We obtained the TCGA Pan-Cancer dataset, which is a harmonized and standardized collection of data from various cancer types in UCSC (https://xenabrowser.net/). The data on the expression of the PRNP gene (ENSG00000171867) was extracted from this dataset for each sample. Subsequently, the samples were filtered based on their source, and we applied a log2 (x+1) transformation to normalize the expression values. A criterion where the expression level was 0 for the samples was utilized to exclude samples. Furthermore, we performed the same log2 (x+1) transformation and excluded cancer types that were affirmed to have <3 available samples.

The R packages “survminer” and “survival” were utilized to perform Kaplan-Meier and Cox regression analyses in order to investigate the impact of PRNP on the prognosis of patients. The evaluation focused on OS. A univariate Cox proportional hazards regression model was employed to determine the hazard ratio (HR) for mortalities linked to the expression of PRNP. The adjusted HR for the expression of PRNP was estimated by the constructed multivariate Cox model. P-values < 0.05 denoted statistical significance.

The PPI network of PRNP was constructed by the STRING database (https://string-db.org/) and enrichment analysis, applying a 0.4 threshold of the minimum interaction score. For the analysis of molecules associated with PRNP, ClusterProfiler packages were used to perform GO and KEGG analyses.

The level of immune cell infiltration for each tumor was examined depending on gene expression using the “donvo_CIBERSOR” method from the R package IOBR. Infiltration scores of cancer-associated fibroblasts (CAFs), B cells, CD4_T cells, CD8_T cells, Macrophages, NK cells, Endothelial cells, and other cells were reassessed in each tumor using the “donvo_epic” method from the R package IOBR. Furthermore, the B cell infiltration score, T cell CD4 infiltration score, T cell CD8 infiltration score, Neutrophil infiltration score, Macrophage infiltration score, and DC (Dendritic Cell) infiltration score were re-examined for each patient’s tumor as per gene expression using the “Timer” approach that is provided by the R package IOBR. PRNP expression data along with 60 marker genes representing 24 Inhibitory and 36 Stimulatory immune checkpoint pathway genes were analyzed in every sample. The “TMB function” provided by the R package was used to compute the TMB of each tumor. An analysis was then performed to calculate individual tumors’ TMB value by combining TMB and gene expression data from samples through the TMB function that is integrated into the R package. Moreover, a previous study availed MSI scores for each tumor (Bonneville et al., 2017).

The National Collection of Authenticated Cell Cultures (Shanghai, China) Supplied the Human CRC cell lines HCT116 and HT29. CCK-8 was purchased from GlpBio (United States), and GEM HCl was obtained from Sigma-Aldrich Chemical Co., Ltd. Materials vital in cell culture, including McCoy’s 5A and fetal bovine serum (FBS) were purchased from Procell Life Science & Technology Co., Ltd. (Wuhan, China) and Serana (Germany). Piperstreptomycin and the Transwell chamber were obtained from Beijing Solarbio Science & Technology Co., Ltd. (Beijing, China) and Merck Millipore (Billerica, Massachusetts, United States), respectively. Cytosine arabinoside (Ara-C) was obtained from Ararat (Canton) Biotechnology Co., Ltd. (Guangzhou, China). PRNP overexpression lentivirus (LV-PRNP), blank control vector lentivirus (LV-control), shRNAs targeting the PRNP gene (shRNA-PRNP), and negative control (shRNA-control) were purchased from Shanghai Nuobai Biotechnology Co., Ltd (NM_000311.5). Supplementary Table S1 shows the sequences of shRNAs. The other reagents used met the analytical grade.

The human CRC cell lines HCT116 and HT29, which exhibit varying levels of PrPC expression, were acquired from the National Collection of Authenticated Cell Cultures and were cultured in McCoy’s 5A medium supplemented with 10% FBS, penicillin (100 U/mL), and streptomycin (100 μg/mL) (P/S). The cells were maintained in the following conditions: a temperature of 37°C and a 5% CO2 humidified atmosphere.

HCT116 in conjunction with HT29 tumor cells was cultured after attaining the logarithmic growth phase. Cells were introduced to each well of a 6-well plate at a density of 5 × 10⁴ cells/mL. The wells were then grouped into four: PrPC-, PrPC+, PrPC- control, and PrPC+ control. Following a 12-h culture period, the cells were infected with the lentiviral vector (PrPC- group add shRNA-PRNP lentiviral vector; PrPC+ group add LV-PRNP lentiviral vector; PrPC- control group add shRNA-control lentiviral vector; PrPC+ control group add LV-control lentiviral vector) when they reached approximately 30%–50% confluence. To ensure consistent infection rates, a multiplicity of infection ratios of 100, 10, 10, and 10 were respectively used for each group. Equal amounts of virus particles, empty vector constructs, and culture medium were added to their corresponding wells. After incubating for another 12 h, 0.5 ug/mL concentration of puromycin was added to each group of supernatant. And then after 12 h of cultivation, the supernatant was discarded and replaced with a complete medium. Four days post-transfection, cell counting was performed using bright field microscopy while the count of green fluorescent protein-positive cells was carried out under fluorescent microscopy. Transfection efficiency was calculated based on these counts. Protein detection analysis was conducted once the transfection efficiency exceeded >80% (Wang S. et al., 2012).

The human CRC cell lines HCT116 and HT29, with different expression levels of PrPC (5 × 10⁵ cells per well) were seeded onto 96-well plates. They were seeded when they were in their logarithmic growth phase and incubated in McCoy’s 5A containing 10% FBS and P/S at 37°C with 5% CO2 saturation. After being cultured in a medium for 6, 12, and 24 h, these CRC cell lines were incubated with 20 μL CCK8 solution for 4 h at 37°C, and the absorbance value at 490 nm was then recorded.

The human CRC cell lines HCT116 and HT29, which exhibited varying levels of PrPC expression, were cultured in six-well plates at a density of 5 × 10⁵ cells per well. After completing 24 h of incubation at 37°C, the cells were subjected to mechanical injury by creating two acellular lanes measuring 1 mm in width using plastic pipette tips and Ara-C (0.2 ug/mL) was added to the medium. Subsequently, any detached cell residue was rinsed from the wells. Changes in the wound area over time were monitored utilizing an inverted microscope, and scratch width was taken at the 24-h mark by ImageJ software as per a previously established protocol (Miyazaki et al., 2015).

The experiments on cell migration assays were conducted following previously described methods (Jia et al., 2017). HCT116 and HT29 cells were cultured for 24 h in a medium that did not have serum. To investigate the migratory behavior of human CRC cell lines HCT116 and HT29 with varying levels of PrPC expression, Transwell cell culture chambers with an 8 mm pore size (BD Bioscience) were employed. In brief, 100 μL of serum-free medium with 80,000 cells and Ara-C (0.2 ug/mL) was introduced to the upper chamber, while 600 μL of culture medium with 20% FBS and Ara-C (0.2 ug/mL) was introduced into the lower chamber. Following a 24-h incubation period, the upper chamber was taken off, and any nonmigrating cells from its surface were gently removed by wiping them off using a cotton swab. The cells that migrated were then fixed utilizing methanol for 20 min, stained with Giemsa for 1 h, and subsequently counted under a microscope. In the invasion experiment, the upper chamber, seeded with 5 ×10⁴ cells and 1% FBS-supplemented medium with Ara-C (0.2 ug/mL), was coated with Matrigel (BD Bioscience); meanwhile, the lower chamber had a 20% FBS-supplemented medium with Ara-C (0.2 ug/mL). After being incubated at 37°C with a CO2 concentration of 5% for 24 h, the Transwell chamber was retrieved, and its contents were discarded before being rinsed with calcium-free PBS solution. Afterward, the cells were treated using methanol and to ensure fixation the treatment was allowed to last for 30 min, followed by staining utilizing a solution containing 0.1% crystal violet for 20 min. Any cells that did not migrate were gently eliminated using a cotton swab and subsequently enumerated under microscopic observation.

As previously described, N-cadherin, E-cadherin, Vimentin and Snail proteins in the CRC cells (HCT116 and HT29) were determined using western blot analysis (Subramanian et al., 2012). SDS-PAGE separated equivalent amounts of proteins, followed by transferring them onto a 4 PVDF membrane. A 5% skimmed milk blockaded the proteins, along with incubation of the proteins with the primary antibody (PrPC, 1:2,000; Cat. No: 12555-1-AP, Proteintech, United States and N-cadherin, 1:1,000, Cat. No: AF5239, Affinity Biosciences, United States and E-cadherin, 1:1,000, Cat. No: AF0131, Affinity Biosciences, United States and Vimentin, 1:1,000, Cat. No: AF7013, Affinity Biosciences, United States and Snail, 1:1,000, Cat. No: AF6032, Affinity Biosciences, United States) in PBST for 2 h at room temperature. The membrane was incubated with the secondary antibody (Goat Anti-Rabbit IgG HRP, 1:10,000, Cat. No: SA00001-2, Proteintech, United States) in PBST for 1 h. β-actin (1:10,000, Cat. No: 20536-1-AP, Proteintech, United States) was the internal control. ECL kit from Beyotime was utilized to visualize the proteins.

The levels of PRNP mRNA expression in 26 distinct types of cancer were initially examined to assess the significance of PRNP in malignancies. This analysis was conducted using a comprehensive dataset obtained from the TCGA Pan-Cancer (PANCAN) project within the UCSC database. Our findings revealed that PRNP is significantly upregulated in most cancers, with only a few exceptions (Figure 1). To further ascertain the prognostic value of PRNP expression across various cancer types, a survival analysis was conducted employing a Cox proportional hazards model. The results affirmed that heightened levels of PRNP expression were linked to poorer OS rates, specifically in COAD, READ, and COADREAD (Figure 2A). These observations were subsequently confirmed through Kaplan-Meier survival analysis (Figures 2B–D).

We aimed to explore the potential of PRNP serving as a therapeutic target for immunotherapy in cancer by investigating its association with tumor immunity. Recognizing the crucial role of TME in tumorigenesis and progression, various algorithms (CIBERSORT, EPIC, and TIMER) were utilized to evaluate the correlation between PRNP expression and immune cell infiltration levels. Findings from these algorithms indicated a positive relationship between PRNP expression and several cell types, including macrophages, neutrophils, CAFs, and CD8⁺ T cells. Conversely, a negative link was observed with plasma cells and regulatory T cells (Figure 3).

Immune checkpoints, a crucial mechanism used by tumor cells to avoid detection and attack by T cells, were the focus of our study. We aimed to explore the associations between PRNP and 60 common immune checkpoint expressions (24 inhibitory and 36 stimulatory) across various cancers. Interestingly, positive correlations were found between PRNP and a majority of immune checkpoint molecules in nearly all types of cancer, particularly with CD274 and C10orf54, as shown in Figure 4A. Strong associations were observed between the expression level of PRNP and MSI as well as TMB across distinct cancer types. Specifically, a positive correlation was identified between PRNP and TMB in HNSC and COAD, while in KIRC, STAD, READ, KIPA, and GBML, it showed a negative correlation. Furthermore, a positive link between PRNP and MSI was found in GBML, THCA, HNSC, and UCEC; conversely, a negative correlation with MSI was observed in STAD, PRAD, LUSC, STES, KIPA, ESCA, PAAD, and LUAD, as depicted in Figures 4B, C.

A PPI network for PRNP was generated, and the STRING was utilized to detect the top 10 associated genes (Figure 5A). GO and KEGG analyses for the genes followed. The KEGG analysis affirmed a significant enrichment of these genes in pathways such as Prion diseases, Viral myocarditis, and Phospholipase D signaling pathway, among others (Figure 5B). Additionally, the GO analysis demonstrated the enrichment of these genes in cellular responses to organonitrogen compounds, nitrogen compounds, and cellular calcium ion homeostasis, among others (Figure 5C).

The transfection efficiency of PrPC- and PrPC+ was analyzed by western blot (Figure 6A). Results showed a statistically substantial decline in the expression of PrPC in HCT116 cells of the PrPC- group compared with the PrPC- control group (p < 0.05, Figure 6B). Moreover, a statistically substantial increase in the expression level of PrPC was observed in the PrPC+ group compared to the PrPC+ control group (p < 0.01). No significant variation was deduced between the PrPC- control group (or PrPC+ control group) and the blank control group (p > 0.05). Similarly, the expression of PrPC in HT29 cells (Figure 6C) decreased remarkably in the PrPC- group (p < 0.001) and increased significantly in the PrPC+ group when compared to the corresponding control group (p < 0.01).

HCT116 and HT29 human CRC cell lines were cultured for 6, 12, and 24 h. A CCK-8 assay explored the influence of different PrPC expression levels on CRC proliferation. As shown in Figure 7A, after 6 h of culturing HCT116 cells, no significant difference in cell proliferation was affirmed between the PrPC- control group and the PrPC- group (p > 0.05). Nevertheless, after 12 and 24 h of culturing, the proliferation capacity decreased by 28.54% and 30.74% in PrPC- group compared to control group (p < 0.05, p < 0.01). Conversely, the proliferation ability of HCT116 cell lines significantly increased after 6, 12, and 24 h of culturing when the PrPC gene was overexpressed, and the proliferation capacity increased by 27.05%, 16.72%, and 15.07% in PrPC+ group compared to control group (p < 0.01, p < 0.01, p < 0.05). Similar findings were noted in HT29 cells, after 12 and 24 h of culturing, the proliferation capacity decreased by 29.33% and 33.46% in PrPC- group compared to control group (p < 0.01, p < 0.05). On the contrary, the proliferation capacity increased by 180.9%, 113.27%, and 39.89% in PrPC+ group compared to control group (p < 0.001, p < 0.001, p < 0.01) (Figure 7B).

In-vitro cell migration assays were utilized to explore the effect of different PrPC expression levels on CRC migration (Figure 8A). As shown in Figure 8B, results from the Wound healing assay indicated a significant decrease in the migration ability of HCT116 cell lines after 12 and 24 h of cell culture when the PrPC gene was silenced, and the migration ability decreased by 11.72% and 19.99%, respectively (p < 0.05, p < 0.01). Conversely, the migration ability of HCT116 cell lines significantly increased after 12 and 24 h of cell culture when the PrPC gene was overexpressed, and the migration ability increased by 37.31% and 57.42%, respectively (p < 0.05, p < 0.001). Similar outcomes were found in HT29 cells, after 12 and 24 h of cell culture, the migration ability decreased by 11.49% and 26.68% in PrPC- group compared to control group (p < 0.01, p < 0.05). On the contrary, the migration ability increased by 38.06% and 60.40% in PrPC+ group compared to control group (both p < 0.01) (Figure 8C). The Transwell cell migration assay was employed to validate the findings of the Wound healing assay regarding the effects of PrPC on CRC (Figure 9A). Data analysis revealed a substantial decline in the number of migrating HCT116 cells in the PrPC- group in comparison to that in the PrPC- control group, and the migration ability decreased by 26% (p < 0.001). Conversely, the number of cells migrating in the PrPC+ group was markedly heightened compared to that in the PrPC+ control group, and the migration ability increased by 138% (p < 0.001) (Figure 9B). Similar trends were observed in HT29 cells, the migration ability decreased by 17.30% in PrPC- group compared to control group (p < 0.05), and the migration ability increased by 75.50% in PrPC+ group compared to control group (p < 0.01) (Figure 9C).

In-vitro cell invasion assays were conducted to explore the influence of varying PrPC expression levels on CRC invasion (Figure 10A). As presented in Figure 10B, the results demonstrated a decrease that is significant in the invasion ability of HCT116 cells in the PrPC- group compared to that in the PrPC- control group, and the invasion ability decreased by 6% (p < 0.01). In contrast, the invasion ability of the PrPC+ group was remarkedly increased compared to that of the PrPC+ control group, and the invasion ability increased by 42% (p < 0.01). Similar patterns were observed in HT29 cells, the invasion ability decreased by 6.33% in PrPC- group compared to control group (p < 0.01), and the migration ability increased by 56% in PrPC+ group compared to control group (p < 0.05) (Figure 10C).

In Figure 11A, the expression of EMT-related proteins (N-cadherin, E-cadherin, Vimentin and Snail) in CRC cell lines with varying PrPC expression levels was detected using western blot. The results (Figures 11B–I) indicated that silencing the PrPC gene led to a declined expression of N-cadherin (both p < 0.001), Vimentin (both p < 0.05) and Snail (both p < 0.001) and a surge in the expression level of E-cadherin (p < 0.05, p < 0.01) in HCT116 and HT29 cells. Conversely, when the PrPC gene was overexpressed, the expression of N-cadherin (p < 0.01, p < 0.001), Vimentin (both p < 0.05) and Snail (p < 0.01, p < 0.001) increased, and E-cadherin decreased (both p < 0.01) in both CRC cell lines. No significant difference was observed in the expression of N-cadherin, E-cadherin, Vimentin and Snail between the blank group and the control group.