Advancing the diagnosis of mitochondrial diseases with genomic sequencing

© 2020 Rocio Rius DominguezMitochondrial diseases are the most common metabolic disease, caused by pathogenic variants in both nuclear and mitochondrial genomes. The molecular diagnosis can be challenging, sometimes taking years and requiring multiple invasive and risky diagnostic techniques. Genome sequencing (GS) is an emerging technology that allows the analysis of nuclear and mitochondrial genomes simultaneously. However, the interpretation of data can be complex, and the use of this technology in mitochondrial diseases is limited. This PhD thesis explores the diagnostic utility of GS by investigating retrospective and prospective cohorts of patients with suspected (but molecularly undiagnosed) mitochondrial disease. The thesis is divided into eight chapters. Chapter 1 begins with a general introduction to mitochondrial diseases and associated diagnostic challenges. Chapter 2 describes the subjects and methods used throughout the PhD. Chapters 3 to 6 use trio genome sequencing to analyse a retrospective cohort of patients with paediatric onset mitochondrial disease. In Chapter 3, trio GS uncovers TEFM and COX11 as new candidate mitochondrial disease genes. In Chapter 4, trio GS was useful to identify a deep intronic variant in NBAS, enabling diagnosis of a patient who showed clinical overlap with mitochondrial disease. In Chapter 5, trio GS was used to identify a novel apparently synonymous variant in PNPT1 where in silico and functional studies revealed a splicing defect instead. Clinical information and functional evidence from other patients with PNPT1 variants were studied through an international collaboration to gain a better molecular and clinical understanding of PNPT1-related mitochondrial disease. In Chapter 6, trio GS data was interrogated to study mtDNA inheritance, showing an absence of biparental mitochondrial DNA transmission in the cohort. In Chapter 7, the use of singleton GS data was studied in a prospective cohort of patients with suspected mitochondrial disease. A molecular diagnosis was made in 37% of cases. Examples that show the utility of GS include the identification of a mtDNA variant with low levels of heteroplasmy in blood (NC_012920.1(MT-TL1):m.3243A>T with 2% heteroplasmy), deletions in mtDNA (9kb deletion with 50% heteroplasmy) and nDNA (heterozygous 4.1kb intragenic deletion in AARS2), and the use of GS data to aid in paternity confirmation and variant phasing in a patient with suspected Perrault syndrome and PEX6 variants. Finally, Chapter 8 concludes that GS is an effective tool for the diagnosis of mitochondrial diseases. The diagnostic potential is increased by combining GS with other techniques such as RNA sequencing. Further research is being conducted to assess the cost-benefit of this technology