Spinal Muscular Atrophy: The Past, Present, and Future of Diagnosis and Treatment

Abstract

Spinal muscular atrophy (SMA) is a lower motor neuron disease with autosomal recessive inheritance. The first cases of SMA were reported by Werdnig in 1891. Although the phenotypic variation of SMA led to controversy regarding the clinical entity of the disease, the genetic homogeneity of SMA was proved in 1990. Five years later, in 1995, the gene responsible for SMA, SMN1, was identified. Genetic testing of SMN1 has enabled precise epidemiological studies, revealing that SMA occurs in 1 of 10,000 to 20,000 live births and that more than 95% of affected patients are homozygous for SMN1 deletion. In 2016, nusinersen was the first drug approved for treatment of SMA in the United States. Two other drugs were subsequently approved: onasemnogene abeparvovec and risdiplam. Clinical trials with these drugs targeting patients with pre-symptomatic SMA (those who were diagnosed by genetic testing but showed no symptoms) revealed that such patients could achieve the milestones of independent sitting and/or walking. Following the great success of these trials, population-based newborn screening programs for SMA (more precisely, SMN1-deleted SMA) have been increasingly implemented worldwide. Early detection by newborn screening and early treatment with new drugs are expected to soon become the standards in the field of SMA.

Keywords: SMN1; SMN2; antisense oligonucleotides; classification; gene therapy; low-molecular-weight compounds; newborn screening; spinal muscular atrophy.

Figure 1

Three periods in the history of SMA research. In 1891, Werdnig reported the first two cases of SMA. Many cases of SMA with different phenotypes have since been reported. In 1995, two SMA-related genes, SMN1 and SMN2, were identified. SMN1 is the gene responsible for SMA, whereas SMN2 is a modifier gene of SMA. The discovery of these genes led to the emergence of novel drugs for SMA. In 2016, nusinersen was the first drug approved for treatment of SMA by the FDA in the US [15]. Two other drugs were thereafter approved by the FDA: onasemnogene abeparvovec (approved in 2019) and risdiplam (approved in 2020) [15].

Figure 2

SMN genes in chromosome 5q13. In 1995, two SMA-related genes, SMN1 and SMN2, were identified in 5q13 region [12]. These two genes were paralogs. The SMN1 gene, located in the telomeric side, is the gene responsible for SMA; its loss or defect causes SMA with different phenotypes. The SMN2 gene, located in the centromeric side, is a modifier gene for SMA; its copy number is associated with the severity of the disease. SMN1 and SMN2 were originally reported to exist in an inverted duplication [12]; however, recently, a tandem duplication model has been presented [36].

Figure 3

Functional domains of SMN protein. The Tudor domain is responsible for an interaction with coilin, a marker of Cajal bodies [49]. This domain also binds to Sm proteins [50,51]. The YG box is a tyrosine/glycine-rich region in the C-terminus of SMN protein that facilitates oligomerization of SMN protein by formation of the glycine zipper structure [52].

Figure 4

Alternative splicing of SMN genes and its protein products. When exon 7 is included in SMN2 mRNA (full-length (FL)-SMN2 mRNA), then a full-length SMN (FL-SMN) protein is produced. When exon 7 is excluded from SMN2 mRNA (Δ7-SMN2 mRNA), then a truncated SMN (Δ7-SMN) protein is produced. Δ7-SMN protein is unstable and almost nonfunctional because of its inability to form oligomers (lack of self-oligomerization).

Figure 5

ESE-targeting and ISS-targeting strategies. To prevent skipping of SMN2 exon 7, exonic splicing enhancer (ESE)-targeting and intronic splicing silencer (ISS)-targeting strategies were established. In 2003, two ESE-targeting compounds, namely exon-specific splicing enhancement by small chimeric effectors (ESSENCE) [102] and targeted oligonucleotide enhancers of splicing (TOES) [103], were devised. In 2002, a cis-acting element (element 1) that regulated SMN2 exon 7 splicing was identified in intron 6 [104]. In 2006, another cis-acting element (ISS-N1) that regulated SMN2 exon 7 splicing was identified in intron 7 [105]. Further details are provided in the text. ASO: antisense oligonucleotide, PNA: peptide nucleic acid.

Figure 6

Mechanism of action of nusinersen. Hua et al. proposed that nusinersen hampers interaction between ISS-N1 in SMN2 pre-mRNA intron 7 and hnRNP A1/A2 [106]. However, Singh et al. proposed a different mechanism. Nusinersen alters the secondary structure of SMN2 pre-mRNA, which promotes recruitment of U1 snRNP at the 5′-splice site of SMN2 pre-mRNA exon 7 [134].

Figure 7

Mechanism of action of onasemnogene abeparvovec. Onasemnogene abeparvovec is an scAAV9 vector-based drug [140]. After entering the host cell, the scAAV vector translocates into the nucleus, where the transgene acts as an episome (a small, stable chromosome separate from the native chromosome). The scAAV ITR increases the speed at which the double-stranded transgene is transcribed and the resulting protein is produced. The hybrid CMV enhancer and CB promoter activates the transgene to allow for continuous and sustained SMN protein expression. Abbreviations are as follows, AAV2 (adeno-associated virus serotype 2); AAV9 (AAV serotype 9); BGH Poly A (bovine growth hormone polyadenylation); CB (chicken β-actin); cDNA (complementary DNA); CMV (cytomegalovirus); ITR (inverted terminal repeat); scAAV (self-complementary AAV); SV (simian virus).

Figure 8

Mechanism of action of risdiplam. According to studies in which risdiplam analogues were used, risdiplam may act on two regions of SMN2 exon 7 for highly selective pre-mRNA splicing [147,148]. The first region is the 5′-splice site of exon 7 (TSL2) of the pre-mRNA transcribed from SMN2. A risdiplam molecule stabilizes the duplex of the 5′-splice site RNA sequence and the U1 snRNP RNA sequence and promotes splicing initiation. The second region is the internal structure around exonic splicing enhancer 2 (ESE2) within SMN2 exon 7. A risdiplam molecule binds to the ESE2 region, which alters the stem-loop structure (TSL1) in the first half of SMN2 exon 7. The risdiplam molecule bound to the ESE2 region inhibits hnRNPG protein binding while promoting the binding of two splicing regulatory proteins: far upstream element binding protein 1 (FUBP1) and its homolog, KH-type splicing regulatory protein (KHSRP). TSL1 includes a cis-acting element that suppresses SMN2 exon 7 splicing, and FUBP1 and KHSRP proteins are trans-acting factors that promote SMN2 exon 7 splicing.