Comprehensive study of the tumor immune microenvironment and the prognostic significance of reticulocalbin-1: a pan-cancer analysis Authors: JiangLi Zhao, XuChen Liu, QingYuan Sun, ShuJie Peng, ZhiWei Xue, JunZhi Liu, Bin Huang, JiWei Wang, Chao Li, Ning Yang, Xinyu Wang Published: 2026 ABSTRACT Reticulocalbin-1 (RCN1), an endoplasmic reticulum (ER)-related gene, demonstrates critical regulatory functions in cellular growth. RCN1 has been previously revealed to be closely associated with initiation, metastasis and chemosensitivity in some tumors. However, the prognostic significance of RCN1 and its influence on the cancer immune microenvironment remains unclear. By leveraging multi-omics data from TCGA and GEO, this study delineated the expression profile of RCN1 across various human malignancies and normal tissues. Analysis of RCN1 expression in correlation with tumor immune microenvironment characteristics was performed using CIBERSORT and ssGSEA. Pan-cancer prognostic value was assessed via multiple survival models. In glioma, immunohistochemistry, co-culture, and qPCR were utilized to further investigate the role of RCN1 in tumor immunity and prognosis. Drug-specific responses to RCN1 expression were probed for cisplatin, irinotecan, carmustine and lomustine, which are the common chemotherapeutic agents used in glioma clinical treatment. The results indicate that RCN1 is a key gene with the potential to promote the progression of multiple tumors and is closely associated with tumor immune response, particularly macrophage activation. In glioma, overexpression of RCN1 promotes tumor proliferation and indicates poor prognosis, and is closely related to endoplasmic reticulum stress. Knockdown of RCN1 enhances the chemotherapeutic efficacy of carmustine and lomustine. In conclusion, our research positioned RCN1 as an important oncogene from a pan-cancer perspective and validated its role in glioma. In particular, RCN1 is closely associated with the immune microenvironment and endoplasmic reticulum stress during tumorigenesis. In addition, RCN1 has broad prospects in guiding tumor chemotherapy pending physiological in vivo validation.Supplementary informationThe online version contains supplementary material available at 10.1007/s10142-026-02017-5. Introduction Endoplasmic reticulum (ER) is a network composed of organelles wrapped in membrane with a unique structure, which plays essential roles in protein handling, modification, and folding to determine cell function, fate, and survival (Schwarz and Blower 2016). Therefore, research on ER-related genes is crucial to disease pathogenesis and therapy, as they are central to maintaining protein homeostasis. Dysregulation of these genes impairs protein folding and processing, leading to misfolded protein accumulation, ER stress, and the subsequent activation of autophagy, oxidative stress, and inflammatory responses (Di Conza and Ho 2020) (Lin et al. 2019). Reticulocalbin (RCN) serves as a calcium-signaling agent, which is characterized as an EF-hand domain-containing protein operating within the ER. Furthermore, RCN1, RCN2, and RCN3 are members of the CREC family (Honoré, 2009). The RCN family has crucial roles in ER regulation and the relationship between the RCN family and diseases is complex. The RCN family member RCN1 contains six conserved regions and binds Ca2+ cooperatively to increase the α-helix content of RCN1. Previous evidence suggested that RCN1 was essentially a survival switch for damaged cells during pathophysiological ER stress events (Gomez et al. 2021). Recent research has proved that RCN1 mediates progression in several tumors, including lung and prostate cancer (Fu et al. 2021) (Liu et al. 2018), and appears as renal cell carcinoma’s biomarker (Giribaldi et al. 2013). Collectively, these findings suggest that RCN1 plays an important role in tumor microenvironment remodeling, and elucidating its function will provide effective insights for the future development of immune checkpoint therapy (Wang et al. 2025). Bioinformatics research based on immunology suggested that RCN1 takes part in tumor cell invasion to accelerate invasion and migration. We speculate that the RCN family, especially RCN1, may be closely related to tumor progression, although research on the RCN family remains insufficient, the exploration of RCN1’s functions in pan-cancer remains limited, and its potential impact on patient prognosis and immune responses has not been thoroughly analyzed. Alterations in ER function are critical, because they simultaneously control the pro-tumor properties of cancer cells and dynamically reprogram immune cell functions (Chen and Cubillos-Ruiz 2021). Studies suggest that RCN1 exerts its functions depending on the regulatory mechanisms of the endoplasmic reticulum system(Gomez et al. 2021). Targeting the ER stress response could inhibit tumor growth and enhance anti-cancer immunity concomitantly, offering a new pathway to cancer immunotherapy (Cubillos-Ruiz et al. 2015). The IRE1α-XBP1s axis, for instance, controls dendritic cell homeostasis and PD-L1 expression in cancer cells(Shen et al. 2025). Whether RCN1, as an upstream regulator of ER function, similarly impacts immune surveillance remains unexplored. Meanwhile, whether RCN1 exerts an influence on endoplasmic reticulum stress regulatory pathways such as the IRE1α-XBP1s axis remains insufficiently investigated. This research revealed unique expression characteristics of RCN1 in pan-cancer. We used multiple algorithms to examine the characteristics of RCN1 in tumor immune regulation and found that RCN1 showed strong specificity in immune checkpoints and immune cell infiltration. Using glioma as an example, we validated the above findings and confirmed that knockdown of RCN1 affects the expression of endoplasmic reticulum stress molecules such as CHOP, IRE1α, and XBP1s, while also increasing apoptosis following chemotherapy. Methods Data acquisition and assessment of rcn1 expression in normal versus tumor tissues The harmonized pan-cancer resource originated from UCSC Xena (https://xenabrowser.net/), utilizing their TCGA Pan-Cancer dataset (N = 10,535, G = 60,499). Expression matrices for ENSG00000049449 (RCN1) were compiled across three source categories: Solid Tissue Normal, Primary Tumor tissues and Primary Blood Derived Cancer- Peripheral Blood. Each value underwent log2(x + 0.001) adjustment before analysis, with each tumor type maintaining ≥ 3 samples. The final cohort comprised 26 cancers with complete expression datasets. For pan-cancer differential analysis of RCN1 in tumor versus normal tissue using TCGA resources, statistical testing was conducted via unpaired Wilcoxon tests implemented in limma (R v3.6.4). Our dual-filtering criteria required: (1) adjusted p-values < 0.05, and (2) absolute log-fold-change |logFC|> 0.35 for significance. Analysis of the immune microenvironment Immune cell infiltration was assessed via CIBERSORT to quantify cell subset fractions from bulk tissue transcriptome data (Chen et al. 2018). Immune function analysis was profiled using ssGSEA implemented (Hänzelmann et al. 2013), adding the expression of genes involved in certain pathways and obtain their relative activity. Furthermore, to obtain an overall landscape of the immune microenvironment, stromal and immune scores were generated with ESTIMATE algorithm accessed through TIMER2.0 (Li et al. 2020) — a computational platform providing comprehensive visualization of immune infiltrates in tumor. Identification of genes correlated with rcn1 Glioma patient cohorts were stratified by RCN1 expression levels into low- and high-RCN1 subgroups. Differential gene expression analysis employed the “limma” package (Ritchie et al. 2015) with significance thresholds of |log2FC|> 1 and false discovery rate (FDR) < 0.05 to define biologically relevant differentially expressed genes. Besides, genes co-expressed with RCN1 were identified through Pearson correlation analysis, considering only those with P < 0.05. Eventually, we took the intersection of DEGs and co-expressed genes of RCN1, as the RCN1-correlated genes. Construction and analysis of the protein–protein interaction (PPI) network Protein–protein interaction (PPI) networks were constructed using the STRING database (https://cn.string-db.org/) and visualized via Cytoscape v3.9.0, enabling integrative analysis of biomolecular interactions within molecular profiling data (Shannon et al. 2003). Gene set enrichment analysis and COX regression analysis Gene set enrichment was assessed using GSEA v4.2.3(Subramanian et al. 2005), which is a method that derives its power by focusing on gene sets, i.e., genes that share a common biological function, chromosomal location, or regulation. All of the selected pathways satisfied P < 0.05. Cox proportional hazards models were fitted using the R package “survival” (v3.8.3). GO and KEGG enrichment analyses To identify the influential functions of RCN1, we conducted GO and KEGG enrichment analyses through the R package “clusterProfiler”, which compared the biological themes among gene clusters (Yu et al. 2012). The GO enrichment analysis comprised three parts: biological process (BP), cellular component (CC), and molecular function (MF) (Ashburner et al. 2000). The KEGG analysis links the genomic information with higher order functional information and represents a knowledge base for the systematic analysis of gene functions (Kanehisa and Goto 2000). Pathways with the lowest P values were displayed for each analysis, all of which were < 0.05. Cell lines and culture The human glioma cell lines U118MG, A172, and LN229 were obtained from the Chinese Academy of Sciences Cell Bank (Shanghai, China) and ATCC Cell Bank (Shanghai, China). T98, derived from primary human glioblastoma (GBM) xenograft tumors, and P3 glioma cells were kindly provided by Prof. Rolf Bjerkvig from University of Bergen. All cell lines were cultured in DMEM (Sigma, USA) containing 10% fetal bovine serum (Thermo Fisher Scientific, USA) under standard conditions: 37 °C with 5% CO2 humidified atmosphere. THP-1 human acute monocytic leukemia cell line was obtained from the American Type Culture Collection and cultured in THP-1 medium containing RPMI-1640 supplemented with 10% FBS. Macrophages were differentiated from THP-1 cells in the aforementioned medium supplemented with 100 nM phorbol 12-myristate 13-acetate (PMA), seeded at a density of 1 × 10⁶ cells per well in 6-well tissue culture plates for 72 h. Subsequently, the PMA-containing medium was removed and replaced with THP-1 medium. To evaluate the involvement of macrophages in glioma progression, an in vitro co-culture model was employed. Macrophages were cultured in 6-well plates, while tumor LN229 cells were subjected to siRNA treatment or transfection. After a 48-h incubation period, glioma cells were introduced into the upper chamber. Targeted gene modulation through sirna and plasmid transfection Targeted RCN1 siRNAs and engineered plasmids for RCN1 overexpression were synthesized from GenePharma (China). Transfections were conducted with Lipofectamine™ 3000 (Thermo Fisher Scientific). For macrophage intrinsic modulation, PMA-differentiated THP-1 macrophages were similarly transfected with si-RCN1 using CALNP™ RNAi reagent (Duona, Beijing, China), following the manufacturer’s instructions. Detailed sequences of primers for PCR and siRNA are provided in Supplementary Table 1. Western blotting Protein extraction from glioblastoma (GBM) cells was performed using RIPA lysis buffer (Beyotime, P0013B) containing 1% protease and phosphatase inhibitor cocktail. Proteins were separated by SDS-PAGE and transferred onto PVDF membranes (Bio-Rad, CA, USA). Membranes were incubated with specific primary antibodies, including RCN1, phosphorylated p53 (p-p53), GAPDH (Proteintech, USA), p53, Caspase-3, and cleaved Caspase-3 (Cell Signaling Technology, USA). RNA extraction and quantitative RT-PCR (qrt-PCR) Total RNA isolation employed the RNA Extraction Kit (YISHAN Biotechnology, Shanghai, China) following manufacturer’s guidelines. cDNA synthesis was carried out with High-Capacity cDNA Reverse Transcription Kits (Applied Biosystems). Subsequently, we conducted quantitative PCR analysis on the Mx-3000P platform (Stratagene) utilizing primers detailed in Supplementary Table 1. Cell viability assay using CCK-8 and dose–response curve analysis Cisplatin and Irinotecan were obtained from Topscience (Shanghai, China T1564 and T6228). Cell proliferative viability was determined with Cell Counting Kit-8 (CCK-8) assays. A172 and U118MG cells were plated in 96-well plates at 4 × 103 cells/well density. Cells were exposed to a series of drug concentrations (5,10, 20, 50, 60, 80, 100, 120, 150, 180 and 200 μmol/ml) for a duration of 24 h. After supplementing each well with 10 μl CCK-8 solution, plates were incubated at 37 °C for 1 h. Absorbance at 450 nm (OD450) was then quantified using a PerkinElmer EnSight instrument. The resulting data were normalized and visualized to obtain IC50 with Graphpad Prism (version 8.02). 5-ethynyl-2’-deoxyuridine (edu) cell proliferation assay Cell proliferation activity was evaluated through EdU cell proliferation assay kit (Yeasen, 40275ES76, China) after cell culture. The proliferation index was calculated with Graphpad prism (version 8.02) via the ratio of EdU-positive (green) cells to total Hoechst 33,342-positive (blue) cells. Cell apoptosis rates via flow cytometry Following the protocol, Annexin V-FITC/PI Apoptosis Detection Kit (Yeasen, China) were used to detect the apoptosis rates of glioma cells. All assays and data processing were conducted on NovoCyte flow cytometer (ACEA, China). For rescue assays, transfected LN229 cells were treated with the IRE1α activator IXA6 (MedChemExpress;10 μM) for 18 h before apoptosis detection. Statistical analysis The correlations among related genes in RCN1 study were predicted using Spearman linear regression analyses. Correlation testing mapped relationships between gene expression levels and immune cells abundance within our analytical process. Differences between groups were detected through Wilcoxon test except for specified illustration. Besides, p value < 0.05 was considered as statistically significant. All statistical analyses were performed based on R software (version 4.2.2). Results Tissue-specific expression patterns of rcn1 We extracted RCN1 tissue-specific expression data using datasets from CCLE and GTEx. The RCN1 low expression tissue included lymphocytes and plasma cells according to CCLE, as well as blood and pancreas based on GTEx. Meanwhile, the CCLE data showed higher RCN1 expression level in the fibroblast and kidney, with higher RCN1 expression level in the pituitary and ovary based on GTEx data (Figs. 1A, B).Fig. 1RCN1 expression among different tissues and pan-cancer. (A, B) Expression level of RCN1 in different tissue types. Data from the CCLE, GTEx database. (C) Differential RCN1 expression between normal and tumor based on TCGA pan-cancer database. (D) Differential RCN1 expression between normal and tumor based on TCGA and GTEx pan-cancer databases. (E, F) IHC staining of RCN1 protein level among normal tissues and various types or grades of common tumors from HPA database and corresponding statistics of RCN1 staining. (G) Tissue heteromorphism score of representative tumor sections in HPA database. (* P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001) RCN1 expression among different tissues and pan-cancer. (A, B) Expression level of RCN1 in different tissue types. Data from the CCLE, GTEx database. (C) Differential RCN1 expression between normal and tumor based on TCGA pan-cancer database. (D) Differential RCN1 expression between normal and tumor based on TCGA and GTEx pan-cancer databases. (E, F) IHC staining of RCN1 protein level among normal tissues and various types or grades of common tumors from HPA database and corresponding statistics of RCN1 staining. (G) Tissue heteromorphism score of representative tumor sections in HPA database. (* P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001) Rcn1 is expressed specifically among pan-cancer We evaluated differences in RCN1 expression level between normal tissues and different tumors using the TCGA database, observing marked upregulation in 19 tumors and downregulation in two (Fig. 1C). RCN1 expression was upregulated in glioblastoma multiforme (GBM), brain lower grade glioma (LGG), GBMLGG, lung adenocarcinoma (LUAD), colon adenocarcinoma/rectum adenocarcinoma (COADREAD), COAD, esophageal carcinoma (ESCA), kidney renal papillary cell carcinoma (KIRP), pan-kidney cohort (KIPAN), stomach and esophageal carcinoma (STES), stomach adenocarcinoma (STAD), prostate adenocarcinoma (PRAD), head and neck squamous cell carcinoma (HNSC), kidney renal clear cell carcinoma (KIRC), lung squamous cell carcinoma (LUSC), liver hepatocellular carcinoma (LIHC), rectum adenocarcinoma (READ), bladder urothelial carcinoma (BLCA), and cholangiocarcinoma (CHOL), and was downregulated in thyroid carcinoma (THCA) and kidney chromophobe (KICH). To address TCGA normal sample limitations, we supplemented our analysis with GTEx-derived normal data integration and discovered that there were seven and one tumors in which the expression level of RCN1 was up- and downregulated, respectively. The former included Wilms tumor (WT), skin cutaneous melanoma (SKCM), pancreatic adenocarcinoma (PAAD), testicular germ cell tumors (TGCT), uterine carcinosarcoma (UCS), acute lymphoblastic leukemia (ALL), and acute myeloid leukemia (LAML), and the latter included adrenocortical carcinoma (ACC) (Fig. 1D). Analysis of core elements related to tumor development Various vital factors associated with tumor progression, including the “T”, “N”, and “M” stages of carcinoma, were subsequently analyzed (Supplementary Figure S1). Subsequently, the relationship of RCN1 transcriptional levels with tumor stage, which reflects the invasion and metastatic ability of tumors, was evaluated. Ultimately, significant differences were found in eight tumors, including CESC and LUAD (Supplementary Figure S2D). Subsequently, correlation analysis between RCN1 expression and tumor progression (T, N, M, and tumor stage) was conducted and the top five are displayed. Obviously, in addition to the positive correlation between RCN1 expression and “M-stage”, the associations between RCN1 and the “T-stage”, “N-stage” and overall tumor stage varied across different tumor types (Supplementary Figure S2). Therefore, we carried out a further investigation below. Relationship between rcn1 expression and tumor grades according to immunohistochemical analysis of pan-cancer To further investigate the clinical relevance of RCN1 expression in tumors, we assessed RCN1 expression level among tumors of varying grades. The results showed valid differences in nine tumors (Supplementary Figure S3), which were verified by Spearman linear analysis further. Combined with the above pan-cancer study, we performed an immunohistochemical analysis of RCN1 expression in classical tumors (Fig. 1E, F). Moreover, RCN1 expression pattern was different from that of normal tissues in many tumors. Simultaneously, we scored representative tissue sections for heterotypicity (Fig. 1G), with scores ranging from 1–5, and the heterotypicity increased with the score. In tumors with high expression of RCN1, such as glioma and renal cell carcinoma, the score of tissue heterotypicity was correspondingly higher. Pan-cancer analysis correlating rcn1 expression levels with features of the immune microenvironment Next, we employed the “ESTIMATE” algorithm to assess RCN1’s influence on the pan-cancer microenvironment. According to Fig. 2A, RCN1 expression displayed tight linkage with tumor stromal components (Stromal score), immune infiltration (Immune score), and global tumor purity (ESTIMATE score). Also, using “CIBERSORT” algorithm, we analyzed immune cells abundance in TCGA patient samples with diverse tumor types, finding that RCN1 expression was linked to the microenvironment in most tumors. Typically, glioma (GBM and LGG) had the highest number of RCN1-regulated immune cells (n = 16; Fig. 2B). Furthermore, RCN1 showed a positive association with most immune checkpoint (IC) regulators across the pan-cancer dataset, including both IC inhibitors and stimulators (Fig. 2C). In detail, we noted that LAG3, CTLA4, and CD274 (PD-L1) expression level were positively correlated with RCN1 in a variety of cancers, including GBMLGG, BRCA. These findings established RCN1 as a valid regulator of immune microenvironment dynamics across human cancers.Fig. 2Heatmaps showing RCN1 role within the pan-cancer immune landscape. (A) Correlation of RCN1 expression and microenvironment of pan-cancer based on ESTIMATE. (B) Co-expression of RCN1 with tumor-infiltrating immune cells in pan-cancer based on CIBERSORT. (C) Association of RCN1 and immune checkpoint regulators in pan-cancer.(* P < 0.05) Heatmaps showing RCN1 role within the pan-cancer immune landscape. (A) Correlation of RCN1 expression and microenvironment of pan-cancer based on ESTIMATE. (B) Co-expression of RCN1 with tumor-infiltrating immune cells in pan-cancer based on CIBERSORT. (C) Association of RCN1 and immune checkpoint regulators in pan-cancer.(* P < 0.05) Rcn1 expression significantly associated with tumor prognosis, mutation, and expression modification To evaluate RCN1’s prognostic impact in pan-cancer, three critical prognostic indicator:overall survival (OS), disease-specific survival (DSS), and progression-free interval (PFI) were analyzed using random forest modeling. The results showed an obvious positive relationship among most carcinomas. Surprisingly, enhanced RCN1 expression in glioma was significantly associated with adverse clinical outcomes, regardless of the evaluation measures (OS, 95% CI: 1.93–3.16; DSS, 95% CI: 1.93–3.32; PFI, 95% CI: 1.53–2.34) (Figs. 3A-C). We plotted the corresponding KM and ROC curves in glioma, and each figure indicated that worse survival was associated with higher RCN1 levels, confirming their ability to stratify patients based on 1-, 3-, and 5-year survival risk. (Figs. 3D–F). We also analyzed the RNA modification, tumor mutation burden (TMB), and microsatellite instability (MSI) of RCN1 across tumor types (Supplementary Figures S4, S5, S6).Fig. 3RCN1 prognostic value in pan-cancer. (A, B, C) Forest plots visualize Cox regression analysis of RCN1 prognostic significance respectively associated with OS, DSS, and PFI. A Hazard ratio (HR) > 1 represents a risk factor. (D, E, F) KM and ROC curves based on OS, DSS, and PFI, respectively, in glioma. (G) RCN1 expression levels and corresponding KM survival curves across glioma grades (G2–G4) in TCGA. (H, I, J, K) Independent expression evaluation of RCN1 in glioma via CGGA, GSE43378, GSE4271, and CBTTC databases RCN1 prognostic value in pan-cancer. (A, B, C) Forest plots visualize Cox regression analysis of RCN1 prognostic significance respectively associated with OS, DSS, and PFI. A Hazard ratio (HR) > 1 represents a risk factor. (D, E, F) KM and ROC curves based on OS, DSS, and PFI, respectively, in glioma. (G) RCN1 expression levels and corresponding KM survival curves across glioma grades (G2–G4) in TCGA. (H, I, J, K) Independent expression evaluation of RCN1 in glioma via CGGA, GSE43378, GSE4271, and CBTTC databases Expression and prognostic features of rcn1 in glioma We selected glioma for use in further validation based on the above research. In glioma samples from TCGA, RCN1 expression was closely correlated with the degree of malignancy, with a statistically significant gradual increase in expression from normal to LGG to GBM tissues. In addition, enhanced RCN1 expression level was greatly associated with unfavorable patient outcomes. (Fig. 3G). Data from the CGGA, GEO, and CBTTC were used for verification. We concluded that RCN1 was closely associated with the malignancy of glioma and that high-RCN1 could be considered a risk factor of prognosis (Figs. 3H-K, Supplementary Figures S7). Transcriptomic landscape analysis of rcn1-associated genes in glioma To explore transcriptional alterations of genes regulated by RCN1, we divided patients with glioma into two different groups based on RCN1 expression. Comparing low- and high-RCN1 patients, genes meeting dual thresholds with |log2FC|> 1 and FDR < 0.05 were identified as differentially expressed genes (DEGs). Concomitantly, RCN1-co-expressed genes were defined using nominal p < 0.05 significance (Figs. 4A, B). Therefore, their shared genes were finally selected, which comprised 653 RCN1-correlated genes, including 258 negatively correlated and 395 positively correlated (Fig. 4C). Their interactions were displayed through a PPI network, with larger nodes representing a closer relationship between RCN1 and its corresponding gene (Fig. 4D). The top ten genes with the largest betweenness centrality and their interactions were also screened (Fig. 4E), among which EGFR, FN1, and IL6 had the highest scores (Fig. 4F).Fig. 4Transcriptomic landscape analysis of RCN1-associated genes in glioma. (A) Differentially expressed genes (DEGs) visualized by Volcano map with |log2FC|> 1 and an adjusted P value < 0.05. (B) Identification of RCN1 co-expressed genes based on the Pearson test. (C) Overlap between DEGs and RCN1-co-expressed genes. (D) Protein–protein interaction (PPI) network of RCN1-correlated genes. (E, F) Top 10 nodes with the largest betweenness centrality in this PPI network. (G, H) Significantly enriched GO (G) and KEGG (H) 10 enrichment pathways Transcriptomic landscape analysis of RCN1-associated genes in glioma. (A) Differentially expressed genes (DEGs) visualized by Volcano map with |log2FC|> 1 and an adjusted P value < 0.05. (B) Identification of RCN1 co-expressed genes based on the Pearson test. (C) Overlap between DEGs and RCN1-co-expressed genes. (D) Protein–protein interaction (PPI) network of RCN1-correlated genes. (E, F) Top 10 nodes with the largest betweenness centrality in this PPI network. (G, H) Significantly enriched GO (G) and KEGG (H) 10 enrichment pathways We next conducted GO and KEGG enrichment analyses. The results indicated that RCN1 may take part in the regulation of ECM and information transmission of cytokines (Figs. 4G-H). In summary, RCN1 may be a significant regulator of the tumor extracellular microenvironment and is closely related to immune functions, where it serves to further influence the properties of the tumor. Analysis of the immune microenvironment of glioma CIBERSORT-based accumulation of 22 immune cell subsets revealed RCN1-associated immune remodeling. Enhanced RCN1 expression correlated with increased immune infiltration, for example elevated CD8+ T lymphocyte abundance. Additionally, we found that some common immune pathways such as immune checkpoint were activated with higher RCN1 expression (Fig. 5B). Figure 5C shows the immune regulators that may be influenced by RCN1, and most of them were positively correlated with RCN1 expression. Additionally, according to the results of “ESTIMATE”, stromal and immune scores exhibited positive correlations with RCN1 expression, while tumor purity was inversely correlated (Fig. 5D). Moreover, checkpoint-associated genes appeared to be differentially expressed between RCN1 expression patterns. Notably, key immunotherapy targets including CTLA-4 and CD274 (PD-L1) demonstrated enhanced expression in tumors with elevated RCN1(Fig. 5E). Collectively, these findings establish RCN1 as a positive modulator of the immune microenvironment in glioma, implicating it in the broad regulation of immune-related pathways. As a supplement, we studied the relevant mutations of RCN1 expression in glioma (Supplementary Figure S8). Meanwhile, according to the TCGA and CGGA databases, (Fig. 5F), and the results of western blot (Fig. 5G), we proved that in high expression samples of RCN1, the expression of PD-L1 increased correspondingly. To further address how RCN1 regulates the behavior of immune cells, we established a co-culture model comprising glioma cells knocked down for RCN1 and THP-1 macrophages (Fig. 5H). Strikingly, RCN1 depletion robustly elevated the expression of M2 macrophage polarization indicators (CD163, CD206, CD301, and ARG1) (Fig. 5I). Consistently, macrophage intrinsic RCN1 depletion also elevated M2 markers (Supplementary Figure S11), suggesting its conserved regulatory role in immune microenvironment homeostasis. Finally, based on the data in TCGA, we also found a significant positive correlation between EGFR and MKI67 genes related to tumor proliferation and prognosis and RCN1(Fig. 5J).Fig. 5Analysis of RCN1 with glioma immune microenvironment. (A) Immune infiltration in glioma samples based on CIBERSORT. (B) Differences in immune function between high- and low-RCN1 expression groups. (C) Spearman correlation between RCN1 and immune-related genes in glioma. (D) RCN1 correlated with a Stromal Score, Immune Score, and Tumor purity. (E) Differential expression of immune checkpoints among RCN1 expression patterns. (F) Validation of RCN1-associated immune checkpoint expression in TCGA and CGGA datasets. (G) PD-L1 Expression after RCN1 knockdown in A172 cells. (H) Diagram detailing the glioma-macrophage in vitro co-culture model. (I) M2-phenotype indicators mRNA expression after co-culturing. (J) Correlation between RCN1 expression and MKI67 and EGFR. (* P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001) Analysis of RCN1 with glioma immune microenvironment. (A) Immune infiltration in glioma samples based on CIBERSORT. (B) Differences in immune function between high- and low-RCN1 expression groups. (C) Spearman correlation between RCN1 and immune-related genes in glioma. (D) RCN1 correlated with a Stromal Score, Immune Score, and Tumor purity. (E) Differential expression of immune checkpoints among RCN1 expression patterns. (F) Validation of RCN1-associated immune checkpoint expression in TCGA and CGGA datasets. (G) PD-L1 Expression after RCN1 knockdown in A172 cells. (H) Diagram detailing the glioma-macrophage in vitro co-culture model. (I) M2-phenotype indicators mRNA expression after co-culturing. (J) Correlation between RCN1 expression and MKI67 and EGFR. (* P < 0.05; ** P < 0.01; *** P < 0.001; **** P < 0.0001) Knocking down rcn1 inhibits tumor cell proliferation To further identify the function of RCN1 in glioma, we constructed glioma cell lines transfected with si-NC, siRCN1-1, and siRCN1-2, including T98 and A172, and checked the knockdown efficiency using western blot and qRT-PCR assay (Supplementary Figure S9A). Subsequently, to evaluate cell viability and proliferative activity, we conducted CCK-8 and EdU assays respectively. The data indicated that both cell viability and proliferation were inhibited after RCN1 knockdown in T98 and A172 cell lines (Figs. 6B-D). The experiments with OE-NC and OE-RCN1 indicated that overexpression of RCN1 induced the proliferation and promoted the viability of tumor cells (Figs. 6B-D, Supplementary Figure S9B).Fig. 6Expression level of RCN1 influenced glioma cell proliferation and apoptosis. (A) GSEA results of related KEGG pathways in the RCN1 high-expression group. (B) cell viability of differrent groups assessed by CCK-8 assay. (C, D) EdU assay to assess the proliferation of the T98, A172 and U118 cell lines. (E Apoptosis rate analyzed by flow cytometry. (F) Expression levels of RCN1, p53, and p-p53 after RCN1 knockdown or overexpression. (G) Expression levels of RCN1 and IRE1α-XBP1s-CHOP axis after RCN1 knockdown in LN229. Quantitative data represent mean ± SD values, with intergroup statistical significance determined by Student’s t-test (* P < 0.05; ** P < 0.01; *** P < 0.001) Expression level of RCN1 influenced glioma cell proliferation and apoptosis. (A) GSEA results of related KEGG pathways in the RCN1 high-expression group. (B) cell viability of differrent groups assessed by CCK-8 assay. (C, D) EdU assay to assess the proliferation of the T98, A172 and U118 cell lines. (E Apoptosis rate analyzed by flow cytometry. (F) Expression levels of RCN1, p53, and p-p53 after RCN1 knockdown or overexpression. (G) Expression levels of RCN1 and IRE1α-XBP1s-CHOP axis after RCN1 knockdown in LN229. Quantitative data represent mean ± SD values, with intergroup statistical significance determined by Student’s t-test (* P < 0.05; ** P < 0.01; *** P < 0.001) Flow cytometry and western blotting were also applied to explore whether RCN1 affected cell apoptosis. Data analysis revealed a marked increase in apoptotic cell proportion following RCN1 depletion (Figs. 6E), while the protein level of cleaved caspase-3 also increased. The apoptotic rate was slightly downregulated in the overexpression group as shown in Supplementary Figure S10. In contrast, immunoblotting analysis demonstrated reduced cleaved caspase-3 intensity upon RCN1 overexpression, with OE-NC controls showing significantly higher levels. According to the results of GSEA enrichment (Fig. 6A), we examined the expression of p53 and phosphorylated p53 after different treatments. The results showed that the P53 levels were higher, while those of phosphorylated P53 were lower in RCN1 knockdown group, while the opposite was observed in overexpression group (Fig. 6F). Given that the natural characteristic of RCN1 is that it is a resident protein within the endoplasmic reticulum (ER) lumen, we hypothesized that these phenotypes and apoptotic changes are essentially due to the disruption of ER homeostasis. Therefore, we analyzed the unfolded protein response (UPR). Specifically focusing on the IRE1α branch due to its known roles in immune modulation, Western blot analysis confirmed that the knockdown of RCN1 significantly impaired the IRE1α-XBP1s-CHOP signaling axis, which was demonstrated by the significant reduction in the expression of IRE1α, XBP1s and CHOP (Fig. 6G). Importantly, treating RCN1-knockdown LN229 cells with the specific IRE1α activator IXA6 partially reversed the elevated apoptosis rate (Supplementary Figure S12). Drug resistance and sensitivity analysis based on rcn1 expression Subsequently, we explore the correlation between the drug sensitivity in glioma patients with glioma and RCN1 expression using the R package “oncoPredict”. Based on this, we obtained the IC50 of different drugs and showed the drugs whose Spearman coefficient with RCN1 expression was > 0.3 or < −0.3. As a result, we found 6 and 23 drugs whose IC50 was positively correlated and negatively correlated with RCN1 expression, respectively (Fig. 7A). Besides, among the 198 drugs included in this algorithm, cisplatin, temozolomide, and irinotecan are commonly used for glioma treatment. Pharmacodynamic analysis established an inverse correlation between RCN1 abundance and chemosensitivity of cisplatin and irinotecan, evidenced with lower IC50 values, rather than temozolomide (Figs. 7B, C). In addition, we next tested the IC50 of cisplatin and irinotecan in glioma cell lines with different RCN1 expression levels. First, we analyzed the expression levels of RCN1 mRNA and protein in common glioma cell lines and normal human brain astrocytes by qRT-PCR and western blot, respectively. The results showed enhanced RCN1 expression in A172 and T98 cell lines, whereas relatively reduced RCN1 levels in U-118MG cell lines (Figs. 7D-F). Concurrently, we quantitatively assessed RCN1 expression at both protein and transcriptional levels in normal brain tissues and clinically stratified glioma specimens through Western blot analysis (Fig. 7H) and qRT-PCR (Fig. 7J). As for the immunohistochemistry staining, our results demonstrated that an enhanced levels of RCN1 expression, which may suggest a poorer prognosis (Fig. 7I). Moreover, the effects of the above two drugs were more obvious in A172 cells with high RCN1 expression, which was consistent with the prediction (Fig. 7G). Considering their unique position in glioma chemotherapy, we studied the relationship between the efficacy of lomustine and carmustine, two commonly used glioma drugs, and the expression of RCN1 (Fig. 7G). IC50 results suggested that the efficacy of both chemotherapy agents decreased to some extent in cells with high RCN1 expression. Therefore, we conducted a combined trial of RCN1 knockdown, using the two drugs at the same time, and preliminarily suggested that RCN1 knockdown might play a role in chemotherapy sensitization of lomustine and carmustine (Figs. 8A-F). The design concept of the entire research is briefly reflected in the Fig. 8G.Fig. 7Correlation between RCN1 expression levels and drug sensitivity. (A) Spearman correlations between RCN1 expression profiles and anti-tumor drug IC50 values. (B, C) Correlations (B) and Boxplots (C) between RCN1 expression levels and the IC50 of cisplatin, temozolomide, and irinotecan. (D, E, F) RCN1 expression in glioma cell lines compared to normal human astrocytes. (G) Dose–response curves for chemotherapeutic agents in glioma cell lines. (H) RCN1 levels in normal brain tissue and glioma samples of different WHO grades with statistical plot. (I) Immunohistochemical results of different grades of glioma patients with RCN1. (J) RCN1 expression in normal brain tissue and LGG, GBM. Quantitative data represent mean ± SD values, with intergroup statistical significance determined by Student’s t-test (* P < 0.05; ** P < 0.01; *** P < 0.001)Fig. 8Knockdown of RCN1 promotes the chemotherapy efficacy of Lomustine and Carmustine. (A, B) RCN1 knockdown enhances Carmustine- and Lomustine-induced cell apoptosis. (C, D, E, F) RCN1 knockdown enhances the anti-proliferative effect of Carmustine and Lomustine in T98 cells. (G) Research flowchart. Quantitative data represent mean ± SD values, with intergroup statistical significance determined by Student’s t-test (* P < 0.05; ** P < 0.01; *** P < 0.001) Correlation between RCN1 expression levels and drug sensitivity. (A) Spearman correlations between RCN1 expression profiles and anti-tumor drug IC50 values. (B, C) Correlations (B) and Boxplots (C) between RCN1 expression levels and the IC50 of cisplatin, temozolomide, and irinotecan. (D, E, F) RCN1 expression in glioma cell lines compared to normal human astrocytes. (G) Dose–response curves for chemotherapeutic agents in glioma cell lines. (H) RCN1 levels in normal brain tissue and glioma samples of different WHO grades with statistical plot. (I) Immunohistochemical results of different grades of glioma patients with RCN1. (J) RCN1 expression in normal brain tissue and LGG, GBM. Quantitative data represent mean ± SD values, with intergroup statistical significance determined by Student’s t-test (* P < 0.05; ** P < 0.01; *** P < 0.001) Knockdown of RCN1 promotes the chemotherapy efficacy of Lomustine and Carmustine. (A, B) RCN1 knockdown enhances Carmustine- and Lomustine-induced cell apoptosis. (C, D, E, F) RCN1 knockdown enhances the anti-proliferative effect of Carmustine and Lomustine in T98 cells. (G) Research flowchart. Quantitative data represent mean ± SD values, with intergroup statistical significance determined by Student’s t-test (* P < 0.05; ** P < 0.01; *** P < 0.001) Discussion RCN1 is a surface-associated protein encoding an ER-resident Ca2+-binding domain (Kurpińska et al. 2019), which is expressed in most fetal and adult organs, including central nervous system (Fukuda et al. 2007). In human diseases, RCN1 is associated with the prognosis of multiple types of cancers and is considered a biomarker for several tumors, including non-small cell lung cancer, oral squamous cell carcinoma, and renal cell carcinoma (Giribaldi et al. 2013; Chen et al. 2019; Ueda et al. 2020). To elucidate the role of RCN1 in pan-cancer, this study analyzed its expression patterns, prognostic implications, and mechanistic pathways, with an emphasis on its crosstalk with the tumor immune microenvironment. Our investigation first examined the correlation between RCN1 expression levels and various clinical parameters. Most previous studies have considered RCN1 as a risk factor for promoting tumor development, including non-small cell lung cancer, hepatocellular carcinoma, and laryngeal cancer (Fu et al. 2021; Wang et al. 2021; Zhang et al. 2021). Our results confirmed this by identifying RCN1 as a risk factor in most cases. The Immunohistochemical analysis further confirmed that elevated RCN1 levels is in almost tumor types relative to normal tissues. We also found that RCN1 was highly associated with the malignant progression of gliomas, both LGG and GBM. However, for certain tumors in the abdominal cavity, such as in UCEC, READ, and COAD, RCN1 expression was negatively correlated with clinical malignant indicators, especially in UCEC, which may be caused by the unique microenvironment and other features. Additionally, RCN1 not only promotes tumor development at the molecular and cellular level but also was found to be take part in the survival of patients in pan-cancer. Recent research indicates that RCN1 overexpression was correlated with unfavorable clinical outcomes in non-small cell lung cancer (Chen et al. 2019). Therefore, we analyzed its hazard ratio based on the TCGA database, revealing it to be a poor prognostic indicator in most cases. Besides, we noticed close correlations between the hazard ratio and RCN1 expression in GBM and LGG. This indicates that the effect of RCN1 on the survival time of patients with glioma is the most typical, which was confirmed by clinical patient data validation from multiple databases. Therefore, RCN1’s role as a prognostic indicator in pan-cancer can be largely understood by studying its relevant mechanisms in glioma. Based on the important role of RCN1 in malignant tumors, especially gliomas, its function and related pathways were identified and researched. The results of enrichment analyses indicated that RCN1 was closely associated with immune-related pathways, including the differentiation processes of Th1, Th2, and Th17 cells, which might be regulated by the extracellular matrix and cytokines. Th1 cells secrete IFN-γ, an important cytokine in macrophage activation (Shapouri-Moghaddam et al. 2018). Th2 cells also interact with macrophages to produce IL-4 and IL-13, promoting the conversion of M1 to M2. M1 macrophages are pro-inflammatory, while M2 macrophages inhibit inflammation and promote cancer development (Martinez et al. 2008). Based on the predictive results, high RCN1 expression is closely associated with M1 polarization of macrophages. Interestingly, our co-culture model indicated that reduced RCN1 expression in tumor cells significantly elevated the expression levels of M2 macrophage polarization markers (CD163, CD206, CD301, and ARG1). Given that RCN1 depletion profoundly disrupts ER homeostasis and induces tumor cell apoptosis, we hypothesize that this robust in vitro M2 polarization may be a compensatory response. Previous literature indicates that acute apoptosis can act as a potent signaling source, polarizing macrophages toward a tissue-clearance M2 phenotype to scavenge cellular debris (Boada-Romero et al. 2020). Furthermore, in glioma and cancers including THYM, MESO, and DLBC, RCN1 expression was significantly correlated with mononuclear phagocytes infiltration. Previous study on the microglia concluded that RCN1 could serve as a phagocyte ligand to bind to the surface of apoptotic neurons, and eventually facilitate microglial phagocytosis (Ding et al. 2015). This suggests that RCN1 is a dynamic regulator whose depletion potentially reshapes the local immune microenvironment. In contrast, our data indicate that established high-RCN1 tumors in vivo are not primarily characterized by M2 polarization, but rather maintain an immunosuppressive microenvironment via the RCN1-mediated upregulation of the PD-L1/CTLA4 checkpoint axis. Furthermore, RCN1 transcriptional activity demonstrated notable co-regulation with key immune checkpoint stimulators including CTLA4 and CD274. CD274, also termed PD-L1, binds with PD1 to inhibit the anticancer effects of T cells (Cha et al. 2019). Similarly, CTLA4, as a negative regulator of T cells, controls T-cell activation by competing with the co-stimulatory molecule CD28 (Chen and Mellman 2017). This finding resolves the apparent divergence from typical oncogenes, indicating that the close relationship between RCN1 and immune checkpoints provides a critical reference for guiding immune-targeted therapy. According to GSEA, RCN-mediated promotion of cancer cell proliferation involved many possible mechanisms and pathways, including the p53 pathway, a key regulator involved in controlling apoptosis and cell cycle progression in cancer (Aubrey et al. 2018). The phosphorylation of p53 leads to changes in the level of apoptosis. Our study preliminarily confirmed that RCN1 knockdown inhibited the phosphorylation of p53, and significantly increased the level of tumor cell apoptosis. This suggests that RCN1 may protect tumor cells from programmed death by regulating p53 phosphorylation. From another aspect, the level of cleaved-Caspase3, which is a critical determinant of apoptosis induced by a wide variety of apoptotic signals (Choudhary et al. 2015), was significantly upregulated after RCN1 knockdown. These findings indicate that RCN1 may regulate cancer cell proliferation through p53-induced apoptosis. Importantly, this RCN1-regulated apoptosis is not merely an isolated intracellular event but actively shapes the immune microenvironment. Since RCN1 can serve as a ligand to facilitate macrophage phagocytosis of apoptotic cells (Ding et al. 2015), this p53-dependent process acts as a signaling source that further links our mechanistic findings with the observed immune remodeling. Previous studies have shown that RCN1 is associated with drug resistance in hepatocellular carcinoma and uterine cancer, indicating its significance in guiding clinical therapy (May et al. 2014; Wang et al. 2021). This investigation examined the responsiveness of glioma cells to pharmacological agents correlated with the expression levels of RCN1. Temozolomide, cisplatin, and irinotecan are common chemotherapy drugs for patients with glioma and have shown great success either alone or in combination (Nicholas et al. 2005). However, in clinical application, different patients have different sensitivities to chemotherapy drugs. According to the R package “oncoPredict”, the IC50 of cisplatin and irinotecan was negatively correlated with RCN1 expression, as verified through cell culture, confirming that cisplatin and irinotecan are more suitable for patients with glioma with high RCN1 expression. Of note, RCN1 exhibited opposing correlations with different chemotherapeutic agents: high RCN1 expression sensitized glioma cells to cisplatin and irinotecan, but conferred resistance to carmustine and lomustine. This dichotomy can be explained by the drugs’ differing mechanisms. Carmustine and lomustine induce severe protein misfolding and ER stress. As shown in Fig. 6G, RCN1 maintains the IRE1α-XBP1s-CHOP pathway. High RCN1 expression utilizes this pathway to clear misfolded proteins, allowing cells to survive and develop resistance. In contrast, cisplatin and irinotecan primarily target DNA replication. Our data indicates that RCN1 promotes rapid cell proliferation, supported by its modulation of p53 and positive correlation with MKI67. This fast replication process makes high-RCN1 cells significantly more sensitive to DNA-targeting agents. These findings suggest that RCN1 represents a significant indicator for projecting rational therapy strategies. Given that RCN1 is functionally considered to be closely associated with the unfolded protein response (UPR), we investigated the effect of RCN1 knockdown on the regulation of endoplasmic reticulum (ER) stress in tumor cells. The results demonstrated that downregulation of RCN1 expression led to decreased expression levels of IRE1α, XBP1s, and CHOP. It is important to note that the UPR comprises three parallel arms: IRE1α, PERK, and ATF6. In this study, we prioritized RCN1’s coordination of the IRE1α branch due to its well-established role in regulating PD-L1 expression and immune cell homeostasis (Cubillos-Ruiz et al. 2015). Because CHOP can also be activated downstream of the PERK-ATF4 axis, determining whether RCN1 concurrently alters the PERK and ATF6 arms remains a crucial subject for future investigations. Nevertheless, our data suggest RCN1 distinguishes itself as a Ca2⁺-binding protein that regulates the IRE1α-XBP1s-CHOP axis. As previous studies indicated that ER function alterations dynamically reprogram immune cells (Chen and Cubillos-Ruiz 2021; Shen et al. 2025), our results suggest that this unique ER-resident protein influences critical immune functions by bridging calcium homeostasis with UPR-mediated transcriptional reprogramming. These include the regulation of immune checkpoints such as PD-L1 and the polarization of macrophages. Notably, PD-L1 also represents an important target for TME-oriented immunotherapy (Liu et al. 2026), further supporting the therapeutic potential of targeting RCN1 in combination with immune checkpoint blockade. This finding, further corroborated by our IXA6 rescue experiments, indicates an impaired regulatory capacity of the IRE1α-XBP1s axis, thereby establishing a causal mechanism by which RCN1 knockdown enhances tumor cell death induced by chemotherapeutic agents such as carmustine. This study has certain limitations. First, the investigation was primarily validated in glioma, particularly glioblastoma. However, RCN1 is also predicted to exhibit oncogenic functions in lung cancer and various kidney cell-derived tumors, where the underlying mechanisms regulating tumor progression and immunity may differ. Meanwhile, the observed changes in IC50 during drug sensitivity analysis may arise from general stress responses or off-target effects of siRNA. Additionally, the specific cytokines mediating the crosstalk in the tumor-macrophage co-culture system remain to be profiled. Most importantly, our mechanistic and chemosensitivity findings are currently restricted to 2D in vitro cell culture models which cannot fully recapitulate the complex physiological and immunological architecture of a living organism. These necessitate subsequent validation through in vivo experiments, large-scale clinical samples from chemotherapy cohorts, and comprehensive secretome analyses. In particular, the development of RCN1-targeted therapeutic strategies could benefit from advanced drug delivery systems. For instance, biomaterial-based approaches such as anti-swelling and ROS-scavenging nanofiber hydrogels (Luo et al. 2025) and engineered extracellular vesicles loaded in ROS-responsive hydrogels (Pan et al. 2025) have shown promising results in modulating local immune environments and could be adapted for localized delivery of RCN1 modulators in glioma. Some findings are based on predictions and require further corroboration through molecular biology experiments. These aspects will constitute the focus of our subsequent research. In conclusion, we analyzed the functions of RCN1 at several levels and found it to have great research value in the development, prognosis, immune microenvironment, and drug sensitivity of pan-cancer. This work integrated the existing RCN1-related cancer research and provided a reference for further study of the value of RCN1 in other common tumors, including its possible mechanisms and pathways. Our findings enhance the comprehension of the roles of RCN1 in cancers, supporting the formulation for clinical chemotherapy strategies. Conclusion This study identified RCN1 as an oncogene in most cancers. Pan-cancer analysis showed that RCN1 expression correlates with immune infiltration and tumor immune microenvironment remodeling. In glioma, RCN1 regulates endoplasmic reticulum stress and apoptosis in vitro; its knockdown enhances chemosensitivity to lomustine and carmustine in cell models. Collectively, RCN1 is a potential key modulator of tumor biology and immunity, and targeting it holds promise for future clinical translation that requires future in vivo validation. Supplementary information Below is the link to the electronic supplementary material.Supplementary file1 (ZIP 21783 KB) Supplementary file1 (ZIP 21783 KB)