Regulatory Effect of Lactobacillus brevis Bmb6 on Gut Barrier Functions in Experimental Colitis

The integrity of gut barrier functions is closely associated with the pathogenesis of colitis. It is speculated that Lactobacillus brevis Bmb6 alleviates colitis by improving the tight junction (TJ) of the inflamed intestinal epithelial layer. In the present study, the regulatory effects of L. brevis Bmb6 on the TJ barrier to ameliorate colitis-symptoms were investigated. Preliminary screening showed that L. brevis Bmb6 exhibited strong acid and bile acid tolerance, along with antioxidants and β-galactosidase activities. In a 14-day dextran sulfate sodium (DSS)-induced colitis mouse model, treatment with L. brevis Bmb6 significantly decreased in the disease activity index score. In addition, histological analyses showed that treatment with L. brevis Bmb6 protected the structural integrity of the intestinal epithelial layer and mucin-secreting goblet cells from DSS-induced damage, with only slight infiltration of immune cells. Interestingly, western blotting analyses showed that the expression of the TJ protein, zona occluden-1, was restored in Bmb6-treated mice, but not in DSS-induced mice. Consistently, the gene expression of inflammatory cytokines (tumor necrosis factor-α and interferon-γ) was also suppressed in the Bmb6-treated mice. Hence, our findings suggest that suppression of inflammatory conditions enhanced expression of TJ protein, ZO-1, or vice versa, contributing to a colitis-ameliorating effect in L. brevis Bmb6.


Introduction
Inflammatory bowel disease (IBD) is a gastrointestinal tract disorder characterized by chronic inflammation of the mucosal cells. The incidence of IBD has been increasing rapidly since 1990, affecting one in 200-300 people in high-income countries. These epidemiological data suggest that a further increase in IBD incidence may pose a major health burden to the community [1]. Although the etiology of IBD is yet to be clarified, evidence suggests genetic elements, dietary patterns, and alteration of the gut microbiome as factors leading to immune-dysregulation and intestinal barrier dysfunction in IBD [2][3][4]. Interestingly, impaired tight junction (TJ) barrier function was observed in patients with IBD and an animal model, which significantly increased the risk of bacterial translocation and infusion of other harmful substances into the bloodstream, leading to bacteremia and organ failure [5][6][7]. Hence, maintaining healthy and functional TJ barrier integrity is of utmost importance to hinder the development of gastrointestinal and systemic disorders.
As vital components of the epithelial barrier, TJ proteins such as claudins, occludin, and zonula occludens (ZO) connect the adjacent epithelial cells and provide mechanical stability to facilitate intercellular communication and paracellular transport [8]. For instance, the absence of ZO-1 can delay the assembly of other TJ proteins in the epithelial layer, and the absence of claudin-1 can increase The SOD assay was performed as previously described with some modifications [22]. First, 2.9 mL reaction mixture [13.3 mM methionine, 63.0 µM nitro-blue tetrazolium chloride, 0.1 mM EDTA, and 1.3 µM riboflavin in 50 mM phosphate buffer (pH 7.0)] was added to 0.1 mL of cell lysate from selected Lactobacillus strains. The reaction mixtures were exposed to an ultraviolet lamp at room temperature for 20 min. A non-irradiated complete reaction mixture was used as a blank. The absorbance was measured at 560nm, and SOD activity was calculated using Formula (2).

Determination of β-Galactosidase Activity
The β-galactosidase activity of cell lysates from selected Lactobacillus strains was determined according to the Miller's method [23]. The reaction mixtures consisted of 2 mM o-nitrophenyl-β-d-galactoside (ONPG) in 0.05 M phosphate buffer (pH 7.0) and cell lysates of selected Lactobacillus strains. The reaction mixtures were incubated at 40 • C for 15 min and the reaction was stopped by adding 1.0 M Na 2 CO 3 . The yellowish end product of ONPG hydrolysis, o-nitrophenol, was collected by centrifugation at 12,000 rpm for 15 min at 4 • C and its absorbance was measured at 420 nm.

16S rDNA Sequencing and Identification
Genomic DNA of the strain with prominent acid and bile tolerance, antioxidant activity, and B-galactosidase activity was extracted using PureHelix™ Genomic DNA Prep Kit (Nanohelix, Daejeon, Korea). The almost complete 16S rDNA region of the selected strain was determined by Macrogen (Seoul, Korea) using the primers 27F and 1482R for polymerase chain reaction (PCR) amplification and sequencing as detailed in the Macrogen service website (https://dna.macrogen.com/eng/support/ces/ guide/ces_sample_prep.jsp). The 16S sequence was analyzed, and the 16S rDNA phylogenetic tree was constructed and visualized using NCBI (https://www.ncbi.nlm.nih.gov). The 16S rDNA sequence of Enterococcus faecalis V538 (AE0168830) was used as an outgroup.

Induction and Assessment of DSS-Induced Colitis
All animal experiments were approved and performed in accordance with the guidelines of the Institutional Animal Care and Use Committee of the Chonnam National University (CNU-IACUC-YB-2016-47). Eighteen five-week-old female C57BL/6J mice were purchased from Daehan Lab (Daejeon, Korea). The mice were housed and acclimatized for one week in the Animal Housing Unit under standard conditions of 22-25 • C, 50-60% humidity, and 12 h light/dark cycle. Standard mouse chow-diet and water were provided ad libitum. The mice were divided into three groups: control, DSS, and Bmb6 groups, with six mice in each group (n = 6). The experiment design is illustrated in Figure 1. In this experiment, PBS which had a similar osmolarity and ion concentration as the animal was used as an adjuvant. Mice in the control and DSS groups were orally administered with 100 µL PBS, while those in the Bmb6 group were orally administered with viable L. brevis BMB6 [10 9 colony forming units (CFU) in 100 µL PBS] daily. At day 7, the drinking water in the DSS and Bmb6 groups was replaced with 4% (w/v) DSS in water until the end of the experiment. The disease activity index (DAI) was assessed daily based on a scoring system (Table 1) [24]. On day-14, the mice were sacrificed. The length of the colons was measured, and the colon contents were carefully collected.

Histological Assessment
The collected colon was washed three times with sterile PBS. A part of the colon tissue was then fixed with 10% phosphate-buffer formalin for 24 h. After fixation, the tissue sample was dehydrated through an ethanol series, followed by embedding in paraffin. The paraffin blocks were then sectioned (5 µm) and stained with hematoxylin-eosin for histological evaluation. For mucosal layer evaluation, Alcian blue was used to stain the mucin in the paraffin-embedded sections and the nuclei were counterstained with nuclear fast red. For immunofluorescence analysis, the cut sections were stained with ZO-1 (#61-7300; Invitrogen, Carlsbad, CA, USA) or claudin-1 antibodies (#71-7800; Invitrogen) and goat anti-rabbit Alexa Flour 488 secondary antibody (A-11008; Invitrogen). 4′,6diamidino-2-phenylindole (DAPI; D-1306; Invitrogen) was used to counterstain cell nuclei. Slides were examined and analyzed using an epifluorescence microscope.

RNA Extraction and Gene Expression Analysis
Total RNA was extracted from the excised mouse colon using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA) according to manufacturers' protocol. Next, 2 μg of total RNA was used to synthesize complementary DNA (cDNA) using Maxime RT Premix Oligo (dT) RT-PCR kit (iNtRON Biotechnology, Inc., Seongnam, Korea). The primers used in the study are listed in Table 2. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control.

Histological Assessment
The collected colon was washed three times with sterile PBS. A part of the colon tissue was then fixed with 10% phosphate-buffer formalin for 24 h. After fixation, the tissue sample was dehydrated through an ethanol series, followed by embedding in paraffin. The paraffin blocks were then sectioned (5 µm) and stained with hematoxylin-eosin for histological evaluation. For mucosal layer evaluation, Alcian blue was used to stain the mucin in the paraffin-embedded sections and the nuclei were counterstained with nuclear fast red. For immunofluorescence analysis, the cut sections were stained with ZO-1 (#61-7300; Invitrogen, Carlsbad, CA, USA) or claudin-1 antibodies (#71-7800; Invitrogen) and goat anti-rabbit Alexa Flour 488 secondary antibody (A-11008; Invitrogen). 4 ,6-diamidino-2-phenylindole (DAPI; D-1306; Invitrogen) was used to counterstain cell nuclei. Slides were examined and analyzed using an epifluorescence microscope.

RNA Extraction and Gene Expression Analysis
Total RNA was extracted from the excised mouse colon using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA) according to manufacturers' protocol. Next, 2 µg of total RNA was used to synthesize complementary DNA (cDNA) using Maxime RT Premix Oligo (dT) RT-PCR kit (iNtRON Biotechnology, Inc., Seongnam, Korea). The primers used in the study are listed in Table 2. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control. Quantitative-PCR was performed on the Bio-Rad thermal cycler (Bio-Rad Laboratory, Hercules, CA, USA). The PCR conditions were as follows: Foods 2020, 9, 864 5 of 17 initial denaturation at 95 • C for 5 min, followed by 30 cycles of denaturation at 95 • C for 30 sec, annealing at 56 • C for 30 s, and extension at 72 • C for 5 min. The relative gene expression levels were determined by comparative analysis, using Formula (3). The relative gene expression was determined by quantitative PCR analysis with the PCR conditions of initial denaturation at 95 • C for 5 min, followed by 30 cycles of denaturation at 95 • C for 30 s, annealing at 56 • C for 30 s, and extension at 72 • C for 5 min.

Western Blotting
Total protein was extracted from the excised mouse colons using PRO-PREP protein extraction solution (iNtRON Biotechnology, Inc.). Briefly, 100 mg of the excised mouse colon tissue was immersed in 500 mL of PRO-PREP solution and homogenized using Q125 sonicator (QSonica, Sonicator, Newtown, CT, USA) for 5 min (pulsing mode of 15 s on and off) and 40% power in the ice bath. The mixture was then centrifuged at 13,000× g at 4 • C for 5 min, and the supernatant was collected as the extracted protein. The concentration of the extracted protein was determined using the Pierce™ BCA Protein Assay Kit (Thermo Fisher Scientific, Rockford, IL, USA). The extracted proteins (50 µg) were separated using 12% sodium dodecyl sulfate-polyacrylamide gel and electroblotted (Mini-PROTEAN ® II Cell Systems; Bio-Rad Laboratories, Hercules, CA, USA) onto the polyvinylidene difluoride membranes (PVDF; Bio-Rad Laboratories). The membranes were blocked with ZO-1 (#61-7800; Invitrogen), claudin-1 (#71-7800; Invitrogen), and β-actin (sc-4778; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) antibodies in 5% (w/v) skim milk-supplemented with Tween 20-Tris-buffered saline (TTBS) overnight at 4 • C. After incubation, membranes were washed three times with TTBS and incubated with horseradish peroxidase-conjugated goat anti-mouse or anti-rabbit antibodies. Protein bands were then developed and detected with enhanced chemiluminescence, and the band density was determined using β-actin as the reference protein.

Statistical Analysis
All data are presented as the mean ± standard deviation from three independent runs or six animals per group (n = 3 or 6). Tukey's multiple comparison test and repeat measure analysis of variance (ANOVA) were performed using SPSS 20 (SPSS, Inc., Chicago, IL, USA) with a p < 0.05 considered to be statistically significant.

In Vitro Characteristic of L. brevis Bmb6
As shown in Table 3, the growth of all strains decreased under acidic environments. The number of viable cells was only reduced by one log 10 in strains 2, 4, R10, P11, and Bmb6, whereas strains 1 and 5 were unable to survive in the acidic environment. For bile acid tolerance assay, strain Bmb6 exhibited prominent tolerance to bile acid, with a growth of two log 10 CFU/mL in the presence of 0.3% (w/v) ox-gall.

Antioxidant Activity
The DPPH radical scavenging assay results (Figure 2a) demonstrated that Lactobacillus strain Bmb6 was the most effective DPPH radical scavenger (59.48%), as compared to strains R10 (23.59%) and P11 (29.85%). The SOD assay results (Figure 2b) showed that both Bmb6 and P11 exhibited similar enzymatic activities of 80.28 and 80.97%, respectively, which were significantly higher than those of R10 (76.41%). Hence, our data suggested that Bmb6 strain exhibited a more significant antioxidant activity than R10 and P11 strains.

β-galactosidase activity
The β-galactosidase activity was determined by measuring the production of yellowish onitrophenyl, the end product of ONPG hydrolysis ( Figure 3). Based on our results, strain Bmb6 showed the highest (434 ± 21 unit/mL) while strain P11 showed the lowest β-galactosidase activity (21 ± 5 unit/mL), revealing the probiotic potential of strain Bmb6.

β-Galactosidase Activity
The β-galactosidase activity was determined by measuring the production of yellowish o-nitrophenyl, the end product of ONPG hydrolysis ( Figure 3). Based on our results, strain Bmb6 showed the highest (434 ± 21 unit/mL) while strain P11 showed the lowest β-galactosidase activity (21 ± 5 unit/mL), revealing the probiotic potential of strain Bmb6.

β-galactosidase activity
The β-galactosidase activity was determined by measuring the production of yellowish onitrophenyl, the end product of ONPG hydrolysis (Figure 3). Based on our results, strain Bmb6 showed the highest (434 ± 21 unit/mL) while strain P11 showed the lowest β-galactosidase activity (21 ± 5 unit/mL), revealing the probiotic potential of strain Bmb6.

16S Identification
16S rDNA analysis using BLAST showed that strain Bmb6 had 100% homology with L. brevis. Moreover, a neighbor-joining tree (data not shown) was constructed based on the 16S rDNA, and a close grouping of Bmb6 with the other eighteen L. brevis strains was observed, which was in agreement with the BLAST results, thus validating its identity as L. brevis Bmb6.

Effects of L. brevis Bmb6 on DSS-Induced Colitis Mice
The control mice exhibited the longest colon length (7.35 ± 0.68 cm), with normal shaped feces in the light red color colon (Figure 4a,b). In comparison, DSS mice had a shorter colon, with an average length of 6.83 ± 0.31 cm and dark red color, with intestinal hemorrhage. Meanwhile, the colons of Bmb6 mice had an average length of 6.77 ± 0.23 cm. There was no significant difference in the colon length among the three groups of mice (control, DSS, and Bmb6). However, the colons of the Bmb6 group had a lighter color than those of the DSS group, indicating that L. brevis Bmb6-treated mice display higher alleviation of intestinal hemorrhage than DSS-induced mice.

Histological Analysis of Colon Sections
Hematoxylin-eosin-stained colon sections showed normal and intact intestinal epithelial structures with no symptoms of immune cell infiltration in the control group ( Figure 5). However, Figure 4. Effects of L. brevis Bmb6 on the (a,b) colon length and (c) disease activity (DAI) score of DSS-induced colitis mice. Data represent the mean ± standard deviation of six mice from each treatment group (n = 6). Repeated measure ANOVA was performed for DAI score, and different lowercase letters indicate statistically significant differences (p < 0.05). On day-10, the DAI score began to rise in the DSS group, and loose, unformed, and bloody stools were visible to the naked eye (Figure 4c). The DAI score of the DSS group further increased to 7.50 ± 0.71 on day 13, following which mice displayed a drastic decrease in body weight and watery diarrhea with blood till the end of day 14. Similar to the DSS group, the DAI score of the Bmb6 group started to increase on day 10. Administration of L. brevis Bmb6 significantly alleviated colitis symptoms, as loose and hemoccult feces and a lower DAI score of 4.33 ± 0.58 was observed at the end of the study, as compared to the DSS group. Meanwhile, the control group showed no colitis symptoms, and the DAI score fluctuated around 1.0.

Histological Analysis of Colon Sections
Hematoxylin-eosin-stained colon sections showed normal and intact intestinal epithelial structures with no symptoms of immune cell infiltration in the control group ( Figure 5). However, intense infiltration of immune cells and severe epithelial structural damage was observed in the colon of DSS-induced mice. In contrast, the colon of L. brevis Bmb6-treated mice was greatly improved, with only slight epithelial structural damage and immune cell infiltration. intense infiltration of immune cells and severe epithelial structural damage was observed in the colon of DSS-induced mice. In contrast, the colon of L. brevis Bmb6-treated mice was greatly improved, with only slight epithelial structural damage and immune cell infiltration. The Alcian blue-stained colon micrographs ( Figure 6) showed that the mucous layer and goblet cells were well protected and preserved in the control group. However, the mucous layer was disrupted, and only a limited number of goblets cells were detected in the DSS group. In contrast, a higher number of goblet cells and only a slight alteration in mucus integrity was observed in the Bmb6 group.   The Alcian blue-stained colon micrographs ( Figure 6) showed that the mucous layer and goblet cells were well protected and preserved in the control group. However, the mucous layer was disrupted, and only a limited number of goblets cells were detected in the DSS group. In contrast, a higher number of goblet cells and only a slight alteration in mucus integrity was observed in the Bmb6 group.

Effects of L. brevis Bmb6 on TJ Proteins
As shown in Figure S1a, ZO-1 protein was detected in the colon sections of the Bmb6 group but was absent in the DSS group. Moreover, western blot analysis showed that the expression of ZO-1 was significantly increased by 192.12 ± 70.44% in the Bmb6 group, but significantly decreased by 89.02 ± 18.92% in the DSS group (Figure 7). Meanwhile, claudin-1 was detected in the colon of control, DSS, and Bmb6 groups ( Figure S1b). As revealed by western blot analysis, a non-significant increase in the expression of claudin-1 was observed in the Bmb6 group, compared to the DSS group (Figure 7). The overall findings indicate the L. brevis Bmb6 ameliorates colitis by partially recovering the expression of ZO-1 protein.

Effects of L. brevis Bmb6 on TJ Proteins
As shown in Figure S1a, ZO-1 protein was detected in the colon sections of the Bmb6 group but was absent in the DSS group. Moreover, western blot analysis showed that the expression of ZO-1 was significantly increased by 192.12 ± 70.44% in the Bmb6 group, but significantly decreased by 89.02 ± 18.92% in the DSS group (Figure 7). Meanwhile, claudin-1 was detected in the colon of control, DSS, and Bmb6 groups ( Figure S1b). As revealed by western blot analysis, a non-significant increase in the expression of claudin-1 was observed in the Bmb6 group, compared to the DSS group ( Figure  7). The overall findings indicate the L. brevis Bmb6 ameliorates colitis by partially recovering the expression of ZO-1 protein. Data represent the mean ± standard deviation (n = 6 mice per group). Tukey's multiple comparison test was performed, and different lowercase letters indicate statistically significant differences (p < 0.05).

Effects of L. brevis on Colonic Inflammatory Cytokine Expression
As shown in Figure 8, the gene expression of interferon (IFN)-γ and tumor necrosis factor (TNF)α was significantly increased upon DSS treatment. However, treatment with L. brevis Bmb6 suppressed IFN-γ expression to a level similar to that in control. Moreover, TNF-α gene expression was significantly suppressed in the Bmb6 group compared to that in the DSS group but remained at a higher level than the control group. Meanwhile, no difference in interleukin (IL)-1β gene expression Data represent the mean ± standard deviation (n = 6 mice per group). Tukey's multiple comparison test was performed, and different lowercase letters indicate statistically significant differences (p < 0.05).

Effects of L. brevis on Colonic Inflammatory Cytokine Expression
As shown in Figure 8, the gene expression of interferon (IFN)-γ and tumor necrosis factor (TNF)-α was significantly increased upon DSS treatment. However, treatment with L. brevis Bmb6 suppressed IFN-γ expression to a level similar to that in control. Moreover, TNF-α gene expression was significantly suppressed in the Bmb6 group compared to that in the DSS group but remained at a higher level than the control group. Meanwhile, no difference in interleukin (IL)-1β gene expression was observed between the control, DSS, and Bmb6 groups. Tukey's multiple comparison test was performed, and different lowercase letters indicate statistically significant differences (p < 0.05).

Discussion
Numerous studies have reported that consumption of fermented food helps in alleviating gastrointestinal disorder symptoms [28][29][30][31][32]. Notably, the functional microorganisms, especially those from the genera Bifidobacterium, Lactobacillus, Lactococcus, and Pediococcus, and their bioactive metabolites found in fermented food, have shown to having been responsible for these functions [19,20,28,[30][31][32][33][34][35]. Hence, in this study, potential lactic acid bacteria were isolated from the local homemade fermented food, kimchi, to investigate their regulatory effects on alleviating colitis symptoms, especially on TJ recovery. Our preliminary studies indicated that L. brevis Bmb6 possess several probiotic traits, including tolerance to acidic and bile acid conditions, which resemble the gut environment, and a prominent β-galactosidase activity.
In addition, L. brevis Bmb6 possesses strong antioxidant potential, as indicated by its prominent radical scavenging activity in the preliminary DPPH scavenging and SOD assays. A growing body of evidence suggests that elevated reactive oxygen and nitrogen species are closely associated with intestinal inflammation [36]. Accumulation of these radicals creates a high-oxidative stress environment, damaging the mucin layer and epithelial cells, subsequently stimulates the infiltration of immune cells and secretion of inflammatory mediators, and initiates gut inflammation [37,38]. Several studies have shown that treatment with SOD or high-SOD-producing Lactobacillus significantly reduces colonic myeloperoxidase level, oxidative stress, and inflammation in DSSinduced colitis mice [39][40][41][42]. Although the prominent in vitro radical scavenging activities of L. brevis Tukey's multiple comparison test was performed, and different lowercase letters indicate statistically significant differences (p < 0.05).

Discussion
Numerous studies have reported that consumption of fermented food helps in alleviating gastrointestinal disorder symptoms [28][29][30][31][32]. Notably, the functional microorganisms, especially those from the genera Bifidobacterium, Lactobacillus, Lactococcus, and Pediococcus, and their bioactive metabolites found in fermented food, have shown to having been responsible for these functions [19,20,28,[30][31][32][33][34][35]. Hence, in this study, potential lactic acid bacteria were isolated from the local homemade fermented food, kimchi, to investigate their regulatory effects on alleviating colitis symptoms, especially on TJ recovery. Our preliminary studies indicated that L. brevis Bmb6 possess several probiotic traits, including tolerance to acidic and bile acid conditions, which resemble the gut environment, and a prominent β-galactosidase activity.
In addition, L. brevis Bmb6 possesses strong antioxidant potential, as indicated by its prominent radical scavenging activity in the preliminary DPPH scavenging and SOD assays. A growing body of evidence suggests that elevated reactive oxygen and nitrogen species are closely associated with intestinal inflammation [36]. Accumulation of these radicals creates a high-oxidative stress environment, damaging the mucin layer and epithelial cells, subsequently stimulates the infiltration of immune cells and secretion of inflammatory mediators, and initiates gut inflammation [37,38]. Several studies have shown that treatment with SOD or high-SOD-producing Lactobacillus significantly reduces colonic myeloperoxidase level, oxidative stress, and inflammation in DSS-induced colitis mice [39][40][41][42]. Although the prominent in vitro radical scavenging activities of L. brevis Bmb6 suggest its potential role in alleviating colitis, its effect on lowering oxidative stress in vivo has not yet been evaluated.
For in vivo evaluation, DSS was used to induce colitis in mice. The clinical symptoms of DSS-induced mice are similar to those of patients with IBD, which include bloody stool, diarrhea, elevated inflammatory biomarkers, and drastic weight loss [43]. In our previous study, we reported that L. brevis Bmb6-containing fermented milk significantly improved the DAI score in DSS-induced mice [20]. This is in agreement with our current results which showed that the DAI score was significantly decreased in the Bmb6 group, with no drastic weight loss, and only non-forming hemoccult stools were observed. Despite weight loss and bloody diarrhea, shortened colon length, colon hemorrhage, and destruction of intestinal epithelial structures are typically observed in DSS-induced colitis mice [44][45][46]. In this study, administration of L. brevis Bmb6 alleviated colon hemorrhage while preserving the intact intestinal epithelial structure but showed no effect on restoring the colon length. Histological analyses revealed that L. brevis Bmb6 treatment alleviated colonic hemorrhage in DSS-induced colitis mice by preserving the integrity of the epithelial structure and preventing mass infiltration of immune cells and destructive damage of goblet cells in the colon, thereby contributing to the lower DAI score and absence of bloody diarrhea in the Bmb6 group. Preservation of goblet cells is crucial for maintaining mucosal barrier integrity via the secretion of mucins and trefoil factors [47]. Therefore, our findings suggest that the colitis-ameliorating effect of L. brevis Bmb6 was partly attributed to its ability to preserve the integrity of the intestinal epithelial structure.
The expression and localization of TJ proteins, such as ZOs, occludin, and claudins play a key role in regulating intestinal barrier functions. Impairment of the TJ barrier increases the paracellular permeation of pro-inflammatory molecules and activates mucosal immune response, resulting in chronic inflammation and tissue damage [6,[48][49][50]. In the current study, the expression and localization of ZO proteins, particularly ZO-1, were investigated. ZO proteins have been known for regulating the assembly and maintenance of TJ structure. For instance, the most well-studied ZO protein, ZO-1, was shown to be involved in the early assembly of TJ proteins into the cells and connecting adjacent cells in both cell cultures and animal models. Moreover, the absence of ZO-1 protein resulted in impaired TJ function, wherein TJ formation was delayed, and no occludin and claudins were detected [10,51]. Consistent with previous studies, the expression of ZO-1 was reduced in the DSS group. However, the administration of L. brevis Bmb6 significantly improved the expression of ZO-1 in the Bmb6 group, indicated by western blot analysis. Hence, our findings indicate a strong correlation between ZO-1, DAI score, and colon tissue histology, highlighting the role of L. brevis Bmb6 in TJ recovery and preservation of the intestinal epithelial structure, thereby ameliorating colitis.
In addition to ZO-1, the expression of claudins, which are critical components of the TJ barrier, was also investigated in the present study. Claudins connect adjacent cells via heterophilic and hemophilic interactions, forming pores and barriers for paracellular permeation of specific molecules in various tissues [52][53][54]. For instance, the critical role of claudin-1 in TJ barrier function was demonstrated using claudin-1 knockout mice, who displayed impaired epidermal barrier function and died within 24 h owning to excessive loss of body electrolytes [9,55]. However, the role of claudin-1 in IBD remains unclear. A number of studies have reported an elevation in claudin-1 expression in both experimental colitis models and patients with IBD [56][57][58]. Meanwhile, studies have also reported a decrease in claudin-1 expression in an experimental colitis model [59][60][61]. Notably, our findings showed that the expression of claudin-1 was not significantly different among control, DSS, and Bmb6 groups. We were unable to correlate the results of claudin-1 expression with the DAI score and histological analyses data. Hence, we were unable to conclude the role of claudin-1 in the experimental colitis model. Further studies are needed to clarify the role of claudin-1 in the pathogenesis of colitis.
Abnormal and uncontrolled secretion of inflammatory cytokines is also commonly seen in patients with IBD. Among the vast number of cytokines, TNF-α plays a key role in the pathogenesis of IBD by inducing inflammation and cell apoptosis, and subsequently intestinal TJ barrier defect in intestinal epithelial cells [62,63]. As a pro-inflammatory cytokine, TNF-α activates NF-κB and initiates the pro-inflammatory cascade by recruiting other pro-inflammatory factors, such as IFN-γ and IL-1β, further intensifying inflammatory reaction and intestinal dysfunction [48,64]. Therefore, elevated levels of IFN-γ and IL-1β are commonly detected in patients with IBD. An increase in IFN-γ and IL-1β levels has been shown to alter the expression and distribution of TJ proteins among the intestinal epithelial cells, resulting in hyper-paracellular permeation [65][66][67]. Hence, in this study, the gene expression of the pro-inflammatory cytokines TNF-α, IFN-γ, and IL-1β was assessed. Treatment with L. brevis Bmb6 significantly suppressed the gene expression of TNF-α and IFN-γ, but not IL-1β in DSS-induced colitis mice. These outcomes were consistent with the histological analyses data showing improved protein expression of the TJ protein, ZO-1, and well-preserved intestinal epithelial structure in L. brevis Bmb6-treated DSS-induced colitis mice, thereby indicating that regulation of inflammation conditions can restore the expression of TJ protein or vice versa.
Taken together, L. brevis Bmb6 isolated from the fermented food, kimchi, exerted significant colitis-ameliorating effects through suppression of pro-inflammatory cytokines, improvement of ZO-1 protein expression, and preservation of the intestinal epithelial structural integrity. Our findings showed the colitis-ameliorating effects in L. brevis Bmb6 could be attributed to a close association between inflammation and TJ barrier functions, wherein a reduction in inflammatory response can restore the expression of TJ protein, ZO-1, and vice versa. Notably, the expression of claudin-1 remained controversial in the present experimental colitis model. Further investigation is needed to clarify the role of claudin-1 in IBD pathogenesis.