Complex SMN Hybrids Detected in a Cohort of 31 Patients With Spinal Muscular Atrophy

Abstract

Background and objectives: Spinal muscular atrophy (SMA) is a recessive neuromuscular disorder caused by the loss or presence of point pathogenic variants in the SMN1 gene. The main positive modifier of the SMA phenotype is the number of copies of the SMN2 gene, a paralog of SMN1, which only produces around 10%-15% of functional SMN protein. The SMN2 copy number is inversely correlated with phenotype severity; however, discrepancies between the SMA type and the SMN2 copy number have been reported. The presence of SMN2-SMN1 hybrids has been proposed as a possible modifier of SMA disease.

Methods: We studied 31 patients with SMA, followed at a single center and molecularly diagnosed by Multiplex Ligand-Dependent Probe Amplification (MLPA), with a specific next-generation sequencing protocol to investigate their SMN2 genes in depth. Hybrid characterization also included bioinformatics haplotype phasing and specific PCRs to resolve each SMN2-SMN1 hybrid structure.

Results: We detected SMN2-SMN1 hybrid genes in 45.2% of the patients (14/31), the highest rate reported to date. This represents a total of 25 hybrid alleles, with 9 different structures, of which only 4 are detectable by MLPA. Of particular interest were 2 patients who presented 4 SMN2-SMN1 hybrid copies each and no pure SMN2 copies, an event reported here for the first time. No clear trend between the presence of hybrids and a milder phenotype was observed, although 5 of the patients with hybrid copies showed a better-than-expected phenotype. The higher hybrid detection rate in our cohort may be due to both the methodology applied, which allows an in-depth characterization of the SMN genes and the ethnicity of the patients, mainly of African origin.

Discussion: Although hybrid genes have been proposed to be beneficial for patients with SMA, our work revealed great complexity and variability between hybrid structures; therefore, each hybrid structure should be studied independently to determine its contribution to the SMA phenotype. Large-scale studies are needed to gain a better understanding of the function and implications of SMN2-SMN1 hybrid copies, improving genotype-phenotype correlations and prediction of the evolution of patients with SMA.

Figure 1. Schematic Representation of Solved Hybrid Alleles

Structure of the alleles in the 11 patients from group 1. Each bar represents 1 allele, SMN1 is represented in orange, and SMN2 in purple. Patient 1 presented 1 SMN1 copy harboring the pathogenic variant c.332G>C. Note that PSV8 does not appear in the figure because it is no longer considered a PSV (see Methods for more details). PSV = paralogous sequence variant.

Figure 2. Schematic Representation of Partially Solved Hybrid Alleles

(A) Depiction of the specific PCRs designed for each SMN gene, one including PSV1 to PSV9 (PCR 1) and the other PSV9 to PSV16 (PCR 2). Primer numbers 1 and 4 are common for both SMN genes, while primer numbers 2 and 3 are specific for SMN1 and primer numbers 5 and 6 for SMN2. For more details, see eTable 1. (B) Schematic representation of the different alleles of patients 27, 28, and 29, whose hybrid structures were partially solved with the methodology used. In patient 27, allele 1 was fully solved as a hybrid structure containing only PSV10 from SMN1. By haplotype phasing, in alleles 2 and 3, we were able to define 4 blocks of PSVs, albeit the connection between the blocks cannot be defined (dashed lines). In patients 28 and 29, haplotype phasing established 3 blocks of PSVs, but the connection between the blocks cannot be solved for any of these alleles. Depending on these connections, both patients could present from 1 to 3 hybrid copies. Note that PSV8 does not appear in the figure because it is no longer considered a PSV (see Methods for more details). PSV = paralogous sequence variant.

Figure 3. Paralogous Sequence Variants (PSVs) of the Different Hybrid Structures

(A) Scheme of the 15 PSVs, which differentiate SMN genes, named from 1 to 16, because PSV 8 has been excluded (for more details, see Methods). SMN1 is colored orange, and SMN2 is purple. (B) Details of the PSVs included in each SMN2-SMN1 hybrid structure reported in our cohort. Hybrid structure numbers 1, 6, 8, and 9 are detectable by MLPA.