Mutations in ATP1A1 Cause Dominant Charcot-Marie-Tooth Type 2

Abstract

Although mutations in more than 90 genes are known to cause CMT, the underlying genetic cause of CMT remains unknown in more than 50% of affected individuals. The discovery of additional genes that harbor CMT2-causing mutations increasingly depends on sharing sequence data on a global level. In this way-by combining data from seven countries on four continents-we were able to define mutations in ATP1A1, which encodes the alpha1 subunit of the Na⁺,K⁺-ATPase, as a cause of autosomal-dominant CMT2. Seven missense changes were identified that segregated within individual pedigrees: c.143T>G (p.Leu48Arg), c.1775T>C (p.Ile592Thr), c.1789G>A (p.Ala597Thr), c.1801_1802delinsTT (p.Asp601Phe), c.1798C>G (p.Pro600Ala), c.1798C>A (p.Pro600Thr), and c.2432A>C (p.Asp811Ala). Immunostaining peripheral nerve axons localized ATP1A1 to the axolemma of myelinated sensory and motor axons and to Schmidt-Lanterman incisures of myelin sheaths. Two-electrode voltage clamp measurements on Xenopus oocytes demonstrated significant reduction in Na⁺ current activity in some, but not all, ouabain-insensitive ATP1A1 mutants, suggesting a loss-of-function defect of the Na⁺,K⁺ pump. Five mutants fall into a remarkably narrow motif within the helical linker region that couples the nucleotide-binding and phosphorylation domains. These findings identify a CMT pathway and a potential target for therapy development in degenerative diseases of peripheral nerve axons.

Keywords: ATP1A1; CMT; Charcot-Marie-Tooth; Mendelian disease; Na(+),K(+) ATPase; axonal neuropathy; genetic matchmaking.

Figure 1

Pedigrees of Families Carrying Mutations in ATP1A1 Autosomal-dominant CMT-affected families show segregation of ATP1A1 variants: family 1 (Czech Republic), family 2 (Italy), family 3 (USA1), family 4 (USA2), family 5 (Australia1), family 6 (Australia2), and family 7 (Korea).

Figure 2

ATP1A1 Mutations Identified in CMT-Affected Families (A) Schematic depicting the position of pathogenic substitutions in the ATP1A1 protein and their associated locations around the globe. (B) ATP1A protein paralogs alignment showing that the ATP1A1 substitutions are located at highly conserved residues across the four different paralogs. (C) 3D structural model of ATP1A1. (D) Electron micrograph of a sural nerve biopsy of affected individual from family 7 (Korea) showing typical signs of chronic axonal degeneration.

Figure 3

Genetic and Functional Studies of ATP1A1 (A and B) Mutational constraint analysis of ATP1A genes and known CMT-associated genes. ATP1A1, ATP1A2, and ATP1A3 have high missense constraint scores (A) and pLI (probability of LoF intolerance) (B); obtained from the ExAC browser. (C) Dominant CMT-associated genes have significantly higher constraint (more intolerant) scores for missense variants than recessive CMT-associated genes. (D) Two electrodes voltage clamp (TEVC) recordings at −50 mV from different CMT-associated mutations in the α1 subunit. For comparison, currents through wild-type α1 are shown in black. Dashed lines represent zero current. Summary of currents from the CMT-associated mutations shown are normalized to WT currents (mean ± SEM; n = 5–7; p < 0.05). (E) Ouabain survival assay of U2OS cells treated with 0.5 μM ouabain. Cell viability represents the luminescence values obtained from the CellTiter-Glo assay (n = 8; t test p < 0.05 compared to oua-WT-ATP1A1). (F) Distinct localization of α1 and α3 in myelinated axons. These are confocal images of teased fibers from an adult rat, immunostained with a mouse monoclonal antibody against α3 (XVIF9-G10; red), and a rabbit antiserum against α1 (NASE; green), as indicated. Two myelinated axons (1,3) have strong α3 staining (and weak α1 staining), and a myelinated axon (4) has strong α1 staining (and weak α3 staining). The incisures of myelin sheaths are α1-positive (inset). A Remak bundle (2) is α1-positive and α3-negative. Apposed arrowheads mark nodes of Ranvier. Scale bar: 10 μm.