The Caco-2 cells were initially treated for 4 h with different concentrations (1%, 2.5%, 5%, 7.5% and 10%) of ethanol, and cell viability was evaluated by the MTT and LDH assays. The results showed that cell viability, evaluated by the MTT assay, was not altered at ethanol concentrations of less than 5% (Figure 1A). Moreover, at concentrations of up to 10%, ethanol did not increase the release of the cytosolic enzyme lactate dehydrogenase into the medium (Figure 1B). Ethanol concentrations greater than 5% caused cell shedding, but not cell fragmentation. At an ethanol concentration of 10%, the ratio of shedding cells was up to 51.67% ± 3.36%. Therefore, subsequent experiments were carried out using ethanol concentrations ranging from 1% to 5%.

As shown in Figure 2A, ethanol induced a significant time-dependent decrease in TEER, with the lowest value obtained after 60 min of ethanol treatment. A time-dependent increase in fluorescent yellow flux resulting from the ethanol treatment was observed, with maximum flux occurring at 60 min (Figure 2C). The decreased TEER value was associated with an increase in the fluorescent yellow flux rate. RhoA shRNA partially inhibited ethanol-induced epithelial leakage and restored the normal TEER values and fluorescent yellow flux rate of the IEB (Figure 2B,D). We achieved significant reductions of RhoA protein upon shRNA transfection (Figure 3), and the result showed the RhoA-381 group had higher transfection efficiency. The most effective shRNA was selected for the subsequent experiments. The experiments using luminal yellow (LY) flux and TEER assays were performed mainly to evaluate IEB permeability [18,19]. Our current results indicate that the ethanol-induced increase in IEB permeability is associated with RhoA. Ma et al.[3] showed that in vitro, ethanol decreased the TEER value in epithelial cells and opened TJ, which is consistent with our findings. Further studies to detect paracellular permeability using the LY flux rate, which is specifically transported through the paracellular space [20], yielded similar TEER results.

As shown in Figure 4, a time-dependent increase in RhoA activity following ethanol administration was observed. The increasing trend was consistent with the ethanol-induced increase in IEB permeability. RhoA shRNA interference inhibited the activation of RhoA at 60 min of treatment. These results suggest that the RhoA pathway may be involved in the ethanol-induced increase in IEB permeability. Rho, a member of a subfamily of small GTPases, is thought to be involved in many cellular functions, including the regulation of actin filament reorganization, cell-shape change and gene expression [21,22]. Extracellular stimuli convert inactive GDP-bound Rho into active GTP-bound Rho. Once activated, Rho interacts with its specific effectors, eliciting a variety of biological functions [23]. Recently, research has demonstrated the important roles of RhoA activation in modulating epithelial permeability [24]. Our studies show that RhoA is required for ethanol-induced changes in IEB permeability.

As shown in Figure 5, a time-dependent increase in myosin phosphorylation by ethanol administration was observed. RhoA shRNA interference decreased myosin phosphorylation after 60 min. Myosin phosphorylation is a downstream effect of Rho [25–27]. Rho may increase the phosphorylation of myosin light chains (MLC) by decreasing MLC phosphatase activity or by directly phosphorylating MLC at serine 19, its primary phosphorylation site [28–30]. It has also been demonstrated that myosin phosphorylation contributes to TJ disassembly [31–33]. Some investigators have also demonstrated that myosin phosphorylation can disrupt TJ and result in TJ opening [34,35].

The integrity of TJ is primarily responsible for epithelial barrier function [36,37]. The strands of TJ are composed of the transmembrane proteins occludin and claudin, polymerizing to form TJs [31]. Occludin is thought to be not only a structural, but also a functional, component of TJs [38]. Occludin, the first identified transmembrane protein of the TJ, is the major regulatory protein of TJs [39–41]. Post-translational modifications of the occludin protein involve phosphorylation events, and multiple phosphorylation sites have been identified on occludin serine and threonine residues [42]. Several proteins, such as Rho kinase, protein kinase C, protein phosphatase 2A and casein kinase 2, may regulate occludin phosphorylation [43,44]. The phosphorylation state of occludin mediates its association with the cell membrane and barrier permeability. Sakakibara et al.[45], using phosphoamino acid analyses, showed that phosphorylated occludin protein was selectively recovered in the NP-40-insoluble fraction and appeared in the higher Mr bands, whereas non-phosphorylated occludin protein was predominantly in the NP-40-soluble fraction and appeared in the lower Mr bands. Soluble fragments of occludin are the low molecular weight fragments that are susceptible to non-ionic detergent extraction. Insoluble fragments of occludin represent phosphorylated forms of high molecular weight fragments that combine with actin microfilaments through zonula occludens (ZOs) and that are resistant to non-ionic detergent extraction. As previously reported, a shift in the occludin distribution from insoluble to soluble fractions suggests that the TJs were destroyed [46,47]. Therefore, we prepared the NP-40-soluble and NP-40-insoluble protein fractions for hybridization with an occludin antibody. Our results demonstrated that ethanol induced a shift in the occludin distribution from insoluble to soluble fractions (Figure 6). At the same time, we found that ethanol induced decreased expression of ZO-1 (Figure 6). In our study, ethanol changed the ratio of occludin (S/IS) and expression of ZO-1 in Caco-2 cells, supporting the conclusion that ethanol can affect the expression of TJ proteins, disrupt the TJs and increase the permeability of the intestinal barrier. Moreover, our current data suggest that RhoA shRNA interference prevents the disruption of TJs induced by ethanol treatment (Figure 6). In order to clarify further the mechanism of TJ opening, the distribution of occludin and ZO-1 were assessed by immunofluorescence microscopy (Figure 7). Ethanol induced a re-distribution of occludin from plasmalemma to cytoplasm and induced distribution of ZO-1 discontinuously at the cellular membrane. Our findings reveal that RhoA shRNA interference prevented the ethanol-induced relocation of occludin and ZO-1. Accordingly, it seems possible for a signal transduction pathway to proceed from RhoA to occludin and ZO-1 disassembly, thereby resulting in IEB opening in our presented paper.

RhoA shRNA interference showed only partial reversal of the aforementioned data by ethanol treatment in our study. This result suggests that other signal molecule mechanisms may be involved in ethanol-induced IEB opening.

Caco-2 cells were acquired from American Type Culture Collection (ATCC) and cultured in Dulbecco’s Modified Eagles Medium (DMEM, Hyclone, Logan, UT, USA) supplemented with 4.5 mg/mL glucose, 50 U/mL penicillin, 50 U/mL streptomycin, 4 mM glutamine and 10% fetal calf serum (FCS, Hyclone, Logan, UT, USA) in a humidified atmosphere (95% air, 5% CO₂) at 37 °C. To establish an in vitro intestinal epithelial barrier model, Caco-2 cells were plated on Transwell filters (Corning, New York, NY, USA) and monitored regularly by visualization with an inverted microscope and by epithelial resistance measurements.

The viability assay was performed as described previously [48]. Briefly, Caco-2 cells were grown in 96-well plates, as described above. After incubation with ethanol (1%, 2.5%, 5%, 7.5%, 10%) for 4 h, the cells were incubated with 100 mL of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT, Sigma-Aldrich, Taufkirchen, Germany) at a concentration of 5 mg/mL MTT diluted in PBS for 1 h at 37 °C. The culture media was removed, and 150 μL of DMSO was added to 96-well plates. The absorbance of the resulting colored solution was measured at 570 nm with a microplate reader (Wallac Victor 2, Perkin-Elmer, Life Sciences, Boston, MA, USA). In this assay, yellow MTT solution is converted into a blue formazan dye within mitochondria and deposited intracellularly. The amount of blue staining is quantitatively detectable by spectrometry and is a measure of cell viability.

At the same time, cell viability was assessed by measuring the release of the cytosolic enzyme. The electrical resistance of Caco-2 cell monolayers cultured on Transwell filters was measured using a Millicell-ERS instrument (Millipore, Bedford, MA, USA). Electrical resistance was expressed in units of Ω cm² using the surface area of the Transwell insert.

Fluorescent yellow (Sigma-Aldrich) at 40 μg/mL in serum-free Dulbecco’s modified Eagle’s medium (DMEM) was added to the upper chamber of the Transwell system. After ethanol was administered for the times indicated, the media from the lower chamber was collected. Absorbance was detected by fluorescence spectrophotometer (excitation wavelength 427 nm, emission wavelength 536 nm), and the fluorescent yellow concentration was calculated according to the standard curve. The fluorescent yellow flux rate (%) = the fluorescent yellow concentration in the lower chamber/the fluorescent yellow concentration in the upper chamber. Caco-2 cells were incubated in different concentrations of ethanol for 4 h. LDH measurement was carried out in 250 μL aliquots using the LDH kit from Doles reagents.

The electrical resistance of Caco-2 cell monolayers cultured on Transwell filters was measured using a Millicell-ERS instrument (Millipore, Bedford, MA, USA). Electrical resistance was expressed in units of Ω cm² using the surface area of the Transwell insert.

Fluorescent yellow (Sigma-Aldrich, Milwaukee, WI, USA) at 40 μg/mL in serum-free Dulbecco’s modified Eagle’s medium (DMEM) was added to the upper chamber of the Transwell system. After ethanol was administered for the times indicated, the media from the lower chamber was collected. Absorbance was detected by fluorescence spectrophotometer (excitation wavelength 427 nm, emission wavelength 536 nm) and the fluorescent yellow concentration was calculated according to the standard curve. The fluorescent yellow flux rate (%) = the fluorescent yellow concentration in the lower chamber/the fluorescent yellow concentration in the upper chamber.

To block the activity of RhoA, Caco-2 cells were transfected with U6/GFP/Neo- RHOA-homo-326 and U6/GFP/Neo-RHOA-homo-381 (U6/GFP/Neo-RHOA shRNA expression plasmid, GenePharma, Shanghai, China) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s protocol. Negative control sequence (pGPU6/GFP/Neo-shNC) was purchased from GenePharma and used as a control. ShRNA interference efficiency was evaluated by Western blot analysis of the target protein (RhoA) expression. The shRNA sequences were shown in Table 1.

Caco-2 cells were plated onto the underside of the Transwell insert with suitable culture medium. When the Caco-2 cells were 80% confluent, they were seeded in the upper chamber of the Transwell insert. The cells became confluent after 21–23 days of culture. Ethanol (5%) was added to the upper chamber. The cultured cells were then rinsed with cold phosphate buffered saline (PBS). Subsequently, the Caco-2 cells in the upper chamber were collected by gently scraping with a cell scraper, and the cells were then stored at −80 °C. There were eight groups in our study, with n = 4 per group: control group (Caco-2 cells monolayer), ethanol (5%) 0 min group, ethanol (5%) 15 min group, ethanol (5%) 30 min group, ethanol (5%) 60 min group, ethanol (5%) 60 min +RhoA-326 group, ethanol (5%) 60 min + RhoA-381 group, ethanol (5%) 60 min + NC group (respectively, Caco-2 cells were transfected with U6/GFP/Neo-RHOA-homo-326, U6/GFP/Neo-RHOA-homo-381, pGPU6/GFP/Neo-shNC in the upper chamber of the Transwell and stimulated with ethanol (5%) for 60 min).

The Caco-2 cells were washed three times with Dulbecco’s phosphate-buffered saline (D-PBS) containing 0.1 mM ethylenediamine tetraacetic acid (EDTA) without calcium and magnesium, and the Caco-2 cells were then homogenized in 1 mL of lysis buffer A (2 mM EDTA, 10 mM ethylene glycol tetraacetic acid (EGTA), 0.4% NaF, 20 mM Tris-HCL, protease inhibitor cocktail, phosphatase inhibitor 1% Triton X-100, pH 7.5) at 4 °C. Samples were centrifuged at 14,000× g for 30 min, and the supernatant was transferred to a separate tube and collected as the soluble fraction (S). Buffer A (150 μL) with 1% sodium dodecyl sulfate (SDS) at 4 °C was then added to the pellet. The pellet was disrupted with an ultrasonic crusher. The samples were then centrifuged at 14,000× g for 30 min at 4 °C, as described above. The supernatant was collected as the insoluble fraction (IS). Equal amounts of proteins (40–50 μg) were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and processed for immunoblotting with antibodies for MLC, myosin phosphorylation, occludin and ZO-1 (diluted 1:1,000, Santa Cruz Biotechnology, San Francisco, CA, USA). All the protein bands were scanned using ChemiImager 5500 V2.03 software (AlPha InnCh, San Leandro, CA, USA, 2002), and the integrated density values (IDV) were calculated by a computerized image analysis system (Fluor Chen 2.0, Bio-Rad, Hercules, CA, USA) and normalized to that of β-actin.

Briefly, the cells were lysed as described above, and the supernatants were collected for protein quantification. Equal amounts of protein were incubated with GST-Rhotekin (Milipore, Billerica, MA, USA) at 4 °C for 1 h with gentle agitation. Affinity-precipitated RhoA proteins were resolved by SDS-PAGE and detected by Western blot analysis. Aliquots of lysates were electrophoresed and blotted in parallel with anti-RhoA (diluted 1:500; Santa Cruz Biotechnology) to determine total RhoA levels (“total RhoA”).

The Caco-2 cell monolayers grown on glass coverslips were fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X-100. After blocking with 2% BSA in PBS, the cells were incubated with rabbit anti-occludin (diluted 1:50; Santa Cruz Biotechnology) and rabbit anti-ZO-1 (diluted 1:50; Santa Cruz Biotechnology) to visualize the distribution of occludin and ZO-1. The glass slides were analyzed using immunofluorescence microscopy (Olympus, Tokyo, Japan).

Experiments were performed at least three times. Data are presented as the mean ± standard deviation (SD). Comparisons between more than two groups were performed using a one-way analysis of variance (ANOVA), followed by Dunnett’s post-test.