Hyperexcitability precedes motoneuron loss in the Smn2B/- mouse model of spinal muscular atrophy.

Spinal motoneuron dysfunction and loss are pathological hallmarks of the neuromuscular disease spinal muscular atrophy (SMA). Changes in motoneuron physiological function precede cell death, but how these alterations vary with disease severity and motoneuron maturational state is unknown. To address this question, we assessed the electrophysiology and morphology of spinal motoneurons of presymptomatic Smn2B/- mice older than 1 wk of age and tracked the timing of motor unit loss in this model using motor unit number estimation (MUNE). In contrast to other commonly used SMA mouse models, Smn2B/- mice exhibit more typical postnatal development until postnatal day (P)11 or 12 and have longer survival (~3 wk of age). We demonstrate that Smn2B/- motoneuron hyperexcitability, marked by hyperpolarization of the threshold voltage for action potential firing, was present at P9-10 and preceded the loss of motor units. Using MUNE studies, we determined that motor unit loss in this mouse model occurred 2 wk after birth. Smn2B/- motoneurons were also larger in size, which may reflect compensatory changes taking place during postnatal development. This work suggests that motoneuron hyperexcitability, marked by a reduced threshold for action potential firing, is a pathological change preceding motoneuron loss that is common to multiple models of severe SMA with different motoneuron maturational states. Our results indicate voltage-gated sodium channel activity may be altered in the disease process.NEW & NOTEWORTHY Changes in spinal motoneuron physiologic function precede cell death in spinal muscular atrophy (SMA), but how they vary with maturational state and disease severity remains unknown. This study characterized motoneuron and neuromuscular electrophysiology from the Smn2B/- model of SMA. Motoneurons were hyperexcitable at postnatal day (P)9-10, and specific electrophysiological changes in Smn2B/- motoneurons preceded functional motor unit loss at P14, as determined by motor unit number estimation studies.

Permittivity of ex vivo healthy and diseased murine skeletal muscle from 10 kHz to 1 MHz.

A better understanding of the permittivity property of skeletal muscle is essential for the development of new diagnostic tools and approaches for neuromuscular evaluation. However, there remain important knowledge gaps in our understanding of this property in healthy and diseased skeletal muscle, which hinder its translation into clinical application. Here, we report the permittivity of gastrocnemius muscle in healthy wild type mice and murine models of spinal muscular atrophy, muscular dystrophy, diabetes, amyotrophic lateral sclerosis and in a model of myofiber hypertrophy. Data were measured ex vivo from 10 kHz to 1 MHz using the four-electrode impedance technique. Additional quantitative histology information were obtained. Ultimately, the normative data reported will offer the scientific community the opportunity to develop more accurate models for the validation and prediction of experimental observations in both pre-clinical and clinical neuromuscular disease research.

Regulation of murine survival motor neuron (Smn) protein levels by modifying Smn exon 7 splicing.

Proximal spinal muscular atrophy (SMA) is caused by mutations in the survival motor neuron gene (SMN1). In humans, two nearly identical copies of SMN exist and differ only by a single non-polymorphic C-->T nucleotide transition in exon 7. SMN1 contains a 'C' nucleotide at the +6 position of exon 7 and produces primarily full-length SMN transcripts, whereas SMN2 contains a 'T' nucleotide and produces high levels of a transcript that lacks exon 7 and a low level of full-length SMN transcripts. All SMA patients lack a functional SMN1 gene but retain at least one copy of SMN2, suggesting that the low level of full-length protein produced from SMN2 is sufficient for all cell types except motor neurons. The murine Smn gene is not duplicated or alternatively spliced. It resembles SMN1 in that the critical exon 7 +6 'C' nucleotide is conserved. We have generated Smn minigenes containing either wild-type Smn exon 7 or an altered exon 7 containing the C-->T nucleotide transition to mimic SMN2. When expressed in cultured cells or transgenic mice, the wild-type minigene produced only full-length transcripts whereas the modified minigene alternatively spliced exon 7. Furthermore, Smn exon 7 contains a critical AG-rich exonic splice enhancer sequence (ESE) analogous to the human ESE within SMN exon 7, and subtle mutations within the mESE caused a variation in Smn transcript levels. In summary, we show for the first time that the murine Smn locus can be induced to alternatively splice exon 7. These results demonstrate that SMN protein levels can be varied in the mouse by the introduction of specific mutations at the endogenous Smn locus and thereby lay the foundation for developing animals that closely 'resemble' SMA patients.

Cloning, characterization, and copy number of the murine survival motor neuron gene: homolog of the spinal muscular atrophy-determining gene.

Because of a 500-kb inverted duplication, there are two copies of the survival motor neuron (SMN) gene in humans, cenSMN and telSMN. Both genes produce identical ubiquitously expressed transcripts; however, only mutations in telSMN are responsible for spinal muscular atrophy (SMA), the second most common autosomal recessive childhood disease. We have cloned the murine homolog Smn and mapped the gene to Chromosome 13 within the conserved syntenic region of human chromosome 5q13. We show that the Smn transcript (1.4 kb) is expressed as early as embryonic day 7. In contrast to humans, we found no evidence of alternative splicing. The predicted amino acid sequence between mouse and human SMN is 82% identical, and a putative nuclear localization signal is conserved. FISH data indicate that the duplication of the SMA region observed in humans is not present in the mouse. We also found no evidence of multiple Smn genes using Southern blot hybridization and single-strand conformation analysis. Using these methods, we detected at least four copies of Naip exon 5 clustering distal to Smn. Finally, three biallelic markers were identified within the Smn coding region; two are silent polymorphisms, whereas the third changes a cysteine residue to a tyrosine residue in exon 7. Overall, our results indicate that Smn is single copy within the mouse genome, which should facilitate gene disruption experiments to create an animal model of SMA.

Deletion and conversion in spinal muscular atrophy patients: is there a relationship to severity?

The spinal muscular atrophy-determining gene, survival motor neuron (SMN), is present in two copies, telSMN and cenSMN, which can be distinguished by base-pair changes in exons 7 and 8. The telSMN gene is often absent in spinal muscular atrophy patients, which could be due to deletion or sequence conversion (telSMN conversion to cenSMN giving rise to two cenSMN genes). To test for conversion events in spinal muscular atrophy, we amplified a 1-kb fragment that spanned exons 7 and 8 of SMN from 5 patients who retained telSMN exon 8 but lacked exon 7. In all patients, sequence analysis demonstrated that cenSMN exon 7 was adjacent to telSMN exon 8, indicating conversion. All 5 patients with this mutation had type II or III spinal muscular atrophy, strongly supporting an association with chronic spinal muscular atrophy. We also identified 3 families in which 2 siblings had no detectable telSMN but presented with markedly different phenotypes. We suggest that sequence conversion is a common event in spinal muscular atrophy and is associated with the milder form of the disease. The severity, however, can be modified in either a positive or negative direction by other factors that influence splicing or expression of the sequence converted SMN gene.

Allelic association and deletions in autosomal recessive proximal spinal muscular atrophy: association of marker genotype with disease severity and candidate cDNAs.

The candidate region for spinal muscular atrophy (SMA) has been defined as a 750 kb interval on 5q13. In this study, we performed allelic association studies in 154 German SMA families with the multicopy markers Ag1-CA (D5S1556); C212 (D5F149S1/S2) and correlated genotype data with deletion of candidate genes. Both multicopy markers recognize 0-3 alleles pro chromosome. Deletions were detected for all copies of the markers Ag1-CA (C272) and C212 in 13 of 88 (15%) type I SMA patients and three of 48 (6%) type II patients. In all informative cases, the deletion was inherited from one parent. In two further cases (one type I and one type III SMA), de novo deletions of only one copy of Ag1-CA and C212 were found. In both cases the patients were homozygously deleted for the survival motor neuron (SMN) gene (exons 7 and 8) but only the type I SMA patient was deleted for the neuronal apoptosis inhibitory protein (NAIP) gene (exons 5 and 6). A third case (type II SMA) showed de novo deletion of SMN, but not of Ag1-CA, C212 and NAIP. Specific alleles of Ag1-CA and C212 showed significant association with SMA, particularly in type I SMA. When the number of marker copies defines genotypes, 1,1 (one allele on each chromosome) is found to be increased in type I SMA (50%) and 1,2 (one allele on one chromosome and two alleles on the other one) in type II SMA (60%). The 2,2 genotype (two alleles on each chromosome) was found in 4% of type I and II patients. By comparison, pooled normal genotype frequencies were 20, 44 and 36%, respectively. These results suggest a strong correlation between genotype and severity of disease. Based on these data we propose a model which indicates that type I SMA patients are composed of two severe alleles, type II of a mild and a severe, and type III of two mild alleles. Correlation of Ag1-CA genotype with deletion of the XS2G3/NAIP genes indicates that most patients with a deletion have a 1,1 genotype. Owing to the physical proximity of these markers, we propose that a large deletion occurs on type I SMA chromosomes that removes DNA between C212 and XS2G3/NAIP and that type II SMA results from compound heterozygosity for mild (small deletion) and severe mutations.

A novel cDNA detects homozygous microdeletions in greater than 50% of type I spinal muscular atrophy patients.

Spinal muscular atrophy (SMA) is the second most common lethal, autosomal recessive disease in Caucasians (after cystic fibrosis). Childhood SMAs are divided into three groups (type I, II and III), which are allelic variants of the same locus in a region of approximately 850 kb in chromosome 5q12-q13, containing multiple copies of a novel, chromosome 5-specific repeat as well as many atypical pseudogenes. This has hampered the identification of candidate genes. We have identified several coding sequences unique to the SMA region. A genomic fragment detected by one cDNA is homozygously deleted in 17/29 (58%) of type I SMA patients. Of 235 unaffected individuals examined, only two showed the deletion and both are carriers of SMA. Our results suggest that deletion of at least part of this novel gene is directly related to the phenotype of SMA.

Association between Ag1-CA alleles and severity of autosomal recessive proximal spinal muscular atrophy.

The gene for autosomal recessive proximal spinal muscular atrophy (SMA) has been mapped to an 850-kb interval on 5q11.2-q13.3, between the centromeric D5S823 and telomeric D5S557 markers. We report a new complex marker, Ag1-CA, that lies in this interval, whose primers produce one, two, or rarely three amplification-fragment-length variants (AFLVs) per allele. Class I chromosomes are those which amplify a single AFLV allele, and class II chromosomes are those which amplify an allele with two or three AFLVs. Ag1-CA shows highly significant allelic association with type I SMA in both the French Canadian (Hôpital Sainte-Justine [HSJ]) and American (Ohio State University [OSU]) populations (P < .0001). Significant association between the Ag1-CA genotype and disease severity was also observed. Type I patients were predominantly homozygous for class I chromosomes (P = .0003 OSU; P = .0012 HSJ), whereas the majority of type II patients were heterozygous for class I and II chromosomes (P = .0014 OSU; P = .001 HSJ). There was no significant difference in Ag1-CA genotype frequencies between type III patients (P = .5 OSU; P = .25 HSJ) and the paired normal chromosomes from both carrier parents. Our results indicate that Ag1-CA is the most closely linked marker to SMA and defines the critical candidate-gene region. Finally, we have proposed a model that should be taken into consideration when screening candidate SMA genes.

A YAC contig of the region containing the spinal muscular atrophy gene (SMA): identification of an unstable region.

We report a 3.0-Mb YAC contig of the region 5q11.2-q13.3, which is where the spinal muscular atrophy gene has been localized. Three total genomic YAC libraries were screened by the polymerase chain reaction (PCR), and 45 YACs were recovered. These YACs were characterized for sequence tag site (STS) content, and overlaps were confirmed by vectorette PCR. Of the 45 YACs, 20 were isolated with the polymorphic marker CATT-1, which demonstrates significant allelic association with the SMA gene and maps within the 850-kb interval defined by the markers D5S557 and D5S823. Haplotyping of these YACs and their mother cell line indicates that the majority of YACs from this region contain deletions. Furthermore, a 1.9-Mb CATT-1 YAC that was negative for MAP1B and D5S435 and nonchimeric by FISH analysis provides a minimum distance between MAP1B and D5S435.

A multicopy dinucleotide marker that maps close to the spinal muscular atrophy gene.

Spinal muscular atrophy (SMA) is a common autosomal recessive disorder resulting in loss of motor neurons. The interval containing the SMA gene has been defined by linkage analysis as 5qcen-D5S435-SMA-D5S557-5qter. We have isolated a new dinucleotide repeat marker, CATT1, that lies between these two closest markers. The marker CATT1 has 16 alleles and is highly polymorphic. The marker can have 1 to 4 (or more) copies per chromosome, giving rise to individuals with up to 8 (or more) alleles. All of the subloci map between the markers D5S557 and D5S435 and lie in close proximity to one another. The marker CATT1 is linked to the SMA gene with a lod score of Zmax = 34.42 at theta = 0 and crosses all available recombinants. Certain alleles occurred more frequently in either the SMA or normal populations, indicating significant allelic association between CATT1 and the SMA locus. Haplotype analysis combining U.S. and Canadian SMA families reveals that one haplotype group (VII) occurs significantly more frequently in the SMA population than in the normal. This confirms the allelic association of CATT1 with the SMA locus.

Linkage mapping of the spinal muscular atrophy gene.

Spinal muscular atrophy (SMA) is a common autosomal recessive disorder resulting in loss of motor neurons. We have performed linkage analysis on a panel of families using nine markers that are closely linked to the SMA gene. The highest lod score was obtained with the marker D5S351 (Zmax = 10.04 at theta = O excluding two unlinked families, and Zmax = 8.77 at theta = 0.007 with all families). One type III family did not show linkage to the 5q13 markers, and in one type I consanguineous family the affected individual did not show homozygosity except for the marker D5S435. Three recombinants were identified with the closet centromeric marker, D5S435, which position the gene telomeric of this marker. These recombinants will facilitate finer mapping of the location of the SMA gene. Lastly, two families provide strong evidence for a remarkable variability in presentation of the SMA phenotype, with the age at onset in one family varying from 17 months to 13 years.