Male NIH Swiss mice aged 6–8 weeks were obtained from the Guangdong Medical Laboratory Animal Center (Guangdong, China). The mice were housed under specific pathogen-free conditions at the Animal Laboratory Center of Tongji Medical College (Wuhan, China). They were housed individually at a temperature of 23–24°C with a light-dark cycle of 12–12 h and were allowed free access to standard mouse food and water. Each mouse was gavaged with 350–400 T. spiralis larvae in 0.2 ml phosphate-buffered saline (PBS). T. spiralis cultures were acquired from the Department of Parasitology and Microbiology, Tongji Medical College (Wuhan, China). All experiments were approved by the Ethics Committee of Tongji Medical College, Huazhong University of Science and Technology.

The CRD protocol was performed as previously described (31). Briefly, a catheter with a balloon was coated with lubricant and inserted 2 cm from the anal verge. The behavioral response to CRD was assessed based on the AWR using a semi-quantitative scoring system (32). AWR scores were recorded following the application of ascending-limit phasic distension (20, 40, 60 and 80 mmHg) for 20 sec every 4 min. Measurement of the AWR score was repeated three times at each pressure value. The CRD stimulus intensity that elicited contraction of the abdominal wall was recorded as the threshold. Each balloon inflation value was recorded five times by an observer in a blinded manner to ensure for accuracy. More than six mice were included in each group.

The mice were sacrificed by cervical dislocation, and ileal samples were fixed in buffered 10% formalin. Paraffin-embedded tissues were cut into 5 µm sections and stained with hematoxylin and eosin (H&E).

The ileum was quickly removed from each mouse and flushed with ice-cold Krebs solution (121 mM NaCl, 25 mM NaHCO₃, 3.8 mM KCl, 1 mM KH₂PO₄, 1.2 mM CaCl₂, 1.2 mM MgSO₄ and 11.1 mM glucose). The external muscle tissue and myenteric plexus were stripped off of the intestinal specimen. Sections of ileal villus epithelium were macroscopically identified as previously described (33,34). Each piece was placed in a Ussing chamber (Physiology Instruments, Santiago, CA, USA), and both sides of the chamber were filled with 5 ml Krebs solution, which was oxygenated, and maintained at 37°C throughout the experiment. The spontaneous potential difference and short circuit current in the Ussing chambers were recorded after a 25 min equilibration period. Transepithelial electrical resistance was calculated with Acquire and Analyze 2.3 software.

Three milliliters of FITC-dextran (FD4, 1 mg/ml; Sigma-Aldrich, St. Louis, MO, USA) were added to the mucosal side of each chamber, and an equivalent volume of Krebs solution was added to the other side. At 30-min intervals, 100 µl samples were collected and transferred to 96-well plates in duplicate. Krebs solution (200 µl) was added to the Ussing chambers after fluid collection to equalize the volumes. FD4 flux in each sample was measured at 520 nm with a fluorescence microplate reader (Molecular Devices, LLC, Sunnyvale, CA, USA). The FD4 concentration was determined based on standard curves as previously described (35). The permeability of each piece of tissue was presented as the calculated FD4 flux over a 60–90 min period. More than six mice were included in each group.

Tissues were processed for transmission electron microscopic observations using modified standard procedures (36). The tissues were fixed in 2.5% buffered glutaraldehyde for 2 h at 4°C and were subsequently fixed in 1% osmium tetroxide for 2 h at room temperature. Then, they were dehydrated in graded alcohol and acetone and embedded in Epon 812 overnight. Next, ultra-thin 60–80 nm sections were cut with a diamond knife using a Leica Ultracut UCT (Leica Microsystems GmbH, Wetzlar, Germany) and stained with uranyl acetate and lead citrate. The ultrastructures of the tight junctions were observed using a HITACHI U8010 transmission electron microscope (Hitachi, Ltd., Tokyo, Japan).

At 8 weeks after Trichinella spiralis infection, CD11c⁺ LPMPs were isolated from the intestines of the PI-IBS and normal mice. The entire small intestine was collected from each mouse, cut open longitudinally, washed with PBS, and then cut into 0.5-cm pieces. Subsequently, the tissue pieces were incubated in PBS supplemented with 5% fetal bovine serum (FBS) (Life Technologies, Gaithersburg, MD, USA), EDTA (1 mM), DTT (1 mM), and penicillin (100 Units/ml)/streptomycin (100 µg/ml) (1%; Life Technologies) at 37°C for 20 min, followed by removal of the epithelium. This incubation/epithelial removal step was repeated twice, and then the tissues were cut into smaller pieces and subsequently digested with 1 mg/ml collagenase IV (Roche Diagnostics, Basel, Switzerland) in RPMI-1640 medium (Life Technologies) at 37°C for 60 min. Next, the cell suspension was sequentially filtered through 100-µm and 400-µm filters and washed with RPMI-1640. LPMCs were harvested using a discontinuous 40%/75% Percoll gradient (GE Healthcare, Chalfont, UK). After being washed and resuspended in MACS buffer, the LPMCs were incubated with a microbead-conjugated anti-CD11c antibody (Miltenyi Biotec GmbH, Bergisch Gladbach, Germany). CD11c⁺ LPMPs were selected using MACS columns (Miltenyi Biotec GmbH). The selected cells were routinely found to contain ~85% CD11c⁺ cells based on flow cytometry. Cell viability did not fall below 90% based on trypan blue exclusion assays.

CD11c⁺ LPMPs were generated in mice with or without Trichinella infection and were then adoptively transferred into NIH Swiss mice. Each mouse was administered 1×10⁶ CD11c⁺ LPMPs via intravenous injection into the tail. At 120 h after the transfer of CD11c⁺ LPMPs, AWR scores were determined. Intestinal tissues were collected for permeability assays and analysis of intestinal inflammation. More than six mice were included in each group.

Intestinal tissues were homogenized via mechanical disruption in RIPA buffer with a protease inhibitor cocktail (EDTA-free; Roche Diagnostics). The samples were then clarified via centrifugation at 12,000 rpm for 10 min at 4°C. Next, the supernatants were diluted with loading buffer and heated to 95°C for 10 min. Subsequently, protein concentrations were determined using a BCA protein assay kit (Thermo Fisher Scientific, Waltham, MA, USA). The samples were separated on 10–12% SDS-PAGE gels and transferred to polyvinylidene fluoride membranes (Millipore, Billerica, MA, USA). The membranes were then probed overnight at 4°C with antibodies against the TNF-α (goat anti-mouse; AF-410-NA), IL-4 (rat anti-mouse; MAB404) and IL-10 (rat anti-mouse; MAB417) cytokines (1 µg/ml, all from R&D Systems, Minneapolis, MN, USA), occludin-1 (rabbit anti-mouse; 71–1500) and claudin-1 (rabbit anti-mouse; 51–9000) (1.25 µg/ml; both from Life Technologies) or Gapdh (rabbit anti-mouse; ABS16; 0.5 µg/ml; Millipore) and were then probed with HRP-conjugated secondary antibodies [goat anti-rabbit IgG (111-035-003), goat anti-rat IgG (112-035-003) and rabbit anti-goat IgG (305-035-003); 0.25 µg/ml; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA, USA] at room temperature for 2 h. The results were analyzed with Quantity One 4.6.2 software (Bio-Rad Laboratories, Inc., Hercules, CA, USA). More than six mice were included in each group.

The AWR scores were compared using the Kruskal-Wallis one-way analysis of variance (ANOVA) on ranks test, and significant results (P<0.05) were further analyzed using the Wilcoxon rank sum test with Bonferroni correction for multiple comparisons (0.05/3). The other data are expressed as the mean ± SE and were analyzed via one-way ANOVA, followed by the least significant difference (LSD) test or Dunnett's T3 multiple range test. P<0.05 was considered to indicate a statistically significant difference. Statistical analyses were performed with SPSS version 19.0 (IBM SPSS, Armonk, NY, USA).

Severe inflammation was observed at 2 weeks post-infection; however, no discernible inflammation was detected at 8 weeks post-infection (Fig. 1A). H&E staining revealed that hyperemia, swelling and severe neutrophilic and eosinophilic infiltration were induced in the whole gut by Trichinella at 2 weeks post-infection. However, no obvious microscopic changes were observed at 8 weeks post-infection.

Next, we examined the cytokine expression levels to determine whether a T helper 1 or T helper 2 immune response was involved in the intestinal inflammatory response. IL-4 expression in the ileum was found to be significantly elevated at 8 weeks post-infection compared with that in the control mice and mice at 2 weeks post-infection (p=0.046; p=0.001); however, no significant difference in IL-10 or TNF-α expression was observed among these groups of mice (Fig. 1B). These results suggested that Trichinella infection in the chronic stage might be associated with a T helper 2 response.

The AWR scores began to increase during the 2nd week post-infection, and they remained elevated during the 8th week at CRD intensities of 40 and 60 mmHg (Fig. 2A). Similarly, the thresholds of the 2- and 8-week post-infection groups were lower than those of the control group (Fig. 2B). These results implied that visceral hyperalgesia was sustained even though intestinal inflammation had subsided by the 8th week after infection, and they further suggested that the 8-week post-infection group was a better model for PI-IBS with visceral hypersensitivity.

Next, we used a Ussing chamber system to examine transepithelial electrical resistance and FD4 flux to evaluate intestinal epithelial barrier function in the ileum after infection (Fig. 3).

The transepithelial electrical resistance of the ileal villus epithelium was decreased in both the 2- and 8-week post-infection groups compared with the control group (42.60±2.51 vs. 53.87±3.55, P=0.008; 40.61±1.95 vs. 3.87±3.55, P=0.002) (Fig. 3A). In contrast, FD4 flux, which reflects the barrier function of the paracellular pathways, exhibited a drastic increase in the ileal villus epithelium (P=0.001) (Fig. 3B). These results suggested that intestinal permeability increased immediately post-infection and that it remained elevated, even at 8 weeks when intestinal inflammation had subsided. To further identify which paracellular pathway was affected, we focused on the tight junctions and analyzed the occludin-1 and claudin-1 expression levels. As shown in Fig. 3C, occludin-1 and claudin-1 expression in the ileum were downregulated in the 2-week post-infection group compared with the control group (p=0.034; p=0.021); further, their expression was slightly but non-significantly downregulated in the 8-week group compared with the control group (Fig. 3C).

CD11c⁺ LPMPs were isolated from the intestines of PI-IBS mice and normal mice at 8 weeks after T. spiralis infection. The mice that received CD11c⁺ LPMPs from the PI-IBS or normal mice displayed inconspicuous microscopic inflammation, as visualized by H&E staining (Fig. 4A). No significant differences in the histological characteristics were observed between the mice that received CD11c⁺ LPMPs from the PI-IBS mice (AT-IBS group), those that received CD11c⁺ LPMPs from the normal mice (ATN group), and the untreated mice (CTR group). The ATN group was included in this study to ensure that CD11c⁺ LPMPs were the only variable in the experiment and to eliminate effects of the adoptive transfer on the experimental results. Although no inflammation or damage was observed in the microscopic examinations, increased IL-4 expression was detected in the ilea of the AT-IBS mice compared with that in the ilea of the CTR and ATN mice (P=0.019; P=0.048). There was no difference in IL-10 or TNF-α expression in the ileum between the CTR, ATN and AT-IBS mice (Fig. 4B).

The transfer of CD11c⁺ LPMPs from PI-IBS mice resulted in decreased transepithelial electrical resistance in the ileal villus epithelium compared with that in the CTR and ATN mice (31.68±5.41 vs. 55.23±3.81, P=0.001; 31.68±5.41 vs. 49.23±4.54, P=0.008) (Fig. 5A). In contrast, the FD4 flux in the ileal villus epithelium was increased following CD11c⁺ LPMP transfer (P=0.043) (Fig. 5B). These results suggested that intestinal permeability was increased in the ileal villus epithelium. Furthermore, tight junction protein expression was slightly but non-significantly downregulated in the AT-IBS mice compared with that in the CTR and ATN mice (Fig. 5C).

However, ultrastructural alterations of tight junction proteins were detected in the mice that received CD11c⁺ LPMPs from the PI-IBS mice. As previously described, the routine H&E staining performed in the present study revealed no differences in epithelial architecture between the control and adoptive transfer groups. Transmission electron microscopy imaging showed that the functional alterations described above were associated with ultrastructural changes in the inter-cellular junctions, including disassembly (Fig. 5D). The mice that received CD11c⁺ LPMPs from the PI-IBS mice exhibited increases in the apical intercellular distance and proportion of dilated junctions compared with the controls. These mice also displayed a greater number of junctions with perijunctional cytoskeletal condensation (Fig. 5D).

Finally, we examined visceral sensation after CD11c⁺ LPMP transfer. As illustrated in Fig. 6, the mice that received CD11c⁺ LPMPs from PI-IBS mice exhibited greater visceral sensitivity. Significant increases in the AWR scores were observed at pressures of 40 and 60 mmHg compared with those of the control mice (Fig. 6A). Similarly, the thresholds were lower following the adoptive transfer of CD11c⁺ LPMPs from the PI-IBS mice (Fig. 6B); consequently, the mice that received these cells exhibited visceral hypersensitivity.