Spinal muscular atrophy within Amish and Mennonite populations: Ancestral haplotypes and natural history

We correlate chromosome 5 haplotypes and SMN2 copy number with disease expression in 42 Mennonite and 14 Amish patients with spinal muscular atrophy (SMA). A single haplotype (A1) with 1 copy of SMN2 segregated among all Amish patients. SMN1 deletions segregated on four different Mennonite haplotypes that carried 1 (M1a, M1b, M1c) or 2 (M2) copies of SMN2. DNA microsatellite and microarray data revealed structural similarities among A1, M1a, M1b, and M2. Clinical data were parsed according to both SMN1 genotype and SMN2 copy number (2 copies, n = 44; 3 copies, n = 9; or 4 copies, n = 3). No infant with 2 copies of SMN2 sat unassisted. In contrast, all 9 Mennonites with the M1a/M2 genotype (3 copies of SMN2) sat during infancy at a median age of 7 months, and 5 (56%) walked and dressed independently at median ages of 18 and 36 months, respectively. All are alive at a median age of 11 (range 2-31) years without ventilatory support. Among 13 Amish and 26 Mennonite patients with 2 copies of SMN2 who did not receive feeding or ventilatory support, A1/A1 as compared to M1a/M1a genotype was associated with earlier clinical onset (p = 0.0040) and shorter lifespan (median survival 3.9 versus 5.7 months, p = 0.0314). These phenotypic differences were not explained by variation in SMN1 deletion size or SMN2 coding sequence, which were conserved across haplotypes. Distinctive features of SMA within Plain communities provide a population-specific framework to study variations of disease expression and the impact of disease-modifying therapies administered early in life.