eNews August 08, 2012

A novel, allelic series of spinal muscular atrophy (SMA) mouse models created

There is no cure for spinal muscular atrophy (SMA), but ongoing studies and the development of novel mouse models are offering hope to SMA-affected children and their families. In 2012, Jackson Laboratory scientist Cathleen Lutz, Ph.D., and Sylvie Ramboz, Ph.D., of PsychoGenics, Inc., Tarrytown, New York, spearheaded the development of a series of new SMA mouse models that can be used to explore the pathogenesis of and develop therapies (or a cure) for SMA (Osborne et al. 2012).


Spinal muscular atrophy is an autosomal recessive disease and the leading genetic cause of infant and toddler death worldwide. Approximately 1 in 6,000-10,000 children are born with SMA. It is characterized by the loss of motor neurons in the spinal cord and the resulting inability to control voluntary muscle movements, muscle wasting and respiratory complications.

Genetically, spinal muscular atrophy is caused by mutations in the survival motor neuron 1 (SMN1) gene on human chromosome 5. The gene's encoded protein, SMN, is evolutionarily conserved, ubiquitously expressed and essential for localizing and processing subcellular RNA. Due to an evolutionarily recent duplication, humans have one or more copies of a gene similar to SMN1 that is, SMN2SMN2, however, is imperfect: a C to T transition in exon 7 causes the aberrant splicing of 80-90% of SMN2 transcripts, resulting in transcripts without exon 7. The absence of both the SMN1 and SMN2 genes is embryonically lethal. Therefore, to survive, people with SMA must have one or more copies of the SMN2 gene. Generally, the more SMN2 copies they have, the less severe their disease. There is mounting evidence that increasing SMN protein production from one or more SMN2 gene copies might compensate for the absence of the SMN1 gene and mitigate, prevent or even cure SMA. Indeed, Smn1-deficient mice that are transgenic for eight copies of the human SMN2 gene do not develop SMA.

Novel allelic series of SMA mouse models

Mice have only a single Smn gene, Smn1, and its absence is embryonically lethal. To produce viable SMA mouse models, scientists have introduced various copies of full-length, exon 7-less or otherwise genetically modified human SMN2 transgenes in the murine Smn1 locus. They have also produced conditional Smn1 knockouts. These spinal muscular atrophy models vary in SMA severity, SMA phenotypes and lifespan. All have increased our understanding of and been instrumental in producing therapies for SMA.  Their utility, however, has been limited to studying either very severe or mild forms of SMA.

Models with intermediate phenotypes would provide more options for developing SMA therapies. To produce new models that exhibit a wider range of SMA phenotypes, and which in turn can be bred together to produce lines with an even broader variety of intermediate SMA phenotypes, Lutz, Ramboz, and colleagues produced a series of four models that harbor one of the four following genetically engineered SMN alleles. Each congenic model is available on either a C57BL/6J (000664) or an FVB/NJ (001800) genetic background.

  • Smn1A (Smn1tm2Mrph) – A knockout allele that completely disrupts the mouse Smn1 locus
    B6.Cg-Smn1tm2Mrph /J (007963) and FVB.Cg-Smn1tm2Mrph/J (007955)
  • Smn1B (Smn1tm4(SMN2)Mrph) – In place of the mouse Smn1 gene, has a hybrid allele fusing mouse Smn1 exons 1-6 to human SMN2 exons 7 and 8
    B6.129-Smn1tm4(SMN2)Mrph/J (008453) and FVB.129(B6)-Smn1tm4(SMN2)Mrph/J (008713)
  • Smn1C (Smn1tm5(Smn1/SMN2)Mrph) – In place of the mouse Smn1 gene, has both the hybrid allele and one copy of the human SMN2 gene
    B6.129-Smn1tm5(Smn1/SMN2)Mrph/J (008714) and FVB.129(B6)-Smn1tm5(Smn1/SMN2)Mrph/J (008604)
  • Smn1D (Smn1tm6(SMN2)Mrph) – In place of the mouse Smn1 gene, has the hybrid allele and three copies of the human SMN2 gene
    B6.129-Smn1tm6(SMN2)Mrph/J (009378) and FVB.Cg-Smn1tm6(SMN2)Mrph/J (009381)

The Lutz and Ramboz team then bred pairs of these models and produced lines with combinations of 0, 1, 2, 3, 4, 5, 6, or 8 relative copies of SMN. They found that, to be viable, a line must have at least four relative SMN copies (either hybrid or full-length SMN2 alleles). For examples, Smn1C/C homozygotes (which have two hybrid and two SMN2 alleles) and Smn1D/D homozygotes (which have two hybrid and six SMN2 alleles) are viable. Each line exhibits a reduced level of SMN, mild neuromuscular deficits and a different SMA severity. 

Because only compound heterozygotes produced by crossing parents with the Smn1C and Smn1D allele to parents with any other allele are viable, Lutz and Ramboz characterized Smn1C/C, Smn1D/A, Smn1D/B, Smn1D/C and Smn1D/D mouse lines. The phenotypes characterized include SMN protein expression levels in the brain, liver and spinal cord, body weight through 60 days of age, necrosis in the tail, hind limbs, and ears, neuromuscular junction formation and maintenance, electrophysiology, synaptic efficacy, behavior (rotarod, distance traveled, open field test, grip strength), nociception (sensitivity to heat and cold), cardiac function and body composition. As hypothesized, Smn1C/C mice with two copies of the hybrid allele and two copies of the full length SMN2 allele have the most severe SMA phenotype, substantiating previous findings that the amount of SMN produced is inversely proportional to SMA severity.

In summary, the Lutz and Ramboz team showed that different mice in the allelic series can be crossed to produce lines that express different levels of SMN, exhibiting different degrees of SMA severity. The models in the allelic series and the lines produced from them can be used to assess the preclinical efficacy of candidate SMA therapies in mice.