In my opinion the major message here is that there is a good deal of heterogeneity in all forms of ALS, even in specific subtypes of familial ALS such as this. This includes variability in penetrance, age and site of onset, comorbidities (for example dementia) and rate of progression. There are several possible reasons for this type of variability, including the specific site of each person's mutation within a given gene, possible modifier genes (for example kifap 3), and also possible environmental influences (for example exercise, lipid status).

Personally I think we now need to focus hard on this variability between patients....mother nature is trying to tell us something here! We could for example use new tools such as whole-genome microarray to compare fast and slow progressors. If we can understand this variability we may be able to shift patients' systems over to a more favorable phenotype even if we don't fully understand why it works.

View all comments by Richard Bedlack

John Landers and colleagues have successfully used exome capture followed by deep sequencing to identify novel mutations in the profilin gene (PFN1) that cause familial amyotrophic lateral sclerosis (FALS). The application of this methodology has greatly speeded up the identification of pathogenic mutations. Data obtained from affected members of two kindreds revealed different coding mutations, C71G and M114T, both being present in PFN1, that segregated with disease and had not been previously reported in available SNP databases. Subsequent screening of this gene for mutations in 273 further FALS and 816 sporadic ALS cases revealed two more FALS index cases with the C71G mutation and one other case with the M114T mutation. Two new mutations, G118V and E117G, were found in familial cases, and the E117G mutation was also found in two sporadic cases. The E117G mutation also occurred in three out of 7,560 controls and must be viewed with caution. No coding changes in PFN2 and PFN3 were seen in FALS cases.

Quite a lot of research has been performed on profilin, and it is known to...  Read more

John Landers and colleagues have successfully used exome capture followed by deep sequencing to identify novel mutations in the profilin gene (PFN1) that cause familial amyotrophic lateral sclerosis (FALS). The application of this methodology has greatly speeded up the identification of pathogenic mutations. Data obtained from affected members of two kindreds revealed different coding mutations, C71G and M114T, both being present in PFN1, that segregated with disease and had not been previously reported in available SNP databases. Subsequent screening of this gene for mutations in 273 further FALS and 816 sporadic ALS cases revealed two more FALS index cases with the C71G mutation and one other case with the M114T mutation. Two new mutations, G118V and E117G, were found in familial cases, and the E117G mutation was also found in two sporadic cases. The E117G mutation also occurred in three out of 7,560 controls and must be viewed with caution. No coding changes in PFN2 and PFN3 were seen in FALS cases.

Quite a lot of research has been performed on profilin, and it is known to function in the regulation of actin structure, transforming a globular monomer, known as G-actin, to long helical polymer, F-actin. This transformation is crucial in cytoskeletal dynamics, important in neurite outgrowth, growing axons, and synapse formation. Furthermore, the FALS-associated mutations lie in close proximity to the actin binding residues. The mutations were convincingly shown in this study to reduce axon outgrowth. Growth cone size and morphology were also affected by the mutations, with greatly reduced F-actin-rich filopodia being present.

Functional studies carried out by the authors showed that the three novel mutations had a propensity to aggregate and, when transfected into a neuronal cell line (Neuro2A) or primary motor neurons, formed ubiquitinated aggregates. Whilst these aggregates did not show co-aggregation with FUS or SMN, co-staining of aggregates with PFN1 and TDP-43 occurred in about one-third of cells. However, no abnormal PFN pathology was seen in spinal cord sections from sporadic cases of ALS.

Abnormalities in a number of cytoskeletal proteins are found in motor neuron diseases, but tend to be quite rare. At present, no autopsy cases harboring these PFN1 mutations are available, but information about the neuropathological features of this condition will be extremely valuable. The pathogenic mechanism remains to be elucidated and could involve effects on actin polymerization, or may yet result from one of the other numerous protein interactions reported for PFN.

View all comments by J. de Belleroche

There remain many families with ALS where the causative genes are yet to be discovered. In our Department of Neurology at the University of Tokyo, causative genes are unidentified in approximately half of ALS families.

In PFN1-associated familial ALS (FALS), the mutations were found in seven out of 274 families, meaning the frequency of families with mutations in PFN1 is rare. Nonetheless, I think this finding is important. Surely, most of the FALS families may have mutations in orphan genes. We need to identify all the causative genes for FALS, which should bring insight into sporadic ALS, which is more common compared to FALS.

I believe this discovery provides better understanding of the pathophysiology of ALS. Abnormality in conversion of monomeric (G)-actin to filamentous (F)-actin is a new mechanism in the disease.

We are pursuing similar approaches, and hope to contribute to better understanding ALS in the very near future.

View all comments by Shoji Tsuji