Impact of hypoxia on remyelination in a mouse model of Multiple Sclerosis

Multiple Sclerosis (MS) is a chronic inflammatory demyelinating disease of the central nervous system (CNS), characterised by focal areas of myelin destruction (demyelination), oligodendrocyte death with variable extent of axon damage contributing to neurodegeneration. The aetiology of MS is not fully understood by the scientific community, with considerations of a primary inflammatory root with an autoimmune component as well of a degenerative disorder with secondary inflammation under investigation. Demyelination leads to conduction block of action potentials and to loss of metabolic and structural support offered by the oligodendrocytes, contributing to axonal vulnerability to damage and may in the long-term lead to neurodegeneration, the main driver of permanent disability in MS patients. Regeneration of the myelin sheaths (remyelination) around axons can help protect demyelinated axons from damage, restore nerve conduction and thus function. Remyelination is carried out by oligodendroglia, but in MS, this process is inefficient and promoting remyelination represents a therapeutic goal due to the hopes to achieve prolonged neuroprotection. In the healthy brain, interstitial tissue oxygen levels are low. Thus, CNS cells, particularly oligodendrocytes and their precursors habitually exist in a physiological low oxygen environment. To withstand shifts in the oxygen tension, such as mild hypoxia, adaptive mechanisms take place in the brain, via the hypoxia inducible factors (HIFs) and its target genes, that help sustain physiological functions with changes in metabolism, angiogenesis, cell proliferation and cell death. In humans and animal models of MS, neuropathological and imaging studies show the presence of tissue hypoxia and evidence of tissue adaptations to this hypoxia. Oligodendrocytes and their precursors are especially susceptible to hypoxia. Remyelination, thus occurs within this hypoxic environment, but we do not understand whether hypoxia worsens tissue repair or whether the tissue adaptations aid remyelination, similarly to the regenerative effects of hypoxia pre-conditioning in other organs and diseases. To test this hypothesis, I used two experimental paradigms: 1) exposure to hypoxia during remyelination and 2) exposure to hypoxia prior to demyelination, as pre-conditioning. To study remyelination, in the absence of an autoimmune response, I used stereotactic injection of lysophosphatidylcholine (LPC) into the mouse corpus callosum as a focal model of demyelination, exposed mice to mild hypoxia (10% normobaric oxygen) and assessed remyelination efficiency at 10- and 15-days post lesion (Dpl) by electron microscopy. Firstly, I found that demyelinating lesions at 3 Dpl are hypoxic, as assessed using the oxygen sensitive immunohistochemical probe, pimonidazole, and this hypoxia persisted during remyelination at 10 and 15 Dpl. I also found that exposure to additional hypoxia during the period of remyelination resulted in a significant decrease in the number of remyelinated axons at 15 Dpl but not at 10 Dpl, compared to mice undergoing remyelination in normoxia (room air). Importantly, the observed decrease in the number of remyelinated axons was not due to a loss of axons. I also noticed a difference in thickness of the inner cytoplasmic tongue (ICT) than expected for a given myelin thickness. The ICT is the inner most uncompacted myelin layer facing the axon. The ICT is in direct connection to the oligodendrocyte soma and via myelin 'channels' which provide passage of metabolites to the underlying axons and this is the site of new myelin membrane deposition during myelin growth. I observed that the ICT was larger than expected for myelin thickness at 10 Dpl in remyelinated axons irrespective whether they were exposed to normoxia or hypoxia, followed by its recovery at 15 Dpl in normoxic mice, compared to non-demyelinated controls. At 15 Dpl, mice that remyelinated under additional hypoxic conditions showed a larger ICT for a given myelin thickness, compared to non-lesioned mice. This enlargement of the ICT may represent a metabolic compensatory mechanism during early remyelination in response to the lesion-associated hypoxia that may help sustain remyelination through provision of metabolites. However prolonged hypoxia leads to decompensation and together with an unfavourable lesion environment results in the loss of myelin and poor remyelination. I next assessed whether preconditioning of brain tissue to mild hypoxia before inducing a focal demyelinating lesion affects subsequent remyelination using the same model. Evidence from preconditioning paradigms in different organs, has shown that this confers resilience to further damage, aiding tissue repair. Lesion size was significantly reduced in mice preconditioned to hypoxia compared to controls, but there was no difference in the number of remyelinated axons at the two post-lesion timepoints. In conclusion, this work provides evidence that additional longer period of hypoxia during the period of remyelination may be detrimental to remyelination and that hypoxia preconditioning warrants future investigation to fully address its possible protective effect in reducing lesion size.