Reply to Millecamps et al.: Elucidating the role of D amino acid oxidase in familial amyotrophic lateral sclerosis

Abstract

We agree with the comment made in the letter from Millecamps et al. (1) that finding a second family with a mutation in D amino acid oxidase (DAO) would strengthen the case for the role of DAO and D amino acids in amyotrophic lateral sclerosis (ALS). However, the identification of a mutation in DAO that segregates with disease (2) has provided a starting point to investigate molecular mechanisms that may be relevant to sporadic cases and animal models of ALS, where D-serine and its synthetic enzyme, serine racemase, are elevated (3). Indeed, we were able to show, using primary motor neuron cultures and a motor neuron cell line, that the R199W DAO mutation promotes the formation of ubiquitinated protein aggregates and decreases cell survival. We recognize that this is a rare variation, because it was absent from 199 ALS families that lack previously identified mutations [superoxide dismutase 1 (SOD1), TAR DNA binding protein (TARDBP), fused in sarcoma (FUS), and vesicle-associated membrane protein (VAMP)-associated protein B and C (VAPB)]. This is also born out in the report from Millecamps et al. (1), showing its absence from 126 individuals with a family history of ALS, whereas a heterozygous variant, arginine 38 histidine (R38H), was detected in one case of familial ALS and also within a control population. Although the authors show that this amino acid is conserved in mammals (but not within vertebrates), this contrasts with the high degree of conservation seen for R199 DAO, which extends across the animal kingdom to fungi and bacteria (2). We also identified 20 further variants in the DAO gene in cases of familial ALS (2), some being present in multiple cases and four found within 33 bp of an exon boundary. Of three previously unreported variants tested to date, two were also present in a control population, and one was absent from 176 control chromosomes (2), which merits more extensive screening. Further investigations are in progress using a transgenic model that we have now established. However, because of the rarity of the R199W DAO mutation, priority in diagnostic screening of familial ALS should be given to the three most common disease-associated genes, SOD1, TARDBP, and FUS.