Multiple sclerosis (MS) is characterized by focal demyelinated inflammatory lesions in the brain and spinal cord [1]. One key event occurring early in the MS pathogenesis is a breach in blood brain barrier (BBB) integrity, even in areas of normal-appearing white matter and especially around periventricular regions [2]. Previously, we reported on an enlargement of cerebral ventricles in experimental autoimmune encephalomyelitis (EAE), an animal model that mimics MS pathogenesis [3]. We observed ventriculomegaly prior to clinical disease manifestation. In this study we investigated the kinetics of BBB leakage in relation to ventricle size changes during the pathogenesis of a relapsing-remitting EAE model. Animal experiments were carried out in accordance with the Animal Welfare Department (Office of Health and Social Affairs, Berlin). EAE was induced in female SJL/J mice (Janvier SAS) following immunization with PLP (250µg, Pepceuticals Ltd) and mycobacterium tuberculosis H37Ra (800µg, Difco). Pertussis toxin (200ng, List Biological Laboratories) was administered intraperitoneally. Mice were daily weighed and scored: 0 (no disease); 1 (tail weakness, righting reflex weakness); 2 (paraparesis); 3 (paraplegia); 4 (paraplegia with forelimb weakness or paralysis) or 5 (dead). Mice with a score of 2.5 or more received intraperitoneal glucose and mice with a score of 3 (>24hr) were sacrificed. Pre- and post-contrast MR imaging and T₁ mapping were performed on day -2, 5, 8, 11, 13, 15, 18, 20 following immunization, using a 9.4 Tesla animal scanner (Biospec 94/20 USR, Bruker Biospin) fitted with warm water circulation (for constant body temperature) and gas anesthesia (isoflurane 1–1.5% mixture in air and O2). A ¹H birdcage coil specially constructed for mouse brain imaging and ideal for T₁ brain mapping was employed [4]. The contrast agent gadolinium-diethylenetriamine-pentaacetate (300 nmol/g Gd-DTPA, Magnevist®, Bayer-Schering) was i.v. infused (over 2min) during each imaging/mapping session. A Modified Driven-Equilibrium Fourier Transformation sequence was used for pre- and post-contrast T₁-weighted imaging (3D MDEFT: TR/TE/TI=2600/3.9/950ms, FA=20°, FOV=19.2×12.8mm², matrix=256×170, slices/thickness=15/500µm). Horizontal entire brain MR images were performed in 2min 5sec. A RARE sequence with variable repetition time (VTR) was used to generate T₁ maps (RARE-VTR: TE=11.53ms, matrix=128×85, slices/thickness=11/500µm, VTR=0.38s, 0.55s, 0.94s, 1.48s, 2.40s and 7.00s, RARE factor=4). Slice positioning was kept fixed throughout study; horizontal slices were positioned parallel to the base of the brain. For histological assessment of pre-onset EAE, mice underwent terminal anesthesia and transcardial perfusion with PBS and zinc fixative. Following extraction, brains were post-fixed in zinc, embedded into paraffin, cut into 5-µm sections, and stained with hematoxylin and eosin according to standard procedures. Analysis of T₁ maps was performed using an in-house software developed in Matlab (MathWorks Inc.). In this study, all mice (n=6) showed clinical symptoms of EAE, on average 9 days following immunization, with an average maximum score of 2.5. We performed pre- and post-contrast T₁-weighted imaging and T₁ mapping in 4 EAE mice (2 mice had to be excluded due to animal welfare reasons). In all cases we observed a clear increase in cerebral ventricle size, which was preceded by an earlier disruption of the BBB as observed in contrast-enhanced MDEFT images (Fig. 1A) and T₁ maps (Fig. 1B), as leakage of contrast agent into the CNS parenchyma. The increase in ventricle size was detected on average 8 days and contrast-enhanced BBB disruption already 5 days following immunization. In histological examinations of a separate cohort of pre-onset EAE mice (day 6 post-immunization), changes in blood vessel architecture were detectable (Fig. 1C). Interestingly, partial normalization of BBB integrity (on day 8) also appears to precede the normalization of ventricle size (Fig. 1B). In this preliminary study we showed that the integrity of the BBB is compromised even earlier than the incidence of ventriculomegaly, which already occurs prior to the occurrence of neurological symptoms. In a spontaneous relapsing-remitting mouse EAE model, it was also recently indirectly shown by histology that focal disturbances in the BBB occur early during the pathogenesis of EAE, prior to the formation of neuroinflammatory lesions [5]. Our results also suggest a partial renormalization and reappearance of BBB disruptions throughout the disease course, particularly in the cerebellum and midbrain regions, and that these changes appear to occur prior to the normalization and re-expansion of ventricle size. Pattern recognition within micro and macroscopic MR changes in the brain might have implications on the decisions made to treat patients since an early identification or even anticipation of the next acute exacerbation of the disease could determine or influence the therapeutic choices that will made.