Remodeling of the mitochondrial network in response to metabolism involves changes to mitochondrial structure from the ultrastructure to the cellular level. Morphological changes in dysfunctional mitochondria that manifest in diseases such as Parkinson’s and Leber’s hereditary optic neuropathy drive the need to have a ’systems’ level understanding of the relationship between mitochondrial structure and function. However, an integrated, quantitative understanding of the mechanisms linking the changes of structure in response to functional state, and vice versa, is lacking in the field. We developed a multi-scale, quantitative image analysis pipeline and database to simultaneously extract structural features and functional markers of mitochondrial networks for further analysis. We applied this pipeline to the budding yeast, Saccharomyces cerevisiae, which we grew in different carbon sources to achieve distinct respiratory states. Our system was able to quantitatively show that the spatial distribution of mitochondrial membrane potential (ΔΨ, an indicator of mitochondrial function) within individual mitochondrial tubules was nonrandom and dependent on the respiratory state of the cell. These differences were consistent with known alterations to the cristae of the mitochondria. We next investigated the relationship between the connectivity of the mitochondrial network and ΔΨ. Network connectivity is generated by fission and fusion events between individual tubules within the mitochondrial network. Mitochondrial fusion and bioenergetic status are known to be interdependent. We were thus surprised that nowhere in our exhaustive network measurement-based analysis was ΔΨ upregulated in more highly connected networks or network regions as we had predicted. We expect that dynamic data will be needed to detect local regions of the network undergoing fusion dynamics and that it may be these highly dynamic regions of the network that could be be upregulated in ΔΨ. We also investigated the asymmetry of ΔΨ between the mother and daughter bud (future daughter cell) and found that ΔΨ was maintained at a higher level in the daughter throughout the cell division cycle. We detected an increasing gradient in the

distribution of ΔΨ along the mother-daughter axis and speculate that this might indicate an increasing ΔΨ-dependent process in the direction of the daughter bud during cell growth.

ABSTRACT OF THE DISSERTATION

Mitochondrial Metabolism and Morphology

in Mitochondrial Disease States

By

Mariella Theresa Simon

Doctor of Philosophy in Biological Sciences

University of California, Irvine, 2016

Professor Susanne Rafelski, Chair

Mitochondria are the hub of cellular metabolism, respiration and energy production. They are sites of reactive oxidative species (ROS), heme and steroid generation. They buffer cellular calcium flux, regulate redox states, control cell cycle and death and are therefore at the nexus of human health and disease. Syndromes encountered in the mitochondrial disease clinic, associated with mutations in mitochondrial genes, are generally rare and require detailed clinical mitochondrial disease workups, including next generation sequencing technology. The discovery of new genes often presents itself with inconclusive findings, which need to be further pursued and delineated in the research setting. It is therefore of great importance to apply the “bedside to bench and back to bedside” principle in the field of mitochondrial medicine to advance diagnostic and treatment modalities. In this dissertation, I demonstrate bioenergetic findings in three novel mitochondrial disease genes, NARS2, TFAM and LonP1, in a collaborative effort. By doing so, I have helped to delineate the underlying cause of pathologies for families with mitochondrial disease. Presented are also two case studies of patients with mutations in POLG and GFM1, with unusual clinical findings, which may point to new, not yet described cellular processes associated with mitochondrial dysfunction. By correlating gene function with patient’s disease, taking into consideration clinical parameters observed, I have also shown new metabolic findings in a primary fibroblast tissue culture system from patients with mutations in the mitochondrial matrix protease LonP1. The patients have mutations in a domain of the gene, which has never been associated with disease. Findings suggest that the patient phenotype depends on the location of the mutations in LonP1 and that clinical expression is domain specific. Findings also show increased susceptibility to stress, elicited by the use of antibiotic/antimycotic in the tissue culture system. In the future, I plan to build on the results from this study, to assess the effects of potential treatment modalities on mitochondrial function in fibroblast culture.