Skip to main content

Treating mitochondrial disease caused by pathogenic mtDNA mutations

Periodic Reporting for period 2 - mtDNA-CURE (Treating mitochondrial disease caused by pathogenic mtDNA mutations)

Reporting period: 2019-07-01 to 2020-12-31

Mutations of mtDNA are important causes of mitochondrial disease and are also thought to play important roles in age-associated diseases and ageing. Human pathogenic mutations of mtDNA frequently affect tRNA genes and are typically heteroplasmic, i.e. present only in a fraction of all mtDNA copies in a cell. If the levels of mutated mtDNA exceed a certain threshold, the oxidative phosphorylation will be impaired and cause dysfunction of affected organs. Remarkably, maternal relatives to mitochondrial disease patients can be perfectly healthy despite having very high, but subthreshold levels, of mutated mtDNA. The corollary to these observations is that a small increase of wild-type mtDNA may prevent mitochondrial dysfunction and disease. It is currently unknown whether an increase of the absolute amount of wild-type mtDNA will prevent disease manifestations even if the proportion of mutated mtDNA remains the same. Pathogenic mtDNA mutations typically act in a “recessive” (loss of function) way and it is therefore reasonable to assume that increase of the absolute amount of wild-type mtDNA may restore respiratory chain function. Based on this hypothesis, we propose that treatments aiming to increase overall mtDNA copy number are a reasonable strategy to treat or even cure a large group of human patients with mitochondrial diseases.

The specific aims of the proposal are:
1. To investigate how mtDNA copy number manipulation affects the phenotypic expression of disease-causing mtDNA mutations in flies and mice.
2. To identify the regulatory mechanism at the end of the displacement (D)-loop region that
controls the switch between abortive and genome-length mammalian mtDNA replication.
3. To develop novel small molecule chemical stimulators that increase mtDNA copy number.
To aim 1:
My lab has previously generated a mouse model with a single pathogenic mutation in the tRNA gene of mtDNA (C5024T in tRNAAla), which recapitulates a pathogenic human mtDNA mutation (G5650A in tRNAAla). We have now systematically characterized mice with different levels of the mutation to determine the proportion of mutated mtDNA needed to induce respiratory chain deficiency in different cell types We found that the maximally tolerated proportion of the pathogenic C5024T mutation is increased in mice with high mtDNA copy number. MtDNA copy number is increased in the colon and heart of C5024T mice at advanced disease stages. We were also able to show that increasing mtDNA copy number has beneficial effects on the pathology associated with the C5024T mutation, whereas reduced mtDNA copy number had tissue-specific effects.

We are currently also generating mutant mice with other single pathogenic mtDNA mutations, which will be analysed in the same way as C5024T mice.

To aim 2:
In order to identify the protein or protein complex regulating mtDNA replication at the end of the displacement loop (D-loop), we will perform a genome wide CRISPR/Cas9 screen. We are currently optimizing the FISH-FACS protocol in order to detect ACR transcript. Once the conditions for the screen are properly optimized, ACR-expressing cells generated during the CRISPR/Cas9 screen will be isolated by FACS sorting followed by sequencing of the gRNAs in order to identify the genes that regulate replication at the end of the D-loop.
My laboratory has generated MGME1 knockout mice. We found that knockout mice display tissue specific replication stalling patterns and sequence coverage patterns of mtDNA. These molecular phenotypes emphasize the importance of the MGME1 knockout mouse line as a valuable tool to investigate tissue-specificity of mtDNA maintenance disorders. To further study those tissue specific molecular changes of MGME1 mutant animals we performed histological stainings of tissue sections from multiple organs in MGME1 knockout mice and controls. Preliminary investagation of the liver, brain, colon and testis of MGME1 knockout animals did not display any obvious hystological changes, whereas we noted immune cell infiltration and focal-segmental sclerosis in kidney of aged MGME1 knockout mice.
In addition, we have used BioID proximity labelling method to screen for the physiologically relevant protein interactions of MGME1 occurring in mitochondria in living cells and are currently analyzing and validating the results.

To aim 3:
We have worked together with the Lead Discovery Center, Dortmund, Germany and have set up a robust drug screening strategy to identify small molecular stimulators of mtDNA transcription. After an initial phase of assay development, we have successfully completed a high throughput screen of 201 000 compounds. After applying a drug-likeness filter and repeated analysis, we identified a list of 528 active compounds with EC30 lower than 10 μM. Following characterization for compound specificity on mitochondrial transcription, we have currently selected 15 compounds for further optimization.

In order to study if replication primer formation can be modulated in vivo, we overexpressed POLRMT in mice using the BAC transgenic strategy. Polrmt overexpressing mice are viable and do not show any detectable effect on mitochondrial metabolism.
My group is developing novel mutant mouse models with single pathogenic mtDNA mutations. Identification of mutations that affect tRNA genes and severely impair respiratory chain function is crucial in order to understand underlying causes for human mitochondrial disease and to develop new treatments. We are also generating various novel mouse models in order to study if autophagy or mitophagy influences random mutations and/or has an impact on tRNAAla mutation load. None of these models have existed prior to this project.

We have also developed a database available via web interface to systematize our proteomic data and to facilitate its accessibility to the research division members and other researchers.
Amongst other features, the database is able to automatically calculate log2-fold change (LFC) and log10 (p) (logP) values and draw volcano plots from the data. Furthermore, the database allows for pairwise and multiple dataset comparison, including drawing LFC/LFC comparison plots or heatmaps or calculation of correlation coefficients.