Spinal muscular atrophy (SMA) is an autosomal recessive genetic disorder characterized by the degeneration of motor neurons in the anterior horn of the spinal cord, leading to severe progressive muscle weakness and atrophy. SMA is one of the most common fatal neurogenetic diseases in infancy, with an incidence of 1/6,000 to 1/10,000 ^[1]. SMA is primarily caused by biallelic loss-of-function mutations in the SMN1 gene. The SMN1 gene is ubiquitously expressed throughout the body, with the highest expression in the spinal cord, and its encoded survival motor neuron (SMN) protein is crucial for the survival and function of motor neurons ^[2]. In addition, humans possess a highly homologous gene to SMN1, called SMN2, with only a few nucleotide differences. The SMN2 gene contains a c.840C>T point mutation at an exon 7 splicing enhancer, which disrupts the splicing enhancer and/or creates a splicing silencer. This mutation results in a different pre-mRNA splicing pattern for SMN2 compared to SMN1, with most mRNA lacking exon 7, leading to the production of a non-functional, truncated SMN protein that is rapidly degraded. Only about 10%-15% of SMN2 pre-mRNA is spliced into full-length mRNA, encoding functional SMN protein ^[2-3]. Approximately 95% of SMA patients have homozygous deletions of exon 7 in SMN1, resulting in the inability of SMN2 to adequately compensate for the deficiency of SMN protein, leading to disease onset ^[2-4]. Therefore, the copy number of SMN2 is currently considered an important modifier influencing SMA disease phenotype. In most patients, the SMN2 copy number ranges from 1 to 6, and a higher copy number correlates with increased production of full-length SMN protein, typically resulting in a milder clinical phenotype ^[5-6]. Consequently, modulating SMN2 splicing to generate normal functional SMN protein has become a major research focus for SMA treatment. Unlike humans, mice have only one Smn1 gene encoding SMN protein, and its homozygous knockout is lethal, which presents limitations in constructing SMA models and mimicking the compensatory mechanism of the human SMN2 gene ^[7]. Therefore, the development of humanized mouse models that can both simulate the pathogenic mechanism of SMA (especially SMN2 copy number) and recapitulate the human disease course is of great significance for the development and validation of SMN2-targeted therapies.

Cyagen Biosciences has developed two foundational humanized SMN2 strains based on different strategies and used them to breed SMA disease models containing varying SMN2 copy numbers. One foundational strain (Smn1^hSMN2/hSMN2) was constructed by in situ replacement of both copies of the endogenous mouse Smn1 gene with the human SMN2 gene, i.e., knocking out the mouse Smn1 gene while simultaneously introducing two copies of the human SMN2 gene, aiming to mimic SMA patients carrying two copies of SMN2. The other foundational strain was constructed by inserting two copies of the human SMN2 gene into the mouse ROSA26 safe harbor locus (ROSA26^hSMN2/hSMN2). Through multiple breeding crosses of Smn1^hSMN2/hSMN2 mice and ROSA26^hSMN2/hSMN2 mice using different strategies, SMA models carrying 2, 3, and 4 copies of the SMN2 gene on an Smn1-deficient background can be obtained. B6-3*hSMN2 mice (genotype: Smn1^hSMN2/hSMN2ROSA26^hSMN2/+) are a humanized disease model carrying three copies of the human SMN2 gene, which can be used to mimic SMA patients with three SMN2 gene copies. Since the SMN2 gene primarily produces SMNΔ7 protein lacking exon 7, rather than full-length SMN protein, the humanized SMN2 gene cannot fully compensate for the abnormalities caused by Smn1 deficiency, resulting in the manifestation of SMA-like phenotypes in this model.