Exome sequencing of the proband resulted in 127X average coverage of targeted regions and 98% of targeted bases having at least 10 reads. Variant analysis revealed 17,328 total coding single nucleotide variants (SNVs) with a transition/transversion ratio of 3.3, and 188 total small insertions and deletions (InDels) (Table
). Due to the patient’s clinical diagnosis of suspected pediatric-onset mitochondrial disease, exome sequence results were analyzed with an assumed recessive genetic model. Fifteen (15) homozygous or hemizygous SNVs and 4 homozygous InDels were not reported in dbSNP137. One of the SNVs is predicted to affect splicing. Four of the SNVs were non-synonymous missense mutations; of these two were predicted to damage protein function due to high conservation across species (Table
Summary of coding variants identified through exome sequencing
Summary of novel non-synonymous coding homozygous variants in patient
Analysis of results of exome sequencing showed that the patient is homozygous for a splice site mutation in senataxin
). This mutation, SETX
A (GenBank Accession Number: NM_015046.5), is in the −1 position of exon 10 of the SETX
has a total of 24 exons encoding a protein of 2677 amino acids long and contains at its C terminus a seven-motif domain that places it in the superfamily 1 of helicases. Recessive mutations in SETX
cause Spinocerebellar Ataxia, Autosomal Recessive 1 (SCAR1, OMIM 606002). Patients typically develop juvenile-onset progressive ataxia, and some but not all patients exhibit oculomotor apraxia, hence the condition has also been termed ataxia with oculomotor apraxia type 2 (AOA2). Other features include increases in serum alpha-fetoprotein and creatine kinase. The clinical phenotype previously reported for SCAR1 parallels the ataxia, tremor, cerebellar atrophy, and hyporeflexia observed in the patient described here, thus, the homozygous SETX
splice site mutation likely accounts for the patient’s ataxia. Heterozygous missense substitutions in SETX
can give rise to dominant juvenile amyotrophic lateral sclerosis (ALS4, OMIM 602433).
The patient also has several clinical features that have not been reported to occur in SCAR1. In particular, the patient does not have ocular apraxia but rather CPEO, ptosis, and a history of bilateral congenital cataracts, features not reported in a recent survey of 90 mutation positive cases of SCAR1/AOA2
]. Given this clinical picture, the possibility that more than one gene locus gave rise to the patient’s phenotype was entertained, and the identified variants were re-analyzed with both recessive and dominant genetic models. Further review of the list of genes harboring variants not found in public databases and predicted to be damaging to protein function failed to identify any genes that have been annotated as part of the mitochondrial proteome, but did reveal an additional missense variant in the known disease gene, OCRL
. Mutations in OCRL
cause Lowe Syndrome as well as Dent2 disease. The oculocerebrorenal syndrome of Lowe includes variable involvement of the eyes, brain, and kidneys that could possibly explain the patient’s historical congenital cataracts. The missense variant, OCRL
A; p.E284K (GenBank Accession Number: NM_000276.3), was identified in exon 10. Most persons with Lowe Syndrome have mutations in exons 10–23 and many of these are missense
]. Two different tests for evolutionary conservation, GERP and PhyloP, both showed that this base position is highly constrained. The GERP test for evolutionary sequence conservation yielded a score of 5.44, which is the maximum or most constrained score possible (Table
). PhyloP analysis also showed significant evidence for conservation (score
1). Software programs that predict whether a missense mutation is damaging to protein function scored this variant with the most damaging score (Sift
1). Additionally, this OCRL
variant is not present in either dbSNP or HGMD.
There is substantial bioinformatics support that this mutation damages protein function and therefore would manifest as Lowe Syndrome in this male, however, we were unable to complete a full clinical assessment for Lowe Syndrome in the patient because the patient was not available for further clinical investigations. Likewise, due to a lack of tissue from the patient we were not able to confirm directly if the SETX
A results in the loss of exon 10 or otherwise affects this protein.
DNA samples from each parent were sequenced. Each was heterozygous for the SETX
splice site mutation, and the mother was found to be a carrier of the OCRL
variant. The patient’s mother did not display the characteristic cortical flecks in the lens typical of Lowe Syndrome carriers, but instead she has bilateral cerulean cataracts that were first discovered when she was in her 20’s. The mother’s pedigree shows three other family members with cataracts, one who had an early age of onset necessitating surgery (Figure
). The possibility of an independent locus segregating in this family that confers autosomal dominant cataract was evaluated. We queried the patient’s exome variants present in genes known to cause all inherited forms of cataract, reviewed in Huang et al.
]. No variants that were predicted to damage protein function were identified in any known cataract gene. This includes four known cerulean cataract genes: MAF
, and GJA3.
Ascertainment of CRYG
coding region was 100% in the exome dataset. However, part of the first exon of MAF
and the last exon of GJA3
are both repeat rich and as a result was not targeted by exome capture. Attempts to sequence these gene regions by Sanger sequencing were unsuccessful.
Figure 1 Pedigree of patient with ataxia and congenital cataracts. Four generations of the family of a proband with ataxia and congenital cataracts. Genotypes for the SETX mutation chr9:135,176,191, C>T and the OCRL mutation is chrX:128695181, (more ...)