Autosomal recessive West Syndrome. Exome sequencing and other high-throughput sequencing technologies work best in the identification of recessive disorders. While many cases of West Syndrome are thought to be the result of de novo mutations, recessive inheritance is seen in a subset of patients. In a recent paper in Epilepsia, Edvardson and colleagues now report mutations in ST3GAL3 in a consanguineous Palestinian family with four affected individuals with West Syndrome. This report takes us deep into the chromosomal anatomy of the linkage region, raising the question at what point we can claim that a gene is found.
Four affecteds. The authors investigate a family with three affected sibs and one affected cousin in a consanguineous Palestinian family. All four affected individuals have West Syndrome starting at the age of 3-4 months, followed by transition to Lennox-Gastaut Syndrome accompanied by intellectual disability. Available metabolic investigations and MRI were normal. The authors performed linkage analysis followed by exome sequencing.
Candidate genes in the region. Amongst the ~190 candidate genes, the 15Mb region on chromosome 1 carries various genes implicated in epilepsy including SNIP1 and SLC2A1 (GLUT1). As biochemical investigations were reported to be normal, I would assume that hypoglychorrhachia (low glucose in cerebrospinal fluid indicative of GLUT1-deficiency) had been excluded. Other genes in the proximity of the linkage region include EIF2B3. Mutations in this gene are known to cause leukencephalopathy with vanishing white matter, a neurodegenerative disease and CLN1, the gene mutated in neuronal ceroid lipofuscinosis type 1, an infantile neurodegenerative disease associated with epilepsy. It might be interesting to know to what extent these genes were covered by exome sequencing, even though they are unlikely to be causative given the otherwise unremarkable investigations. In situations like this, it might be important not to become a victim of the exome fallacy, i.e. the intuitive believe that a normal result in exome sequencing automatically excludes a given candidate gene or other genes in a candidate region.
The DECIPHER story. The UCSC genome browser offers the possibility to assess known microdeletions and microduplications in a given chromosomal region using the DECIPHER database. There are two de novo microdeletions reported in patients with autism, intellectual disability and other features. As these deletions were de novo, they are considered to be pathogenic. However, these microdeletions also include other genes besides ST3GAL3. Nevertheless, in conjuction with an earlier report that recessive variants in ST3GAL3 maybe involved in intellectual disability, this gene is the most likely culprit.
From ID to West Syndrome. In the family reported by Edvardson and colleagues, ST3GAL3 has been identified as the only gene with a potentially relevant recessive mutation. The authors investigate the effect of this mutation in vitro and show that is virtually abolishes ST3GAL-III activity. This enzyme is a sialyl-transferase and is important in the modification of brain gangliosides, highly complex oligosaccharide molecules important for neuronal cell-cell communication. The effect of the mutation identified by Edvardson and colleagues may be more severe than the mutations found in ID and the deletions found in patients with autism. This difference in dosage may account for the different phenotypes.
Conclusion. Even though gene identification in recessive families may appear straightforward, studies using exome sequencing always have to consider that a potentially causative mutations was not detected. Adding experiments that underline the biological consequence of the mutation adds certainty, as performed by the authors. However, even with added biological data, single families still lack some evidence with respect to the role of the candidate gene. Validation in other families still is the best proof of pathogenicity.