Clinical and genetic characterisation of dominant congenital spinal muscular atrophy

Abstract

The spinal muscular atrophies (SMAs) are a clinically and genetically heterogeneous group of disorders caused by aberrant development and/or early loss of spinal cord anterior horn motor neurons, resulting in muscle weakness and wasting. The most common form of SMA is caused by recessive mutations in SMN1 on chromosome 5q. Dominant congenital spinal muscular atrophy (DCSMA), also known as SMA with lower extremity predominance (SMALED), is a “non-5q” form of SMA characterized by congenital or early-onset lower limb-predominant muscle weakness and wasting, which often results in significant mobility impairment. The genetic basis of DCSMA and the pathological mechanisms that underlie this disorder have not been fully elucidated. The initial aim of this project was to define the clinical, MRI, histopathological and post mortem features of DCSMA, and to establish the genetic basis of the disorder in a four generation Australian DCSMA kindred. We used linkage analysis and direct sequencing to exclude involvement of all known dominant SMA and hereditary motor neuropathy genes. Analysis of whole exome sequencing-derived variants in genes within linked candidate regions resulted in the identification of a cytoplasmic dynein 1 motor complex adaptor gene, BICD2 (which encodes bicaudal-2 protein), as the causative gene in our research family. At the time of the discovery mutations in BICD2 were not an established cause of human disease. It was, however, known that mutations in a different dynein motor complex gene, DYNC1H1, which encodes a core structural component of the dynein motor complex, were responsible for a subset of DCSMA cases. We confirmed the pathogenicity of BICD2 by identifying additional cases and families with mutations in this gene, and by assessing the functional effects of BICD2-DCSMA mutations. With the help of an international team of collaborators we successfully identified a total of 32 affected individuals from nine different families. Functional studies demonstrated that BICD2 mutations altered the ability of BICD2 to bind with known binding partners (e.g. dynein intermediate chain and Rab6), and resulted in abnormal BICD2 cellular localisation. We subsequently analysed the clinical and pathological features of our BICD2-DCSMA cohort to comprehensively characterise the clinical, lower limb MRI, and histopathological features of this disorder. As a result of this research project, we have established that BICD2-DCSMA affects the development and/or early survival of anterior horn lower motor neurons (LMNs) in the lumbar spinal cord, with minor additional involvement of cervical cord LMNs. The underlying pathological process appears to be largely limited to early life which accounts for the lack of clinical progression after early childhood. The vast majority of affected individuals present with a “typical DCSMA” lower limb predominant clinical phenotype. Upper motor neuron features are present in a significant subset of affected individuals, and a BICD2 mutation has been identified in one member of a family segregating for a purely upper motor neuron disorder; hereditary spastic paraplegia (HSP), rather than DCSMA. This suggests that in some affected individuals abnormal development and/or loss of the upper motor neurons that populate the motor cortex is a feature of the disorder, and is sometimes the predominant pathological abnormality. By defining the phenotypic features of BICD2-disease we have facilitated focused genetic testing and exome variant filtering in individuals with typical DCSMA features. This is likely to result in improved clinical and genetic diagnosis rates for patients with this disorder. In addition, by screening cohorts of patients with HSP, and a closely related disorder, amyotrophic lateral sclerosis (ALS), we hope to further-clarify BICD2’s contribution to upper motor neuron disorders. We are also in the process of exploring whether other dynein motor complex genes might be responsible for genetically uncharacterised DCSMA cases. The discovery that mutations in two dynein motor complex-related genes cause two forms of DCSMA suggests that the dynein motor complex and the proteins that regulate this complex play critical roles in LMN development and/or survival. Components of this complex may therefore be important targets for future therapies aimed at treating motor neuron disorders. This is a possibility that will be explored further as we gain further insight into the mechanisms that underlie dynein-mediated motor neuron pathology.