Unconjugated bilirubin (bilirubin IXα) was obtained from Sigma-Aldrich (St. Louis, MO, United States), and the stock solution was freshly prepared in 0.1 M potassium phosphate (pH 12) as previously described (Huang et al., 2017). DSS (36–50 kDa) was obtained from MP Biomedical (Solon, OH, United States).

Eighty-nine UC patients that were evaluated at the Department of Gastroenterology and Gastrointestinal Surgery, Guangzhou First People’s Hospital, between January 2017 and December 2019 were recruited. UC was diagnosed in accordance with the guidelines of the European Crohn’s and Colitis Organization (ECCO). Patients with Gilbert syndrome and primary sclerosing cholangitis, or any other autoimmune disease that might influence serum total bilirubin (sTB) levels, were excluded from the study. Healthy subjects who visited the hospital for routine physical examination during the same period were included in the control group. The clinical and laboratory data were retrieved from the medical record systems, and sTB levels were measured by standard methods. The study was approved by the Institutional Review Board and Ethics Committee, and all participants signed the informed consent form.

Male C57BL/6 mice aged 8–12 weeks (weight approximately 25 g) were purchased from the Guangdong Laboratory Animal Monitoring Institute, and housed under 12 h light/dark cycles with ad libitum access to food and water. Colitis was induced in the adult mice by administering 3% DSS (w/v) in the drinking water. The control group received filtered water without DSS. Twenty-four hours before commencing DSS treatment, the animals were given daily intraperitoneal (ip) injections of bilirubin (30 mg/kg) or an equivalent volume of the potassium phosphate for indicated days. All animal experiments were approved by the Institutional Animal Care and Use Committee of South China University of Technology.

The mice were weighed and examined for signs of colitis on a daily basis. The disease activity index (DAI) was graded as follows: a) Body weight loss: 0—none, 1—1–5%, 2—5–10%, 3—10–15%, 4—>15%; b) stool consistency: 0—normal, 2—loose stools, 4—diarrhea; c) blood in stool: 0—negative, 2—positive, 4—gross rectal bleeding. The mean scores were calculated and recorded as the DAI. After the 7-day treatment, the animals were euthanized by CO₂ inhalation, and blood was immediately drawn by cardiac puncture. The small and large intestines were resected, and the colon length was measured.

Longitudinal sections of proximal and distal colon were fixed in 4% paraformaldehyde and stained with hematoxylin and eosin as per standard protocols. The specimens were analyzed independently by three investigators blinded to the grouping. At least 10 low-power non-overlapping fields were examined per section, and histologically scored on the basis of the following parameters: a) severity of inflammation: 0—no inflammation. 1—mild, 2—moderate and 3—severe; b) depth of inflammatory involvement: 0 - no inflammation, 1—mucosa, 2—mucosa and submucosa, and 3—transmural; c) crypt damage: 0—intact crypts, 1—loss of the basal one third, 2—loss of the basal two thirds, 3—entire crypt loss but intact epithelial surface, and 4—entire crypt loss and erosion of epithelial surface. The scores of all three parameters were added, and the maximum score was 40.

The colon pieces were homogenized in lysis buffer, and then centrifuged at 12,000 × g at 4°C for 30 min. The protein content in the supernatant was determined using bicinchoninic acid (BCA)™ protein assay kit (Thermo, MA, United States). IL-1β, IL-6, TNF-α, IFN-γ and MPO levels were measured by ELISA using specific kits (Abcom; Cambridge, United States) as per the manufacturer’s instructions, and expressed as pg/mg or U/g (MPO).

Colonic tissues were homogenized in ice-cold lysis buffer, and then centrifuged at 12,000 × g at 4°C for 30 min. Levels of malondialdehyde (MDA) in colon tissues were measured by the kits according to the manufacturer’s instructions from Beyotime Institute of Biotechnology (Haimen, China). The total protein content was determined by a BCA™ protein kit (Thermo, MA, United States).

ROS levels were measured using the fluorescent 2, 7-dichlorofluorescein-diacetate probe (DCFH-DA, Beyotime Institute of Biotechnology, China). The colon specimens were homogenized with PBS into a single-cell suspension. The cells were harvested and incubated with DCFH-DA for 30 min at 37°C in the dark. The samples were acquired by flow cytometry, and the fluorescence intensity was measured.

Colonic tissues were homogenized in cold PBS. The antioxidant activity in the supernatant was evaluated using a specific kit from Beyotime Institute of Biotechnology (Haimen, China) according to the manufacturer’s instructions.

The frozen colon tissues were embedded in OCT and cut into 8 μm sections. After staining with 30 μM DHE (Invitrogen Molecular Probes) for 10 min at 37°C, the sections were counterstained with Vectashield containing DAPI (Vector Laboratories, Inc.). The slides were observed with a Zeiss 510 upright confocal microscope, and the fluorescence intensity of at least 100 nuclei per sample was scored in at least three mice.

The bilirubin levels in serum and colon homogenates were measured by ion-pair extraction using 0.1 mol/L methanolic di-n-octylamine acetate. Briefly, the supernatants were injected into the liquid chromatography mass spectrometer (LC-MS/MS) system was equipped with an Agilent 1200 liquid chromatograph and a 6410B triple quadrupole mass spectrometer with an ESI source. Data were analyzed using the MassHunter software (Agilent Corporation, Lexington, MA, United States). Chromatography was performed on a Poroshell 120 EC-C18 2.1 μm 30 × 50 mm column (Agilent).

Colonic tissues were collected immediately and homogenized in ice-cold PBS, and centrifugation at 12,000 × g at 4°C for 30 min and kept at −80°C until they were assayed. Superoxide anion (O₂ ^·−) production was detected using a LumiMax Superoxide Anion Detection kit (Agilent Technologies, La Jolla, CA, United States) according to the manufacturer’s protocol. Briefly, 50 μg colon proteins were suspended in 100 μl assay medium and then mixed with 100 μl detection reagent containing 0.2 mM luminol and 0.25 mM enhancer. The luminescence was measured using a luminometer, and the O₂ ^·− content was expressed as relative light units (RLU)/μg protein/min. The total protein content was determined by a BCA™ protein kit (Thermo, MA, United States).

The colon tissues were fixed in 4% paraformaldehyde, embedded in paraffin and sectioned as previously described (Lu et al., 2011). The sections were immunostained using MaxVision kit (Maixin Biol, Fuzhou, Fujian, China) according to the manufacturer’s instructions. The primary bilirubin was diluted 1:100 in blocking solution. Color was developed using 0.05% diaminobenzidine and 0.03% H₂O₂ in 50 mM Tris–HCl (pH 7.6), and the sections were counterstained with hematoxylin. Pre-immune rabbit serum was used as the negative control. The anti-bilirubin antibody was purchased from LifeSpan BioSciences (Seattle, WA, United States).

All statistical analyses were performed using GraphPad Prism 5 (GraphPad Software, United States) and SPSS version 21.0 (IBM SPSS Statistics, United States). For the clinical study, non-normally distributed variables are expressed as median (interquartile range) and categorical variables as absolute numbers. For the animal study, the data are presented as mean ± standard error of the mean (SEM). Differences between groups were determined using one-way analysis of variance followed by Tamhane’s multiple comparisons post-hoc tests using SPSS. p < 0.05 was considered statistically significant.

A total of 89 UC patients (average age - 43.37 years, male/female ratio - 48/) and 95 healthy controls (average age - 38.76 years, male/female ratio - 50/39) were included in this study. The electronic medical records of the case and control populations were screened for total serum bilirubin levels. As shown in Table 1, the UC patients had significantly lower median serum bilirubin levels compared to the controls (9.8 vs. 12.5 μM, p < 0.0001). When stratified by age and sex, the reduction in serum bilirubin levels was more pronounced in males compared to females, and in the middle-aged patients (30–50 years old).

Given the reduced bilirubin levels in UC patients, we assessed the potential therapeutic effects of exogenous UCB in a mouse model of DSS-induced colitis. The animals received UCB for 24 h prior to colitis induction (Figure 1A). As shown in Figures 1B–D, DSS increased DAI, and reduced the colon length and body weight. In contrast, pre-treatment with UCB improved the DAI, and restored colon length and body weight compared to the vehicle-treated DSS mice. Consistent with the clinical findings, UCB also reduced the infiltration of inflammatory cells in the colon mucosa, and alleviated the mucosal/epithelial damage as per the histological scores (Figure 1E), although the effect was less pronounced in the distal compared to the proximal colon (Figure 1F). In addition, UCB had no impact on any of the above parameters in healthy mice. Taken together, UCB alleviated DSS-induced inflammation and tissue damage.

The severity of acute colitis often correlates with increased levels of pro-inflammatory cytokines. The mice pre-treated with UCB had significantly lower levels of TNF-α, IL-6, INF-γ and IL-1β in the serum as well as colonic tissues compared to the PBS-treated group (Figures 2A,B). In addition, the colonic myeloperoxidase (MPO) activity, a measure of neutrophil infiltration, was considerably increased in the untreated model group and decreased markedly in the mice pre-treated with UCB (Figure 2C). Finally, high levels of MDA were detected in the colon of untreated mice with colitis (Figure 2D), indicating significant oxidative injury. Pre-treatment with UCB reduced the MDA content to near baseline levels. Taken together, UCB has a potent anti-inflammatory effect in colitis.

The gut-liver reabsorption of UCB is essential for normal gut function, and is a potential factor in the development of intestinal diseases. UCB is the β-glucuronidase hydrolysis product of bilirubin glucuronides, and can be absorbed passively from both the large and small intestines. In the human digestive system, the cecum and proximal colon are more acidic compared to the other parts of the intestinal bowel, and therefore more favorable sites for the absorption of diacidic UCB. To determine whether reabsorption of UCB plays a role in DSS-induced colitis, we assessed its localization in situ. Bilirubin was mainly localized in the proximal colon tissues of mice pre-treated with UCB, with considerable enrichment in the intestinal epithelial cells (Figure 3A). UCB levels were also measured using LC-MS/MS, which showed that mice pre-treated with UCB had respectively 2-fold and 1.5-fold higher UCB in their proximal colon tissues and serum compared to the vehicle-treated controls (Figures 3B,C). Taken together, UCB supplementation increased its reabsorption from the colonic contents.

Overproduction of ROS is involved in the progression and pathogenesis of various acute and chronic inflammatory diseases. There is evidence suggesting that bilirubin has antioxidant activity, especially against O₂ ^·−. Therefore, we also evaluated ROS levels in colon tissues using the fluorescent DCFH-DA probe. UCB administration decreased ROS generation in the colon tissues of DSS-treated mice (Figure 4A). In addition, UCB also restored TAC levels that was markedly reduced in the DSS-treated mice to baseline (Figure 4B). The O₂ ^·− level in colon tissues was next measured by DHE staining, which revealed increased O₂ ^·− generation in the vehicle group which declined upon UCB pre-treatment (Figure 4C). Likewise, the specific detection kit also indicated a significant accumulation of O₂ ^·− in the colitis model, which was markedly reduced by UCB pre-treatment (Figure 4D). To summarize, exogenous UCB alleviated oxidative stress in the colitis model by directly scavenging O₂ ^·− from the inflamed colon tissues.