Anti-integrin αvβ6 Antibodies Predict Pouchitis in Patients With Ulcerative Colitis After Restorative Proctocolectomy With Ileal Pouch-Anal Anastomosis

Abstract

Background

Pouchitis is the most common complication of restorative proctocolectomy (RPC) with ileal pouch-anal anastomosis (IPAA) in patients with ulcerative colitis (UC). We previously reported the presence of anti-integrin αvβ6 antibodies in the serum of patients with UC. This study investigated the association between anti-integrin αvβ6 antibodies and the development of pouchitis in patients with UC.

Methods

Serum levels of anti-integrin αvβ6 antibodies were measured by enzyme-linked immunosorbent assay in 16 patients with UC who underwent RPC with IPAA. Integrin αvβ6 expression in the colonic, terminal ileal, and pouch epithelium was examined using immunohistochemistry and western blot analysis.

Results

Anti-integrin αvβ6 antibody levels in patients with UC were significantly decreased at 3, 9, and 12 months after RPC (P < .05). However, in patients who developed pouchitis, antibody levels remained high. The antibody levels at the time of RPC were significantly higher in patients who developed pouchitis compared to those who did not. Kaplan-Meier analysis revealed a significantly higher incidence of pouchitis in patients with antibody levels above the cutoff at the time of RPC. Although integrin αvβ6 was not expressed in the terminal ileal epithelium at the time of RPC, expression became positive in the pouch epithelium of patients with pouchitis.

Conclusions

The anti-integrin αvβ6 antibody levels in patients with UC were decreased after RPC but remained high in patients who developed pouchitis. The antibody levels at the time of RPC may serve as a potential prognostic biomarker for predicting the risk of pouchitis in patients with UC.

Introduction

Ulcerative colitis (UC) is an inflammatory bowel disease characterized by chronic colonic mucosal inflammation. Despite medical advances, approximately 10%-30% of patients with UC eventually require surgical intervention, such as restorative proctocolectomy (RPC) with ileal pouch-anal anastomosis (IPAA), because of medically refractory disease or the development of colitis-associated cancer. Recent studies have reported that the cumulative colectomy rates in patients with UC are 7.0% and 9.6% at 5 and 10 years after diagnosis, respectively.¹ Although these surgeries alleviate various UC-related symptoms, patients often develop pouchitis as a postoperative complication.

Pouchitis, the most common complication of RPC with IPAA, is characterized by inflammation of the ileal pouch, resembling UC. The incidence of pouchitis in patients with UC is 48% within the first year after IPAA, and almost 80% throughout the follow-up period after IPAA.^2-4 Interestingly, the incidence of pouchitis in patients with familial adenomatous polyposis, another condition that can be treated with IPAA, is remarkably lower at only 6%.⁵ Therefore, immunological abnormalities may play important roles in the development of pouchitis in patients with UC. The endoscopic findings of pouchitis, such as mucosal redness, edema, erosions, and ulcers, resemble those observed in patients with UC. Moreover, treatments for pouchitis, including 5-aminosalicylic acid and corticosteroids, are similar to those for UC. Although the exact mechanism underlying pouchitis remains unknown, these similarities suggest a shared pathophysiology with UC.

We have previously identified anti-integrin αvβ6 antibodies in the serum of patients with UC.⁶ Integrin αvβ6, a member of the integrin family, is expressed in the colonic epithelium and plays a crucial role in maintaining the epithelial barrier function through its interaction with extracellular matrix proteins, such as fibronectin.^6-8 Inhibition of integrin αvβ6-fibronectin binding by anti-integrin αvβ6 antibodies may lead to epithelial barrier dysfunction and contribute to the development of UC.

Considering the similarities in clinical presentation and treatment response between UC and pouchitis, and the potential role of anti-integrin αvβ6 antibodies in the pathophysiology of UC, we hypothesized that anti-integrin αvβ6 antibodies might also be involved in the development of pouchitis. This study aimed to investigate the association between serum anti-integrin αvβ6 antibody levels and the development of pouchitis in patients with UC who underwent RPC with IPAA.

Methods

Patients

We enrolled all 16 patients with UC aged ≥18 years who underwent RPC with IPAA between September 2017 and January 2023 at Kyoto University Hospital (Table 1, Supplementary Table S1) and assessed the development of pouchitis until December 2023 (6 patients with pouchitis and 10 patients without pouchitis). Patients were censored at the time of last follow-up or the study end date (December 2023). Patients were diagnosed based on symptoms, endoscopic and histological findings, and the absence of alternative diagnoses.^9,10 Pouchitis was defined as a modified Pouchitis Disease Activity Index of ≥5.^11,12 Serum samples were collected from September 2017 to December 2023 and stored at −80 °C until assayed. This study was performed in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Kyoto University Graduate School and Faculty of Medicine (protocol number: R1004). Patients were informed through our hospital website so that they could opt out by contacting the research team at any time and also we directly explained this issue to the patients.

Enzyme-Linked Immunosorbent Assay

To examine the serum anti-integrin αvβ6 antibody levels, we used Anti-Integrin αvβ6 ELISA Kit (5288, Medical and Biological Laboratories, Tokyo, Japan) according to the manufacturer’s instructions, with minor modifications. The standard material (recombinant monoclonal human anti-integrin αvβ6 antibody) was serially diluted with the reaction buffer from 400.0 to 0.781 U/mL. The reaction buffer served as the zero standard. The αvβ6-coated microplates were incubated with 100 µL of diluted serum (1:100) from patients or standard material in each well for 60 minutes at room temperature. After washing with wash solution, the plates were incubated with 100 µL of horseradish peroxidase-conjugated antibody at room temperature for 60 minutes. After another washing with wash solution, the bound reactants were detected by incubation for 20 minutes with 100 µL of substrate reagent. The absorbance was measured at the dual wavelengths of 450 and 620 nm after adding 100 µL of stop solution. The antibody levels were determined using a calibration curve generated based on the optical density (OD) value of standard materials and blanks through a sigmoidal 4-parameter logistic equation. When the antibody levels were higher than 400 U/mL, we retested with further diluted serums to ensure that the antibody levels of the diluted serums were less than 400 U/mL.

Preparation of Human Immunoglobulin G

We used Ab-Rapid SPiN EX (P-014-10, ProteNova, Higashikagawa, Japan), according to the manufacturer’s instructions, to purify immunoglobulin (Ig) G from the serum of patients with UC. Purified IgG was stored at −20 °C.^6,13

Solid-Phase Integrin αvβ6 Binding Assay

A solid-phase integrin αvβ6 binding assay was performed according to the previously described method, with minor modifications, using ELISA Starter Accessory Kit (E101, Bethyl Laboratories, Montgomery, TX).^6,14-16 A 96-well microlevel plate was coated with 100 µL per well of 2 µL/mL fibronectin (F0895, Sigma-Aldrich, St Louis, MO) overnight at 4 °C and blocked for 30 minutes at room temperature. Subsequently, 100 µL of diluted patient or control IgG (1:20) solution premixed with 2 µg/mL His-tagged integrin αvβ6 (IT6-H52E1, ACROBiosystems, Newark, NJ) at 4 °C overnight was added and incubated for 60 minutes at room temperature. After washing 5 times with the wash solution, 100 µL of peroxidase-conjugated anti-His-tag monoclonal antibody (1:5000, D291-7, Medical and Biological Laboratories) was added, followed by incubation at room temperature for 60 minutes. After washing another 5 times with the wash solution, bound reactants were detected by incubation with 3,3ʹ,5,5ʹ-tetramethylbenzidine for 10 minutes. The absorbance was measured at 450 nm. The assays were performed in the presence of MgCl₂ and CaCl₂ (1 mM each).

To calculate the inhibition rate, we used control wells coated with integrin αvβ6 and incubated with fibronectin in the absence of patient or control IgG. The inhibition rate was calculated using the following formula: (control OD − sample OD)/control OD.

Immunohistochemical Analysis

Immunohistochemical analysis was performed according to standard methods for human formalin-fixed paraffin-embedded samples. We obtained ascending colonic and terminal ileal tissues from patients with UC at RPC and patients with colorectal cancer (CRC) at colectomy as controls, and ileal pouch tissues from patients with UC who developed pouchitis at pouch endoscopy. The primary antibody was anti-integrin β6 (1:3,000; NBP2-14136, Novus Biologicals, Centennial, CO) and the secondary antibody was peroxidase-conjugated anti-rabbit IgG (K4003, Dako, Santa Clara, CA). We enrolled 5 patients with CRC who underwent colectomy between October and December 2022 at Kyoto University Hospital (Supplementary Table S2).

Western Blot Analysis

Protein extracts from human descending and ascending colonic, and terminal ileal mucosal samples of patients with UC at RPC and CRC at colectomy were boiled in Laemmli sample buffer with 2.5% mercaptoethanol, fractionated on 4%-15% sodium dodecyl sulfate polyacrylamide gels (4561086, Bio-Rad Laboratories, Hercules, CA), and transferred to nitrocellulose membranes (10600003, Cytiva, Marlborough, MA) according to standard protocols. Blocking and antibody reactions were performed using iBind Flex Western System (SLF2000, Thermo Fisher Scientific, Waltham, MA) for 2.5 hours at room temperature, following the manufacturer’s protocol. The primary antibodies were anti-integrin αv (1:5000; ab179475, Abcam, Cambridge, UK), anti-integrin β6 (1:10 000; ab187155, Abcam), and anti-β actin (1:10 000; ab6276, Abcam). The secondary antibodies were peroxidase-conjugated anti-mouse IgG (1:10 000; A28177, Thermo Fisher Scientific) and anti-rabbit IgG (1:10 000; 31458, Thermo Fisher Scientific). Protein-antibody complexes were visualized using a chemiluminescent substrate (34095, Thermo Fisher Scientific). Immunoreactive protein bands were detected using Amersham Imager 600 (Cytiva). As controls, we enrolled 5 patients with CRC (Supplementary Table S2).

Statistical Analysis

Statistical analysis was performed using GraphPad Prism version 10 (GraphPad Software, San Diego, CA). For all boxplots, the box represents the interquartile range, the centerline represents the median, and the whiskers indicate the minimum to maximum value. Serum anti-integrin αvβ6 antibody levels at 0, 3, 6, 9, and 12 months after RPC and the inhibition of integrin αvβ6-fibronectin binding at 1 year after RPC were compared between patients who developed pouchitis and those who did not using Mann-Whitney U test. Changes in serum anti-integrin αvβ6 antibody levels over time (0, 3, 6, 9, and 12 months after RPC) were compared using Wilcoxon matched-pairs signed-rank test. The association between serum antibody levels and future development of pouchitis was assessed using receiver operating characteristic curve analysis. The optimal cutoff value was determined by the Youden index. Kaplan-Meier analysis was performed to compare the pouchitis-free survival probability between patients with serum antibody levels above and below the cutoff value at the time of RPC by the log-rank test. Patients were censored at the time of last follow-up or the study end date (December 2023). The correlation between serum anti-integrin αvβ6 antibody levels and the inhibition of integrin αvβ6-fibronectin binding was evaluated by means of the Pearson product moment correlation. Differences were considered statistically significant at P <.05.

Results

Serum Anti-integrin αvβ6 Antibody Levels in Patients With UC After RPC

We examined the serum anti-integrin αvβ6 antibody levels of 16 patients with UC who underwent RPC with IPAA by enzyme-linked immunosorbent assay. Serum anti-integrin αvβ6 antibody levels of patients with UC significantly decreased after RPC (P < .05, Figure 1A). However, in patients who developed pouchitis, the serum antibody levels remained high even after RPC (Figure 1B). Subsequently, we compared the serum anti-integrin αvβ6 antibody levels at the time of RPC in patients who developed pouchitis and those who did not. The serum anti-integrin αvβ6 antibody levels in patients who developed pouchitis were significantly higher than those in patients who did not (P < .05, Figure 1C). Table 1 presents the clinical characteristics of patients who developed pouchitis and those who did not. Patients who developed pouchitis were significantly younger at the time of RPC compared to those who did not (median age: 41.5 vs 55.0 years, P = .0005). No patients in this study had extraintestinal manifestations such as primary sclerosing cholangitis (PSC). We also examined whether the treatment modalities before and at the time of RPC influenced the development of pouchitis and found no significant differences in either previous treatments or treatments at the time of RPC between patients who developed pouchitis and those who did not (Table 1 and Supplementary Table S1).

Prediction of the Risk of Pouchitis Based on Serum Anti-integrin αvβ6 Antibody Levels

We compared serial changes of the serum anti-integrin αvβ6 antibody levels at 0, 3, 6, 9, and 12 months after RPC between patients who developed pouchitis and those who did not. In addition to the serum antibody levels at the time of RPC, significant differences were observed in serum anti-integrin αvβ6 antibody levels at 3, 9, and 12 months after RPC (Figure 2A). To establish a predictive marker for the risk of pouchitis, we determined the optimal cutoff value for serum anti-integrin αvβ6 antibody levels at the time of RPC through receiver operating characteristic curve analysis, which was found to be 106.5 U/mL (Figure 2B, Supplementary Figure S1). The sensitivity and specificity of this predictive marker for the onset of pouchitis were 75% and 100%, respectively. Patients with serum anti-integrin αvβ6 antibody levels above the cutoff had a significantly higher incidence of pouchitis than those below the cutoff, as demonstrated by Kaplan-Meier analysis (Figure 2C). The median period from RPC to the onset of pouchitis in patients with serum anti-integrin αvβ6 antibody levels above the cutoff was 1.95 years. These findings suggest the usefulness of serum anti-integrin αvβ6 antibody levels at the time of RPC as a predictive marker for development of pouchitis.

Inhibition of Integrin αvβ6-Fibronectin Binding by IgG From Patients With Pouchitis

We previously reported that serum anti-integrin αvβ6 antibodies in patients with UC inhibit integrin αvβ6-fibronectin binding.⁶ Therefore, we examined whether serum anti-integrin αvβ6 antibodies in patients with pouchitis also have an inhibitory activity on integrin αvβ6-fibronectin binding. Using purified IgG from patients with UC at 1 year after RPC, we conducted solid-phase integrin αvβ6-fibronectin binding assays. The results revealed that IgG from patients who developed pouchitis effectively blocked integrin αvβ6-fibronectin binding (Figure 3A and B). Importantly, the inhibitory activity of IgG correlated with the levels of serum anti-integrin αvβ6 antibodies, as in patients with UC.⁶

Expression of Integrin αvβ6 in Ascending Colonic, Terminal Ileal, and Ileal Pouch Epithelium

Colonic metaplasia of the ileal pouch has been proposed as one of the causes of pouchitis.^17,18 Integrin αvβ6 is known to be expressed in the colonic epithelium, but its expression in the terminal ileal epithelium remains unclear.^6,7 To examine the expression of integrin αvβ6 in the ascending colonic, terminal ileal, and ileal pouch epithelium, we conducted immunohistochemical analysis. At the time of RPC, integrin αvβ6 was expressed in the ascending colonic epithelium, but not in the terminal ileal epithelium in all patients with UC that were examined (Figure 3C). Interestingly, however, integrin αvβ6 expression could be observed in the ileal pouch epithelium of the patients who developed pouchitis (Figure 3D). In addition, we performed western blot analysis to quantitatively evaluate integrin αvβ6 expression at the time of RPC. The results showed that integrin αvβ6 was highly expressed in the ascending colonic mucosa of patients with UC, but it was barely expressed in the terminal ileal mucosa (Figure 3E).

Discussion

We previously reported the presence of serum anti-integrin αvβ6 antibodies in patients with UC with high sensitivity and specificity. In this study, we found that serum anti-integrin αvβ6 antibody levels in patients with UC significantly decreased after RPC; however, the serum antibody levels in patients who developed pouchitis remained high, even 1 year after RPC. Interestingly, although integrin αvβ6 was not expressed in the terminal ileal mucosa of patients with UC, expression became positive in the ileal pouch epithelium of the patients who developed pouchitis. Most importantly, preoperative serum anti-integrin αvβ6 antibody levels in patients who developed pouchitis were significantly higher than those in patients who did not develop pouchitis. These data suggest that serum anti-integrin αvβ6 antibody levels may be a useful marker for predicting the development of pouchitis. Previously, we reported that IgG in patients with UC inhibit integrin αvβ6-fibronectin binding.⁶ In this study, we confirmed that postoperative IgG levels in patients with UC have similar inhibitory effects.

To date, the precise mechanism for the decrease of serum anti-integrin αvβ6 antibodies after RPC in patients with UC remains unknown. However, integrin αvβ6 is expressed in the colonic epithelium, and its expression is increased in inflammatory conditions.^6,7 Thus, removal of the inflammatory colon with resulting disappearance of integrin αvβ6 as the autoantigen may have resulted in the decrease of serum anti-integrin αvβ6 antibody levels after RPC.

However, it should be emphasized that, when focusing on the patients who developed pouchitis, the serum antibody levels remained high even after RPC with IPAA. In this regard, it is noteworthy that, although integrin αvβ6 was not expressed in the ileal mucosa of patients with UC at the time of RPC, expression became positive in the ileal pouch mucosa of the patients who developed pouchitis in this study. Previous reports showing that the ileal pouch epithelium undergoes colonic metaplasia following pouch formation in patients with UC offer consistent results with the present data.^17,18 Taken together, the results suggested that expression of integrin αvβ6 in the ileal pouch mucosa after RPC with IPAA may continuously stimulate production of anti-integrin αvβ6 antibody.

In previous studies, we found that serum anti-integrin αvβ6 antibodies in patients with UC had inhibitory effects on integrin αvβ6-fibronectin binding and showed a positive correlation between serum anti-integrin αvβ6 antibody levels and disease activity, suggesting involvement of the antibody in the pathophysiology of UC.⁶ In this study, we observed that postoperative patients’ serum also had inhibitory effects on integrin αvβ6-fibronectin binding, and these effects correlated with the levels of serum anti-integrin αvβ6 antibodies, similar to the findings in patients with UC before RPC. Thus, it may be possible that anti-integrin αvβ6 antibodies are involved in the development of pouchitis by disrupting the epithelial barrier function through inhibition of integrin αvβ6-fibronectin binding. Taken together, these findings suggest that antibody production induced by integrin αvβ6 expressed in the pouch mucosa, and the inhibition of integrin αvβ6-fibronectin binding by anti-integrin αvβ6 antibodies together might create a vicious cycle in the development of pouchitis.

A recent study has shown that serum anti-integrin αvβ6 antibodies could be detected up to 10 years before the diagnosis of UC, indicating that the serum antibodies can be a biomarker for predicting the disease onset.¹⁹ Therefore, we investigated the usefulness of serum anti-integrin αvβ6 antibodies as a biomarker for predicting the development of pouchitis following RPC with IPAA. First, we found that serum anti-integrin αvβ6 antibody levels before RPC in patients who developed pouchitis were significantly higher than those who did not. Moreover, using the optimal cutoff value of serum anti-integrin αvβ6 antibody levels determined through receiver operation characteristic analysis at the time of RPC, the sensitivity and the specificity of the serum antibody levels for predicting the development of pouchitis were 75% and 100%, respectively. Furthermore, Kaplan-Meier analysis demonstrated a significantly higher incidence of pouchitis in patients with serum antibody levels above the cutoff. These data suggested that serum anti-integrin αvβ6 antibodies can be a good biomarker for predicting the development of pouchitis after RPC with IPAA in patients with UC. Whether anti-integrin αvβ6 antibodies may have some roles in the development of pouchitis remains to be clarified in future. Finally, we recently reported the presence of serum anti-integrin αvβ6 antibodies in patients with PSC, and that the serum antibody levels are higher in PSC patients with UC than those without UC.²⁰ Many patients with PSC also have UC.^21,22 In this study, however, no patient with PSC was included. Thus, it may be interesting to examine in future studies the relationship between serum anti-integrin αvβ6 antibody levels and development of pouchitis in terms of presence or absence of PSC.

The study has limitations that should be considered. As a single-center study with a relatively small sample size, the generalizability of our findings may be limited. Additionally, the retrospective nature of this study means we can only establish association, not causality. Despite these limitations, our study provides valuable insights into the potential role of anti-integrin αvβ6 antibodies in pouchitis development. The observed association between serum antibody levels and pouchitis risk suggests a promising avenue for future research. Larger prospective studies are warranted to confirm these findings and explore their clinical implications.

In conclusion, serum anti-integrin αvβ6 antibody levels in patients with UC were significantly decreased after RPC. However, the serum antibody levels remained high in patients who developed pouchitis. Since serum anti-integrin αvβ6 antibody levels before RPC in patients who developed pouchitis were significantly higher than those who did not, the serum antibody levels may be useful for predicting development of pouchitis after RPC with IPAA in patients with UC.

Acknowledgments

We thank the patients who provided serum samples for this study. We would also like to thank Shino Yamaguchi and Taichi Ito for their valuable technical support. We are grateful to Medical and Biological Laboratories Co., Ltd. for providing the Anti-Integrin αvβ6 ELISA Kit used in this study. Finally, we thank Editage for English language editing.

Author Contributions

Conceptualization: R.N., T.K., M.S., and Y.N.; Methodology: R.N. and T.K.; Formal analysis: R.N. and H.Y.; Investigation: R.N. and T.K.; Funding acquisition: T.K.; Resources: R.N., T.K., and S.O.; Supervision: T.C. and H.S.; Visualization: R.N.; Writing—original draft preparation: R.N. and T.K.; Writing—review and editing: M.S., Y.N., S.O., H.Y., T.Y., K.S., A.H., M.Y., I.T., K.C., M.Y., Y.M., S.M., T.M., N.U., T.C., and H.S.

Funding

This work was supported by the Japan Society for the Promotion of Science (Grant-in-Aid for Early-Career Scientists 22K16017 to T.K.).

Conflicts of Interest

T.K., M.S., T.C., and H.S. licensed a patent to Medical and Biological Laboratories Co., Ltd. for the anti-integrin αvβ6 antibody test employed in this study. The remaining authors declare no conflicts of interest.

References