Simultaneous Amelioratation of Colitis and Liver Injury in Mice by Bifidobacterium longum LC67 and Lactobacillus plantarum LC27

Disturbances in the gut microbiota composition are associated with chronic inflammatory diseases of the intestine and the liver. In a preliminary study, Lactobacillus plantarum LC27 and Bifidobacterium longum LC67 could inhibit Escherichia coli growth and lipopolysaccharide-induced NF-κB activation linked to gut inflammation. Here, we investigated their effects on 2,4,6-trinitrobenzesulfonic acid (TNBS)-induced colitis and liver damage in mice. First, oral administration of LC27 or LC67 (1 × 109 CFU/mouse) inhibited TNBS-induced colon shortening [F(5,30) = 100.66, P < 0.05] and myeloperoxidase activity [F(5,30) = 56.48, P < 0.05]. These probiotics restored TNBS-induced disturbance of gut microbiota, leading to the suppression of Proteobacteria to Bacteroidetes ratio and fecal and blood lipopolysaccharide levels. Second, LC27 and LC67 inhibited TNBS-induced NF-κB activation, reversed TNBS-suppressed tight junction protein expression, and restored Th17/Treg balance. Also, treatment with LC27 or LC67 significantly decreased TNBS-induced alanine transaminase [ALT, F(5,30) = 3.50, P < 0.05] and aspartate transaminase [AST, F(5,30) = 12.81, P < 0.05] levels in the blood, as well as t-butylhydroperoxide-induced ALT and AST levels. Finally, the mixture of LC27 and LC67 (0.5 × 109 CFU/mouse, respectively) synergistically attenuated TNBS- or t-butylhydroperoxide-induced colitis and liver damage. The capability of LC27 and LC67 to reverse TNBS-mediated microbiota shift and damage signals suggests that these probiotics may synergistically attenuate colitis and liver injury by alleviating gut microbiota imbalance.

macrophage activation and Th17 cell and Treg differentiation via the gut microbiota-liver axis could simultaneously inhibit colitis and prevent liver injury.
Numerous studies have shown that functional foods including probiotics are beneficial for reducing the risks of metabolic and degenerative diseases and promoting good health 18,19 . Of these, lactobacilli and bifidobacteria have been reported to be beneficial microbes, as they support the maintenance of gut microbiota homeostasis in humans and animals 20,21 . These probiotics restore balance to gut microbiota composition 22 , induce host immune systems 23 , and have anti-obesity 24 , anti-hepatoprotective 25 , and anti-colitic effects 26 . Bifidobacterium longum alleviates dextran sulfate sodium (DSS)-induced colitis by suppressing IL-17A responses 27 . Bifidobacterium infantis inhibits colitis in mice by inducing Treg differentiation 28 . Lactobacillus plantarum C29 ameliorates age-dependent colitis in aged mice via inhibition of the NF-κB signaling pathways 29 . Lactobacillus rhamnosus GG also attenuates ethanol-induced liver injury in mice by restoring the gastrointestinal barrier via regulation of tight junction proteins and miR122a expression 30 . Lactobacillus casei MYL01 attenuates ethanol-induced liver damage in vitro by regulating the expression of proinflammatory cytokines such as tumor necrosis factor (TNF)-α and anti-inflammatory cytokine such as IL-10 31 . However, the effects of probiotics against both colitis and liver injury have not been thoroughly investigated.
Therefore, to understand the simultaneous effects of probiotics against colitis and liver injury, we screened for probiotics that can potently suppress bacterial growth and LPS production in Escherichia coli, and inhibited NF-κB activation in LPS-stimulated macrophages in vitro. Here, two probiotics, Bifidobacterium longum LC67 isolated from human fecal microbiota, and Lactobacillus plantarum LC27 isolated from kimchi, were selected to investigate their effects against 2,4,6-trinitrobenzenesulfonic acid (TNBS)-induced colitis and liver injury in mice.

Effects of LC27, LC67, and PM on TNBS-induced gut microbiota disturbance in mice.
Previous studies have shown that TNBS treatment increases the ratio of Firmicutes/Bacteroidetes, and elevates LPS production in the gut microbiota 32 . Moreover, the overexpression of LPS in the gut microbiota leads to gastrointestinal inflammation 24 and excessive exposure to LPS causes inflammation in the liver 5,7 . Therefore, we investigated the effects of LC27, LC67, and PM on the gut microbiota composition and LPS production in mice with TNBS-induced colitis and liver injury (Fig. 8). TNBS treatment increased the ratio of Firmicutes/Bacteroidetes, like previously reported 32 . Furthermore, TNBS treatment increased the number of Proteobacteria but reduced the number of Bacteroidetes, resulting in a reduced Proteobacteria/Bacteroidetes ratio [F(5,30) = 5.22, P < 0.05] (Fig. 8B). Treatment with LC27, LC67, or PM significantly inhibited the TNBS-induced Proteobacteria level, and increased the TNBS-suppressed Bacteroidetes level (Fig. 8A). Moreover, TNBS treatment significantly increased the LPS level [F(5,30) = 3.85, P < 0.05] in the colonic fluid and blood of mice, whereas LC27, LC67, or their mixture significantly decreased the TNBS-induced LPS production (Fig. 8C). TNBS treatment also increased Enterobacteriaceae levels including Escherichia coli. However, these probiotics reversed the suppression of lactobacilli [F(5,30) = 31.36, P < 0.05] and bifidobacteria [F(5,30) = 4.08, P < 0.05] by TNBS, and reduced the amount of Enterobacteriaceae and Escherichia coli [F(5,30) = 18.34, P < 0.05], which belong to phylum Proteobacteria (Fig. 8B,C). Furthermore, these treatments increased the populations of Lactobacillus plantarum and Bifidobacterium longum. Collectively, these data suggest that LC27 and LC67 can alleviate TNBS-induced gut microbiota imbalance.

Discussion
Acute and chronic inflammations are the body's response to injuries and infections 33,34 . Acute inflammation is a normal and beneficial response to injury, whereas chronic inflammation is persistent and excessive. Inflammatory reactions in the gastrointestinal tract can be activated by a variety of stresses such as excessive ROS, alcohol, and LPS owing to disturbance of gut microbiota 34 . Exposure to alcohol or high fat diet causes dysbiosis via alteration in the gut microbiota, including an increase in Proteobacteria and a decrease in Bacteroidetes 5 . An increase in the Gram-negative Proteobacteria reduces the expression of cellular tight junctions and increases gut permeability through overexpression of LPS, resulting in increased absorption of LPS into the blood. LPS, a major driver of systemic inflammation 35 , increases blood TNF-α levels via Toll-like receptor 4-associated NF-κB signaling pathway to cause inflammation, even though blood TNF-α level is barely detectable in mice in absence of any stimuli or treatment 36 . Therefore, chronic inflammatory responses lead to progressive damage to the body, resulting in a variety of chronic inflammatory diseases, such as colitis, hepatitis, and rheumatoid arthritis 37 . To regulate these inflammatory diseases, probiotics may be used to control TNF-α expression via regulation of NF-κB in immune cells, as well as to inhibit LPS production induced by the gut microbiota imbalance.
In the present study, we found that LC27 and LC67 inhibited TNBS-induced colitis via inhibition of NF-κB in macrophages and epithelial cells, similar to previous reports 38 . Furthermore, treatment with these probiotics restored TNBS-disturbed gut microbiota composition: they suppressed the population of Enterobacteriaceae, particularly Escherichia coli, which is belonging to Proteobacteria, and gut microbiota LPS levels and increased the populations of lactobacilli and bifidobacteria, including Lactobacillus plantarum and Bifidobacterium longum. Additionally, it has been shown that Lactobacillus brevis G-101 inhibits TNF-α and IL-1β expression in macrophages, leading to attenuation of colitis 39 . In addition, Bifidobacterium longum CH57 was found to attenuate colitis by inhibiting NF-κB signaling pathways and TNF-α expression. Lactobacillus plantarum C29 also ameliorates colitis in aged mice by inhibiting NF-κB signaling 29 . Another study has demonstrated that Lactobacillus casei DN-114001 inhibits DSS-induced colitis by inhibiting gut membrane permeability and NF-κB activation 40 . Furthermore, some probiotics were also shown to restore composition of gut microbiota and fecal LPS level in mice with colitis 29,40 . These results suggest that probiotics can inhibit NF-κB activation and restore disrupted gut microbiota composition to attenuate colitis. Furthermore, treatment with these probiotics significantly suppressed blood LPS and TNF-α levels in mice with TNBS-induced colitis. These treatments also reduced liver MDA and myeloperoxidase activity, and blood AST and ALT levels, resulting in attenuation of TNBS-induced liver injury in mice. In spite of short-term treatment with these probiotics, their effects were comparable to those of sulfasalazine, a positive agent. Another study showed that L. rhamnosus CCFM1107 reduces oxidative stress and restores intestinal flora in ethanol-treated mice, which ameliorates liver injury 41 . In addition, L. rhamnosus GG reestablishes the gastrointestinal barrier via the suppression of tight junction proteins and miR122a in mice, leading to the alleviation of ethanol-induced liver injury 30 . Lastly, L. acidophilus CSG exhibited hepatoprotective  26 . These results suggest that probiotics such as LC27 and LC67 are effective against colitis, and could attenuate oxidative stress-induced liver injury and colitis by inhibiting NF-κB activation, scavenging ROS, and alleviating gut microbiota imbalance.
Activated macrophages secrete IL-23, which induces expression of proinflammatory cytokines such as TNF-α and IL-6, and also promotes the differentiation and activation of Th17 cells 10,33,34 . Th17 cells suppress Treg the differentiation, resulting in the onset of chronic inflammatory diseases such as colitis 13 . Conversely, activated Tregs inhibit Th17 cell the differentiation via secretion of IL-10 and TGF-β, which results in attenuation of chronic colitis. Therefore, the Th17/Treg cell balance is important for the development of colitis.
In the present study, we found that LC27 and LC67 inhibited Th17 cell differentiation and RORγt expression, and enhanced Treg differentiation and Foxp3 expression, leading to attenuation of colitis and liver injury. Furthermore, an increase in IL-10 expression was also observed both in vitro and in vivo. It has been previously shown that Lactobacillus brevis CH23 restores Th17/Treg balance via regulation of the transcription factors Foxp3 and RORγt, as well as the cytokines IL-17 and IL-10, resulting in recovery from colitis 28 . Similarly, Lactobacillus TNBS, except in the normal control group, was intrarectally administered to mice and test agents [saline, LC5, LC27, LC67, LC68 (2 × 10 9 CFU/mouse), or sulfasalazine (SUL; 50 mg/kg)] were orally administered for 3 days. Cytokines were determined by ELISA. iNOS, COX-2, and NF-κB signaling molecules were determined by immunoblotting. All values are shown as the mean ± SD (n = 6). # p < 0.05 vs. normal control group. *p < 0.05 vs. group treated with TNBS alone.  suggest that LC23 and LC67 may inhibit colitis and liver injury by correcting the imbalance of Th17/Treg cells involved in adaptive immunity, via regulation of innate immune cells and gut microbiota composition, and by increasing the expression of colonic tight junction proteins.
Although commercial probiotic products contain a combination of various probiotics such as Lactobacilli, Bifidobacteria, and Streptococci the combined effects of these probiotic mixtures have not been thoroughly investigated 20 . For example, Bifidobacterium longum CH57 and Lactobacillus brevis CH23 synergistically inhibit colitis by inhibiting macrophage activation and restoring Th17/Treg balance 26 . In the present study, the LC27 and LC67 mixture PM, synergistically rather than additively, attenuated TNBS-induced colitis such as colon shortening, myeloperoxidase activity, TNF-α and IL-10 expression, and Th17 and Treg cell differentiation. However, TNBS-induced liver damage was significantly attenuated by treatment with PM. This involved correcting the Th17/Treg imbalance via regulation of innate immune cells and gut microbiota composition and increasing the expression of colonic tight junction proteins. Furthermore, these probiotics also attenuated t-BHP-induced liver injury. These results suggest that synergistic probiotic products containing a combination of bacterial species, such as PM, may be more effective in protecting diseased caused by imbalance of the gut microbiota.
In conclusion, Lactobacillus plantarum LC27, isolated from kimchi, suppressed gut bacterial LPS production and NF-κB activation in macrophages. Bifidobacterium longum LC67, isolated from the human gut microbiota, inhibited gut bacterial LPS production and differentiation of splenic T cells into Th17 cells. It also increased Treg cell differentiation via up-regulation of IL-10 and Foxp3 expression. These probiotics synergistically attenuated TNBS-induced colitis and liver injury as well as t-BHP-induced liver injury by correcting the gut microbiota composition and inhibiting inflammatory responses involved in innate and adaptive immunity. Collectively, our study supports that LC27 and LC67 could be effective tools to control colitis and liver damage induced by altered gut microbiota landscape.   coli, assessed by the culture of selective media. Test agents and saline were orally administered for 3 days after TNBS treatment. TNBS, except in the normal control group, was intrarectally administered to mice and test agents [saline, LC27, LC67, PM (1 × 10 9 CFU/mouse), or sulfasalazine (SS; 50 mg/kg)] were orally administered for 3 days. All values are shown as mean ± SD (n = 6). # p < 0.05 vs. normal control group. *p < 0.05 vs. group treated with TNBS alone.
Animals. Male C57BL/6 (21-23 g, 6-weeks old) were supplied from RaonBio Inc. (Seoul, Korea). All animals were housed in wire cages at 20-22 °C and 50 ± 10% humidity, fed standard laboratory chow and water ad libitum. After the acclimation for 7 days, mice were used in experiments.
All animal experiments were approved by the Committee for the Care and Use of Laboratory Animals in the Kyung Hee University and performed in accordance with the Kyung Hee University Guidelines for Laboratory Animals Care and Usage (IRB No., KHUASP(SE)-16-049).
Preparation of macrophages. Macrophages were prepared according to the method of Jeong et al. 29 . Mice were intraperitoneally injected with 4% (w/v) thioglycolate solution (2 mL) and killed 4 days after the injection 29 . Cells were removed with RPMI 1640 in the peritoneal cavity, centrifuged (300 × g, 10 min), and washed with RPMI 1640 twice. Collected cells (1.5 × 10 6 cells/well) were incubated in RAF at 37 °C for 20 h and washed three times. The attached cells were used as macrophages. To evaluate the anti-inflammatory effect of probiotics, macrophages (1 × 10 6 cells/well) were treated with LPS (100 ng/mL) in the absence or presence of each probiotic (1 × 10 3 or 1 × 10 5 CFU/mL) for 90 min (for p65 and p-p65) or 24 h (for TNF-α and IL-1β).
Preparation of mice with experimental colitis and liver injury. First, mice were randomly divided into 7 groups: normal control, TNBS-induced colitic control groups treated with vehicle, four probiotics (1 × 10 9 CFU), or sulfasalazine (50 mg/kg). Each group consisted of six mice.
Second, mice were randomly divided into 6 groups: normal control, TNBS-induced colitic control groups treated with vehicle, LC27 (1 × 10 9 CFU/mouse), LC67 (2 × 10 9 CFU/mouse), and their mixture (1:1, each 1 × 10 9 CFU/mouse), or sulfasalazine (50 mg/kg). Each group consisted of six mice. Colitis was induced by the intrarectal injection of 2.5% (w/v) TNBS solution (100 μL, dissolved in 50% ethanol) into the colon of mice anesthetized with ether 43 . Normal control group was treated with vehicle alone instead of TNBS. To entirely distribute TNBS within the colon, mice were held in a vertical position for 30 s after the TNBS injection. Test agents (probiotics or sulfasalazine dissolved in 1% glucose) were orally administered once a day for 3 days after TNBS treatment. Mice were killed 18 h after the final administration of test agents. Normal control group was treated with vehicle alone instead of test agents. Whole-blood samples were immediately withdrawn from carotid artery. Sera were prepared by centrifugation (10 min, 250 × g) and ALT, AST, and TNF-α levels were then determined according to the method of Lee et al. 44 . The colon was removed and opened longitudinally. The colitis grade was macroscopically scored (0, no ulcer and no inflammation; 1, no ulceration and local hyperemia; 2, ulceration with hyperemia; 3, ulceration and inflammation at one site only; 4, two or more sites of ulceration and inflammation; 5, ulceration extending more than 2 cm). The colons were gently washed by ice-cold phosphate buffered saline (PBS) and were stored at -80 °C until used in the experiment for, myeloperoxidase activity assay, ELISA, and immunoblotting.

Preparation of mice with t-BHP-induced hepatic injury. Mice with t-BHP-induced hepatic injury
were prepared according to the method of Lee et al. 44 . Firs, mice were randomly divided into 6 groups: normal control, t-BHP-treated control groups treated with vehicle, probiotics (LC27, LC67, or their mixture (1:1): 1 × 10 9 CFU/mouse), or silymarin (50 mg/kg). Each group consisted of six mice. Mice were intraperitoneally treated with 1.5 mmol t-BHP/kg. Test agents were orally administered once a day for 3 days from 24 h after treatment with t-BHP. Control group was given with saline instead of the sample compounds. Blood samples were collected 18 h after the final administration of test agents by cardiac puncture under ether anesthesia. Sera were obtained by centrifugation (1000 × g, 15 min) and ALT, AST, and TNF-α levels were then determined according to the method of Lee et al. 44 .

Histological examination.
Colons or livers were fixed in 50 mM phosphate buffer (pH 7.4) containing 4% paraformaldehyde overnight, frozen in optimal cutting temperature solution, cut into 15 μm section using a cryostat, stained with hematoxylin-eosin, and then observed under a light microscopy 43 . Assay of myeloperoxidase (MPO) activity. Mouse colon or liver tissues were homogenized in 10 mM potassium phosphate buffer (pH 7.0) containing 0.5% hexadecyl trimethyl ammonium bromide, and centrifuged for 10 min at 20,000 × g at 4 °C 44 . The resulting supernatants (50 μL) were added to the reaction mixture containing 0.1 mM H 2 O 2 and 1.6 mM tetramethyl benzidine preincubated at 37 °C for 2 min, and sequentially monitored the absorbance (650 nm) at 37 °C for 5 min. Myeloperoxidase activity was calculated as the quantity of enzyme degrading 1 μmol/mL of peroxide, and expressed in unit/mg protein protein 43 . The amount of protein was determined by the method of Bradford.
Quantitative real time -polymerase chain reaction (qPCR). Reverse transcription was performed with 2 μg of total RNA isolated from the colon. Real time PCR for IL-10, IL-17, Foxp3, RAR-related orphan receptor gamma t (RORγt), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was performed as described previously 43 , utilizing Qiagen thermal cycler, which used SYBER premix agents. Thermal cycling conditions were as follows: activation of DNA polymerase at 95 °C for 5 min, followed by 36 cycles of denaturation and amplification at 95 °C for 5 s and 63 °C for 30 s, respectively. The normalized expression of the assayed genes, with respect to GAPDH, was computed for all samples by using the Microsoft Excel data spreadsheet. Primers  TAA CCC TTA AAG TCC TGC-3′; IL-17 forward, 5′-TTT AAC TCC CTT GGC GCA AAA-3′ reverse, 5′-CTT  TCC CTC CGC ATT GAC AC-3′; RORγt forward, 5′-ACAGCCACTGCATTCCCA GTTT-3′, reverse, 5′-TCTCGGAAGGACTTGCAGACAT-3′; Foxp3 forsward, 5′-CCC ATC CCC AGG AGT CTT-3′, reverse, 5′-ACC  ATG ACT AGG GGC ACT GTA-3′; and GAPDH forward, 5′-TGC AGT GGC AAA GTG GAG AT-3′, reverse,  5′-TTT GCC GTG AGT GGA GTC AT-3′. Real time PCR for Firmicutes, Proteobacteria, Actinobacteria, and Bacteroidetes was performed with 100 ng of total DNA isolated from the colon fluid, utilizing Takara thermal cycler, which used SYBER premix agents 45 . Thermal cycling conditions were as follows: activation of DNA polymerase for 30 s at 95 °C, followed by 35 cycles of denaturation and amplification at 95 °C for 5 s and 63 °C for 30 s, respectively. The normalized expression of the assayed genes, with respect to bacterial rRNA, was computed for all samples using the Microsoft Excel data spreadsheet. Primers were used as follows 45,46  Determination of LPS. The content of LPS was determined using a LAL assay kit according to manufacturer's protocol 29 . For the assay of culture medial LPS contents, each probiotic (1 × 10 6 CFU/mL) and E. coli (1 × 10 6 CFU/mL) was simultaneously inoculated in GAM (5 mL) and anaerobically cultured for 37 °C for 24 h. The cultured suspension was sonicated for 1 h on ice, centrifuged at 5,000 × g for 10 min, filtrated through a 0.45 μm filter followed by re-filtration through a 0.22 μm filter, and inactivated at 70 °C for 10 min.
For the assay of fecal LPS contents, colon content from mice (100 mg) were placed in 50 mL of PBS in a pyrogen-free tube and sonicated for 1 h on ice. After centrifugation at 400 × g for 10 min, the supernatant was collected, sterilized by filtration through a 0.45 μm filter followed by re-filtration through a 0.22 μm filter, and inactivated at 70 °C for 10 min.
For the assay of blood LPS contents, serum (5 μL) was diluted 1:10 in pyrogen-free water, inactivated for 10 min at 70 °C, and incubated with LAL solution for 30 min at 37 °C.
Each filterate or serum (50 μl) was incubated with LAL solution at 37 °C for 30 min, added additional reagents to formation of a magenta derivative, and measured the absorbance at 545 nm.

Flow cytometry of Th17 and Treg cells in the lamina propria of colons.
For the assay of Th17 cells and Tregs, colons were cut into small pieces, incubated with 2.5 mM EDTA at 37 °C with agitation for 20 min, minced, and digested for 20 min with RPMI containing 1 mg/mL collagenase type VIII at 37 °C. Lamina propria cells were then prepared 44 . T cells were isolated using a Pan T cell Isolation Kit II, fixed and stained with anti-FoxP3 or anti-IL-17A antibodies, and then analyzed by flow cytometry (C6 Flow Cytometer ® System, San Jose, CA, USA). ELISA and immunoblotting. Colon, liver tissues, and cultured cells were homogenized in the RIPA lysis buffer (1 mL) containing 1% phosphatase inhibitor cocktail and 1% protease inhibitor cocktail at 4 °C and centrifuged at 15,000 × g for 15 min.
For the determination of cytokines, the supernatants of tissue homogenates and cultured cells were transferred to a 96-well microplate. IL-1β, IL-10, IL-17, and TNF-α expression levels were determined using ELISA kits 44 .
For the immunoblotting, the supernatants of tissue and cultured cell homogenates were subjected to electrophoresis on sodium dodecyl sulfate-polyacrylamide gel, transferred to nitrocellulose membrane, blocked with non-fat dried-milk proteins, probed with antibodies, and washed with PBS with tween 20 43 . Proteins were detected with horseradish peroxidase-conjugated secondary antibodies. Protein bands were visualized with an enhanced chemiluminescence detection kit. Statistical analysis. All data are indicated as the mean ± standard deviation (SD), with statistical significance analyzed using one-way ANOVA followed by Duncan's multiple range test (P < 0.05).