A Type 2 Ryanodine Receptor Variant in the Helical Domain 2 Associated with an Impairment of the Adrenergic Response

Abstract

Catecholaminergic polymorphic ventricular tachycardia (CPVT) is triggered by exercise or acute emotion in patients with normal resting electrocardiogram. The major disease-causing gene is RYR2, encoding the cardiac ryanodine receptor (RyR2). We report a novel RYR2 variant, p.Asp3291Val, outside the four CPVT mutation hotspots, in three CPVT families with numerous sudden deaths. This missense variant was first identified in a four-generation family, where eight sudden cardiac deaths occurred before the age of 30 in the context of adrenergic stress. All affected subjects harbored at least one copy of the RYR2 variant. Three affected sisters were homozygous for the variant. The same variant was found in two additional CPVT families. It is located in the helical domain 2 and changes a negatively charged amino acid widely conserved through evolution. Functional analysis of D3291V channels revealed a normal response to cytosolic Ca²⁺, a markedly reduced luminal Ca²⁺ sensitivity and, more importantly, an absence of normal response to 8-bromo-cAMP and forskolin stimulation in both transfected HEK293 and HL-1 cells. Our data support that the D3291V-RyR2 is a loss-of-function RyR2 variant responsible for an atypical form of CPVT inducing a mild dysfunction in basal conditions but leading potentially to fatal events through its unresponsiveness to adrenergic stimulation.

Keywords: CPVT; RYR2; adrenergic stimulation; arrhythmia; calcium; sudden death.

Figure 1

Pedigrees of two families. (a) Pedigree of Family 1; (b) Pedigree of Family 2. Squares and circles indicate males and females, respectively, and slashes indicate deceased individuals. For Family 1, only deceased and living D3291V carriers were indicated; thirty healthy genotype negative carriers were excluded. Genotypes and phenotypes are indicated with black and grey boxes. SD: Sudden death, ASD: Aborted Sudden Death.

Figure 2

Abnormal ECGs during ESTs in D3291V heterozygous carriers. (a) EST performed in an asymptomatic untreated subject (III.14, family 1) after genetic diagnosis at the age of 17 showing bigeminy, bi-directional doublets, and ventricular tachycardia at 125 W; (b) Polymorphic ventricular tachycardia during EST in a 44-year-old subject treated with 80 mg of nadolol (III.2, family 1). Her daughter (IV.3) was also treated with 80mg of nadolol and developed PVT; (c) Holter recording of an untreated child (IV.13, family 1) at the age of 2 with AVB and bi-directional ventricular doublets at 110 bpm.

Figure 3

Location of the D3291V-RyR2 variant. (a) The D3291V variant is located in the HD2 outside the 4 mutation hotspots (red double arrows) where CPVT mutations are mostly found [39,40]. (b) Sequence alignment of RyR orthologues and paralogues. The Uniprot accession numbers are the following: Homo sapiens (RyR1 P21817, RyR2 Q92736, RyR3 Q15413), Pan troglodytes (RyR2 A0A2I3RIJ4), Sus scrofa (RyR2 F1RHM3), Mus musculus (RyR2 E9Q401), Gallus gallus (RyR2 AOA1D5PAZ1), Xenopus tropicalis (RyR1 F7E4CO), Lepisosteus oculatus (RyR2 W5NIQ4, RyR1 W5NAB3, RyR3 W5N4D0), Drosophila melanogaster (RyR Q24498), Musca domestica (RyR A0A1I8MDA8), Caenorhabditis elegans (I2HAA6_CAEEL, unc-68). Sequence alignment was done at http://www.ebi.ac.uk/Tools/msa/clustalo/ (accessed on 1 January 2018). The Aspartate 3291 is boxed in red. The variant N3308S (in blue) was identified in a CPVT patient [41].

Figure 4

Functional study of D3291V-RyR2 channel in HEK293 cells. (a) Representative traces of Ca²⁺ release under basal conditions and after cAMP treatment. (b) Percentages of oscillating WT-RyR2 (n = 29 dishes) and D3291V-RyR2 cells (n = 23 dishes) at different Ca²⁺ concentrations (0.1–1 mM) under basal conditions or challenged by 250 µM 8-bromo-cyclic AMP (cAMP). WT: black squares and dotted lines; WT + cAMP: black triangles and solid lines; D3291V: red squares and dotted lines; D3291V + cAMP: red triangles and solid lines. (c) Number of oscillations per min of WT-RyR2 (n = 300 cells) or D3291V-RyR2 cells (n = 265 cells) at different Ca²⁺ concentrations (0.1–2 mM) under basal conditions or challenged by 250 µM cAMP. (d) Percentages of oscillating WT-RyR2 (n = 29 dishes) (black squares) and D3291V-RyR2 (n = 23 dishes) (red circles) cells at different concentrations of Ca²⁺ (0.1–1 mM) challenged by 5 µM forskolin. (e) Number of oscillations per minute of WT-RyR2 (n = 300 cells) (black squares) or D3291V-RyR2 cells (n = 265 cells) (red circles) at different Ca²⁺ concentrations (0.1–2 mM) under basal condition or challenged by 5µM forskolin. (f) Representative Western blot of total proteins extracted from HEK293 cells transfected with RYR2-WT or RYR2-D3291V plasmids (n = 12). (g) Western blot quantification of RyR2 expression normalized to α-tubulin expression. Statistical significance denoted as WT compared to WT+ cAMP: * p < 0.05, ** p < 0.01, *** p < 0.001. WT + cAMP compared to D3291V + cAMP: ^# p < 0.05, ^## p < 0.01, ^### p < 0.001.

Figure 5

S2808 phosphorylation of both WT and D3291V in presence of cAMP: (a) Western blot of total proteins extracted from HEK293 cells transfected with eGFP-WT-hRyR2 or eGFP-D3291V-hRyR2 plasmids and treated or not with 250 μM 8-Bromo-cyclic AMP (cAMP) (n = 6) revealed with anti-RyR2 (A.a.) or anti-pSer2808 (A.b.). (b) Quantification of Ser2808 phosphorylation normalized to RyR2 total expression. ^# p < 0.05, WT + cAMP compared to D3291V + cAMP.

Figure 6

Dominant effect of D3291V in HL-1 cardiomyocytes: (a) Representative traces of Ca²⁺ release under basal conditions and after cAMP treatment. (b) Percentages of oscillating WT-RyR2 cells (n = 4 dishes) and D3291V-RyR2 cells (n = 5 dishes) at different Ca²⁺ concentrations (0.1, 0.5, 1 mM) under basal conditions or challenged by 250 µM cAMP. WT: black squares and dotted lines; WT + cAMP: black triangles and solid lines; D3291V: red squares and dotted lines; D3291V + cAMP: red triangles and solid lines. (c) Number of oscillations per minute of WT-RyR2 (n = 90 cells) or D3291V-RyR2 cells (n = 209 cells) at different Ca²⁺ concentrations (0.1, 0.5, 1 mM) under basal conditions or challenged by 250 µM cAMP. (d) Amplitude of the Ca²⁺ released at each spontaneous oscillation at different Ca²⁺ concentrations (0.1–1 mM) under basal conditions or challenged by 250 µM cAMP. (e) Time to peak at different Ca²⁺ concentrations (0.1–1 mM) under basal conditions or challenged by 250 µM cAMP. (f) Calcium transient duration at 50% (CTD50) at different Ca²⁺ concentrations (0.1–1 mM) under basal conditions or challenged by 250 µM cAMP. (g) Decay time 50% at different Ca²⁺ concentrations (0.1–1 mM) under basal conditions or challenged by 250 µM cAMP. (h) Representative trace of Ca²⁺ release of D3291V-RyR2 cells after cAMP treatment at 0.5 mM of extracellular Ca²⁺. WT compared to D3291V: ^§ p < 0.05. WT compared to WT + cAMP: * p < 0.05, ** p < 0.01, *** p < 0.001. D3291V compared to D3291V + cAMP: ^¥ p < 0.05. WT + cAMP compared to D3291V + cAMP: ^# p < 0.05, ^## p < 0.01, ^### p < 0.001.

Figure 7

Cytosolic and luminal Ca²⁺ regulation of the D3291V-RyR2 channels: (A) Single-channel recordings from WT-RyR2 (black) and D3291V-RyR2 (red) were recorded at a membrane holding potential of 0 mV at three different cytosolic free Ca²⁺ concentrations. Open events are upward deflections from zero current level (blue dotted line, c). The cytosolic solution contained 120 mM Tris and 250 mM HEPES (pH 7.4), and the cytosolic Ca²⁺ levels were adjusted with 0.2 mM BAPTA and 1 mM Dibromo-BAPTA in Cis. Trans solution contained 50 mM Ca²⁺ and 250 mM HEPES (pH 7.4). (B) Single-channel traces from WT-RyR2 (black) and D3291V-RyR2 (red) were recorded in symmetrical conditions of 250 mM CsCH₃O₃S, 20 mM HEPES, pH 7.4. In Trans, the desired luminal Ca²⁺ levels were adjusted with 1 mM BAPTA with the cytosolic free Ca²⁺ concentration set to 5 µM in Cis. Channel activity was taken at +20 mV holding potential. (C) Open probabilities of WT-RyR2 (black squares) and D3291V-RyR2 (red squares) obtained, under the same conditions as in A, and challenged with increasing free [Ca²⁺] in Cis. The sigmoidal curves resulted from the fitting using a Hill function: y = Vmax × xⁿ/(kⁿ + xⁿ) where Vmax is the max velocity, k is Michaelis constant, and n is the number of cooperative sites. At the same concentrations of free Ca²⁺, the open probability of WT-RyR2 and D3291V-RyR2 channels were not significantly different. (ANOVA, one-way analysis at p < 0.05). The EC_50′s for the WT-RyR2 and D3291V-RyR2 were 438 ± 75nM (n = 2–7) and 320 ± 41nM (n = 3–5), respectively. (D) Open probabilities of WT-RyR2 (black squares) and D3291V-RyR2 (red squares) obtained under the same conditions as in C, and challenged with increasing free [Ca²⁺] in Trans. The open probabilities of WT and D3291V-RyR2 channels were significantly different at low luminal concentrations of free Ca²⁺ (ANOVA one-way test at p < 0.05). Currents traces were filtered at 800 Hz.