Multiple sclerosis (MS) is an immune-mediated disease of the central nervous system that involves inflammation, demyelination, and neurodegeneration. Different therapeutic strategies aim at targeting these processes and brain MRI has helped greatly to understand the contribution of these processes to the pathology of MS. A harmonization of standardized MR data acquisition and analysis protocols with fine-tuned therapeutic protocols will be valuable to identify the exact time window for therapeutic action during disease progression. In this study we aimed to further understand the morphological changes in the brain that accompany encephalomyelitis. To mimic the relapsing-remitting form of MS (RRMS), we employed the SJL/J experimental autoimmune encephalomyelitis (EAE) animal model [1].

Animal experiments were carried out in accordance to guidelines from the Animal Welfare Department of the LAGeSo State Office of Health and Social Affairs Berlin. In two separate studies 7 female SJL/J mice (Janvier SAS) were immunized subcutaneously with 250μg proteolipid protein (Pepceuticals) and 800μg mycobacterium tuberculosis H37Ra (Difco) in 200μL emulsion containing equal volumes of phosphate/buffered saline (PBS) and complete Freunds adjuvant (BD-Difco). On days 0 and 2, 200ng pertussis toxin (List Biological Laboratories) was administered intraperitoneally. Mice were imaged on day -1, 5, 8, 11, 13, 15, 18, 20, 26, 29, 32, 34, 36, 39, 43, 46, 48, 50 and 64 after EAE induction and weighed and scored daily [2]. MRI was performed using a 9.4 Tesla animal scanner (Biospec 94/20 USR, Bruker Biospin) and an in-house built shingled-leg mouse brain birdcage coil [3]. Mice were placed on a water-circulating heated holder to ensure constant body temperature (37°C) and kept anesthetized using a mixture of isoflurane 1–1.5% (Abbott GmbH & Co. KG), air and oxygen. Body temperature and breathing rate were constantly monitored (PCSAM, SA Instruments). Horizontal fat-suppressed turbo spin echo T₂-weighted TurboRARE (TE=14.345ms, TR=3000ms, matrix=512×512, in plane resolution=32μm, repetitions=16, slices=15, thickness=500μm, acquisition time=33min 36s) brain images were acquired. Slice positioning was kept fixed through the longitudinal brain examination: horizontal slices were positioned parallel to the base of the brain. Quantification of ventricle size was performed using FSL 5.0 (FMRIB’s Software Library). The T₂-weighted images were corrected for bias field inhomogeneity. Non-brain tissue was cleared using the brain extraction tool (BET). The images were registered to a reference brain with FLIRT and FNIRT. An inverse transformation matrix was generated and applied to the reference ventricle mask. The ventricle volume was calculated and the volume change was estimated as a ratio of the ventricle volume to pre immunization ventricle volume.

In the first EAE study, 5 EAE mice out of 6 started showing clinical symptoms on average 11 days following EAE induction with an average maximum score of 2.75 (Fig. 1). On average, all mice started losing weight on day 11 (Fig. 1). We followed brain modifications in 3 EAE mice. In all cases we observed a clear increase in cerebral ventricle size on T₂-weighted images (Fig. 2 A) upon clinical manifestation with ensuing fluctuations during the next relapses. The increase in ventricle size was only detected 11 days following immunization, on average 1 day prior to symptom start. The ventricle size of all mice started returning to normal upon remission of disease, in some instances prior to this (i.e. at peak of disease). Interestingly, the next ventricle enlargements occurred in parallel to the next clinical exacerbations. In this first cohort of animals, the pronounced second and third ventricular expansions (Fig. 2 A) sometimes occurred at earlier/later time-points or to a lesser extent, similar to clinical disease progression, thereby preventing clear-cut second/third ventricle size peak quantification upon averaging all data (Fig. 2 B).In this study we could confirm our previous results of an early ventricular enlargement prior to disease manifestation and added on to our current knowledge, by showing a dynamic fluctuation in ventricle size that runs ahead of relapses and remissions during disease progression [2]. We could identify these findings by following ventricle size for a long period of time (64 days) during the progression of encephalomyelitis. We hypothesize that standard therapeutic compounds used in MS and its animal model exert their therapeutic benefit during different time points of disease progression. Micro and macro MR observations alongside neurological symptom assessment will help us determine the best therapeutic window. Further work needs to be done to identity the appropriate MR methods and computational analysis to deliver a clear-cut therapeutic assessment during the different stages of the pathology.