A recurrent missense variant in ITPR3 causes demyelinating Charcot-Marie-Tooth with variable severity

Abstract

Charcot-Marie-Tooth (CMT) disease is a neuromuscular disorder affecting the peripheral nervous system. The diagnostic yield in demyelinating CMT (CMT1) is typically ∼80%-95%, of which at least 60% is due to the PMP22 gene duplication. The remainder of CMT1 is more genetically heterogeneous. We used whole exome and whole genome sequencing data included in the GENESIS database to investigate novel causal genes and mutations in a cohort of ∼2670 individuals with CMT neuropathy. A recurrent heterozygous missense variant p.Thr1424Met in the recently described CMT gene ITPR3, encoding IP3R3 (inositol 1,4,5-trisphosphate receptor 3), was identified. This previously reported p.Thr1424Met change was present in 33 affected individuals from nine unrelated families from multiple populations, representing an unusual recurrence rate at a mutational hotspot, strengthening the gene-disease relationship (gnomAD v4 allele frequency 1.76 × 10-6). Sanger sequencing confirmed the co-segregation of the CMT phenotype with the presence of the mutation in autosomal dominant and de novo inheritance patterns, including a four-generation family with multiple affected second-degree cousins. Probands from all families presented with slow nerve conduction velocities, matching the diagnostic category of CMT1. Remarkably, we observed a uniquely variable clinical phenotype for age at onset and phenotype severity in p.Thr1424Met carrying patients, even within families. Finally, we present data supportive of a dominant-negative effect of the p.Thr1424Met mutation with associated changes in protein expression in patient-derived cells.

Keywords: demyelinating Charcot-Marie-Tooth disease; inositol 1,4,5-trisphosphate; next-generation sequencing.

Figure 1

Pedigrees of families carrying the p.Thr1424Met variant. Family pedigrees demonstrating dominant (Families 1–4 and 6–9) and de novo (Family 5) inheritance of the p.Thr1424Met variant in ITPR3. Carriers are shown by ‘+/−’, non-carriers by ‘−/−’. Individuals indicated with a number sign show very mild signs of neuropathy without fulfilling the clinical diagnosis of Charcot-Marie-Tooth disease, hearsay affected individuals are indicated in grey.

Figure 2

Intrafamilial variability of phenotype associated with the p.Thr1424Met ITPR3 variant. (A) Demonstrated by clinical pictures of four individuals from Family 1. (B) Plotted by Charcot-Marie-Tooth Disease Examination Score (CMTES-2) versus age (in years); higher score reflects a more severe disease. Individuals from the same family shown in the same colour, demonstrating the overall associating between increased age and disease severity (r² = 0.475, P < 0.001), meanwhile showing the extreme variability between aged-matched individuals (red box as example). Solid line = linear regression; dotted line = standard deviation.

Figure 3

Nerve biopsy and IP₃R3 expression in Schwannoma, induced pluripotent stem cell-derived neurons and p.Thr1424Met patient fibroblasts. (A) Epoxy section of a radial nerve biopsy of Individual 5:II:1 (age 21 years), showing severe loss of myelinated large axons, and most myelinated axons (red asterisks) are partially surrounded by circumferential cellular processes (‘onion bulbs’). There is prominent endoneurial collagen (C) and two clusters of regenerated axons are marked by arrowheads, two small axons with ‘redundant myelin loops’ are marked by arrows. Scale bar = 10 µm. (B) Comparative protein abundance by immunoblotting of induced pluripotent stem cell (iPSC)-derived motor neurons and rat Schwannoma cells probed for IP₃R3 and loading-control beta-tubulin. (C) Immunoblotting of patient-derived fibroblasts of two biological replicate controls and one p.Thr1424Met ITPR3 mutation carrier (Patient 1) (two technical replicates each) probed IP₃R3 protein, normalized to GAPDH expression, showing reduced IP₃R3. Relative expression differences were determined across three replicate blots from separate lysates by one-way ANOVA. ns = not significant. *P < 0.05, ***P < 0.001.

Figure 4

p.Thr1424Met mutation in human IP₃R3 augments the channel activity under non-optimal ligand conditions. (A) Representative traces (2 s long) of continuous single-channel recordings at V_m = −30 mV, for wild-type and p.Thr1424Met h-IP₃R3 channels. Ligand conditions ([IP₃] and free [Ca²⁺]_free) indicated. Arrows indicate the channel closed state. (B) Open probability (P_o) values for the wild-type and p.Thr1424Met mutant channels in various ligand conditions (indicated on abscissa).