Abstract
Background/Aim: The 8q24 chromosomal region, which contains the MYC and PVT1 candidate oncogenes, is amplified in carcinomas. Both genes have been involved in the etiopathogenesis of ovarian cancer (OC). In this study, we used an in vitro OC model with a known 8q24 copy number increase and in silico tools to investigate the expression of MYC/PVT1 loci and copy number variation in OC. We also assessed the effects of rucaparib (a PARP inhibitor) in the presence or absence of 10058F4 (a MYC inhibitor) on the expression of MYC/linear PVT1/circular PVT1. Materials and Methods: Tissue culture, chromosome preparation, RNA extraction, RT-qPCR, FISH, and wound healing assays were employed. OncoDB, cBioportal, UALKAN, and ROC Plotter in silico tools were also utilized. Results: Although PVT1 and MYC expression levels remained unaltered in OC, putative copy number alterations across all cancers showed a marked difference between the two genes, particularly in gain and amplification for MYC. PVT1 expression demonstrated prognostic value for the treatment of patients with serous and endometrioid OC. Both genes correlated with PARP10, FAM83H, and DEPTOR. The use of rucaparib in the presence or absence of the MYC inhibitor (10058F4) in vitro, led to a significant down-regulation in the expression of MYC, linear, and circular PVT1. Conclusion: Our data provide a novel insight into the potential interactions of MYC and PVT1 with other genes. Moreover, we identified a new PARP inhibition mechanism down-regulating MYC, as well as the linear and circular PVT1 transcripts. Future work should expand on clinical studies to better understand the prognostic role of PVT1 in OC.
Deletions and amplifications are commonly found in advanced tumor stages. In particular, the 8q24 chromosomal region is frequently amplified in carcinomas. Emerging studies demonstrate that this amplification in ovarian and breast cancers is associated with reduced patient survival (1). This effect is primarily due to the role of the MYC oncogene, mapping at 8q24.21, encoding for a nuclear transcription factor; it has been implicated in the malignant progression of various human tumors (2), including ovarian cancer (OC). Indeed, numerous studies have investigated MYC copy number variation (CNV) in OC and their correlation with clinicopathological parameters using fluorescence in situ hybridization (FISH). For example, in 30% of gynecological malignancies, like endometrioid and epithelial OC, MYC amplification is evident, and there is an association between OC malignancy and MYC CNV (3).
Another emerging candidate oncogene mapping at the same chromosomal region is the long non-coding RNA gene plasmacytoma variant translocation 1 (PVT1). Consistent with its association with various types of cancer, its expression is regulated by TP53 through a canonical p53-binding site (4) and has been implicated in regulating MYC levels (5) to promote tumorigenesis (6, 7). PVT1 maps to 52 Kb downstream of the MYC locus, and it is about 300 Kb pairs in size (8). Interestingly, patients and cell lines with a high copy number gain of MYC and PVT1 show a higher expression of the hsa_circ_0001821 circular RNA (circRNA) obtained from the back splicing of PVT1 exon 2 (circPVT1) than in samples without 8q24 high-copy number gain (9, 10). circPVT1 is presently described as involved in inducing cell proliferation and tumorigenesis. Notably, it is reported as crucial in multiple tumor types, including OC, and there is evidence that it enhances cell proliferation and drug resistance, and inhibits apoptosis. Indeed, it might have a role as a miRNA sponge, as documented for several miRNAs, considering it is enriched in the cytoplasm of tumor cells. Notably, circPVT1 could play an essential role in OC as a sponge of miR-149-5p, leading to the Forkhead Box M1 (FOXM1) over-expression. Furthermore, recent evidence excluded its translation into a functional protein since its longest ORF did not produce a detectable protein at the western blotting level (10).
The role of PVT1/circPVT1 and the interplay with MYC is still under investigation (11). As mentioned, MYC and PVT1 contribute independently to ovarian pathogenesis when over-expressed due to genomic abnormalities (1). Moreover, silencing PVT1 in vitro impaired cell proliferation, migration, and invasion (12). These data are of particular significance, given that OC affects over 300,000 women globally, accounting for more deaths than any other cancer of the female reproductive system (13). Here, we used an in vitro OC model with a known 8q24 low-copy number amplification and in silico tools to investigate the expression of MYC/PVT1 loci and CNV in OC. Furthermore, we assessed the effects of rucaparib (a PARP inhibitor) in the presence or absence of 10058F4 (a MYC inhibitor) on the expression of MYC/PVT1/circPVT1 using RT-qPCR.
Materials and Methods
Tissue culture. In this study, we used the SKOV3 cell line (ECACC 91091004), considered a serous OC cell line characterized by adherent and hypo-diploid cells derived from a patient with OC. The cell line was cultured in T75 cell flasks with a filter head (Nunc; Life Technologies Ltd, Paisley, UK), supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin (Fisher Scientific, Leicestershire, UK). Cells were incubated at 37°C in a humidified atmosphere of 5% CO2 in the air. Cells were subcultured every two or three days at 80-90% confluency by trypsinization with Tryple Express (TrypLE Express, Gibco).
Chromosome preparations. Chromosome suspensions were obtained from SKOV-3 cultured cells using standard protocols (14). Briefly, colcemid (0.05 μg/ml) was added to cell cultures 1 h before harvesting. Then, cells were treated with hypotonic solution (KCl, 0.075 M) and fixed with methanol-acetic acid (3:1). Fixed chromosome suspensions were stored at −20°C until spread on microscope slides.
RNA extraction/RT-qPCR. Cells were plated in a 6-well plate and then treated in a dose- and time-dependent manner with DMSO (0.1%), PARP inhibitor (rucaparib), and MYC inhibitor (10058F4). RNA was extracted using the RNeasy Mini Kit (Qiagen, Manchester, UK). cDNA was synthesized from mRNA utilizing cDNA reverse transcription (Life Technologies). cDNA concentration was measured using RNA concentrations defined by Nano-Drop 2000C (Life Technologies). Relative expression of the genes of interest was measured using quantitative PCR (qPCR) on QPCR QuantStudio 7 Flex Real-Time PCR machine using SYBR® Green PCR Master Mix (Life Technologies) using primers detailed in Table I for PVT1 and MYC, using YWHAZ as a housekeeping gene. The three primer pairs used for the MYC gene could detect all possible splicing variants so far mapped for MYC (UCSC GRCh38/hg38 release). Relative quantities (RQ) values were calculated using the comparative 2−∆∆Ct analysis method (15).
Primers used in RT-qPCR experiments.
Wound-healing assay. Wound healing assay was performed to assess the cell ability to close a created gap in the cell growth area. A ‘scratch’ was made, as a line drawn on one part of the well, by using a 200 μl pipette tip. In the following step, the closure of the scratch was monitored, and images were taken after 24, 48 and 72 h using a Leica DMi1 inverted microscope (40× magnifications; Leica Microsystems, Milton Keynes, UK). The perpendicular line of the marker was used as a landmark to ensure that an image of the same area was taken at each time point. All images were analyzed with ImageJ.
In silico analyses. Xena Functional Genomics Explorer from the University of California Santa Cruz (UCSC) was accessed, and in particular, The Cancer Genome Atlas (TCGA) dataset, to study correlations between the OC cohort samples and healthy ovarian tissue cohort (GTEX-samples). These datasets are enriched with information from studies from different research groups all over the world, thanks to the project called Genomic Data Common Data Portal (GDC) by the National Cancer Institute (NIH), to collect information for each cancer type. The analysis focused on assessing differential gene expression and copy number variations between cancer and the healthy situation, underlining the status of MYC and PVT1 loci at 8q24. Dataset outputs were investigated using RStudio data analysis software. Further analyses were performed using OncoDB (16) and UALKAN (17, 18) online resources that utilize GTEX and TCGA datasets. Copy number variation for MYC and PVT1 was assessed using cBioportal (19). Finally, the ROC Plotter was used to link gene expression with response to therapy (20).
Fluorescence in situ hybridization (FISH). FISH experiments on metaphase chromosomes and interphase nuclei were performed using a selection of probes mapping at different sites on the long arm of chromosome 8 (see Figure 1 for a list of probes and corresponding chromosomal locations). FISH experiments were carried out as previously described (14, 21, 22). Hybridization signals on metaphase chromosomes and interphase nuclei were analyzed using an Olympus AX70 fluorescence microscope, and images were captured using MacProbe v4.3 software (Applied Imaging, Newcastle, UK).
Fluorescence in situ hybridization on SKOV-3 metaphase chromosomes. Localization of genomic regions covered by FISH probes used in this study as shown on the ideogram of chromosome 8 (A) and with reference to the genomic positions according to the GRCh37/hg19 release of the UCSC Human Genome Browser (B). Dual color FISH using probes RP11-440N18-FITC (green) and RP11-125A17-CY3 (red) shows the presence of signals for both probes in the three copies of chromosome 8 (C). Single color FISH using probe RP11-946L14 (detected using streptavidin-Cy3, visible in red) also shows the presence of signals in each one of the three copies of chromosome 8 (D).
RP11-440N18 bacterial artificial chromosome (BAC) targeting MYC, RP11-125A17, and RP11-946L14 BAC probes targeting PVT1 were used in FISH experiments. Probes were labelled directly with fluorochromes or indirectly with biotin or digoxigenin, using nick translation (Roche, Mannheim, Germany) and detected according to previously described standard methods (23). RP11-440N18 was directly labelled with FITC, RP11-125A17 labelled with CY3, whereas RP11-946L14 was labelled with biotin and detected with streptavidin-CY3.
Statistical analysis. The method used for differential analysis in the present study was one-way ANOVA, using disease state (Tumor or Normal) as variables for calculating differential expression: Gene expression disease state. The expression data were first log2 (TPM+1) transformed for differential analysis, and the log2FC was defined as median (Tumor) – median (Normal). Genes with higher |log2FC| values and lower q values than pre-set thresholds were considered differentially expressed genes.
Results
Expression and biomarker utility of MYC and PVT1 in ovarian cancer. Using the TCGA and GTEX datasets, it was demonstrated that, compared to the normal controls (n=180), PVT1 and MYC expression levels remained unaltered in OC patients (n=418) (Figure 2A and Figure 3A). Putative copy number alterations across all cancers showed a marked difference between the two genes, particularly in terms of gain and amplification for MYC (Figure 2B and Figure 3B). The expression of both genes was not dependent on the stage or TP53 status (Figure 2C and D, Figure 3C and D). Finally, only PVT1 demonstrated prognostic value for the treatment of patients with serous (n=1,624) and endometrioid (n=343) OC (Figure 2E and F, Figure 3E and F).
PVT1 expression in ovarian cancer (OC). A) PVT1 expression between normal (n=180) and OC patients (OV, n=418)); B) putative copy-number of PVT1 alterations from GISTIC (across all cancers); C) PVT1 expression in OC across different stages (I-IV); D) PVT1 expression in OC according to p53 status; E) non-responders (n=492) to treatment serous OC patients have higher expression of PVT1 compared to responders (n=1138); and F) non-responders (n=142) to treatment of endometrioid OC patients have higher expression of PVT1 compared to responders (n=201).
MYC expression in ovarian cancer (OC). A) MYC expression between normal and OC (OV) patients; B) putative copy-number of MYC alterations from GISTIC (across all cancers); C) MYC expression in OC across different stages (I-IV); D) MYC expression in OC according to p53 status; E) non-responders to treatment serous OC patients have higher expression of MYC; and F) non-responders to treatment endometrioid OC patients have higher expression of MYC.
Correlation between MYC, PVT1, and chromosome 8 gene expression. To better understand any further “cross-talk”, gene expression correlation was performed among all genes mapping on chromosome 8. By characterizing this chromosomal set, two genes, namely PARP10 and FAM83H, were identified (Figure 4A). We further expanded our observations by creating a PVT1-MYC correlation heatmap in the TCGA cohort for the HUGO dataset, where a correlation with the DEPTOR gene was also noted (Figure 4B).
Heatmap for Chromosome 8 gene expression correlation in TCGA (A); and HUGO (B).
The MYC and PVT1 copy number status assessment and its variation (CNV) in OC (24) was based on the TCGA ovarian gene levels dataset. The GISTIC2 method was applied on entire genome microarray datasets to produce segmented CNV data, which was then mapped to genes to make gene-level estimates. Genes were mapped onto the human genome coordinates using UCSC Xena HUGO probeMap. The information was obtained from 579 OC samples (Figure 5).
Copy number variation of MYC and PVT1. Analyses of 15,241 tumors from the TCGA database showed that 18.02% (2,821 tumors) displayed 8q24 copy-number increase and that 2,746 out of 2,821 tumors (97.34%) showed co-gain of both MYC (A) and PVT1 (B). CNV: Copy number variation; OV: ovarian cancer.
MYC and PVT1 had the same CNV trend, which leads to rearrangements of the 8q24 chromosome band, harboring both genes. This result is in accordance with the literature; in fact, previous analyses of 15,241 tumors from the TCGA database showed that 18.02% (2,821 tumors) displayed 8q24 copy-number increase and that 2,746 out of 2,821 tumors (97.34%) showed co-gain of both MYC and PVT1. In addition, less than 0.15% of tumors showed an increased copy number of MYC but not of PVT1 (25).
Effects of rucaparib (PARP inhibitor) and MYC inhibitor in vitro. Three copies of chromosome 8 in the SKOV-3 cell line were observed using FISH with whole chromosome painting probes (data not shown). Furthermore, FISH using probes specific for the 8q24.21 region, encompassing the MYC and PVT1 loci, showed that those genomic regions were retained in all three copies of chromosome 8 in the SKOV-3 cell line (Figure 1). Studies from our group have shown that 10 μM rucaparib can induce apoptosis in vitro, not only in BRCA2 mutant cells but also in those exhibiting “BRCAness”, like those of the SKOV-3 cell line (26). This serous ovarian adenocarcinoma cancer cell line is one of the most frequently used in the scientific literature (27). Moreover, frequently mutated OC driver genes and copy number alterations have been described for this cell line (28).
Following the validation of SKOV3 as a preclinical model, we performed wound healing assays for up to 72 h to assess the combined impact of a PARP inhibitor (PARPi; rucaparib) and a MYC inhibitor (MYCi; 10058-F4). A number of different concentrations of rucaparib and MYCi were tested to determine whether these agents, alone or in combination, can exert a cytostatic or cytotoxic effect. It was evident that either inhibitor alone (data not shown) or in combination (i.e., PARPi 12.5 μM and MYCi 2.5 μM) were capable of arresting cell growth (Figure 6). Higher concentrations resulted in complete cell death.
Schematic representation showings the healing of the artificial wound on all cell line surfaces (40× magnification). The figure shows the cells before (0-h) and after treatment at 24, 48, and 72 h. The images show the extent of the scratch closure developed under control conditions compared to those following treatment with PARPi (12.5 μM) and MYCi (2.5 μM), and treated cells with PARPi (25 μM) and MYCi (5.0 μM).
Following the in vitro studies, we assessed the impact of both rucaparib (25 μM) and MYCi (10058F4; 10 μM) on the expression of linear (PVT1) and circular PVT1 (circPVT1), as well as the different MYC splicing variant transcripts. Both treatments alone or in combination reduced the expression of PVT1 and circPVT1 (Figure 7), as well as all MYC isoforms (Figure 8), compared to no treatment (basal) levels (p<0.05).
Effect of the rucaparib (25 μM) and MYCi (10058F4; 10 μM) on the gene expression of PVT1 (A) and circPVT1 (B). In both cases, both treatments alone or in combination reduced significantly (p<0.005) the expression of PVT1 transcripts. For the PVT1, rucaparib alone exerted a more significant decrease compared to MYCi or in combination with MYCi (p=0.0024; Panel A). Similarly, for the circPVT1, rucaparib alone exerted a significant decrease compared to MYCi or in combination with MYCi (p=0.0009; Panel B). Error bars: SEM.
Effect of the rucaparib (25 μM) and MYCi (10058F4; 10 μM) on the gene expression of all MYC splicing variants (panels A, B, C). In all cases, both treatments alone or in combination significantly reduced the expression of the MYC isoforms (*p<0.05; ***p<0.001). Error bars: SEM.
Discussion
In this study, we provide a comprehensive overview of the role of MYC and PVT1 in OC, using both in silico and pre-clinical models. While neither of these genes appeared to have an altered expression in OC compared to controls or any TP53- or stage-dependent changes, their role is well-documented in cancer progression (29).
Although PVT1 was not shown to have a diagnostic potential, it might have a clinical utility as a prognostic biomarker, given that patients with OC who do not respond to treatment have higher PVT1 expression. Several studies have indicated an involvement of PVT1 in radio- and chemoresistance (30). For example, when PVT1 was over-expressed in vitro (in SKOV3 and A2780 cell lines), it promoted cisplatin resistance in a pathway that involved apoptotic components like caspase-3 (31). A similar study corroborated these findings, although the authors have demonstrated the involvement of the miR-370/FOXM1 pathway (32). These data collectively argue for targeting PVT1 to reverse chemoresistance in OC.
It is well known that PVT1 expression correlates with that of MYC (30). Here, we provide further insight into chromosome 8 gene correlations. Three genes, namely PARP10, FAM83H, and DEPTOR, demonstrated a good correlation. PARP10 is over-expressed in numerous cancers, including OC, and can be a driver of tumorigenesis and promote cellular proliferation (33). Moreover, Saha et al. have identified PARP10 as a novel marker of platinum response in OC patients (34). Similarly, FAM38H appears to be significantly over-expressed in OC, correlated with the FIGO stage, and involved in cancer progression via a cell-cycle signaling pathway (35). Using esophageal cancer cell lines, Feng et al. have shown that FAM38H-AS1 is involved in epithelial-to-mesenchymal transition (EMT), a fundamental process for cancer progression (36). To date, there is no data on the involvement of this lncRNA in the EMT process in OC. Finally, a correlation with DEPTOR was also noted. This result is increasingly important, given the involvement of mTOR signaling in cancer. We studied extensively the role of mTORC1 and mTORC2 components in OC and how paralogues can be used therapeutically (37-39). Although initial studies suggested that DEPTOR can act as a suppressor of mTOR, the relationship with mTORC1 and mTORC2 appears to be far more complex (40). Moreover, in colorectal cancer, DEPTOR is a target for MYC since it can bind directly to the DEPTOR promoter region and subsequently regulate its transcription (23). Despite the lack of studies showing any mechanistic involvement of PVT1 in the DEPTOR expression, a survey on patients with Li-Fraumeni-like syndrome (LFLS) showed co-amplification of both genes at 8q24.2, suggesting their involvement in LFLS-associated tumors (24).
In the second part of the study, we assessed the effects of rucaparib (a PARP inhibitor) in the presence or absence of 10058-F4 (a MYC inhibitor) in vitro. Poly(ADP-ribose) polymerase (PARP) inhibitors are now used therapeutically in patients with OC harboring homologous recombination repair deficiencies (HRD). Rucaparib is an oral PARP inhibitor that has shown promising results. In a recent randomized, phase III clinical trial, rucaparib monotherapy appeared effective as first-line maintenance therapy in HRD-positive and negative PC patients (41). However, to date, no clinical trials have used 10058-F4 for intervention in any cancer (source: clinicaltrials.gov).
When we performed a wound healing assay, both compounds appeared to exert a strong, not additive, cytostatic/cytotoxic effect in vitro. Data from rucaparib treatment corroborates our previous study on a wider repertoire of OC cell lines, where we have demonstrated that rucaparib significantly decreased cell proliferation (26). Similar results have been previously recorded with the use of the MYC inhibitor. For example, treatment of 2008C13 OC cells with 10058-F4 induced cell cycle arrest at the G1 phase and stimulated the expression of cell-cycle related genes (e.g., p15, p21) (42).
Study limitations. We only performed the experiments in one cell line. Future studies should use a broader range of concentrations and more cell lines representative of high-grade and low-grade serous cancers, as well as cells that have mutations in either BRCA1 or BRCA2, where rucaparib should exert a more significant effect via synthetic lethality. In silico tools can also be restrictive in the case of MYC, due to the expression of multiple isoforms. Future investigations should also concentrate on further functional studies in combination with omics approaches to define the exact pathway(s) involved in these responses. Furthermore, our data further underpin a role for PVT1 in the response to treatment. It is necessary, therefore, to develop PVT1 inhibitors that can be used in combination with other chemotherapeutic agents (e.g., cisplatin, taxol) that can reduce the onset of chemoresistance. More work is needed to gain a much deeper insight into how druggable these genes are prior to embarking on clinical trials.
Conclusion
We assessed the effect of both PARP- and MYC-inhibitors on PVT1, circPVT1, and MYC transcript isoforms using RT-qPCR. All treatments, in combination or as monotherapies, significantly decreased all tested transcripts of the genes mentioned above. Again, similarly to the wound healing assay, the effect was not synergistic when both inhibitors were used. To the best of our knowledge, this is the first time that the impact of these inhibitors is assessed in terms of PVT1 or MYC expression and suggests another way that they can exert their anti-tumor effects.
Footnotes
Authors’ Contributions
Conceptualization, ST, EK, CTS; Methodology, AM, SS, CS, ST, EK; Validation, ST.; Formal Analysis, AM, CS, SS, EK; Draft preparation, ST, EK, CTS; writing – review and editing, AM, SS, CS, MH, ST, EK, CTS; visualization, AM, SS, CS; supervision, MH, ST, EK, and CTS. EK and CTS are joint senior and corresponding co-authors; AM and SS are joined first co-authors.
Conflicts of Interest
The Authors declare no competing interests in relation to this study.
Funding
This study was partially funded through the Cancer Treatment & Research Trust (CTRT), and the Global Thesis Program 2017-2018 (University of Bari Aldo Moro).
- Received January 23, 2024.
- Revision received March 6, 2024.
- Accepted March 8, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.