From Network to Neurotransmitter: Understanding cognitive impairment in multiple sclerosis: an interdisciplinary approach

The central aim of this thesis was to better understand the underlying mechanisms of cognitive impairment in MS and its relationship to brain network functioning. This relationship was studied from three perspectives: a cognitive rehabilitation perspective (i.e., the role of physical exercise on network function and cognition), a longitudinal network perspective and from a cellular and molecular perspective (i.e., the role of the glutamatergic and GABAergic systems). The three perspectives resulted in the following sub-questions that I have tried to answer in this thesis: 1) What are the effects of an exercise intervention (running training) on cognitive and network functioning in PwMS? 2) How does network functioning evolve over time in PwMS that convert to a worse cognitive status? 3) What is the role of the glutamatergic and GABAergic systems in relation to cognitive performance and network functioning in PwMS? Physical exercise For the first question above, we found that: • MS patients who followed a 12-week running training showed increased visuospatial memory functioning, which was not seen in the control group. • In the intervention group, improvement on visuospatial memory correlated with change in hippocampus-DMN FC, indicating that stronger improvements on visuospatial memory coincided with larger increases in hippocampus-DMN FC • Stronger, more synchronized communication between the hippocampus and the DMN may thus be involved in the effects of a running training on visuospatial memory. Cognitive and network evolution over five years Regarding question 2 above, we found that: • Almost 20% of MS patients converted to a (mild) cognitively impaired phenotype over the course of five years • At baseline, the DMN of cognitively impaired patients showed higher centrality, while over time the VAN of cognitively preserved patients increased in centrality • This could indicate that the VAN starts to show changes first (either to regulate the DMN/FPN or is becoming dysfunctional) and that the DMN becomes dysfunctional when cognitive impairment is overt. The glutamatergic and GABAergic systems: in vivo neurotransmitters and receptors Finally, for the third question we found that: • No differences in glutamate and GABA concentrations were observed in the hippocampus and thalamus of MS patients, regardless of cognitive status (i.e., CP or CI). • However, CP patients showed increased GABAA-receptor density in the cortical and deep GM, which correlated to information processing speed. • This GABAA-receptor upregulation in the CP phase of MS could be a compensatory mechanism to preserve network or cognitive functioning, or could be secondary to inflammation or glutamatergic changes. And for the ex vivo and in silico study we conclude the following: • Both excitatory and inhibitory synaptic loss were observed in cortical layer VI, with losses of up to 15% in NAGM • In demyelinated cortex, inhibitory synaptic loss was more severe than excitatory and reached levels of up to 30% • Modeling synaptic loss showed that reducing inhibitory synapses impacted the network most, resulting in disinhibitory increases in network measures. • This could mean that inhibitory synapses are of particular importance in MS-related network disturbances.