Sex as a biological variable. This study included samples from 4 male and 3 female patients with UC. Sex was not considered a biological variable in the analyses because the study focused on the molecular and functional characterization of anti–integrin αvβ6 antibodies, which do not show sex-related differences. Therefore, the findings are expected to be relevant to both sexes.

Study design. The purpose of this study was to generate and characterize patient-derived anti–integrin αvβ6 antibodies. We used PBMC or mesenteric lymph node samples from patients with UC to identify B cells expressing anti–integrin αvβ6 antibody by B cell immortalization or single-cell sorting using flow cytometry. The gene of the variable region of the antibody was sequenced to produce an anti–integrin αvβ6 mAb. These sequences were annotated using IgBLAST (28) to identify gene usage and amino acid sequences. Based on our previous reports (4), we performed experiments such as ELISA and solid-phase integrin αvβ6 binding assay to characterize the antibodies.

Patients. A total of 7 patients with UC undergoing treatment at Kyoto University Hospital participated in this study. The clinical characteristics of the patients are summarized in Supplemental Table 1. Diagnosis of patients was based on a combination of their symptoms, endoscopic findings, histological features, and the lack of alternative diagnoses (29, 30). Patients were considered eligible if they were positive for anti–integrin αvβ6 antibodies, regardless of clinical severity or treatment. The study was conducted in accordance with the Declaration of Helsinki and was authorized by the Ethics Committee of the Graduate School of Medicine, Kyoto University. Written informed consent was obtained from all patients after an explanation of the nature and the possible outcomes of the study. Sample sources (lymph nodes or PBMCs) were selected solely on the basis of clinical availability and procedural feasibility. Lymph node samples were obtained from patients undergoing surgical procedures, which facilitated the collection of a sufficient number of B cells. PBMC samples were obtained from patients who did not undergo surgery. Each patient contributed only one type of sample. No notable differences were observed in the characteristics of antibodies derived from PBMCs and those derived from lymph nodes.

Sample preparation. To generate mAbs, blood or lymph node samples were collected from the patients. PBMCs were isolated from whole blood by gradient centrifugation using BD Vacutainer CPT (Becton, Dickinson and Co.). The lymph node samples were obtained from the mesenteric lymph nodes attached to the surgical specimen during total colectomy. Lymph node samples were chopped with a clean scalpel, mashed on a 70 μm mesh using a plunger, and washed with PBS. After centrifugation of the cell suspension at 300g for 5 minutes, the supernatant was removed, and 1× RBC Lysis Buffer (pluriSelect) was added. After 3 minutes, the reaction was stopped by addition of PBS with 0.1% BSA. The resulting cell suspension was subjected to centrifugation at 300g for 5 minutes to obtain lymph node–derived cells, which were preserved at –80°C for further use.

Preparation of Epstein-Barr virus reagent. The Epstein-Barr virus (EBV) reagent was prepared and stocked as a culture supernatant harvested from the B95-8 marmoset cell line obtained from the Japanese Collection of Research Bioresources Cell Bank (cell ID: JCRB9123). B95-8 cells were cultured in RPMI 1640 medium (Fujifilm Wako Pure Chemical) containing 10% FBS at 37°C in 5% CO2 until reaching confluence. The culture medium was centrifuged at 400g for 10 minutes to remove all residue, and the supernatant was divided into aliquots and stored at –80°C until used for transduction.

Establishment of lymphoblastoid cell lines by EBV transduction. Lymphoblastoid cell lines (LCLs) were established as previously reported (31, 32). Briefly, IgM+ B cells were removed from PBMCs using magnetic cell sorting with anti-human IgM MicroBeads following the manufacturer’s instructions (Miltenyi Biotec). PBMCs obtained after removal of IgM+ B cells were suspended in EBV stock (1 × 107 cells/mL) containing 2.5 μg/mL ODN2006 (Alpha Diagnostic International) and kept at 37°C under 5% CO2 for 1 hour. The cells were suspended in RPMI 1640 (2 × 105 cells/mL) containing 20% FBS, streptomycin/penicillin, 2.5 μg/mL ODN2006 CpG, 50 IU recombinant human IL-2 (R&D Systems), and 500 ng/mL cyclosporin A (Tokyo Chemical Industry Co) and seeded at 200 μL/well in round-bottomed 96-well plates. After culture for 2 weeks, ELISA was performed using culture supernatants to identify LCLs producing anti–integrin αvβ6 antibodies, and RNA isolation was performed from these LCLs. Total RNA was extracted from LCLs using an RNeasy Mini Kit (catalog 74106, QIAGEN) following the manufacturer’s protocol.

An alternative method was to scale up the LCLs producing the antibody of interest and then use a cell array method to identify single cells producing anti–integrin αvβ6 antibodies. This method was supported by EVEC Inc. Briefly, antigen-immobilized microbeads were seeded onto microarray chips together with LCLs and allowed to react for several hours. Array chips were washed and stained with anti-human IgG-R-phycoerythrin, and wells with positive beads were identified using fluorescence microscopy. RNA isolation was performed from cells in positive wells.

Single-cell sorting. MAbs were generated from antigen-specific B cells using a single-cell sorting protocol based on a previously reported method (33), with minor modifications as detailed below. PBMCs or lymph node cells were resuspended at 1 × 106 cells per 100 μL in FACS buffer, which comprised 49 mL of Dulbecco’s PBS and 1 mL of FBS. Cells were stained on ice for 20 minutes with the following antibodies: Alexa Fluor 700–mouse anti–human CD20 (1:80; clone L27; catalog 560631, Becton, Dickinson and Co.), APC–mouse anti-human IgG (1:20; clone G18-145; catalog 550931, Becton, Dickinson and Co.), and DAPI (1:100; Thermo Fisher Scientific).

To detect integrin αvβ6–specific B cells, a bait complex was prepared by preincubation of biotinylated integrin αvβ6 (IT6-H82E4, ACROBiosystems) with phycoerythrin-conjugated (PE-conjugated) NeutrAvidin (Thermo Fisher Scientific) in FACS buffer at a fixed volumetric ratio (31:1:18 μL, respectively). This bait complex was applied at an approximate cell-to-bait ratio of 20:1. It enables the detection of integrin αvβ6–specific B cells through fluorescent signal amplification. Appropriate negative controls were included to assess background staining. The gating strategy for detection of integrin αvβ6–specific B cells using PBMCs or lymph node cells by flow cytometry is shown in Supplemental Figure 4.

Before sorting, each well of a 96-well PCR plate was preloaded with 4 μL of sorting buffer, comprising 3.1 μL of nuclease-free H2O, 0.2 μL of RNasin (40 U/μL; Promega), 0.1 μL of RNaseOUT (40 U/μL; Thermo Fisher Scientific), 0.2 μL of 10× PBS (resulting in 0.5× PBS), and 0.4 μL of 100 mM DTT (final concentration 10 mM; Thermo Fisher Scientific). Single viable B cells (CD20+, IgG+, DAPI–, and PE+ for αvβ6 binding) were individually sorted into the wells using a BD FACSAria II cell sorter (Becton, Dickinson and Co.). After sorting, the plates were immediately frozen and stored at –80°C until RNA extraction.

Single-cell cDNA synthesis and PCR for amplifying variable regions of BCRs. To characterize the antibody repertoire at the single-cell level, reverse transcription (RT) and semi-nested PCR were performed to amplify the variable regions of Ig heavy and light chain genes from individual B cells.

Sorted single B cells were thawed on ice and first incubated at 65°C for 2.5 minutes with a random-hexamer primer mix comprising 5.6 μL of nuclease-free H2O, 0.75 μL of random-hexamer primers (200 ng/μL; Thermo Fisher Scientific), 0.5 μL of NP-40 (10%; Thermo Fisher Scientific), and 0.15 μL of RNaseOUT (40 U/μL; Thermo Fisher Scientific). After incubation, the samples were placed on ice for at least 2 minutes.

Subsequently, an RT mix containing 2.05 μL of nuclease-free H2O, 3 μL of 5× RT buffer (Thermo Fisher Scientific), 0.5 μL of dNTP mix (25 mM; Thermo Fisher Scientific), 1 μL of DTT (100 mM), 0.1 μL of RNasin (40 U/μL; Promega), 0.1 μL of RNaseOUT, and 0.25 μL of SuperScript IV reverse transcriptase (200 U/μL; Thermo Fisher Scientific) was added to each well (final volume of RT mix: 7 μL). The RT reaction was performed using the following thermal protocol: 42°C for 10 minutes, 25°C for 10 minutes, 50°C for 10 minutes, and 94°C for 5 minutes, followed by a hold at 4°C. Following RT, the resulting cDNA was diluted with 16 μL of nuclease-free water before PCR amplification.

The variable regions of Ig heavy and light chains were amplified using a semi-nested PCR strategy with Platinum Taq DNA Polymerase or Platinum Taq Green Hot Start DNA Polymerase (Thermo Fisher Scientific), using exactly the same optimized primer set described in the previous study (33), which was specifically designed for efficient amplification of Ig variable regions.

For both the first- and second-round PCRs, reactions were conducted in a 25 μL total volume using Platinum Taq DNA Polymerase (Thermo Fisher Scientific). The master mix comprised the following components per reaction: 14.68 μL of nuclease-free H2O, 2.05 μL of 10× Platinum Taq PCR buffer, 1.23 μL of KB Extender (6%), 0.61 μL of MgCl2 (50 mM), 0.16 μL of dNTP mix (25 mM), and 0.09 μL each of forward and reverse primers (50 μM), using the same primer sets previously optimized and described in detail (33). For the first-round PCR, 6 μL of diluted cDNA was used as a template; subsequently, 1 μL of the first-round PCR product was used as the template in the second-round PCR. For second-round PCR, Platinum Taq Green Hot Start DNA Polymerase (Thermo Fisher Scientific) was used to enable direct visualization of PCR products on agarose gels via the included tracking dye.

The first-round PCR was performed under the following conditions: 94°C for 2 minutes, followed by 50 cycles at 94°C for 30 seconds, 57°C for 30 seconds, and 72°C for 55 seconds. The second-round PCR used similar conditions, with an extended step involving 72°C for 45 seconds. Second-round PCR products were analyzed by agarose gel electrophoresis, and samples of the correct size (approximately 500 bp for heavy chains and 450 bp for light chains) were subjected to Sanger sequencing. The same protocol was applied to RNA derived from LCLs.

Analysis of the V region gene sequence of BCRs. Aliquots of the second-round PCR products were analyzed by 2% agarose gel electrophoresis to confirm the presence of amplicons of the expected size — approximately 500 bp for heavy chain and 450 bp for light chain variable regions. PCR products of the correct size were subjected to Sanger sequencing, and the resulting sequences were annotated using IgBLAST (28) with the IMGT reference database to determine V(D)J gene usage, CDR3 sequences, and repertoire characteristics and to infer clonal relationships. Sequences containing stop codons or out-of-frame rearrangements (i.e., nonproductive sequences) were excluded from further analysis.

Cloning of the V region gene sequence of BCRs. For cloning of BCR variable regions, the first-round PCR product was reamplified using KOD -Plus- Neo polymerase (TOYOBO) and the same optimized primer set with vector-compatible overhangs as described in a previous report (33). PCR comprised the following thermal cycle: 98°C for 30 seconds; 35 cycles of 98°C for 10 seconds, 65°C for 30 seconds, and 72°C for 30 seconds; and 72°C for 2 minutes. Before cloning, amplified PCR products were purified using a PCR purification kit (QIAquick PCR Purification Kit, QIAGEN).

AbVec2.0-IGHG1 (IgG1) (Addgene plasmid 80795), AbVec1.1-IGKC (Igk) (Addgene plasmid 80796), and AbVec1.1-IGLC2-XhoI (Igl) (Addgene plasmid 99575) were used as human antibody expression vectors as previously reported (33). EcoRI and SalI were used for IgG1, EcoRI and BsiWI for IgK, and EcoRI and XhoI for IgL to linearize each vector. PCR products and expression vectors were cloned using In-Fusion Snap Assembly Master Mix (Takara Bio Inc.).

Expression plasmids were obtained by transforming of Stellar Competent Cells (Takara Bio) and purified using the QIAprep Spin Miniprep Kit (QIAGEN). To screen for reactivity to integrin αvβ6, antibodies were produced in HEK293 cells (originally obtained from ATCC, CRL-1573) by transfection with Lipofectamine 3000 (Thermo Fisher Scientific). Transfected cells were maintained in DMEM containing 2% FBS for 5 days, and then the supernatant was used for ELISA. Plasmid DNA from reactive clones was transformed into Stellar Competent Cells and purified using the QIAGEN Plasmid Maxi Kit.

Antibody production. Recombinant antibodies were transiently expressed using the ExpiCHO Expression System (Thermo Fisher Scientific). ExpiCHO cells were cotransfected with a mixture of expression vectors for the heavy and light chains of the antibodies following the manufacturer’s protocol.

After 10 days of culturing of the transfected cells, the clarified culture supernatant was loaded into Ab-Capcher (ProteNova), and mAbs were purified in accordance with the manufacturer’s instructions. Disposable plastic columns (Thermo Fisher Scientific) were used according to the manufacturer’s recommended protocol to obtain solubilized recombinant mAbs.

Preparation of human IgG. Ab-Rapid SPiN (P-013, ProteNova) was used to purify IgG from the sera of patients with UC and controls. The purified IgG was then dialyzed against PBS (pH 7.2), and concentrated by ultrafiltration with an Amicon Ultrafilter (UFC805024, Millipore) followed by storage at –20°C. Purified IgG concentrations were measured using a NanoDrop 2000c spectrophotometer (Thermo Fisher Scientific).

ELISA. ELISA starter accessory kits (E101, Bethyl Laboratories) were used in accordance with the instructions of the manufacturer. Microtiter plates were coated using carbonate-bicarbonate buffer (coating buffer) with 100 μL of 2 μg/mL recombinant human integrin αvβ6 heterodimer proteins (IT6-H52E1, ACROBiosystems), incubated overnight at 4°C, washed 3 times with TBS containing 0.05% Tween 20 (wash solution), and blocked with TBS containing 1% BSA for 30 minutes at approximately 25°C. After 3 washes with wash solution, 10-fold serial dilutions of mAbs starting at 10 μg/mL and 3 μg/mL (diluted with TBS with 0.05% Tween 20 and 1% BSA) were added, and plates were incubated for 1 hour at approximately 25°C. Plates were washed 5 times and incubated with 100 μL goat anti-human IgG antibody conjugated with horseradish peroxidase (1:50,000; ab6759, Abcam) at 25°C for 60 minutes. After washing, the bound antibodies were detected by incubation with 3,30,5,50-tetramethylbenzidine for 10 minutes. Absorbance was noted at 450 nm. EC50 values were calculated by nonlinear regression analysis on the binding curves using GraphPad Prism version 10 (GraphPad Software). ELISA was carried out in the presence or absence of MgCl2 and CaCl2 (1 mM each). MgCl2 and CaCl2 were added to buffer for washing, blocking, and dilution of antibodies. The same method was used to assess reactivity with other integrins. The mAb 10D5 (ab77906, Abcam) was used as the positive control and IgG derived from healthy individuals (143-09501, Fujifilm Wako Pure Chemical) as the negative control.

To examine whether the RGD (Arg-Gly-Asp) peptide blocked the binding of each mAb to integrin αvβ6, the RGDS (Arg-Gly-Asp-Ser) peptide (A9041, Sigma-Aldrich) or the control RGES (Arg-Gly-Glu-Ser) peptide (A5686, Sigma-Aldrich) was added to each mAb before incubation. In experiments with peptides, the final concentration of mAb was 3 μg/mL, and each peptide was adjusted to 5 concentrations: 0 μg/mL, 12.5 μg/mL, 25 μg/mL, 50 μg/mL, and 100 μg/mL.

We used an anti–integrin αvβ6 ELISA kit (catalog 5288, Medical & Biological Laboratories) for detecting anti–integrin αvβ6 IgG antibody titers from patients with UC according to the manufacturer’s instructions.

Biolayer interferometry. The affinity between each mAb and integrin αvβ6 was measured using biolayer interferometry (BLI) with an Octet RED96 (ForteBio). Biotinylated integrin αvβ6 (IT6-H82E4, ACROBiosystems) was loaded at 25 nM in kinetics buffer (0.1% BSA, 0.6 M sucrose, 0.02% Tween 20, 1 mM MgCl2, and CaCl2 in TBS) for 300 seconds onto a SAX2 biosensor (ForteBio). The association of integrin αvβ6 and mAbs at 200, 50, 12.5, 3.13, and 0.78 nM was measured in kinetics buffer for 300 seconds. The measurement range was adjusted according to the dissociation constant (KD) value of each mAb. Dissociation in kinetics buffer was measured for 300 seconds. The on-rate constant (Kon), off-rate constant (Koff), and KD values were calculated using a global fit to a 1:1 binding model.

Solid-phase integrin αvβ6 binding assay. For this assay, 96-well microtiter plates were previously coated using coating buffer with either 5 μg/mL of fibronectin (F0985, Sigma-Aldrich) or 0.5 μg/mL of LAP (LAP-H5213, ACROBiosystems) (100 μL/well, 4°C, overnight). After removal of the coating solution, the plates were blocked by TBS containing 1% BSA. In another 96-well plate, 60 μL/well of a 2× stock (4 μg/mL of integrin αvβ6 with His-tag for fibronectin or 0.4 μg/mL integrin αvβ6 with His-tag for LAP) was combined with 60 μL/well of a 2× stock of each mAb diluted in the same way as for ELISA and incubated for 1 hour. After the ligand-coated plates were washed, 100 μL of the mAb–integrin αvβ6 mixture was transferred to the ligand-coated plate and incubated for 1 hour. After washing of the plate with wash solution, an Anti-His-tag mAb-HRP-DirecT (1:5,000; D291-7, Medical & Biological Laboratories) was added followed by incubation for 60 minutes. After the wash, bound antibodies were incubated with 3,30,5,50-tetramethylbenzidine for 10 minutes for detection. The absorbance was measured at 450 nm.

Statistics. Statistical analysis was conducted using GraphPad Prism (version 10). The correlation between anti–integrin αvβ6 IgG titers and blocking activity of integrin αvβ6–fibronectin or integrin αvβ6–LAP binding was evaluated using the Pearson product-moment correlation. A P value less than 0.05 was considered to indicate statistical significance. For experiments using patient-derived IgG, the cutoff OD values for the antibody titer and the inhibitory effect were defined as the mean value of control IgG plus 3 SD.

Study approval. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethics Committee of the Graduate School of Medicine, Kyoto University, Kyoto, Japan (protocol R1004). Written informed consent was obtained from all patients after they were provided with a full explanation of the nature and possible outcomes of the study. Recombinant DNA experiments were performed following approval by the Kyoto University Recombinant DNA Experiment Safety Committee (approval 230103).

Data availability. Values for all data points shown in graphs and values behind any reported means are provided in the Supporting Data Values XLS file accompanying this article. No next-generation sequencing data were generated in this study. All BCR variable region sequences were obtained by Sanger sequencing, which does not fall under the MINSEQE guidelines. All data supporting the findings of this study are included in the article or its supplemental materials. Additional information is available upon reasonable request.