Dig Dis Sci DOI 10.1007/s10620-015-3770-1

ORIGINAL ARTICLE

Lactobacillus paracasei Induces M2-Dominant Kupffer Cell Polarization in a Mouse Model of Nonalcoholic Steatohepatitis Won Sohn1 • Dae Won Jun2 • Kang Nyeong Lee2 • Hang Lak Lee2 Oh Young Lee2 • Ho Soon Choi2 • Byung Chul Yoon2

•

Received: 3 October 2014 / Accepted: 15 June 2015 Ó Springer Science+Business Media New York 2015

Abstract Background and Aims Gut microbiota may be associated with the pathogenesis of nonalcoholic steatohepatitis (NASH). This study aimed to investigate the protective effects and possible mechanisms of Lactobacillus paracasei on NASH. Methods Thirty male C57BL/6 mice were randomized into three groups and maintained for 10 weeks: control group (standard chow), NASH model group (high fat ? 10 % fructose diet), and the L. paracasei group (NASH model with L. paracasei). Liver histology, serum aminotransferase levels, and hepatic gene expression levels were measured. Intestinal permeability was investigated using urinary 51Creatinine Ethylenediaminetetraacetic acid (51Cr-EDTA) clearance. Total Kupffer cell counts and their composition (M1 vs. M2 Kupffer cells) were measured using flow cytometry with F4/80 and CD206 antibodies. Results Hepatic fat deposition, serum ALT level, and 51 Cr-EDTA clearance were significantly lower in the L. paracasei group than the NASH group (p \ 0.05). The L. paracasei group had lower expression in Toll-like receptor4 (TLR-4), NADPH oxidase-4 (NOX-4), tumor necrosis factor alpha (TNF-a), monocyte chemotactic protein-1

Electronic supplementary material The online version of this article (doi:10.1007/s10620-015-3770-1) contains supplementary material, which is available to authorized users. & Dae Won Jun [email protected] 1

Department of Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea

2

Department of Internal Medicine, Hanyang University School of Medicine, 17 Haengdang-dong Sungdong-gu, Seoul 133-792, Korea

(MCP-1), interleukin 4 (IL-4), peroxisome proliferator activated receptor gamma (PPAR-c), and PPAR-d compared with the NASH group (p \ 0.05). The total number of F4/80? Kupffer cells was lower in the L. paracasei group than the NASH group. L. paracasei induced the fraction of F4/80?CD206? cells (M2 Kupffer cells) while F4/80?CD206- cells (M1 Kupffer cells) were higher in the NASH group (F4/80?CD206? cell: 44 % in NASH model group vs. 62 % in L. paracasei group, p \ 0.05). Conclusions Lactobacillus paracasei attenuates hepatic steatosis with M2-dominant Kupffer cell polarization in a NASH model. Keywords Nonalcoholic steatohepatitis  Kupffer cells  Lactobacillus  Probiotics Abbreviations NAFLD Nonalcoholic fatty liver disease TLR Toll-like receptors NASH Nonalcoholic steatohepatitis L. paracasei Lactobacillus paracasei AST Aspartate aminotransferase ALT Alanine aminotransferase H&E Hematoxylin and eosin stain NOX NADPH oxidase Cr Creatinine EDTA Ethylenediaminetetraacetic acid CD Cluster of differentiation TNF-a Tumor necrosis factor alpha MCP-1 Monocyte chemotactic protein-1 IL Interleukin PPAR-c Peroxisome proliferator activated receptor gamma PPAR-d Peroxisome proliferator activated receptor delta

123

Dig Dis Sci

RT-PCR GAPDH SEM COX

Reverse transcription polymerase chain reaction Glyceraldehyde-3-phosphate dehydrogenase Standard error of the mean Cyclooxygenase

Introduction The spectrum of nonalcoholic fatty liver disease (NAFLD) histologically varies from simple steatosis to cirrhosis, including steatohepatitis [1]. The pathogenesis of NAFLD is associated with free fatty acids, insulin resistance, oxidative stress, lipotoxicity, and others [2]. Recently, the intestinal microbiota was highlighted as having an important role in the pathogenesis of NAFLD [3]. The ‘‘gut–liver axis,’’ the interaction between the gut and liver, is an essential player in chronic liver disease, including NAFLD, because various lipid metabolites and bacterial products flow into the liver via the portal vein [4]. Kupffer cells are the resident macrophages of the liver and are involved in hepatic injury, inflammation, infection, immune response, ischemia, and stress [5]. Macrophage has a crucial role in the pathogenesis of obesity including NAFLD. Adipose tissue macrophages accumulate in obesity and control inflammation and insulin resistance [6]. Macrophages are classified according to their phenotype and function. Classically activated macrophages (M1 macrophage) play a role in defense of the host from bacteria, viruses, and protozoa. Meanwhile, alternatively activated macrophages (M2 macrophage) have a function as the anti-inflammation and the regulation of wound healing [7]. There is more prominent activation of M1 macrophages than M2 macrophages in obesity, while the M2 macrophage is more dominant than the M1 macrophage in the lean condition. As such, the composition of M1 and M2 macrophages changes during various disease states. The change in macrophage phenotypes is called ‘‘macrophage polarization’’ [8]. The polarization and regulation of adipose tissue macrophages is well known in obesity [9]; however, there is a lack of data on the polarization and regulation of Kupffer cells in NAFLD or NASH. Probiotics are live microorganisms that are beneficial to a person’s health. They are widely used for gastrointestinal diseases such as antibiotic-associated diarrhea, irritable bowel syndrome, inflammatory bowel disease, and others. However, there is a lack of data on the effect of probiotics on Kupffer cells in the pathogenesis of NAFLD or NASH [10]. Lactobacillus paracasei is one of the lactic acid bacteria species that are used in the fermentation of food and probiotics. L. paracasei enhances the phagocytic

123

activity of macrophage [11] and it has immune-modulatory properties to prevent intestinal inflammation [12]. Furthermore, L. paracasei decreased fat storage in high-fat fed mice and reduced oxidative stress and hepatic inflammation in obese mice [13, 14]. The aim of this study was to investigate the effect of probiotics on NASH in regards to the polarization and regulation of Kupffer cells. We examined the anti-inflammatory and anti-obesity effects of L. paracasei with regard to alteration in Kupffer cell composition (M1 vs. M2 Kupffer cells) in a NASH mouse model.

Materials and Methods Animals and Treatments Thirty male C57BL/6 mice (Oriental Yeast, Tokyo, Japan), aged 6–8 weeks, were housed in an individual pathogenfree barrier facility. All procedures in this study were approved by the Institutional Animal Care and Use Committee of Hanyang University (HY-IACUC-11-007). After a 1-week adaptation period, a total of 30 mice were randomized into three groups (each group n = 10): control group (fed with standard chow), NASH model group (high fat (40 %) ? 10 % fructose diet), and L. paracasei group (NASH model with L. paracasei supplement). The mice in each group were maintained for 10 weeks with the planned foods or probiotics. L. paracasei LPC4 (KCTC 11866BP) was administered to mice with a dosage of 1 9 108 viable cells in a lyophilized powder. To monitor the amount of food and water intake, we kept a diary of the amount of food intake or supplied probiotics for a whole period. L. paracasei was administered into mice using a gastric gavage needle without drinking water. The amount of food intake or supplied probiotics was the same for each group. Body weight was checked once a week. Aminotransferase Activity Serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels were measured using an autonomic biochemical analytical system (Hitachi-747; Hitachi, Tokyo, Japan). Liver Histology Thin slices of harvested liver specimens were fixed with 10 % neutral formalin. Formalin-fixed liver sections embedded in paraffin were stained with H & E (hematoxylin and eosin). Liver histology of the tissue was evaluated by one pathologist who has over 10 years of

Dig Dis Sci

experience in the interpretation for liver histology and pathology. He scored the liver histology without knowing any information about the experiment condition (i.e., the group of mouse). To minimize the intra-observer variation, liver histology was assessed based on the mean value regarding at least six different areas in high power field and all scoring of tissue histology was performed on the same day. The criteria on hepatic steatosis and inflammatory cell infiltration were based on Standardized Guideline Proposed by the Korean Study Group for the Pathology of Digestive Diseases [15]. Hepatic fat accumulation (steatosis) and inflammatory cell infiltration were assessed on each section with the following scoring criteria. Hepatic steatosis was scored from 0 to 3: score 0, no fat; score 1, fatty hepatocytes occupying less than 33 % of the hepatic parenchyma; score 2, fatty hepatocytes occupying 34–66 % of the hepatic parenchyma; score 3, fatty hepatocytes occupying more than 66 % of the hepatic parenchyma. Inflammatory cell infiltration was scored from 0 to 2: score 0, none; score 1, up to 4 foci/field; score 2, [4 foci/field. Immunohistochemistry The expression of F4/80, the macrophage cell surface marker, was immunohistochemically investigated to assess the infiltration of Kupffer cells in liver tissue. Anti-mouse F4/80 antibody (BD Bioscience Mountain View, CA, USA) was diluted 1:50 for 30 min, and each section was incubated with the F4/80 antibody. In addition to F4/80, immunohistochemical staining for Toll-like receptor 4 (TLR-4) and NADPH oxidase 4 (NOX4) was performed in liver tissue. Polyclonal anti-TLR4 antibody (dilution 1/50; Abcam, Cambridge, UK) and polyclonal anti-NOX4 antibody (concentration of 5 lg/ml; Abcam, Cambridge, UK) were both made in rabbits. The staining of F4/80, TLR-4, and NOX-4 was assessed by the following scoring criteria. The mean intensity of staining was scored semiquantitatively from 0 to 3: score 0, none stained; score 1, mild intensity of staining; score 2, moderate intensity of staining; score 3, high intensity of staining. Intestinal Permeability Intestinal permeability was performed to assess epithelial barrier function. It was quantified by measuring the 51CrEDTA clearance of blood and urine in the small intestine of mice [16]. A jugular cannulation was done to administer 51 Cr-EDTA. The bilateral renal hili were ligated to prevent the excretion and loss of 51Cr-EDTA into the urine. A small intestine loop was fitted with both inflow and outflow tubes and consistently perfused with Tyrode’s solution

(0.25 ml/min). For 90 min, the collection of luminal perfusate was done at 10-min intervals with 20-min equilibration time. Thereafter, the loop of small intestine was removed. A Cobra II autogamma counter (Packard, Virginia Beach, VA, USA) was used to evaluate the activity of 51 Cr-EDTA in luminal and plasma perfusates. Intestinal permeability was evaluated as the plasma to lumen clearance of 51Cr-EDTA (in ml/min/100 g tissue 9 103). The calculation was performed in the following manner: clearance = (counts/min/ml in perfusate) 9 (perfusion rate in ml/min) 9 100/(counts/min/ml in plasma 9 weight of the intestinal segment in grams). Flow Cytometric Analysis Liver tissue was minced and digested in DME/HEPES (10 ml/2 g) with 10 mg/ml BSA (Calbiochem, San Diego, CA, USA), 60 U/ml DNAse I (Sigma-Aldrich, St. Louis, MO, USA), and 35 lg/ml liberase blendzyme 3 (Roche, Indianapolis, IN, USA) in a 37° water bath. The stromovascular cell pellet was treated with a red cell lysis solution (155 mmol/l NH4Cl, 90 lmol/l EDTA, and 10 mmol/l KHCO3), which was applied to a 70-lm filter and treated with a fluorescence-activated cell sorting (FACS) buffer (PBS with 5 mmol/l EDTA and 0.2 % fatty acid–poor BSA). Flow cytometry (BD FACS Calibur, CA, USA) was used to perform cell sorting for the immunofluorescent method. Flow cytometric identification of the cells was performed using simultaneous labeling with an APClabeled antibody against F4/80 (BD Biosciences, Mountain View, CA, USA), and an Alexa Flour 647-labeled antibody against CD206 (BD Biosciences, Mountain View, CA, USA). Semi-Quantitative RT-PCR For complementary DNA synthesis, 3 lg of RNA samples were reverse-transcribed using a PrimeScript 1st strand cDNA Synthesis Kit (Takara Bio, Otsu, Japan). The tissue for mRNAs was harvested at the same time of the day (early morning). The hepatic mRNA expression of tumor necrosis factor alpha (TNF-a), monocyte chemotactic protein-1 (MCP-1), interleukin 4 (IL-4), peroxisome proliferator activated receptor gamma (PPAR-c), and PPAR-d was measured by semi-quantitative reverse transcription (RT)-PCR using glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a reference. Total RNA was extracted from liver tissues by TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) following the manufacturer’s protocol. TRIzol reagent (Invitrogen Life Technologies, Carlsbad, CA, USA) was used to extract total RNA from liver tissue. The reaction was performed at

123

Dig Dis Sci

42 °C for 1 h, and thereafter heat inactivation of the enzymes was performed at 95 °C for 5 min. The cDNA was amplified using Taq polymerase (Universal PCR, Invitrogen Life Technologies, Carlsbad, CA, USA). The reaction mixture was pre-denatured at 95 °C for 5 min, and thereafter a cyclic procedure was repeated for 40 cycles (denaturation process at 95 °C for 30 s, annealing process for 30 s, and elongation process for 40 s). The procedure was performed using an Eppendorf Mastercycler gradient thermal cycler (Eppendorf Scientific, Westbury, NY, USA). A final extension process was performed at 72 °C for 5 min. RT-PCR products were electrophoresed on a 1.5 % (w/v) agarose gel, the gel was stained with ethidium bromide, and bands were visualized by ultraviolet light. We presented the results of RT-PCR in the following manner. The amount of mRNA for each hepatic gene was standardized to that for GAPDH in each sample. Finally, the relative expression of mRNA for each gene was calculated as compared with the relative mRNA value of hepatic genes to GAPDH in the control group. Real-Time PCR Real-time PCR was performed in white 96-well plates on a LightCycler 480 instrument with LightCycler 480 SYBR Green I Master (Roche Diagnostics, Mannheim, Germany). The hepatic mRNA expression of MCP-1, TNF-a and PPAR-c was measured using real-time PCR. Expression levels of each sample were normalized relative to expression of b-actin. Data were analyzed using the LightCycler 480 software, release version 1.5.0 (Roche Diagnostics, Mannheim, Germany). The primers sequences were as following: MCP-1 forward, 50 -GAA GCT GTA GTT TTT GTC ACC AAG C-30 ; reverse, 50 -TTT AAT GTA TGT CTG GAC CCA TTC C-30 ; TNF-a forward, 50 -ACC CCT TTA CTC TGA CCC CTT TAT T-30 ; reverse, 50 -TGA GCC ATA ATC CCC TTT CTA AGT T-30 ; PPAR-c forward, 50 -GAG TAT GCC AAA AAT ATC CCT GGT T-30 ; reverse, 50 -TGA TCT CAT GGA CAC CAT ACT TGA G-30 ; b-actin forward, 50 -ATC TTG ATC TTC ATG GTG CTA GGA-30 ; reverse 50 -GAC AGG ATG CAG AAG GAG ATT ACT G-30 .

SPSS for Windows release 18.0 (SPSS Inc, Chicago, IL, USA). All significance tests were two-sided, and significance was less than 5 %.

Results Liver Histology and Serum Aminotransferase Level Body Weight None of the 30 mice died during the experiment. After 10 weeks, the mean body weight in the NASH model group and the L. paracasei group were higher than the control group at week 11 (26 ± 0.6 vs. 38 ± 1.5 g, p \ 0.05), while the L. paracasei group did not have a lower body weight than the NASH model group (38 ± 1.5 vs. 43 ± 1.7 g, p [ 0.05) (Fig. 1). The amounts of food intake in the NASH model group and the L. paracasei group were measured weekly (Supplement Table 1). There was no statistical difference in the amount of food intake between two groups (p [ 0.05). In this study, the change of body weight was not checked between the mice treated with normal chow and those treated with normal chow and L. paracasei. Instead, we investigated the change of body weight in rats treated with normal chow and those treated with normal chow and L. paracasei (our internal preceding data). There was no statistical difference in body weight

Statistical Analysis All results were described as mean ± SEM (standard error of the mean). One-way analysis of variances with the post hoc test of Tukey was performed for the differences between the three groups. The data were tested using Levene’s test for the homogeneity for variances. Unless the values followed a normal distribution or equal variance, the Kruskal–Wallis test was followed by Bonferroni multiple comparisons. All data were statistically analyzed using

123

Fig. 1 Changes of body weight in the control, NASH model, and L. paracasei groups over 11 weeks. After a 1-week adaptation period, a total of 30 mice were appropriately fed in each group for 10 weeks. Body weight in the NASH model group and L. paracasei group were higher than in the control group at week 11 (26 ± 0.6 vs. 38 ± 1.5 g, *p \ 0.05), while the group with L. paracasei did not have a lower body weight than the NASH model group (38 ± 1.5 vs. 43 ± 1.7 g, *p [ 0.05)

Dig Dis Sci

Fig. 2 Effect of L. paracasei on hepatic histology in NASH (H & E staining, 9200 magnification). After 10 weeks, hepatic steatosis and inflammatory cell infiltration were histologically evaluated in the control (a), NASH model (high fat and high fructose) (b), and NASH with L. paracasei groups (c). Hepatic steatosis was graded as 0 (none), 1 (\33 % high power field (HPF)), 2 (33–67 % HPF), or 3 ([67 % HPF). Inflammatory cell infiltration was graded as 0 (none), 1 (B4 foci/HPF), or 2 ([4 foci/HPF). While the NASH model group

had more hepatic fat than the control group (2.0 ± 0.3 vs. 1.0 ± 0, p \ 0.05), steatosis in the L. paracasei group was significantly lower than in the NASH model group (2.0 ± 0.3 vs. 1.6 ± 0.2, p \ 0.05). There were no significant differences in inflammatory cell infiltration between the NASH model group and the L. paracasei group. Arrows indicate the areas of inflammatory cell infiltration. Data are presented as mean ± SEM

Fig. 3 Effects of L. paracasei on ALT (a) and AST (b) activities in NASH. After 10 weeks, serum ALT levels were higher in the NASH model than the control group unlike AST level (*p \ 0.05). Meanwhile, ALT levels in the L. paracasei group were significantly lower than in the NASH model group unlike AST level (*p \ 0.05). Data are presented as mean ± SEM

between two groups (normal chow vs. normal chow with L. paracasei) after 12 weeks (Supplement Table 2).

ALT, there was no difference in AST levels between the three groups (p [ 0.05; Fig. 3).

Liver Histology

Intestinal Permeability

Histologic analysis of fat deposition and inflammatory cell infiltration in the liver was performed. While the NASH model group had much more hepatic fat compared with the control group (2.0 ± 0.3 vs. 1.0 ± 0, p \ 0.05), the L. paracasei group showed lower levels of liver steatosis than the NASH model group (2.0 ± 0.3 vs. 1.6 ± 0.2, p \ 0.05). However, there were no significant differences in inflammatory cell infiltration between the NASH model group and the L. paracasei group histologically (Fig. 2).

To assess the effect of L. paracasei on intestinal permeability, the 51Cr-EDTA clearance test was performed (Fig. 4). While the urinary excretion of 51Cr-EDTA in the NASH model group was significantly higher than the control group, that of 51Cr-EDTA was significantly lower in the L. paracasei group compared with the NASH model group (1107.4 ± 134.9 vs. 1794.6 ± 227.5 ml/min/100 g tissue 9 103, p \ 0.05). Total Kupffer Cell Counts and Their Phenotypes

Serum ALT and AST The L. paracasei group had significantly lower levels of serum ALT compared with the NASH model group (135.5 ± 25.3 vs. 92.9 ± 21.8 U/l, p \ 0.05). Unlike

Immunohistochemical staining using the Kupffer cell marker F4/80 was performed in liver tissue (Fig. 5). Compared with the control group, the staining for F4/80? Kupffer cells was significantly more intense in the NASH

123

Dig Dis Sci

fraction of F4/80?CD206- cells compared with the NASH model group (e.g., F4/80?CD206?: 44 % in NASH model group vs. 62 % in L. paracasei group, p \ 0.05). Toll-Like Receptor 4 and NADPH Oxidase 4

Fig. 4 Effect of L. paracasei on intestinal permeability by 51Cr-EDTA clearance in NASH. After 10 weeks, intestinal permeability was evaluated by the 51Cr-EDTA clearance test. While the 51Cr-EDTA clearance in the NASH model was higher than the control group (1794.6 ± 227.5 vs. 941.0 ± 178.5 ml/min/100 g tissue 9 103, *p \ 0.05), the L. paracasei group had lower urinary excretion of 51 Cr-EDTA compared with the NASH group (1107.4 ± 134.9 vs. 1794.6 ± 227.5 ml/min/100 g tissue 9 103, *p \ 0.05). Data are presented as mean ± SEM

model group. However, F4/80? Kupffer cells in the L. paracasei group were less stained than in the NASH mode group. Flow cytometry was performed to examine Kupffer cell infiltration of the liver in more detail. Anti-F4/80 and antiCD206 antibodies were used to identify M1 and M2 Kupffer cells. The cell count and fraction of Kupffer cells were demonstrated in Fig. 6. The fraction of total F4/80? Kupffer cells was significantly higher in the NASH model group than the control group (32 vs. 16 % of total stromavascular fraction cells, p \ 0.05). Compared with the control group, the CD206- fraction of F4/80? Kupffer cells was high and F4/80?CD206? cells were low in the NASH model group (e.g., F4/80?CD206-: 25 % in the control group vs. 56 % in the NASH model group, p \ 0.05). Meanwhile, the L. paracasei group showed an increased fraction of F4/80?CD206? cells and a decreased

Fig. 5 Immunohistochemical staining of F4/80 for Kupffer cells (9200 magnification). a Control group; b NASH model group; c L. paracasei group. Compared with the control group, staining of F4/80 was significantly more intense in the NASH model group (p \ 0.05).

123

The expression of TLR-4 and NOX-4 was evaluated by immunohistochemistry in the liver (Fig. 7). Compared with the control group, the staining of TLR-4 was significantly more intense in the NASH model group; however, TLR-4 in the L. paracasei group was less stained than in the NASH model group. The expression pattern of NOX-4 was similar to TLR-4 between the control, NASH model, and L. paracasei groups. TNF-a, MCP-1, IL-4, PPAR-c, and PPAR-d Several inflammatory cytokines were measured using RTPCR in the three groups (Fig. 8). Hepatic expression of TNF-a and MCP-1 was significantly higher in the NASH model group compared to the control group. However, the L. paracasei group showed significantly lower expression of TNF-a and MCP-1 than the NASH model group. Also, the hepatic expression of IL-4 was lower in the L. paracasei group than the NASH model group. Compared with the control group, the activity of PPAR-c in the liver was significantly higher in the NASH model group; however, PPAR-c activity in the L. paracasei group was significantly lower than in the NASH model group. PPAR-d activity was similar to PPAR-c in the control group, NASH model group, and L. paracasei group. The hepatic mRNA expression of MCP-1, TNF-a, and PPAR-c was measured using real-time PCR (Fig. 9). While the mRNA expressions of MCP-1 and PPAR-c were significantly increased in NASH group compared to control group (p \ 0.05), the expressions of those were significantly decreased in L. paracasei group compared to NASH group

However, the expression of F4/80 in the L. paracasei group was significantly lower than in the NASH model group (p \ 0.05). Arrows indicate the areas of F4/80 (?) staining

Dig Dis Sci

Fig. 6 Effect of L. paracasei on Kupffer cell polarization in NASH. Cells in the stroma-vascular fraction (SVF) of liver tissue from each group of mice (control group, NASH model group, and L. paracasei group) were analyzed using flow cytometry. F4/80? cells were subsequently analyzed with CD206 antibody. a The cell numbers in total F4/80? fractions with CD206?/CD206- are shown according to the percentage of a fraction per total SVF cells in liver tissue. b The fraction of total F4/80? Kupffer cells was significantly higher in the NASH model group than the control group (32 vs. 16 % of total stroma-vascular fraction cells, p \ 0.05). c The percentages of

CD206- and CD206? cells out of total F4/80? cells are demonstrated in the control group, NASH model group, and the L. paracasei group. Compared with the control group, the CD206- fraction of F4/80? Kupffer cells was higher and F4/80?CD206? cells were lower in the NASH model group (e.g., F4/80?CD206-: 25 % in control group vs. 56 % in NASH model group, p \ 0.05). Meanwhile, the L. paracasei group showed an increased fraction of F4/80?CD206? cells and a decreased fraction of F4/80?CD206- cells in comparison with the NASH model group (e.g., F4/80?CD206?: 44 % in NASH model group vs. 62 % in L. paracasei group, p \ 0.05). c *p \ 0.05

Fig. 7 Effect of L. paracasei on immunohistochemical TLR-4 and NOX-4 expression in NASH (upper panels TLR-4 and lower panels NOX-4 immunohistochemistry at 9200 magnification). a, d Control group; b, e NASH model group; c, f L. paracasei group. Compared with the control group, the staining of both TLR-4 and NOX-4 was

significantly more intense in the NASH model group (p \ 0.05). However, the expression of TLR-4 and NOX-4 in the L. paracasei group was significantly lower than in the NASH model group (p \ 0.05). Arrowheads indicate the areas of TLR-4 or NOX-4 stainings

123

Dig Dis Sci

Fig. 8 Effects of L. paracasei on hepatic mRNA expression of TNFa (a), MCP-1 (b), TLR-4 (c), IL-4 (d), PPAR-c (e), and PPAR-d (f) in NASH. The amount of mRNA for each hepatic gene was measured using RT-PCR and standardized to the amount of mRNA for GAPDH in each sample. The levels of mRNA in the NASH model group and the L. paracasei group were presented as a value relative to the level of the control group. Data are presented as mean ± SEM. Con,

control group; NASH, NASH model group; Paracasei, L. paracasei group. Hepatic expression of TNF-a, MCP-1, TLR-4, IL-4, PPAR-c, and PPAR-d were significantly higher in the NASH model group compared to the control group. However, the L. paracasei group showed a significantly lower expression of TNF-a, MCP-1, TLR-4, IL-4, PPAR-c, and PPAR-d than in the NASH model group. *p \ 0.05

(p \ 0.05). However, there was no significant difference in the mRNA expression of TNF-a among three groups.

pro-inflammatory M1 Kupffer cell response and activated the anti-inflammatory M2 response, thereby inducing M2 Kupffer cell polarization in NASH. In terms of the ‘‘gut–liver axis,’’ there is a correlation between gut microbiome, bacterial translocation, and hepatic steatosis [17]. Intestinal dysbiosis is observed in patients with NAFLD or NASH [18–20]. We speculated possible mechanisms of probiotic action in NAFLD. First, probiotics could induce the inhibition of invading bacteria by modulating the composition of intestinal microbiome and the production of antimicrobial factors such as short-

Discussion The present study demonstrated that L. paracasei supplements diminished hepatic fat accumulation and serum ALT level with preventing intestinal permeability in a NASH model. Also, L. paracasei supplements reduced Kupffer cell infiltration caused by NASH. L. paracasei inhibited the

123

Dig Dis Sci

Fig. 9 Real-time PCR of MCP-1 (a), TNF-a (b), and, PPAR-c (c) in three groups. The amount of mRNA for each hepatic gene was measured using real-time PCR and standardized to the amount of mRNA for b-actin in each sample. The levels of mRNA in the NASH model group and the L. paracasei group were presented as a value relative to the level of the control group. Data are presented as

mean ± SEM. Hepatic expressions of MCP-1, and PPAR-c were significantly higher in the NASH model group compared to the control group. However, the L. paracasei group showed a significantly lower expression of MCP-1, and PPAR-c than in the NASH model group. *p \ 0.05

chain fatty acids [21]. Second, probiotics could modify intestinal epithelial permeability and reduce endotoxemia [22, 23]. Third, probiotics have immune-modulatory and anti-inflammatory effects on the intestinal tract in NAFLD [24, 25]. In this study, we used L. paracasei for probiotic therapy in NASH. The reason is as follows. We paid attention to the role of Kupffer cell as an immune-modulatory function as well as ‘‘gut–liver axis’’ in the pathogenesis of NAFLD. We would find appropriate probiotics that have good immune-modulatory properties and a favorable effect on the intestine. Previous studies showed that L. paracasei enhances the phagocytic activity of macrophage [11], and it was superior in the immunemodulatory response in a colitis model as compared with L. plantarum and L. rhamnosus [24]. L. paracasei has immune-modulatory properties to prevent intestinal inflammation [12]. Furthermore, L. paracasei decreased fat storage in high-fat fed mice and reduced oxidative stress and hepatic inflammation in methionine/cholinedeficient diet mice [13, 14]. Based on these considerations, L. paracasei was selected for our study. In NAFLD or NASH, there were significant changes in the composition in gut microflora, including dysbiosis. Increased gut permeability and dysbiosis-related endotoxemia were demonstrated in obesity as well as in NAFLD [3, 26]. It seems that probiotics may prevent intestinal permeability through increased mucin release and improved tight junction function [27]. The present study showed that L. paracasei prevents intestinal permeability using the urine 51Cr-EDTA test in a NASH model. In NAFLD or NASH, dysbiosis-induced increased production of endotoxins and increased intestinal permeability, which caused an inflow of endotoxins to the liver via the portal vein [28]. Kupffer cells, the tissue macrophages of the

liver, particularly recognized the endotoxin. Absorbed endotoxins and other bacterial products are recognized by TLR in the liver, and TLR has a crucial role in the control of inflammation. TLR-4 is the main receptor detecting gutderived endotoxins and regulating hepatic inflammation in NASH. TLR-4 is abundant in Kupffer cells in charge of phagocytic function [28]. In this study, TLR-4 was highly expressed in NASH while L. paracasei suppressed its expression compared with NASH. Kupffer cells play a pivotal role in the pathogenesis of NAFLD or NASH [29]. In the present study, total Kupffer cells counts were high in the NASH model group compared with the control group, while they were low in the L. paracasei treatment group compared with the NASH model group. In addition to total Kupffer cell count, we paid attention to the phenotype (M1 vs. M2 Kupffer cells) in NASH. Macrophages are classified into pro-inflammatory M1 macrophages and anti-inflammatory M2 macrophages. The stimulating cytokines of each macrophage are very different in obesity (i.e., adipose tissue macrophages). TNF-a, interferon-gamma, and endotoxin induce M1 macrophages. In contrast, M2 macrophages are induced by IL-4 and IL-13 in adipose tissue macrophages [30]. The composition of M1 and M2 macrophages varies according to disease status. While M1 macrophages are prominently induced in tissue inflammation, early sepsis, cancer onset, and obesity, M2 macrophages are dominantly induced in tissue resolution, late sepsis, established cancer, and lean status [8]. However, an analysis of Kupffer cell polarization (the phenotype switch of M1 and M2 Kupffer cells) was lacking in liver diseases, including NAFLD or NASH. Louvet et al. [31] showed that cannabinoid CB2 induced M2 Kupffer cell polarization with M1 cell inhibition in alcoholic liver disease. Leroux et al. [32] reported that proinflammatory Kupffer cells with a fat-laden phenotype

123

Dig Dis Sci

were activated in early stages of steatohepatitis. Wan et al. [33] showed that M2 Kupffer cells promoted M1 Kupffer cells apoptosis in alcoholic and nonalcoholic fatty liver disease. The present study showed the beneficial effect of L. paracasei on NAFLD or NASH. However, the role of Lactobacillus species in NAFLD or NASH seems to be not clear because previous studies revealed that the Lactobacillus family is highly diverse with some species providing protection while others harm host metabolism. Some studies showed that a decrease in Lactobacillus species was associated with the protection against obesity or NAFLD [19, 34–36], whereas several studies showed that the supplement of the Lactobacillus species is beneficial in the reduction of hepatic steatosis, anti-inflammation, and improvement of insulin resistance in experimental models of NAFLD [37–39]. Further studies are needed to clarify the role of lactobacillus species in NAFLD or NASH. In this study, we did not show a significant improvement of hepatic inflammation in liver histology in the probiotics group compared with the NASH model group. Also, there was no significant difference in weight gain between the NASH model group and the probiotics group although caloric intake was similar between the two groups. In other words, the probiotics group had no significant improvement of hepatic inflammation in liver histology although a variety of inflammatory cytokines or biochemical marker was improved with probiotics supplement. We speculated about the reasons for this finding. First, the development of intrahepatic inflammation was not prominent in the NASH model group of this study. Therefore, there was no statistical difference in histologic inflammation between the NASH model group and the probiotics group. Second, the duration of probiotics supplement was a little short (10 weeks) in this study. Thus, there was a possibility that a significant improvement of hepatic inflammation did not still develop in histology although molecular biologic tests showed the improvement of indicators related to anti-inflammation in the probiotics group. This study has some limitations. First, our study did not include whether the mice might lose weight after treating with normal chow and L. paracasei. We did not conduct an experiment for treating mice with normal chow and L. paracasei because preceding research showed that the body weight of mice did not change after treating with normal chow and L. paracasei as compared with the baseline level. More precise results for the effect of L. paracasei could be gained if we conducted an experiment for mice treated with normal chow and L. paracasei. Second, to assess the function of the gut epithelial barrier in NASH or probiotics supplement, intestinal permeability only was performed. We did not assess the effect of probiotics on the tight

123

junction of intestinal epithelium (e.g., tight junction protein such as ZO-1, occludin, and claudin 1). Third, changes in intestinal microflora were not assessed in this study. If fecal microflora analysis or intestinal tissue culture were performed, more information could be elucidated on the effect of L. paracasei on gut microbiota because probiotic supplements essentially altered the composition of intestinal microflora by supplying beneficial organisms to the gut. Despite the above limitations, we believe that this study suggests the effects and action mechanism of probiotics on NASH, especially focusing on the role and phenotype of Kupffer cells. In conclusion, L. paracasei prevented intestinal permeability and attenuated hepatic steatosis with an M2 prominent Kupffer cell polarization pattern in a NASH model. Acknowledgments This study was supported by a Grant from the Korea Healthcare Technology R&D Project, Ministry of Health & Welfare, Republic of Korea (A121185). Conflict of interest The author(s) declare that they have no competing interests.

References 1. Kopec KL, Burns D. Nonalcoholic fatty liver disease: a review of the spectrum of disease, diagnosis, and therapy. Nutr Clin Pract. 2011;26:565–576. 2. Tiniakos DG, Vos MB, Brunt EM. Nonalcoholic fatty liver disease: pathology and pathogenesis. Annu Rev Pathol. 2010;5: 145–171. 3. Abu-Shanab A, Quigley EM. The role of the gut microbiota in nonalcoholic fatty liver disease. Nat Rev Gastroenterol Hepatol. 2010;7:691–701. 4. Compare D, Coccoli P, Rocco A, et al. Gut–liver axis: the impact of gut microbiota on non alcoholic fatty liver disease. Nutr Metab Cardiovasc Dis. 2012;22:471–476. 5. Bilzer M, Roggel F, Gerbes AL. Role of Kupffer cells in host defense and liver disease. Liver Int. 2006;26:1175–1186. 6. Bourlier V, Bouloumie A. Role of macrophage tissue infiltration in obesity and insulin resistance. Diabetes Metab. 2009;35: 251–260. 7. Murray PJ, Wynn TA. Protective and pathogenic functions of macrophage subsets. Nat Rev Immunol. 2011;11:723–737. 8. Biswas SK, Chittezhath M, Shalova IN, Lim JY. Macrophage polarization and plasticity in health and disease. Immunol Res. 2012;53:11–24. 9. Chinetti-Gbaguidi G, Staels B. Macrophage polarization in metabolic disorders: functions and regulation. Curr Opin Lipidol. 2011;22:365–372. 10. Lirussi F, Mastropasqua E, Orando S, Orlando R. Probiotics for non-alcoholic fatty liver disease and/or steatohepatitis. Cochrane Database Syst Rev. 2007;24:1–15. 11. Kapila R, Sebastian R, Varma DV, et al. Comparison of innate immune activation after prolonged feeding of milk fermented with three species of Lactobacilli. Microbiol Immunol. 2013;57: 778–784. 12. Oliveira M, Bosco N, Perruisseau G, et al. Lactobacillus paracasei reduces intestinal inflammation in adoptive transfer mouse

Dig Dis Sci

13.

14.

15.

16.

17.

18.

19.

20.

21. 22.

23.

24.

25.

model of experimental colitis. Clin Dev Immunol. 2011;2011: 1–13. Aronsson L, Huang Y, Parini P, et al. Decreased fat storage by Lactobacillus paracasei is associated with increased levels of angiopoietin-like 4 protein (ANGPTL4). PLoS One. 2010;5: e13087. Nardone G, Compare D, Liguori E, et al. Protective effects of Lactobacillus paracasei F19 in a rat model of oxidative and metabolic hepatic injury. Am J Physiol Gastrointest Liver Physiol. 2010;299:G669–G676. Yu E, Korean Study Group for the Pathology of Digestive D. Histologic grading and staging of chronic hepatitis: on the basis of standardized guideline proposed by the Korean Study Group for the Pathology of Digestive Diseases. Taehan Kan Hakhoe Chi. 2003;9:42–46. McCafferty DM, Miampamba M, Sihota E, Sharkey KA, Kubes P. Role of inducible nitric oxide synthase in trinitrobenzene sulphonic acid induced colitis in mice. Gut. 1999;45:864–873. Miele L, Marrone G, Lauritano C, et al. Gut-liver axis and microbiota in NAFLD: insight pathophysiology for novel therapeutic target. Curr Pharm Des. 2013;19:5314–5324. Mouzaki M, Comelli EM, Arendt BM, et al. Intestinal microbiota in patients with nonalcoholic fatty liver disease. Hepatology. 2013;58:120–127. Raman M, Ahmed I, Gillevet PM, et al. Fecal microbiome and volatile organic compound metabolome in obese humans with nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol. 2013;11:868–875 e861–863. Zhu L, Baker SS, Gill C, et al. Characterization of gut microbiomes in nonalcoholic steatohepatitis (NASH) patients: a connection between endogenous alcohol and NASH. Hepatology. 2013;57:601–609. Carr FJ, Chill D, Maida N. The lactic acid bacteria: a literature survey. Crit Rev Microbiol. 2002;28:281–370. Ramakrishna BS. Probiotic-induced changes in the intestinal epithelium: implications in gastrointestinal disease. Trop Gastroenterol. 2009;30:76–85. Rivera CA, Adegboyega P, van Rooijen N, Tagalicud A, Allman M, Wallace M. Toll-like receptor-4 signaling and Kupffer cells play pivotal roles in the pathogenesis of non-alcoholic steatohepatitis. J Hepatol. 2007;47:571–579. Mileti E, Matteoli G, Iliev ID, Rescigno M. Comparison of the immunomodulatory properties of three probiotic strains of Lactobacilli using complex culture systems: prediction for in vivo efficacy. PLoS ONE. 2009;4:e7056. Petrof EO, Kojima K, Ropeleski MJ, et al. Probiotics inhibit nuclear factor-kappaB and induce heat shock proteins in colonic

26.

27.

28. 29. 30. 31.

32.

33.

34.

35.

36.

37.

38.

39.

epithelial cells through proteasome inhibition. Gastroenterology. 2004;127:1474–1487. Lam YY, Ha CW, Campbell CR, et al. Increased gut permeability and microbiota change associate with mesenteric fat inflammation and metabolic dysfunction in diet-induced obese mice. PLoS ONE. 2012;7:e34233. Iacono A, Raso GM, Canani RB, Calignano A, Meli R. Probiotics as an emerging therapeutic strategy to treat NAFLD: focus on molecular and biochemical mechanisms. J Nutr Biochem. 2011;22:699–711. Machado MV, Cortez-Pinto H. Gut microbiota and nonalcoholic fatty liver disease. Ann Hepatol. 2012;11:440–449. Baffy G. Kupffer cells in non-alcoholic fatty liver disease: the emerging view. J Hepatol. 2009;51:212–223. Cassetta L, Cassol E, Poli G. Macrophage polarization in health and disease. Sci World J. 2011;11:2391–2402. Louvet A, Teixeira-Clerc F, Chobert MN, et al. Cannabinoid CB2 receptors protect against alcoholic liver disease by regulating Kupffer cell polarization in mice. Hepatology. 2011;54:1217–1226. Leroux A, Ferrere G, Godie V, et al. Toxic lipids stored by Kupffer cells correlates with their pro-inflammatory phenotype at an early stage of steatohepatitis. J Hepatol. 2012;57:141–149. Wan J, Benkdane M, Teixeira-Clerc F, et al. M2 Kupffer cells promote M1 Kupffer cell apoptosis: a protective mechanism against alcoholic and nonalcoholic fatty liver disease. Hepatology. 2014;59:130–142. Joyce SA, MacSharry J, Casey PG, et al. Regulation of host weight gain and lipid metabolism by bacterial bile acid modification in the gut. Proc Natl Acad Sci USA. 2014;111:7421–7426. Li F, Jiang C, Krausz KW, et al. Microbiome remodelling leads to inhibition of intestinal farnesoid X receptor signalling and decreased obesity. Nat Commun. 2013;4:2384. Million M, Maraninchi M, Henry M, et al. Obesity-associated gut microbiota is enriched in Lactobacillus reuteri and depleted in Bifidobacterium animalis and Methanobrevibacter smithii. Int J Obes (Lond). 2012;36:817–825. Lee HY, Park JH, Seok SH, et al. Human originated bacteria, Lactobacillus rhamnosus PL60, produce conjugated linoleic acid and show anti-obesity effects in diet-induced obese mice. Biochim Biophys Acta. 2006;1761:736–744. Wang Y, Xu N, Xi A, Ahmed Z, Zhang B, Bai X. Effects of Lactobacillus plantarum MA2 isolated from Tibet kefir on lipid metabolism and intestinal microflora of rats fed on high-cholesterol diet. Appl Microbiol Biotechnol. 2009;84:341–347. Yadav H, Jain S, Sinha PR. Antidiabetic effect of probiotic dahi containing Lactobacillus acidophilus and Lactobacillus casei in high fructose fed rats. Nutrition. 2007;23:62–68.

123