Structural model of the OPA1 GTPase domain may explain the molecular consequences of a novel mutation in a family with autosomal dominant optic atrophy

Autosomal dominant optic atrophy (ADOA) is the most frequent hereditary optic neuropathy. Three loci have been reported for ADOA: a major locus, harboring all identified mutations to date, maps to 3q28 (OPA1), a second locus is linked to 18q12.2–q12.3 (OPA4) and a third locus on 22q12.1–q13.1 (OPA5) has been reported recently. We describe a six-generation Iranian family in which optic atrophy runs as an autosomal dominant trait with an age of onset at 14–15 years. We performed linkage analysis with markers mapping to 3q28 and 18q12.2–q12.3 and found linkage to 3q28. Subsequent sequencing of OPA1 identified a novel heterozygous missense mutation (c.1313A>G) replacing aspartic acid by glycine (p.D438G) in the GTPase domain of OPA1. Interestingly, another missense mutation at the same position (c.1313A>T, D438V) has been reported before in two unrelated German families, indicating a possible mutation hot spot. Further evidence supporting the importance of D438 is its conservation from human to acoelomata. OPA1 is believed to be the human orthologue of yeast MGM1, a dynamin-related protein required for the integrity of mitochondrial DNA. Homology modeling of the OPA1 GTPase domain revealed extensive structural similarity to the Dictyostelium dynamin A GTPase domain and showed that D438 may interact with residues of the G1 and the G4 motifs, which are crucial in coordinating GTP. Based on this analysis, we propose a mechanism which explains the gradual decline of vision in ADOA patients with OPA1 mutations at position 438.