Mouse experimental autoimmune encephalomyelitis (EAE) has long been the favored animal model of multiple sclerosis (MS). However, Theiler's Murine Encephalomyelitis Virus (TMEV) allows for the investigation of cerebral atrophy and iron accumulation, alongside inflammation, which offers parallels to neurodegeneration in MS.

In genetically susceptible breeds (such as SJL/J) TMEV infection gives way to a progressive demyelinating course (starting at about 35 days post injection¹) and a chronic, immune-mediated, demyelinating, neurodenegerative condition that persists for the remainder of the natural life of the animal. While post mortem tissue and motor disability are well-characterized in TMEV, structure and metabolite tissue damage associations are not thoroughly understood.

In this work, we studied the TMEV model over 2 months after the infection using advanced MRI with a cryogenic brain coil at 9.4 Tesla.

Animals: 23 SJL/J mice were infected with TMEV at 8 weeks of age and scanned scanned twice, directly before injection at 7 weeks of age (baseline) and again 8 weeks later. The study was approved by our Institutional Animal Care & Use Committee (IACUC).

Data acquisition: Experiments were performed on a 9.4 Tesla Bruker BioSpec 94/20 USR with a gradient coil providing 440 mT/m and a two-element transmit-receive ¹H cryogenically cooled MRI coil. Animals were anesthetized using 1-3% isoflurane under monitoring of respiration rate and body temperature. The protocol involved a 3D multi-echo gradient-echo (MGE) sequence (TR/TE₁/ΔTE=90ms/2.38ms/4.4ms, 9 monopolar echoes, FA=18°, 80μm isotropic resolution, TA=27min) and an optimized ultra-short echo time STEAM sequence (TE/TR/TM=2.6ms/2s/10ms, NEX=512, TA=17min) providing a spatial shift of less than 750μm. The STEAM sequence was applied in the cortex (1.1x2.9x1.2mm³), thalamus (2.1x1.6x1.1mm³), and the basal ganglia (BG; 2.0x1.5x1.2mm³).

Analysis: Volumetry was performed based on the MGE magnitude images (]all echoes averages) using a dedicated model-specific atlas-based approach. Metabolites were quantified using LCModel (v6.3) with water-reference and concentrations were normalized to creatine. A p<0.05 was considered statistically significant.

Volumes: We detected significantly (p<0.001) increased volumes of the cortex (+3.6%), BG (+9.9%), corpus callosum (+11.8%), and lateral ventricles (LV, +90%). The thalamus showed a decreased volume (-6.2%).

Creatine reference-concentration: Creatine did not change significantly between the timepoints, justifying the use of the metabolite as an internal reference for MRS.

Cortex-metabolites: Alanine was reduced (-21.5%; p=0.024), Glucose was increased (+80.3%; p=0.015).

BG-metabolites: GABA was reduced (-7.5%; p=0.04).

Thalamus-metabolites: NAA was increased (+5.2%; p=0.04).

This is the first characterization of the TMEV model with advanced quantitative MRI at 9.4T. The use of a cryogenic coil, highly optimized sequences, and sophisticated post-processing and analysis strategies allowed a region-specific assessment of the metabolite spectrum as well as an accurate volume determination in this model.

Reported increased regional volumes can be attributed to normal growth over 8 weeks.

Shrinkage of the Thalamus, especially considering possible normal growth (above), is strongly indicative for atrophy, which is also supported by the large increase of the LV.

The increase in glucose concentration could indicate a loss in cortex astrocytes, a down-regulation of glycolysis, decreased glutamatergic neuron activity, or a loss of glutamatergic neurons entirely.

Decreased alanine (usually associated to a change in cellular protein composition) could simply be a product of down-regulated glycolsis. If lactate is in high demand as an energy source in the cortex, glycolytic reactions may be regulated to produce more lactate from pyruvate, limiting alanine synthesis. Alternatively, glycolysis may be downregulated altogether.

Due to connectivity of these structures², decreased GABA in the BG could lead to increased activation and/or excitotoxicity of the interneurons between nuclei in the BG. In particular, it could lead to decreased inhibition of the thalamus and/or excitotoxicity in the thalamus (in line with the detected Thalamus-atrophy).

The increased thalamic NAA requires special attention, because NAA is traditionally localized within neurons and commonly considered a marker of neuronal viability. Increased NAA could be explained by developmental myelination (likely still ongoing after baseline-MRI), counteracting an atrophy-related NAA decrease. It has also been hypothesized that a dysfunctional neuron-to-oligodendrocyte NAA-transport or NAA-catabolism in oligodendrocytes can lead to an accumulation of NAA in neurons, e.g. caused by absence/dysfunction of mature myelinating oligodendrocytes³. Another explanation could be a massive apoptosis of differentiating oligodendrocytes followed by a proliferation of oligodendrocyte progenitor cells (OPC), which contain higher NAA concentrations than neurons³.