Multi-omics, Epigenomics and Computational Analysis of Neurodegenerative Disorders

Loss-of-function mutations in the gene that encodes TYRO protein kinase-binding protein (TYROBP) cause Nasu-Hakola disease, a heritable disease resembling Alzheimer’s disease (AD). Methylation of N6 methyl-adenosine (m6A) in mRNA plays essential roles in learning and memory. Aberrant m6A methylation has been detected in AD patients and animal models. In the present study, Tyrobp^–/– mice showed learning and memory deficits in the Morris water maze, which worsened with age. Tyrobp^–/– mice also showed elevated levels of total tau, Ser202/Thr205-phosphorylated tau and amyloid β in the hippocampus and cerebrocortex, which worsened with aging. The m6A methyltransferase components METTL3, METTL14, and WTAP were downregulated in Tyrobp^–/– mice, while expression of demethylases that remove the m6A modification (e.g., FTO and ALKBH5) were unaltered. Methylated RNA immunoprecipitation sequencing identified 498 m6A peaks that were upregulated in Tyrobp^–/– mice, and 312 m6A peaks that were downregulated. Bioinformatic analysis suggested that most of these m6A peaks occur in sequences near stop codons and 3′-untranslated regions. These findings suggest an association between m6A RNA methylation and pathological TYROBP deficiency.

The rate of disease progression in amyotrophic lateral sclerosis (ALS) is highly variable, even between patients with the same genetic mutations. Metabolic alterations may affect disease course variability in ALS patients, but challenges in identifying the preclinical and early phases of the disease limit our understanding of molecular mechanisms underlying differences in the rate of disease progression. We examined effects of SOD1^G93A on thoracic and lumbar spinal cord metabolites in two mouse ALS models with different rates of disease progression: the transgenic SOD1^G93A-C57BL/6JOlaHsd (C57-G93A, slow progression) and transgenic SOD1^G93A-129SvHsd (129S-G93A, fast progression) strains. Samples from three timepoints (presymptomatic, disease onset, and late stage disease) were analyzed using Gas Chromatography-Mass Spectrometry metabolomics. Tissue metabolome differences in the lumbar spinal cord were driven primarily by mouse genetic background, although larger responses were observed in metabolic trajectories after the onset of symptoms. The significantly affected lumbar spinal cord metabolites were involved in energy and lipid metabolism. In the thoracic spinal cord, metabolic differences related to genetic background, background-SOD1 genotype interactions, and longitudinal SOD1^G93A effects. The largest responses in thoracic spinal cord metabolic trajectories related to SOD1^G93A effects before onset of visible symptoms. More metabolites were significantly affected in the thoracic segment, which were involved in energy homeostasis, neurotransmitter synthesis and utilization, and the oxidative stress response. We find evidence that initial metabolic alterations in SOD1^G93A mice confer disadvantages for maintaining neuronal viability under ALS-related stressors, with slow-progressing C57-G93A mice potentially having more favorable spinal cord bioenergetic profiles than 129S-G93A. These genetic background-associated metabolic differences together with the different early metabolic responses underscore the need to better characterize the impact of germline genetic variation on cellular responses to ALS gene mutations both before and after the onset of symptoms in order to understand their impact on disease development.