Probiotics Prevent Intestinal Barrier Dysfunction in Acute Pancreatitis in Rats via Induction of Ileal Mucosal Glutathione Biosynthesis Authors: Femke Lutgendorff, Rian M. Nijmeijer, Per A. Sandström, Lena M. Trulsson, Karl-Eric Magnusson, Harro M. Timmerman, L. Paul van Minnen, Ger T. Rijkers, Hein G. Gooszen, Louis M. A. Akkermans, Johan D. Söderholm Published: 2009, PLOS ONE DOI: 10.1371/journal.pone.0004512 ABSTRACT During acute pancreatitis (AP), oxidative stress contributes to intestinal barrier failure. We studied actions of multispecies probiotics on barrier dysfunction and oxidative stress in experimental AP. Fifty-three male Spraque-Dawley rats were randomly allocated into five groups: 1) controls, non-operated, 2) sham-operated, 3) AP, 4) AP and probiotics and 5) AP and placebo. AP was induced by intraductal glycodeoxycholate infusion and intravenous cerulein (6 h). Daily probiotics or placebo were administered intragastrically, starting five days prior to AP. After cerulein infusion, ileal mucosa was collected for measurements of E. coli K12 and 51 Cr-EDTA passage in Ussing chambers. Tight junction proteins were investigated by confocal immunofluorescence imaging. Ileal mucosal apoptosis, lipid peroxidation, and glutathione levels were determined and glutamate-cysteine-ligase activity and expression were quantified. AP-induced barrier dysfunction was characterized by epithelial cell apoptosis and alterations of tight junction proteins (i.e. disruption of occludin and claudin-1 and up-regulation of claudin-2) and correlated with lipid peroxidation (r>0.8). Probiotic pre-treatment diminished the AP-induced increase in E. coli passage (probiotics 57.4±33.5 vs. placebo 223.7±93.7 a.u.; P <0.001), 51 Cr-EDTA flux (16.7±10.1 vs. 32.1±10.0 cm/s10 −6 ; P <0.005), apoptosis, lipid peroxidation (0.42±0.13 vs. 1.62±0.53 pmol MDA/mg protein; P <0.001), and prevented tight junction protein disruption. AP-induced decline in glutathione was not only prevented (14.33±1.47 vs. 8.82±1.30 nmol/mg protein, P <0.001), but probiotics even increased mucosal glutathione compared with sham rats (14.33±1.47 vs. 10.70±1.74 nmol/mg protein, P <0.001). Glutamate-cysteine-ligase activity, which is rate-limiting in glutathione biosynthesis, was enhanced in probiotic pre-treated animals (probiotics 2.88±1.21 vs. placebo 1.94±0.55 nmol/min/mg protein; P <0.05) coinciding with an increase in mRNA expression of glutamate-cysteine-ligase catalytic (GCLc) and modifier (GCLm) subunits. Probiotic pre-treatment diminished AP-induced intestinal barrier dysfunction and prevented oxidative stress via mechanisms mainly involving mucosal glutathione biosynthesis. INTRODUCTION Multi-organ-failure and systemic inflammatory response syndrome (SIRS) remain major causes of mortality at intensive care units [1] . There is compelling evidence for an important role of the gut in the origin and development of critical illness [2] , [3] . Gut barrier dysfunction can propagate a pathophysiological state leading to increased mortality. Deitch et al. [4] demonstrated for example, that shock-induced intestinal hypoperfusion leads to release of reactive oxygen species (ROS) and oxidative stress resulting in barrier failure and release of pro-inflammatory mediators, enhancing a subsequent SIRS. Evidence suggests that ROS disrupt epithelial tight junctions (TJs) 5 , 6 leading to barrier dysfunction [7] . Furthermore, ROS cause epithelial cell apoptosis [8] contributing to mucosal barrier failure [9] – [11] and associated mortality [12] , [13] in experimental studies. Moreover, clinical evidence shows that increased intestinal apoptosis is a prominent event in patients who succumb from sepsis [14] . The mucosal barrier may be further compromised by overgrowth of enteric pathogens e.g. Escherichia coli [15] or by other opportunistic pathogens which switch on their virulence genes upon intestinal hypoxia [16] , suggesting an important role for intestinal microbiota in gut-derived sepsis [2] . Taken together in critically ill patients, SIRS may be driven by an oxidative stress-induced disruption of the equilibrium of the otherwise symbiotic three-way partnership between intestinal microbiota, epithelium, and immune system. Conversely, a moderate increase in intracellular ROS concentrations may paradoxically afford protection against oxidative stress via upregulation of oxidative defense mechanisms. Indeed, de novo synthesis of the most important endogenous antioxidant, glutathione (GSH) is found to be enhanced after low dose H 2 O 2 [17] and is also increased by other weak oxidative agents [18] . Severe acute pancreatitis (AP), which is characterized by intestinal barrier dysfunction and not seldom leading to SIRS, represents a clinical disease in which maintenance of this equilibrium is severely disturbed [3] . Since commensal bacteria are believed to be a crucial part of host homeostasis, recent studies have looked at effects of probiotics in recreating equilibrium [19] – [21] . Our group previously developed a probiotic combination designed to prevent infectious complications in critical illness based on anti-inflammatory and microbiota modulating capacities [22] . Five-day pre-treatment with these multispecies probiotics attenuated bacterial translocation and reduced the mortality in experimental AP in rats [23] , but recently we also demonstrated in a double-blind clinical study that these probiotics, contrary to any expectations, doubled the mortality compared with placebo in 298 patients with predicted severe AP [24] . These results painfully showed the need to study mechanisms of action of probiotics in critical illnesses. The objective of this study was to characterize the intestinal mucosal barrier in experimental AP and to explore mechanisms by which multispecies probiotics affect barrier function under these circumstances. We found that probiotics maintained the mucosal barrier in AP by up-regulation of the rate-limiting step in glutathione (GSH) biosynthesis. MATERIALS AND METHODS Male specific pathogen-free Sprague-Dawley rats (B&K, Sollentuna, Sweden, 250–350 g, 50–70 days of age) were maintained under constant conditions with a 12-hour light/dark cycle and free access to water and standard rat pellets. Rats were acclimatized for one week prior to surgery and randomly allocated into five groups: 1) non-operated controls (n = 5); 2) sham-procedure (n = 12); 3) AP (n = 12); 4) AP, placebo (n = 12); 5) AP, probiotics (n = 12). The experimental design ( fig 1 ) was in accordance with guidelines of the Linköping University Animal Welfare Committee, following European legislation (2003/65/EC). At the start of the experiment, animals were fitted with a gastric cannula, except for control animals. Probiotics and placebo were administered daily to the probiotics and placebo group, starting 5 days prior to induction of acute pancreatitis (AP). At day 0, AP or sham-procedure was performed. After the six hours of cerulein infusion, animals were anesthetized for removal of organ samples. PROBIOTICS The multispecies probiotics consisted of six viable, freeze-dried strains: Lactobacillus acidophilus (W70), Lactobacillus casei (W56), Lactobacillus salivarius (W24), Lactococcus lactis (W58), Bifidobacterium bifidum (W23), and Bifidobacterium lactis (W52) (previously classified as Bifidobacterium infantis ) (Ecologic® 641, Winclove Bio Industries, Amsterdam, the Netherlands). Placebo, that consisted of the carrier of the probiotic product, i.e. cornstarch and maltodextran, was packed in identical coded sachets to guarantee blinding. Before daily administration, probiotic or placebo formulas were reconstituted in sterile water, for 15 min at 37°C. A single probiotic dose of 1.0 ml contained 5×10 9 colony forming units (CFU) of bacteria. SURGICAL PROCEDURES Under general anesthesia (2% isoflurane), a permanent gastric cannula was fitted in all rats, as performed previously [23] . Animals were allowed to recover for four days prior to the start of daily probiotics or placebo administrations through the cannula and then after five days of pre-treatment AP was induced as described by Schmidt et al . [25] . Briefly, pressure controlled (35