ISM2 as a prognostic biomarker and mediator of immune infiltration in colorectal cancer: evidence from bioinformatics and experimental analysis

Introduction

In the United States, the incidence of colorectal cancer (CRC) is approximately 38.7 per 100,000 individuals (1). Reports indicate a higher prevalence of CRC in men compared to women, although the total number of cases among women remains substantial. Annually, CRC accounts for a significant number of new diagnoses and fatalities. For instance, in 2020, approximately 147,950 new CRC cases were reported in the United States, with a higher incidence in men than women, and an estimated 53,200 deaths were attributed to the disease (2,3).

In China, the incidence of CRC is notably high, with an increasing trend toward earlier onset in younger populations. This underscores the critical need for effective prevention and treatment strategies. Current research efforts are focused on identifying biomarkers associated with CRC to inform clinical decision-making and optimize treatment plans (4).For example, identifying biomarkers linked to tumor metastasis could enable clinicians to personalize comprehensive treatment approaches, including chemotherapy, radiotherapy, or surgical interventions, based on the expression profiles of these markers in CRC tissues. Consequently, the search for reliable biomarkers that can accurately predict CRC prognosis has emerged as a key area of research (5).

Isthmin 2 (ISM2), a member of the Isthmin protein family, encodes proteins containing thrombospondin type 1 repeat (TSR1) and adhesion-associated domain in MUC4 and other proteins (AMOP) domains (6). Studies have demonstrated its expression in various tissues, including elevated levels in choriocarcinoma and reduced concentrations in the serum of patients with preeclampsia. However, its role in the immune system remains to be fully elucidated (7). ISM2 is closely linked to tumor biology, with abnormal expression observed in certain tumor cell lines compared to normal tissue cells. This aberrant expression influences tumor cell proliferation and contributes to shaping the tumor microenvironment (TME) by regulating cytokine secretion, promoting tumor angiogenesis, and thereby facilitating tumor growth and metastasis (8,9).

The relationship between ISM2 and CRC was investigated using data from The Cancer Genome Atlas (TCGA), focusing on its potential role as a prognostic biomarker. Analysis of RNA sequencing (RNAseq) data from TCGA revealed differential ISM2 expression between CRC and normal tissues. Further stratification revealed variations in ISM2 expression across different patient groups and its correlation with prognosis. The biological functions of ISM2 were examined using Linked Omics and STRING databases (10). Additionally, the relationship between ISM2 expression and immune infiltration as a potential approach to predicting CRC prognosis was examined. To validate bioinformatics findings, in vitro experiments were conducted, confirming that ISM2 serves as a promising diagnostic and prognostic marker for CRC. We present this article in accordance with the REMARK and MDAR reporting checklists (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-295/rc).

Methods

RNA-seq data acquisition and pan-cancer analysis

Transcriptome RNA sequencing data and corresponding clinical information from 698 individuals diagnosed with colon and rectum adenocarcinoma (COADREAD) were retrieved from TCGA database (https://portal.gdc.cancer.gov/). The dataset included 647 tumor tissue samples and 51 tumor-adjacent normal tissue samples. Associations between ISM2 expression and clinicopathological features, such as pathological stage (I and II vs. III and IV), T stage (T1 and T2 vs. T3 and T4), N stage (N0 vs. N1 and N2), and M stage (M0 vs. M1), were analyzed using non-parametric tests. Paired samples (n=50) were selected based on matching clinical stage (I–IV) and pathological type (adenocarcinoma), with strict exclusion of samples lacking survival data or with ambiguous pathological diagnosis.

Data processing and visualization were conducted using R software (version 4.2.1). The stats and car packages were utilized for statistical evaluations, with methods selected based on dataset characteristics. Analyses were conducted only when relevant statistical assumptions were met. Data visualization was carried out using the ggplot2 package. Additionally, ISM2 expression profiles across various cancers and corresponding normal tissues were downloaded from the TIMER database (http://timer.cistrome.org/). A significance threshold of P<0.05 was applied for all analyses. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Kaplan-Meier curve analysis

Data processing and visualization were carried out using R software (version 4.2.1). Survival curves were generated through the Kaplan-Meier method, and their statistical significance was assessed using the log-rank test. The results were visualized with the R package ’survminer’.

ISM2 protein expression

The protein expression of ISM2 in CRC and normal tissues was analyzed using data from the Human Protein Atlas database (https://www.proteinatlas.org/). The expression of ISM2 in CRC cell lines was examined through the Cancer Cell Line Encyclopedia (CCLE) (https://sites.broadinstitute.org/ccle).

Immune infiltration analysis

Immune infiltration matrix data were obtained from the Xiantao Academic Cloud Database (https://www.xiantao.love). The correlation between the primary variables in the preprocessed dataset and the immune infiltration matrix was analyzed using R software (version 4.2.1). The results were visualized with the ggplot2 package.

LinkedOmics database analysis

The LinkedOmics database (http://www.linkedomics) was used to identify differentially expressed genes (DEGs) associated with ISM2 in the TCGA COADREAD group through the “Link Finder” module. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses were performed using the “Functions” module.

Search Tool for the Retrieval of Interacting Genes (STRING) database analysis

The STRING online database (http://string-db.org) was used to predict the protein-protein interaction (PPI) network of ten hub genes associated with ISM2. Interactions with a combined score >0.4 were considered statistically significant.

Cell culture and transfection

Human CRC cell lines (SW1116, LOVO, and RKO) and normal human colonic epithelial cells (NCM460) were obtained from the American Type Culture Collection. Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin. Cultures were maintained in sterile petri dishes and passaged every three days using 0.25% trypsin. SW1116 cells were seeded into 6-well plates 24 hours prior to transfection.

Small interfering RNA (siRNA) targeting ISM2 and a control siRNA were synthesized. The sequences of ISM2 siRNAs were as follows: 5'-GCAGGCAGAGCTACACCAA-3', antisense: 5'-TTGGTGTAGCTCTGCCTGC-3'; ISM2 siRNA#2 sense: 5'-CATGGATGTTGGACTGTCA-3', antisense: 5'-TGACAGTCCAACATCCATG-3'; ISM2 siRNA#3 sense: 5'-GCGACTTCCTAATCAAGTA-3', antisense: 5'-TACTTGATTAGGAAGTCGC-3'. Cells were transfected with plasmids or siRNA using Lipofectamine^® 2000 (Invitrogen, California, USA) according to the manufacturer’s protocol.

Western blotting

Total cellular proteins were extracted using a protein lysis buffer (Biyuntian, Shanghai, China). Proteins were separated on 10% SDS-PAGE and subsequently transferred onto a polyvinylidene difluoride (PVDF) membrane (Millipore, Massachusetts, USA). The membrane was blocked with 5% skim milk for 2 hours at room temperature, followed by overnight incubation at 4 ℃ with the respective primary antibodies. After washing, the membrane was incubated with diluted secondary antibodies for 2 hours. Enhanced chemiluminescent reagent was applied evenly to the PVDF membrane for signal detection.

The primary antibodies used were ISM2 (1:500, Proteintech, Wuhan, China), β-catenin (1:10,000, Proteintech, China), c-Myc (1:1,000, ABclonal, Wuhan, China) and GAPDH (1:5,000, Santa Cruz Biotechnology, California, USA), with GAPDH serving as the loading control. Band intensities were quantified using ImageJ software.

Cell Counting Kit-8 (CCK-8) assay

Transfected SW1116 cells were seeded in 96-well plates at a density of approximately 3,000 cells per well. After incubation for 12, 24, and 36 hours, 10 µL of CCK-8 reagent was added to each well, followed by a 1-hour incubation at 37 ℃. The optical density of each well was measured at a wavelength of 450 nm using a microplate reader.

Wound healing assay

Approximately 2×10⁵ transfected cells were seeded into six-well plates and incubated at 37 ℃ for 48 hours until reaching 100% confluence. An artificial wound was created by scratching the cell monolayer with a sterile 200 µL pipette tip. The cells were then washed with 1× phosphate-buffered saline (PBS) to remove any floating debris. Images of the scratched area were captured using an inverted optical microscope at 100× magnification at 0 and 36 hours post-scratch. The cell migration rate was calculated using ImageJ software.

Cell invasion assay

A 24-well Transwell assay with an 8 µm polycarbonate membrane was used to assess cell migration and invasion. Approximately 1×10⁵ cells were suspended in 200 µL of serum-free medium and seeded into the upper chamber of the Transwell assay. The lower chamber was filled with 600 µL of 10% DMEM to serve as a chemoattractant. After 24 hours of incubation, the cells that migrated or invaded through the membrane were fixed with 4% paraformaldehyde for 30 minutes and subsequently stained with 0.1% crystal violet for 30 minutes. Images of the stained cells were captured using an inverted research microscope, and cells were counted in no fewer than three random fields at 100× magnification.

Statistical analysis

ISM2 expression in CRC tissues was analyzed using univariate analysis. The Chi-squared test and Fisher’s exact test were applied to assess significant differences between categorical variables. Kaplan-Meier survival curves and Cox regression analysis were conducted to assess the prognostic potential of ISM2 in predicting overall survival (OS). In statistical analyses, the Benjamini-Hochberg false discovery rate (FDR) method was used for multiple testing correction (FDR <0.05). Clinical confounders [age, gender, tumor-node-metastasis (TNM) stage] were adjusted as covariates in Cox regression, with multivariate progress free interval (PFI) analysis accounting for age (>65 vs. ≤65 years), M stage, and pathological stage (independent risk factors, P<0.05). Analyses were conducted using R 4.2.1, with key results validated by Student’s t-test/Chi-squared test and effect sizes [odds ratio (OR), hazard ratio (HR)] reported with 95% confidence intervals (CIs). GraphPad Prism (version 8) was used for statistical analysis, with a significance threshold set at P<0.05. Graphical illustrations were generated using Adobe Illustrator CC, Adobe Photoshop CC, and ImageJ software.

Results

Differential expression of ISM2 in CRC versus normal tissues

Protein levels of ISM2 were analyzed across several public databases to examine its differential expression between CRC and normal tissues. Data from the TIMER database using TCGA results revealed that ISM2 exhibited differential expression in 15 of 33 tumor types. Significant differences in expression were observed in breast invasive carcinoma (BRCA, P<0.001), cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC, P<0.05), cholangiocarcinoma (CHOL, P<0.01), colon adenocarcinoma (COAD, P<0.001), glioblastoma multiforme (GBM, P<0.05), head and neck squamous cell carcinoma (HNSC, P<0.001), kidney renal clear cell carcinoma (KIRC, P<0.001), kidney renal papillary cell carcinoma (KIRP, P<0.001), liver hepatocellular carcinoma (LIHC, P<0.001), lung adenocarcinoma (LUAD, P<0.001), lung squamous cell carcinoma (LUSC, P<0.001), pheochromocytoma and paraganglioma (PCPG, P<0.05), prostate adenocarcinoma (PRAD, P<0.001), rectal adenocarcinoma (READ, P<0.001), and uterine corpus endometrial carcinoma (UCEC, P<0.001) (Figure 1A). These findings highlight the involvement of ISM2 in tumorigenesis and progression across various cancers.

Figure 1 ISM2 is upregulated in COAD. (A) Differential expression of ISM2 in tumor and normal tissues in TCGA pan-cancer data. (B) Difference in ISM2 expression between patients with colorectal cancer and corresponding normal tissues in TCGA. n=50 paired samples. (C) Difference in ISM2 expression between patients with colorectal cancer and normal individuals. n=647 CRC samples and n=50 normal samples from TCGA. (D) TCGA database screening for significant differences in differential genes ISM2 in the COAD dataset. The criteria for defining significant differenced genes were: |log₂ (fold change)| >1 and P value <0.05. (E) ROC curve showing the efficiency of ISM2 in distinguishing colorectal cancer tissues from normal tissues. (F,G) Representative immunohistochemistry images of ISM2 in normal and CRC tissues (https://www.proteinatlas.org/ENSG00000100593-ISM2/cancer/colorectal+cancer#img). Normal: HPA039314, CRC: HPA039314; magnification ×20. *, P<0.05; **, P<0.01; ***, P<0.001. AUC, area under the curve; CI, confidence interval; COAD, colon adenocarcinoma; CRC, colorectal cancer; FPR, false positive rate; ROC, receiver operating characteristic; sig, significant; TCGA, The Cancer Genome Atlas; TPM, transcripts per million; TPR, true positive rate.

To further confirm the association between ISM2 expression and CRC, gene expression data from 647 CRC patients and 51 healthy individuals were retrieved from the TCGA database (Table S1). A paired analysis of 50 patients with CRC and 50 healthy individuals demonstrated a significant increase in ISM2 expression in CRC tissues (P<0.001, Figure 1B). ISM2 expression was also significantly higher in CRC patients compared to healthy individuals (P<0.001, Figure 1C). Both comparative approaches confirm the correlation between ISM2 expression and CRC.

Analysis of DEGs in the TCGA database revealed significant upregulation of ISM2 in CRC tissues (Figure 1D). The discriminatory power of ISM2 in differentiating tumor tissues from non-tumor tissues was assessed using a receiver operating characteristic (ROC) curve. The area under the curve for ISM2 was 0.755 [95% confidence interval (CI): 0.709–0.800], indicating its potential as a diagnostic marker (Figure 1E).

Immunohistochemical staining images from The Human Protein Atlas database further substantiated these findings, showing negligible ISM2 expression in normal peritumoral tissues, while significantly elevated levels were observed in CRC tumor tissues (Figure 1F,1G). Collectively, these results demonstrate that ISM2 is a reliable biomarker for CRC diagnosis and prognosis.

Relationship between ISM2 expression and clinicopathological features in CRC

To further explore the role of ISM2 in CRC, the relationship between ISM2 expression and clinicopathological characteristics was analyzed. Baseline data from 644 patients with CRC in the TCGA database were statistically assessed (Table 1). Patients were categorized into low-expression and high-expression groups based on the median ISM2 mRNA expression level, with 322 cases in each group. No statistically significant differences were observed between the two groups in terms of gender (P=0.07) or age (P=0.34). However, significant differences were identified in TNM staging [T stage (P=0.004), N stage (P<0.001), M stage (P<0.001)], pathological staging (P<0.001), and histological type (P<0.001).

Logistic regression analysis was conducted to further explore the relationship between ISM2 expression and clinicopathological characteristics (Table 2). Consistent with the baseline analysis, no significant differences were observed for gender (P=0.07), age (P=0.34), or T stage (P=0.21). However, significant associations were detected for N1 & N2 stages compared to N0 (OR =2.055, 95% CI: 1.494–2.827, P<0.001), M1 stage compared to M0 (OR =2.260, 95% CI: 1.406–3.635, P<0.001), and between stage III and IV versus stage I and II (OR =2.098, 95% CI: 1.520–2.894, P<0.001).

ISM2 expression across various pathological stages was analyzed to confirm its correlation with CRC progression. ISM2 expression was significantly elevated in T-staged CRC samples compared to normal tissues (P<0.001), with significant differences observed among T3, T4, and T1 stages (P<0.05) (Figure 2A). Similarly, ISM2 expression was significantly higher in N-staged CRC samples compared to normal tissues, with notable differences between N0 and N1 (P<0.01) and N0 and N2 stages (P<0.001) (Figure 2B). For M stages, ISM2 expression was significantly higher in M-staged tissues compared to normal tissues, with significant differences observed between M0 and M1 stages (P<0.01) (Figure 2C). Significant differences were identified in ISM2 expression between CRC samples and normal tissues across various pathological stages (P<0.001), including between Stage IV and Stage I (P<0.001) (Figure 2D). These findings collectively indicate that increased ISM2 expression is closely associated with the malignant progression of CRC.

Figure 2 Relationship between ISM2 expression and clinicopathological features of colorectal cancer. Differences in ISM2 expression across T stages (A), N stages (B), M stages (C), and pathological stages (D) are shown, with statistical significance denoted. *, P<0.05; **, P<0.01; ***, P<0.001. M, metastasis; N, node; T, tumor; TPM, transcripts per million.

Role of ISM2 expression in the survival of patients with CRC

The correlation between ISM2 expression and survival outcomes in patients with CRC was analyzed. Kaplan-Meier survival curves, derived from the TCGA dataset, demonstrated that patients with high ISM2 expression had significantly lower OS compared to those with low ISM2 expression (HR =1.59, 95% CI: 1.11–2.27, P=0.01; Figure 3A). A similar trend was observed in the progression-free interval (PFI) analysis, where high ISM2 expression was associated with poorer outcomes (HR =1.52, 95% CI: 1.10–2.10, P=0.01; Figure 3B).

Figure 3 Survival analysis of ISM2 in colorectal cancer. (A) Kaplan-Meier curve for OS in TCGA COADREAD cohort (n=698). Patients with high ISM2 expression showed significantly shorter OS (HR =1.59, 95% CI: 1.11–2.27, P=0.01, log-rank test). (B) PFI analysis demonstrating poorer outcomes in high ISM2 group (HR =1.52, 95% CI: 1.10–2.10, P=0.01, log-rank test). CI, confidence interval; HR, hazard ratio; OS, overall survival; PFI, progress free interval; TCGA, The Cancer Genome Atlas.

Univariate and multivariate Cox regression analyses based on PFI were conducted to identify independent prognostic factors influencing survival in patients with CRC. Univariate analysis revealed significant associations between PFI and several clinicopathological factors, including T stage, N stage, M stage, pathological stage, lymphatic invasion, and neural invasion (Table 3). However, multivariate analysis indicated that only pathological M stage (M0 vs. M1) remained significantly associated with PFI (OR =5.285, 95% CI: 2.317–12.051, P<0.001), while ISM2 lost significance as an independent predictor.

These findings indicate that elevated ISM2 expression is associated with poorer survival outcomes and serve as a prognostic marker for CRC, though its association is confounded by metastatic status (M stage). ISM2 may act as a prognostic indicator in the context of metastatic progression rather than an independent factor.

Correlation between ISM2 expression and immune cell infiltration

The correlation between ISM2 expression and immune infiltration across 24 types of immune cells was assessed using the single-sample gene set enrichment analysis (ssGSEA) algorithm. The results were visualized as depicted in Figure 4A, where the size of each circle represents the strength of the correlation between ISM2 expression and the enrichment of a specific immune cell type, and the color indicates the corresponding P value. The distance of each circle from the baseline and its size further reflect the magnitude of the correlation.

Figure 4 Relationship between ISM2 expression and immune invasion in colorectal cancer microenvironment. (A) Correlation between the relative abundance of 24 immune cells and ISM2 expression. The size of the point represents the absolute value of Spearman’s correlation coefficient R (ssGSEA analysis). (B) Correlation between ISM2 expression level and the infiltration level of aDC, CD8⁺ T cells, cytotoxic cells and Th2 cells. (C-F) Correlation of ISM2 expression with aDC cells, CD8⁺ T cells, cytotoxic cells, and Th2 cells. R=−0.223, P<0.001 (aDC); R=−0.244, P<0.001 (CD8⁺ T cells); R=−0.212, P<0.001 (cytotoxic cells); R=−0.368, P<0.001 (Th2 cells) (ssGSEA analysis). (G) Correlation analysis of ISM2 expression with multiple immune cell subsets. The correlation coefficients (R) and corresponding P values are displayed (CIBERSORT analysis). (H,I) Scatter plot illustrating the correlation between ISM2 expression [log₂ (TPM +1)] and infiltration level of T cells CD8 (R=−0.087, P=0.03) and dendritic cells activated (R=−0.143, P<0.001). Spearman’s correlation and Wilcoxon test were used for analysis. (J) Correlation analysis of ISM2 expression with immune-related pathways and molecules (ssGSEA analysis). ns, no significance; *, P<0.05; **, P<0.01; ***, P<0.001. APC, antigen-presenting cell; aDC, activated dendritic cell; CCR, C-C chemokine receptor; HLA, human leukocyte antigen; iDC, immature dendritic cell; IFN, interferon; MHC, major histocompatibility complex; NK, natural killer; pDC, plasmacytoid dendritic cell; ssGSEA, single-sample gene set enrichment analysis; TFH, T follicular helper cell; TPM, transcripts per million.

Further statistical analysis indicated that the proportions of Th2 cells, CD8⁺ T cells, aDCs, and Cytotoxic cells were significantly lower in the ISM2 high-expression group compared to the low-expression group (P<0.001) (Figure 4B). Significant negative correlations were observed between ISM2 expression and the immune infiltration of Th2 cells (R=−0.368, P<0.001), CD8⁺ T cells (R=-0.244, P<0.001), activated dendritic cells (aDC) (R=−0.223, P<0.001), and cytotoxic cells (R=−0.212, P<0.001) (Figure 4C-4F). To continue to evaluate the relationship between ISM2 expression and immune-cell infiltration, we recalcitated the correlations of immune infiltration for 22 immune-cell types by transforming the CIBERSORT algorithm (Figure 4G). Consistently, we found that both T cells CD8 (R=−0.087, P=0.03) and Dendritic cells activated (R=−0.143, P<0.001) were negatively correlated with ISM2 expression (Figure 4H,4I). Further analysis showed that the expression of ISM2 was potentially associated with immune checkpoint-related pathways (such as APC_co_inhibition, CCR) and inflammatory regulatory pathways. Among immune-related molecules, the expression of ISM2 was correlated with the expression of HLA and inflammation-promoting MHC_class_I in a dose-dependent manner. Numerical analysis showed that its expression level may be positively correlated with the inflammation-promoting function of MHC_class_I (Figure 4J).

Functional analysis of ISM2-related genes

To further elucidate the biological roles of ISM2 in CRC, DEGs associated with ISM2 in the TCGA-COADREAD dataset were identified using the LinkFinder module on the LinkedOmics platform. As depicted in Figure 5A, 2,855 genes were positively correlated and 1,323 genes were negatively correlated with ISM2 (FDR <0.001), as determined by Spearman’s test. The top 50 positively and negatively correlated genes are depicted in Figure 5B,5C, respectively.

Figure 5 Analysis of genes and functions associated with ISM2. (A) The genes that are significantly associated with ISM2, as identified by the Spearman’s test in the COADREAD cohort from the LinkedOmics database. (B,C) Heatmaps of the top 50 genes positively and negatively associated with ISM2 in the COADREAD cohort are shown. (D-F) GO term annotation: BP, CC, MF. (G) KEGG pathway analysis. (H) Ten protein interaction networks associated with ISM2 in the STRING database. (I) GO term and KEGG pathway analysis of nine partner genes of ISM2. BP, biological process; CC, cellular component; FDR, false discovery rate; GO, Gene Ontology; KEGG, Kyoto Encyclopedia of Genes and Genomes; IgA, immunoglobulin A; MF, molecular function; NAD, nicotinamide adenine dinucleotide; NADP, nicotinamide adenine dinucleotide phosphate.

GO and KEGG analyses were conducted to explore the biological functions of ISM2-associated DEGs. GO analysis revealed that these DEGs were involved in processes such as stem cell proliferation, branching morphogenesis, cardiac morphogenesis, maintenance of cell number, WNT-mediated cell-cell signaling, and mesenchymal development (Figure 5D). They were primarily localized in collagen trimers, synaptic membranes, contractile fibers, and postsynaptic specializations (Figure 5E). Functionally, these genes contributed to structural components of muscles, transmembrane receptor protein kinase activity, β-catenin binding, and extracellular matrix structure (Figure 5F).

KEGG pathway analysis further indicated that these DEGs were significantly enriched in the WNT signaling pathway [Normalized Enrichment Score (NES) =1.82, FDR <0.001, Figure 5G] suggesting a potential role of ISM2 in modulating Wnt/β-catenin signaling. To validate this, we performed Western blot analysis in SW1116 cells transfected with ISM2 siRNA#2, which showed significant downregulation of nuclear β-catenin and c-Myc compared to NC siRNA controls (Figure S1A). Densitometric quantification of Figure S1B revealed that ISM2 knockdown reduced nuclear β-catenin expression by 42% and c-Myc by 51% compared to the control group [n=3, mean ± standard error of the mean (SEM)]. These results confirm that ISM2 depletion suppresses Wnt pathway activation at the post-translational level, aligning with the bioinformatics predictions and establishing a mechanistic link between ISM2 and Wnt-mediated tumor progression.

A PPI network analysis of ISM2 was conducted using the STRING online database. The 10 genes with the highest connectivity in the network—SORD, MRTFA, SPTLC1, ALAD, KDSR, APPBP2, MUC4, NOP53, TGFB3, and C14orf178—were identified (Figure 5H). GO and KEGG enrichment analyses of these functionally interacting genes indicated that they were predominantly localized in the cytoplasmic vesicle cavity and secretory granule lumen, involved in responses to hypoxia, oxidoreductase activity targeting the CH-OH group of donors, and functioning as nicotinamide adenine dinucleotide (NAD) or nicotinamide adenine dinucleotide phosphate (NADP) receptors. They participated in the negative regulation of proteasomal protein catabolic processes and sphincter metabolic processes, among other roles (Figure 5I).

These genes were predominantly localized in the cytoplasmic vesicle cavity and secretory granule lumen and were involved in responses to hypoxia. They exhibited oxidoreductase activity, specifically targeting CH-OH groups of donors, and functioned as receptors for NAD or NADP. Additionally, they participated in processes such as the negative regulation of proteasomal protein catabolism and sphincter metabolic regulation, among other biological functions.

ISM2 is involved in the proliferation and invasion of CRC

To further validate the role of ISM2 in the tumorigenesis and progression of CRC, ISM2 protein expression was examined in normal human colon epithelial cells and three CRC cell lines (RKO, LOVO, and SW1116). As depicted in Figure 6A, data obtained from the CCLE database indicated elevated ISM2 expression in the three CRC cell lines compared to normal colon epithelial cells. Western blot analysis confirmed these findings, revealing upregulated ISM2 expression in all three CRC cell lines, with a particularly significant increase observed in SW1116 cells (Figure 6B,6C).

Figure 6 ISM2 promotes the migration and invasion of colon cancer cells. (A) Relative expression of ISM2 in three colon cancer cells in CCLE database. (B) Western blot analysis of ISM2 expression in normal colon epithelial cells and three types of colon cancer cells. (C) Quantitative analysis of the gray value of strip ISM2, mean ± SEM are displayed (n=3). (D) Western blot analysis of ISM2 protein knockdown inefficiency in SW1116 cells transfected with ISM2 SiRNA#1, 2, 3 and NC SiRNA. (E) Quantitative analysis of the gray value of strip ISM2, mean ± SEM are displayed (n=3). (F) Wound healing assay for cell migration in SW1116 cells; the images of wound closure are shown for the indicated number of hours after scratching (0, 36 h) (red lines indicate the wound boundary, magnification ×50). (G) The quantification for wound healing assay. Mean ± SEM are displayed (n=3). (H) CCK8 analysis of SW1116 cell survival (12, 24 and 36 h). (I,J) Transwell migration assay showing the effect of ISM2 knockdown on SW1116 cell migration after 24 hours. Migrated cells were stained with 0.1% crystal violet and counted in at least three random fields (original magnification ×100). Data are presented as mean ± SEM (n=3). ns, no significance; *, P<0.05; **, P<0.01; ***, P<0.001. CCK8, Cell Counting Kit-8; CCLE, Cancer Cell Line Encyclopedia; NC, negative control; OD, optical density; SEM, standard error of the mean; TPM, transcripts per million.

To examine the functional role of ISM2, three siRNAs targeting ISM2 were synthesized for gene silencing. Following transfection of SW1116 cells, protein extraction from control and knockdown groups demonstrated that siRNA#2 was the most effective in inhibiting ISM2 expression in SW1116 cells (Figure 6D,6E).

Functional assays were then conducted to assess the impact of ISM2 knockdown. The wound healing assay showed a significant reduction in wound closure at 36 hours in the ISM2 knockdown group compared to the control (Figure 6F,6G), indicating impaired cell motility. However, this assay reflects both migration and proliferation. Additionally, the CCK-8 assay showed a significant decrease in cell viability following ISM2 silencing (Figure 6H), consistent with its oncogenic function. To specifically evaluate cell migration, we performed Transwell migration assays with a 24-hour incubation period. The results demonstrated that ISM2 knockdown significantly reduced the number of migrated cells (Figure 6I,6J), confirming its role in promoting CRC cell migration. These findings collectively indicate that ISM2 plays a key role in the proliferation, migration, and invasion of CRC cells.

Discussion

CRC ranks among the most prevalent malignant tumors globally, with notably high incidence and mortality rates within the spectrum of digestive system tumors (11). Prior studies have extensively examined the mechanisms underlying CRC tumorigenesis and progression, along with efforts to identify reliable diagnostic biomarkers and prognostic indicators (12).

Currently, colonoscopy combined with pathological biopsy remains the gold standard for CRC diagnosis. However, this method is invasive and has limitations, particularly in screening asymptomatic patients in the early stages of the disease (13). Consequently, ongoing research focuses on developing non-invasive or minimally invasive diagnostic strategies, including the use of blood-based biomarkers and gene expression.

In terms of treatment, multimodal approaches—encompassing surgery, radiotherapy, chemotherapy, and emerging targeted and immunotherapy techniques—have been used. Despite these advancements, significant variability persists in patient prognoses. These challenges continue to drive investigations into factors influencing CRC outcomes, with the goal of enhancing therapeutic strategies and patient management (14,15).

The ISM protein family, encompassing ISM1 and ISM2, plays essential roles in various biological processes (16). ISM1, encoded by the ISM1 gene, regulates the cell cycle, influences cellular differentiation and apoptosis, and exhibits aberrant expression in several cancers (6). Similarly, the ISM2 gene displays abnormal expression across various tumors. In esophageal cancer, ISM2 is implicated in immune responses involving tumor-associated antigens, as evidenced by differential expression of autoantibodies against ISM2 in the serum of affected patients. This indicates its potential involvement in the development and progression of esophageal cancer (9). In breast cancer, the gene expression profile of MDA-MB-231 cells under hypoxic conditions reveals upregulation of ISM2, indicating a possible role in the response of breast cancer cells to hypoxic environments (17).

In this study, ISM2 was identified as a DEG in CRC based on TCGA database analysis. Its expression differences indicate that ISM2 may serve as a biomarker for CRC diagnosis and prognostic evaluation. ROC curve analysis further demonstrated the potential of ISM2 in distinguishing between tumor and non-tumor tissues. Consistent with findings from Wang et al. in their analysis of CRC using TCGA and Gene Expression Omnibus (GEO) datasets, we confirmed that ISM2 is highly expressed in CRC and associated with advanced disease stages, reinforcing its role as a driver of malignant progression. Wang et al. further showed that silencing ISM2 reduces tumor growth and enhances CD8⁺ T cell infiltration, which aligns with our observations of ISM2’s impact on immune microenvironment regulation (18). Additionally, ISM2 expression varied significantly across T, N, and M stages, pathological stages, and histological types. Higher ISM2 expression was associated with advanced disease stages, indicating a role in promoting CRC malignancy. In vitro experiments corroborated these findings, showing that ISM2 knockdown significantly inhibited colon cancer cell proliferation and migration. Notably, Wang et al. also reported that ISM2 silencing improves the efficacy of PD1 antibodies, which supports our hypothesis that ISM2 modulation could enhance immunotherapeutic responses in CRC (18). Similarly, Huang et al. identified ISM2 as a key immune-related gene in hepatocellular carcinoma, where it was associated with poor prognosis and immune infiltration, further underscoring its potential as a pan-cancer prognostic biomarker (19). However, this study only validated ISM2 function in SW1116 cells and relied on a single TCGA database. Further confirmation of its biological role in more cell lines and clinical cohorts is needed.

The TME comprises of a complex network of tumor cells, immune cells, fibroblasts, vascular endothelial cells, and the extracellular matrix (20). In CRC, the TME exerts diverse influences on immune cell infiltration (21). Tumor cells secrete chemokines and cytokines that can modulate immune cell infiltration, including promoting tumor-associated macrophages toward the M2 phenotype, changing immune cell composition, and reducing cytotoxic T lymphocyte infiltration. In hypoxic regions, hypoxia-inducible factors trigger the secretion of immunosuppressive cytokines, further inhibiting immune cell activation and infiltration (22-24).

In this study, ISM2 expression was negatively associated with the infiltration of Th2 cells, CD8⁺ T cells, aDC, and cytotoxic cells. These findings indicate that ISM2 contributes to the CRC progression by interfering with the anti-tumor immune response, thereby impacting patient prognosis. Previous research on early-stage lung squamous cell carcinoma has demonstrated the association of immune-related genes, including ISM2, with the tumor immune microenvironment, indicating a regulatory role (25). Additionally, one report highlighted that CRC following human papillomavirus (HPV) infection exhibited enhanced anti-tumor immune responses and increased immune cell infiltration, providing a novel perspective on the involvement of ISM2 in immune regulation (8). While Th2 cells typically exert immunosuppressive effects, their role in CRC is context-dependent. Recent studies indicate that in advanced CRC, Th2 cell infiltration may diminish due to tumor-driven metabolic reprogramming or cytokine dysregulation, limiting their immunosuppressive capacity (26). Thus, ISM2 likely drives CRC progression through Wnt-mediated immune evasion, with Th2 cell reduction reflecting a broader immunosuppressive TME remodeling (27). Future single-cell RNA sequencing will dissect ISM2’s impact on Th2 cell differentiation and TME crosstalk, resolving this complexity. Current research on the specific signaling pathways involving ISM2 in tumors remains limited. However, insights can be drawn from studies on its family member ISM1. For instance, ISM1 engages in signal transduction through interactions with various molecules, such as binding to activin receptor IB to regulate the NODAL signaling pathway, thereby influencing processes like cell proliferation and differentiation (28).

In this study, GO and KEGG enrichment analyses indicated that ISM2 contributes to tumorigenesis through the Wnt/β-catenin signaling pathway, and is associated with angiogenesis and hypoxic responses. The Wnt/β-catenin signaling pathway, a well-established pathway in cancer biology, has been demonstrated to play a key role in CRC progression. Activation of this pathway enhances the invasive and metastatic potential of tumor cells (29). Mechanistically, ISM2 may promote CRC progression via Wnt/β-catenin pathway activation, as supported by KEGG enrichment and reduced β-catenin signaling upon ISM2 knockdown. Its negative correlation with CD8⁺ T cell infiltration suggests a role in immune evasion, possibly through TGF-β3-mediated immunosuppression.

The role of ISM2 in tumor biology is likely not independent but involves interactions or synergistic effects with other genes. Tumorigenesis and progression are complex processes regulated by multiple genes influencing various cellular biological functions (30). For example, under hypoxic conditions in breast cancer, a network of hypoxia-responsive genes collaborates to regulate tumor cell behavior. Within this regulatory network, ISM2 acts in concert with other hypoxia-related genes to affect the biological characteristics of breast cancer cells (17).

Further exploration of the interactions between ISM2 and other genes is necessary to deepen the understanding of its functional roles in tumor biology and to elucidate the molecular mechanisms underlying its involvement in tumorigenesis and progression. Such studies provide valuable insights into its potential as a therapeutic target. Potential strategies for targeting ISM2 include the development of monoclonal antibodies aimed at blocking its signaling pathways or mediating immune-mediated cytotoxicity, the design of small molecule drugs that inhibit its interaction with the Wnt pathway, and the combination with immune checkpoint inhibitors to reshape the TME. The development of anti-ISM2 monoclonal antibodies can learn from the success of anti-epidermal growth factor receptor (EGFR) antibodies by blocking their binding to receptors to inhibit the Wnt pathway or using antibody-dependent cytotoxicity to eliminate tumor cells (31). Small molecule inhibitors like XAV939 can stabilize Axin protein and block β-catenin nuclear translocation, and combining these with ISM2 targeting may yield synergistic effects (32). In immunotherapy, previous studies have shown that combining ISM2 silencing with programmed death-1 (PD-1) antibody treatment enhances anti-tumor efficacy in CRC, which aligns with clinical trial observations that PD-1 inhibitors combined with chemotherapy yield synergistic benefits. Similar strategies could further explore the combination of ISM2 inhibition with immune checkpoint blockade to reshape the anti-tumor immune microenvironment, building on the evidence that ISM2 regulates immune cell infiltration (18,33). In the future, it will be essential to validate the specificity of this target in a broader range of models, optimize multi-target combination therapy regimens, and utilize clinical biomarkers to identify populations that would benefit from this approach, thereby facilitating its transition towards precision medicine.

In conclusion, this study demonstrates that ISM2 expression is significantly associated with clinical outcomes and immune cell infiltration in CRC. These findings highlight the potential of ISM2 as a diagnostic and prognostic biomarker for CRC. However, several limitations should be acknowledged.

First, while ectopic expression of ISM2 influences cell proliferation and invasion, the underlying molecular mechanisms are being illuminated by our supplementary findings. Our research demonstrates that ISM2 knockdown suppresses nuclear β-catenin and c-Myc expression, supporting its role in activating the Wnt/β-catenin pathway. This mechanistic insight highlights ISM2 as a potential upstream regulator of Wnt-mediated tumor progression. However, the full spectrum of its molecular functions remains undefined. Future investigations should delve into ISM2’s involvement in tumor angiogenesis, immune evasion, and crosstalk with other signaling networks, preferably using in vivo models to validate its functional relevance in TMEs. Second, the statistical analyses in this study may be subject to bias, as the data were derived exclusively from the TCGA database. A more comprehensive assessment of the clinical significance of ISM2 in tumors will require multicenter clinical studies with larger sample sizes. Such studies would enable the collection of data from patients across various tumor types and stages, allowing for a more thorough validation of the correlation between ISM2 expression and clinicopathological features, as well as prognosis. This would provide a stronger foundation for its potential application in clinical settings.

It is also important to note that the in vitro functional assays were conducted using a single CRC cell line (SW1116) and one siRNA sequence targeting ISM2. While this may limit the generalizability of the findings, SW1116 is a well-established cell line commonly used in CRC research, and the siRNA used was validated for its knockdown efficiency. Moreover, our experimental results are consistent with bioinformatics analyses based on large-scale clinical datasets, which enhances the reliability of our observations. In summary, ISM2 shows potential as a prognostic biomarker in CRC. While this study did not directly assess immune cell function, our in vitro findings confirm the oncogenic role of ISM2 in promoting CRC cell proliferation, migration, and invasion-effects that may contribute to immune evasion. Further studies are warranted to investigate the direct impact of ISM2 on immune cell recruitment and function, such as through co-culture models or cytokine profiling, to clarify its role in immune modulation.

Conclusions

This study focused on the role of ISM2 in CRC, confirming its elevated expression in clinicopathological settings and its function as a potential oncogene. ISM2 expression is associated with prognosis and immune cell infiltration in CRC. ISM2 has emerged as a promising biomarker for prognostic evaluation in CRC, while further research is necessary to comprehensively validate its role.

Acknowledgments

We are particularly grateful to all the people who have given us help on our article.

Footnote

Reporting Checklist: The authors have completed the REMARK and MDAR reporting checklists. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-295/rc

Data Sharing Statement: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-295/dss

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-295/prf

Funding: This study was supported by the Scientific Research Project of Health Commission of Jiangsu Province, China (No. Y12023020); and Scientific Research Project of Nantong Health Commission (No. MSZ2024034).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2025-295/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Farvid MS, Sidahmed E, Spence ND, et al. Consumption of red meat and processed meat and cancer incidence: a systematic review and meta-analysis of prospective studies. Eur J Epidemiol 2021;36:937-51. [Crossref] [PubMed]
2. Siegel RL, Miller KD, Goding Sauer A, et al. Colorectal cancer statistics, 2020. CA Cancer J Clin 2020;70:145-64. [Crossref] [PubMed]
3. Kuiper JG, van Herk-Sukel MPP, Lemmens VEPP, et al. A steep increase in healthcare seeking behaviour in the last months before colorectal cancer diagnosis. BMC Fam Pract 2021;22:121. [Crossref] [PubMed]
4. Cao Y, Wang X. Effects of molecular markers on the treatment decision and prognosis of colorectal cancer: a narrative review. J Gastrointest Oncol 2021;12:1191-6. [Crossref] [PubMed]
5. Al-Maghrabi JA, Qureshi IA, Khabaz MN. Expression of leptin in colorectal adenocarcinoma showed significant different survival patterns associated with tumor size, lymphovascular invasion, distant metastasis, local recurrence, and relapse of disease in the western province of Saudi Arabia. Medicine (Baltimore) 2018;97:e12052. [Crossref] [PubMed]
6. Shakhawat HM, Hazrat Z, Zhou Z. Isthmin-A Multifaceted Protein Family. Cells 2022;12:17. [Crossref] [PubMed]
7. Martinez C, González-Ramírez J, Marín ME, et al. Isthmin 2 is decreased in preeclampsia and highly expressed in choriocarcinoma. Heliyon 2020;6:e05096. [Crossref] [PubMed]
8. Zhou J, Liu Y, Zhang Y, et al. Comprehensive analysis of a novel subtype of immune microenvironment-derived HPV-infected colorectal cancer. Microbes Infect 2024;26:105315. [Crossref] [PubMed]
9. Li T, Sun G, Ye H, et al. ESCCPred: a machine learning model for diagnostic prediction of early esophageal squamous cell carcinoma using autoantibody profiles. Br J Cancer 2024;131:883-94. [Crossref] [PubMed]
10. Chen J, Yang P, Li S, et al. Increased FOXM1 Expression was Associated with the Prognosis and the Recruitment of Neutrophils in Endometrial Cancer. J Immunol Res 2023;2023:5437526. [Crossref] [PubMed]
11. Scheurlen KM, Billeter AT, O'Brien SJ, et al. Metabolic dysfunction and early-onset colorectal cancer - how macrophages build the bridge. Cancer Med 2020;9:6679-93. [Crossref] [PubMed]
12. de Souza JB, de Almeida Campos LA, Palácio SB, et al. Prevalence and implications of pKs-positive Escherichia coli in colorectal cancer. Life Sci 2024;341:122462. [Crossref] [PubMed]
13. Parsa N, Byrne MF. Artificial intelligence for identification and characterization of colonic polyps. Ther Adv Gastrointest Endosc 2021;14:26317745211014698. [Crossref] [PubMed]
14. Yang Y, Meng WJ, Wang ZQ. Immunotherapy with Immune Checkpoint Inhibitors for Advanced Colorectal Cancer: A Promising Individualized Treatment Strategy. Front Biosci (Landmark Ed) 2023;28:69. [Crossref] [PubMed]
15. Shi J, Sun Z, Gao Z, et al. Radioimmunotherapy in colorectal cancer treatment: present and future. Front Immunol 2023;14:1105180. [Crossref] [PubMed]
16. Venugopal S, Chen M, Liao W, et al. Isthmin is a novel vascular permeability inducer that functions through cell-surface GRP78-mediated Src activation. Cardiovasc Res 2015;107:131-42. [Crossref] [PubMed]
17. Han D, Li Z, Luo L, et al. Targeting Hypoxia and HIF1α in Triple-Negative Breast Cancer: New Insights from Gene Expression Profiling and Implications for Therapy. Biology (Basel) 2024;13:577. [Crossref] [PubMed]
18. Wang Y, Wang P, Liu J, et al. ISM2 Is a Novel Prognostic Biomarker and Correlates with Tumor Immune Microenvironment in Colorectal Cancer. Iran Red Crescent Med J 2024; [Crossref]
19. Huang R, Chen Z, Li W, et al. Immune system-associated genes increase malignant progression and can be used to predict clinical outcome in patients with hepatocellular carcinoma. Int J Oncol 2020;56:1199-211. [Crossref] [PubMed]
20. Prasad S, Saha P, Chatterjee B, et al. Complexity of Tumor Microenvironment: Therapeutic Role of Curcumin and Its Metabolites. Nutr Cancer 2023;75:1-13. [Crossref] [PubMed]
21. Waldner MJ, Neurath MF. TGFβ and the Tumor Microenvironment in Colorectal Cancer. Cells 2023;12:1139. [Crossref] [PubMed]
22. Lim GY, Cho SW, Ka NL, et al. IFI16/Ifi202 released from breast cancer induces secretion of inflammatory cytokines from macrophages and promotes tumor growth. J Cell Physiol 2023;238:1507-19. [Crossref] [PubMed]
23. Su MT, Kumata S, Endo S, et al. LILRB4 promotes tumor metastasis by regulating MDSCs and inhibiting miR-1 family miRNAs. Oncoimmunology 2022;11:2060907. [Crossref] [PubMed]
24. Jakubowska K, Kisielewski W, Kańczuga-Koda L, et al. Diagnostic value of inflammatory cell infiltrates, tumor stroma percentage and disease-free survival in patients with colorectal cancer. Oncol Lett 2017;14:3869-77. [Crossref] [PubMed]
25. Fan T, Lu Z, Liu Y, et al. A Novel Immune-Related Seventeen-Gene Signature for Predicting Early Stage Lung Squamous Cell Carcinoma Prognosis. Front Immunol 2021;12:665407. [Crossref] [PubMed]
26. Fiegle E, Doleschel D, Koletnik S, et al. Dual CTLA-4 and PD-L1 Blockade Inhibits Tumor Growth and Liver Metastasis in a Highly Aggressive Orthotopic Mouse Model of Colon Cancer. Neoplasia 2019;21:932-44. [Crossref] [PubMed]
27. Katoh M, Katoh M. WNT signaling and cancer stemness. Essays Biochem 2022;66:319-31. [Crossref] [PubMed]
28. Zhou X, Zhang K, Wang C, et al. Isthmin-1 promotes growth and progression of colorectal cancer through the interaction with EGFR and YBX-1. Cancer Lett 2024;590:216868. [Crossref] [PubMed]
29. Ji R, Ji Y, Ma L, et al. Keratin 17 upregulation promotes cell metastasis and angiogenesis in colon adenocarcinoma. Bioengineered 2021;12:12598-611. [Crossref] [PubMed]
30. Kontomanolis EN, Koutras A, Syllaios A, et al. Role of Oncogenes and Tumor-suppressor Genes in Carcinogenesis: A Review. Anticancer Res 2020;40:6009-15. [Crossref] [PubMed]
31. Filmus J, Capurro M. Glypican-3: a marker and a therapeutic target in hepatocellular carcinoma. FEBS J 2013;280:2471-6. [Crossref] [PubMed]
32. Luo F, Li J, Liu J, et al. Stabilizing and upregulating Axin with tankyrase inhibitor reverses 5-fluorouracil chemoresistance and proliferation by targeting the WNT/caveolin-1 axis in colorectal cancer cells. Cancer Gene Ther 2022;29:1707-19. [Crossref] [PubMed]
33. Lin KX, Istl AC, Quan D, et al. PD-1 and PD-L1 inhibitors in cold colorectal cancer: challenges and strategies. Cancer Immunol Immunother 2023;72:3875-93. [Crossref] [PubMed]