Search

Introduction High mortality from high-grade serous ovarian cancer (HGSOC) is caused by its late detection due to the absence of early symptoms and lack of early detection tests. The Hedgehog-GLI (Hh-Gli) signaling pathway is involved in ovarian embryonic development and its atypical activation can lead to different types of ovarian tumors. Our microarray-based transcriptome profiling on smaller number of HGSOC tissue samples and healthy Fallopian tubes, from which most of HGSOC originate, has shown that expression of BCL2, one of the Hh-Gli pathway members, was significantly decreased in HGSOC. On the contrary, the expression of two miRNAs that target BLC2, miR-21- 5p and miR-96-5p, was increased in HGSOC. Furthermore, the expression of two lncRNAs that can also bind those two miRNAS, MALAT1 and XIST1, was also decreased. We hypothesize that BCL2 expression could be regulated by a network of endogenous competitive RNA molecules (ceRNAs), which compete for binding available miRNAs and thus regulate the expression of miRNA target- genes. Material and Methods On an expanded set of 62 HGSOC and 19 healthy Fallopian tube tissue samples, we determined by qPCR the expression of BCL2, its transcription factors GLI1 and GLI2, miR-21-5p, miR-96-5p, MALAT1 and XIST1. Based on the obtained expressions we conducted a correlation analysis to build a network of potential ceRNAs. We also performed the ROC curve analysis to determine a diagnostic value of studied RNAs for distinguishing between cancer and healthy tissue. Results and Discussions The qPCR results confirmed results obtained with microarrays: BCL2, MALAT1 and XIST were significantly down-regulated, while miR-21-5p and miR-96-5p were up-regulated in HGSOC. In addition, GLI1 was also decreased in HGSOC. The BCL2 expression was in an expected correlation with the expressions of all other studied RNAs: negative between BLC2 and miRNAs expression, and positive between BCL2 and lncRNAs or GLi1/2. Positive correlation between BCL2 and GLI1 expression was stronger than for GLI2, potentially meaning that GLI1 is more important regulator of BCL2 transcription in HGSOC. Regarding the lncRNAs and miRNAs, there was only a significant negative correlation between XIST and miR-96-5p. Expression of BCL2 and XIST showed the highest area under ROC curve (AUC), thus potentially having an excellent diagnostic value. Conclusion Based on our expression analyses, it seems that BCL2—miR-96-5p— XIST axis could be an epigenetic cause of a decreased BCL2 gene expression in high-grade serous ovarian cancer.

Ovarian carcinoma is the deadliest of all gynecologic cancers. Its high death rate is a result of the fact that >60% patients are diagnosed in advanced stage. Hedgehog-GLI signaling pathway (HH-GLI) is involved in embryonal ovariogenesis but its atypical activity can lead to different types of ovarian tumors. Aberrant HH-GLI activity was observed in some ovarian tumor types previously, molecular causes of which are not jet fully known. We hypothesize that changes in the expression of miRNA molecules related to HH-GLI genes could contribute to the development of high-grade serous ovarian carcinoma (HGSOC), the most malignant and most difficult to detect at an earlier stage. Gene and miRNA expression profiling of 16 HGSOC and 8 healthy fallopian tube tissue samples was performed using Agilent microarray platforms. Data was analyzed using R/Bioconductor packages. Four tools were used to predict target genes of differentially expressed miRNAs. Expression of selected miRNAs and genes was verified using qPCR on an extended set of 10 fallopian tube and 55 HGSOC samples. Various cell-based assays were used to examine biological functions of discovered miRNA/mRNA pairs. Analysis gave a list of 55 miRNAs and 2782 genes differentially expressed in HGSOC. By comparing a list of predicted target genes and differentially expressed HH-GLI genes we revealed several potential miRNA/mRNA interactions. Co-expression of the following combinations has been verified: GLI3/miR-200b-3p ; BCL2/miR-96-5p and miR-21-5p, and IHH/miR-107 and miR-16-5p. Luciferase assays proved physical interactions of miRNAs and GLI3 and IHH 3’UTRs, and their impact on ovarian carcinoma cell biology was identified. We discovered several new miRNAs targeting HH-GLI genes and further studies would verify their usefulness as potential diagnostic and prognostic markers or even therapeutic targets for HGSOC. This study was funded by Terry Fox Foundation, City of Zagreb and Croatian Science Foundation (IP-2016-06-1268).

INTRODUCTION Signalling pathway Hedgehog-Gli (Hh- Gli) is involved in embryonal ovarian development, but its atypical activation can lead to different types of ovarian tumours. Our previous studies showed aberrant Hh-Gli activity in some ovarian tumour types and hypermethylation in a promoter of tumour suppressor gene PTCH1 in the CpG islands near the GLI-binding site (Sabol et al. Int J Oncol 2012, 2017 ; Musani et al. Gene 2013 ; Maurac et al. Int J Gynecol Pathol 2012). We and other are aware that various genetic and epigenetic alterations contribute to aberrant Hh- Gli pathway activity in ovarian cancer, which could serve as interesting targets for cancer treatment and therapy, especially if can be used to bypass resistance and to target cancer in more efficient way. We hypothesize that changes in the expression of miRNA molecules related to the Hh- Gli signalling pathway genes contribute to the development of high-grade serous ovarian carcinoma (HGSOC). MATERIAL AND METHODS We conducted a miRNA profiling of HGSOC and healthy Fallopian tube control samples with Agilent SurePrint G3 Human miRNA 8 × 60K Microarray Kit containing probes for 2549 human miRNAs. Gene expression profiling was preformed using Agilent SurePrint G3 Human Gene Expression v3 8 × 60K Microarray Kit, which covers 37 756 known RefSeq coding transcripts. Data were analysed using R/Bioconductor packages AgiMicroRna and limma. Online DIANA Tools (TarBase, microT-CDS and miRPath) were used to find which Hh-Gli genes are targets of miRNAs differentially expressed in HGSOC. RESULTS AND DICSCUSSION Data filtration gave 55 miRNAs: 32 up- and 23 down-regulated in HGSOC. Out of 47 genes involved in Hh-Gli pathway in humans (KEGG Pathway hsa04340), 35 are known targets for 27 over-expressed miRNAs, 22 are known targets for 16 under-expressed miRNAs, while 28 genes are potential targets for 26 up-regulated miRNAs and 24 are potential targets for 19 down- regulated miRNAs in HGSOC. In addition, 1090 genes were significantly over-expressed and 1692 were under-expressed in HGSOC. Further analysis revealed a couple of potential combinations of miRNAs and their target genes which are members of Hh-Gli pathway: ADRBK2/hsa-miR-96–5 p, BCL2/hsa- miR-16–5 p and hsa-miR-96–5 p, BOC/hsa-miR-224–5 p, and IHH/hsamiR-16–5 p, hsa-miR-107 and hsa-miR- 103a-3p. CONCLUSION Our results highlighted several candidate miRNAs targeting Hh-Gli signalling pathway genes, which, when additionally verified, could be potentially used for better therapeutic and early prevention approaches for HGSOC.

Ovarian cancer is the leading cause of death from gynecological malignancies in the Western world. Its high death rate is a result of the fact that >60% patients are diagnosed in advanced stage of the disease. Hedgehog-GLI (HH-GLI) signaling pathway is involved in embryonal ovarian development, but its atypical activation can lead to different types of ovarian tumors. Our previous studies showed aberrant HHGLI activity in some ovarian tumor types and hypermethylation in the promoter of tumor suppressor gene PTCH1 (Ozretić et al. Int J Oncol. 2017 ; Musani et al. Gene. 2013 ; Sabol et al. Int J Oncol. 2012 ; Maurac et al. Int J Gynecol Pathol. 2012 ; Cretnik et al. Int J Mol Med. 2007). We hypothesize that changes in the expression of miRNA molecules related to the HH-GLI signaling pathway genes contribute to the development of high-grade serous ovarian cancer (HGSOC ; the most malignant type and most difficult to detect at an earlier stage. We conducted miRNA and gene expression profiling of 16 HGSOC and 8 healthy Fallopian tube fresh frozen tissue samples as a control (Reade et al. J Obstet Gynaecol Can. 2014) with Agilent SurePrint G3 Human miRNA 8x60K Microarray Kit and Agilent SurePrint G3 Human Gene Expression v3 8x60K Microarray Kit, respectively. Microarray data was analyzed using R/Bioconductor packages AgiMicroRna and limma. Online tools DianaMicroT, miRDB, RNA22- HSA and TargetScan were used to predict target genes of discovered differentially expressed miRNAs. Expression of selected miRNAs and genes was verified using TaqMan and SYBR Green qPCR, respectively, on an extended set of 10 Fallopian tube and 47 HGSOC samples. Data filtration gave a list of 55 miRNAs differentially expressed in HGSOC: 32 up- and 23 downregulated. In addition, 1, 090 genes were significantly over- and 1, 692 were under-expressed in HGSOC. By comparing a list of predicted miRNA target genes and differentially expressed HH-GLI pathway genes we revealed a couple of combinations of miRNAs and their target genes: GRK3 / hsa-miR-513a-5p ; BCL2 / hsa-miR-96- 5p and hsa-miR-21-5p ; IHH / hsa-miR-103a-3p, hsa- miR-107 and hsa-miR-16-5p ; and GLI3 / hsa-miR- 200b-3p. Co-expression of all those target genes / miRNAs has been confirmed in an extended set of samples. Additional in vitro verification will be continued with in vivo approach. Our results highlighted several candidate miRNAs potentially targeting HH-GLI pathway genes. When additionally verified, they could be used as potential diagnostic and prognostic markers or even therapeutic targets for HGSOC. This study was funded by the Terry Fox Foundation, City of Zagreb and Croatian Science Foundation (IP-2016-06-1268).

High-grade serous ovarian cancer (HGSOC), the most common ovarian cancer, is urgent clinical problem (high mortality, no early signs of disease, diagnosis at an advanced stage), and its molecular background is still unknown. Signaling pathway Hedgehog-Gli (Hh-Gli) is involved in embryonal ovarian development, but its atypical activation can lead to different types of ovarian tumors. Our previous studies showed its aberrant activity: in benign ovarian tumors, but not in ovarian cancer, hypermethylation in a promoter of tumor suppressor gene PTCH1 in the CpG islands near the GLI-binding site (Sabol et al 2012, 2017 Int J Oncol, Musani et al 2013 Gene, Maurac et al 2012 Int J Gynecol Pathol). We and others claim that aberrant activity of Hh-Gli pathway in ovarian cancer is result of various genetic and epigenetic alterations. Our hypothesis is that changes in the expression of miRNA molecules related to the Hh-Gli signaling pathway contribute to the development of HGSOC. We did microRNA profiling of HGSOC (Agilent Human miRNA Microarray platform), with 2, 549 human miRNAs represented in miRBase database (Release 21.0) and healthy Fallopian tube samples as a control (Reade et al., J Obstet Gynaecol Can 2014). Data were analyzed (AgiMicroRna in R/Bioconductor), p-values by Benjamini-Hochberg method, and on-line analyses by DIANA Tools (diana.imis.athena-innovation.gr). Data filtration (IQR > 0.1, FDR < 0.1, logFC > 0.58 and < -0.58) gave 55 microRNAs: 32 up- and 23 down-regulated in HGSOC. Out of 47 genes involved in Hh-Gli pathway in humans (KEGG Pathway hsa04340), 35 are known targets for 27 over-expressed observed microRNAs, while 22 are known targets for 16 under-expressed microRNAs, 28 genes are potential targets for 26 up-regulated microRNAs while 24 are potential targets for 19 down-regulated microRNAs in HGSOC. Those results highlighted candidate microRNAs potentially targeting Hh-Gli signaling pathway genes, when additionally verified, could be better therapeutic and early prevention approaches.

Ovarian carcinoma is the leading cause of death from gynecological malignancies in the Western world. Its high death rate is a result of the fact that most patients (>60%) are diagnosed in advanced stage of the disease. Our experience with ovarian cancer so far came out from the research on genetic (mutations and polymorphisms) and epigenetic (promoter methylation) mechanisms of Hedghog-Gli signaling pathway in the development of various types of ovarian malignancies. The purpose of this study was to find out which Hedgehog-Gli (Hh-Gli) signaling pathway genes could be potential targets of microRNA molecules (miRNAs) differentially expressed in high-grade serous ovarian cancer, which is the most malignant type and most difficult to detect at an earlier stage. We conducted a miRNA profiling on 8 fresh-frozen high-grade serous ovarian cancer tissue samples using Agilent SurePrint Human miRNA Microarray Kit 8x60K which contains probes for 2, 549 human miRNAs represented in miRBase database (Release 21.0). In addition, 8 healthy Fallopian tube samples were used as a control tissue to find differentially expressed miRNAs, since this type of tissue is presumed origin of high-grade serous ovarian cancer (Reade et al. J Obstet Gynaecol Can. 2014). Acquired miRNA expression data were analyzed using LIMMA and AgiMicroRna tools from R/Bioconductor software package. Resulting p-values were adjusted for multiple hypothesis testing based on false discovery rate by the Benjamini-Hochberg method. Furthermore, on-line DIANA Tools (diana.imis.athena-innovation.gr) were used to find which Hh-Gli pathway genes are experimentally proven or potential targets of observed differentially expressed miRNAs. Data filtration (IQR>0.1, FDR0.58 and