IEC-6 is a rat ileal epithelial cell line obtained from the Korean Cell Line Bank (No. 21592). Lipopolysaccharides (LPS, L4524) from Escherichia coli 055:B5 were purchased from Sigma (St. Louis, MO, USA). Ad/LacZ (adenoviral vectors expressing E. coli β-galactosidase) and adenoviral vector system expressing IGFBP-3 (Ad/IGFBP-3) were made using the AdEasy vector system (Quantum Biotechnologies, Montreal, CA, USA).13 To create Ad/LacZ, a SalI-NotI fragment from pcDNA3.1/LacZ was ligated into the cytomegalovirus (CMV) vector. As for Ad/IGFBP-3, NotI-XbaI fragments from pcDNA3/IGFBP-3 DNA were ligated into CMV.14 These recombinations and the pAdEasy-1 viral backbone were created by homologous recombination in E. coli (strain BJ5183). Recombinant DNA of adenovirus was verified and transferred to DH5α (E. coli strain). After linearizing by PacI-digested DNA, the recombinant CMV-cDNA was transferred to QBI293A (human embryonic kidney cells) using the calcium phosphate transfection method (Promega, Madison, WI, USA). Recombinant viruses were amplified in QBI293A and purified by CsCl centrifugation (100,000 g at 4°C for 20 hours). The adenoviruses were titrated by plaque analysis using serial dilution infection. The infection was measured by the multiplicity of infection (MOI) in subconfluent cells in % fetal bovine serum (FBS) medium.

The cells were cultured in the Dulbecco's Modified Eagle medium (DMEM; Life Technologies, Grand Island, NY, USA), supplemented with L-glutamine (300 μg/mL), penicillin (100 U/mL), streptomycin (100 μg/mL), and 5% FBS. One million cells were seeded in 10 mL culture dishes at 37°C and incubated overnight in a humidified atmosphere of 5% carbon dioxide. Once they reached confluence, the cells were detached from the plate by trypsinization, washed, and replated in DMEM. In this study, we used cells between passages 5 and 10. IEC-6 cells were divided into 6 groups: 3 normal groups without LPS stress and 3 with LPS stress. To establish the LPS stress model, when the cells reached over 90% confluence, the medium was replaced with DMEM containing 0.5% FBS and the cells were incubated for 18 hours. After a starvation period, the cells were infected with 100 MOI of Ad/LacZ or Ad/IGFBP-3. After a 2 hours incubation, the medium was replaced with DMEM containing 0.5% FBS, and the cells were stimulated with 10 μg/mL LPS for 24 hours.

Cell viability was measured using the MTT assay as described previously. The MTT solution (Sigma) was used to evaluate IEC-6 cell viability or cytotoxicity after treatments. The treated cells were distributed in 96-well plates. After removing the DMEM, 20 μL of MTT solution (200 μg/mL) was added to each well and incubated for 3 hours at 37°C. The supernatant was then removed and 150 μL of dimethyl sulfoxide (DMSO) was added. After vigorous shaking to dissolve the formazan crystals, absorbance of the wells was measured using a spectrophotometer (ELX 800; Bio-Tek Instruments, Winooski, VT, USA) at a wavelength of 540 nm. Cell viability was calculated using the following equation:

At the indicated time, ROS generation was analyzed by FACScan flow cytometry (Becton Dickinson, Franklin Lakes, NJ, USA) using a 2′,7′-dichlorofluorescin diacetate (DCFH-DA) (Fluka, Molecular Probes, Eugene, OR, USA) fluorescent probe. IEC-6 cells in 10 mL plates were incubated with 5 μmol/L DCFH-DA for 30 minutes at 37°C in the dark. After washing two times with phosphate buffered saline (PBS), the cells were collected by centrifugation (900 rpm for 4 minutes). The mean fluorescence at the emission wavelength of 520 nm was measured after excitation at 480 nm. The data were analyzed by Becton Dickinson Partec software and expressed as a representative histogram.

Antibodies against COX-2 (sc-1745), interleukin (IL)-1β (sc-7884), TNF-α (sc-1350), NF-κB (sc-109), IGFBP-3 (sc-9028), and lamin B (sc-6216) were purchased from Santa Cruz (Santa Cruz, CA, USA). Antibodies against β-actin (#A-2066) were purchased from Sigma, phospho-p38 MAPK (#9211) and phospho-ERK (#9101) from Cell Signaling (Beverly, MA, USA), and Anti-Cu/Zn (#SOD-100) and Anti-Mn SOD (#SOD-110) were from Stressgen (Ann Arbor, MI, USA).

Western blotting was performed according to the standard protocol.13 Proteins were extracted from cells by immersing the cells in non-denaturing lysis buffer (Tris-HCl, NaH₂PO₄, NaCl, Triton X-100, sodium pyrophosphate, pH 7.5) for 10 minutes on ice. Cellular debris was eliminated by centrifugation. The samples were stored at −80°C. The protein concentrations were quantified using the Bradford dye-binding assay kit (Bio-Rad, Hercules, CA, USA). Twenty micrograms of protein were diluted, loaded, and separated by 8% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) without reducing conditions. Proteins were electroblotted onto Hybond filters (Amersham Biotech, Piscataway, NJ, USA). Membranes were blocked with 5% nonfat milk/Tris-buffered saline and Tween (TBST) for 30 minutes at room temperature and incubated with the appropriate antibody for 2 hours. Protein bands were detected using enhanced chemiluminescence (NEN, Boston, MA, USA).

Eight-week old female mice (C57BL6) weighing 25–30 g were purchased from Samtako (Osan, Korea). All mice were placed in plastic cages and had access to water and food, under 12-hour dark/light cycles at controlled temperatures (25°C–27°C).

Dextran sodium sulfate (DSS) was purchased from MP Biomedical (MW: 36,000–50,000; Solon, OH, USA). The mice were sorted into 4 experimental groups: 1 control group and 3 colitis groups. To induce colitis in mice,15 the mice were given drinking water mixed with 3% DSS for 7 days (day 1 to 7). On day 7, the DSS-mixed water was replaced with purified water. Meanwhile, the control group received normal water. The mice were monitored carefully and their body weights were measured daily. Ad/IGFBP-3 or Ad/LacZ was administered into the rectosigmoid colon by enema on day 3 and day 6. For this procedure, mice were anesthetized using diluted ketamine without any complication. After PBS enema and washing, 200 μL of Ad/LacZ or Ad/IGFBP-3 (10¹¹ pfu) were added using a 5 Fr catheter (inserted up to 4 cm in length from anus). The other 2 groups (control and DSS colitis) were intracolonically injected with normal saline using the same protocol. On day 10, the mice were sacrificed by vertebral dislocation, the colons were then removed and rinsed in PBS. To evaluate colitis severity, colon length was measured macroscopically, and histological studies, including microscopic scoring, were performed. Histological evaluation was performed by two scoring systems. The crypt scoring was based on pathological changes as follows: score 0 = intact crypt, score 1 = loss of one-third of the crypts, score 2 = loss of two-thirds of the crypts, score 3 = loss of whole crypt with mild inflammatory cell infiltration, score 4 = erosions with marked inflammatory cell infiltration. Involvement scoring means percentage of total mucosal slide involved as follows: score 0 = normal colonic mucosa, score 1 = 1%–25% involvement, score 2 = 26%–50% involvement, score 3 = 51%–75% involvement, score 4 = 76%–100% involvement.16

Results were expressed as the means ± standard error, where applicable. Data analyses were performed using unpaired Student's t-test (parametric) and the Mann-Whitney U test (nonparametric) for comparison. P value < 0.05 was considered statistically significant. SPSS software v21.0 (IBM, Armonk, NY, USA) was used for statistical analysis.

All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. This research was performed by the approval of the Chonbuk National University Institutional Animal Care and Use Committee (IACUC; approval No.: CBNU 2015-094).

To confirm the expression of Ad/LacZ or Ad/IGFBP-3 in mouse colonic mucosa, mice were administered with the adenoviral particles through three administration routes: tail vein injection, peritoneal injection, and enema. The same amount of Ad/LacZ (10¹¹ pfu, PBS dilution) was injected into the tail vein (50 μL) and intraperitoneal space (100 μL), and infused into rectosigmoid colonic lumen (200 μL) with an enema using a 5 Fr catheter. The 4 groups of animals including the control group were sacrificed after 3 days. The liver and colon were isolated and stained to detect the level of LacZ expression (Fig. 1A). We found that the rectal route was the most effective method of administration for colon-specific expression. The same protocol was applied for Ad/IGFBP-3 administration. Immunohistochemical staining was performed using an IGFBP-3 antibody purchased from Bioworld Technology (Louis Park, MN, USA). Comparing the 4 groups for IGFBP-3 expression (Fig. 1B), the intracolonic administration was the most effective method for expressing IGFBP-3 using the adenoviral expression system in mice.

IEC-6 cells were sorted into 6 groups. To determine the effect of adenovirus infection on IEC-6 cells without LPS injury, two cell groups were treated with Ad/LacZ and Ad/IGFBP-3 viral particles, respectively. Cell viability or cytotoxicity was measured using an MTT assay 24 hours after treatment. The adenoviral infection had no cytotoxic effect on the cells without LPS stimulation (i.e., control, Ad/LacZ, and Ad/IGFBP-3 treated groups) (Fig. 2A). On the contrary, decreased cell viability was observed in the other 3 LPS-treated groups. However, pretreatment with Ad/IGFBP-3 significantly reduced LPS-induced cytotoxicity (Fig. 2A, P = 0.001).

ROS levels were detected using the fluorescent probe DCFH-DA. After 24 hours of 10 μg/mL of LPS stress, LPS-induced ROS formation was observed through increased fluorescence levels compared to levels in the three non-treated groups. There was no significant difference among the control, Ad/LacZ, and Ad/IGFBP-3 groups without LPS treatment. Compared to the 2 LPS-treated groups, pretreatment with Ad/IGFBP-3 showed marked protective effects with reduced ROS production in cells (Fig. 2B).

Cells were exposed to 10 μg/mL LPS for 24–48 hours. Inflammatory cytokines such as COX-2 or IL-1 were expressed by LPS stimulation in IEC-6 cells (Fig. 3A). Superoxide dismutase (SOD) is an important antioxidant enzyme that catalyzes and removes superoxide radicals generated by an inflammatory response. The expression of SOD enzyme and protein in the LPS-treated cells was gradually downregulated as time elapsed. In contrast, pretreating IEC-6 cells with Ad/IGFBP-3 sustained the level of Cu/Zn and Mn SOD (Fig. 3B). We next examined whether LPS-stimulated NF-κB would be influenced by IGFBP-3. Ad/IGFBP-3-treated cells showed reduced phospho-p38 MAPK, phospho-ERK (Fig. 3C), and nuclear NF-κB (Fig. 3D) levels in response to LPS. Upon LPS stimulation, the expression of pro-inflammatory cytokines IL-1β, COX-2, and TNF-α was upregulated. However, treatment with Ad/IGFBP-3 inhibited the expression of these cytokines (Fig. 3E). Phospho-IκB-α levels were also reduced by Ad/IGFBP-3 treatment (Fig. 3F). These results suggest that IGFBP-3 expression by Ad/IGFBP-3 suppresses inflammatory responses in LPS-treated IEC-6 cells.

The inflammation severity can be characterized by weight loss and shortened colonic length in DSS-induced colitis mice.17 The range of measured body weight of mice was 25–30 g on day 1, and there were no significant median weight differences among the groups. Sixteen mice were divided into 4 groups, and the mean values were compared among each group. DSS-induced colitis groups were characterized by loss of body weight (Fig. 4). On day 10, statistically significant differences in body weight measurements between the control and colitis groups were obtained, but the Ad/IGFBP-3 treated group showed a significant improvement in weight gain compared to the Ad/LacZ-treated group (Fig. 4, P = 0.028). After the end of the experiment (day 10), all mice were sacrificed by cervical dislocation. The colonic lengths were measured and the longest and the shortest lengths from each group were excluded. In the DSS-induced colitis groups, significant shortening of the colonic length was observed. This shortening was suppressed by IGFBP-3 expression through Ad/IGFBP-3 treatment (Fig. 5, P = 0.032).

Paraffin-embedded colon tissues were sliced and stained. Immunohistochemical staining was performed with IGFBP-3 antibodies from Bioworld Technology. IGFBP-3 expression was compared among the 4 experimental groups. DSS with Ad/IGFBP-3 treated group showed greater IGFBP-3 expression than was observed in the other 3 groups (Fig. 6). The tissues were stained with hematoxylin-eosin for microscopic examination. In the control group, the mucosa and the submucosa showed normal crypts and goblet cells without thickening, edema, or infiltration of inflammatory cells. The colitis mucosa presented crypt distortion or absence, infiltration of inflammatory cells, edema, and erosions or ulceration in histological studies. Treatment with Ad/IGFBP-3 markedly reduced these inflammatory injuries, showing relatively preserved crypts (Fig. 7A-D). In the crypt and the involvement scoring system, the Ad/IGFBP-3 therapy group showed significantly lower damage scores compared to scores of other colitis groups (Fig. 7E, P < 0.05).