Mitochondria and energy metabolism in cell culture models of motor neuron disease

Abstract

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease characterised by the selective loss of motor neurons. Although a relatively small proportion of all ALS cases are caused by familial mutations in proteins such as superoxide dismutase 1 (SOD1) and transactive response DNA binding protein 43 (TDP43), shared clinical and pathological features across sporadic and familial cases of ALS suggest that both may share a common underlying mechanism. The reasons why motor neurons are primarily affected in ALS, particularly in those cases caused by a ubiquitously expressed mutation, remain unknown. Given the disproportionately high energy demand of motor neurons when compared to other cell types, and the wealth of evidence demonstrating the role of impaired energy metabolism in both sporadic and familial ALS, it is possible that the selectivity of motor neuron death in ALS involves impaired energy metabolism. To investigate this possibility, a better understanding of the mechanisms leading to impaired energy metabolism in ALS is needed, and this is dependent upon the availability of valid models.

Although neurons in the body are largely dependent on mitochondrial oxidative phosphorylation (OXPHOS) to meet the bulk of their energy demands, most cultured cells generate the bulk of their energy via glycolysis. Substituting medium glucose with galactose is one way of increasing OXPHOS in cultured cells, however galactose is not normally present in the central nervous system (CNS). Commonly used cell culture conditions, therefore, limit the ability to extrapolate results obtained under such conditions to in vivo neurons. As such, the hypothesis of this thesis was that increasing reliance on mitochondrial OXPHOS in cultured cells in a way that more closely replicates the in vivo energy metabolism of neurons will expose energy metabolism deficits due to ALS-causing mutations.

To establish a cell culture model in which cells have an increased reliance on mitochondrial OXPHOS in a way that more closely replicates the in vivo energy metabolism of neurons (Aim 1), primary cortical neurons were initially utilised for this study. However, due to the slow consumption of extracellular glucose cultured neurons could not be easily driven towards dependence on extracellular lactate as the primary fuel of mitochondrial OXPHOS. In subsequent experiments an alternative cell type, primary mouse embryonic fibroblasts (pMEFs), were grown in medium containing 3.5 mM glucose and two phases of growth were identified: the initial glucose-consuming (i.e. glycolytic) phase followed by the lactate-consuming (i.e. OXPHOS) phase. The lactate-consuming phase was characterised by increased Mito-Tracker Deep Red staining intensity, increased expression of nuclear-encoded mitochondrial proteins, and increased sensitivity to the OXPHOS inhibitor rotenone. Thus, by inducing cell autonomous depletion of glucose from the culture medium cells were forced cells to utilise extracellular lactate via mitochondrial OXPHOS to supply their energy needs.

To determine the effects of ALS-causing mutations on the mitochondria and energy metabolism of cells that have an increased reliance on mitochondrial OXPHOS (Aim 2), pMEFs derived from mice expressing mutant SOD1-G37R or mutant TDP43-A315T were grown under conditions established in Aim 1. No difference between control and mutant pMEFs (SOD1-G37R or TDP43-A315T) was observed with respect to any of the mitochondrial and energy metabolism parameters investigated. By contrast, when human fibroblasts derived from an ALS patient expressing mutant TDP43-M337V were grown under the same conditions, expression of TDP43-M337V suppressed upregulation of several nuclear-encoded mitochondrial proteins. However, this was also observed in human fibroblasts that were grown under conditions whereby they continued to be glucose-consuming (i.e. glycolytic) throughout the culture period. Common to both culture conditions was the fact that irrespective of metabolic state, cell proliferation markedly decreased with time in culture. This indicates that, in contrast to the need to grow cells under conditions whereby an increased reliance on OXPHOS is achieved via manipulating the availability of glucose, the proliferative state of the cells appeared to be a greater determinant of mutant TDP43-mediated effects than the relative availability of glucose or lactate.

Therefore, increased reliance of cultured cells on mitochondrial OXPHOS in a way that more closely replicates the in vivo energy metabolism of neurons does not appear to be necessary to expose energy metabolism deficits due to ALS-causing mutations. Rather the findings presented in this thesis indicate that it is necessary to analyse less proliferative cells, which more closely replicate the non-proliferative state of terminally differentiated post-mitotic neurons, in order to expose mitochondrial changes due to ALS-causing mutations. This is the first study to show a connection between a TDP43 mutation and nuclear-encoded mitochondrial protein alterations. Given the high energetic demand of motor neurons, this finding may explain why terminally differentiated post-mitotic neurons are more sensitive to ubiquitously expressed mutant TDP43.