Ethical Considerations

Animal experiments were approved by the Animal Care and Use Committee of Nagoya University Graduate School of Medicine (approval number: M230086) and performed in compliance with the regulations and guidelines of animal care and use of Nagoya University. All the experiments were performed and reported in accordance with ARRIVE guidelines. The human study was conducted following the Declaration of Helsinki for Human Research, and protocols were approved by the Research Ethics Committee of Nagoya University (approval number: 2018–0316). All participants provided written informed consent to participate in the study.

Human Samples

Serum, colonic biopsy, and surgical samples were obtained from 31 patients with active CD (CD activity index [CDAI] scores between 150 and 450) who were resistant to anti-TNF-α antibodies and treated with ustekinumab at Nagoya University Hospital in Japan from October 2017 to September 2020. The patients were divided into two groups (responders and non-responders), according to their response to ustekinumab treatment. Responders were those with a CDAI score that decreased by ≥ 100 from baseline or those with a CDAI score < 150 at 24 weeks after ustekinumab induction.

Measurement of Human TNF-α and Full-Length and Mature IL-18 Serum Concentrations

Serum concentrations of full-length and mature human IL-18 and TNF-α were measured using an enzyme-linked immunosorbent assay (ELISA) kit (#7620; MBL, Nagoya, Japan; #E-I-002 mAbProtein, Shimane, Japan, #DTA00D; R&D, Minneapolis). Full-length and mature IL-18 and TNF-α concentrations were calculated using standard curves.

Immunohistochemistry

For immunohistochemical studies, formaldehyde-fixed, paraffin-embedded biopsy and surgical specimens were deparaffinized, and antigen retrieval was performed in a target retrieval solution (Agilent Technologies, Santa Clara, CA, USA). After cooling to 20 °C, tissue sections were washed with phosphate-buffered saline (PBS), blocked with normal goat serum (#MP7451, Vector Laboratories, Burlingame, CA, USA), and incubated with anti-full-length and anti-mature IL-18 antibodies (mAbProtein) diluted 1:100 in PBS. The sections were treated with a 3% hydrogen peroxide/ethanol solution and incubated at 20 °C for 60 min with anti-rabbit IgG secondary antibody (#MP7451, Vector Laboratories), followed by signal detection using diaminobenzidine solution. The software ImageJ was used to quantify the stained areas (National Institute of Health, USA).

Immunofluorescence Staining

To retrieve antigen, the samples that were fixed with formaldehyde were deparaffinized and boiled in a target retrieval solution (Agilent Technologies) at a pH of 6. For mouse samples, after washing with PBS, samples were permeabilized in 0.05% Triton X-100 (Sigma-Aldrich, St. Louis, MO, USA)/PBS solution for 20 min at 20 °C. Samples were blocked with normal goat serum and incubated with primary antibody (anti-MUC-2, 1:100; Invitrogen, Waltham, MA, USA) at 4 °C for 16 h. Samples from patients with CD were blocked and permeabilized with 0.1% Triton X-100 diluted in 5% bovine serum albumin for 1 h, and then incubated with primary antibody (anti-Muc2, sc-515032, 1:100, Santa Cruz Biotechnology, Dallas, TX, USA) at 4 °C for 16 h. Both mouse and human slides were incubated with Alexa Fluor 488-conjugated secondary antibodies (#4412; Cell Signaling Technology) for 1 h. After washing, the slides were mounted in 4ʹ,6-diamidino-2-phenylindole Fluoromount-G (#010020, Southern Biotech, Birmingham, AL, USA) and then coverslips added. The stained areas were quantified using ImageJ software (National Institute of Health, USA).

Multiple immunofluorescence staining was conducted using an Opal assay kit (NEL810001KT, Akoya Biosciences, Marlborough, MA, USA) according to the manufacturer’s recommendations. Briefly, anti-human CD68 antibody (PG-M1, 1:100, DAKO) and anti-full-length and anti-mature IL-18 antibodies (mAbProtein) were incubated in 4 °C for 16 h and recognized by fluorescence of Opal570 and Opal520, respectively.

Alcian Blue and Periodic Acid-Schiff Staining

To perform Alcian Blue-periodic acid-Schiff (PAS) staining, a PAS and Alcian Blue staining kit (#40,582, Muto Pure Chemicals, Tokyo, Japan) was used. Images were obtained under a universal fluorescence microscope (BZ-9000, Keyence). The number of goblet cells was determined after hematoxylin and eosin staining by counting five high-power fields (400 ×) in crypts.

Mouse Model of Acute Colitis

Male C57BL/6 J mice, 7–9 weeks of age, were purchased from CLEA Japan (Tokyo, Japan). The mice were maintained under specific pathogen-free conditions with a 12-h day/night cycle and controlled humidity and temperature. Temperature was maintained at 18–23 °C with 40–60% humidity. Each experimental group comprised seven mice. For the 2,4,6-trinitrobenzene sulfonic acid (TNBS) acute colitis model, we first conducted a preliminary study, which indicated low survival rates at high TNBS concentrations. Because exploring the efficiency of drug administration is difficult with early deaths, mice in the IL-18 mAb and isotype IgG groups were intrarectally administered TNBS (Wako Chemicals, Osaka, Japan) dissolved in 50% ethanol at a dose of 1 mg on day 1.

Mice in the control group were administered intrarectally without TNBS on day 1 without intervention.

Mice in the isotype IgG and IL-18 mAb groups, which modeled TNBS-induced colitis, were treated intraperitoneally with IgG (200 μg) (Bio X Cell, Lebanon, NH, USA) or anti-mature IL-18 mAb (5-4.1) (200 μg) daily for 5 d. All mice were euthanized using carbon dioxide on day 5, and colon tissues were collected. The severity of colitis was evaluated based on changes in body weight, disease activity score, colon length, and histological score. Histological evaluation was performed blinded, as described previously.

Quantitative PCR

Total RNA was extracted using the acid guanidinium thiocyanate-phenol–chloroform extraction method, and purified RNA samples were reverse-transcribed into cDNA using ReverTra Ace (Toyobo, Tokyo, Japan). Quantitative PCR (qPCR) was performed using an Mx3000P thermal cycler (Agilent Technologies). TaqMan probes and primers for mouse CXCL2 (Mm00436450_m1), mouse glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Mm99999915_g1), and mouse IL-6 (Mm00446190_m1) were purchased from Life Technologies (Carlsbad, CA, USA). Data were analyzed using the comparative threshold cycle (CT) method and normalized to GAPDH levels.

Measurement of Tissue Cytokine Concentrations

Mouse CXCL2 and IL-6 levels in colon tissues were measured using an ELISA. Distal colons were dissected, opened longitudinally, and washed twice with PBS. Colon homogenates were prepared using tissue homogenizers in a lysis buffer containing a protease inhibitor cocktail (#5871, Cell Signaling Technology, Danvers, MA, USA), and protein concentrations were measured. The samples were stored at –80 °C until assayed. Cytokine concentrations were determined using a Quantikine ELISA kit for mice (#M6000B-1; #MM200; R&D Systems, Inc., Minneapolis, MN, USA).

Microbiome Analysis

Fecal samples were collected, and DNA was extracted using a DNeasy PowerSoil Kit (Qiagen, Hilden, Germany). Bacterial DNA was amplified using universal primers targeting the V3–4 region of the 16S rRNA gene (F: 5ʹ-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAGCCTACGGGNGGCWGCAG-3ʹ, R: 5ʹGTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATCC-3ʹ) with KAPA HiFi HotStart ReadyMix (KAPA Biosystems, Boston, MA, USA). PCR products were pooled to create a library and then sequenced on an Illumina MiSeq sequencer (Illumina, San Diego, CA, USA). Paired-end reads were generated using a MiSeq Reagent Kit v3 with 2 × 300 reads and 600 cycles (Illumina). For basic analysis of the 16S rRNA gene sequence data, Quantitative Insights Into Microbial Ecology (QIIME 2–2021.4 with DADA2) [16] and SILVA (version 138) were used. Linear discriminant analysis effect size (LEfSe) [17] was used to compare intestinal microbiome compositions.

Statistical Analyses

Statistical analyses were conducted using GraphPad Prism 9 (IBM Corp., Armonk, NY, USA). Data are presented as means ± standard deviations (SD). Statistical comparisons were conducted using the Mann–Whitney U test or analysis of variance (ANOVA) with Tukey’s multiple comparison post-hoc test. Statistical significance was defined as P < 0.05.

Databases

The IBD Transcriptome and Metatranscriptome Meta-Analysis (IBD TaMMA; https://ibd-meta-analysis.herokuapp.com) database was used to analyze IL-18 expression based on RNA-sequence data from patients with CD. To determine the origin of IL-18 secretion cell types in patients with CD, we searched the scRNA-seq database derived from the Single Cell Portal (SCP1423, https://singlecell.broadinstitute.org/single_cell).