Effect of Enterococcus faecalis 2001 on colitis and depressive-like behavior in dextran sulfate sodium-treated mice: involvement of the brain–gut axis

Background Patients with inflammatory bowel disease (IBD), including those with ulcerative colitis and Crohn’s disease, have higher rates of psychiatric disorders, such as depression and anxiety; however, the mechanism of psychiatric disorder development remains unclear. Mice with IBD induced by dextran sulfate sodium (DSS) in drinking water exhibit depressive-like behavior. The presence of Lactobacillus in the gut microbiota is associated with major depressive disorder. Therefore, we examined whether Enterococcus faecalis 2001 (EF-2001), a biogenic lactic acid bacterium, prevents DSS-induced depressive-like behavior and changes in peripheral symptoms. Methods We evaluated colon inflammation and used the tail suspension test to examine whether EF-2001 prevents IBD-like symptoms and depressive-like behavior in DSS-treated mice. The protein expression of tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), X-linked inhibitor of apoptosis protein (XIAP), and cleaved caspase-3 in the rectum and hippocampus was assessed by western blotting. Hippocampal neurogenesis, altered nuclear factor-kappa B (NFκB) p65 morphometry, and the localization of activated NFκB p65 and XIAP were examined by immunohistochemistry. Results Treatment with 1.5% DSS for 7 days induced IBD-like pathology and depressive-like behavior, increased TNF-α and IL-6 expression in the rectum and hippocampus, activated caspase-3 in the hippocampus, and decreased hippocampal neurogenesis. Interestingly, these changes were reversed by 20-day administration of EF-2001. Further, EF-2001 administration enhanced NFκB p65 expression in the microglial cells and XIAP expression in the hippocampus of DSS-treated mice. Conclusion EF-2001 prevented IBD-like pathology and depressive-like behavior via decreased rectal and hippocampal inflammatory cytokines and facilitated the NFκB p65/XIAP pathway in the hippocampus. Our findings suggest a close relationship between IBD and depression.


Background
Inflammatory bowel disease (IBD), which comprises ulcerative colitis and Crohn's disease, affects approximately 2.2 million people in Europe and 1.4 million people in the USA. Recent studies have demonstrated a connection between intestinal inflammation and changes in brain function [1]. Inflammation in the bowel is associated with alterations in the central nervous system, as revealed by the activation of tumor necrosis factor-α (TNF-α) signaling and microglia in the brain [2]. Other researchers have demonstrated that chronic experimental colitis increases anxiety behavior in mice [3]. Further, peripheral inflammation may account for at least some of the neurological and behavioral changes associated with chronic inflammatory diseases. Indeed, patients with IBD have higher rates of obsessive-compulsive disorder, panic disorder, depression, and anxiety [4][5][6][7].
A well-characterized mouse model of IBD is produced by repeated administration of dextran sulfate sodium (DSS) in drinking water [8]. DSS does not cross the blood-brain barrier because of its higher molecular weight. Epithelial cell toxicity, increased intestinal permeability, and macrophage activation are implicated in the deleterious effects of DSS. The DSS model is characterized by colonic epithelial cell lesions and acute (7)(8)(9)(10)(11)(12)(13)(14) days after the beginning of the treatment) intestinal inflammation [9]. Repeated DSS cycling in combination with treatment with azoxymethane, a genotoxic agent, induced colitis-dependent neoplasia, generating a commonly used model for colorectal neoplasia and cancer in humans [10]. Recent studies have reported that DSS-treated rodents exhibit anxiety-and depressive-like behavior [11] and reduction of hippocampal neurogenesis [12].
Decreased adult hippocampal neurogenesis is associated with depression in rodents and humans [13][14][15]. Moreover, depression is associated with altered inflammation [16], which manifests due to increased inflammatory cytokine expression [17]. Pro-inflammatory cytokines also inhibit adult neurogenesis in the hippocampus [18][19][20]. Therefore, cytokine-induced disruption of neurogenesis might be a key link between inflammation and depression. Antidepressants enhance hippocampal neurogenesis [21] and regulate several apoptotic factors, which are involved in cell survival pathways [22]. Treatment of mood disorders, including depression and anxiety, is critically dependent on intact adult neurogenesis in the hippocampal dentate gyrus (DG) [23,24]. Thus, the stimulation of neurogenesis and reduction of apoptosis may constitute important drug targets in the modulation of depressive symptoms [25].
Enterococcus faecalis 2001 (EF-2001) is a biogenic lactic acid bacterium that is used as a biological response modifier (BRM). Live E. faecalis 2001 can be heattreated to produce a BRM containing high levels of β-glucan, named EF-2001. β-Glucan, one constituent of EF-2001, is a ligand for toll-like receptor 2 (TLR2) and activates nuclear factor-kappa B (NFκB) p65, which controls spontaneous apoptosis and anti-apoptotic effects. NFκB p65 activation inhibits apoptosis by increasing Xlinked inhibitor of apoptosis protein (XIAP), an antiapoptotic factor that exerts its action by regulating caspase-3 activity [26,27]. EF-2001 can decrease serum inflammatory cytokines in a contact dermatitis model mouse [28], has anti-tumor effects [29], and protects chemical-induced colitis [30]. Therefore, we hypothesized that EF-2001 may attenuate inflammation and apoptosis in DSS-treated mice. Additionally, reports indicate that E. faecalis modulates inflammation in colitis models [31,32]. However, the effect of EF-2001 on colitis-induced depression is unclear.
Taken together, we examined whether EF-2001 prevents DSS-induced depressive-like behaviors and changes in peripheral symptoms and investigated the neuroprotective molecular mechanisms underlying these effects.

Materials and methods
All experiments were performed following approval of the Ethics Committee of Animal Experiments in Tohoku Medical and Pharmaceutical University (approval numbers: 16023 cn, 17015 cn, and 18031 cn) and according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize suffering and reduce the number of animals used.

Compounds
Commercially available heat-treated EF-2001 was originally isolated from healthy human feces. It was supplied as a heat-killed, dried powder by Nihon BRM Co. (Tokyo, Japan). DSS (0.75%, 1.5%, or 3%; Wako Pure Chemical Industries Ltd., Osaka, Japan) and EF-2001 (250 mg/kg) were dissolved in drinking water. Mice were given drinking water containing DSS ad libitum for 7 days to induce colitis. Dexamethasone (Dex; 0.1 mg/kg; Aspen Japan, Tokyo, Japan) and imipramine (Imi; 20 mg/kg; Sigma-Aldrich, St-Louis, USA) were dissolved in saline. EF-2001 was administered orally (per os [p.o.]) from 14 days before the beginning of DSS administration until the day prior to the last DSS treatment. Dex and Imi were administered intraperitoneally (i.p.) starting on the same day as the first DSS administration until the last day of DSS treatment. The dose for each drug used was calculated from previous reports [29,33,34].

Evaluation of colon inflammation
This evaluation was conducted according to the experimental protocol shown in Fig. 1a, b. Disease activity index (DAI) scores are well correlated with pathological findings in a DSS-induced model of IBD [35]. DAI scores were calculated as described previously [36]. DAI is the combined score of stool consistency and bleeding, as detailed in Table 1. When mice were sacrificed, the colon length, starting above the anus to the top of the cecum, was measured. All parameters were scored on days 3, 5, and 7 during DSS treatment.

Tail suspension test
This test was conducted according to the experimental protocol shown in Fig. 1a, b. The tail suspension test was conducted to assess depressive-like behaviors and antidepressant effects. The procedure was performed as described previously [37]. Mice were considered Fig. 1 Experimental time course for the behavioral tests, assessment of inflammation, histology, western blotting, and immunohistochemical of the experimental protocol A, experimental protocol B, and experimental protocol C immobile only when they hung completely and passively motionless. Mice were suspended 30 cm above the floor by means of adhesive tape placed approximately 1 cm from the tip of the tail. The duration of time spent immobile was quantified during a test period of 10 min.

Locomotor activity
This test was conducted according to the experimental protocol shown in Fig. 1a. The locomotor activity of mice was evaluated using the multichannel activitycounting system SUPERMEX (Muromachi Kikai Co., Tokyo, Japan). The procedure and instrument have been reported previously [38]. This instrument can monitor even minute movements in all three planes of motion (sagittal, vertical, and horizontal) as one movement, owing to its infrared sensor with multiple Fresnel lenses that can be moved close enough to the cage to capture multidirectional locomotor alterations in a single mouse. This system interprets each movement as a count. Thus, the vertical movements, such as jumping, and horizontal movements, such as walking and running, can be counted. Activity measurements were conducted between 11:00 and 15:00 during the light phase. Mice were divided into three groups (water, DSS 1.5%, and DSS 3%) and were placed in activity boxes during 15 min for adaptation. Locomotor activity was recorded for 60 min.

Histology
The histological analysis was conducted according to the experimental protocol shown in Fig. 1b. Histology was assessed in mice following DAI and behavioral evaluations. The entire colon was collected and fixed in 4% buffered formalin for 24 h at room temperature, imbedded in paraffin, and sliced. Samples were sectioned into 5-μm slices and subjected to staining with hematoxylin and eosin (HE). Finally, sections were examined under a light microscope to evaluate the histopathologic changes in colon tissue.

Western blotting
Western blotting was performed according to the experimental protocol shown in Fig. 1b. Mice were divided into four groups (water/water, water/EF-2001, DSS/water, and DSS/EF-2001). Mice were sacrificed by decapitation after 20 days of water or EF-2001 administration. Protein isolation and western blots were performed as described previously [37,39]. Sectioning was performed as described previously [40,41]. After sacrifice, the rectum of each mouse was washed in ice-cold phosphate-buffered saline (PBS). The rectal tissue, 8 mm from the edge of the cecum (side of the anus), was carefully cut into 5-mm slices on ice. The brain was removed and sectioned on ice using a mouse brain slicer (Muromachi Kikai) to produce 1-mmthick coronal sections. To ensure precise regions, the cerebral peduncle was regarded as a landmark, and the five edge blades were anteriorly placed from this landmark. We visually confirmed the dorsal hippocampal location using Paxinos and Franklin mouse brain atlas [42]. After electrophoresis, proteins were transferred to a PVDF membrane, which was then incubated with blocking solution [10 mM Tris-HCl (pH 7.4), 100 mM NaCl, 0.01% Tween 20, and 5% skim milk] for 1 h. Next, membranes were probed with antibodies against TLR2 (1:100; Cell Signaling Technology, Danvers, USA), TNF-α (1:1000; Cell Signaling Technology), interleukin-6 (IL-6; 1:1000; Cell Signaling Technology), XIAP (1:200; Abcam Ltd., Cambridge, UK), brain-derived neurotrophic factor (BDNF; 1:100; Abcam Ltd.), and β-actin (1:1000; Cell Signaling Technology) overnight at 4°C. Membranes were washed with blocking solution without milk and incubated with horseradish peroxidase-conjugated secondary antibody (Cell Signaling Technology) for 2 h, followed by visualization of the immunoreactive species with ECL Western Blotting Detection Reagent (Amersham Life Science, Piscataway, USA). Band densities were analyzed with ImageJ 1.43 (National Institutes of Health).

Immunohistochemical analysis
Immunohistochemical analysis was conducted according to the experimental protocol shown in Fig. 1c. To assess neurogenesis, on day 20 after EF-2001 administration, 5bromo-2′-deoxyuridine (BrdU; Sigma-Aldrich; 75 mg/ kg, i.p.) was injected three times every 2 h after the last administration of water or EF-2001. Animals were subsequently sacrificed 24 h after the last injection. Brain samples were collected as described previously [37,38]. The brains were cut into 40-μm sections from bregma − 2.20 to − 2.80 mm using a cryostat (MICROM HM560, Mikron Instrument, Inc., California, USA).
Frozen sections were mounted on glass slides (Matsunami Glass, Osaka, Japan). Sections were treated with HCl (2 N) at 37°C for 30 min, followed by neutralization with sodium borate buffer (0.15 M) at room temperature twice every 10 min. After three washes every 5 min, the sections were incubated with PBS containing 1% normal goat serum (Life Technologies Corporation, Carlsbad, USA) and 0.3% Triton X-100 (PBSGT) at room temperature for 2 h. The sections were incubated

Double immunofluorescence staining
Immunofluorescence analysis was conducted according to the experimental protocol shown in Fig. 1b. The brain samples were collected as described previously [43,44]. When double labeling was performed using two primary antibodies from the same host species (rabbit anti-TLR2, rabbit anti-NFκB p65, rabbit anti-XIAP, and rabbit anti-Iba1 antibodies), the detection of each antigen was performed sequentially and labeled goat anti-rabbit IgG Alexa Fluor 488 AffiniPure Fab fragments (Jackson ImmunoResearch Laboratories, USA), instead of whole antibodies, were used in the first detection (Iba1). The immunohistochemical staining with two primary antibodies from the same host species was carried out as described previously in detail [45]. Immunofluorescent images were analyzed with a confocal laser-scanning microscope (A1Rsi; Nikon).

Neuromorphometrical study
Morphometric assessment of the brain was conducted according to the experimental protocol shown in Fig. 1b. The brain samples were collected as described previously [37,38]. The sections were incubated overnight at 4°C with rabbit anti-NFκB p65 antibody (1:500; Cell Signaling Technology). Sections were washed and incubated for 2 h at room temperature with goat anti-rabbit IgG Alexa Fluor 568 (1:200; Molecular Probes) in PBSGT. We observed alterations in activation of NFκB p65positive cells in the hippocampal DG area with a confocal laser-scanning microscope. We then evaluated the activation of NFκB p65-positive cells by observing translocation to cell nuclei.

Statistical analysis
Normality and homoscedasticity assumptions were verified before the use of any parametric tests (Shapiro-Wilk normality test and equality of variances F test). Results are expressed as mean ± standard error of the mean (SEM). The significance of differences was determined by the Student's t test for two-group comparisons or by one or two-way analysis of variance (ANOVA), followed by Tukey-Kramer tests for multigroup comparisons using GraphPad Prism 7 (GraphPad Software, California, USA) and StatView 5.0 (HULINKS, Tokyo, Japan). For the DAI scores, statistical significance of differences was assessed with a non-parametric Mann-Whitney test for two-group comparisons or Kruskal-Wallis test followed by Steel's test for multigroup comparisons. In some cases, when a main effect was significant without interaction effect, we did an exploratory and limited pairwise post hoc comparison consistent with our a priori hypothesis. Results were considered statistically significant if p < 0.05.

Results
Concentration-dependent effect of DSS on DAI scores, colon length, immobility time, and locomotor activity in mice DAI scores of both stool consistency and rectal bleeding in DSS-treated mice (1.5% and 3%) were significantly increased compared with those in the control group [ Fig. 2a, b, Kruskal-Wallis test, stool consistency: p < 0.0001, rectal bleeding: p < 0.0001]. The colon length in DSS-treated mice (0.75%: p = 0.0007, 1.5%: p < 0.0001, and 3%: p < 0.0001) was significantly shorter than in control mice [Fig. 2c We observed that DSS 3% caused the death of a few mice (data not shown). Based on these results, DSS 1.5% was the appropriate dose to investigate in the IBD model with depression.

Time-dependent effects of DSS on DAI scores, colon length, and immobility time in mice
As shown in Fig. 3, diarrhea and shortened colon length were observed on day 7 of DSS treatment, but not on days 3 and 5 [ Fig. 3a, Mann-Whitney test, day 3: p > 0.9999, day 5: p = 0.3173, day 7: p = 0.0009; Fig. 3c Fig. 4i-l).

Effect of EF-2001 on TNF-α and IL-6 levels in the rectum and hippocampus of DSS-treated mice
As shown in Fig. 5, the immunocontent of TNF-α and IL-6 in the rectum and hippocampus of DSStreated mice was significantly increased compared with controls. Interestingly, these changes were reversed by treatment with EF-2001 [two-way ANOVA,

NFκB p65-positive astrocytes and microglia in the hippocampal DG after treatment with EF-2001
NFκB p65 may control spontaneous apoptosis, antiapoptotic gene expression, and translocation to cell nuclei during activation. Two-way ANOVA showed significant main effects for the treatment factor but not an interaction [ Fig. 8f, two-way ANOVA, group: F (1, 15) = 3.237, p = 0.0922, treatment: F (1, 15) = 34.35, p < 0.0001, group × treatment: F (1, 15) = 0.4169, p = 0.5282]. Therefore, we focused our analysis on the effects of treatment. EF-2001 treatment significantly increased activation of NFκB p65positive cells in the DG compared with the water-treated group. To determine which cell types were involved, dual immunofluorescence staining for NFκB p65 was performed in conjunction with cell-specific markers, such as DCX, NeuN (a marker for mature neurons), GFAP (an astrocyte marker), and Iba1 (a microglia marker). Activated NFκB p65-positive cells were identified as astrocytes and microglia (Fig. 8g).

Effect of EF-2001 on TLR2 levels in the hippocampus of DSS-treated mice
Two-way ANOVA showed statistical significance for the main effects of treatment [F (1, 14) = 16.49, p = 0.0012] but no significance for the TLR2 interaction [group × treatment: F (1, 14) = 0.13, p = 0.7239] (Fig. 9a). Thus, we focused our analysis on the major effects of EF-2001 administration. EF-2001 significantly increased TLR2 expression in the hippocampus of DSS-treated mice. Moreover, to identify the cell types involved in the production of TLR2, dual immunofluorescence staining was performed for the localization of TLR2 and cell-specific  production of XIAP, dual immunofluorescence staining for XIAP was performed in conjunction with cellspecific markers, such as GFAP and Iba-1. XIAP was localized in the microglia (Fig. 10d).

Discussion
Patients with IBD have higher rates of psychiatric disorders, such as depression and anxiety; however, the mechanisms underlying a link between intestinal inflammation and depressive-like symptoms are largely unknown. In this study, we investigated the effect of EF-2001 in IBD-like physiological changes and depressive-like behavior in DSS-treated mice. Chronic administration of EF-2001 prevented such changes. In addition, EF-2001 attenuated the increase of inflammatory cytokines in the rectum and hippocampus, attenuated the reduction of neurogenesis in the hippocampus, and facilitated the NFκB p65/XIAP pathway in the hippocampus of DSS-treated mice. This is the first report that the antidepressant effect of EF-2001 may involve hippocampal neuroprotection via decreased inflammatory cytokine expression in the rectum and hippocampus as well as apoptotic cell death regulation via inhibition of caspase-3 activity through facilitation of the NFκB p65/XIAP pathway in the hippocampus.
IBD, including Crohn's disease and ulcerative colitis, is a chronic relapsing condition characterized by intestinal damage (barrier disruption, altered microbiota) and high levels of inflammation such as elevated inflammatory cytokines [43,46,47]. Cytokines induce extensive inflammation in the colon, which has a negative impact on epithelial cells, resident and recruited immune cells, and stromal cells [44]. Specifically, inflammation can cause damage to epithelial cells, and activate and recruit immune and stromal cells, ultimately leading to nonresolving chronic inflammation and the development of IBD [44]. Moreover, during acute or chronic inflammation, inflammatory cytokines can induce the development of depression [48,49]. Clinical studies have reported that patients with IBD often exhibit obsessivecompulsive disorder, panic disorder, depression, and anxiety [4][5][6][7]. DSS treatment induces colonic epithelial cell lesions and intestinal inflammation, including elevated inflammatory cytokines, via epithelial cell toxicity, increased intestinal permeability, and macrophage activation [9]. It has been suggested that E. faecalis improves colitis by increasing interleukin-10 (IL-10), a factor that inhibits cytokine synthesis, in colonic epithelial cells [50,51]. The present study showed that EF-2001 reduced inflammatory cytokines in the rectum (Fig. 5). This effect was similar to that of steroids, which are commonly prescribed for treating IBD [52]. Recently, other researchers have reported that EF-2001 protects dinitrobenzene sulfonic acid-induced colitis, a chemically induced colitis model, via a decrease in inflammatory cytokines [30]. Thus, the anti-inflammatory effect of EF-2001 may also be related to reduced inflammatory cytokines. Peripheral inflammation is a risk factor for developing mood or psychotic disorders, such as depression [53][54][55][56], and may affect hippocampal neurogenesis, including the proliferation, differentiation, and survival of newborn neurons [54,55]. Adult neurogenesis occurs in two main regions of the brain, one of which is the subgranular zone of the DG [57,58]. In the present study, we observed that DSS treatment significantly increased TNF-α and IL-6 levels in the rectum and hippocampus. A previous study has reported that DSS treatment significantly increases rectal TNF-α and IL-6 levels [59]. Interestingly, hippocampal TNF-α and IL-6 mRNA in DSS-treated mice were unchanged compared with controls (Fig. 6). Depression is closely associated with altered inflammation [16], manifested by increased expression of inflammatory cytokines such as TNF-α and IL-6 [17]. Neuroinflammatory factors, such as TNFα and IL-6, can negatively affect many stages of neurogenesis in the adult mammalian brain, including the proliferation, differentiation, and survival of newborn neurons [18-20, 54, 55]. Therefore, cytokine-induced reduction of neurogenesis might establish a key link between inflammation and depression. In this study, DSS-treated mice showed a significant decrease in neurogenesis in the DG, consistent with a previous study [12]. Likewise, DeCarolis and Eisch reported a reduction in neurogenesis in the hippocampus of patients with depression [60]. These findings suggest that DSS-induced depressive-like behavior may be associated with the reduction of neurogenesis in the DG via the release of inflammatory cytokines derived from peripheral inflammation. Moreover, several studies have suggested that antidepressant effects are critically dependent on intact adult neurogenesis and may be mediated by the enhancement of neurogenesis in the hippocampal DG [21,23,24]. We observed that administration of EF-2001 significantly attenuated the enhancements of rectal and hippocampal inflammation and reduction of newborn neurons in the hippocampus of DSS-treated mice. Therefore, we suggest that the antidepressant effect and enhanced neurogenesis observed upon EF-2001 administration are partly independent effects, resulting from the EF-2001-mediated reduction of peripheral inflammation. Although the mechanism by which pro-inflammatory cytokines reduce neurogenesis is not fully understood, we believe that inflammatory cytokines in peripheral tissue might be key mediators of this process. In human neutrophils, the activation of NFκB p65 seems to control spontaneous apoptosis and anti-apoptotic effects. Unexpectedly, we found that the activation of NFκB p65-positive cells were increased by EF-2001 administration in water-and DSS-treated groups compared with control groups. Moreover, we found that activation of NFκB p65 was localized to hippocampal astrocytes and microglia; however, the mechanisms underlying this activation remain unclear. We hypothesized that β-glucan may be associated with activation of NFκB p65 via TLR2. β-Glucan activates TLR2 and increases TLR2 expression via NFκB p65 activation [61][62][63]. Interestingly, we found that hippocampal TLR2 was activated by EF-2001, which was localized in all cell types, including immature and mature neurons, astrocytes, and microglia (Fig. 9). We did not assess the migration of EF-2001 or β-glucan into the brain; we will examine them in a future study. Modulation of this pathway most likely regulates the balance between pro-and anti-apoptotic factors [64], thus affecting neutrophil survival. In this study, EF-2001 attenuated DSSinduced neuroinflammation and we hypothesized that NFκB p65 may play a role in anti-apoptosis. The activation of NFκB p65 inhibits apoptosis via a mechanism involving upregulation of various anti-apoptotic genes, such as cellular FLICE-inhibitory protein, Bcl-xL, and XIAP [26,27]. XIAP, a key member of the inhibitors of apoptosis protein family, can inhibit apoptosis by directly binding to the initiator caspases: caspase-3, -7, and -9 [65]. Moreover, hippocampal XIAP regulates synaptic plasticity, which is associated with the development of depression [66,67]. In this study, XIAP was significantly increased by EF-2001. Further, XIAP was localized in hippocampal microglia. Moreover, cleaved caspase-3, which is crucial in the process of apoptosis and contributes to the irreversible stage of apoptosis [68], was significantly increased in the hippocampus of DSS-treated mice. In contrast, EF-2001 suppressed the increased levels of cleaved caspase-3 in the hippocampus of DSS-treated mice. These results suggested that EF-2001 might partly modulate apoptosis via regulation of the microglial NFκB p65/XIAP pathway and caspase-3 in the hippocampus of DSS-treated mice.
As summarized in Fig. 11, EF-2001 had antiinflammatory and antidepressant effects in DSS-treated mice, and hippocampal neuroprotection is a key factor mediating the antidepressant effect of EF-2001. Other researchers have reported that administration of Lactobacillus strains had antidepressant effects and enhanced neurogenesis in the hippocampus via the vagus nerve [69,70]. Hence, further extensive experiments examining the relationship between the vagus nerve and effects of EF-2001 will be presented in a subsequent paper.

Conclusions
The present study showed that DSS-treated mice exhibited IBD-like physiological changes and depressive-like behaviors (Figs. 2 and 3). This result was consistent with previous studies [11,12]. In this study, we evaluated the predictive validity of the IBD animal model by using a classic antidepressant drug, Imi, or a steroid, Dex. Administration of Imi significantly improved depressivelike behavior, whereas Dex significantly improved both IBD-like peripheral symptoms and depressive-like behavior. These results demonstrated that DSS-treated mice provided a model of IBD with depression.
Our results indicate that EF-2001 attenuated IBD-like symptoms and depressive-like behavior in DSS-treated mice. EF-2001 prevented DSS-induced colitis and the mechanism may also involve the suppression of inflammatory cytokines in the rectum. The antidepressant effect of EF-2001 may involve neuroprotection in the hippocampus via decreased TNF-α and IL-6 expression in the rectum and hippocampus and facilitation of the NFκB p65 pathway in the hippocampus. This process is likely mediated by modulation of XIAP, which is involved in the regulation of apoptotic cell death via caspase-3 activity. Moreover, our findings suggest a close relationship between IBD and depression.