Rare, low-frequency and common coding variants of ARHGEF28 gene and their association with sporadic amyotrophic lateral sclerosis

Abstract

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease. Over 90% of cases are sporadic (sALS) and 5%–10% are familial (fALS). So far, more than 20 genes/loci have been linked to ALS. C9orf72, SOD1, TARDBP, and FUS are noted as the most common ALS genes; however, mutations of these genes explain <10% of sALS cases. Recently, Rho guanine nucleotide exchange factor, encoded by ARHGEF28, has been linked to the ALS pathogenesis, possibly by binding low-molecular-weight neurofilament mRNA and affects its stability. However, a systemic screening of ARHGEF28 mutations in ALS is lacking. In this study, we sequenced the entire coding sequence of ARHGEF28 in a Chinese cohort of 399 sporadic ALS and 327 elderly controls. A total of 73 coding variants were identified, including 26 synonymous and 47 nonsynonymous. Among the nonsynonymous variants, 33 were rare (minor allele frequency [MAF]<0.01), in which 18 were only identified in cases and 12 were only in controls. Three loss-of-function mutations were identified, including 2 truncations (p.Arg231Ter and p.Ser561Ter) and a frameshift deletion (p.Lys1070fs) in 2 cases and 1 control subject. The frequency of total and case-only rare variants was 7.5% (30/399) and 5.0% (20/399), respectively, in the patients. SKAT-O test suggested that the novel coding variants were marginally enriched in the cases (p = 0.049). Single-variant analysis suggested that the p.Asn1046Ser variant had a higher frequency in cases (8/399, 0.02) than in controls (1/327, 0.003) (OR: 6.67, 95% CI: 0.83–53.61; p = 0.046). By contrast, none of the low-frequency (MAF: 0.01–0.05) or common (MAF > 0.05) variants was associated with ALS (p > 0.05). Among all patients, 9 (2.3%) carried rare variants predicted to be deleterious, and the age at onset of these carriers (45.6 ± 10.9 years) was marginally younger than noncarriers (51.9 ± 10.7 years) (p = 0.11). Our results supported a possible genetic contribution of rare but not low-frequency and common coding variants to ALS. These data may have implications in the mechanisms and genetic counseling of the disease.