001 two-way ANOVA). (C) Current-voltage relationship of ouabain-sensitive currents normalized to total Na+ charge transfer (Qtot) in absence of K+, as a measure of functional protein expression (p.S779N n = 7; wild-type n = 13; P = 0.001, two-way ANOVA). Qtot was estimated in a subset of cells where lower amplifier gain was used to minimize current saturation because of capacitive artefacts. Ten traces were averaged to reduce the noise and then used to determine the span of the Boltzman curve (Qtot: S779N 5.4 ± 0.9 nC, n = 7; wild-type 5.5 ± 0.7 nC, n = 13). (D) Extracellular apparent K+ affinity of the ATPases: current-voltage relationships of wild-type (i) and mutant pumps (ii) in various [K+]o conditions normalized to the outward current at +20 mV in 15 mM K+ (p.S779N n = 5–6; wild-type n = 3–5). (E) Concentration-response curves obtained from the data in D at three representative voltages. Dotted lines represent fits to a modified Hill equation (wild-type EC50 at −100, −60, 0 mV: 3.5 ± 0.3, 2.2 ± 0.2, 1.3 ± 0.1 mM, n = 5; P < 0.0001 repeated measures ANOVA; p.S779N EC50 at −100, −60, 0 mV: 6.9 ± 0.4, 6.9 ± 0.2, 7.5 ± 0.7 mM, n = 5; P = 0.7). Curves obtained from separate experiments were normalized to the top and bottom values of the fit and averaged across oocytes. Hill coefficients were voltage independent and did not change significantly between pump variants: wild-type 1.54 ± 0.06, S779N 1.56 ± 0.06 (P = 0.3, two-way ANOVA across the voltages). (F) Overall voltage dependence of apparent [K+] affinity (wild-type n = 3–5; p.S779N n = –5, P < 0.001 two-way ANOVA).|
|Figure 4. Ionic contributions to the p.S779N ATPase leak current. (A) Representative current traces in 0 [K+]o for wild-type (black) and p.S779N (red) illustrating changes in leak currents with varying [H+]o. (B) Steady state current-voltage relationships in different pH conditions normalized to current at +20 mV in pH 7.4 (n = 4 for both). Linear fits to raw leak currents (not shown) in p.S779N indicate an increase in slope conductance with acidification from 0.54 ± 0.1 to 0.85 ± 0.2 nA/mV (P = 0.03, paired t-test) and a shift in Erev from −12.7 ± 2.5 to −2.2 ± 2.6 mV (P = 0.01). For pH 8.2 the slope was 0.43 ± 0.1 (P = 0.15) and Erev = −17.8 ± 2.2 mV (P = 0.06). (C) Increase in inward leak currents in p.S779N pump with acidification in the presence of 15 mM [K+]o; comparison with linear leak in Na+-only (pH 7.4) conditions (n = 4). Currents have been normalized to amplitude at +20 in pH 7.4. (D) Average steady state currents for wild-type (left) or p. S779N (right) pumps in control conditions (115 mM Na+, 0 K+) and when 86 mM Na+ is substituted with NMDG+. Currents have been normalized to control value at +20 mV (wild-type n = 6, P = 0.008; p.S779N n = 7, P = 0.001 repeated measures ANOVA). For the mutant pump, Erev shifted from −16.2 ± 1.5 mV in Na+ to −48.8 ± 10 mV in Na/NMDG+ (n = 7, P = 0.016, Wilcoxon paired t-test).|