A number sign (#) is used with this entry because of evidence that susceptibility to amyotrophic lateral sclerosis-25 (ALS25) is conferred by heterozygous mutation in the KIF5A gene (602821) on chromosome 12q13.

For a phenotypic description and a discussion of genetic heterogeneity of amyotrophic lateral sclerosis, see ALS1 (105400).

Nicolas et al. (2018) reported 19 patients from 9 unrelated families with ALS associated with heterozygous loss-of-function (LOF) variants in the KIF5A gene. The median age at disease onset was 46.5 years, which is lower than the age of onset reported for ALS in epidemiologic studies (65.2 years). The median survival time for ALS25 patients was nearly 10 years (117 months), which is much longer than that reported in epidemiologic studies (20 to 36 months). Two patients with sporadic disease were also reported: they had onset at ages 50 and 52 years, respectively.

Brenner et al. (2018) reported 3 unrelated probands with ALS25. The patients had typical features of the disorder, including adult onset of focal asymmetric involvement of upper and lower motor neuron systems with later generalization, bulbar motor involvement, rapid disease progression, and early death.

The transmission pattern of ALS25 in the families reported by Brenner et al. (2018) was consistent with autosomal dominant inheritance with possible incomplete or age-dependent penetrance.

In 12 patients from 9 unrelated families with ALS25, Nicolas et al. (2018) identified heterozygous loss-of-function (LOF) variants in the C-terminal region of the KIF5A gene (see, e.g., 602821.0014-602821.0016). The KIF5A gene was chosen as a candidate gene for study based on the results of a genomewide association study (GWAS) of several cohorts encompassing 20,806 ALS cases and 59,804 controls. In the GWAS, there was a significant association between a P986L missense variant (rs113247976) in the KIF5A gene and ALS (OR = 1.38, p = 6.4 x 10(-10)). The ALS association with P986L was replicated in 4,159 additional ALS cases and 18,650 controls. The authors noted that this missense variant may not be the primary risk factor and may be in linkage disequilibrium with other causative variants, or that it may be a common, low-penetrance risk allele. The LOF point mutations were initially found among a cohort of 1,138 probands with familial ALS and 19,494 controls who underwent exomewide rare variant burden (RVB) analysis for association of LOF variants. Nicolas et al. (2018) identified 6 heterozygous LOF variants in the KIF5A gene in patients (0.53%) compared to 3 such variants among the controls (0.015%). Further analysis of this patient cohort identified 2 small indels in the KIF5A gene (Asp996fs and Asn999fd) that were predicted to result in a frameshift. Variation in the KIF5A gene reached exomewide significance (OR = 41.16, p = 3.8 x 10(-9)). Two of the variants (602821.0014 and 602821.0015) were confirmed by Sanger sequencing and segregated with the disorder in the 2 unrelated families. All ALS-associated KIF5A variants occurred within a 34-bp stretch of DNA and were predicted to affect splicing of exon 27, which encodes amino acids 998-1007. The mutations were predicted to result in the complete skipping of exon 27, yielding a transcript with a frameshift at residue 998, the deletion of the normal C-terminal 34 amino acids of the cargo-binding domain, and the extension of an aberrant 39 amino acids to the C terminus. The 2 small indels were predicted to have a similar effect near exon 27. Mutational screening of KIF5A in an additional cohort of 9,046 ALS cases, mainly sporadic patients, identified 3 additional carriers of C-terminal variants. Splicing abnormalities were confirmed in cells derived from 2 unrelated patients with mutations. Nicolas et al. (2018) noted that KIF5A mutations associated with SPG10 are almost exclusively missense mutations that affect the N-terminal motor domain, whereas mutations associated with ALS are found predominantly in the C-terminal cargo-binding region, suggesting a genotype/phenotype correlation. Nicolas et al. (2018) speculated that KIF5A variants cause disease by disrupting axonal transport. The variants identified, in the cargo-binding domain, may cause accumulation of cytoplasmic protein aggregates at the neuronal cell body, resulting in a deficiency of certain cargo proteins at neurite terminals.

In 3 unrelated probands with familial ALS25, Brenner et al. (2018) identified 3 different mutations in the KIF5A gene (see, e.g., 602821.0017). The mutations, which were found by whole-exome sequencing of a cohort of 426 probands, were confirmed by Sanger sequencing. In each case, family segregation studies suggested that the mutation segregated with the disorder, but DNA was not available from other affected family members to confirm. There was also evidence of incomplete penetrance. All 3 mutations (c.2993-1G-A, c.3019A-G, and 3020+2T-C) occurred at the C terminal in or near exons 26 and 27, and all were predicted to result in splicing abnormalities. Aberrant splicing was confirmed in lymphoblasts from 1 of the patients. The variants identified were similar to a c.3020+1G-A variant reported in the ALS Variant Server database. Brenner et al. (2018) postulated haploinsufficiency as the pathogenic mechanism. In addition, Brenner et al. (2018) found the P986L variant in the KIF5A gene in 29 of the 426 patients with familial ALS (allele frequency of 3.4%) compared to 123 of 6,137 controls (allele frequency of 1%) and the gnomAD database (allele frequency of 1.13%). However, 11 of the 29 patients carrying this missense variant also had heterozygous variants in other ALS-associated genes. Analysis of lymphoblasts carrying the P986L variant showed that the variant did not have any effect on splicing, and mRNA levels were similar to those of controls.