Naringin (purity ≥ 98%) was purchased from Beijing Solarbio Co. Ltd. (Beijing, China). Sulfasalazine (SASP) was purchased from Tian-jin Kingyork Group Co. Ltd. (Tianjin, China). Antibodies against PPAR-γ (A0270) were from ABclonal Co. Ltd. (Wuhan, China). Antibodies against caspase-3 (19677-1-AP), cleaved caspase3 (cl-caspase3) (19677-1-AP), and iNOS (14142-1-AP) were from Proteintech Co. Ltd. (Wuhan, China). Antibodies against NF-κB-p65 (66535-1-Ig), phosphorylated NF-κB-p65 (p-NF-κB-p65) (WL02169) and PPAR-α (WL00978) were from Wanleibio Co. Ltd. (Shenyang, China). A cell counting kit-8 (WLA074b) was from Wanleibio Co. Ltd. BADGE (B6691) was obtained from Apexbio (Houston, TX, United States). Chemicals were obtained from Sigma–Aldrich (St. Louis, MO, United States), unless otherwise indicated.

Eighty male C57BL/6 mice (6–8 weeks old, weighing 18–20 g) were obtained from the laboratory animal center, Dalian Medical University, Dalian city, China [Certificate of Conformity: No. SYXK (Liao) 2018–0007]. The experimental protocol was approved by Dalian Medical University Animal Care and Ethics Committee (No. AEE20046). The animals were acclimatized to laboratory conditions (23°C, 12 h/12 h light/dark, 50% humidity, ad libitum access to food, and water) for 2 weeks prior to experiments. The animal protocol was designed to minimize pain and discomfort to the animals. Mice were housed one per cage and were deprived of food for 12 h before experiments. All mice were euthanized by barbiturate overdose (intravenous injection, 150 mg/kg pentobarbital sodium) for intestinal tissue collection. All animal experiments were carried out in accordance with the National Institutes of Health guide for the care and use of laboratory animals. Animal studies are reported in compliance with the ARRIVE guidelines (Kilkenny et al., 2010). All applicable institutional and/or national guidelines for the care and use of animals were followed.

Murine RAW264.7 cells and rat intestinal IEC-6 epithelial cells were purchased from the cell bank of the Shanghai Institute (Shanghai, China). Cells used in this study were evaluated before conducting experiments, and no significant interspecies variations in PPAR-γ signaling that might have affected the results of the current study were observed in preliminary tests. The cells were maintained at 37°C in a 5% CO₂ environment. The RAW264.7 cells were cultured in RPMI-1640 medium with 10% top fetal bovine serum. Rat intestinal IEC-6 epithelial cells were cultured in DMEM (Invitrogen, Waltham, MA, United States) medium with 4.5 mg/ml glucose, 50 U/mL penicillin, 50 U/mL streptomycin, 4 mM glutamine, 25 mM HEPES, and 10% fetal bovine serum (Invitrogen) (Sui et al., 2020). RAW264.7 and IEC-6 cells were transfected using Lipofectamine 2000 (Invitrogen) with PPAR-γ-targeted or control small interfering RNA (siRNA) oligos (GenePharma, Suzhou, China), according to the manufacturer’s instructions (Takara Biotechnology (Dalian) Co., Ltd.). The siRNA sequences for PPAR-γ interfering RAW264.7 cells and IEC-6 cells, respectively were:

Sense: 5′-GGU​GCU​AAG​AGA​UUG​CCU​UTT-3′;

For the DSS–colitis model, mice were divided randomly into five groups of six mice each. The mice were treated as follows: group I was a sham-operated control with gavage administration of saline; group II was untreated colitis; and groups III, IV, and V were treated with the following: SASP (500 mg/kg body weight, gavage administration, dissolved in saline), low-dose naringin (20 mg/kg body weight, gavage administration, dissolved in saline), and high-dose naringin (40 mg/kg body weight, gavage administration, dissolved in saline), respectively. All treatments occurred 1 day after colitis induction with DSS. SASP is an anti-inflammatory drug that is, widely used for the clinical treatment of diseases such as IBD and, therefore, it was used as a positive control for the effects of naringin on colitis. Mice in group III, IV, and V were administered SASP or naringin by gavage once a day for seven successive days. Mice in groups II–V were given drinking water containing DSS (4% w/v) dissolved in autoclaved distilled water, to induce colitis symptoms (Feng et al., 2020). Control group I mice were given autoclaved distilled water and otherwise, treated in the same way. No abnormal moribund mice were found during the study. Food intake and bodyweight of mice were recorded once a day. On the 8th day, mice in group I–V were given autoclaved distilled water as drinking water. On the 9th day, distal colon samples in groups I–V were harvested for biochemical studies. Mice in the PPAR-γ inhibition group were injected intraperitoneally with 30 mg/kg BADGE solution per day for seven consecutive days, the experimental designs are the same as above.

For the TNBS–colitis model study, mice were divided randomly into five groups of six mice each. The mice were treated as follows: group I was a sham-operated control with gavage administration of saline; group II was untreated colitis; and groups III, IV, and V were treated with the following: SASP (500 mg/kg body weight, gavage administration, dissolved in saline), low-dose naringin (20 mg/kg body weight, gavage administration, dissolved in saline), and high-dose naringin (40 mg/kg body weight, gavage administration, dissolved in saline), respectively. All treatments occurred 1 day after colitis induction. Mice in group III, IV, and V were administered SASP (Xiong et al., 2017), naringin by gavage once a day for seven successive days. Colitis was induced with TNBS, as described previously (Dong et al., 2020). A catheter was inserted through the anus to approximately the level of the splenic flexure under urethane anesthesia. The colon was then infused with 0.1 ml of TNBS dissolved in ethanol (50% v/v) at a dose of 125 mg/kg. The mice could eat and drink ad libitum starting 1 h after the operation. On the 8th day, mice in groups I–V were given autoclaved distilled water as drinking water. On the 9th day, distal colon samples in groups I–V were harvested for biochemical studies. Protein extraction, western blotting, and ELISA were performed as previously described (Dong et al., 2020).

Mice body weights were monitored daily. The DAI scores were assessed according to our previous study (Melgar et al., 2005). Briefly, an observer blinded to the treatments combined the scores for weight loss, stool bleeding, and stool consistency on the last day.

Colonic epithelial cells were isolated according to a previous study (Dong et al., 2020). Briefly, distal colons were isolated from euthanized mice and immediately rinsed with ice-cold phosphate-buffered saline (PBS) to clear luminal contents. The distal colon was then opened longitudinally along the mesenteric border. Tissue was cut into approximately 2 mm long pieces and submerged in 40 ml of ice-cold PBS with 5 mM EDTA in a 50 ml Falcon tube. The pieces of tissue in PBS-EDTA were then incubated at 37°C with gentle rocking for 30 min. Following incubation, colonic tissue was vigorously shaken to disperse colonic crypts and surface epithelium in solution. Supernatant was then loaded into 1.5 ml microcentrifuge tubes and spun at 1 × 10³ g for 5 min to pellet suspended cells. The isolated epithelial cells were used in downstream applications.

Three naringin treated DSS-induced C57BL/6 mice, three DSS-induced C57BL/6 mice and three untreated mice were used for mRNA profile detection. Total RNA was extracted using TRIzol reagent (Invitrogen), and quality was evaluated using the RNA Nano 6000 Assay Kit of the Bioanalyzer 2,100 system (Agilent Technologies, CA, United States).

A total of 3 µg RNA per sample was used to generate an mRNA sequencing library using NEBNext^® Ultra™ RNA Library Prep Kit for Illumina^® (NEB, United States) following manufacturer’s recommendations and index codes were added to each sample for identification (Gao et al., 2012). Briefly, mRNA was purified and fragmented. Then, first and second strand cDNA was synthesized, ends were repaired and adenylated, adapters were ligated, and fragments were enriched with PCR amplification. Library quality was assessed on the Agilent Bioanalyzer 2,100 system. Libraries were sequenced on an Illumina platform.

Raw data (raw reads) of fastqc format were first processed through Trimmomatic (Bolger et al., 2014). In this step, clean data (clean reads) were obtained by removing reads containing adapter, reads containing poly-N, and low-quality reads from raw data. Reference genome and gene model annotation files were downloaded from the genome website directly. Index of the reference genome was built and paired-end clean reads were aligned to the reference genome using Hisat2 (v2.0.5), the selected mapping tool. The mapped reads of each sample were assembled by StringTie (v1.3.3b) in a reference-based approach. FeatureCounts v1.5.0-p3 was used to count the reads mapped to each gene. Then, FPKM of each gene was calculated based on the length of the gene and read counts mapped to this gene (Conesa et al., 2016). FPKM, expected number of Fragments Per Kilobase of transcript sequence per Million base pairs sequenced, considers the effect of sequencing depth and gene length for the read counts at the same time. Differential expression analysis of two conditions/groups (two biological replicates per condition) was performed using the DESeq2 R package (1.16.1). The resulting p-values were adjusted using the Benjamini and Hochberg approach. Genes with an adjusted p-value < 0.05 found by DESeq2 were assigned as differentially expressed. We used the clusterProfiler R package to test the statistical enrichment of differential expression genes in KEGG pathways. The raw sequencing data sets were deposited in the Sequence Read Archive of the NCBI (https://www.ncbi.nlm. nih.gov/sra) under accession number PRJNA741857.

Validation of differentially expressed genes was performed by qRT-PCR using SuperReal PreMix Plus (SYBR Green, FP205, Tiangen) with an Applied Biosystems StepOnePlus Real-Time PCR System, according to the manufacturer’s instructions. Total RNA was isolated from colonic epithelial cells using RNAiso Plus (9108, TaKaRa) and reverse-transcribed to single-stranded cDNA using a reverse transcription system (KR116, Tiangen). Primer sequences are listed in Table 1. A separate qRT-PCR using a primer for the detection of GAPDH was used as a control.

Colon tissues were fixed with 4% paraformaldehyde at 4°C. The samples were dehydrated, embedded in paraffin, and sectioned into 3-µm-thick transverse sections. For H and E staining, the sections were dewaxed, rehydrated, and stained with hematoxylin and eosin. After being washed with distilled water and dehydrated, the sections were treated with xylene. Colitis activity scores were assessed in a blinded fashion, by combining the scores of inflammation severity, inflammation extent, crypt damage, and percent involvement (Kihara et al., 2003). Inflammation severity score was assessed according to the following criteria: 0: none; 1: mild; 2: moderate; and 3: severe. Inflammation extent score was assessed according to the following: 0: none; 1: mucosa; 2: submucosa; and 3: transmural. Crypt damage score was assessed by the following criteria: 0: none; 1: basal 1/3 damage; 2: basal 2/3 damage; 3: crypt lost; surface epithelium present; and 4: crypt and surface epithelium lost. Percent involvement score was determined as follows: 0: 0%; 1: 1–25%; 2: 26–50%; 3: 51–75%; and 4: 76–100%. For IHC staining, the sections were dewaxed, rehydrated, and washed with distilled water three times. The sections were then treated with 3% H₂O₂ for 10 min, washed in PBS containing Tween-20 (PBST), treated with 5% bovine serum albumin (BSA) (Solarbio, A8020) and then incubated with rabbit anti-PPAR-γ (ABclonal, A0270) (1:100 dilution) overnight at 4°C. After three washes with PBST, the sections were incubated with secondary antibody for 30 min at 37°C. After being rinsed, the sections were incubated with diaminobenzidine. After counterstaining with hematoxylin, washing with distilled water, and dehydration, the sections were treated with xylene. For IF staining, sections were dewaxed, dehydrated, and subjected to antigen retrieval. 3% BSA (Solarbio, A8020) was utilized to block nonspecific binding. Then, the sections were incubated with rabbit anti-PPAR-γ (ABclonal, A0270) (1:100 dilution) overnight at 4°C. Later, these sections were rinsed with PBS and incubated with Alexa Fluor 488 (FITC) secondary antibody (Proteintech, SA00013-2) in the dark for 60 min at room temperature. 4,6-diamidino-2-phenylindole (Invitrogen, P36941) was later applied to dye the nuclei. Fluorescence photographs were obtained under a fluorescence microscope DM5000 B (Leica, Germany).

The animal experiments, in vitro experiments, and data analysis were conducted according to a single-blind study design. Data were compared between three or more groups using one-way ANOVA, and between two groups using Student’s t-test. Data were expressed as the mean ± standard deviation. Data were normally distributed, and each group showed similar variances. Further evaluation was carried out using Kruskal–Wallis rank sum tests. All experiments were repeated at least three times and a p value < 0.05 was considered statistically significant.

The chemical structure of naringin is shown in Figure 1A. Naringin had no significant effect on body weight or food intake in normal mice (Figures 1B,C). Cytokine profiles [TNF-α, interleukins (IL)-1β, and interferon (INF)-γ] in colon tissue were not affected by naringin (Figures 1D–F). There was no significant effect on expression of PPAR-γ in colon tissue after gavage administration of 10 and 40 mg/kg naringin for 7 days, while the expression of PPAR-γ was decreased by 160 mg/kg naringin (Figure 1G). Naringin (5–160 μM, 12 h) did not significantly affect cell viability in RAW264.7 cells. Naringin (5–80 μM, 12 h) exerted no significant effects on cell viability in IEC6 cells, while 160 μM significantly inhibited cell viability (Figure 1H). According to our preliminary experiments and previous reports (Ahmad et al., 2015; Cao et al., 2018), the maximum doses of naringin were 40 mg/kg in vivo and 20 μM in vitro.

The significant weight loss, high colon weight-to-length ratio, decreased food intake and high DAI scores in the untreated model group showed that instillation of DSS to the colon resulted in reproducible colitis in mice (Figures 2A–D). The overall morphology of the colon is shown in Figure 2E. Histopathological damage was characteristically induced by DSS (Figures 2F,G). A variety of colitis symptoms, including damaged villi structure (green ring), infiltration of inflammatory cells (black arrow), submucosal edema (blue ring), and muscle fiber separation (blue arrow) were visible in the H and E staining (Figure 2G). Naringin and SASP reversed the pathological changes in colitis.

As shown in Figures 3A–D, MPO activity and expression of pro-inflammatory factors (TNF-α, IL-1β, and INF-γ) were significantly increased in DSS-colitis, and naringin significantly reversed these changes. DSS-induced colitis up-regulated inflammation-related proteins (iNOS and p-NF-κB-p65) and an apoptosis-related protein cl-caspase3 expression. Finally, an attenuated trend was observed in groups treated with naringin (Figures 3E–G). These results confirmed that naringin has anti-inflammatory and anti-apoptotic activities in DSS-colitis.

We applied transcriptome sequencing to investigate the possible mechanisms of naringin in the treatment of colitis. A total of 753 differentially expressed mRNAs before and after naringin treatment were identified with fold changes > 2 and p < 0.05 and some representative genes were shown in heatmap (Figures 4A,B). In addition, compared with the control group, 834 differentially expressed mRNAs were observed in the model group (Figure 5A).

The KEGG enrichment results showed the top 20 enriched KEGG pathways in the model group vs. naringin treatment group (Figure 4C) and the control group vs. model group (Figure 5B). There are three common enriched KEGG pathways including the neuroactive ligand-receptor interaction, calcium signaling, and PPAR signaling among the top 20 enriched KEGG pathways between the model group vs. naringin treatment group and the control group vs. model group (Figure 4C, Figure 5B). Based on the genes contained in the top 11 pathways, we constructed a KEGG-target network of naringin-induced treatment including 133 targets (Figure 4D). Most targets were associated with different pathways, suggesting synergistic effects are involved in naringin-induced treatment. Therefore, naringin may target different pathways and targets.

According to reference validation, we subsequently studied proteins in the PPAR signaling pathways to uncover naringin-induced treatment effects on colitis. We used qRT-PCR to detect the roles of some mRNAs in PPAR signaling pathways, including PPAR-α, PPAR-γ, PPAR-δ, Scd1, and Hmgcs2 in naringin-induced treatment effects (Figure 6). There were significant differences in the expressions of PPAR-α and PPAR-γ between the naringin treatment and DSS-colitis groups. The expressions of PPAR-δ, Scd1, and Hmgcs2 were not significantly affected by naringin in the colitis model.

We further studied the protein expressions of PPAR-α and PPAR-γ. Compared with the sham group, PPAR-γ expression was significantly decreased in DSS-colitis, but naringin treatment significantly stimulated PPAR-γ expression (Figures 7B–D). However, naringin treatment did not significantly affect PPAR-α protein expression (Figure 7A). The results suggest that PPAR-α and PPAR-γ may be involved in naringin-induced treatment effects, however, PPAR-γ plays a more important role.

Next, we examined the role of PPAR-γ for naringin-induced treatment effects. In vivo, the PPAR-γ inhibitor BADGE was used to investigate the role of PPAR-γ in naringin-induced improvement of colitis. As shown in Figure 8A, PPAR-γ expression was significantly inhibited by BADGE, and BADGE significantly exacerbated colitis symptoms including weight loss (Figure 8B), high DAI, increased colon weight-to-length ratio, and histological score (Figures 8C,D,F). The overall morphology of the colon is shown in Figure 8E. Every colitis symptom was reversed by naringin, but this affect was significantly inhibited by BADGE.

Both RAW264.7 and IEC-6 cells were used to confirm the stimulatory effects of naringin on PPAR-γ. In RAW264.7 cells, pro-inflammatory cytokines, including TNF-α, IL-1β, and INF-γ, were significantly increased by LPS, and the increase was reversed by naringin (Figures 9A–C). P-NF-κB-p65, a target gene of PPAR-γ, was also significantly increased by LPS (Figures 9D–F). The LPS induced increase in p-NF-κB-p65 was significantly reversed by naringin, but was significantly inhibited by siRNA targeting PPAR-γ. Similar results were also obtained using IEC-6 cells (Figures 9G–I). These results suggested that PPAR-γ is a potential target of naringin.

A TNBS-colitis model was also established to study the effects of naringin. Compared with the sham group, significant weight loss, high colon weight-to-length ratio, lower food intake, high DAI scores, and increased histological score were observed in the untreated colitis model group (Figures 10A–D,F). The overall morphology of the colon is shown in Figure 10E. Naringin significantly reversed these colitis symptoms. Higher levels of MPO activity and the expression of pro-inflammatory cytokines (IL-1β, INF-γ, and TNF-α) were also reversed by naringin treatment (Figures 10H–K). Naringin ameliorated the above colitis symptoms as evidenced by H and E staining (Figure 10G). Damaged villi structure (green ring), infiltration of inflammatory cells (black arrow), and submucosal edema (blue ring) shown in H and E staining were alleviated by naringin.

Naringin can reverse the increasing expression of inflammation-related proteins (iNOS and p-NF-κB-p65), and apoptosis-related protein cl-caspase3 in TNBS-colitis (Figures 11D–F). Furthermore, decreased PPAR-γ expression in TNBS-colitis was also stimulated by gastric administration of naringin (Figures 11A–C). In accordance with our DSS-colitis results, naringin also significantly alleviated TNBS-colitis.