Male C57BL/6 mice weighing 20±2 g were purchased from the Laboratory Animal Center of the Army Medical University (Chongqing, China). All animals were 6–8 weeks of age and specific pathogen-free. At the beginning of the study, the animals were housed under temperature-, humidity-, and light-controlled conditions. The laboratory animals were cared for according to the Guidelines for the Care and Use of Laboratory Animals, as set forth and approved by the University Committee at the Army Medical University.

After a 12-h fast, the mice were anesthetized with an intraperitoneal injection of sodium pentobarbital (50 ml/kg). The abdomens were opened via midline incisions. The mice were randomly assigned to the sham group and two I/R groups, as follows: i) sham group, no intestinal ischemia treatment; ii) I/R group, the superior mesenteric vessels (SMVs) were occluded for 30 min and an intraperitoneal injection of saline (0.3 ml) was administered 6 h before reperfusion; and iii) I/R+FICZ group, an intraperitoneal injection of FICZ [1 µg/mouse (0.3 ml)] was given before reperfusion. Then, the mice in the three groups were sacrificed by cervical dislocation. The mice were sprayed with 70% ethanol, an abdominal midline incision was performed and then the mesenteric lymph nodes were isolated and collected from the entire small intestine. The same individual performed these manipulations. The experiment was repeated at least three times.

The same intestinal segment (2 cm) from each animal was removed and immediately placed in 4% paraformaldehyde. Paraffin blocks were prepared from the tissue pieces. The blocks were cut into sections (5-µm thick), stained with hematoxylin and eosin (H&E) and observed by light microscopy. Histologic evaluations were graded from 0–5 by a single observer, as follows: grade 0, normal morphology; grade 1, sub-epithelial edema and partial separation of apical cells; grade 2, moderate lifting of enterocytes from the tips of the villi; grade 3, lifting of enterocytes from both the tips and sides of the villi, including the superficial crypts; grade 4, partial mucosal necrosis of the lamina propria; and grade 5, total mucosal necrosis.

After the small intestine was harvested, a 3-cm section of the Peyer's patch-free jejunum along the mesenteric border was opened and gently washed three times with cool RPMI-1640 medium to cleanse the luminal contents. The full-thickness small intestine was fixed in a modified Ussing chamber (Physiologic Instruments, San Diego, CA, USA). Each half cell (mucosal and serosal) was filled with 5 ml of pre-heated 37°C Krebs-Ringer solution buffer containing 110.0 mM NaCl, 5.5 mM KCl, 3.0 mM CaCl₂, 1.2 mM MgCl₂, 1.4 mM KH₂PO₄ and 29.0 mM NaHCO₃, adjusted to pH 7.4. The cells were continuously oxygenated with O₂/CO₂ (95%/5%) and stirred by gas flow in the chambers. One pair of Ag/AgCl electrodes (Physiologic Instruments) with 3 mM KCl in 3% agar bridges was used to calculate the transepithelial potential difference and another pair of Pt electrodes was used for current passage. The transmembrane resistance (TER) was determined using Ohm's law. The baseline short circuit current (ISC) was recorded for up to 120 min after a 20-min equilibration period.

The entire small intestine was removed and placed in tissue culture media (RPMI-1640). The small intestine was cut longitudinally into 5-mm pieces, washed 3 times with the same tissue culture media, and incubated in isolation buffer (54 ml of PBS, 22.8 mg of EDTA, 9.6 mg of DTT, and 6 ml of 10% fetal calf serum) with continuous gentle shaking at 37^ºC for 30 min. After centrifugation, the pellets were purified in 40% isotonic Percoll (GE Healthcare Biosciences) and reconstituted in RPMI-1640 tissue culture media. The viability of the IELs exceeded 96% based on trypan blue exclusion staining.

Total RNA was isolated from the sorted IELs using Trizol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturer's instructions. Briefly, RNA was reverse-transcribed into complementary DNA (cDNA) using a SuperScript First-Strand Synthesis System RT-PCR kit (Invitrogen; Thermo Fisher Scientific, Inc.), and this cDNA was used as a template for the amplification of β-actin, TNF-α, IL-1β, IL-6, Bcl2, Bcl2l1, KGF, CYP1A1 and CD127. The primer sequences are presented in Table I (gene symbol: forward, reverse). Gene expression was determined by qPCR, which was performed using a SYBR PrimeScript RT kit, as described by the manufacturer (Takara Bio Inc., Kusatsu, Japan). The PCR mixture (20 µl final volume per reaction) was prepared per the manufacturer's protocol. Amplifications were performed under the following conditions on a 7500 Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.): 94^ºC for 5 min; 35 cycles of 94^ºC for 30 sec, 59^ºC for 30 sec, and 72^ºC for 30 sec; and 72^ºC for 10 min. The expression of each target gene was calculated by the 2^−ΔΔCq method (19).

The IEL subtype was confirmed by fluorescent antibody staining, which was detected using flow cytometry. The following anti-mouse monoclonal antibodies (eBioscience, Inc., San Diego, CA, USA) were used: CD45-PE (dilution 1:100; cat. no. 12-0451-82); CD45-PerCP-Cy5.5 (dilution 1:100; cat. no. 45-0451-82); CD4-PE (dilution 1:100; cat. no. 12-0041-82); CD69-FITC (dilution 1:100; cat. no. 11-0691-82); CD8α-APC (dilution 1:100; cat. no. 17-0081-82); CD8β-PE (dilution 1:100; cat. no. 12-0083-82); TCRβ-APC (dilution 1:100; cat. no. 17-5961-82); CD127-PE-CY⁷ (dilution 1:100; cat. no. 25-1271-82); and TCRγδ-FITC (dilution 1:100; cat. no. 11-5711-82). Isotype-matched, irrelevant antibodies (eBioscience, Inc.) were used as negative controls: Rat IgG2b kappa Isotype Control (eB149/10H5), PE (cat. no. 12-4031-82); Rat IgG2b kappa Isotype Control (eB149/10H5), PerCP-Cyanine5.5 (cat. no. 45-4031-80); Armenian Hamster IgG Isotype Control (eBio299Arm), FITC (cat. no. 11-4888-81); Rat IgG2a kappa Isotype Control (eBR2a), APC (cat. no. 17-4321-81). The apoptotic ratios of γδIELs among the different groups were detected by flow cytometry per the manufacturer's protocol. γδIELs were stained with PE-Annexin V and cellular DNA was stained with7-aminoactinomycin D (7-AAD), as described in the protocol. The acquisition and analysis were performed using MoFlo XDP (Beckman Coulter, Inc., Brea, CA, USA).

To study associated gene expression in γδIELs, total IELs were stained with the following anti-mouse monoclonal antibodies (eBioscience, Inc.): FITC anti-mouse TCRγδ (dilution 1:100; cat. no. 11-5711-82); and PE anti-mouse CD8β (dilution 1:100; cat. no. 12-0083-82). γδIELs were purified using a cell sorter (Beckman Coulter). The TCRδ⁺/CD8β⁻ subsets were isolated from γδIELs and the purity was ≥98%.

All data are presented as the mean ± standard deviation (SD). The results were analyzed by one-way analysis of variance (ANOVA) followed by Tukey's post-hoc test using SPSS statistical software package. A P-value <0.05 was considered to indicate statistical significance.

Based on our previous study, the destruction of intestinal morphology was more severe 6 h after intestinal I/R. Therefore, this time-point was chosen for subsequent experiments. H&E was used to stain the small intestines from mice in the sham, I/R and I/R+FICZ groups. Representative images from each group are presented in Fig. 1A. The intestines of the sham group showed an intact epithelium, a normal villous length, well-defined glands, and minimal leukocyte infiltration in the mucosa. In contrast, severe mucosal damage, including villous blunting, epithelial denudation, gland distortion, leukocyte inflammation, and lamina propria disintegration, were observed in the I/R group (Fig. 1B). FICZ attenuated these changes. The histologic scores of intestinal mucosal injury are presented as the mean ± SD (Fig. 1D). Moreover, changes in intestinal permeability were detected using Ussing chambers. Compared with the I/R+FICZ group, the TER of the I/R group was significantly lower (P<0.01; Fig. 1C). These results showed that FICZ attenuates the IR-induced destruction of intestinal morphology and intestinal barrier function.

During reperfusion injury, pro-inflammatory molecules, such as IL-1β, IL-6 and TNF-α, directly induce tissue damage to affect intestinal barrier function and are potent activators of other neutrophils. Based on the attenuation of decreased intestinal barrier function, the effect of FICZ was confirmed on IL-1β, IL-6 and TNF-α gene expression. The expression of IL-1β, IL-6 and TNF-α mRNA in intestinal IELs is shown. Real-time PCR showed that IEL-derived IL-1β (P<0.05; Fig. 2), IL-6 (P<0.05; Fig. 2), and TNF-α (P<0.01; Fig. 2) mRNA expression was significantly increased after I/R compared with the sham group. Importantly, the expression of IL-1β (P<0.05; Fig. 2), IL-6 (P<0.05; Fig. 2) and TNF-α (P<0.01; Fig. 2) was significantly decreased in the I/R+FICZ group when compared with the I/R group.

FICZ reduces the percentage of activated CD4⁺ and CD8⁺ IEL subpopulations. CD69, an early T cell activation marker, is expressed in most IELs, indicating the intensity of inflammatory responses. In the present study, CD69 was used to assess CD4⁺ and CD8⁺ IEL subsets. Compared with the sham group, CD4⁺CD69⁺ and CD8⁺CD69⁺ IEL subpopulations were significantly increased in the I/R group (55.30±1.92% vs. 49.40±2.22%, P<0.05; 85.60±3.22% vs. 77.40±2.00%, P<0.05; Fig. 3). Notably, the I/R+FICZ group had a decreased population of IELs compared with the I/R group (43.60±2.90% vs. 55.30±1.92%, P<0.01; 72.10±2.86% vs. 85.60±3.22%, P<0.01; Fig. 3).

Reductions in intestinal inflammation suggested that FICZ may alter IEL phenotypes. Thus, a series of phenotypic analyses were performed by flow cytometry. The T-cell receptor (TCR) subpopulations were examined. Compared with the sham group, intestinal I/R treatment resulted in lower percentages of small intestinal IEL TCRγδ⁺/CD45⁺ T cells (29.20±2.43% vs. 51.30±1.87%, P<0.001; Fig. 4A) and CD127⁺/TCRγδ⁺ T cells (2.31±0.26% vs. 4.65±0.32%, P<0.05; Fig. 4B). The percentage of TCRγδ⁺/CD45⁺ T cells (40.80±.72% vs. 29.20±2.43%, P<0.01; Fig. 4A) and CD127⁺/TCRγδ⁺ T cells (4.82±0.50% vs. 2.31±0.26%, P<0.01; Fig. 4B) among small intestinal IELs was significantly higher in the I/R+FICZ group than the I/R group. Considering the decreased percentage of TCRγδ cells, the survival of γδIELs was subsequently investigated by assaying I/R apoptosis using flow cytometry. There were significant changes in the apoptosis of these cells. Compared with the sham group, the apoptosis rate (early and late) of γδIELs was significantly increased after I/R (3.26±0.51% vs. 0.78±0.23%, P<0.01; Fig. 4C). FICZ administration significantly attenuated apoptosis induced by intestinal I/R (1.53±0.33% vs. 3.26±0.51%, P<0.05; Fig. 4C).

The expression of CD127, which binds to IL-7, is regulated throughout T cell development, activation, and survival, and CD127 signals are thus required for γδ T cell development and survival. Since the proportion of TCRγδ⁺/CD45⁺ T cells was decreased after I/R, it was determined whether or not the expression of CD127 mRNA in small intestinal γδIELs was altered. CD127 gene expression was significantly lower in the I/R group than that noted in the sham group (P<0.05; Fig. 5B). Bcl2 and Bcl2l1 expression is maintained by IL-7R signaling. Thus, we detected the expression of Bcl2 and Bcl2l1 mRNA in small intestinal γδIELs. The expression of Bcl2 and Bcl2l1 was significantly decreased after I/R (P<0.05; Fig. 5B). KGF, which plays a critical role in intestinal epithelial growth and maintenance, is only expressed by γδIELs. Intestinal I/R resulted in a significant decrease in the expression of KGF mRNA (P<0.05; Fig. 5A). As a standard indicator of AhR activation, the expression of CYP1A1 mRNA in γδIELs was significantly decreased after I/R when compared with the sham group (P<0.01; Fig. 5A). Interestingly, the I/R+FICZ group had significantly higher CYP1A1expression when compared with the I/R group (P<0.01; Fig. 5A). Bcl2 and Bcl2l1 gene expression in the I/R+FICZ group increased significantly compared with the I/R group (P<0.05; Fig. 5B). Moreover, compared with the I/R group, CD127 expression was significantly increased after FICZ treatment (P<0.01; Fig. 5B). As expected, KGF gene expression was increased after FICZ administration compared with I/R alone (P<0.01; Fig. 5A).