Lipidomic analysis reveals differences in the extent of remyelination in the brain and spinal cord.

Mohotti NDS, Kobayashi H, Williams JM, Binjawadagi R, Evertsen MP, Christ EG, and Hartley MD. (2023) Lipidomic analysis reveals differences in the extent of remyelination in the brain and spinal cord. Cold Spring Harbor Laboratory

Myelin sheaths are a layer that encompass neuronal axons in the central nervous system (CNS) and act as an insulator for signal transduction. It is formed by an extension of oligodendrocyte lipid bilayers and forms a multilamellar myelin sheath consisting of cholesterol, glycolipids, sphingomyelin, phospholipids, and plasmalogens. Demyelination can occur when an oligodendrocyte or myelin or the myelin itself is damaged, and the sheath degrades. For remyelination to occur following this degradation, any myelin debris need to be removed from the site by microglia phagocytosis, however it is unclear whether these lipids are recycled into new myelin and cell membranes or if they are further metabolized and broken down. Mohotti et al were interested in determining how the lipidome of the brain and spinal cord change throughout demyelination and remyelination. A genetic mouse model of both processes, Plp1-iCKO-Myrf, was used by inducing the knockout of Myrf, myelin regulatory factor, in mature oligodendrocytes which results in widespread demyelination in the brain, spinal cord, and optic nerve. Importantly, in this model Myrf is not knocked out in oligodendrocyte precursor cells (OPC) so they are still able to proliferate, differentiate into oligodendrocytes, and eventually remyelinate the CNS. As lipids are the primary component of myelin, in this study the authors performed lipidomics on the brain and spinal cord from these mice to study the lipid dynamics of myelin when it is damaged and repaired.

Throughout the study, the authors chose specific timepoints based on earlier work in this mouse model to represent active demyelination (6 weeks post-tamoxifen), peak demyelination (12 weeks post-tamoxifen), and remyelination (18 and 24 weeks post-tamoxifen). Previous work in these animals had not looked at demyelination in the spinal cord, therefore, to do so Mohotti et al stained brain and spinal cord tissue sections with Black-Gold, which stains myelinated tracts, from mice at peak motor disability, week 12, and peak motor recovery, week 24. The brain showed demyelination at week 12 with 39% of the levels seen in the control animals, but by week 24 remyelination returned these levels to 66% of that seen in the control mice. Compared to the brain, though, the spinal cord showed 30% of control levels at week 12 and 27% at week 24, indicating that remyelination was not able to occur in the spinal cord, but was able to in the brain. This difference in the ability to remyelinate provided the authors with the opportunity to investigate whether any lipidome differences between the brain and spinal cord were responsible for the impaired remyelination in the latter.

Another characteristic of this model is that when the Myrf knockdown is induced, it results in a loss of motor function which can be assessed by clinical scoring and the rotarod test. Typically, this begins at 5 weeks post-tamoxifen, continues to decline until 10-14 weeks post- tamoxifen, then partial recover can occur gradually through 14-24 weeks post-tamoxifen. Consistent with the model, when tested at 6, 12, 18, and 24 weeks post-tamoxifen on the rotarod motor test, mice with Myrf knocked out had reduced time on the rod (latency) at all time points compared to the control animals, with the greatest loss of function showing at 12 weeks and recovery beginning at week 24.

To determine whether there were any changes to the lipidome in the brain and spinal cord tissues of the demyelinated mice and healthy controls, samples were taken at 6, 12, 18, and 24 weeks post-tamoxifen. Many similarities were seen between the brain and spinal cord including reductions in ether-linked phosphatidylethanolamines (ethanolamine plasmalogens) and phosphatidylserines at all timepoints, even when remyelination was seen. Ethanolamine plasmalogens and sphingomyelin were decreased at weeks 12 and 18 but increased to normal levels by week 24. These both are important structural lipids that promote interactions within myelin and help maintain more stable myelin. As these recovered in the stage associated with remyelination, it was suggested that this indicates that these lipids have important roles in myelin. However, lipids such as phosphatidylcholines, lysophosphatidylcholine in the brain, and lysophosphatidylethanolamine in the brain and spinal cord were less effected by demyelination in the earlier timepoints and likely are less abundant in myelin. Interestingly, plasmalogens with either a choline or serine headgroup are less abundant but were found to be increased at 6 weeks post tamoxifen in the Myrf knockdown animals when demyelination was occurring, but then is normalized or reduced at the following timepoints.

Mohotti et al were interested in determining the role of plasmalogens in the process of demyelination and remyelination. This was accomplished using a mouse model where tamoxifen administration would induce loss of Myrf in mature oligodendrocytes while OPCs are still able to replicate, differentiate, and eventually form myelin. This causes a stage of demyelination followed by remyelination once the OPCs have developed into mature oligodendrocytes with functioning Myrf. Black-Gold staining was used to show myelinated tracts in the brain and spinal cord and found that the brain had demyelination at 12 weeks post-tamoxifen with 39% of levels found in controls but following remyelination at week 24 they were up to 66% of myelination seen in control animals. The spinal cord did not show this improvement with levels being 30% and 27% at weeks 12 and 24, respectively, indicating remyelination did not occur in this region. The author’s evaluated whether these stages of myelination were associated with alterations in motor function through the rotarod test and found that the Myrf knocked out animals had reduced latency scores at all timepoints compared to the control animals with the greatest decrease seen at 12 weeks and recovery apparent at 24 weeks. When looking at the lipidomics, early demyelination (6 weeks) was shown to have increased choline and serine plasmalogens, but later demyelination had decreases in phosphatidylethanolamines and phosphatidylserines while ethanolamine plasmalogens were decreased at all timepoints. Further work is needed to determine the biological mechanisms behind these lipid changes following demyelination and remyelination.