Data collectionThe transcriptome data and corresponding clinical data were downloaded from the Gene Expression Omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/) database. All transcriptome data in Fragments Per Kilobase Million (FPKM) format underwent conversion to Transcripts Per Kilobase Million (TPM) normalized format for subsequent analysis using RStudio software. After excluding cases with missing prognostic information, the GSE31312 dataset comprised 471 cases and served as the training cohort for the prognostic model. Additionally, the GSE87371 and GSE10846 datasets, with 121 and 233 cases respectively, were utilized as external validation cohorts. The baseline characteristics of DLBCL patients from different cohorts are summarized in Table 1.Table 1 The characteristics of the DLBCL patients in GEO datasets.Identification of differentially expressed BRRGsTo identify key genes associated with ibrutinib resistance, we downloaded the transcriptome data of ibrutinib-resistant and non-resistant DLBCL cell lines (GSE138126) from the GEO database. All cell line was cultured with a medium containing ibrutinib for over 8 months, and resistant cell clones were selected. The ‘limma’ package (version 3.5.1) was used to identify differentially expressed BRRGs (DEBRRGs), and DEGs threshold was set as follows: |log2-fold change (FC)|> 2 and an adjusted P-value < 0.05. Volcano Plot and heatmap were conducted with the ‘pheatmap’ package for visualizing the gene expression differences.Functional enrichment analysisThe ‘cluster Profiler’ package was used to gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of DEBRRGs involved in disease progression. The annotated gene sets of GSEA were selected, c2.cp.kegg.v2023.1.Hs.entrez and c5.go.bp.v2023.1.Hs.entrez sets from the Molecular Signature Database (MSigDB) (https://www.gseamsigdb.org/gsea/msigdb/index.jsp). The number of permutations was set to 1,000. The criteria for screening statistically significant pathways were set as adjusted P-value less than 0.05. Visualization was performed using the ‘enrichplot’ package.Establishment and validation of BRRGs prognostic signatureThe prognostic signature was constructed using the GSE31312 dataset as the training cohort, with the GSE87371 and GSE10846 datasets serving as external validation cohorts. Univariate and LASSO regression analysis was utilized to screen the risk model containing the BRRGs, and the risk model was defined as the BRRGs signature. The resistance score for each patient in the training cohort was computed using the following formula: \(\text{Resistance score}={\sum }_{i=1}^{N}(\text{exp}\times coef)\), where N is the number of model genes; exp represents the gene expression value of each gene; coef represents the coefficient index.To validate the prognostic model, the resistance score for each individual was calculated through the regression coefficients and their expression. All cohorts of DLBCL patients were divided into high- and low-resistance score groups by the median resistance score. The disparities in overall survival (OS) among the groups were assessed utilizing the Kaplan–Meier survival analysis. The ‘timeROC’ package was utilized to construct the time-dependent receiver operating characteristic (ROC) curve and determine the area under the curve (AUC) of the 2-year, 3-year, and 5-year OS in the DLBCL patients. The predictive efficacy of the model was evaluated based on the AUC.Immune infiltration assessmentThe ‘GSVA’ package was used to perform a gene set enrichment analysis ssGSEA algorithm to unambiguously present the infiltrating score of 29 tumor-infiltrating immune cells and pathways in each sample (aDCs, APC co-inhibition, APC co-stimulation, B cells, CCR, CD8+ T cells, Check-point, Cytolytic activity, DCs, HLA, iDCs, Inflammation-promoting, Macrophages, Mast cells, MHC class I, Neutrophils, NK cells, Parainflammation, pDCs, T cell co-inhibition, T cell co-stimulation, T helper cells, Tfh, Th1 cells, Th2 cells, TIL, Treg, Type I IFN Response, and Type II IFN Response). Furthermore, we further analyzed the differences between the high- and low-resistance score groups.Chemotherapeutic response predictionThe ‘pRRophetic’ package (version 0.5) was used to predict chemosensitivity between different groups, primarily mainly construct ridge regression model to infer half-maximal inhibitory concentration (IC50) values based on gene expression levels through ten-fold cross-validation19,20. The dataset within the ‘pRRophetic’ package is derived from the “cgp2016” initiative, encompassing gene expression matrices and drug treatment information. We analyzed common chemotherapy drugs and used boxplots to illustrate differences in drug sensitivity between the two groups.Weighted gene co‑expression network analysis (WGCNA)The ‘WGCNA’ package was used to identify BTKi-resistance gene clusters highly associated with DLBCL based on the GSE138126 dataset. The weighted adjacency matrix was converted into a topological overlap matrix (TOM) according to the optimal soft threshold (β = 10), and then hierarchical clustering analysis was performed to detect the correlation between gene modules (minmodulesize = 100; mergecutheight = 0.2). Interaction strength was assessed using the heatmap toolkit, and gene significance (GS) and module membership (MM) were calculated to assess the relationship between module and resistance characteristics.Statements, patient samples and cell cultureAll experiments were performed in accordance with relevant guidelines and regulations. 20 DLBCL tissues and 5 lymph node reactive hyperplasia tissues were obtained from the Second Affiliated Hospital of Anhui Medical University. All DLBCL patient’s baseline characteristics were summarized in Supplementary Table 1. This study was approved by the Ethics Committee of the Second Hospital of Anhui Medical University, and informed consent was obtained from the patients. Human DLBCL cell lines SU-DHL-2 (RRID:CVCL_9550) and SU-DHL-4 (RRID:CVCL_0539) were purchased from the Institute of Biochemistry and Cell Biology of the Chinese Academy of Science. Cells were cultured in RPMI-1640 (Hyclone, Logan, UT, USA) containing 10% FBS at 37 °C cell incubators with 5% CO2.SiRNA transfectionCells were transfected with TRIP13 siRNA and negative control siRNA (GenePharma, China) using LipofectamineTM 3000 (Invitrogen, USA) reagent following the manufacturer’s instructions. Briefly, 4 × 105 cells were seeded in a 24-well plate, dilute 4 µL siRNA (20 µM) with 50 µL Opti-MEM, and dilute 1uL LipofectamineTM 3000 with 50 µL Opti-MEM. Mix the transfection reagent and siRNA diluent and add 24-well plate, continue to incubate for 6 h and then replace the medium, after 24 h of transfection, verify the expression of TRIP13 through subsequent assays. The sequences of siRNAs are listed in Supplementary Table 2.RNA extraction and quantitative PCRTotal RNA was extracted using Trizol reagent (sangon Biotech, Shanghai) according to the manufacturer’s protocol, and cDNA was synthesized using RevertAid First Strand cDNA Synthesis Kit (Thermofisher, USA). mRNA levels were measured by qPCR using an ABI 7500 System (Life Technologies, USA). The relative expression was determined using the 2−ΔΔCt method, with GAPDH as an endogenous control. The primers are listed in Supplementary Table 3.Western blotTotal protein was obtained as previously described21. Proteins were separated by SDS-PAGE and transferred to NC membranes. Incubate the membrane overnight with one of the antibodies listed in Supplementary Table 4. Then incubate with corresponding HRP-conjugated secondary antibodies. To save costs and reduce antibody loss, PVDF membranes were cut based on the molecular size of the target protein. Original blots are presented in Supplementary Fig. 1. Finally, immunoreactive bands were detected with WesternBright ECL kit (Advansta, USA). Grayscale analysis of WB bands was performed using ImageJ software.Cell viability assayCell viability was detected using CCK8 reagent (Beyotime, China) according to the manufacturer’s protocol. The absorbance of the cells was measured using multiscan spectroscopy at a wavelength of 450 nm.Chemosensitivity assay in vitroHuman DLBCL cell lines SU-DHL-2 and SU-DHL-4 were seeded into 96-well plates at a density of 1 × 104 cells per well after siRNA transfection. The cells were treated with various concentrations (0, 0.01, 0.1, 1, 2, 5, 10, 20 µg/mL) of ibrutinib (Aladdin) for 24 h. Cell viability was subsequently assessed using the CCK8 assay. The log(inhibitor) versus response model in GraphPad Prism 6 was employed to generate the fitted curve and calculate the half-maximal inhibitory concentration (IC50).Apoptosis analysisApoptosis was detected by Annexin V-APC/PI Apoptosis Detection Kit (BestBio, China). The fluorescence of at least 5,000 cells per sample was measured on a Cytomics FC 500 flow cytometer (Beckman Coulter, USA) for further calculations.Xenograft mouse modelFemale BALB/c nude mice were purchased from Nanjing Jicui Yaokang Biotechnology Co., Ltd. This experiment was undertaken according to the guidelines for the Animal Care and Use Committee of the Anhui Medical University. The in vivo experiments were authorized by the Animal Care and Use Committee of the Anhui Medical University. All animal procedures were performed in specific pathogen-free (SPF) barrier facilities. We used sodium pentobarbital anesthesia to minimize the pain of nude mice during the experiment. 2 × 107 TRIP13 knockdown and control SU-DHL-4 cell lines were injected into subcutaneous of 5-week-old female BALB/c nude mice (n = 3 for each group). The tumor volume was measured every 3 days after injection. After 30 days, the mice were sacrificed under deep anesthesia induced by intraperitoneal injection of sodium pentobarbital, and the tumor weights were recorded.immunohistochemistry testingThe experiment of immunohistochemistry (IHC) was executed as previously described22. All antibodies used in the study are listed in Supplementary Table 4. Typical images (40×) were captured using a microscope system (ZEISS, Germany).Statistical analysisAll statistical analyses were performed using the R software (v.4.2.1). Student’s t-tests were used to compare the differences between the two groups. The results were presented as the mean ± standard deviation (SD) of at least three independent experiments. P < 0.05 was considered statistically significant.
