A de novo SPAST mutation leading to somatic mosaicism is associated with a later age at onset in HSP
- First Online:
- Cite this article as:
- Depienne, C., Fedirko, E., Faucheux, J. et al. Neurogenetics (2007) 8: 231. doi:10.1007/s10048-007-0090-4
- 90 Views
SPG4/SPAST, the gene-encoding spastin, is responsible for the most frequent form of autosomal dominant hereditary spastic paraplegia (HSP). SPG4-HSP is a heterogeneous disorder characterized by both interfamilial and intrafamilial variation, especially regarding the severity and the age at onset. In this study, we investigated the origin of the mutation and the factors involved in intra-familial heterogeneity in a family with a SPG4 mutation. We demonstrated that the mutation occurred de novo and show evidence of somatic mosaicism in the grandfather, who was the only affected member of six siblings. His disease began at age 55, much later than in his daughter, who had onset at age 18, and his grandson, in whom onset was at age 5. These observations indicate that de novo mutations can occur in SPG4, and that somatic mosaicism might account for intra-familial variation in SPG4-linked HSP.
Mutations in SPG4/SPAST, encoding spastin, are responsible for the most frequent form of autosomal dominant hereditary spastic paraplegia (HSP), a group of disorders characterized by progressive spasticity and weakness in the lower limbs (MIM#182601). SPG4-HSP is heterogeneous and characterized by inter- and intrafamilial variation in severity and age at onset, which ranges from 1 to 80 years .
Clinical features of patients 1, 4, and 6
Age at onset
Signs at beginning
Stiffness in the legs
Stiffness in the legs
Needs help walking at 59 years; walks with a cane at 74 years
Help walking at 30 years
Walking still unlimited at age 17 years
+ in all limbs
+ in lower limbs
We followed the transmission of the alleles using microsatellite markers located within and flanking the SPAST gene. This analysis revealed that the haplotype segregating with the disease and the mutation in patients 1, 4, and 6 was transmitted by the mother, but was also present in the unaffected brother and two unaffected half-sisters of patient 1 who did not have the mutation (Fig. 1a). This strongly suggests that the mutation occurred de novo in patient 1. Another striking feature was the difference in the DHPLC elution profile of patient 1 compared to that of his affected offspring: it showed more homoduplexes and less heteroduplexes, indicating a different ratio between the normal and mutated alleles (Fig. 1b). Sequencing confirmed that he had a lower level of mutant allele in his blood cells (Fig. 1c). A quantitative PCR assay that specifically amplifies the c.1684C>T mutation  showed that patient 1 had 25–30% less of the mutated allele in his blood cells and fibroblasts than his daughter and grandson (p < 0.01, Fig. 1d and e), providing evidence of somatic mosaicism in this patient.
The high proportion of the mutation in both lymphoblasts and fibroblasts suggests that the mutation occurred at an early stage of embryogenesis. Whether patient 1 had mosaicism in the central nervous system could not be investigated. However, this patient had later onset than his offspring (see clinical features and ages at onset in Table 1), which is compatible with a smaller number of motor neurons containing mutated spastin. It has been shown that reduced intracellular levels of functional spastin are not well tolerated since leaky splice site mutations (creating both normal and aberrant mRNA) were reported to be pathogenic . Our results now suggest that reducing the proportion of neurons containing functional spastin is sufficient to cause HSP, although with a later onset. The proportion of de novo mutations and somatic mosaicism in an adult-onset disease is probably low and not the only cause of intrafamilial disease variability, but this possibility should be considered. The histopathologic analysis of such cases would help to understand the pathophysiological mechanisms of HSP.
The authors thank the family for its participation and the DNA and cell bank of IFR70 for DNA extraction. This work was supported by the VERUM foundation and the Programme Hospitalier de Recherche Clinique AP-HP (n°AOM03059, to AD).