Multidimensional analyses of the pathomechanism caused by the non-catalytic GNE variant, c.620A>T, in patients with GNE myopathy

Abstract Background GNE myopathy is an autosomal recessive distal myopathy caused by biallelic variants in GNE, which encodes a protein involved in sialic acid biosynthesis. Compound heterozygosity of the second most frequent variant among Japanese patients with GNE myopathy, GNE c.620A>T, occurs in the expected number of patients; however, homozygotes for this variant are rare. The aim of this study was to elucidate the pathomechanism caused by c.620A>T.Methods Genome sequencing, identity-by-descent mapping, and RNA-seq analyses were conducted for GNE c.620A>T homozygotes. Blood and muscle sialylation levels were measured, and three-dimensional structural modeling of GNE oligomers was conducted using Alphafold2.Results The estimated frequency of the c.620A>T variant in the Japanese general population is 0.00161, equating to 238 homozygotes in total, but only three homozygotes with GNE myopathy have been identified. Identity-by-descent mapping indicated two distinct c.620A>T haplotypes, which were not correlated with patient clinical characteristics. Patients homozygous for c.620A>T had mildly decreased sialylation, and no additional pathogenic variants in GNE or abnormalities in transcript structure or expression of other genes related to sialic acid biosynthesis in skeletal muscle. Structural modeling of full-length GNE dimers revealed that the variant amino acid localized close to the monomer interface, but far from catalytic sites, suggesting functions in enzymatic product transfer between the epimerase and kinase domains on GNE oligomerization.Conclusion Homozygotes for c.620A>T rarely develop myopathy, while symptoms occur in compound heterozygotes, likely because of mildly decreased sialylation, due to partial defects in oligomerization and product trafficking by the mutated GNE protein.