Multi-omics analysis identifies PTTG1 as a prognostic biomarker associated with immunotherapy and chemotherapy resistance

Abstract

Background

Pituitary tumor-transforming gene 1 (PTTG1) is an important gene in tumour development. However, the relevance of PTTG1 in tumour prognosis, immunotherapy response, and medication sensitivity in human pan-cancer has to be determined.

Methods

TIMER, GEPIA, the human protein atlas, GEPIA, TISCH2, and cBioportal examined the gene expression, protein expression, prognostic value, and genetic modification landscape of PTTG1 in 33 malignancies based on the TCGA cohort. The association between PTTG1 and tumour immunity, tumour microenvironment, immunotherapy response, and anticancer drug sensitivity was investigated using GSCA, TIDE, and CellMiner CDB. Molecular docking was used to validate the possible chemotherapeutic medicines for PTTG1. Additionally, siRNA-mediated knockdown was employed to confirm the probable role of PTTG1 in paclitaxel-resistant cells.

Results

PTTG1 is overexpressed and associated with poor survival in most tumors. Functional enrichment study revealed that PTTG1 is involved in the cell cycle and DNA replication. A substantial connection between PTTG1 expression and immune cell infiltration points to PTTG1’s possible role in the tumour microenvironment. High PTTG1 expression is associated with tumour immunotherapy resistance. The process could be connected to PTTG1, which mediates T cell exhaustion and promotes cytotoxic T lymphocyte malfunction. Furthermore, PTTG1 was found to be substantially linked with sensitivity to several anticancer medications. Suppressing PTTG1 with siRNA reduced clone formation and migration, implying that PTTG1 may play a role in paclitaxel resistance.

Conclusion

PTTG1 shows potential as a cancer diagnostic, prognostic, and chemosensitivity marker. Increased PTTG1 expression is linked to resistance to cancer treatment. The mechanism could be linked to PTTG1’s role in promoting cytotoxic T lymphocyte dysfunction and mediating T cell exhaustion. It is feasible to consider PTTG1, which is expressed on Treg and Tprolif cells, as a new therapeutic target for overcoming immunotherapy resistance.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-024-13060-5.

Introduction

The pituitary tumor-transforming gene 1 (PTTG1), also known as SECURIN was isolated from pituitary tumors and identified as a gene with tumorigenic effect in vivo. It encodes a homolog of the yeast Securin protein and acts as a sister chromatid regulator that inhibits separase function and thus prevents sister chromatid separation [1, 2]. As a transcriptional activator, PTTG1 can activate C-Myc and up-regulate the expression of cyclinB1 and CDK1 [3]. PTTG1 plays a role in chromosomal stability maintenance, orderly cell cycle monitoring, genetic stability and mitotic fidelity assurance, malignant transformation, tumour induction, invasion, and metastasis [4, 5].

Research studies indicate that overexpression of PTTG1 is associated with unfavorable tumor phenotype and adverse prognosis. This shows that PTTG1 plays a critical role in the occurrence and progression of tumours. For instance, hepatocellular carcinoma has an elevated overexpression of PTTG1, which enhances the synthesis of asparagine to activate mTOR and advance the tumour [6]. Furthermore, it was shown that a higher level of PTTG1 was associated with poorer prognosis for LUAD patients [7], and targeted PTTG1 inhibition improves radiation-induced antitumour immunity in lung adenocarcinoma [8]. Additionally, the predictive values of PTTG1 for immunotherapy response [8] and chemosensitivity [9] were also revealed. Although the importance of PTTG1 in tumors is becoming more widely known, there have been no comprehensive studies of its function in tumor occurrence, recurrence, immunotherapy or chemotherapy.

In this work, we evaluated the gene expression, prognostic value, and relate to immunotherapy response of PTTG1 in many well-established databases. We discovered that PTTG1 is overexpressed in several tumours and is related with poor survival in a variety of malignancies. PTTG1 promotes cytotoxic T cell malfunction, which contributes to immunotherapy resistance. PTTG1 was found to be significantly related with sensitivity to several anticancer medicines, particularly MET inhibitors. The current findings highlighted the critical function of PTTG1 in carcinogenesis, immunotherapy, and chemotherapy response, identifying a new target for cancer treatment.

Materials and methods

Gene and protein expression analysis

The differential expression of PTTG1 in tumor and neighboring normal tissues across all TCGA tumors was analyzed by TIMER (Tumour Immune Estimation Resource, http://timer.cistrome.org/) [10]. The Wilcoxon test was used to determine the statistical significance of differential expression. The PTTG1 gene expression profile and pathological stage plot across all tumor samples and paired normal tissues were analyzed by GEPIA2 (Gene Expression Profiling Interactive Analysis 2, http://gepia2.cancer-pku.cn/#index) [11]. The immunohistochemistry-based protein expression profiles of PTTG1 (Antibody No. HPA008890) were analyzed by the Human Protein Atlas (HPA, https://www.proteinatlas.org/) [12] in LUAD (3 normal, 6 LUAD), LIHC (3 normal, 11 LIHC), KIRC (3 normal, 12 KIRC), and BRCA (2 normal, 12 LUAD). PTTG1 staining intensity scores were calculated as follows: 0 (negative), 1 (weak positive), 2 (moderate positive), and 3 (strong positive). And the staining quantity was scored as 0 (none), 1 (< 25%), 2 (25–75%), and 3 (> 75%). The staining score was defined as staining quantity score multiplied by staining intensity score.

Kaplan–Meier test

Survival analysis of PTTG1 in tumor patients was obtained from the GEPIA2 based on the TCGA datasets. Kaplan-Meier Plotter [13] (http://kmplot.com/analysis/) was used to validate the prognostic value of PTTG1 in NSCLC patients. The Kaplan-Meier method was performed to plot the overall survival (OS) or disease-free survival (DFS, also called relapse-free survival and RFS) curves. Survival differences were assessed by the log-rank test using the median of PTTG1 as a cutoff value. The cox proportional hazard ratio (HR) with 95% confidence intervals and log‑rank P‑values were calculated.

Genetic alteration analysis

The PTTG1 alteration landscape in human cancers was depicted based on TCGA pan-cancer atlas studies by cBioPortal (https://www.cbioportal.org/) [14]. To determine the correlation with copy number variation and survival, patients were divided into delete, amplification, and wild-type groups based on PTTG1 mutation type. The Kaplan-Meier method was performed to plot the overall survival curves in TCGA datasets by GSCA (Gene Set Cancer Analysis, https://guolab.wchscu.cn/GSCA/#/) [15]. The variant functional annotation of PTTG1 SNV was analyzed by the Functional Annotation of Variants Online Resources (FAVOR, https://favor.genohub.org/) [16].

Similar genes detection and PPI network construction

To identify genes with a similar expression pattern to PTTG1 in pan-cancer, the Pearson correlation coefficients were analyzed using the GEPIA2 database. The similar genes were analyzed using the STRING database (https://string-db.org/) to generate the PPI network. PPI pairs with a combined score > 0.4 were extracted. The PPI network was visualized using Cytoscape 3.7.2 software [17], and the most important module was performed using the MCODE plug-in in Cytoscape software. The Metascape web-based tool (https://metascape.org/gp/index.html) [18] was used for functional enrichment analysis (GO analysis and KEGG pathway analysis) of these genes.

PTTG1 and tumor immune microenvironment

The correlation of PTTG1 expression with immune infiltration level in diverse cancer types was analyzed by TIMER (Tumor IMmune Estimation Resource, https://cistrome.shinyapps.io/timer/) [10]. The TCGA pan-cancer data set was retrieved from the UCSC database. The PTTG1 expression data in individual samples was retrieved, and a log2(x + 1) transformation was applied. The immune, stromal, and ESTIMATE scores for each sample were determined using the R package ESTIMATE (https://bioinformatics.mdanderson.org/estimate/rpackage.html). Pearson correlation analysis was used to identify significantly linked immune infiltration scores.

PTTG1 and immunotherapy response

TIDE (Tumor Immune Dysfunction and Exclusion, http://tide.dfci.harvard.edu/) is a computational framework created to assess the potential of tumour immune escape from the gene expression profiles of cancer samples [19]. The ‘Biomarker Evaluation’ module was used to compare PTTG1 to other reported biomarkers in terms of their predictive ability for response outcome and overall survival. The ‘Query Gene’ module calculated the PTTG1 gene signature in T-cell dysfunction using data from the TCGA, PRECOG, and METABRIC databases. The T cell dysfunction score of PTTG1 is defined as the Wald test z score [19]. The two-sided Wald test was used to calculate the relationship between cytotoxic T lymphocyte CTL levels and overall survival. The Kaplan-Meier plot divides tumours into two groups: ‘High CTL’ has above-average CTL values across all samples, and ‘Low CTL’ has values below average. To demonstrate the relationship between CTL levels and survival outcomes, samples were divided based on their PTTG1 expression levels.

Expression level of PTTG1 at the single‑cell level

TISCH2 (Tumor Immune Single-cell Hub 2, http://tisch.comp-genomics.org/) is a scRNA-seq database that focusses on the tumour microenvironment. It provides precise cell-type annotation at the single-cell level, allowing the analysis of TME across various cancer types [20]. The expression level of PTTG1 at the single-cell level was visualized by TISCH2 in the NSCLC_GSE139555, KIRC_GSE111360, and LIHC_GSE140228 datasets. The PTTG1 expression between immunotherapy responders and non-responders was investigated in the LIHC_GSE125449_aPDL1aCTLA4 dataset.

PTTG1 and chemo response

CellMiner Cross-Database (CellMinerCDB, discover.nci.nih.gov/cellminercdb) [21] enables the analysis of molecular and pharmacological data across cancer cell line databases to identify medications that match genomic determinants of response. CellMinerCDB investigated the relationship between PTTG1 expression and chemotherapy response using the GDSC (Sanger/Massachusetts General Hospital Genomics of Drug Sensitivity in Cancer) dataset.

Moleular Docking

The X-ray crystal structure of PTTG1 (7nj0) was obtained from the RCSB PDB protein databank (http://www.rcsb.org/). The 3D formats of ABT-263 (ZINC150338726) and NSC-207,895 (ZINC5180959) were obtained from ZINC15 drug database (http://zinc15.docking.org/) [22]. Discovery Studio 4.5 Client deleted the water and ligand molecules from the crystal structures before starting the docking simulation. PyRx software was used to simulate drug docking and PTTG1 interactions. We calculated interaction energies to predict docking positions and pick the binding pose with the lowest binding energy (kcal mol^− 1). The results were visualised and analysed with Discovery Studio 4.5 Client.

Cell lines

Human lung cancer cell lines A549, H358 and the paclitaxel resistant cell lines A549-TXR, H358-TXR were kindly provided by Professor Wang Luo (University of Michigan, USA). Cells were maintained in RPMI 1640 or F12K supplemented with 10% FBS and 1% antibiotic-antimycotic. All cell lines were cultured at 37 °C in a 5% CO2 cell culture incubator. Mycoplasma contamination was excluded in these cell lines.

Real-time quantitative PCR

Total RNA was reverse transcribed into cDNA using Reverse Transcription Kit (Thermo Fisher Scientific). qRT-PCR was performed using Power SYBR™ Green (Thermo Fisher Scientific) and ABI 7300 detection system (Applied Biosystems). qRT-PCR data were normalized to the expression of housekeeping gene GAPDH, and relative expressions were calculated using the 2^−ΔΔCt method. The oligonucleotide primers were shown as follows: PTTG1 (forward: 5’- ATGAATGCGGCTGTTAAGACCTG-3’, reverse: 5’- TCCCATCTAAGGCTTTGATTGAAGG-3’). All tests were performed in triplicate, and the data were presented as mean ± SD.

siRNA-mediated knockdown

Cells were seeded at the desired concentration in 60 mm plates and then transfected with 10 nM experimental siRNA oligonucleotides or non-targeting controls 24 h later. Lipofectamine^® RNAiMax Reagent (Invitrogen) was used to knockdown cells in OptiMEM medium, following the manufacturer’s recommendations. The knockdown efficiency was measured using qPCR. PTTG1 siRNA sequences for knockdown tests are listed below: sense: 5’-UGUGGUUGCUAAGGAUGGGCUTT-3’; antisense: 5’- AGCCCAUCCUUAGCAACCACATT-3’.

Western blot analysis

Cells were homogenized in RIPA buffer (SIGMA). Concentration of protein was measured using BCA Assay Kit (Beyotime). Equivalent protein was separated by polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membranes (Millipore). After blocking the membranes with 5% nonfat milk, incubating membranes with primary antibodies against Cyclin B1 (1:1000, Cat. 4138, Cell signaling), CDC2 (1:1000, Cat. 9116, Cell signaling), PTTG1 (1:1000, Cat. 18040-1-AP, Proteintech), GAPDH (1:1000, Cat. 2118, Cell signaling) overnight at 4 °C. Membranes were incubated with HRP conjugated secondary antibody (1:2000 dilution) and the signals were detected by MiniChemi 610 Plus (SageCreation Science Co.).

Clone formation assay

A549-TXR and H358-TXR cells (250 per well) were planted in 12-well plates and grown overnight. Cells were transfected with 10 nM experimental siRNA oligonucleotides or non-targeting controls 24 h after plating. After two weeks, the cells were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet. Each well’s clone count was recorded.

Basement membrane migration assays

Cells were treated with the siRNAs listed above to perform invasion tests. After 48 h of transfection, cells were trypsinised, and diluted to the desired concentration. Cells were seeded into basement membrane matrix Boyden chambers (8-mm pore size, BD) located in the insert of a 24-well culture plate. The lower compartment received 20% FBS as a chemoattractant. After 48 h, the non-migrating cells were gently removed using a cotton swab. Cells on the chamber’s lower side were stained with Diff-QuikTM Stain Set (SIEMENS), then air dried and photographed.

Statistical analysis

ANOVA was used to compare PTTG1 expression levels in tumour and normal samples from the TCGA and GTEx datasets. Pearson’s correlation coefficient was used to analyse the correlation between Immuno Score, Stromal Score, and ESTIMATE Score, as well as immunological checkpoints. The P-value < 0.05 was judged significant.

Results

PTTG1 is overexpressed in pan cancer and associated with tumor stages

We performed a gene expression examination for PTTG1 on the gene expression matrix derived from the TCGA datasets using TIMER. PTTG1 mRNA expression was considerably higher in BLCA, BRCA, CHOL, ESCA, HNSC, KICH, KIRC, KIRP, LIHC, LUAD, LUSC, PRAD, READ, STAD, and UCEC, but significantly lower in THCA. The remaining cancer types exhibited no significant variation in PTTG1 mRNA levels between normal and tumor cells (Fig. 1A). The PTTG1 mRNA expression in pan-cancer was then analyzed using RNA sequencing datasets from the TCGA and GTEx projects by GEPIA. In addition to the 14 cancer types listed above, PTTG1 was significantly overexpressed in ACC, CESC, DLBC, GBM, OV, PAAD, SKCM, THYM, and UCS while being underexpressed in LAML and TGCT (Fig. 1B). To validate the diagnostic and prognostic relevance of PTTG1 in pan cancer, we used GEPIA to investigate its expression in tumor pathological stages. The findings revealed that PTTG1 was overexpressed in samples from advanced cancer patients with KIRC, KIRP, ACC, LUAD, LIHC, and BRCA (Fig. 1C). We then obtained the immunohistochemistry staining of PTTG1 from the HPA database. Compared with normal tissues, the PTTG1 protein stained darker in BRCA (P < 0.0001), LIHC (P = 0.0099), and KIRC (P = 0.00052) samples. PTTG1 protein stained darker in LUAD, but the results were not statistically significant (P = 0.097). Weak to moderate cytoplasmic and/or nuclear immunoreactivity was observed in most tumour cells (Fig. 1D).

Fig. 1.

The mRNA and protein expression of PTTG1 in pan-cancer. (A) The mRNA expression level of PTTG1 in tumor tissues and adjacent normal tissues in 33 TCGA dataset. The red or purple plots represent the tumor sample and blue plots represent normal sample, respectively. (B) The mRNA expression level of PTTG1 in tumor and normal tissues from TCGA and GTEx data sets. The red letters represent significantly high expression in tumor and green letters represent low expression, respectively. (C) Pan-cancer analysis of PTTG1 gene expression at different tumor stages. (D) Immunohistochemical staining images of PTTG1 in human cancers and normal tissue from HPA database. *P < 0.05; **P < 0.01; ***P < 0.001

PTTG1 is a predictor of poor patient survival

To investigate the prognostic benefit of PTTG1, we compared survival (OS and DFS) differences between high and low PTTG1 expression groups in pan-cancer. The high PTTG1 expression group had substantially lower overall survival (P < 0.05) compared to the low expression group in ACC, KIRC, KIRP, LGG, LIHC, LUAD, MESO, PAAD, THCA, and UVM (Fig. 2A). Increased PTTG1 expression was linked to decreased disease-free survival in ACC, KIRC, KIRP, LGG, LIHC, MESO, PAAD, PRAD, SARC, and UVM (Fig. 2B). These findings indicate that PTTG1 is overexpressed and linked with poor outcomes in ACC, KIRC, KIRP, LIHC, LUAD, and PAAD, and has the potential to be a predictive biomarker.

Fig. 2.

Survival map of PTTG1 in pan-cancer. GEPIA2 survival analysis including OS. (A) and DFS (B) demonstrated the survival map (upper panel) and Kaplan–Meier survival plots (lower panel) of PTTG1 in TCGA dataset

Associations of PTTG1 expression with genomic alterations in pan-cancer

cBioPortal was used to analyse genetic variations (mutation, structural variant, amplification, and deep deletion) in PTTG1 across malignancies. In the majority of cancer types, PTTG1 gene amplification was obvious. High amplification was observed in KIRC, CHOL, and UCS (Supplementary Fig. 1A, B). The GSCA database was then used to analyse the frequency of change in single nucleotide variation (SNV) and copy number variation (CNV) in the PTTG1 gene. The findings revealed that SNV of PTTG1 is more frequently found in UCEC (0.88%, Fig. 3B) and STAD (0.68%, Fig. 3B), with low frequency in the other cancer types (Fig. 3A). Survival analysis between SNV and wide type of PTTG1 in TCGA UCEC and STAD cohorts was performed by GSCA. Survival analysis was performed for groups with at least two samples. The results revealed that the patient’s survival rate has no significant difference between SNV and wide type of PTTG1 groups in UCEC and STAD (P > 0.05, Supplementary Fig. 2). The most common detrimental mutation (Missense_Mutation, Splice_Site, Frame_Shift_Del, etc.) in the PTTG1 gene is missense mutation. Point mutation analysis revealed that the SNV of PTTG1 had C > T and C > A transversions (Fig. 3C). Then we analyzed the variant functional annotation of PTTG1 SNV by FAVOR. The results showed that most C > T and C > A transversions occurring in the exons of the PTTG1 gene can enhance the function of the PTTG1 protein, with 5-160427736-C-T being the most relevant (Supplementary Table 1). Next, we looked for CNV of the PTTG1 gene across malignancies. The findings revealed that CNV of PTTG1 was common in KIRC, ACC, CHOL, BLCA, TGCT, and LUSC, but uncommon in THCA, LAML, and PRAD (Fig. 3D). We investigated at the Kaplan-Meier survival curves for PTTG1 CNV (homo amplification and homo deletion) and broad type (Fig. 3E). KIRP patients with homo deletion and homo amplification of PTTG1 genes exhibited a poor prognosis compared to the wild type group (P < 0.05, Fig. 3F).

Fig. 3.

Genetic alteration analysis of PTTG1 in various TCGA tumors. (A) The profile of SNV (Single Nucleotide Variation) of PTTG1 in the TCGA cohorts. (B) The lollipop plot presents the mutation site, type, and count of PTTG1 in the sample set of TCGA UCEC and STAD cohorts. (C) The SNV classes of PTTG1 in the TCGA cohorts. (Variant Classification): the count of each type of deleterious mutation; (Variant Type): the count of SNP and DEL of PTTG1; (SNV class): the count of each SNV class of PTTG1. (D) Pie plot summarizes the CNV (Copy Number Variation) of PTTG1 in the TCGA datasets. (E) The difference of survival between PTTG1 CNV and wide type in the TCGA datasets. (F) Overall survival (OS), progression free survival (PFS), disease free interval (DFI) and disease specific survival (DSS) of PTTG1 CNV and wide type in the TCGA KIRP cohort

Protein-protein interactions of PTTG1 and similar genes in pan-cancer

In order to investigate the probable mechanism of PTTG1 in carcinogenesis, we identified the top 100 genes with similar expression patterns to PTTG1 in the TCGA cancer cohort using GEPIA2. The top five genes with similar expression patterns ordered by Pearson correlation coefficient were AURKB (R = 0.69, P < 0.0001), CDC20 (R = 0.70, P < 0.0001), CCNB1 (R = 0.69, P < 0.0001), KIF2C (R = 0.66, P < 0.0001), and RAD54L (R = 0.64, P < 0.0001) (Fig. 4A). Metascape was used to visualise and analyse the interactome network composed of the top 100 PTTG1 related genes. The PPI network has 90 nodes and 599 edges, indicating a wide range of interactions between these proteins (Fig. 4B). The MCODE algorithm was then applied to this network to find sites where proteins are highly linked. Following that, each MCODE network underwent GO enrichment analysis. The results revealed that five MCODE complexes were discovered in the PPI network, with ‘cell cycle’ being the most common biological meaning (Fig. 4C). To further study the functions of the top 100 PTTG1 similar genes, GO enrichment analysis and KEGG pathway analysis were used. These proteins have a significant role in various cellular processes, including nuclear division, organelle fission, microtubule binding, and tubulin binding (P < 0.0001) (Fig. 4D). Pathway enrichment revealed that PTTG1-related genes function in multiple pathways, including cell cycle and DNA replication (P < 0.0001) (Fig. 4D). These findings indicated that cancer-related genes and pathways were common in PTTG1 similar genes.

Fig. 4.

PPI, GO and KEGG analysis of PTTG1 and similar genes. (A) Correlation analysis of PTTG1 and top 5 similar genes in TCGA database. (B) PPI network of PTTG1 and top 100 similar genes, where the MCODE complexes are colored according to their identities. (C) The five MCODE complexes identified in PPI network by Metascape, colored by their identities. The top-three functional enriched terms of each MOCDE network were listed at the right bottom. (D) Cellular component, molecular function, biological process, and KEGG analysis of PTTG1 and similar genes

Correlations between PTTG1 expression and immune cell infiltration in pan-cancer

We investigated the relationship between immune cell infiltration and PTTG1 gene expression in 33 malignancies using GSCA. ImmuCellAI was performed to get the infiltrates of immune cells in each TCGA sample. In most types of cancer, PTTG1 correlated positively with T cell exhaustion, Th1, B cells, DC, effector memory T cells, Infiltration Score, CD8 naive T cells, CD8 T cells, cytotoxic T cells, and nTreg, but negatively with CD4_T, CD4_naive, central memory T cells, MAIT, neutrophil, NKT, and Th17 cells (Fig. 5A, Supplementary Table 2). The ESTIMATE package in R was used to determine the connection between ESTIMATEScore, ImmunoScore, and StromalScore and PTTG1 expression in TCGA pan-cancer datasets. PTTG1 expression showed a positive correlation with ImmunoScore, StromalScore, and ESTIMATEScore in GBMLGG and KIPAN (Pearson R > 0.2 and Pearson P value < 0.05). PTTG1 expression had a negative connection with ImmunoScore, StromalScore, and ESTIMATEScore in GBM, TGCT, READ, STES, and LUSC (Pearson R<-0.2 and Pearson P value < 0.05, Fig. 5B, Supplementary Table 3). The connection of PTTG1 with immune cell infiltration may help to explain the poor prognosis caused by its high expression.

Fig. 5.

Correlation analysis between PTTG1 expression and immune infiltration. (A) Heatmap summarizes the significance of P value and FDR for the spearman correlation analysis between PTTG1 expression and immune cells’ infiltrates (*: P value < 0.05; #: FDR ≤ 0.05). (B) The correlation between PTTG1 expression and ImmunoScore, StromalScore or ESTIMATE Score

The expression level of PTTG1 is associated with response to immunotherapy treatment

To investigate the possibility of PTTG1 as an immune checkpoint blockade response biomarker, the relationship between PTTG1 and immunotherapy response was examined using the TIDE database. The results showed that high expression of PTTG1 significantly affected the efficacy of immune checkpoint blockade (anti-PD1) and adoptive T cell therapy (ACT), reducing the OS of patients in the Braun2020_PD1_Kidney_Clear cohort, Lauss2017_ACT_Melanoma cohort, and Riaz017_PD1_Melanoma_lpi. Prog cohort (Fig. 6A.B). To assess the accuracy of PTTG1 as an immunotherapy response biomarker, we examined it with other biomarkers previously associated with tumour immune evasion by TIDE. The area under the receiver operating characteristic curve (AUC) was used to assess the predictive ability of these biomarkers. PTTG1 produced an AUC greater than or equal to 0.5 in 11 of the 16 immune checkpoint blockade sub-cohorts (Fig. 6C). PTTG1 outperformed previously published biomarkers in predicting immunotherapy prognosis for melanoma patients (Nathanson2017_CTLA4_Melanoma_Post cohort, AUC = 0.8636 and Gide2019_PD1 + CTLA4_Melanoma cohort, AUC = 0.7000) (Fig. 6C). These findings revealed that PTTG1 played a role in antitumor immune response and promoted immunotherapy resistance.

Fig. 6.

Correlations between PTTG1 expression and immunotherapy response. (A) The plot showed the prognostic value of PTTG1 versus published biomarkers in melanoma and kidney renal clear cell carcinoma cohort. The x-axis shows the z-score on Cox-PH regression and the y-axis indicates its significance level (two-sided Wald test) (B) The association of PTTG1 with patients’ overall survival through Kaplan-Meier curves in melanoma and kidney renal clear cell carcinoma patients with immunotherapy treatment. (C) The bar chart showed the correlation between PTTG1 and published biomarkers in the immunotherapy cohorts. The AUC was used to evaluate the predictive performance of the test biomarker on the immunotherapy response state

PTTG1 exacerbates CTL dysfunction and contributes to resistance against immunotherapy interventions

The relationship between PTTG1 and Cytotoxic T lymphocyte (CTL) dysfunction was evaluated by TIDE. The results showed that a higher CTL level indicates better patient survival, but only when PTTG1 has a low expression level in glioma (OS, z = 2.81, P = 0.00501), myeloma (OS, z = 3.66, P = 0.00025), LUAD (OS, z = 2.25, P = 0.0241), LUSC (OS, z = 2.02, P = 0.0432), COAD (OS, z = 3.08, P = 0.0021), and KIRC (PFS, z = 2.75, P = 0.00604) (Fig. 7A and Supplementary Table 4). This finding suggests that a higher PTTG1 level in tumours reduces the positive connection between CTL and survival. Gene co-expression study revealed that the expression of PTTG1 was positively correlated with immune inhibitory marker such as EDNRB and C10orf54. And negatively correlated with immune stimulatory marker such as HMGB1, CD70, TNFSF9 and TNFRSF18 (Fig. 7B). Tumour mutation burden (TMB) and homologous recombination deficit (HRD) are two clinically important immunological markers that are closely related to tumour immunotherapy. The link between PTTG1 expression and TMB or HRD scores was examined utilising the Sangerbox platform. We discovered a strong positive connection between PTTG1 mRNA expression and TMB or HRD scores in the majority of malignant tumours, particularly KICH (Fig. 7C and D). These results suggested that PTTG1 was related to promoting tumor immune escape and immunotherapy resistance.

Fig. 7.

Mechanisms of PTTG1 promoting immunotherapy resistance. (A) The relationship between PTTG1 and CTL dysfunction. (B) Association of PTTG1 with immune checkpoint genes in pan-cancer. (C) The correlations between PTTG1 expression and TMB scores with Pearson correlation. (D) The correlations between PTTG1 expression and HRD scores with Pearson correlation. *P < 0.05

PTTG1 mainly expressed in regulatory T cells and proliferative T cells

To investigate the probable ways by which PTTG1 affects theted tumor immune microenvironment, public single-cell RNA-seq (scRNA) datasets were used to determine the expression level of PTTG1 on various immune cells. We discovered that PTTG1 was extensively expressed in immune cells, but it was particularly high in regulatory T cells (Treg) and proliferative T cells (Tprolif) in NSCLC, KIRC, and LIHC cohots (Fig. 8A, B, C). Interestingly, immunotherapy responders’ hepatic progenitor cells and malignant cells expressed PTTG1 more than non-responders in LIHC patients, indicating that PTTG1 was related with immune suppressive cells and immunotherapy resistance (Fig. 8D).

Fig. 8.

Single-cell data analysis of PTTG1 expression by TISCH2. Cellular clusters and single-cell PTTG1 expression profile in different cellular types of NSCLC (A), KIRC (B), and LIHC (C). (D) Single-cell PTTG1 expression profile of non-responder and responder in LIHC patients with PD-L1/CTLA-4 treatment

Correlation of PTTG1 and drug response

The CellMinerCDB was used to look into the relationship between PTTG1 expression and drug sensitivity (IC50) in the GDSC dataset. Pearson’s correlation analysis demonstrated a negative connection between high PTTG1 expression and the -log10(IC50) values of 22 drugs (Fig. 9A, Supplementary Table 5), particularly MEK1/MEK2 inhibitors such as RDEA119, trametinib, CI-1040, selumetinib, and PD-0325901 (Fig. 9B). These data showed that increasing PTTG1 could reduce drug sensitivity for these small compounds. PTTG1 showed a substantial positive correlation with the -log10(IC50) values of 13 drugs (Fig. 9A, Supplementary Table 5), including NSC-207895 (Fig. 9C) and ABT-263 (Navitoclax) (Fig. 9E). The results suggested that these 13 small compounds have the potential to be PTTG1 target medicines. A molecular docking model was developed to investigate the mechanism of interaction between PTTG1 and prospective target medicines. We discovered that NSC-207895 could bind to PTTG1 and remain in the binding pocket surrounded by critical residues (Lys1189, Pro1144, Leu1140, His1142, Cys1160, and Arg1197) (Fig. 9D). ABT-263 could attach to PTTG1 and remain in the binding pocket surrounded by critical residues (Ala1671, Arg1675, Cys928, Glu925, Ala929, Leu945, Arg1638, Lys1645, Asp941, Gly931, Gln933, and Gln1674) (Fig. 9F). Our findings suggested that PTTG1 could be a therapeutic target for MEK1/MEK2 inhibitor resistance. ABT-263 and NSC-207895 may be effective anticancer drugs that target PTTG1.

Fig. 9.

PTTG1 increased the drug resistance of cancer cells. (A) The correlation between PTTG1 expression and the sensitivity [-log10(IC50)] of GDSC drugs. A depiction of the correlation between PTTG1 expression and MEK1/MEK2 inhibitor. (B) NSC-207895 (C) and ABT-263 (E) response obtained from the GDSC panel. (D) The predicted interaction of NSC-207895 with PTTG1 protein. Left: NSC-207895 binding with the pocket of PTTG1 is composed of hydrogen bonds. Right: the 2D hydrogen bond interaction pattern of NSC-207895 upon binding to PTTG1 protein. (E) The predicted interaction of ABT-263 with PTTG1 protein. Left: ABT-263 binding with the pocket of PTTG1 is composed of hydrogen bonds. Right: the 2D hydrogen bond interaction pattern of ABT-263 upon binding to PTTG1 protein

Validation of PTTG1 expression and function in NSCLC

We further confirmed the expression of PTTG1 in tumours by utilising the GEO database. The findings showed that NSCLC tissues had considerably greater levels of PTTG1 expression than did normal lung tissues (P < 0.01, Fig. 10A). Additionally, we used the online Kaplan Meier plotter database to determine the predictive significance of PTTG1 in 1161 patients with LUAD. Poor patient survival was observed to be substantially correlated with high expression of PTTG1 (low vs. high: 110.27 vs. 60.73, P < 0.01) (Fig. 10B). In the multi-drug resistance lung cancer dataset GSE77209, which included the parental cell lines H1299, H1355, and paclitaxel-carboplatin-resistant cells, we examined the expression of PTTG1. In H1299 and H1355, resistant cells expressed more PTTG1 than the parental cells did (Fig. 10C). The expression of PTTG1 in chemoresistance cells was confirmed by qRT-PCR and Western Blot. A higher expression of PTTG1 was found in the paclitaxel-resistant cell lines A549-TXR and H358-TXR than in the parental cells, A549 and H358 (Fig. 10D and F). We employed short interfering RNAs to knock down PTTG1 in A549- TXR and H358- TXR in order to ascertain the biological roles of PTTG1 in paclitaxel-resistant cells (Fig. 10E). The result showed that PTTG1 knockdown reduced A549-TXR and H358-TXR’s capacity to generate clones (Fig. 10G) and migrating (Fig. 10H). Our results revealed that PTTG1-related genes function in cell cycle (Fig. 4D). Then we verified the cell cycle related protein CCNB1, CDC2 expression which is in the PTTG1 PPI network. Western blot analysis showed that the expressions of PTTG1, CCNB1, and CDC2 were upregulated in PTX resistance cells (Fig. 10F). PTTG1 siRNA treatment inhibited the expression of CCNB1, CDC2 in A549-TXR and H358-TXR cells (Fig. 10I), indicating that a role of PTTG1 in cell cycle regulation. These results imply that PTTG1 is involved in both chemoresistance and carcinogenesis, cell cycle regulation may be the mechanism.

Fig. 10.

The validation of PTTG1 characterization in lung adenocarcinoma. (A) The expression of PTTG1 was analyzed using the Wilcoxon signed-rank test in GSE32863, GSE19188 and GSE31210 NSCLC datasets. (B) Kaplan–Meier curves and log-rank test of PTTG1 in NSCLC data sets. (C) The mRNA expression of PTTG1 in H1299, H1355 parental and resistance cells in GSE77209 dataset. T[n]: Resistant cells generated after ‘n’ cycles of paclitaxel-carboplatin treatment. (D) The mRNA expression levels of PTTG1 in parental cells A549, H358 and paclitaxel resistant cells A549- TXR, H358- TXR were detected by RT‑qPCR. (E) PTTG1 siRNA knockdown efficiency in A549- TXR and H358- TXR measured. (F) PTTG1, CDC2 and CCNB1 were upregulated in A549- TXR and H358- TXR. (G) Clone formation assays with A549- TXR and H358- TXR cells transfected with PTTG1 siRNA. (H) Transwell assays with A549- TXR and H358- TXR cells transfected with PTTG1 siRNA. (I) PTTG1, CDC2 and CCNB1 were decreased after PTTG1 siRNA treatment measured by Western blot in A549-TXR and H358-TXR cells. *P < 0.05, **P < 0.01

Discussion

Human life is seriously threatened by cancer. It is useful to employ tumour biomarkers which are produced by tumours or the body’s response to tumours during their genesis and development, as targets for tumour screening, early diagnosis, and prognosis prediction [23, 24]. PTTG1 is involved in the growth of tumours and may function as an oncogene in the development and growth of tumours [25, 26]. Qi Zhou et al. found that PTTG1 is involved in the reprogramming of asparagine metabolism by binding to its promoter to promote hepatocellular carcinoma (HCC) progression [6]. PTTG1 is highly expressed in HCC and is associated with poor prognosis in patients. Its mechanism may be related to stabilizing β-catenin and promoting its accumulation in the nucleus, leading to abnormal activation of the Wnt/β-catenin signaling pathway [27]. We analyzed the role of PTTG1 in pan-cancer and confirmed that PTTG1 was overexpressed in 15 of 33 tumor types of TCGA datasets. The expression of PTTG1 was related to tumor classification in several cancers like ACC, LUAD, LIHC, BRCA, KIRC and KIRP. Moreover, high PTTG1 expression was related to poor patient survival, consistent with the previously reported role of PTTG1 in kidney renal clear cell carcinoma [28] and hepatocellular carcinoma [6]. An increasing number of research have investigated the relationship between genetic changes and cancer progression. We discovered that PTTG1 gene amplification resulted in significant protein expression in most tumours. Pathway enrichment revealed that PTTG1 comparable genes were involved in a variety of pathways, including cancer-related pathways such as ‘cell cycle’ and ‘DNA replication’. These findings imply that PTTG1 has the potential to serve as a tumour diagnostic and prognostic marker. Furthermore, inhibiting aberrant PTTG1 expression at the genetic and protein levels could be a possible therapeutic method for reversing carcinogenesis.

Tumor-produced chemicals and non-cancerous elements make up the tumour microenvironment. It is essential for the initiation, growth, metastasis, and reaction to treatment of tumours. One of the main mechanisms of tumour cell immune escape that leads to immunological dysfunction in cancer patients is T cell depletion [29, 30]. Recent research has shown that T cell exhaustion is considered a mechanism of resistance for cellular immunotherapies [31–33] and a significant marker of poor outcomes in many cancers including breast cancer [34] and renal cell carcinoma [35]. Revitalization of exhausted T cells might improve immunity. In this context, recovered fatigued T cells may be a useful source of predictive biomarkers for a possible target [36]. In this study, we discovered that PTTG1 is positively connected with T cell exhaustion but negatively correlated with the ImmunoScore, StromalScore, and ESTIMATEScore in most types of tumours. Furthermore, PTTG1 was shown to be strongly expressed in immune-modulating cells such as Treg and Tprolif cells from NSCLC, KIRC, and LIHC patients. Treg cells produce an immuno-suppressive environment and are less sensitive to immune checkpoint inhibitors [37, 38]. It was noteworthy that Tprolif cells were able to exhibit high expression levels of immunological exhaustion markers, including PDCD1, HAVCR2, CTLA4, LAG3, and TIGIT [39]. In both KIRC and LIHC, our results demonstrated a strong positive correlation between PTTG1 and the immune-suppressive genes HAVCR2, CTLA4, LAG3, TIGIT, and PDCD1. These findings imply that the immune-suppressive microenvironment and T cell exhaustion may be connected to the elevated expression of PTTG1.

The immuno-suppressive tumour microenvironment makes immune checkpoint drugs significantly less effective than anticipated in the treatment of cancer [40]. In lung adenocarcinoma, PTTG1 has been shown to induce senescence and play a critical role in immunotherapy response [41]. Additionally, PTTG1 knockdown has been found to enhance radiation-induced antitumor immunity, further highlighting its involvement in therapeutic resistance [8]. Furthermore, PTTG1 has been linked to poor prognosis in kidney renal clear cell carcinoma, with implications for drug resistance and immunotherapy responses [28]. The gene has also been associated with enhancing oncolytic adenovirus entry into pancreatic cancer cells, suggesting a potential mechanism for immune modulation in this context [42]. To investigate PTTG1’s involvement in immunotherapy resistance, the link between PTTG1 and immunotherapy response was examined. According to the results, patients with KIRC and melanoma undergoing immune checkpoint inhibition (anti-PD1) and adoptive T cell treatment (ACT) had a worse prognosis when their expression of PTTG1 was elevated. An important part of the antitumor immune system is played by cytotoxic T lymphocytes (CTLs) [43]. Immunotherapy resistance and tumour immune evasion are significantly aided by CTL malfunction [44]. We found that higher CTL levels predicted longer patient life only when PTTG1 expression was low in glioma, myeloma, NSCLC, COAD, and KIRC. As a result, increased PTTG1 levels in tumours will reduce the positive correlation between CTL and survival. TMB and HRD are prognostic indicators for immune checkpoint inhibitor clinical effectiveness [45]. Particularly in KICH, we discovered a significant positive relationship between PTTG1 and TMB or HRD scores in most malignant tumours. These findings imply a relationship between tumour immunotherapy resistance and PTTG1 overexpression. The process might be connected to PTTG1’s role in increasing cytotoxic T lymphocyte malfunction and mediating T cell exhaustion. It is possible to view PTTG1, which is expressed on Treg and Tprolif cells, as a novel therapeutic target to overcome immunotherapy resistance.

An increasing amount of research indicates that PTTG1 plays a crucial role in chemosensitivity [46]. Pancreatic ductal adenocarcinoma patients with decreased PTTG1 have good treatment response with great sensitivity and selectivity [47]. In ovarian cancer cell lines, downregulating PTTG1 improved saracatinib sensitivity [48]. PTTG1 gene suppression reduces prostate cancer cell sensitivity to paclitaxel-induced apoptosis [49]. Targeting the STAT3/PTTG1 pathway, as demonstrated by Falcarindiol (FAD), may offer a promising approach to enhance chemosensitivity in HCC cells and potentially other cancer types [9]. Consistent with earlier research, we found that PTTG1 was linked to chemosensitivity, particularly in MEK1/MEK2 inhibitors such as PD-0325901, RDEA119, trametinib, CI-1040, and selumetinib. ABT-263 and NSC-207,895 may be employed as efficient anticancer medication by targeting PTTG1, according to the molecular docking studies. Furthermore, we confirmed the expression and function of PTTG1 in NSCLC. We discovered that PTTG1 was considerably greater in NSCLC tissues compared to non-tumor lung tissues. Furthermore, elevated PTTG1 expression was found to be substantially associated with poor patient survival in NSCLC. PTTG1 expression levels in paclitaxel-resistant cell lines A549-PTX and H358-PTX were much higher than in the parental cells A549 and H358, which is consistent with recent studies in LUAD [50]. PTTG1 knockdown inhibited clone formation and migration ability of A549- TXR and H358- TXR cells. We also found that PTTG1 regulated the cell cycle related gene CCNB1 and CDC2 expression in A549- TXR and H358- TXR cells, which was consistent with our previous study. Our study indicated that PTTG1 may be associated with chemoresistance and carcinogenesis, cell cycle regulation may be the mechanism.

Conclusion

This study shows that PTTG1 plays an important role in tumor progression. High expression of PTTG1 holds the potential to be a tumour diagnostic and prognostic biomarker. In addition, increased PTTG1 expression is associated with resistance to tumour immunotherapy and chemotherapy treatment. The process might be connected to PTTG1’s role in increasing cytotoxic T lymphocyte malfunction, mediating T cell exhaustion and cell cycle regulation. It is possible to view PTTG1, which is expressed on Treg and Tprolif cells, as a novel therapeutic target to overcome immunotherapy resistance. Further research into the molecular mechanisms underlying PTTG1’s involvement in drug resistance and tumor growth is required to understand the mechanisms of PTTG1 regulation. PTTG1-based evaluation of individual tumors enables personalized treatment of tumor patients in the future. The main restriction of this study was only at the level of bioinformatics analysis. Therefore, more cytological experiments, animal experiments and drug trials are urgently needed to clarify the role of PTTG1 in chemotherapy and immunotherapy.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

ACC

Adrenocortical carcinoma

BLCA

Bladder Urothelial Carcinoma BRCA, Breast invasive carcinoma

CESC

Cervical squamous cell carcinoma and endocervical adenocarcinoma

CHOL

Cholangio carcinoma

COAD

Colon adenocarcinoma

DLBC

Lymphoid Neoplasm Diffuse Large B-cell Lymphoma

ESCA

Esophageal carcinoma

GBM

Glioblastoma multiforme

HNSC

Head and Neck squamous cell carcinoma

KICH

Kidney Chromophobe

KIRC

Kidney renal clear cell carcinoma

KIRP

Kidney renal papillary cell carcinoma

LAML

Acute Myeloid Leukemia

LGG

Brain Lower Grade Glioma

LIHC

Liver hepatocellular carcinoma

LUAD

Lung adenocarcinoma

LUSC

Lung squamous cell carcinoma

MESO

Mesothelioma

OV

Ovarian serous cystadenocarcinoma

PAAD

Pancreatic adenocarcinoma

PCPG

Pheochromocytoma and Paraganglioma

PRAD

Prostate adenocarcinoma

READ

Rectum adenocarcinoma

SARC

Sarcoma

SKCM

Skin Cutaneous Melanoma

STAD

Stomach adenocarcinoma

TGCT

Testicular Germ Cell Tumors

THCA

Thyroid carcinoma

THYM

Thymoma

UCEC

Uterine Corpus Endometrial Carcinoma

UCS

Uterine Carcinosarcoma

UVM

Uveal Melanoma

Author contributions

Lihui Wang and Chunlin Zou designed the study. Handong Wei, Yaxin Ma and Shuxing Chen collected the data. All Authors analyzed the data and wrote the manuscript. All authors contributed to the article and approved the submitted version.

Funding

This work was supported by the Joint Project on Regional High-Incidence Diseases Research of Guangxi Natural Science Foundation under Grant (No. 2023GXNSFAA026338), the National Natural Science Foundation of China (81803564), and the National College Students Innovation and Entrepreneurship Training Program (No. 202210598040;No. 202210598038).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.