Dysfunction of NaV1.4, a skeletal muscle voltage-gated sodium channel, in sudden infant death syndrome: a case-control study

Abstract

Background: Sudden infant death syndrome (SIDS) is the leading cause of post-neonatal infant death in high-income countries. Central respiratory system dysfunction seems to contribute to these deaths. Excitation that drives contraction of skeletal respiratory muscles is controlled by the sodium channel NaV1.4, which is encoded by the gene SCN4A. Variants in NaV1.4 that directly alter skeletal muscle excitability can cause myotonia, periodic paralysis, congenital myopathy, and myasthenic syndrome. SCN4A variants have also been found in infants with life-threatening apnoea and laryngospasm. We therefore hypothesised that rare, functionally disruptive SCN4A variants might be over-represented in infants who died from SIDS.

Methods: We did a case-control study, including two consecutive cohorts that included 278 SIDS cases of European ancestry and 729 ethnically matched controls without a history of cardiovascular, respiratory, or neurological disease. We compared the frequency of rare variants in SCN4A between groups (minor allele frequency <0·00005 in the Exome Aggregation Consortium). We assessed biophysical characterisation of the variant channels using a heterologous expression system.

Findings: Four (1·4%) of the 278 infants in the SIDS cohort had a rare functionally disruptive SCN4A variant compared with none (0%) of 729 ethnically matched controls (p=0·0057).

Interpretation: Rare SCN4A variants that directly alter NaV1.4 function occur in infants who had died from SIDS. These variants are predicted to significantly alter muscle membrane excitability and compromise respiratory and laryngeal function. These findings indicate that dysfunction of muscle sodium channels is a potentially modifiable risk factor in a subset of infant sudden deaths.

Funding: UK Medical Research Council, the Wellcome Trust, National Institute for Health Research, the British Heart Foundation, Biotronik, Cardiac Risk in the Young, Higher Education Funding Council for England, Dravet Syndrome UK, the Epilepsy Society, the Eunice Kennedy Shriver National Institute of Child Health & Human Development of the National Institutes of Health, and the Mayo Clinic Windland Smith Rice Comprehensive Sudden Cardiac Death Program.

Figure 1

The location of mutations in the NaV1.4 channel The location of mutations in the NaV1.4 channel Transmembrane helices S1–S6 are labelled in domains I–IV. The S4 helices contain positively charged arginine residues (+). The S1–S4 helices form voltage-sensing domains. The S5–S6 helices are pore-forming. The variants in the SIDS cohort (from N-terminus to C-terminus) are: Ser682Trp, Gly859Arg, Val1442Met, Arg1463Ser, Met1493Val, and Glu1520Lys. The variants in the control cohort (from N-terminus to C-terminus) are: Arg179Gln, Arg190Trp, Leu227Phe, Asp334Asn, Gly863Arg, Ala870Thr, Met897Lys, and Val1590Ile. SIDS=sudden infant death syndrome.

Figure 2

Activation properties of NaV1.4 variants (A) Representative current traces of wild type, and Glu1520Lys channels in response to test voltages in 10 mV increments from −100 mV to + 50 mV (the voltage protocol is shown on the right). Dashed lines indicate 0 current level. (B,C) Peak current density in response to test voltages ranging from −100 to + 50 mV in 10 mV increments is plotted against the test voltage for variants in the SIDS cohort (B) and in controls (C). Current densities of Glu1520Lys (orange) and Arg1463Ser (blue) were significantly lower than wild type (black; appendix p 6–7). Data for the other variants that did not differ significantly from wild type are shown in grey. (D,E) Voltage dependence of activation. Normalised conductance (peak current/[test voltage – reversal voltage]) is plotted against the test voltage for variants in the SIDS cohort (D) and in controls (E). Individual data were normalised to maximum and minimum amplitude of the Boltzmann fit and averaged. The solid lines represent the fit of Boltzmann equation to mean data. Voltage of half-maximal activation (V_1/2) did not differ significantly from wild type (black) for any of the variants (appendix p 6–7) and the data are illustrated with grey symbols. The statistical analysis is shown in the appendix (p 6).

Figure 3

Fast inactivation properties of NaV1.4 variants (A–C) Voltage dependence of fast inactivation. (A) Representative current traces in response to tail voltage step to −10 mV following 150 ms pre-pulse voltage steps ranging from −150 mV to 0 mV in 10 mV increments for wild type, Arg1463Ser, and Val1442Met variants. First 5 ms of the voltage step to −10 mV are shown. Current response following pre-pulse step to −60 mV is highlighted in red. The voltage protocol is shown to the right. Dashed lines indicate 0 current level. (B,C) The peak tail current amplitude at −10 mV is plotted against the pre-pulse voltage for variants in the SIDS cohort (B) and in controls (C). Individual data were normalised to maximum and minimum values of the Boltzmann equation and averaged. Voltage of half-maximal inactivation was shifted significantly to more hyperpolarised voltages for Val1442Met channels (red) and to more depolarised voltages for Arg1463Ser channels (blue). The solid lines represent the fit of Boltzmann equation to mean data for wild type, Val1442Met and Arg1463Ser channels. The data for the variants that did not differ from wild type channels are shown in grey. (D–F) Recovery rate from fast inactivation. (D) Representative current traces for wild type and Arg1463Ser channels illustrating the recovery from inactivation. Channels were inactivated by a 10 ms voltage step to 0 mV and then stepped to recovery voltage of −80 mV for an increasing duration of time. A second voltage step to 0 mV was then applied to see how much the channels had recovered from fast inactivation. Only traces with 0, 1, 2, 5, 10, and 20 ms duration at recovery voltage of −80 mV are shown. The voltage protocol is shown to the right. (E,F) The peak current amplitude during the second step (P2) is divided by the peak current amplitude during the first voltage step (P1) and plotted against the duration of the recovery period at −80 mV. Data for variants in the SIDS cohort (E) and in controls (F) is colour coded as in (B) and (C). The solid lines represent fit of exponential function to the mean data. (G–I) Rate of fast inactivation. (G) Time constant (τ) of inactivation at voltages ranging from −20 mV to +20 mV for wild type (black) and Ser682Trp (magenta) channels. The voltage protocol was as in figure 2A. Only the time constant of the fast component that carries roughly 95% of the amplitude of the inactivating current is analysed. (H) Representative current traces at −20 mV to +20 mV for wild type (top) and Ser682Trp (bottom) channels, showing the slower inactivation of the Ser682Trp variant. (I) Overlay of mean normalised current traces for wild type (black) and Ser682Trp (magenta) channels in response to voltage step to 0 mV.

Figure 3

Fast inactivation properties of NaV1.4 variants (A–C) Voltage dependence of fast inactivation. (A) Representative current traces in response to tail voltage step to −10 mV following 150 ms pre-pulse voltage steps ranging from −150 mV to 0 mV in 10 mV increments for wild type, Arg1463Ser, and Val1442Met variants. First 5 ms of the voltage step to −10 mV are shown. Current response following pre-pulse step to −60 mV is highlighted in red. The voltage protocol is shown to the right. Dashed lines indicate 0 current level. (B,C) The peak tail current amplitude at −10 mV is plotted against the pre-pulse voltage for variants in the SIDS cohort (B) and in controls (C). Individual data were normalised to maximum and minimum values of the Boltzmann equation and averaged. Voltage of half-maximal inactivation was shifted significantly to more hyperpolarised voltages for Val1442Met channels (red) and to more depolarised voltages for Arg1463Ser channels (blue). The solid lines represent the fit of Boltzmann equation to mean data for wild type, Val1442Met and Arg1463Ser channels. The data for the variants that did not differ from wild type channels are shown in grey. (D–F) Recovery rate from fast inactivation. (D) Representative current traces for wild type and Arg1463Ser channels illustrating the recovery from inactivation. Channels were inactivated by a 10 ms voltage step to 0 mV and then stepped to recovery voltage of −80 mV for an increasing duration of time. A second voltage step to 0 mV was then applied to see how much the channels had recovered from fast inactivation. Only traces with 0, 1, 2, 5, 10, and 20 ms duration at recovery voltage of −80 mV are shown. The voltage protocol is shown to the right. (E,F) The peak current amplitude during the second step (P2) is divided by the peak current amplitude during the first voltage step (P1) and plotted against the duration of the recovery period at −80 mV. Data for variants in the SIDS cohort (E) and in controls (F) is colour coded as in (B) and (C). The solid lines represent fit of exponential function to the mean data. (G–I) Rate of fast inactivation. (G) Time constant (τ) of inactivation at voltages ranging from −20 mV to +20 mV for wild type (black) and Ser682Trp (magenta) channels. The voltage protocol was as in figure 2A. Only the time constant of the fast component that carries roughly 95% of the amplitude of the inactivating current is analysed. (H) Representative current traces at −20 mV to +20 mV for wild type (top) and Ser682Trp (bottom) channels, showing the slower inactivation of the Ser682Trp variant. (I) Overlay of mean normalised current traces for wild type (black) and Ser682Trp (magenta) channels in response to voltage step to 0 mV.