Differential expression of P4HB in pan-cancer analysis

TIMER 2.0 was used to analyze the differential expression of P4HB in pan-cancer. As shown in Fig. 1a, the expression of P4HB was upregulated in Bladder Urothelial carcinoma (BLCA), Breast invasive carcinoma (BRCA), Cervical squamous cell carcinoma and endocervical adenocarcinoma (CESC), Cholangiocarcinoma (CHOL), Colon adenocarcinoma (COAD), Esophageal carcinoma (ESCA), Glioblastoma multiforme (GBM), Head and Neck squamous cell carcinoma (HNSC), KIRC, Liver hepatocellular carcinoma (LIHC), Lung adenocarcinoma (LUAD), Lung squamous cell carcinoma (LUSC), Prostate adenocarcinoma (PRAD), Rectum adenocarcinoma (READ), Uterine Corpus Endometrial carcinoma (UCEC) compared to the corresponding adjacent normal tissues. The results of combined UCSC XENA datasets and TCGA datasets further validated the aberrant expression of P4HB in pan-cancer (Fig. 1b). To investigate the relationship between P4HB expression and molecular subtypes in each tumor, we performed an online analysis with TISIDB. In Fig. 1c, elevated P4HB expression is significantly correlated with HM-indel status in COAD and HER-2 status in BRCA tumors.

Fig. 1

RNA expression levels of P4HB in pan-cancer. a Differential expression of P4HB across all TCGA tumors between the tumor and adjacent normal tissue. b Differential expression analysis between tumor and normal tissues using UCSC XENA dataset and TCGA dataset. c Molecular subtype analysis of multiple tumor types using TISIDB. *P < 0.05, **P < 0.01, ***P < 0.001

For tumors (COAD, UCEC, PRAD, BRCA, LUSC, BLCA, LIHC, KIRC, LUAD) differentially expressing P4HB, we performed box plotting using GEPIA2 (Fig. 2). Subsequently, the protein expression level of P4HB was verified by the HPA database. Most of the cancers showed moderate immunoreactivity. Liver, colon, kidney and endometrial cancers showed strong staining in several cases. Testicular cancer, gliomas and lymphomas were negative or weakly stained in a majority of cases (Fig. 2). Immunohistochemical staining showed that P4HB protein was mainly located in the cytoplasm or cell membrane of the cells.

Fig. 2

P4HB protein expression levels in tumors with differential expression of P4HB. a–i Box plots (left) demonstrated the RNA level of tumors differentially expressing P4HB using GEPIA2. Cutoff value is set to |Log2FC| > 1, P value < 0.05. The HPA database was used to analyze the protein expression level (right) of P4HB in a COAD; b UCEC; c PRAD; d BRCA; e LUSC; f BLCA; g LIHC; h KIRC; i LUAD. *P < 0.05

Survival analysis of P4HB expression in pan-cancer

To investigate the prognostic value of P4HB in pan-cancer, we performed a comprehensive survival analysis. KIRC patients with elevated expression of P4HB had poorer OS, DSS and PFI than those with low expression. In addition, BLCA patients with high expression of P4HB had poorer PFI. However, COAD patients with high expression of P4HB had better DFS compared to the group with low expression of P4HB (Fig. 3, Figs. S1, S2). In summary, P4HB may be a potentially specific prognostic biomarker for patients with KIRC.

Fig. 3

Overall survival analysis of P4HB in tumors. a–i Survival analysis comparing P4HB high and P4HB low expression groups in a BRCA; b BLCA; c UCEC; d PRAD; e LUSC; f LUAD; g LIHC; h COAD; i KIRC. ​The survival outcomes were analyzed using the log-rank test

Clinical relevance and prognostic role of P4HB in patients with KIRC

The baseline characteristics of KIRC patients in TCGA based on P4HB expression are summarized in Table 1. In KIRC patients, high expression of P4HB was strongly associated with advanced T-stage, positive lymph nodes, distant metastases, advanced TNM stage, and advanced histological stage. However, no obvious correlation between P4HB expression and the age and gender of KIRC patients were observed (Fig. 4a–g). Subsequently, we performed ROC analysis to investigate the diagnostic value of P4HB expression at T, N, M, TNM, histological stage and tumor status. The results showed that P4HB had good diagnostic value for tumor status (AUC = 0.951 95% CI 0.927–0.975). It also had diagnostic value for positive lymph nodes (AUC = 0.728 95% CI 0.610–0.846) (Fig. S3).

Table 1 Baseline characteristics of KIRC patients in TCGA according to P4HB expressionFig. 4

Clinical relevance and prognostic role of P4HB in KIRC patients. a–g Box plots of P4HB expression between T stages (a); N stages (b); M stages (c); pathological stages (d); histologic stages (e); gender (f); age (g). h Univariate Cox regression analysis of P4HB in KIRC. Prognostic model was constructed using Nomogram (i) and verified by Calibration (j). n.s, not significant, *P < 0.05, **P < 0.01, ***P < 0.001

To explore the risk factors involved in the prognosis of KIRC patients, we performed a univariate Cox regression analysis on the expression of P4HB, as well as other clinical variables. As shown in Fig. 4h, P4HB (HR = 1.582, 95% CI 1.258–1.989, P < 0.001), TNM stage (HR = 3.299, 95% CI 2.342–4.648, P <0.001) and age (HR = 1.765, 95% CI 1.298–2.398, P < 0.001) are risk factors for OS in patients with KIRC.

Furthermore, we performed multivariate Cox regression analysis and found that P4HB (HR = 1.372, 95% CI 1.047–1.681, P = 0.019) was an independent risk factor for OS (Table 2). Based on the results of multivariate cox regression analysis, we integrated multiple clinical variables and plotted the Nomogram (Fig. 4i). We use the calibration curve to assess the difference between the predicted and true values of this Cox regression model. As shown in Fig. 4j, the predicted 3-year and 5-year survival probabilities for this model are close to the true survival probabilities, suggesting that this Nomogram model has clinical prognostic value.

Table 2 Univariate and multivariate cox regression analyses on the expression of P4HB and other clinical variables in KIRCScRNA-seq, functional analysis, and gene set enrichment of P4HB

We applied CancerSEA to explore which functional states P4HB is associated with in pan-cancer at single-cell resolution. Figure 5a is a heatmap of the 14 different functional states of P4HB involved in pan-cancer. The results showed that P4HB was associated with a variety of malignant biological processes, such as being involved in regulating hypoxic processes in renal cell carcinoma (RCC); promoting distant metastasis in non-small cell lung cancer (NSCLC); and mediating DNA repair in BRCA. Figure 5b, c list 10 biological processes in which P4HB is involved in RCC. In addition to hypoxic processes (r = 0.678, P < 0.001), P4HB also regulates angiogenesis (r = 0.517, P < 0.001), regulates differentiation (r = 0.432, P < 0.001), stemness (r = 0.411, P < 0.001), promotes proliferation (r = 0.340, P = 0.002) and metastasis (r = 0.403, P < 0.001) in RCC. However, P4HB expression was negatively correlated with cell cycle (r = −0.271, P = 0.013), DNA repair (r = −0.427, P < 0.001) and inflammation (r = −0.359, P < 0.001) in RCC.

Fig. 5

ScRNA-seq and functional analysis. a Heatmap of the P4HB functional pathways in pan-cancer. b Relationship between P4HB expression and the top 10 enriched pathways in KIRC. c Spearman's method was used to assess the correlation between P4HB expression and enrichment pathways. *P < 0.05, **P < 0.01, ***P < 0.001

We obtained 50 proteins from the STRING database with experimentally validated binding to P4HB and constructed a PPI network (Fig. 6a). Additionally, we used the Metascape database for functional annotation of P4HB related genes and constructed network components (Fig. 6b–d). The Molecular Complex Detection (MCODE) algorithm has been applied to identify densely connected network components. Component 1 is mainly enriched in “protein folding”, “response to endoplasmic reticulum stress” and “protein processing in endoplasmic reticulum”. Component 2 is involved in “hydroxylation of the molecule”. Component 4 participate in “organelle biogenesis and maintenance”. Component 5 regulates “processing of intronless pre-mRNAs”, “processing of capped intronless pre-mRNA” and “mRNA polyadenylation”. Figure 6e lists the top-level Gene Ontology biological processes associated with the P4HB related genes. Interestingly, these genes are enriched mainly in cellular processes, metabolic processes, biological regulation, and immune system processes.

Fig. 6

PPI networks and Gene Ontology analysis. a The PPI network is constructed by the STRING database with 50 proteins which were experimentally verified binding to P4HB. b Densely connected network components had been identified by Molecular Complex Detection algorithm (MCODE). c, d Constructed network components (c) and functional annotation (d) of P4HB related genes using Metascape database. e Top-level Gene Ontology biological processes associated with P4HB-related genes

Figure 7A–D presents the results of GSEA. The results demonstrated that the gene sets were mainly enriched in oncogenic signaling pathways, such as P53 pathway; PTEN pathway; MTOR pathway. Similar to the results of scRNA-seq, the association between P4HB and hypoxia signaling pathway (HIF pathway) was also obtained in this analysis. P4HB gene set was also enriched in cytokines and immune response pathways, such as IL-6 pathway; MMP cytokine connection; sumoylation of immune response proteins. Metabolism-related signaling pathways were also found in this analysis: fatty acid metabolism; glycolysis and iron metabolism. Interestingly, PD-1 pathway and VEGF pathway were enriched in this analysis which indicated that P4HB may be a potential biomarker for ICB or molecular-targeted therapies in KIRC patients

Fig. 7

GSEA functional enrichment analysis of P4HB expression in KIRC (a–d). The Y-axis represents one gene set and the X-axis is the distribution of log FC corresponding to the core molecules in each gene set

Tumor microenvironment and immune checkpoint molecules analysis in KIRC

We aimed to investigate whether P4HB is involved in the tumor microenvironment to regulate the malignancy progression of KIRC. CIBERSORT was used to analyze the immune cell infiltration of KIRC in different P4HB expression groups (Fig. 8a–c). B cells, macrophages and T cells are the main immune infiltrating cells in KIRC (Fig. 8b). The goal of tumor immunotherapy is to activate cytotoxic T lymphocytes (CTLS) in tumors, initiate tumor-specific CTLS in lymphoid organs, and establish an effective immune response to tumors [26]. An important function of CD8+ and CD4+ T cells is to optimize the size and quality of the CTL response and provide a long-term protective immunity [37]. Interestingly, CD8+ T cell infiltration levels were lower in the high P4HB expression group than in the low P4HB expression group (Fig. 8c). Macrophages play a crucial role in cancer development and metastasis. Compared to proinflammatory M1 macrophages which phagocytose tumor cells, anti-inflammatory M2 macrophages, including tumor-associated macrophages (TAMs), promote tumor growth [38]. In this study, we found that high expression of P4HB was significantly associated with higher levels of M2 macrophage infiltration (Fig. 8c). In summary, P4HB is involved in the regulation of the microenvironment of KIRC. Overexpression of P4HB promotes tumorigenesis by up-regulating the infiltration levels of M2 macrophages and down-regulating the infiltration levels of CD8+T cells, resulting in tumor immune escape.

Fig. 8

P4HB expression and tumor microenvironment analysis in KIRC. a Immune cell score heatmap, different colors represent different expression distribution in different samples. b Percentage abundance of tumor infiltrating immune cells in each sample. Different colors represent different types of immune cells. The abscissa represents the sample, and the ordinate represents the percentage of immune cell content in a single sample. c Box plots of immune cell score using CIBERSORT. d Heatmap of immune-checkpoint-related gene expression. The different colors represent the trend of gene expression in different samples. e Expression distribution of immune checkpoints gene in tumor tissues and normal tissues. G1, P4HB high group; G2 P4HB low group. *P < 0.05, **P < 0.01, ***P < 0.001

Cancer therapy has been revolutionized by discovering that overexpression of immune checkpoint molecules in TME plays a crucial role in antitumor immune evasion [39]. Therefore, understanding the relationship between P4HB and immune checkpoint molecules is conducive to exploring the potential mechanisms of P4HB in regulating the TME. In this study, higher expression of P4HB in KIRC was significantly associated with higher expression of immune checkpoint molecules HAVCR2 and PDCD1LG2 compared to lower expression of P4HB. However, no obvious association between P4HB and CD274 (PD-L1) or CTLA4 was found (Fig. 8d, e). Therefore, we hypothesize that P4HB promotes tumor immune escape through immunosuppression induced by the over-expression of HAVCR2 and PDCD1LG2 on T cells.

To further explore the association between P4HB expression level and immune checkpoint blockade (ICB) response in KIRC, we performed tumor mutational burden analysis. There was a positive correlation between P4HB expression and TMB score in the results (Fig. 9a). Furthermore, we analyzed the response to ICB treatment in the P4HB high and low expression groups and found that the P4HB high expression group had a higher response to ICB treatment (Fig. 9b). In addition, we also conducted a comprehensive analysis (chemokines, chemokine receptors and immunostimulators) to thoroughly understand the role of P4HB in immunotherapy response and found that P4HB was positively correlated with CXCR6, CXCL16 and CD70 (Fig S4a-f). It has been reported that CXCR6 and CXCL16 to be involved in the efficacy of anti-PD-1 cancer immune checkpoint therapy [40], and CD70 is an effective target for chimeric antigen receptor T (CAR-T) cell therapy [41].

Fig. 9

TMB and ICB analysis of P4HB in KIRC. a Correlation analysis between P4HB gene expression and TMB was performed using the Spearman’s method. b Statistical table of immune response of samples in different groups in the prediction results (above). The distribution of immune response scores in different groups in the prediction results (below). G1, P4HB high group; G2 P4HB low group. *P < 0.05

Analysis of molecular-targeted therapies

Molecular targeted therapy is the first-line treatment for advanced renal cell carcinoma. Studying the expression of P4HB and IC50 for molecularly targeted therapeutics may help guide clinical treatment strategies. The IC50 of sorafenib, sunitinib, axitinib, imatinib, pazopanib, and erlotinib decreased with the increase of P4HB expression (Fig. 10a–f). It is worth noting that P4HB expression was significantly correlated with the IC50 of sorafenib (r = −0.34, P < 0.001). The results suggest that KIRC patients with high expression of P4HB can derive better drug sensitivity from molecularly targeted therapies, suggesting P4HB as a potential biomarker for molecularly targeted therapy selection.

Fig. 10

Molecular targeted therapy analysis in KIRC. a–f Spearman correlation analysis of P4HB gene expression and IC50 score a Axitinib; b Imatinib; c Sorafenib; d Sunitinib; e Pazopanib; f Erlotinib. The abscissa represents different groups of samples, and the ordinate represents the distribution of the IC50 score. The density curve on the right represents the trend in distribution of the IC50 score, the upper density curve represents the trend in distribution of the gene expression

P4HB promotes the proliferation and migration of RCC cells

We validated the expression level of P4HB in RCC cells and found that P4HB was highly expressed in ACHN and 769-P cells (Fig. 11a). To further explore the role of P4HB in RCC cell proliferation and migration, we transfected siRNA into ACHN and 769-P cells and finally decided to take the next step with si-P4HB#1(Fig. 11b, Fig. S5a). The results showed that knockdown of P4HB inhibited the proliferation and migration of RCC cells (Fig. 11c–f). In addition, we further performed IHC analysis on clinical RCC tissues and found that P4HB expression was significantly higher in RCC than in paracancer tissues (Fig S5b).

Fig. 11

Knockdown of P4HB inhibits cell proliferation and migration of 769-P and ACHN. a P4HB expression was examined in HK-2 and RCC cells lines via quantitative reverse transcription polymerase chain reaction (RT qPCR); b Efficiencies of P4HB inhibition was examined by RT qPCR; c, d The effects of P4HB inhibition on cell proliferation were detected by clone formation in 769-P cell (c) and ACHN cell (d); e, f The effects of P4HB knockdown on migration were observed in769-P (e) and ACHN cell (f) by wound healing assays