A homozygous missense variant in the YG box domain in an individual with severe spinal muscular atrophy: a case report and variant characterization

Abstract

The vast majority of severe (Type 0) spinal muscular atrophy (SMA) cases are caused by homozygous deletions of survival motor neuron 1 (SMN1). We report a case in which the patient has two copies of SMN1 but clinically presents as Type 0 SMA. The patient is an African American male carrying a homozygous maternally inherited missense variant (c.796T>C) in a cis-oriented SMN1 duplication on one chromosome and an SMN1 deletion on the other chromosome (genotype: 2*+0). Initial extensive genetic workups including exome sequencing were negative. Deletion analysis used in the initial testing for SMA also failed to detect SMA as the patient has two copies of SMN1. Because of high clinical suspicion, SMA diagnosis was finally confirmed based on full-length SMN1 sequencing. The patient was initially treated with risdiplam and later gene therapy with onasemnogene abeparvovec at 5 months without complications. The patient's muscular weakness has stabilized with mild improvement. The patient is now 28 months old and remains stable and diffusely weak, with stable respiratory ventilatory support. This case highlights challenges in the diagnosis of SMA with a non-deletion genotype and provides a clinical example demonstrating that disruption of functional SMN protein polymerization through an amino acid change in the YG-box domain represents a little known but important pathogenic mechanism for SMA. Clinicians need to be mindful about the limitations of the current diagnostic approach for SMA in detecting non-deletion genotypes.

Keywords: African American; YG Box; c.796T>C variant; g.27134T>G polymorphism; modeling; non-deletion; spinal muscular atrophy (SMA).

Figure 1

Parental and proband SMN1 genotypes. The father (blue) has two copies of SMN1 without the c.796T>C variant or the g.27134T>G polymorphism (1+1 or 2+0). The mother (orange) has a cis-oriented SMN1 duplication with the homozygous c.796T>C variant (marked by “*”) on one allele and a wild-type SMN1 on the other allele (2*+1). The proband has a cis-oriented SMN1 duplication with the homozygous c.796T>C variant on one allele and a null SMN1 on the other allele (2*+0). The proband’s null SMN1 on the paternal chromosome was presumed to have arisen from a de novo deletion. Alternatively, the father may also be a 2+0 carrier, but is without the g.27134T>G polymorphism.

Figure 2

Structures of human wild-type and mutant SMN YG-box (residues 255–281) tetramers. Ser266 in the wild-type peptide and Pro266 in the mutant peptide are shown in ball-and-stick representations. Predicted solution structures of the wild-type YG-box tetramer (A, golden/wheat) and the mutant tetramer (B, blue/cyan). Root mean squared fluctuations of the helix backbones in the wild-type YG-box (C) and the mutant YG-box (D) from molecular dynamics simulations. The vertical dotted lines indicate the location of the substitution. The error bars indicate standard deviations. A kink is noticeable in the mutant peptide but not in the wild-type peptide.

Figure 3

The timeline of clinical presentations, diagnosis, and treatment for the patient from birth to current age (28 months old).