Application of emerging sequencing and genomic technologies to diagnostic molecular genetics.

Abstract

The advent of massively parallel sequencing (MPS) has provided much broader opportunities in the field of molecular genomics. Previously, sequencing approaches were limited to a small number of single genes or exons. MPS has enabled assessment of hundreds or thousands of genes associated with a particular phenotype simultaneously, providing the clinical scientist with more information than ever before. This thesis examines the processes and challenges of introducing whole exome sequencing (WES) technologies into a clinical laboratory in three families, each with a distinct segregation pattern of an undiagnosed clinical condition. The challenges associated with the sequencing, analysis and interpretation of sequence variants, as well as elucidating mechanism(s) of pathogenicity of detected variants, were the focus of this thesis.

One family with five male siblings affected with intellectual disability were initially investigated for a suspected X-linked disorder. Although candidate variants were identified, the results did not ultimately detect any clearly pathogenic variants or therefore a diagnosis.

In a second family, a single affected daughter presented with an undefined ageing disorder. We identified a novel de novo X-chromosome variant in the gene BCL-6 co-repressor (BCOR), a G to A transition at c.3907G (p.Gly1303Ser). Although functional studies were not performed and definitive pathogenicity was not established during the course of this work, the BCOR variant was considered a strong candidate for this phenotype. Skewed X-chromosome inactivation was identified in this individual which is suggestive of an X-linked disorder, further supporting the role of BCOR dysfunction in her symptoms.

The bulk of this thesis then focused on the analysis of a third pedigree in which four children have acute alcohol sensitivity and pronounced cardiac fibrosis, two of whom died suddenly. Exome sequencing revealed compounding mutations in the PPA2 gene, inherited recessively in all four children. PPA2 had not been previously associated with human disease, however, the enzyme encoded by PPA2 localises to the mitochondria, which implicated this enzyme in this family's disorder. In silico algorithms supported a damaging effect of both variants upon the protein, and early functional experiments in yeast and zebrafish were suggestive that PPA2 was important in cardiac function. This finding led to the identification of three additional families with PPA2 mutations in collaboration with laboratories in Germany and Austria. The phenotype of affected individuals in the European families was more severe resulting in infantile death, whereas the NZ family experienced death in the second decade of life following ethanol exposure. This work has established PPA2 as a new cardiac disease gene, and ended a diagnosis odyssey for a NZ family. It has also instigated an ongoing investigation into the mechanism of PPA2-associated disease.

Exome sequencing in these three families has highlighted both the power of MPS to provide a diagnosis, and the challenges associated with analysing such complex data. In particular, the work has established a pipeline for the clinical laboratory which will need an associated laboratory for extensive follow-up functional analysis of candidate variants where a clinically significant impact has not previously been proved.