Axonal Degeneration and Cortical Abnormalities in Multiple Sclerosis

Post by Stephanie Williams

What's the science?

Within the brains of multiple sclerosis (MS) patients, some portions of cortex appear normal, while others show obvious demyelinating lesions. Normal and lesioned cortex from MS brains each show diffusion tensor abnormalities that have yet to be explained mechanistically. This week in Brain, Preziosa and colleagues analyzed normal-appearing cortex and lesioned cortex to understand the substrates of diffusion tensor microstructure tissue abnormalities in MS.

How did they do it?                             

The authors used diffusion tensor imaging to image the brains of 16 deceased individuals from the Netherlands who had longstanding MS (median duration of the disease was 31.4 years) along with 10 healthy age-matched controls. The MR images were used to calculate a diffusion tensor imaging metric called fractional anisotropy (FA), which ranges from 0 to 1, and is a measure of how water molecules are able to move in a preferential axis along the tracts of the brain. A value of zero indicates that diffusion occurs in all directions, while a value close to 1 indicates diffusion occurs along the preferential axis. These values are relevant as they are correlated with MS-related disability, and differentiate between MS phenotypes. Next, the authors dissected, stained and examined slices of brain to count the relative densities of different brain components (eg. myelin, microglia, astrocytes, axons and neurons), and used a statistical analysis (mixed models) to understand how these histological markers were related to the diffusivity findings. To understand how neuronal and non-neuronal cell density and volume contribute to fractional anisotropy differences, the authors quantified the neuronal density and volume of neurons in all cortical layers.

What did they find?

On average, the FA values calculated from diffusion imaging differed between normal-appearing MS cortex, lesioned MS cortex, and non-neurological (control participants) cortex. The authors found 1) a decrease in FA in normal appearing MS cortex compared to non-neurological controls’ cortex and 2) a significant increase in FA values in lesioned MS cortex compared to normal appearing MS cortex, which is consistent with previous reports of FA values in MS patient brains. To understand the cause of the FA changes, the authors analyzed the density of several neural and non-neural components and cells. They did not find any significant association between myelin density and FA. Microglia, astrocyte and neuron density and volume were also not significantly different between non-neurological controls and MS patients, suggesting that they do not contribute to cortical FA differences.

Deb (2).png

The authors found that the FA abnormalities were best explained by reduced densities of axons perpendicular to the cortex in both normal-appearing MS cortex and lesioned MS cortex, and reduced axons parallel to the cortex in lesioned cortex only. Usually, perpendicular axons contribute to a higher FA value, which explains why the reduction in perpendicular axons within normal-appearing cortex showed lower FA values. The reduction of parallel cortical axon density only in MS lesioned cortex could explain the increased FA values on lesioned cortex. The authors suggest that despite the loss of perpendicular axonal density in lesioned cortex, the concurrent reduction of parallel axon density could increase the coherence of the tissue containing the lesions, and could therefore increase FA.

What's the impact?

The authors collected evidence that advances our understanding of the pathological substrates of the diffusion tensor MRI-derived abnormalities in MS cortex. They suggest that the reduced parallel axon density could explain the increase in FA observed in lesioned portions of MS brains.


Preziosa et al. Axonal degeneration as substrate of fractional anisotropy abnormalities in multiple sclerosis cortex. Brain (2019). Access the original scientific publication here.