Transcriptome profiling in ovarian cancer cells treated with platelets reveals that TGFBI as a novel prognostic indicator

doi:10.21203/rs.3.rs-2230378/v1

Download PDF

Research Article

Transcriptome profiling in ovarian cancer cells treated with platelets reveals that TGFBI as a novel prognostic indicator

https://doi.org/10.21203/rs.3.rs-2230378/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Ovarian cancer is a gynecologic malignancy with poor prognosis. Present prognostic models for ovarian cancer focus on clinico-pathological parameters, quantifiable prognostic biomarkers at molecular level are urgently needed. Platelets contribute to ovarian cancer progression, thus we aimed to search for new predictors in platelet-treated ovarian cancer cells.

Methods

Microarrays analysis was done with platelet-treated SKOV3 cells and controls (4 replicates in each group). Studies on ovarian cancer cells co-incubated with platelets were searched in the Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) were identified by R language. Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment were conducted using online software Metascape. Venn diagram was generated to present common DEGs. Candidate genes were determined by protein-protein interaction (PPI) network, Cox proportional hazards model and Kaplan-Meier analysis. The functions of candidate genes were predicted using data from TCGA by R software, and validated by in vitro experiments.

Results

One dataset (GSE155546) met the inclusion criteria and were analyzed with our microarray data. A total of 4553 mRNAs were differentially expressed between the two groups from our own data, whereas 260 genes exhibited significantly differential expression in GSE155546. DEGs involved in extracellular matrix (ECM) organization and system development were found in both datasets. There were 88 overlapping genes between the two datasets. TGFBI was proved to be an independent adverse factor for ovarian cancer. In addition, high expression of AFT3 and CXCL1 showed worse prognosis in ovarian cancer, while IGFBP7 behaved as a protective predictor. Only increased expression of TGFBI led to significant decrease of overall survival (OS), progression-free survival (PFS) and post-progression survival (PPS), therefore TGFBI was selected as the candidate gene. Functionally, TGFBI was predicted to be significantly correlated with epithelial mesenchymal transition (EMT) markers, degradation of ECM, collagen formation and ECM-related genes. In vitro experiments demonstrated that TGFBI could affect the migration and invasiveness of ovarian cancer cells by regulation E-cadherin, Vimentin, N-cadherin and MMP2.

Conclusion

We found TGFBI as a novel prognostic indicator using platelet-treated ovarian cancer model. Functionally, TGFBI could promote ovarian cancer progression by EMT induction and ECM remodeling.

Ovarian cancer

Platelets

Prognosis

TGFBI

Epithelial mesenchymal transition

Extracellular matrix

Ovarian cancer is a gynecologic malignancy with high death rates worldwide, resulting in 210,000 deaths per year [1]. In western countries, ovarian cancer is the most fatal gynecologic neoplasms, with a 5-year cause-specific survival of 47% on average [2, 3]. The drift of ovarian cancer mortality was 0.76% per year from 1990 to 2019 in China, which exhibited a sustained growth trend as well [4]. Present prognostic models for ovarian cancer survival mainly focus on clinico-pathological parameters like age at diagnosis, tumor stage and histological subtype [5]. Nonetheless, ovarian cancer possesses complex cell-cell interactions and molecular regulation mechanisms, thus qualitative predictors are limited by their efficiency and consistency among patients [6]. Alternatively, quantifiable prognostic biomarkers at cellular and molecular level should be explored for identification of the high-risk patients and personalized medicine [7].

Platelet count in peripheral blood has gradually been recognized as an extracellular predictor for ovarian cancer. An early study of 70 patients in Israel found that ovarian cancer patients with thrombocytosis (༞400×10⁹/L) had worse overall survival (OS) [8]. In a large study of 816 patients in China, thrombocytosis was associated with significantly increased hazards of disease progression and death in ovarian cancer [9]. A multi-institution study that included 1308 patients with clear cell or serous ovarian cancer reported that thrombocytosis was an independent predictor of both progression-free survival (PFS) and OS [10]. However, clinical guidelines for monitoring platelets as a prognostic indicator in ovarian cancer have yet to be determined, since platelet levels can be easily affected by many factors, such as thromboembolism, acute inflammation and stress response [11].

Recently, our group has published a relevant experimental study, revealing that platelets could promote the invasiveness of ovarian cancer cells by inducing a unique transcriptome profiling [12]. Hence, we aimed to perform a microarrays analysis for platelet-treated ovarian cancer cells to find out endogenous genes as stable predictors. To avoid false findings generated by limited samples size in a single independent study, integrated analysis with collective datasets was recommended [13]. Unlike other publicly available databases mainly including tumor samples and normal tissues, Gene Expression Omnibus (GEO) also contains tumor cells with platelet-treatment. Thus, comprehensive analysis was carried out between our microarray results and GEO sample series.

Generally, we first explored differentially expressed genes (DEGs) using data from our microarray and GEO. This was followed by Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways enrichment. Next, core genes were identified by protein-protein interaction (PPI) network and Cox proportional hazards regression analysis, and those showing significant hazard ratios for ovarian cancer patients in The Cancer Genome Atlas (TCGA) were considered as key genes. Among them, those significantly affecting prognosis of patients in Kaplan-Meier plotter database were selected as candidate genes. Finally, the functions of candidate genes were predicted using data from TCGA by R software, and validated by in vitro experiments.

2.1. Cell Line

Human ovarian cancer cell line SKOV3 was acquired from the National Infrastructure of Cell Line Resource. The Cell line was genetically validated by STR profiling using a 20-STR amplification protocol on the ABI 3730XL Genetic Analyzer (Applied Biosystems, Paris, France). SKOV3 cells were grown in RPMI Medium 1640 (Gibco BRL, NY, USA) containing 10% fetal bovine serum (FBS) (Biochannel Biological Technology Co., Ltd., Nanjing, China) and 1% Penicillin-Streptomycin (Gibco BRL, NY, USA).

2.2. Platelets Preparation

Fresh whole blood samples from healthy donors were collected in ethylene diamine tetraacetic acid (EDTA)-containing tubes and centrifuged at 100g for 10 min. The supernatant plasma was named as platelet-rich plasma (PRP), and was aspirated. Platelets were then pelleted from PRP for 10 min at 1000g, and re-suspended in RPMI Medium 1640.

2.3. Ovarian Cancer Cells Treatment

SKOV3 were seeded in six-well plates at a density of 2.0 × 10⁵ /well. After the cells attached to the wall, they were treated platelets prepared from 1 mL PRP (named as group P). Untreated cells were served as control (named as group C). After 48 h, the cells were harvested and washed for subsequent experiments.

2.4. Gene Microarray

Total RNA was isolated from platelet-treated SKOV3 cells or controls (4 replicates in each group) using Trizol Reagent. The complementary DNA (cDNA) was synthesized, marked and hybridized onto GeneChip PrimeView™ Human Gene Expression Array (Affymetrix, Santa Clara, USA) according to the manufacture’s protocol. The GeneChip Scanner 3000 (Affymetrix, Santa Clara, USA) was used to scan the gene chip, and the data files were obtained using GeneChip Operating Software (GCOS) (Affymetrix, Santa Clara, USA).

2.5. Database Retrieval Strategy

We systematically searched for the relevant studies on the GEO DataSets (https://www.ncbi.nlm.nih.gov/gds/). Our search strategy was shown as follow: [("Ovarian Neoplasms") OR ("Ovarian Neoplasm") OR ("Ovary Neoplasms") OR ("Ovary Neoplasm") OR ("Ovary Cancer") OR ("Ovary Cancers") OR ("Ovarian Cancer") OR ("Ovarian Cancers") OR ("Cancer of Ovary") OR ("Cancer of the Ovary")] AND [("Blood Platelets") OR ("Blood Platelets") OR ("Blood Platelet") OR ("Thrombocytes") OR ("Thrombocyte") OR ("Platelet")]. The inclusion criteria for the eligible studies were that (1) ovarian cancer cells were divided into two groups: one group was treated platelets, the other group was set as the control, with no added platelets; (2) RNA sequencing or gene chip were performed on the ovarian cancer cells; (3) platforms used for sequencing or gene chip were introduced. The exclusion criteria were that (1) samples contained other kinds of cells, e.g. stromal cells, besides cancer cells; (2) ovarian cancer cells and their co-incubated platelets were derived from different species; (3) the data in the downloaded files were incomplete or invalid.

2.6. Differentially Expressed Genes (DEGs) Selection

The raw data expression matrices were normalized using log₂-transformed function. The probe IDs were converted to corresponding gene symbols according to the annotation information of each platform. The DEGs analysis was achieved using the “limma” package in R language (version 4.2.1). The thresholds for DEGs identification were set as |fold change| ༞ 1.50 and p value ༜ 0.05. Volcano plots and heat maps were generated to visualize the distribution of DEGs using TBtools (version 1.049), which is a canonical toolkit developed for integrative data mining and interpretation [14].

2.7. Function and Pathway Analysis

GO provides structured terms to describe gene product attributes based on cellular components (CC), molecular functions (MF) and biological processes (BP), while KEGG database is widely used to explore information about biological pathways involving DEGs. We used Metascape (https://metascape.org/) to perform GO annotation and KEGG pathway enrichment analysis. Metascape is an effective and efficient web-based tool designed for experimentalists to interpret large-scale datasets [15]. Minimum overlap ≥ 3 and p-value < 0.01 were considered statistically significant.

2.8. Integration of Protein-Protein Interaction (PPI) Network Analysis

Venn diagram was generated online (http://jvenn.toulouse.inra.fr/app/example.html) to present common DEGs in the included datasets. The STRING online database (version 11.5) (http://string-db.org/) was used to evaluate the potential interactions among the common DEGs. Interactive pairs with middle confidence (interaction score > 0.4) were selected to construct the PPI network using Cytoscape software (www.cytoscape.org/). Additionally, we used CytoHubba, a plugin of Cytoscape, to obtain the degree of each protein node and identify core genes, where genes of the top 20 degrees were considered as core genes.

2.9. Prognostic and Survival Analysis for Core Genes

Firstly, the prognostic values of the core genes in patients with ovarian cancer were tested on samples from The Cancer Genome Atlas (TCGA) database by Cox proportional-hazard model. The “forestplot” package from R programming language (version 4.2.1) was applied to complete the univariate and multivariate Cox regression analysis. Core genes that showed significant hazard ratio (p value < 0.05) were considered as key genes for further analysis. The nomogram was then plotted using the “rms” package in R software (version 4.2.1) to predict the 1, 3, and 5-year overall survival. Finally, survival curve analysis was performed to validate the effect of key genes on OS, PFS and post-progression survival (PPS) of ovarian cancer patients using samples from Kaplan-Meier Plotter database (http://kmplot.com/analysis/). Key genes that significantly affected ovarian cancer prognosis were regarded as candidate genes.

2.10. Mechanisms Exploration for Candidate Genes

We further predicted functions for candidate genes using samples from TCGA database. The “GSVA” package in R software (version 4.2.0) was employed to present correlation between candidate genes and pathway scores by Spearman correlation, and p values >0.05 were considered statistically significant.

2.11. Lentiviral Infection

The lentiviral vector (GV248) containing shRNA targeting TGFBI and its corresponding scramble control were constructed by Genechem (Shanghai, China), and their sequences were listed in Table 1. The virus infection experiment was performed according to the manufacturer's instructions. Clones of stably transfected cells were obtained through puromycin selection, and the knockdown efficiency was evaluated by western blotting.

2.12. Quantitative Real-time PCR

Total RNA was extracted from cell samples according to the manufacturer's instruction of Trizol reagent (Invitrogen, NY, USA). The reverse transcriptase kit (Takara, Shiga, Japan) was used to synthesize cDNA. Quantitative real-time PCR was performed with Talent SYBR Green Kit (Tiangen, Beijing, China) on a LightCycler 480 analyzer (Roche, Shanghai, China). The primer pair sequences of Vimentin, E-cadherin, N-cadherin and MMP-2 were listed in Table 1, and GAPDH was served as a housekeeping control.

Table 1 Sequences of primer pairs and shRNA

Gene	Sequence (5'-3')
Vimentin	F: TGAAGGAGGAAATGGCTCGTC
	R: GTTTGGAAGAGGCAGAGAAATCC
E-cadherin	F: TGGAGGAATTCTTGCTTTGC
	R: CGCTCTCCTCCGAAGAAAC
N-cadherin	F: ACCAGGTTTGGAATGGGACAG
	R: ATGTTGGGTGAAGGGGTGCTTG
MMP-2	F: CCAACTACAACTTCTTCCCTCGC
	R: GCAAAGGCATCATCCACTGTCTC
GAPDH	F: TGACAACTTTGGTATCGTGGAAGG
	R: AGGGATGATGTTCTGGAGAGCC
TGFBI-shRNA	GGTAACGGCCAGTACACGC
Sracmble-shRNA	GCGTGTACTGGCCGTTACC

2.13. Western Blotting

Cell samples were lysed with cold RIPA lytic buffer (Solarbio, Beijing, China) supplemented with 1% PMSF (Solarbio, Beijing, China). Protein concentration of cell lysate was determined using a BCA protein assay kit (Beyotime, Beijing, China). Protein samples were loaded in equal amounts (20 μg) and separated by sodium dodecyl sulfate polyacrylamide (SDS-PAGE) gel electrophoresis (5% stacking gel at 100 V for 15 min and 10% separation gel at 130 V for 90 min). Protein blots were transferred to polyvinylidene difluoride (PVDF) membranes, and incubated with 5% skimmed milk diluted by Tris-buffered saline with 0.05% Tween 20 (TBST) for 1h at room temperature. The membranes were incubated with primary antibodies (1:1000 dilution for E-cadherin, N-cadherin and MMP-2, 1:5000 dilution for GAPDH) (Proteintech, Wuhan, China) for at 4 °C with gentle shaking overnight. After washed three times in TBST, membranes were incubated with the horseradish peroxidase-conjugated secondary antibodies (1:2500 dilution) (Affinity Biosciences, Changzhou, China)) for 1 h at room temperature. After the membranes were washed four times, signals of specific proteins were detected using a chemiluminescence kit (ThermoFisher, Shanghai, China).

2.14. Wound Healing Experiment

Cells were seeded into 6-well plates at the density of 2.5×10⁵/well and cultured to 100% confluence. A 10 µl sterile pipette tip was used make a straight scratch line in the cell monolayer to create an artificial wound. Thereafter, the cells were washed with PBS and suspended in the RPMI-1640 medium without FBS. Migrating cells were observed for 24 h under a light microscope. For every time point, three pictures were taken for wound gap widths measurement.

2.15. Transwell Assay

Transwell inserts (Costar, MA, USA) with polycarbonate membranes of 8.0-μm pore size were placed in 24-well plates (Costar, MA, USA). The upper chamber was coated with 80 μL mixture composed of Matrigel (Corning, Wiesbaden, Germany) and culture medium (1:7). SKOV3 cells were seeded into the upper chamber with serum-free medium with the amount of 1×10⁵. Subsequently, 600 μl complete medium was added to the lower chamber. After 24 h incubation at 37 °C, the non-invading cells on the upper surface of the filters were wiped off using cotton swabs, and the invading cells on the lower surface were fixed with methanol for 20 min and stained with 0.1% crystal violet (Solarbio, Beijing, China) for 5 min. Cell invasion ability was presented as the number of cells on the lower surface, and the average number of cells from three randomly selected visual fields was calculated. Results were expressed as percentage relative to the control.

2.16. Statistical Analysis

Relative fold-change expressions of real-time PCR were calculated by the 2 ^–ΔΔCt method. Quantification of western blotting and wound healing assay were performed by Image J software (National Institute of Health, Bethesda, USA). Differences between two groups of data were statistically evaluated using Statistical Package for the Social Sciences (SPSS) software 13.0 (SPSS, Chicago, USA) with Student’s unpaired t-test or Mann-Whitney U test. Data were shown as mean ± standard error, and p value <0.05 was considered statistically significant.

3.1. The Included Items from GEO Database

Fifty-one studies were listed by GEO repository (updated to November 2022) according to our search strategy. After screening titles and abstracts, 3 studies met the inclusion criteria. For others, 2 studies did not treat ovarian cancer cells with platelets in accordance with the grouping principles mentioned above, 44 studies only presented a single sample in the GSE datasets, and 2 studies performed RNA sequencing on platelets. Two of the remaining three studies were excluded, since GSE155547 contained mixed-cell samples and human ovarian cancer cells were treated with mouse platelets in GSE 95275. Therefore, GSE155546 were finally included and analyzed with our gene chip data.

In the GSE155546, G164 malignant cells, a kind of high-grade serous ovarian cancer cell line, were treated with or without platelets for 24h (named as group P and C, respectively), and each group had 3 samples for biological repeats. Illumina HiSeq 4000 (GPL20301) was, then, applied for the high-throughput RNA sequencing for ovarian cancer cells.

3.2. Gene Expression Changes in GSE155546 and our own dataset (named as P vs. C)

The volcano plot showed that a total of 4553 mRNAs were differentially expressed between the two groups from our own microarray analysis with a significant level of p value < 0.05 and absolute fold change ≧ 1.5, including 1930 up-regulated and 2405 down-regulated genes (Fig. 1A Right).Whereas, only 260 genes exhibited significantly differential expression in GSE155546, of which 152 ones were highly expressed and 108 were down-expression (Fig. 1A Left). The top 50 DEGs were determined in the two sets respectively, and showed as heat maps (Fig. 1B).

3.3. GO Analysis and KEGG Enrichment for DEGs

GO annotations for DEGs from GSE155546 and our own dataset were categorized into three hierarchical frameworks as CC, MF and BP (Fig. 2A and B). Notably, DEGs located in the extracellular matrix (ECM), related to ECM structural constituent and involved in extracellular matrix organization were found in both datasets. This result strongly indicated that during the process of metastasis promotion, platelets resulted in matrisome alteration in ovarian cancer cells. KEGG cluster also presented that ECM-related pathways were abnormally activated in platelet-treated ovarian cancer cells (Fig. 2C and D). Besides, terms related to development, such as heart and vasculature development, were commonly enriched in the two sets (Fig. 2A), which suggested that platelets facilitated metastasis by reversing development procedure and cells underwent mesenchymal transition [12, 16]. GO enrichment and KEGG pathway analysis provided matrisome as research direction for us to explore key genes in the ovarian cancer progression primed by platelets.

3.4. Core Genes Selection

Altogether, we found 88 overlapping DEGs between the two datasets (Fig. 3A). Among them, 35 genes were commonly up-regulated, 13 genes were commonly down-regulated (Fig. 3B). Moreover, 40 genes showed opposite directions of change: 29 genes were up-regulated in GSE155546 but down-regulated in P vs. C dataset, while 13 genes were down-regulated in GSE155546 but up-regulated in P vs. C dataset (Fig. 3B). The specific gene names and their expression in the platelet-treated groups and control groups were represented as volcano plots (Fig. 3C and D). PPI network analysis showed that there were 42 nodes and 116 edges among the common DEGs (Fig. 4A). The top 20 genes rated by the algorithm in Cytoscape were SERPINE1, COL1A2, MMP2, IL6, F3, IL32, CXCL5, CXCL2, CXCL3, FOSL1, IL11, TNFAIP6, CXCL8, ATF, CXCL1, PPBP, PLAT, IGFBP7, TNFRSF11B and TGFBI (Fig. 4B), and they were selected as core genes for further analysis. It is worth noting that COL1A2, MMP2 and TGFBI were the three genes related to ECM.

3.5. Candidate Genes Determination

Cox proportional-hazard model was used to test the prognostic value of the 20 core genes on 374 patients from TCGA database. Univariate analysis indicated that TGFBI was a prognostic predictor for ovarian cancer patients (HR = 1.16, 95% CI: 1.04–1.29, p < 0.01), and it was still proved to be an independent adverse factor in the multivariate analysis (HR = 1.35, 95% CI: 1.15–1.58, p < 0.01) (Fig. 5B). In addition, high expression of AFT3 and CXCL1 showed worse prognosis in ovarian cancer, while IGFBP7 behaved as a protective predictor according to multivariable analysis (Fig. 5B). Therefore, TGFBI, AFT3, CXCL1 and IGFBP7 were screen out as key genes. The nomogram was plotted to predict 1-, 3-, and 5-year OS of a patient with ovarian cancer using the total points calculated by her key genes, tumor stage, age and race. The C-index of this algorithm reached to 0.65(95%CI: 0.612-1, p < 0.01) (Fig. 5C). The calibration curves of the nomogram showed satisfactory consistency between predicted and actual results (Fig. 5D).

The prognostic values of the 4 key genes were further certified using a cohort of ovarian cancer patients from Kaplan-Meier Plotter database. Patients were spilt by median gene expression, and only high expression of TGFBI led to significant decrease of three prognostic index [OS: HR = 1.35, 95%CI: 1.17–1.56, p < 0.01; PFS: HR = 1.37, 95%CI: 1.19–1.58, p < 0.01; PPS: HR = 1.34, 95%CI: 1.10–1.62, p < 0.01]. Other details could be seen in the Fig. 6. Thus, TGFBI was prioritized as candidate gene for function exploration.

3.6. Function prediction for the candidate gene TGFBI

Spearman correlation analysis was applied to explore the potential mechanisms of TGFBI, which was up-regulated by platelet-treatment, in ovarian cancer metastasis. There was a significant positive correlation between the expression of EMT markers and TGFBI levels (ρ = 0.65, 95%CI: 0.59–0.71, p < 0.01) (Fig. 7A), which indicated that TGFBI might induce ovarian cancer cells to lose adhesive epithelial-phenotype and acquire movable features of mesenchyme. Degradation of ECM (ρ = 0.61, 95%CI: 0.54–0.67, p < 0.01), collagen formation (ρ = 0.61, 95%CI: 0.54–0.67, p < 0.01) and ECM-related genes (ρ = 0.56, 95%CI: 0.48–0.62, p < 0.01) were also significantly positively correlated with TGFBI expression (Fig. 7B, C and D), suggesting that TGFBI might regulate ECM remodeling in ovarian cancer microenvironment.

3.7. Experimental validations for the functions of TGFBI

To verify the bioinformatics prediction of potential functions for the selected candidate gene, we knocked down the expression of TGFBI in the SKOV3 cells using shRNA (Fig. 8A), and performed in vitro experiments to support the findings. Real-time PCR analysis revealed that there was a significant decline for the expression of Vimentin in the shRNA-treated groups (Fig. 8B). Significantly increased expression of E-cadherin and significantly decreased expression of N-cadherin were demonstrated by real-time PCR and western blotting following the TGFBI knockdown (Fig. 8B, C and D). These results suggested that TGFBI had the capacity to affect the motility of ovarian cancer cells via EMT regulation. Besides, stimulation of SKOV3 cells with lentivirus containing TGFBI-shRNA plasmid significantly down-regulated the MMP-2 expression both at the mRNA and protein expression levels in comparison with the scramble control cells (Fig. 8B, C and D), indicating that TGFBI could facilitate tissue barrier destruction by degrading ECM. The conjectures mentioned above were confirmed by wound healing experiment and Transwell assay, in which the TGFBI-shRNA treatment resulted in a significant decrease in the migratory and invasive abilities of ovarian cancer cells compared to the scramble control groups (Fig. 8E and F).

In the current study, we did a bioinformatics analysis for the alterations of transcriptome signature after ovarian cancer cells were co-incubated with platelets. Although much is known about the close association between platelet counts and poor outcome in ovarian cancer, information available about prognostic genes regulated by platelets is still limited. Therefore, we searched for new predictors for ovarian cancer using highly metastatic cell model constructed by platelets, and probe into mechanisms by which they were involved in cancer progression. As a result, we found out TGFBI as a novel gene affecting ovarian cancer patients’ survival, and it might promote metastasis by increasing migration and invasion of ovarian cancer cells via EMT induction and ECM remodeling.

Accumulating evidence suggests that platelets actively contribute to ovarian cancer progression. Patients with thrombocytosis were more likely to show the characteristics of aggressive malignancy in the setting of an adnexal mass at the time of diagnosis, including advance stages and high grades [11]. A prospective study for patients who underwent follow-up found that those with higher platelet counts (༞380×10⁹/L) after completing chemotherapy were also more likely to have evidence of disease progression at the time of second-look surgery [17]. On the one hand, the worsened conditions result from the contribution of platelets to ovarian cancer proliferation. Bottsford-Miller et al. has found that platelet depletion was associated with significant reductions in tumor weight, while platelet transfusion significantly increased tumor weight in orthotopic nude mice [18]. Evidence from another study suggested that the proliferation rate of human and murine ovarian cancer cells increased significantly after co-incubation with platelets as well. On the other hand, platelets play an important role in ovarian cancer metastasis, leading to further aggravation for the patients’ conditions. Haemmerle et al. constructed murine models using intraperitoneal injection of cancer cells, and found that reduced platelet counts significantly diminished the number of metastatic nodules detected in abdominal organs [19]. Malacrida et al. developed an artificial tissue model consisted of mesothelial cells, fibroblasts, adipocytes, and ovarian cancer cells, and demonstrated that platelets could stimulate malignant cell invasion into omental structure by activation of mesothelial cells [20]. Moreover, the invasive capacity of ovarian cancer cells could be improved directly by exposure to activated platelets [12, 21]. Therefore, we applied platelet-treated ovarian cancer cells as model to explore the candidate genes associated with ovarian cancer prognosis.

Finally we identified TGFBI as candidate prognostic factor for ovarian cancer through a series of bioinformatics analysis. The TGFBI gene was firstly identified as a novel gene in the human adenocarcinoma cell line A549 after treatment with transforming growth factor-beta (TGF-β) [22]. It is located on region q31 in chromosome 5 and composed of 17 exons of almost similar sizes that span approximately 34kb [23]. The protein encoded by TGFBI is named as transforming growth factor beta induced protein (TGFBIp), which comprises 683 amino acids with a predicted molecular mass of 68 kDa [24]. Structurally, TGFBIp contains a secretory signal followed by an EMI domain, four tandem consecutive FAS1 domains and a carboxy-terminal RGD motif [24]. FAS1 domains and the RGD motif have been known to bind to integrins on the cell membranes and proteins in extracellular matrix [25, 26], so the physiological functions of TGFBIp are to mediate cell to cell contacts and cell-matrix adhesion [24]. The first reported pathological manifestation related to TGFBI was corneal dystrophies due to its mutations [27]. Recently, growing evidence supported the abnormal TGFBI expression in cancer. The positive rate of TGFBIp expression was significantly higher in tumor tissues than in normal tissues, such as colorectal cancer and esophageal carcinoma [28, 29]. In contrast, loss of TGFBI expression caused by promoter hypermethylation was reported in malignant lung tissue specimens [30]. Subsequently, researchers began to employ TGFBI as a biomarker for cancer prognosis. In a study of 94 patients with oropharyngeal carcinoma, high TGFBI expression in tumor stroma was an independent prognostic factor even after statistical adjustment for age, gender, smoking status, stage, grade and treatment [31]. Cancer-specific survival was significantly better for patients with no cytoplasmic TGFBI expression in clear cell renal cell carcinoma as well [32]. We, for the first time, proposed that high TGFBI expression was associated with poor prognosis in ovarian cancer according to our bioinformatics data.

Although TGFBI showed controversial functions in tumorigenesis [33, 34], it was agreed that TGFBI derived pro-metastasis activity for cancer development. Fico et al. crossed a TGFBI knockout mouse with the MMTV-PyMT model, and their experiments illustrated that deletion of TGFBI led to a dramatic decrease in cancer stem cell content and following lung metastasis [35]. Ween et al. suggested that TGFBIp served an important role in peritoneal metastasis in ovarian cancer, since it could promote ovarian cancer cells adhesion to peritoneal cells [36]. Costanza et al. found that TGFBI depletion directly reduced migration of pancreatic cancer cells [37]. Our in vitro assays demonstrated that TGFBI expression influenced not only migratory abilities but invasive capabilities of ovarian cancer cells. However, the detailed mechanism by which TGFBI facilitated ovarian cancer metastasis remained unclear. Since our bioinformatic results revealed that TGFBI expression was positively correlated with the levels of EMT markers and ECM-related genes, we inferred that TGFBI was involved in the regulation of EMT and ECM remodeling in ovarian cancer. Indeed, we observed that silencing TGFBI in SKOV3 ovarian cancer cells resulted in increased expression of epithelial marker E-cadherin and decreased expression of mesenchymal marker N-cadherin and ECM degradation protein MMP2. Similarly, Chen et al. demonstrated that TGFBI knockdown decreased Vimentin increased E-cadherin in RasB1 prostate cancer cells [38], and Ma et al. found that reduced TGFBI expression in U87 astrocytoma cells impaired their secretion of MMP2 and MMP9 [39], which rendered our results more reliable.

The study surely has several limitations. Firstly, only two databases were employed for initial screen, which might result in overrepresented genes for further analysis. We then had to targeted candidate genes through multiple steps, such as overlapping DEGs identification, PPI network construction and cox proportional-hazard model analysis. Secondly, clinical research should be carried out to verify the prognostic effect of TGFBI on ovarian cancer. Finally, the specific pathways by which TGFBI regulates EMT and ECM degradation need to be explored by more laboratory experiments.

In summary, we screened out TGFBI gene in platelet-treated ovarian cancer cell model through combination of our private microarray results with GEO data, and verified its prognostic values for ovarian cancer patients using a public cohort from TCGA. Mechanically, we proposed that TGFBI could promote the invasiveness of ovarian cancer cells by inducing EMT and reprogramming ECM based on the function prediction and in vitro experiments (Fig. 9). Hence, TGFBI might be a promising prognostic biomarker of ovarian cancer and could be used as a potential target for metastasis treatment.

Acknowledgements

We would like to express our thanks to the generous researchers who have uploaded their experimental data onto online databases.

Authors’ contributions

HW performed the data mining and drafted the manuscript. Yin-hai Xu participated in the data analysis and interpretation. Yi Guo was involved in the study design and critical revision of the manuscript. All authors have read and approved the final manuscript. As the corresponding author, Yi Guo and Yin-hai Xu had the final responsibility for the decision for the article publication.

Funding

This work was supported by the Foundation Research Project (Natural Science Foundation) of Jiangsu Province in China (Grant No. BK20201013), Key Research and Development Program (Social Development) of Xuzhou City in China (Grant No. KC20158) and “Shuang‑Chuang Doctor” Project of Jiangsu Province in China (Grant No. 2020204003).

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F: Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: a cancer journal for clinicians 2021, 71(3):209-249.
Siegel RL, Miller KD, Fuchs HE, Jemal A: Cancer statistics, 2022. CA: a cancer journal for clinicians 2022, 72(1):7-33.
Torre LA, Trabert B, DeSantis CE, Miller KD, Samimi G, Runowicz CD, Gaudet MM, Jemal A, Siegel RL: Ovarian cancer statistics, 2018. CA: a cancer journal for clinicians 2018, 68(4):284-296.
Wang Z, Guo E, Yang B, Xiao R, Lu F, You L, Chen G: Trends and age-period-cohort effects on mortality of the three major gynecologic cancers in China from 1990 to 2019: Cervical, ovarian and uterine cancer. Gynecologic oncology 2021, 163(2):358-363.
Gaitskell K, Hermon C, Barnes I, Pirie K, Floud S, Green J, Beral V, Reeves GK, Million Women Study C: Ovarian cancer survival by stage, histotype, and pre-diagnostic lifestyle factors, in the prospective UK Million Women Study. Cancer Epidemiol 2022, 76:102074.
Hu Y, Zheng M, Wang S, Gao L, Gou R, Liu O, Dong H, Li X, Lin B: Identification of a five-gene signature of the RGS gene family with prognostic value in ovarian cancer. Genomics 2021, 113(4):2134-2144.
Kamble PR, Breed AA, Pawar A, Kasle G, Pathak BR: Prognostic utility of the ovarian cancer secretome: a systematic investigation. Arch Gynecol Obstet 2022, 306(3):639-662.
Menczer J, Schejter E, Geva D, Ginath S, Zakut H: Ovarian carcinoma associated thrombocytosis. Correlation with prognostic factors and with survival. Eur J Gynaecol Oncol 1998, 19(1):82-84.
Chen Y, Zhang L, Liu WX, Liu XY: Prognostic significance of preoperative anemia, leukocytosis and thrombocytosis in chinese women with epithelial ovarian cancer. Asian Pacific journal of cancer prevention : APJCP 2015, 16(3):933-939.
Matsuo K, Hasegawa K, Yoshino K, Murakami R, Hisamatsu T, Stone RL, Previs RA, Hansen JM, Ikeda Y, Miyara A et al: Venous thromboembolism, interleukin-6 and survival outcomes in patients with advanced ovarian clear cell carcinoma. European journal of cancer (Oxford, England : 1990) 2015, 51(14):1978-1988.
Hufnagel DH, Cozzi GD, Crispens MA, Beeghly-Fadiel A: Platelets, Thrombocytosis, and Ovarian Cancer Prognosis: Surveying the Landscape of the Literature. Int J Mol Sci 2020, 21(21).
Guo Y, Cui W, Pei Y, Xu D: Platelets promote invasion and induce epithelial to mesenchymal transition in ovarian cancer cells by TGF-beta signaling pathway. Gynecologic oncology 2019, 153(3):639-650.
Qiang R, Zhao Z, Tang L, Wang Q, Wang Y, Huang Q: Identification of 5 Hub Genes Related to the Early Diagnosis, Tumour Stage, and Poor Outcomes of Hepatitis B Virus-Related Hepatocellular Carcinoma by Bioinformatics Analysis. Comput Math Methods Med 2021, 2021:9991255.
Chen C, Chen H, Zhang Y, Thomas HR, Frank MH, He Y, Xia R: TBtools: An Integrative Toolkit Developed for Interactive Analyses of Big Biological Data. Mol Plant 2020, 13(8):1194-1202.
Zhou Y, Zhou B, Pache L, Chang M, Khodabakhshi AH, Tanaseichuk O, Benner C, Chanda SK: Metascape provides a biologist-oriented resource for the analysis of systems-level datasets. Nat Commun 2019, 10(1):1523.
Mitra S, Tiwari K, Podicheti R, Pandhiri T, Rusch DB, Bonetto A, Zhang C, Mitra AK: Transcriptome Profiling Reveals Matrisome Alteration as a Key Feature of Ovarian Cancer Progression. Cancers (Basel) 2019, 11(10).
Bozkurt N, Yuce K, Basaran M, Kose F, Ayhan A: Correlation of platelet count with second-look laparotomy results and disease progression in patients with advanced epithelial ovarian cancer. Obstetrics and gynecology 2004, 103(1):82-85.
Bottsford-Miller J, Choi HJ, Dalton HJ, Stone RL, Cho MS, Haemmerle M, Nick AM, Pradeep S, Zand B, Previs RA et al: Differential platelet levels affect response to taxane-based therapy in ovarian cancer. Clin Cancer Res 2015, 21(3):602-610.
Haemmerle M, Taylor ML, Gutschner T, Pradeep S, Cho MS, Sheng J, Lyons YM, Nagaraja AS, Dood RL, Wen Y et al: Platelets reduce anoikis and promote metastasis by activating YAP1 signaling. Nat Commun 2017, 8(1):310.
Malacrida B, Nichols S, Maniati E, Jones R, Delanie-Smith R, Roozitalab R, Tyler EJ, Thomas M, Boot G, Mackerodt J et al: A human multi-cellular model shows how platelets drive production of diseased extracellular matrix and tissue invasion. iScience 2021, 24(6):102676.
Holmes CE, Levis JE, Ornstein DL: Activated platelets enhance ovarian cancer cell invasion in a cellular model of metastasis. Clin Exp Metastasis 2009, 26(7):653-661.
Skonier J, Neubauer M, Madisen L, Bennett K, Plowman GD, Purchio AF: cDNA cloning and sequence analysis of beta ig-h3, a novel gene induced in a human adenocarcinoma cell line after treatment with transforming growth factor-beta. DNA Cell Biol 1992, 11(7):511-522.
Puthdee N, Sriswasdi S, Pisitkun T, Ratanasirintrawoot S, Israsena N, Tangkijvanich P: The LIN28B/TGF-beta/TGFBI feedback loop promotes cell migration and tumour initiation potential in cholangiocarcinoma. Cancer gene therapy 2022, 29(5):445-455.
Thapa N, Lee BH, Kim IS: TGFBIp/betaig-h3 protein: a versatile matrix molecule induced by TGF-beta. Int J Biochem Cell Biol 2007, 39(12):2183-2194.
Son HN, Nam JO, Kim S, Kim IS: Multiple FAS1 domains and the RGD motif of TGFBI act cooperatively to bind alphavbeta3 integrin, leading to anti-angiogenic and anti-tumor effects. Biochimica et biophysica acta 2013, 1833(10):2378-2388.
Reinboth B, Thomas J, Hanssen E, Gibson MA: Beta ig-h3 interacts directly with biglycan and decorin, promotes collagen VI aggregation, and participates in ternary complexing with these macromolecules. The Journal of biological chemistry 2006, 281(12):7816-7824.
El Kochairi I, Letovanec I, Uffer S, Munier FL, Chaubert P, Schorderet DF: Systemic investigation of keratoepithelin deposits in TGFBI/BIGH3-related corneal dystrophy. Mol Vis 2006, 12:461-466.
Zhu J, Chen X, Liao Z, He C, Hu X: TGFBI protein high expression predicts poor prognosis in colorectal cancer patients. Int J Clin Exp Pathol 2015, 8(1):702-710.
Ozawa D, Yokobori T, Sohda M, Sakai M, Hara K, Honjo H, Kato H, Miyazaki T, Kuwano H: TGFBI Expression in Cancer Stromal Cells is Associated with Poor Prognosis and Hematogenous Recurrence in Esophageal Squamous Cell Carcinoma. Annals of surgical oncology 2016, 23(1):282-289.
Seok Y, Lee WK, Park JY, Kim DS: TGFBI Promoter Methylation is Associated with Poor Prognosis in Lung Adenocarcinoma Patients. Mol Cells 2019, 42(2):161-165.
Kim HJ, Ahn D, Park TI, Jeong JY: TGFBI Expression Predicts the Survival of Patients With Oropharyngeal Squamous Cell Carcinoma. In Vivo 2020, 34(5):3005-3012.
Lebdai S, Verhoest G, Parikh H, Jacquet SF, Bensalah K, Chautard D, Rioux Leclercq N, Azzouzi AR, Bigot P: Identification and validation of TGFBI as a promising prognosis marker of clear cell renal cell carcinoma. Urol Oncol 2015, 33(2):69 e11-68.
Li B, Wen G, Zhao Y, Tong J, Hei TK: The role of TGFBI in mesothelioma and breast cancer: association with tumor suppression. BMC Cancer 2012, 12:239.
Chiavarina B, Costanza B, Ronca R, Blomme A, Rezzola S, Chiodelli P, Giguelay A, Belthier G, Doumont G, Van Simaeys G et al: Metastatic colorectal cancer cells maintain the TGFbeta program and use TGFBI to fuel angiogenesis. Theranostics 2021, 11(4):1626-1640.
Fico F, Santamaria-Martinez A: TGFBI modulates tumour hypoxia and promotes breast cancer metastasis. Mol Oncol 2020, 14(12):3198-3210.
Ween MP, Lokman NA, Hoffmann P, Rodgers RJ, Ricciardelli C, Oehler MK: Transforming growth factor-beta-induced protein secreted by peritoneal cells increases the metastatic potential of ovarian cancer cells. Int J Cancer 2011, 128(7):1570-1584.
Costanza B, Rademaker G, Tiamiou A, De Tullio P, Leenders J, Blomme A, Bellier J, Bianchi E, Turtoi A, Delvenne P et al: Transforming growth factor beta-induced, an extracellular matrix interacting protein, enhances glycolysis and promotes pancreatic cancer cell migration. Int J Cancer 2019, 145(6):1570-1584.
Chen WY, Tsai YC, Yeh HL, Suau F, Jiang KC, Shao AN, Huang J, Liu YN: Loss of SPDEF and gain of TGFBI activity after androgen deprivation therapy promote EMT and bone metastasis of prostate cancer. Sci Signal 2017, 10(492).
Ma J, Cui W, He SM, Duan YH, Heng LJ, Wang L, Gao GD: Human U87 astrocytoma cell invasion induced by interaction of betaig-h3 with integrin alpha5beta1 involves calpain-2. PLoS One 2012, 7(5):e37297.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Transcriptome profiling in ovarian cancer cells treated with platelets reveals that TGFBI as a novel prognostic indicator

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1. Introduction

2. Materials And Methods

2.1. Cell Line

2.2. Platelets Preparation

2.3. Ovarian Cancer Cells Treatment

2.4. Gene Microarray

2.5. Database Retrieval Strategy

2.6. Differentially Expressed Genes (DEGs) Selection

2.7. Function and Pathway Analysis

2.8. Integration of Protein-Protein Interaction (PPI) Network Analysis

2.9. Prognostic and Survival Analysis for Core Genes

2.10. Mechanisms Exploration for Candidate Genes

2.11. Lentiviral Infection

2.12. Quantitative Real-time PCR

2.13. Western Blotting

2.14. Wound Healing Experiment

2.15. Transwell Assay

2.16. Statistical Analysis

3. Results

3.1. The Included Items from GEO Database

3.2. Gene Expression Changes in GSE155546 and our own dataset (named as P vs. C)

3.3. GO Analysis and KEGG Enrichment for DEGs

3.4. Core Genes Selection

3.5. Candidate Genes Determination

3.6. Function prediction for the candidate gene TGFBI

3.7. Experimental validations for the functions of TGFBI

Discussion

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1