CSRP1 is a potential therapeutic targets for DLBCL: A computational analysis and in vitro experiments Authors: Wanlu Tian, Fengbo Jin, Yingying Chen, Mingzhen Yang Published: 2025 ABSTRACT At present, the role of Cysteine and glycine rich protein 1 (CSRP1) in the occurrence and development of diffuse large B lymphoma (DLBCL) has not been reported. We conducted a comprehensive analysis of the potential value of CSRP1 in DLBCL using multiple independent datasets and analytical methods, including differential expression analysis, pathway analysis and immune analysis. Then, cell experiments were used to confirm the regulatory role of CSRP1 in DLBCL. The analysis results show that in DLBCL patients, the expression of CSRP1 is significantly upregulated. In addition, CSRP1 is associated with the cell cycle related pathways and immune microenvironment in DLBCL. Cell experiments confirmed that CSRP1 can significantly regulate the apoptosis and cycle progression of DLBCL cells. In conclusion, CSRP1 may be a potential therapeutic target for DLBCL patients, our research provides a theoretical basis for improving the clinical treatment of DLBCL. Introduction The incidence rate of diffuse large B-cell lymphoma (DLBCL) has increased significantly in the past two decades [1–3]. Despite advances in treatment methods (including aggressive surgical interventions), some lymphoma patients exhibit resistance to treatment, while others may experience disease recurrence after treatment [4]. Therefore, a deep understanding of the molecular basis of DLBCL is crucial for designing effective treatment methods and strengthening clinical decision-making processes. The abnormal expression of cysteine and glycine rich protein 1 (CSRP1) plays an important role in the tumorigenesis of many cancers, the expression of CSRP1 is positively correlated with the progression of colorectal cancer and can promote tumor growth and migration. A higher CSRP1 is a new independent prognostic factor for colorectal cancer [5]. Immunohistochemical analysis showed abnormal expression of CSRP1 in cholangiocarcinoma and adjacent tissues [6]. In hematological tumors, CSRP1 plays an important role in the prognosis of acute myeloid leukemia patients, and CSRP1 is an independent prognostic factor for acute myeloid leukemia [7]. However, the mechanism of CSRP1 in DLBCL is still unclear. Based on these limited studies, the role of CSRP1 in the mechanism of DLBCL is not yet clear, therefore further research is needed to investigate the role of CSRP1. Therefore, in this study, we used a comprehensive bioinformatics approach to evaluate the role of CSRP1 in the DLBCL mechanism. Materials and methods Microarray data acquisition and processing To assess the expression levels of CSRP1, multiple gene expression microarray datasets were utilized, including TCGA-DLBC, GSE56315, GSE25638 and GSE83632. These datasets were sourced from The Cancer Genome Atlas (TCGA) database (https://www.cancer.gov/ccg/research/genome-sequencing/tcga) and the Gene Expression Omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/). The TCGA-DLBC dataset comprises 48 DLBCL samples, while the Genotype-Tissue Expression (GTEx) dataset (https://www.gtexportal.org) provides 929 normal tissue samples for comparison. The GSE56315 dataset, derived from the GPL570 platform, contains 55 DLBCL samples and 33 normal samples (normal B-cell subset samples). Similarly, the GSE25638 dataset, also from the GPL570 platform, includes 26 DLBCL samples and 13 matched normal samples. The GSE83632 dataset, obtained from the GPL5175 platform, consists of peripheral blood samples from 76 DLBCL patients and 87 healthy donors. Additionally, single-cell data from GSE175510 were incorporated into the analysis. Gene expression levels were evaluated using the R 4.2.1 software, which facilitated the processing and analysis of raw microarray data [8]. Differential expression and clinical features of CSRP1 in DLBCL samples DLBCL samples from the TCGA-DLBC dataset were used to identify differential expression, clinical staging and prognostic differences (grouping based on the optimal cut-off value) of CSRP1 in disease states. Differential expression subsequently validated using the GSE83632 and GSE25638 datasets. To further eliminate the differences in homology, we matched normal B cell subpopulation samples and DLBCL samples based on the GSE56315 dataset and conducted differential analysis of CSRP1 expression and clinical features. To explore CSRP1 expression at the single-cell transcriptome level in DLBCL, single-cell data from GSE175510 were analyzed [9]. The criteria for identifying differentially expressed genes were set as “P < 0.05 and Fold change > 1.50,” which served as the screening threshold. Correlation analysis of CSRP1 and DLBCL key pathogenic genes and pathway analysis TCGA data was used to analyze the differential genes (Log2 > 4 and P < 0.01) that were up-regulated in DLBCL ranking, and then used R soft 4.2.1 to analyze the GSE56315 dataset to verify the results of TCGA dataset. Utilize STRING database(https://string-db.org/)to obtain the network diagram of differential genes in DLBCL, and obtain key functional gene modules through the MCODE plug-in the Cytoscape software. Then Spearman correlation was used to analyze the correlation between DLBCL key MCODE genes and CSRP1. Metascape (https://metascape.org/gp/index.html#/main/step1) was a website for analyzing protein lists or gene, which was used to analyze the functional clustering of gene sets. P < 0.05 is considered significant [10]. Analysis of CSRP1 related protein-protein interaction network and pathway analysis We use GeneMANIA (https://genemania.org/), A protein-protein interaction network centered on CSRP1 was constructed, which included the correlation data of protein and genetic, co-expression, interactions, co-localization, pathways and protein domain similarity. Then, Gene Ontology (GO) function enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were carried out for the gene centered on CSRP1 constructed by GeneMANIA [11]. Correlation analysis between CSRP1, immune checkpoints and immune characteristic genes The correlation between immune checkpoints and immune characteristic genes in CSRP1 and TCGA-DLBC was studied through Spearman correlation analysis. A value with p < 0.05 is considered statistically significant, and absolute value of the correlation coefficient is close to 1, indicating a stronger correlation [12]. Construction of cell model for CSRP1 interference OCI-LY1 cells are sourced from the cell bank of The First Affiliated Hospital of Anhui Medical University in Anhui Province. OCI-LY1 cells are cell lines of germinal center B-cell-like (GCB) type DLBCL. We constructed three siRNAs (OCI-LY1 + CSRP1-siRNA1(171), OCI-LY1 + CSRP1-siRNA2(262) and OCI-LY1 + CSRP1-siRNA3(497)) based on the CSRP1 sequence. After transfection into OCI-LY1 cells, the interference effect of these three siRNAs on CSRP1 was verified through quantitative real-time PCR. A random sequence was used as the control group (OCI-LY1 + NC-siRNA), and the cell model of CSRP1 interference was constructed by selecting one of the three siRNAs with the best interference effect and significant interference effect. Each group had three replicates, and an ordinary one-way ANOVA was performed for multi-group comparative analysis. p < 0.05 was considered significant [13]. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns P ≥ 0.05. Apoptosis experiment Blank control group cells (blank control refers to OCI-LY1 cell samples without any fluorescent markers added—this control was used to set the background signal of the flow cytometer and adjust the instrument settings to eliminate spontaneous fluorescence and other non-specific signals), OCI-LY1 + NC-siRNA group cells, and OCI-LY1 + CSRP1-siRNA3(497) group cells (1 × 106 cells/time) were harvested and washed with pre-cooled PBS. Then we resuspended the cells using 1 ml 1X Binding Buffer and achieved a density of 1 × 106 cells/ml in the tube. Then we added 5uL Annexin V-FITC to the tube and gently mix for 10 min at room temperature and in dark conditions. Finally, we add 5uL PI to the tube was incubated at room temperature and in dark conditions for 5 min, and then detected by flow cytometry within 1 h. After counting total cells and apoptotic cells, we analyzed three independent repeated data and plotted them. If the data follows a normal distribution, Student’s t-test (unpaired) was used to check the significance of the statistical mean between the two groups. Otherwise, a non-parametric test (such as Mann Whitney U test) will be used to perform a significance test between two groups. P < 0.05 was considered significant [14]. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns P ≥ 0.05. Cell cycle experiment After corresponding culture stimulation, OCI-LY1 cells and culture medium will be transferred to centrifuge tubes. The centrifuge tubes will be centrifuged at 4℃ for 5 min (1000 rpm) and the supernatant will be removed, 3 ml pre-cooled phosphate buffer saline (PBS) was added to a centrifuge tube to resuspend cells. The centrifuge tube was centrifuged at 4℃ for 5 min (1000 rpm) and the supernatant was removed, pre-cooled 75% alcohol was added to a centrifuge tube to resuspend and fix the cells, and placed in a refrigerator at 4℃ for overnight fixation. Subsequently, pre-cooled PBS was added to the centrifuge tube and washed three times. The centrifuge tube was centrifuged at 4℃ for 5 min (1000 rpm) before removing the supernatant. Finally, Propidine iodide (PI) staining solution was added and stained in the dark at 37℃ for 30 min before flow cytometry detection. We analyzed three independent repeated data and plotted them, if the data follows a normal distribution, Student’s t-test (unpaired) was used to check the significance of the statistical mean between the two groups. Otherwise, a non-parametric test (such as Mann Whitney U test) will be used to perform a significance test between two groups. P < 0.05 was considered significant [15]. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns P ≥ 0.05. Result Expression and clinical features of CSRP1 in DLBCL samples This study’s workflow is shown in Fig. 1. We further analyzed the gene expression level of CSRP1 in DLBCL using the TCGA-DLBC dataset. According to this result, CSRP1 is upregulated in DLBCL patients (Fig. 2 A). Meanwhile, based on the staging analysis of DLBCL patients, it was found that as the DLBCL staging increased, the expression level of CSRP1 also increased (Fig. 2B). Overall survival analysis showed that high expression of CSRP1 represents poor prognosis (P < 0.05, HR = 6.342) (Fig. 2 C). We validated the differential expression of CSRP1 in GSE83632 and GSE25638 (Fig. 3A-B) and further analyzed it in the GSE56315 dataset (Fig. 3 C). Meanwhile, compared to normal B cell subpopulation samples, both ABC and GCB samples showed significant upregulation (P < 0.05), with ABC samples showing a more pronounced upregulation than GCB samples (Fig. 3D).Fig. 1The flowchart of the present study designFig. 2Expression and clinical features of CSRP1 in DLBCL samples. (A) We further analyzed the gene expression level of CSRP1 in DLBCL using the TCGA-DLBC dataset. According to this result, CSRP1 is upregulated in DLBCL patients. (B) Based on the staging analysis of DLBCL patients, it was found that as the DLBCL staging increased, the expression level of CSRP1 also increased. (C) Overall survival analysis showed that high expression of CSRP1 represents poor prognosis (C)Fig. 3Independent dataset to validate CSRP1 in DLBCL. (A-C) We validated the differential expression of CSRP1 in GSE25638, GSE83632 and GSE56315 (D) Compared to normal B cell subpopulation samples, both ABC and GCB samples showed significant upregulation (P < 0.05), with ABC samples showing a more pronounced upregulation than GCB samples. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns P ≥ 0.05 The flowchart of the present study design Expression and clinical features of CSRP1 in DLBCL samples. (A) We further analyzed the gene expression level of CSRP1 in DLBCL using the TCGA-DLBC dataset. According to this result, CSRP1 is upregulated in DLBCL patients. (B) Based on the staging analysis of DLBCL patients, it was found that as the DLBCL staging increased, the expression level of CSRP1 also increased. (C) Overall survival analysis showed that high expression of CSRP1 represents poor prognosis (C) Independent dataset to validate CSRP1 in DLBCL. (A-C) We validated the differential expression of CSRP1 in GSE25638, GSE83632 and GSE56315 (D) Compared to normal B cell subpopulation samples, both ABC and GCB samples showed significant upregulation (P < 0.05), with ABC samples showing a more pronounced upregulation than GCB samples. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns P ≥ 0.05 Expression of CSRP1 in DLBCL scRNA-seq DLBCL’s independent dataset was subjected to single cell sequencing to explore the correlation between immune cell distribution and CSRP1 expression level at the single cell level. Based on the single-cell sequencing dataset (GSE175510), all cells were classified into 4 types, including Malignant cells, Mono/Macro, B cells and CD8T cells (Fig. 4A). High levels of CSRP1 expression were found in two cells (CD8T cells and Malignant cells) (Fig. 4B).Fig. 4CSRP1 is a potential biomarker for DLBCL scRNA-seq (A) Based on the single-cell sequencing dataset (GSE175510), all cells were classified into 4 types, including Malignant cells, Mono/Macro, B cells and CD8T cells (B) High levels of CSRP1 expression were found in two cells (CD8T cells and Malignant cells) CSRP1 is a potential biomarker for DLBCL scRNA-seq (A) Based on the single-cell sequencing dataset (GSE175510), all cells were classified into 4 types, including Malignant cells, Mono/Macro, B cells and CD8T cells (B) High levels of CSRP1 expression were found in two cells (CD8T cells and Malignant cells) Correlation analysis of CSRP1 and DLBCL key pathogenic genes and pathway analysis We analyzed the differential genes (Log2 > 4 and P < 0.01) that were up-regulated in DLBCL ranking by TCGA-DLBC dataset, and then verified them by GSE56315, and the key functional gene modules were obtained through the MCODE plug-in in the Cytoscape software. We took the module with the highest total score (Score > 14), and a total of 16 key MCODE genes were obtained and spearman correlation analysis was used to analyze the correlation between CSRP1 and the 16 key MCODE genes of DLBCL (Fig. 5 A), the results showed that all these genes had a positive correlation with CSRP1, indicating that the upregulation of the key genes of this module promoted the occurrence and development of DLBCL, and there was a significant positive correlation between CSRP1 and 5 key MCODE genes (CCNB1, CDK1, CENPW, KPNA2, UBE2C). The GO terms for five gene significance include Cyclin B1-CDK1 complex, Kinetochore and Viral life cycle (Fig. 5B). Suggest that CSRP1 may play a role in DLBCL through the cell cycle pathway.Fig. 5Correlation analysis of CSRP1 and DLBCL key pathogenic genes. (A) A total of 16 key MCODE genes were obtained and spearman correlation analysis was used to analyze the correlation between CSRP1. (B) The GO terms for five gene significance include Cyclin B1-CDK1 complex, Kinetochore and Viral life cycle. (C) GeneMANIA was used to build a PPI network of 21 genes centered on CSRP1. (D) GO function enrichment and KEGG pathway analysis were carried out for these 21 genes Correlation analysis of CSRP1 and DLBCL key pathogenic genes. (A) A total of 16 key MCODE genes were obtained and spearman correlation analysis was used to analyze the correlation between CSRP1. (B) The GO terms for five gene significance include Cyclin B1-CDK1 complex, Kinetochore and Viral life cycle. (C) GeneMANIA was used to build a PPI network of 21 genes centered on CSRP1. (D) GO function enrichment and KEGG pathway analysis were carried out for these 21 genes Analysis of CSRP1 related protein-protein interaction network and pathway analysis GeneMANIA was used to build a protein-protein interaction network of 21 genes centered on CSRP1 (Fig. 5C). GO function enrichment and KEGG pathway analysis were carried out for these 21 genes. Significantly rich GO terms include actin cytoskeleton, contractile fiber, actin binding and focal adhesion, while the significantly rich KEGG pathway is vascular smooth muscle contraction (Fig. 5D). Correlation analysis between CSRP1, immune checkpoints and immune characteristic genes The Spearman correlation analysis results showed that CSRP1 was correlated with 4 immune checkpoints (CD276, CD40, CD80 and TGFB1) significantly positively (P < 0.05) (Fig. 6 A). CD40, CD80 and TGFB1 have been reported to be associated with macrophages [16, 17]. CSRP1 with macrophage M1immune characteristic genes (NOS2) and macrophage M2 immune characteristic genes (CD163 and MS4A4A) significantly positively (P < 0.05) (Fig. 6B). CSRP1 therefore is potential immunotherapy target and a key immune gene for DLBCL.Fig. 6Correlation analysis between CSRP1, immune checkpoints and immune characteristic genes. (A) The Spearman correlation analysis results showed that CSRP1 was correlated with 4 immune checkpoints significantly positively (P < 0.05) (B) CSRP1 with macrophage M1 immune characteristic genes (NOS2) and macrophage M2 immune characteristic genes (CD163 and MS4A4A) significantly positively (P < 0.05) Correlation analysis between CSRP1, immune checkpoints and immune characteristic genes. (A) The Spearman correlation analysis results showed that CSRP1 was correlated with 4 immune checkpoints significantly positively (P < 0.05) (B) CSRP1 with macrophage M1 immune characteristic genes (NOS2) and macrophage M2 immune characteristic genes (CD163 and MS4A4A) significantly positively (P < 0.05) Construction of cell model for CSRP1 interference The quantitative real-time PCR results showed that compared with the control group, the cells transfected with OCI-LY1 + CSRP1-siRNA3(497) had the most significant interference effect on CSRP1 (Fig. 7A). Therefore, the cells transfected with OCI-LY1 + CSRP1-siRNA3(497) were selected as the cell model for CSRP1 interference.Fig. 7Construction of cell model for CSRP1 interference and apoptosis experiments of OCI-LY1 cells. (A) The quantitative real-time PCR results showed that the cells transfected with OCI-LY1 + CSRP1-siRNA3(497) had the most significant interference effect on CSRP1. (B-C) The results of cell apoptosis showed the apoptosis rate of the OCI-LY1 + CSRP1-siRNA3(497) group cells was significantly increased Construction of cell model for CSRP1 interference and apoptosis experiments of OCI-LY1 cells. (A) The quantitative real-time PCR results showed that the cells transfected with OCI-LY1 + CSRP1-siRNA3(497) had the most significant interference effect on CSRP1. (B-C) The results of cell apoptosis showed the apoptosis rate of the OCI-LY1 + CSRP1-siRNA3(497) group cells was significantly increased Apoptosis experiments Student’s t-test (unpaired) was used to check the significance of the statistical mean between the two groups. The results of cell apoptosis showed that compared with the OCI-LY1 + NC-siRNA group cells, the apoptosis rate of the OCI-LY1 + CSRP1-siRNA3(497) group cells was significantly increased (Fig. 7B-C). Therefore, the low expression of CSRP1 significantly promoted apoptosis of OCI-LY1 cells. Cell cycle experiments Student’s t-test (unpaired) was used to check the significance of the statistical mean between the two groups. The cell cycle results showed that compared with the OCI-LY1 + NC-siRNA group, the OCI-LY1 + CSRP1-siRNA3(497) group showed a significant increase in G1 phase cells and a significant decrease in G2 phase cells (Fig. 8A-B). Therefore, the low expression of CSRP1 significantly inhibited the progression of OCI-LY1 cell cycle.Fig. 8DLBCL cell cycle experiment (A-B) The cell cycle results showed the OCI-LY1 + CSRP1-siRNA3(497) group showed a significant increase in G1 phase cells and a significant decrease in G2 phase cells DLBCL cell cycle experiment (A-B) The cell cycle results showed the OCI-LY1 + CSRP1-siRNA3(497) group showed a significant increase in G1 phase cells and a significant decrease in G2 phase cells Discussion The main purpose of this article is to explore the mechanism of CSRP1 in DLBCL. Firstly, we analyzed the significant high expression of CSRP1 in the lymphoma group based on the TCGA-DLBC dataset and validated it in three other independent datasets. Based on the single-cell sequencing dataset (GSE175510), high levels of CSRP1 expression were found in two cells (CD8T cells and Malignant cells). Secondly, the transcriptional mechanism analysis shows that CSRP1 may play a role in DLBCL through the Cyclin B1-CDK1 complex pathway. CDK1 is the only CDK strictly required for cell cycle [18], Cyclin B1-CDK1 complex is the main mitotic promoting factor [19]. Our results found a significant positive correlation between CSRP1 and CDK1, suggesting that CSRP1 may regulate the cycle progression of DLBCL cells through the Cyclin B1-CDK1 complex pathway. Subsequently, our DLBCL cell experiments confirmed that low expression of CSRP1 can significantly inhibit the cycle progression of DLBCL. Therefore, CSRP1 may therefore be potential therapeutic target and a key gene for DLBCL patients. Our research provides a theoretical basis for improving DLBCL’s clinical treatment. The Spearman correlation analysis results showed that CSRP1 was correlated with 4 immune checkpoints (CD276, CD40, CD80 and TGFB1) significantly positively (P < 0.05). Tumor cells induce immune tolerance and immune escape in the body by upregulating the expression of immune checkpoint molecules, promoting the growth and metastasis of tumor cells [20]. Therefore, CSRP1 may cause an imbalance in the immune microenvironment of DLBCL cells by upregulating the expression of immune checkpoint molecules. CD40, CD80, and TGFB1 have been reported to be associated with macrophages [17]. CSRP1 with macrophage M1 immune characteristic genes (NOS2) and macrophage M2 immune characteristic genes (CD163 and MS4A4A) significantly positively(P < 0.05). CSRP1 therefore is potential immunotherapy target and a key immune gene for DLBCL. However, the specific mechanism needs to be further elucidated. There inevitably are several limitations to this study. One limitation of this article is the small sample size, which requires further validation using a multi-center large sample queue. This study comes from public databases mainly and is retrospective. The number of clinical information datasets that is available is limited, therefore the clinical parameters that are analyzed in this study are not comprehensive. Secondly, although CSRP1’s function has been confirmed by DLBCL cells, its downstream and upstream mechanisms are not yet clear, and further experiments are needed to study its mechanism, which will be the research we need in the future. In addition, immune related mechanisms are at the level of “correlation analysis” and require further functional validation. Our study only used one DLBCL cell line for research, and the experimental results of a single cancer cell line may have limitations and cannot fully reflect the heterogeneity and complexity of cancer cells. In order to gain a more accurate and comprehensive understanding of the biological characteristics of CSRP1 in cancer cells, biological experiments on multiple cancer cell lines will be one of our next research goals. Conclusion The main purpose of this article is to explore the mechanism of CSRP1 in DLBCL. Firstly, we analyzed the significant high expression of CSRP1 in the DLBCL group based on the TCGA-DLBC dataset and validated it in three other independent datasets. Based on the single-cell sequencing dataset (GSE175510), high levels of CSRP1 expression were found in two cells (CD8T cells and Malignant cells). Secondly, the transcriptional mechanism analysis shows that CSRP1 may play a role in DLBCL through the cell cycle pathway. The cell cycle experiment results showed that compared with the OCI-LY1 + NC-siRNA group, the OCI-LY1 + CSRP1-siRNA3(497) group showed a significant increase in G1 phase cells, a increase in S phase cells, and a significant decrease in G2 phase cells. Further immune analysis results showed that CSRP1 is associated with polarization of macrophages in DLBCL. Therefore, CSRP1 may therefore be potential therapeutic target and a key gene for DLBCL patients. Our research provides a theoretical basis for improving DLBCL’s clinical treatment.