Introduction

The incidence of inflammatory bowel diseases (IBDs), including both Crohn’s disease (CD) and ulcerative colitis (UC), is relatively high in North America and Europe, and its morbidity in China is increasing.1,2,3,4,5,6 IBD is of pediatric interest because it is an important reason for failure to thrive and decline in the quality of life.7,8,9 The etiology and pathogenesis of IBD are not clearly understood but depend on the interaction of genetic, environmental, and immunological factors leading to intestinal epithelial barrier dysfunction.10,11,12 The diagnosis of IBD is challenging and requires laboratory testing, imaging, endoscopy, and pathological evaluation in addition to evaluation of clinical manifestations.13,14 The goal of treatment is mucosal healing rather than achieving and maintaining clinical remission.15,16 Patients without mucosal healing have an increased probability of recurrence, and about 30% of patients ultimately require surgery.17,18 Endoscopy provides an objective assessment of IBD status, but it is invasive, with heavy economic and psychological burden if it needs to be repeated frequently. It is also not free of adverse events. The risk of anesthesia is higher in children than in adults. The intestinal wall of children is thin and fragile and the lumen is narrow, both of which increase the risk of severe complications. Novel reliable and noninvasive markers are urgently needed. Recently, fecal calprotectin (FC) has been used to screen IBD.19 Calprotectin belongs to the S100 family and is derived predominantly from neutrophils. Thus, the presence of elevated FC is an indicator of mucosal neutrophil infiltration.20 In addition, various factors may affect FC results, including age, medication and day-to-day variation.21 Finding a specific biomarker expressed in the diseased intestine may have more diagnostic value. Trefoil factor peptide 3 (TFF3) is a peptide specifically secreted by intestinal goblet cells. It promotes cell migration and proliferation, repair of mucosal epithelium, and angiogenesis, which act to maintain the intestinal epithelial barrier and normal intestinal function.22 Both in vitro and animal studies indicate that TFF3 can maintain the integrity of the intestinal epithelium and regulate mucosal inflammation.23,24,25 TFF3 may have therapeutic potential for IBD, and serum TFF3 levels have been investigated as a biomarker of mucosal healing in adults with UC.26,27 As studies in children are lacking, investigations of TFF3 as a pediatric biomarker are warranted.

Materials and methods

Animal model of IBD

The increase in serum TFF3 following damage to the intestinal mucosal barrier was investigated in a TNBSA-induced mouse IBD model, which is widely used to replicate impairment of the intestinal mucosal barrier. Intestinal permeability was determined by measuring the fluorescence of isothiocyanate-conjugated dextran (FD-4) following gavage into the lumen of the intestine. Serum TFF3 concentration was also assayed.

Procedures for care and handling of animals used in this study were approved by the ethics committee of China Medical University. Twenty female BALB/C mice (SCXK shen 2015-0001) between 5 and 8 weeks of age were used. The mice were randomly assigned in equal numbers to an IBD group given a single intracolonic dose of 0.2 mL of 50% ethanol (vol/vol) containing 25 mg/mL TNBSA (Sigma-Aldrich) or a control group given the same volume of normal saline instead of TNBSA in ethanol.28 On day 7 after induction of colitis, 0.5 mL FD-4 (molecular weight 4000, 40 mg/mL) was gavaged into the lumen of the intestine using a gavage needle. The mice were killed after 30 min, a 100 µL blood sample was collected, mixed with 1.9 mL 50 mol/L Tris solution and centrifuged at 4000 × g for 10 min at 4 °C. The FD-4 that entered the blood from the intestine was measured spectrofluorometrically with an excitation wavelength of 480 nm and emission wavelength of 520 nm using a microplate fluorescence reader (FLX-800; BioTek Instruments, Winooski, VT, USA). FD-4 permeability was reported as µg/mL. Tissue samples were collected for assays and serum TFF3 concentration was measured with a commercial sandwich enzyme-linked immunosorbent assay (Mouse TTF3 ELISA, Boster Biological) kit, following the manufacturer’s instructions.

Pediatric IBD patients

The study cohort included 51 children, 28 boys and 23 girls from 5 to 17 years of age with IBD and treated at Shengjing Hospital of China Medical University between March 2017 and September 2018. Twenty healthy children were selected during the study period as the normal control group. All children were diagnosed at our hospital and all met the 2010 diagnostic criteria established by the Digestive Science Group, Pediatrics Branch, of the Chinese Medical Association.29 The growth and clinical characteristics, laboratory investigations, endoscopic examination, histological features, therapy and outcome were reviewed.

Blood samples were collected the morning after admission for assay of serum TFF3 using a commercial sandwich ELISA kit (Human TTF3 UCSN Life Science İnc. USA). Clinical disease activity of UC was described with the Pediatric Ulcerative Colitis Activity Index (PUCAI).30 The index is scored from 0 to 85, with clinical remission as <10, mild activity as 10–34, moderate activity as 35–64, and severe activity as >65. The clinical disease activity of CD was described by the pediatric Crohn’s Disease Activity Index (PCDAI); 0–10 indicated inactive disease, 10–30 indicated mild disease, and >30 indicated moderate-to-severe disease.31 In patients with luminal Crohn’s disease, activity was evaluated endoscopically using the simple endoscopic score for Crohn’s Disease (SES-CD); 0 or 1 indicated mucosal healing.32 Mucosal changes associated with UC were evaluated by Baron’s scoring system.33 Grade 0 or 1 changes indicated mucosal healing.

Ethical approval and consent to participate

The study protocol was approved by the Ethics Committee of Shengjing Hospital of China Medical University (approval no. 2015PS281K Shenyang China). Informed consent was obtained from each patient following the institutional guidelines.

Statistical analysis

Statistical analysis was performed with SPSS 22.0 (IBM Corp., Armonk, NY, USA). Measurement data were reported as means ± standard deviation. Differences in patient group means were compared with the Mann−Whitney U test. One-way analysis of variance and independent t tests were used to compare differences between the mouse groups. Correlations were tested for significance by calculating the Spearman’s coefficient. Receiver operating characteristic (ROC) curves were derived from the original data and the maximum value of the Jordan index (sensitivity + specificity – 1) was used as the diagnostic cutoff value. P values < 0.05 was considered statistically significant.

Results

Animal model of IBD

Mice in the TNBSA group had diarrhea and severe weight loss. Histological manifestations included transmural ulceration with complete loss of epithelium and extensive lymphocyte, neutrophil, monocyte, and eosinophil infiltration of the mucosa (Fig. 1).

Fig. 1: Histopathology of the colon.
figure 1

Histopathology of the colon mucosa showing extensive infiltration of the lamina propria between intestinal glands by lymphocyte and neutrophils (100×).

Full size image

Serum TFF3 was significantly higher in the TNBSA-treated than the control mice (P < 0.05). Serum FD-4 fluorescence was significantly higher in the TNBSA-treated than the control mice (P < 0.01, Table 1).

Table 1 Serum TFF3 in TNBS-treated and control mice.
Full size table

IBD patients

In this cohort study, 33 of the 51 IBD patients (64.7%) had CD; 18 (35.3%) had UC; 24 (47%) had active IBD, and 27 (53%) were in remission. The PCDAI scores indicated that 15 CD patients had active disease and 18 were in the remission stage. The PUCAI scores indicated active disease and remission in nine patients each. There were no significant differences in age or sex in the healthy control and IBD groups (Table 2). There were no significant differences in serum TFF3 in CD and UC patients. Serum TFF3 was significantly higher in patients with active IBD than in both the controls and the IBD patients in remission. Serum CRP was significantly higher in patients with active CD than in both the controls and remission group (P < 0.01, Table 3).

Table 2 Characteristics of patients.
Full size table
Table 3 Serum TFF3 and CRP in patients with Crohn’s disease, ulcerative colitis, and healthy children.
Full size table

Correlation of serum TFF3 and CRP with PCDAI and SES-CD scores in CD patients

Serum TFF3 was positively correlated with the SES-CD score (r = 0.659, P = 0.007), but not with PCDAI (r = 0.472, P = 0.076). CRP was not significantly correlated with the SES-CD score (r = 0.412, P = 0.082). ROC curve analysis yielded an area under the ROC curve of 0.926 for serum TFF3. The sensitivity of TFF3 to detect active IBD in this pediatric sample was 100% and the specificity was 76.2% (Fig. 2 and Table 4). The TFF3 concentration reflected the extent of mucosal healing in the CD patients.

Fig. 2: ROC curve analysis.
figure 2

ROC curve analysis of TFF3 for the evaluation of CD activity.

Full size image
Table 4 Ability of serum TFF3 to detect CD activity.
Full size table

Correlation of serum TFF3 and CRP with PUCAI and Baron’s scores in UC patients

Serum TFF3 was positively correlated with PUCAI (r = 0.994, P = 0.000) but not with serum CRP (r = 0.419, P = 0.301), or Baron’s score (r = 0.478, P = 0.230).

Discussion

In this study, serum TFF3 was assayed in both IBD model mice and pediatric IBD patients. Serum TFF3 was significantly higher in model mice with high epithelial permeability. Serum TFF3 levels were also elevated in patients in the active disease stage and in those who had not achieved mucosal healing. We propose TFF3 as a new biomarker in pediatric IBD.

In this study, the colon mucosa was destroyed in the IBD model mice. The significant increase of serum FD-4 fluorescence confirmed that the intestinal mucosa had been damaged and the TFF3 concentration in peripheral blood significantly increased in parallel with damage to the intestinal barrier. TFF3 is secreted by intestinal goblet cells and is normally present in very small amounts in the blood. Can large amounts of TFF3 come from the damaged intestinal mucosal? We believe that in the IBD model mice, the goblet cells secreted large amounts of TFF3 that entered the blood through the damaged intestinal mucosa. That can be confirmed by following the movement of tracer-labeled TFF3 from damaged intestinal segments into the peripheral blood, which is planned.

An increase in TFF3 in the peripheral blood of pediatric IBD patients that exceeds a threshold level can be considered as a marker of disease activity or recurrence. This hypothesis and potential clinical application of serum TFF3 was tested in a cohort of pediatric IBD patients. Serum TFF3 was significantly higher in patients with active CD and UC disease than those in the remission stage and in the healthy control group. Serum TFF3 was positively correlated with the SES-CD score. Its sensitivity for monitoring CD activity was 100% and its specificity was 76.2%. The predicted cutoff value of serum TFF3 was 6.01 ng/mL. Serum TFF3 level may thus reflect the extent of mucosal healing or damage and serve as a biomarker of mucosal healing in children with CD. A previous study in adults reported a correlation of serum TFF3 and Baron’s score in UC patients, but that was not observed in this group of pediatric patients.24 Possible reasons for the difference are the small number of included UC patients in this study and differences unknown between adult and pediatric patients.

Performing only one serum TFF3 assay was a study limitation. Serial measurements would have allowed monitoring changes that occurred during the progression of disease and possibly determining whether patients with low serum TFF3 at onset or remission had a better prognosis than those with a high initial TFF3 concentration. Additional clinical studies in larger groups of patients are desirable to evaluate the significance of TFF3 as a biomarker in PIBD.

Conclusion

Serum TFF3 level significantly increased with CD activity, which is of significance for diagnosis and evaluation of mucosal healing. TFF3 could also be a marker in pediatric UC with TFF3 positively correlated with UCAI.