Lack of C9ORF72 coding mutations supports a gain of function for repeat expansions in amyotrophic lateral sclerosis

Abstract

Hexanucleotide repeat expansions in C9ORF72 are a common cause of familial and apparently sporadic amyotrophic lateral sclerosis (ALS) and frontal temporal dementia (FTD). The mechanism by which expansions cause neurodegeneration is unknown, but current evidence supports both loss-of-function and gain-of-function mechanisms. We used pooled next-generation sequencing of the C9ORF72 gene in 389 ALS patients to look for traditional loss-of-function mutations. Although rare variants were identified, none were likely to be pathogenic, suggesting that mutations other than the repeat expansion are not a common cause of ALS, and providing supportive evidence for a gain-of-function mechanism. We also show by repeat-primed PCR genotyping that the C9ORF72 expansion frequency varies by geographical region within the United States, with an unexpectedly high frequency in the Mid-West. Finally we also show evidence of somatic instability of the expansion size by Southern blot, with the largest expansions occurring in brain tissue.

Figure 1. Identification and sizing of C9ORF72 repeat expansions

A) A sample was considered positive for a C9ORF72 repeat expansion when repeat-primed PCR (RP-PCR) showed >30 decrementing peaks with 6bp periodicity and deemed negative in all other cases. However, all expansion positive cases in this study showed >60 repeats as shown in this example tracing. B) Southern blot analysis of C9ORF72 repeat sizes in nine unrelated patients with ALS. Fibroblast-derived DNA was available for subjects 1-6 (lanes 1-6) and brain-derived DNA for subjects 7-9 (lanes 7–9). All samples except the negative control in lane 8 were positive for the C9ORF72 expansion by repeat-primed PCR (RP-PCR). Fibroblasts show dominant bands at 6-7kb (600-800 GGGGCC units), two with multiple bands, while brain-derived samples show a smear of sizes near 12 kb (1600 GGGGCC units). Open arrow-head marks the location of non-expanded alleles. Clinical characteristics of subjects 1-9 are found in the Supplementary Table.

Figure 2. Distribution of longest normal C9ORF72 allele in SALS and controls

All non-C9ORF72 SALS patients and controls had their longest repeat allele measured by repeat-primed PCR (RP-PCR). The distribution of alleles was identical between the two groups, p=0.7.