The oncogenic transcription factor ERG represses the transcription of the tumour suppressor gene PTEN in prostate cancer cells

Multi-Omics Analysis Reveals Clinical Value and Possible Mechanisms of ATAD1 Down-Regulation in Human Prostate Adenocarcinoma

Prostate adenocarcinoma (PRAD) is the most common histological subtype of prostate cancer. Post-treatment biochemical recurrence is a challenging issue. ATAD1 (ATPase Family AAA Domain Containing 1) plays a vital role in mitochondrial proteostasis and apoptosis activity, while its clinical value in PRAD and its impact on the tumor microenvironment (TME) remain unanswered. In this study, we aimed to investigate the clinical value and possible mechanisms of ATAD1 in PRAD via multi-omics analysis. Using cBioPortal, we confirmed that ATAD1 alteration was associated with gene expression and unfavorable DFS. Deep deletion predominantly occurred in PRAD. By integrating DriverDBv3 and GEPIA2, we noted ATAD1 downregulation in PRAD tissues compared to normal tissues, associated with unfavorable DFS in PRAD patients. DNA repair genes ATM, PARP1and BRCA2 had positive associations with ATAD1 expression. We found that the generalization value of ATAD1 could be applied to other cancers such as KIRC and UCEC. In addition, LinkedOmics identified that the functional involvement of ATAD1 participates in mitochondrial structure and cell cycle progression. Using TIMER analysis, we demonstrated that ATAD1 downregulation correlated with an immunosuppressive TME. Furthermore, we accessed a GSE55062 dataset on UALCAN and discovered the involvement of ERG-mediated transcriptional repression on ATAD1 downregulation. Cross-association screening of shATAD1 efficacy vs. altered mRNAs identified 190 perturbed mRNAs. Then, functional enrichment analysis using the Metascape omics tool recognized that shATAD1-perturbed mRNAs are primarily in charge of the activation of Wnt/β-catenin pathway and lipid metabolic processes. In conclusion, multi-omics results reveal that ATAD1 downregulation is a clinical biomarker for pathological diagnosis and prognosis for patients with PRAD. Reduced ATAD1 may be involved in the enhanced activity of mitochondria and cell cycle, as well as possibly shaping an immunosuppressive TME. ERG serves as an upstream transcriptional repressor of ATAD1. Downstream mechanisms of ATAD1 are involved in Wnt/β-catenin pathway and lipid metabolic processes.

Ras/ERK and PI3K/AKT signaling differentially regulate oncogenic ERG mediated transcription in prostate cells

The TMPRSS2/ERG gene rearrangement occurs in 50% of prostate tumors and results in expression of the transcription factor ERG, which is normally silent in prostate cells. ERG expression promotes prostate tumor formation and luminal epithelial cell fates when combined with PI3K/AKT pathway activation, however the mechanism of synergy is not known. In contrast to luminal fates, expression of ERG alone in immortalized normal prostate epithelial cells promotes cell migration and epithelial to mesenchymal transition (EMT). Migration requires ERG serine 96 phosphorylation via endogenous Ras/ERK signaling. We found that a phosphomimetic mutant, S96E ERG, drove tumor formation and clonogenic survival without activated AKT. S96 was only phosphorylated on nuclear ERG, and differential recruitment of ERK to a subset of ERG-bound chromatin associated with ERG-activated, but not ERG-repressed genes. S96E did not alter ERG genomic binding, but caused a loss of ERG-mediated repression, EZH2 binding and H3K27 methylation. In contrast, AKT activation altered the ERG cistrome and promoted expression of luminal cell fate genes. These data suggest that, depending on AKT status, ERG can promote either luminal or EMT transcription programs, but ERG can promote tumorigenesis independent of these cell fates and tumorigenesis requires only the transcriptional activation function.

ERG is the most common oncogene in prostate cancer. The ERG protein can bind DNA and can activate some genes and repress others. Previous studies indicated that ERG cannot promote cancer by itself, but that ERG works together with mutations that activate the protein AKT. In this study we found that activation of AKT changes the genes that ERG regulates, leading to luminal epithelial differentiation, which is a hallmark of most prostate tumors. However, we also found that a mutant version of ERG that can activate, but cannot repress genes, can drive prostate tumorigenesis without activation of AKT, but this mutant ERG cannot promote luminal differentiation. Our findings suggest that ERG mediated tumorigenesis only requires ERG’s activation function and can occur independent of luminal cell differentiation.

Targeting the ERG oncogene with splice-switching oligonucleotides as a novel therapeutic strategy in prostate cancer

BACKGROUND

The ERG oncogene, a member of the ETS family of transcription factor encoding genes, is a genetic driver of prostate cancer. It is activated through a fusion with the androgen-responsive TMPRSS2 promoter in 50% of cases. There is therefore significant interest in developing novel therapeutic agents that target ERG. We have taken an antisense approach and designed morpholino-based oligonucleotides that target ERG by inducing skipping of its constitutive exon 4.

METHODS

We designed antisense morpholino oligonucleotides (splice-switching oligonucleotides, SSOs) that target both the 5' and 3' splice sites of ERG's exon 4. We tested their efficacy in terms of inducing exon 4 skipping in two ERG-positive cell lines, VCaP prostate cancer cells and MG63 osteosarcoma cells. We measured their effect on cell proliferation, migration and apoptosis. We also tested their effect on xenograft tumour growth in mice and on ERG protein expression in a human prostate cancer radical prostatectomy sample ex vivo.

RESULTS

In VCaP cells, both SSOs were effective at inducing exon 4 skipping, which resulted in a reduction of overall ERG protein levels up to 96 h following a single transfection. SSO-induced ERG reduction decreased cell proliferation, cell migration and significantly increased apoptosis. We observed a concomitant reduction in protein levels for cyclin D1, c-Myc and the Wnt signalling pathway member β-catenin as well as a marker of activated Wnt signalling, p-LRP6. We tested the 3' splice site SSO in MG63 xenografts in mice and observed a reduction in tumour growth. We also demonstrated that the 3' splice site SSO caused a reduction in ERG expression in a patient-derived prostate tumour tissue cultured ex vivo.

CONCLUSIONS

We have successfully designed and tested morpholino-based SSOs that cause a marked reduction in ERG expression, resulting in decreased cell proliferation, a reduced migratory phenotype and increased apoptosis. Our initial tests on mouse xenografts and a human prostate cancer radical prostatectomy specimen indicate that SSOs can be effective for oncogene targeting in vivo. As such, this study encourages further in vivo therapeutic studies using SSOs targeting the ERG oncogene.

Upregulation of Phosphatase 1 Nuclear-Targeting Subunit (PNUTS) Is an Independent Predictor of Poor Prognosis in Prostate Cancer

Protein phosphatase 1 nuclear-targeting subunit (PNUTS) is ubiquitously expressed and associates with PTEN and protein phosphatase 1 (PP1) to control its activity. The role of PNUTS overexpression has hardly been studied in cancer. In this study, we used immunohistochemistry to quantitate PNUTS expression on a tissue microarray containing 17,747 clinical prostate cancer specimens. As compared to normal prostate epithelium, PNUTS expression was often higher in cancer. Among 12,235 interpretable tumors, PNUTS staining was negative in 21%, weak in 34%, moderate in 35%, and strong in 10% of cases. High PNUTS expression was associated with higher tumor stage, classical and quantitative Gleason grade, nodal stage, surgical margin, Ki67 labeling index, and early biochemical recurrence (p < 0.0001 each). PNUTS expression proved to be a moderate prognostic parameter with a maximal univariable Cox proportional hazard for PSA recurrence-free survival of 2.21 compared with 5.91 for Gleason grading. It was independent from established prognostic parameters in multivariable analysis. Comparison with molecular data available from earlier studies using the same TMA identified associations between high PNUTS expression and elevated androgen receptor expression (p < 0.0001), presence of TMPRSS2:ERG fusion (p < 0.0001), and 8 of 11 chromosomal deletions (3p13, 5q21, 8p21, 10q23, 12p13, 13q14, 16q24, and 17p13; p < 0.05 each). Particularly strong associations with PTEN and 12p13 deletions (p < 0.0001 each) may indicate a functional relationship, which has already been established for PNUTS and PTEN. PNUTS had no additional role on outcome in PTEN-deleted cancers. In conclusion, the results of our study identify high PNUTS protein levels as a predictor of poor prognosis possibly linked to increased levels of genomic instability. PNUTS measurement, either alone or in combination, might be of clinical utility in prostate cancers.

C-MYC, HIF-1α, ERG, TKT, and GSTP1: an Axis in Prostate Cancer?

To analyze putative biomarkers for prostate cancer (PCA) characterization, the second leading cause of cancer-associated mortality in men. Quantification of the expression level of c-myc and HIF-1α was performed in 72 prostate cancer specimens. A cohort of 497 prostate cancer patients from The Cancer Genome Atlas (TCGA) database was further analyzed, in order to test our hypothesis. We found that high c-myc level was significantly associated with HIF-1α elevated expression (p = 0.008) in our 72 samples. Statistical analysis of 497 TCGA prostate cancer specimens confirmed the strong association (p = 0.0005) of c-myc and HIF-1α expression levels, as we found in our series. Moreover, we found high c-myc levels significantly associated with low Glutatione S-transferase P1 (GSTP1) expression (p = 0.01), with high Transketolase (TKT) expression (p < 0.0001). High TKT levels were found in TCGA samples with low GSTP1 mRNA (p < 0.0001), as shown for c-myc, and with ERG increased expression (p = 0.02). Finally, samples with low GSTP1 expression displayed higher ERG mRNA levels than samples with high GSTP1 score (p < 0.0001), as above shown for c-myc. Our study emphasizes the notion of a potential value of HIF-1α and c-myc as putative biomarkers in prostate cancer; moreover TCGA data analysis showed a putative crosstalk between c-myc, HIF-1α, ERG, TKT, and GSTP1, suggesting a potential use of this axis in prostate cancer.