Recent research brings more evidence that dysbiosis accompanies many diseases: atherosclerosis, diabetes, obesity, colon cancer, chronic liver disease (CLD)^1-3 and plays an important role in their evolution/progression. Consequently appropriate modulation of intestinal flora could improve the course or prognosis of patients with CLD by reducing pro-inflammatory mechanisms in the gut, intestinal permeability and consequent bacterial translocation as well as by decreasing intestinal ammonia production (main toxin in CLD-induced Hepatic Encephalopahy, HE). In addition, inflammation, acting alone or in concert with ammonia, may play an important role in HE. Therefore, we aimed to investigate potential therapeutic effect of probiotic treatment with anti-inflammatory proprieties (VSL#3) in CLD-induced HE.Bile duct ligated (BDL) rats were used to study CLD-induced HE^4,5. Probiotic treatment (VSL#3) started 2 weeks before BDL-surgery ('week -2') and was administrated daily by gavage (60 billion bacteria/kg of rat) during the whole study. 38 adult Wistar male rats were studied, divided into 4 groups: BDL-rats without treatment (‘BDL’,n=14), BDL-rats+VSL#3 (‘BDL+VSL#3’,n=14), sham-operated animals without treatment (‘sham’,n=5) and shams+VSL#3 (‘sham+VSL#3’,n=5). Longitudinal in vivo ¹H-MRS was performed on a 9.4T system (Varian/Magnex Scientific) using a home-built 14mm diameter quadrature ¹H-coil as a transceiver. First and second order shims were adjusted using FASTMAP (linewidth=9-11Hz). A VOI of 2×2.8×2mm³ localized in the hippocampus (due to its role in memory - symptom in HE) was measured using the SPECIAL⁶ spectroscopy sequence (TE/TR=2.8/4000ms,160averages). Metabolite concentrations were calculated by LCModel using water as reference. ¹H-MRS and plasma measurements of bilirubin were performed before the BDL-surgery ('week 0') and 2,4,6 and 8 weeks after BDL. Open field test (motor activity) was performed at 4,6 and 8 weeks after BDL-surgery to evaluate the degree of HE⁷. Faeces were collected before the treatment and after every two weeks for microbiota analysis.Plasma measurements of bilirubin confirmed the presence of CLD in both, BDL+VSL#3 and BDL rats, with no differences between the two groups (Fig.1A). BDL rats showed brain glutamine (Gln) increase (due to ammonia detoxification) and subsequent diminution of other osmolytes (mIns,Tau,tCho) as an osmoregulatory, often incomplete response (i.e. continuous increase of osmolyte sum and presence of mild brain edema)^8-10. In this study, ¹H-MRS revealed a significantly smaller increase in brain Gln in BDL+VSL#3 (+107%) compared to BDL (+138%) at week8 (p=0.03) together with a constant sum of osmolytes in BDL+VSL#3 (Fig.2D,F). This pattern is also visible in Fig.3. Differences among other brain metabolites are presented in Fig.2. The increase of genus Bifidobacteria in BDL+VSL#3 is shown in Fig.1B. In addition, a negative correlation between increase of Bifidobacteria at week4 and increase of Gln at week6 in BDL+VSL#3 was found (r=-0.64, p=0.05), suggesting a positive treatment effect. There was no significant difference in behavioural test performance between BDL+VSL#3 and BDL at week8. However, the course of the disease varied between both groups: BDL+VSL#3 did not show a significant decrease in motor activity while the BDL group did (Fig.4A,B). This finding is in agreement with the negative correlation between Gln and motor activity (r=-0.52, p=0.004, valid in BDL+VSL#3 as well as in BDL), suggesting that increased Gln is associated with the decrease in motor activity. Interestingly, ~30% of treated animals did not display the characteristic increase in brain Gln (Fig.2C). They also showed slight diminution in the sum of brain osmolytes (Fig.2D) and very stable performance in behavioural tests (Fig.4C,D), suggesting that in some individuals, VSL#3 may be protective. Fig.5 shows the relation between Gln and mIns/osmolytes in different groups. Since bacterial load could contribute to this effect, we looked at the amount of Bifidobacteria in 'low-Gln' compared to other BDL+VSL#3 rats: the most striking difference was at week4 (Fig.1C). Regarding the amount of Bifidobacteria in 'low-Gln' compared to other BDL+VSL#3 rats, the most glaring difference was at week4 (Fig.1C) that could stand behind the difference in brain Gln. Fig.5 shows the relation between Gln and mIns/osmolytes in different groups.In BDL rats VSL#3 treatment was associated with positive effect on brain Gln, osmotic equilibrium and motor activity. Some individuals showed no increase in Gln during the whole course of disease and displayed significantly better performance on behavioural tests. These results are promising. However further studies in which probiotics would act synergistically with non-absorbable disaccharides (e.g. lactulose¹¹), or poorly or non-absorbable antibiotics (e.g. rifaximin¹²) are needed to further potentiate the effect of treatment. Additionally, probiotics could be used in an alternating fashion with antibiotics in order to avoid multiresistant organisms¹³. Nevertheless probiotics bring the advantage of very mild treatment, easy administration, possibility to take on long term and represent a safer, better tolerated therapy¹⁴.